sustainable expansion of irrigated agriculture and ... nap task2...tomato, capsicum and eggplant was...

Sustainable expansion of irrigated agriculture and horticulture

in Northern Adelaide Corridor: Task 2 ‐ Modelling nutrient and

chemical fate to support the long‐term sustainability of the use

of recycled water

APPENDICES 8 to 12

Dirk Mallants, Vinod Phogat, Danni Oliver, Jackie Ouzman,

Yousef Beiraghdar, Jim Cox

Goyder Institute for Water Research

Technical Report Series No. 19/15

www.goyderinstitute.org

Appendix 9 Management of irrigation associated hazards for greenhouse crops | 326

Appendix 9: Management of irrigation associated

hazards for greenhouse crops

Contents Executive summary ........................................................................................................................ 327

1. Introduction ............................................................................................................................ 329

2. Materials and methods .......................................................................................................... 332

3. Results and Discussion ........................................................................................................... 335

4. Conclusions ............................................................................................................................. 342

References ...................................................................................................................................... 343


Executive summary

Long-term evaluation of irrigation induced transformations in the soil is essential for optimal

water management and devising effective irrigation scheduling for crops, including protected

crops. In this study, the multi-component numerical model HYDRUS-1D UNSATCHEM was

used to assess the effects of long-term (2018-2050) irrigation on salt build-up in soil under

greenhouse conditions. Blended water (recycled water and harvested rainwater) was used to

irrigate soil grown greenhouse vegetables (tomato, cucumber, capsicum, and eggplant).

Simulations provided insight into the development of irrigation induced chemical

transformations in the soils and which management options provide for a sustainable use of

irrigation water. The management scenarios include 4 leaching fractions (LF), i.e. LF0, LF0.15,

LF0.2, and LF0.3, accounting for 0, 15, 20, and 30% excess water applied, respectively. We

also consider four annual levels of gypsum application, i.e. 0 (G0), 10 (G10), 15 (G15) and 20

(G20) meq/kg soil, respectively. The model simulated annual root water uptake by cucumber,

tomato, capsicum and eggplant was 303, 476, 642 and 649 mm, respectively, in response to a

temperature based irrigation schedule that did not account for a leaching fraction. Whereas,

annual drainage component accounted for 4.1- 6.1% of the amount of irrigation to different

crops. On the other hand, the average salinity in the soil solution (ECsw) at the end of the

simulation (year 2050) increased to 6.5, 7.6, 8.7 and 9.3 dS/m, respectively, for cucumber,

tomato, capsicum and eggplant. Results further indicate that salinity continues to build up in

the soil at a nearly constant rate. Similarly, the ESP values at the end of the simulation were

30.8, 27.1, 33.2, and 31.4 % for tomato, capsicum, eggplant and cucumber crops, respectively.

Based on initial ESP values in the range 19-25, this increase indicates a high sodicity

development in the soil. Results show that temperature based irrigation scheduling for

greenhouse crops (with a zero leaching fraction) could lead to the emergence of salinity and

sodicity hazards in the soil. Analysis of further management scenarios suggested that a 15-20%

(LF0.15- LF0.2) higher irrigation amount coupled with an annual gypsum addition of 10

meq/kg soil (G10, 1.7 t/ha), kept the salinity and sodicity (SAR, ESP) below critical thresholds.

This scenario created a favourable soil environment for long-term sustainable vegetable

production under glasshouse condition. It reduced the soil pH by 7.5-8 % and maintaining this

within an adequate range (7.74-7.8), allowing for optimal plant nutrient availability. Overall,

our modelling results indicate that inappropriate long-term irrigation applications can

significantly impact soil health. Therefore, adoption of informed irrigation scheduling for the


long-term use of blended water is essential for sustainable glasshouse vegetable production.

Apart from salinity, other constraints such as high ESP of the soils needs to be adequately

addressed by applying the proper amount of soil amendments.


1. Introduction

Protected agriculture is a steadily growing sector all over the world (FAO, 2017), with

greenhouse vegetable production as one of the most important and profitable agricultural

systems. Greenhouses essentially provide an opportunity to partially modify the climatic

parameters for developing an effective growing system under otherwise adverse conditions,

including climate variability and change. In Australia, the protected cropping industry had an

estimated value of $1.8billion at the farm-gate, which accounts for approximately 20% of the

total gross value of production of vegetables and cut flowers (PCA, 2019). Main crops grown

in greenhouses in Australia are tomato, capsicum, eggplant, cucumber and melons, with

tomatoes accounting for about half the area. Nevertheless, most greenhouses (up to 80%) in

Australia are low cost structures with soil grown crops (Parks et al., 2009), though the number

of high technology structures on soilless growing media are increasing in recent years to meet

specific market requirements.

Water availability and its quality are the most important factors that determine the expansion

of protected agriculture. It is well understood that greenhouse crops cannot be grown profitably

with highly saline water (Stanghellini et al., 2007). To cope with these problems, many

irrigators have to make use of several water sources, which are either conventional (surface

and groundwater), or non-conventional (recycled waste water, desalinated water, brackish

water, etc.). The heterogeneous quality of different water sources adds a higher level of

complexity to irrigation water management and their use poses a new challenge to managers

and irrigators (Reca et al., 2018). Therefore, in many arid and semiarid regions, protected

horticulture is increasingly facing the salinization of soil and water resources (Pardossi et al.,

2004). On the other hand, most of the crops including vegetables grown in the greenhouses

are sensitive to water and salinity; therefore, the knowledge of crop response to salinity is

important for optimal crop management (Stanghellini et al., 2007). Similarly, in the Northern

Adelaide Plains (NAP) region of South Australia, yields of soil grown tomato and cucumber

crops had been steadily declining for several years due to high salinity (high sodium and

chloride) levels in the soil (Ferguson et al., 2012). The reason for the decline was found to be

poor quality of the primary water source, which was either saline bore or recycled water with

an EC of 2-4 dS/m. Due to limited storage capacity, only a very limited amount of rain water

was captured from greenhouse roofs that could be blended to dilute the salinity of irrigation

water. Likewise, Shi et al. (2009) observed high nutrient accumulation which leads to higher


soil salinity in long-term greenhouse cultivation. Some nutrients that accumulate in the root

zone may leach downward with the application of irrigation water, possibly causing soil and

groundwater contamination (Kim et al., 2008). On the other hand, most greenhouse farmers

are following irrigation and fertigation practices based on their experiences (Thompson et al.,

2007), without monitoring or controlling the soil water, nutrient and salt status. Therefore,

accurate estimations of crop water requirement are needed to avoid, on the one hand, salt build

up in the soil profile, and on the other hand, excess water application that would negatively

impacts on nutrient availability for plants and leaching of agrochemicals to groundwater

(Blanco and Folegatti, 2004).

Irrigation application decisions within a closed growing structure depends on the accurate

estimation of the crop evapotranspiration (ETc) for effective irrigation scheduling. However,

direct measurement methods for crop water requirement (e.g., weighing lysimeters, stem

gauges) are either expansive or abrasive and may not be practical to implement in commercial

greenhouse settings (Villarreal-Guerrero et al., 2012). Alternatively, simulation models offer

cost effective and non-destructive techniques for crop water use. Among the models, the FAO-

56 methodology (Allen et al., 1998) has been used worldwide to determine crop water

requirements under open field conditions and this methodology can also be applied to

greenhouse crops (Bonachela et al. 2006; Qiu et al., 2015). In order to use the FAO-56

methodology for greenhouse crops, measurement of greenhouse reference evapotranspiration

(ET0) and crop coefficients (Kc) are required (Fernández et al., 2010). Möller and Assouline

(2007) observed that the daily ET0 under a cropped screen house, calculated following the

FAO-56 methodology, averaged 62% of the estimated outdoor value. The main reason for the

reduction in ET0, compared to outdoors, was the significant reduction in solar radiation inside

the greenhouse (56% of open field conditions), while the lower wind speeds inside contributed

to a smaller extent. Farias et al. (1994) observed that the reference evapotranspiration (ET0)

inside greenhouses was always lower than outdoors, ranging on 45 to 77% of that verified

outside. Braga and Klar (2000) observed that the values of ET0 were 85 and 80% of the outdoor

ET0 for greenhouses oriented east/west and north/south, respectively. However, Harmanto et

al. (2005) found that the crop evapotranspiration (ETC), estimated based on the measured

climate data within the greenhouse, matched the 75–80% of the ETC computed with the

climatic parameters observed in the open environment. Several software have been developed

that are based on the FAO-56 method and use a single Kc approach, such as PrHo V2.0


(Fernández et al., 2009). The VegSyst model simulates ETC, crop growth and crop N uptake

of greenhouse-grown vegetable crops (Gallardo et al., 2011, 2016).

Due to the complexity involved in measuring the climate variables and crop data, such as leaf

area index, other simple methods have been explored such as the temperature-based Hargreaves

equation for accurately estimating greenhouse ETo (Fernández et al., 2010). Empirical

approaches widely used under open field conditions that only require temperature and

greenhouse radiation data (Allen et al., 1998; Itenfisu et al., 2003; Gavilán et al., 2006), could

be easily applied for determining greenhouse ET0 (Boulard and Wang, 2000). Nikolaou et al.

(2017) found a good correlation between leaf temperature and transpiration for soilless

greenhouse cucumber crop, with the prediction equation being validated under different

greenhouse climatic conditions.

However, irrigation practices on the ground have often oversimplified the complex nature of

the interacting climate, soil and crop variables involved in the estimation of crop water

requirement. For example, outdoor temperature based irrigation scheduling is practiced by the

growers for irrigating greenhouse crops in the Northern Adelaide Plains by considering only

60% of outdoor ETC (Awad et al., 2019). While this simplified approach requires less data,

there is need to test its effectiveness on a long-term basis. Essentially, observations of soil

health over one or several seasons are unable to reveal the potential adverse impacts of

irrigation induced problems such as salinity and sodicity, which usually arises only after many

years of adopting a particular irrigation schedule.

Therefore, in the present study we evaluate the impact of long-term use of a temperature based

irrigation schedule to soil grown tomato, cucumber, capsicum and eggplant under greenhouse

conditions. We explore management options such as increased leaching fractions and annual

gypsum applications, and seek to identify how to make best use of blended water for

sustainable irrigation in the NAP region. To achieve these objectives, we used the multi-

component major ion chemistry module UNSATCHEM of the HYDRUS-1D numerical model

(Šimůnek et al., 2016). We focus our simulations on understanding the long-term irrigation

induced salinity and sodicity risks to soil and shallow groundwater, and assess the efficacy of

practical management options, including gypsum amendments and increased leaching. In this

study, we are also implementing a new capability in the UNSATCHEM module, i.e. applying

a solid form of gypsum annually for reducing the ESP levels in the soils. The outcomes of these


long-term simulations, accounting for the future climate, would help in devising better

irrigation guidelines for growing vegetable crops under protected environments while also

coping with the adverse impact of climate change.

2. Materials and methods

The study focussed on conducting long-term (1970-2050) simulations using the HYDRUS-1D

simulator with the UNSATCHEM module (Šimůnek et al., 2016) for evaluating the impact of

long-term use of blended water for irrigation of crops grown under low-cost greenhouse

conditions in the NAP region of South Australia. UNSATCHEM is a multi-component major

ion chemistry process based numerical model which requires numerous input data for soil,

crop, water and climate conditions of the study site. Therefore, soil samples from two

greenhouses were collected from the NAP region (Awad et al., 2019) from 3 locations at two

depths (0-20 and 20-30 cm). These samples were analysed for their physico-chemical

properties (Oliver et al., 2019) to generate soil hydraulic and chemical properties for the

UNSATCHEM model. The average van Genuchten-Mualem (van Genuchten, 1980) soil

hydraulic parameters for two locations are given in Table 1 and soil exchange properties are

shown in Table 2. Although a 200-cm-deep domain was adopted for the simulations, soil

hydraulic parameters obtained in 20-30 cm layer of the greenhouse soils were extended to the

bottom of the 200-cm-deep domain.

Table 1 Soil hydraulic properties of two soils (S1 and S2) for two depths (0-20, 20-30 cm) used in the modelling simulations (this report, Appendix 5).

Soil Depth (cm)

Θr (cm3/cm3) Θs (cm3/cm3)

α (cm-1)

n (-)

Ks (cm/d)

l (-)

Bulk density (g/cm3)

S1 0-20 cm 0.1843 0.4581 0.0881 1.1699 23.49 0.50 1.4 S1 20-30 cm 0.2001 0.4222 0.0484 1.1656 46.10 0.50 1.5 S2 0-20 cm 0.1528 0.4896 0.0302 1.2735 49.22 0.50 1.4 S2 20-30 cm 0.1458 0.4376 0.0245 1.2620 75.25 0.50 1.5

Table 2 Soil chemical properties used in the modelling simulations. Gapon coefficients are for sand over clay soil (this report, Appendix 6).

Soil Texture Exch Ca Exch Mg Exch Na Exch K CEC Gapon Coefficient (KG)

-------------------meq/kg-------------------- meq/kg Mg/Ca Ca/Na Ca/K S1 Scl* 62.05 45.47 26.83 17.61 161.52 0.02 1.88 0.014 S1 Scl* 72.72 40.86 21.97 18.85 153.61 0.02 1.88 0.014 S2 Sl# 72.07 35.71 12.68 14.85 133.26 0.02 1.88 0.014 S2 Sl# 53.32 32.39 14.83 13.88 107.36 0.02 1.88 0.014

*sandy clay loam, #sandy loam


A survey of the existing greenhouses conducted in the NAP region (Awad et al., 2019)

indicated four crops are commonly grown under greenhouse conditions (tomato, capsicum,

cucumber and eggplant). Therefore, these crops were considered for the long-term simulations.

The water quality parameters and irrigation schedules for the test crops were obtained from the

IQ-QC2 model developed by Awad et al. (2019). This model generates daily crop water

requirements for crops and the associated water quality parameters based on a user-defined

mixing of available waters for irrigation, such as recycled water, harvested rain water and storm

water. Monthly averaged irrigation water quality data for multiple years was used as shown in

Table 3. Other crop specific inputs such as root water uptake parameters (Feddes, 1974) and

salinity threshold-slope functions (Maas and Hoffman, 1977) are given in Appendix 1 of this

report.

Table 3 Average monthly blended water quality parameters used in the UNSATCHEM simulations.

Month EC Ca Mg Na K Alk SO4 Cl SAR

dS/m ----------------------------------meq/L-------------------------------------- January 0.30 0.37 0.53 2.39 0.02 0.38 0.73 2.24 3.6 February 0.42 0.52 0.74 3.34 0.02 0.54 1.02 3.13 4.2 March 0.41 0.50 0.72 3.26 0.02 0.52 0.99 3.05 4.2 April 0.37 0.45 0.65 2.93 0.02 0.47 0.89 2.75 3.9 May 0.30 0.36 0.52 2.35 0.02 0.38 0.72 2.20 3.5 June 0.22 0.27 0.39 1.77 0.01 0.28 0.54 1.65 3.1 July 0.16 0.20 0.28 1.27 0.01 0.20 0.39 1.19 2.6 August 0.17 0.20 0.29 1.32 0.01 0.21 0.40 1.24 2.7 September 0.23 0.29 0.41 1.85 0.01 0.30 0.57 1.74 3.1 October 0.21 0.26 0.38 1.70 0.01 0.27 0.52 1.59 3.0 November 0.15 0.18 0.26 1.17 0.01 0.19 0.36 1.09 2.5 December 0.11 0.14 0.20 0.90 0.01 0.14 0.28 0.84 2.2

The IQ-QC2 model adopted a temporally uniform 0.6 factor of open field crop

evapotranspiration (ETc) to derive the crop water requirement of the glasshouse crops. The

calculated ETc matches the temperature based irrigation schedule adopted by the local growers,

therefore, we used these values as daily potential crop transpiration in the HYDRUS-

UNSATCHEM simulations. Rainfall and soil evaporation were not considered in the

simulations as these are supposed to be controlled by the presence of closed growing structures

(greenhouse). The daily ETc of all the crops was estimated from the reference crop

evapotranspiration (ET0) and the crop coefficient (Kc) approach (Allen et al., 1998) used for

open field crops. Monthly Kc values for tomato, cucumber, capsicum and eggplant are


provided in the IQ-QC2mixing model as described by Awad et al. (2019). Climate parameters

for the ETc estimation were obtained from the Bureau of Meteorology station at Edinburg Raaf

(34.71°S, 138.62°E; BOM station number 023083) for the historical climate (1970-2018) and

from the Goyder climate change projections for the future climate (2018-2050) (Charles and

Fu, 2015).

More details on the crops, their irrigation schedule, mixing strategy, and quality of blended

water are given in Awad et al. (2019). A schematic representation of the temperature based

irrigation schedule for greenhouse cultivation is shown in Figure 1.

Figure 1 Schematic representation of the temperature based irrigation schedule followed by growers in the NAP region. Initially the model was run on the crop specific temperature based irrigation schedule adopted

by the growers in the NAP region (Awad et al., 2019). Later on the model was run on numerous

management scenarios developed on the basis of the outcome of the initial scenario. These

simulations include 4 leaching fractions (LF, in % excess water applied), i.e. 0, LF0.15 (15%),

LF0.2 (20%), LF0.3 (30%) and four annual gypsum application levels, i.e. 0, 10, 15, and 20

meq/kg soil equivalent to 0, 1.7, 2.6 and 3.4 t/ha designated as G0, G10, G15 and G20,

respectively. These simulations were designed to generate information on managing the

available irrigation waters for their sustainable use under greenhouse cultivation in the NAP

region.


3. Results and Discussion

Model simulated water balance components obtained from the base scenario based on current

farmers’ irrigation practices for different greenhouse crops are shown in Figure 2. The annual

values of root water uptake for all crops (tomato, cucumber, capsicum and eggplant) are close

to the annual irrigation application. The average annual root water uptake by tomato, capsicum,

eggplant and cucumber was 476, 642, 649, and 303 mm, respectively. However, the

corresponding drainage component was only 4.6, 5.4, 6.1 and 4.1% of the total irrigation

application. Such low values are below the required leaching fraction (LF) needed to flush the

accumulated salts from the crop root zone. Typically, open field crop irrigation designs allow

a 10-20% higher water application for salinity control depending on crop, soil and water

quality. In some cases this fraction can be higher than 20% if climate, soil or crop grown

encourage salinity build in the soils. Rhoades and Loveday (1990) suggested a leaching

fraction of 20-50% as an ideal fraction for such conditions including for irrigation with recycled

water.

Figure 2 Model simulated long-term (2018-2050) annual water balance components for a) tomato, b) capsicum, c) eggplant, and d) cucumber crops. Root = root water uptake; Drain = drainage to groundwater.

The simulated annual values of pH, ECsw, SAR and ESP under different crops are shown in

Figure 3. The pH values decreased gradually, although the magnitude of reduction was very

small. At the end of the simulation at year 2050, the average pH values in the soils varied in a

narrow range (8.2-8.5) for all the crops. However, pH values were relatively higher in

cucumber as compared to other crops. These changes in soil pH might be related to a low pH

0

200

400

600

800

Root Drain Irrigation

0

200

400

600

800

0

200

400

600

800

a)

c)

b)

0

200

400

600

800

d)

Ann

ual w

ater

bal

ance

com

pone

nts (

mm

)


of the irrigation water as compared to the soil solution, which gradually brought down the pH

to achieve a quasi-equilibrium.

On the other hand, there was a sharp increase in the profile average ECsw values under all

crops. However, the ECsw build up was the lowest in cucumber, followed by tomato; these

results are consistent with the lower amount of seasonal irrigation applied to those two crops.

Meanwhile, the highest annual ECsw was observed for capsicum and eggplant; this is the result

of the higher irrigation application for these crops and a longer cropping season (Jan to Oct-

Nov) as compared to cucumber (Jan- May) and tomato (Jan- Sept). The profile average ECsw

at year 2050 rose to 6.5, 7.6, 8.7 and 9.3 dS/m for cucumber, tomato, capsicum and eggplant,

respectively. These values are above salinity tolerance threshold ECsw (ECsw was derived

from published ECe values as ECsw = 2ECe) for these crops, i.e. 5, 5, 3.4 and 2 dS/m (Maas

and Hoffman, 1977; Sonneveld and Vogt, 2009).

Bonachela et al. (2018) observed that mean EC of the soil solution extracted with suction cups

under a greenhouse tomato crop ranged between 6 to 7 dS/m, which is similar to the current

study. However, in some cases, farmers may over irrigate to overcome the salinity problem as

leaching practices are not well documented (Bonachela et al., 2018). However, soil salinization

mostly depends on the amount of saline water applied (Meiri, 1984; Shalhevet, 1994).

Therefore, a leaching fraction should be applied well ahead of the salt accumulation in the root

environment reaching hazardous levels (Meiri, 1984; Shalhevet, 1994). Another potential

factor contributing to salt accumulation is an increase in the evaporative demand due to climate

variability (Fernández et al., 2010). Therefore, a soil with continued high salinity values above

crop thresholds can create considerable osmotic impacts which reduces the crop water uptake

and renders the soil unfit for crop production.

On the other hand, soilless production systems typically adopt a large leaching fraction (30-

40%) (Bonachela et al., 2018), which is a fundamental management consideration for these

growing systems to maintain the salinity of the root zone solution at levels which are not

detrimental to optimal crop production (Sonneveld, 2000). To avoid problems caused by

accumulation of salts, an appreciable proportion of applied nutrient solution has to be drained,

the so-called “drain-to-waste” system (van Os, 1995). Therefore, soilless systems such as

hydroponics represent both a substantial loss of water and potentially a major input of salts and

agrochemicals to groundwater.


Figure 3 Simulated annual average values of a) pH, b) EC (dS/m), c) SAR, and d) ESP in the sandy clay loam soil (S1) under tomato (t), cucumber (cu), capsicum, and eggplant crops irrigated with a blend of recycled water (Rw) and rain water (Gw) following a temperature based irrigation schedule.

Similarly, annual average SAR and ESP values in the soil also showed an increasing trend for

all crops. The ESP values at the end of simulation (year 2050) were 30.8, 27.1, 33.2, and 31.4

% under tomato, capsicum, eggplant and cucumber, respectively. However, the initial ESP

values (13-18 %) in the soil were also high, which had increased over the model warming up

period (1970- 2018) and varied for different crops in response to the amount of irrigation water

applied. Undoubtedly, the final ESP values are much higher than the critical ESP (>6), which

leads to the development of sodic soil conditions. Therefore, an annual addition of gypsum is

essential to reclaim the soil from the high ESP values. Alternatively, a soluble Ca application

as part of the irrigation above the crop requirement also helps in keeping the ESP under control.

However, rapid leaching of soluble Ca or likely precipitation in the soil as calcite at high pH

may reduce the effectiveness of added soluble Ca.

Similar changes in soil pH, ECsw, SAR and ESP were also observed in the sandy loam soil

(S2) for all crops (Figure 4). However, the magnitude of these parameters was different for

different crops. These results suggest that the temperature based irrigation schedule for

0

5

10

15

EC

(dS/

m)

ECt ECcuECca ECeg

7

7.5

8

8.5

9

pH

pHt pHcupHca pHeg

0

2

4

6

8

SAR

SARt SARcuSARca SAReg

0

9

18

27

36

ESP

ESPt ESPcuESPca ESPeg

a)

b)

d)

c)


greenhouse crops in the NAP region appears not to suffice to control the buildup of salinity

and sodicity hazards in the soil, especially if such schedule is followed for a long time without

any management options. Additionally, soil salinity normally arises if moderately brackish or

saline waters are used for irrigation, as frequently occurs when recycled water or groundwater

are used for irrigation or blending. Therefore, there is need to optimize the irrigation schedule

with an appropriate leaching fraction (LF) as well as by applying suitable amounts of gypsum

(G) for controlling salinity and sodicity.

Figure 4 Simulated annual average values of a) pH, b) EC (dS/m), c) SAR, and d) ESP in the sandy loam soil (S2), respectively under tomato (t), cucumber (cu), capsicum, and eggplant crops irrigated with a blend of recycled water (Rw) and rain water (Gw) following a temperature based irrigation schedule.

Additional scenarios were performed with leaching fractions (LF0.15, LF0.2, and LF0.3) and

annual gypsum applications (G10, G15, G20) for soil grown greenhouse tomato crop. Results

show that an increased leaching fraction alone (i.e. without gypsum) has a small impact on soil

pH, with values being reduced by 0.3-2.4% only when the LF increases from 0.15 to 0.3,

respectively (Figure 5). However, gypsum addition at an annual rate of 10 meq/kg soil (G10,

1.7 t/ha) reduced the soil pH at all depths by 7.5-8 %. However, further increase in the gypsum

application (G15, G20) had little impact on subsequent reductions in the soil pH. Thus, the

average pH has been reduced to 7.74-7.8, which may not have any adverse impact on the

0

5

10

15

EC

(dS/

m)

ECt ECcuECca ECeg

7

7.5

8

8.5

9

pH

pHt pHcupHca pHeg

0

1

2

3

4

5

6

7

SAR

SARt SARcuSARca SAReg

0

9

18

27

36

ESP

ESPt ESPcuESPca ESPeg

a)

b)

d)

c)


nutrient availability for crops. Therefore, an annual gypsum application of 10 meq/kg soil (1.7

t/ha) seems adequate, at least to maintain the soil pH within a suitable range.

Figure 5 Model simulated annual profile average a) pH, b) ECsw, c) SAR, and d) ESP in the sandy clay loam soil (S1) under different leaching fractions (LF0, LF0.15, LF0.2, and LF0.3) and annual gypsum application (G0, G10, G15 and G20) scenarios for tomato cultivation under greenhouse condition.

Similarly, the ECsw in the soil decreased below the tomato threshold (ECsw = 5 dS/m) for all

LFs (0.15 to 0.3), which suggests that a minimum LF of 0.15 is adequate to maintaining the

soil salinity below the threshold (Figure 5). Interestingly, application of gypsum increased the

soluble ion concentration in the soil, leading to an increase in the ECsw at all depths, although

it was still below the crop threshold. However, it is worth noting that increasing the LF alone

didn’t reduce the SAR and ESP in the soil. Apparently, these impairments require addition of

soil amendments such as gypsum to reduce their impact on soil as well as on crops. Therefore,

an addition of gypsum at an annual rate of 10 meq/kg soil (1.7 t/ha) along with the leaching

fraction LF 0.15 reduced the SAR and ESP below threshold within the 30 cm soil depth.

However, this combination was unable to bring down SAR and ESP below threshold levels at

60-90 cm depth. Further increasing the gypsum levels to 15 (G15) and 20 meq/kg (G20)

combined with the leaching fraction LF 0.15 also did not further reduce SAR and ESP in the

soil at deeper depths. This suggests that larger leaching fractions are required to remove excess

salt and obtained acceptable SAR and ESP values at lower soil depths.

A further increase in the LF to 0.2 combined with and annual addition of gypsum at 10 meq/kg

soil (1.7 t/ha) has reduced the average SAR and ESP below thresholds, also at lower soil depths.

7.0

7.5

8.0

8.5

9.0G

0LF0

G0L

F0.1

5G

10LF

0.15

G15

LF0.

15G

20LF

0.15

G0L

F0.2

G10

LF0.

2G

15LF

0.2

G20

LF0.

2G

0LF0

.3G

10LF

0.3

G15

LF0.

3

pH

0-15 cm 15-30 cm 60-90 cm

0

2

4

6

8

SAR

0

3

6

9

G0L

F0G

0LF0

.15

G10

LF0.

15G

15LF

0.15

G20

LF0.

15G

0LF0

.2G

10LF

0.2

G15

LF0.

2G

20LF

0.2

G0L

F0.3

G10

LF0.

3G

15LF

0.3

ECsw

(dS/

m)

0

9

18

27

36

ESP

(%)

a)b)

c) d)


Other scenarios were also explored in which leaching fractions less than 0.15 were used

together with different gypsum levels (G10, G15 and G20): while the ECsw dropped below

threshold, they were unable to bring down the ESP below the critical threshold (data not

shown). The conclusion is that, for shallow rooted soil grown vegetables, addition of gypsum

at a rate of 10 t/ha with 0.15 LF is able to keep the soil free from salinity and sodicity hazards

under greenhouses conditions. On the other hand, for deep rooted crops, a combination of 0.2

LF with a 10 meq/kg soil or higher annual gypsum addition would help in maintaining

sustainable production under long-term greenhouse conditions.

An annual comparison of profile average pH, ECsw, SAR and ESP values obtained with and

without 0.2 LF along with different annual gypsum additions (0, 10, 15 meq/kg soil) for tomato

cultivation is shown in Figure 6. The G10 with 0.2 LF scenario initially had average profile

SAR and ESP values higher than the threshold (< 6). These reduced gradually to the values

similar to those obtained with 15 meq/kg soil (G15) gypsum at the end of the simulation at year

2050. Similarly, the average SAR values were also reduced below the threshold, the pH

reduced by 8.7% with average values around 7.7, and ECsw was much lower than the tomato

tolerance threshold. Similar results were obtained in the sandy loam soil (S2) with cucumber

(Figure 7). However, the magnitude of variation between G10LF0.2 and G15LF0.2 scenarios

was much lower as compared to tomatoes. Apparently, the amount of irrigation application and

soil properties also influences the effectiveness of LF and gypsum addition for reclaiming the

soils. Other factors such as purity of gypsum, particle size, uniformity of application and

mixing in the soil also influence the effectiveness of gypsum in reducing the ESP in the soil.

These factors were not evaluated in this study.


Figure 6 Effect of leaching fraction (LF0 and LF0.2) and gypsum application [0 (G0), 10 (G10), and 15 (G15) meq/kg soil] on the changes in annual profile average a) pH, b) ECsw, c) SAR, and d) ESP in the sandy clay loam soil (S1) during 2018-2050 under tomato cultivation in greenhouse condition.

Figure 7 Effect of leaching fraction (LF0 and LF0.2) and gypsum application [0 (G0), 10 (G10), and 15 (G15) meq/kg soil] on the changes in annual profile average a) pH, b) ECsw, c) SAR, and d) ESP in the sandy loam soil (S2) during 2018-2050 under cucumber cultivation in greenhouse condition.

Overall, the results from the current modelling exercise suggest that a long-term irrigation

schedule without leaching fraction may lead to salinity and sodicity problems in the soil. There

0

5

10

15

EC

(dS/

m)

7

7.5

8

8.5

9pH

LF0G0 LF0.2G0LF0.2G10 LF0.2G15

0

2

4

6

8

SAR

0

9

18

27

36

ESP LF0G0 LF0.2G0

LF0.2G10 LF0.2G15

a)b)

d)c)

0

5

10

15

EC

(dS/

m)

LF0G0 LF0.2G0

LF0.2G10 LF0.2G15

7

7.5

8

8.5

9

pH

0

2

4

6

8

SAR

LF0G0 LF0.2G0

LF0.2G10 LF0.2G15

0

9

18

27

36

ESP

a)

b)

d)c)


is clearly a need to apply a leaching fraction of 15- 20% for leaching the salts below the crop

root zone. Even the blended water used for irrigation has a strong tendency to build up sodicity

hazards in the soil. Therefore, annual applications of gypsum at 1.7-2.6 t/ha (or preferably on

the basis of soil a test) or other sources such as a Ca solution should be applied to overcome

the sodium hazard in the soil. The survey conducted by Awad et al. (2019) found that many

growers apply organics/compost annually and also apply Ca through nutrient solutions. Both

of these additions help in alleviating sodicity hazards in the soil. However, continued

monitoring of the salinity and sodicity status of the soil is an essential aspect of soil

management and allows one to intervene well before issues occur.

The results obtained from the current study are specific to the crops, soils, climate, and water

quality considered. Several other aspects of protected crops were not considered in this study,

such as different substrates, effect thof ventilation and evaporative cooling. These can also

impact the extent of water use and its management, and related issues such as salinity and

sodicity development. Concentration of nutrient solution, quality of substrate and local issues

such as leachate management may also impact the plant water requirement and its effective

management. Simulation tools such as those used in this study can account for these additional

conditions and therefore help develop guidelines for practitioners aimed at maintaining

sustainable production systems.

4. Conclusions

Protected agriculture is a popular cultivation system especially for vegetable crops. They are

highly efficient systems in term of energy and other input consumption such as water and

fertilizers. However, suitable water is often scarce in the regions where these production

systems exist, such as in the vicinity of urban, arid or semi-arid environments. Therefore, in

most cases, low quality water is used directly or in combination with other available water

resources such as harvested rainwater. Unfortunately, low quality waters are often used without

knowing the harmful impacts on crop productivity, soil health, and the wider environment.

This study used the multi-component major ion chemistry module UNSATCHEM of the

HYDRUS-1D model to evaluate the effects of long-term (2018-2050) irrigation with blended

water for soil grown tomato, cucumber, capsicum and eggplant under unheated greenhouse

conditions. The results revealed that irrigation schedules that do not apply a significant leaching


fraction may lead to high salt build up and ESP development in the soil while accounting for

future climate projections. The soil solution salinity (ECsw) can increase to 6.5-9 dS/m at year

2050 and ESP can rise to 27-33% for all crops considered. These conditions could render the

soil unfit for crop production and could potentially degrade the associated environment.

Therefore, appropriate management options should be implemented to keep the irrigation

induced harmful impacts under control.

The study evaluated the efficacy of increased leaching and gypsum addition to control salinity

and sodicity. For this purpose the UNSATCHEM module was extended with a new capability

to allow annual gypsum applications. Management scenarios with different leaching fractions

for salinity control showed that 15-20% more water per irrigation would be required to keep

the salinity under control for soil grown greenhouse vegetables. Results obtained in various

scenarios for amelioration of soil with high ESP suggest that annual gypsum application at a

rate of 1.7 t/ha was adequate for managing this hazard. However, both management options

(i.e. leaching fraction and gypsum use) need to be implemented simultaneously. Finally, long-

term monitoring of highly efficient greenhouse production systems is essential for early

identification of irrigation induced soil issues.

References

Allen R.G., Pereira L.S., Raes D., Smith M. 1998. Crop evapotranspiration. Guidelines for

computing crop water requirements. FAO irrigation and drainage paper 56. FAO, Rome.

Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: Guidelines for

computing crop water requirements. FAO Irrigation and Drainage Paper No. 56, FAO,

Rome, Italy.

Boulard, T., Wang, S., 2000. Greenhouse crop transpiration simulation from external climate

conditions. Agric. Forest Meteor. 100, 25-34.

Awad, J., Vanderzalm, J., Pezzaniti, D., Olubukoa Esu, O.-O., van Leeuwen, J., 2019.

Sustainable Expansion of Irrigated Agriculture and Horticulture in Northern Adelaide

Corridor: Source water options/water availability, quality and storage consideration.

Goyder Institute for Water Research Technical Report Series No.xx/18, Adelaide, South

Australia.


Blanco F.F, Folegatti M.V. 2004. Evaluation of evaporation measuring equipments for

estimating evapotranspiration within a greenhouse. Rev. Bras. Eng. Agri. Ambiental 8,

184-188.

Bonachela, S., Fernández, M.D., Cabrerac, F.J., Granados, M.R., 2018. Soil spatiotemporal

distribution of water, salts and nutrients in greenhouse, drip-irrigated tomato crops using

lysimetry and dielectric methods. Agric. Water Manage. 203, 151–161.

Bonachela, S., Gonza´ lez, A.M., Ferna´ndez, M.D., 2006. Irrigation scheduling of plastic

greenhouse vegetable crops based on historical weather data. Irrig. Sci. 25 (1), 53–62.

Braga, M.B., Klar, A.E., 2000. Evaporation and evapotranspiration in the field and in

greenhouse oriented north/south and east/west directions. Irrigation 5, 222–228 (in

Portuguese).

Charles, S.P., Fu, G., 2015. Statistically Downscaled Projections for South Australia – Task 3

CSIRO Final Report. Goyder Institute for Water Research Technical Report Series No.

15/1 Adelaide, South Australia.

FAO, 2017. Good agricultural practices for greenhouse vegetable production in the South East

European countries: Principles for sustainable intensification of smallholder farms. Food

and Agriculture Organization of the United Nations, Plant Production and Protection

Paper No 230, FAO, Rome. http://www.fao.org/3/a-i6787e.pdf.

Farias, J.R.B., Bergamaschi, H., Martins, S.R., 1994. Evapotranspiration in plastic

greenhouses. Braz. J. Agrometeorol. 2, 17–22 (in Portuguese).

Feddes, R.A., Kowalik, P.J., Zaradny, H., 1978. Simulation of field water use and crop yield.

Simulation Monographs, Pudoc, Wageningen, The Netherlands.

Ferguson, K., Hall, B., Parker, J.B., Webb, K., Burfield, T., Cavallaro, D., Parks, S., 2012.

Improving greenhouse systems and production practices (Greenhouse production

practices component). Final Report of Project No. VG07144, Horticulture Australia Ltd,

Sydney, Australia.

Fernandez, M.D., Baeza, E., Cespedes, A., Perez-Parra, J., Gazquez, J.C., 2009. Validation of

on-farm crop water requirements (pRho) model for horticultural crops in an unheated

plastic greenhouse. Acta Hortic. 807, 295–300.

Fernández, M.D., Bonachela, S., Orgaz, F., Thompson, R.B., López, J.C., Granados, M.R.,

Gallardo, M., Fereres, E., 2010. Measurement and estimation of plastic greenhouse

reference evapotranspiration in a Mediterranean climate. Irrig. Sci. 28, 497–509.

http://www.fao.org/3/a-i6787e.pdf


Gallardo, M., Fernández, M.D., Giménez, C., Padilla, F.M., Thompson, R.B., 2016. Revised

VegSyst model to calculate dry matter production, critical N uptake and ET c of several

vegetable species grown in Mediterranean greenhouses. Agric. Syst. 146, 30–43.

Gallardo, M., Giménez, C., Martínez-Gaitán, C., Stöckle, C.O., Thompson, R.B., Granados,

M.R., 2011. Evaluation of the VegSyst model with muskmelon to simulate crop growth,

nitrogen uptake and evapotranspiration. Agric. Water Manage. 101, 107–117.

Gavilán, P., Lorite, I.J., Tornero, S., Berengena, J., 2006. Regional calibration of Hargreaves

equation for estimating reference ET in a semiarid environment. Agric. Water Manage.

81, 257–281.

Harmanto, V.M. Salokhe, M.S. Babel, H.J. Tantau (2005) Water requirement of drip irrigated

tomatoes grown in greenhouse in tropical environment. Agricultural Water Management

71, 225–242.

Itenfisu, D., Elliott, R.L., Allen, R.G., Walter, I.A., 2003. Comparison of reference

evapotranspiration calculation as part of the ASCE standardization effort. J. Irrig. Drain.

Eng. ASCE 129(6), 440–448.

Kim, J.H., Choi, C.M., Lee, J.S., Yun, S.G., Lee, J.T., Cho, K.R., Lim, S.J., Choi, S.C.,

Lee,G.J., Kwon, Y.G., Kyung, K.C., Uhm, M.J., Kim, H.K., Lee, Y.S., Kim, C.Y., Lee,

S.T.,Ryu, J.S., 2008. Characteristics of groundwater quality for agricultural irrigation in

plastic film house using multivariate analysis. Korean J. Environ. Agric. 27 (1), 1–9 (in

Korean with English abstract).

Maas, E.V., Hoffman, G.J., 1977. Crop salt tolerance: current assessment. J. Irrig. Drain. Div.

Am. Soc. Civil Eng. 103 (IR2), 116–134.

Meiri, A., 1984. Plant response to salinity: experimental methodology and application to the

field. In: In: Shainberg, I., Shalhevet, J. (Eds.), Soil Salinity under Irrigation: Processes

and Management. Ecological Studies 51. Springer-Verlag, Berlin, pp. 284–297.

Möller M, Assouline S. 2007. Effects of a shading screen on microclimate and crop water

requirements. Irrigation Science 25(2):171-181.

Nikolaou, G., Neocleous, D., Katsoulas, N., Kittas, C., 2017. Modelling transpiration of

soilless greenhouse cucumber and its relationship with leaf temperature in a

Mediterranean climate. Emirates J. Food Agric. 29(12), 911-920.

Oliver, D.P., Fruzangohar, M., Johnston, C., Ouzman, J. and Barry, K., 2018. Sustainable

expansion of irrigated agriculture and horticulture in northern Adelaide corridor. Task 1:

Development and optimisation of modelling domain and impact assessment of irrigation


expansion on the receiving environment, Goyder Institute for Water Research Technical

Report Series No. 18/xx, Adelaide, South Australia.

Pardossi, A., Tognoni, F., Incrocci, L., 2004. Mediterranean greenhouse technology. Chronica

Hortic. 44, 28–34.

Parks, S.E., Worrall, R.J., Low, C.T., Jarvis, J.A., 2009. Initial efforts to improve the

management of substrates in greenhouse vegetable production in Australia. Acta Hort.

819, 331–336.

Qiu, R., Du, T., Kang, S., Chen, R., Wu, L., 2015. Assessing the SIMDualKc model for

estimating evapotranspiration of hot pepper grown in a solar greenhouse in Northwest

China. Agric. Syst. 138, 1–9.

PCA. 2019. Our Industry. Protected cropping Australia.

http://www.protectedcroppingaustralia.com/?page_id=94.

Reca, J., Trillo, C., Sánchez, J.A., Martínez, J., Valera, D., 2018. Optimization model for on-

farm irrigation management of Mediterranean greenhouse crops using desalinated and

saline water from different sources. Agric. Syst. 166, 173-183.

Rhodes JD, Loveday J 1990: Salinity in irrigated agriculture. In Agronomy Monograph No.

30. Ed. BA Stewart and DR Nielsen., pp. 1089–1142. ASA Publishers, Madison.

Shalhevet, J., 1994. Using water of marginal quality for crop production: major issues. Agric.

Water Manag. 25, 233–269.

Shi, W.M., Yao, J., Yan, F., 2009. Vegetable cultivation under greenhouse conditions leads to

rapid accumulation of nutrients, acidification and salinity of soils and groundwater

contamination in South-Eastern China. Nutr. Cycling Agroecosyst.83, 73–84.

Šimůnek, J., van Genuchten M.T., Šejna, M. 2016. Recent developments and applications of

the HYDRUS computer software packages. Vadose Zone J. 15, 25.

Sonneveld, C. (2000). Effects of salinity on substrate grown vegetables and ornamentals in

greenhouse horticulture. Thesis (The Netherlands: Wageningen University), 151 pp.

https://doi. org/10.1007/978-90-481-2532-6.

Sonneveld, C., Vogt, W., 2009. Plant Nutrition of Greenhouse Crops. Springer Science and

Business Media B.V., Dordrecht, The Netherlands.

Stanghellini, C., Pardossi, A., Sigrimis, N., 2007. What limits the application of wastewater

and/or closed cycle in horticulture? Acta Hortic. 747, 323–330.

Thompson, R.B., Martinez-Gaitan, C., Gallardo, M., Gimenez, C., Fernandez, M.D., 2007.

Identification of irrigation and N management practices that contribute to nitrate leaching

http://www.protectedcroppingaustralia.com/?page_id=94


loss from an intensive vegetable production system by use of a comprehensive survey.

Agric. Water Manage. 89, 261–274.

van Genuchten, M.Th., 1980. A closed-form equation for predicting the hydraulic conductivity

of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898.

van Os, E.A., 1995. Engineering and environmental aspects of soilless growing systems. Acta

Hortic. 396, 25–32.

Villarreal-Guerrero, F., Kacira, M, Fitz-Rodríguez, E., Kubota, C., Giacomelli, G.A., Linker,

R., Arbel, A., 2012. Comparison of three evapotranspiration models for a greenhouse

cooling strategy with natural ventilation and variable high pressure fogging. Scientia

Hortic. 134, 210–221.

Appendix 10 Optimising the riparian zone width along a river | 348

Appendix 10 Optimising the riparian zone width near

a river for controlling lateral migration of irrigation

water and solutes

Contents Executive Summary ........................................................................................................................ 349

1. Introduction ............................................................................................................................ 350

2. Materials and Methods .......................................................................................................... 352

2.1. Experimental site ................................................................................................................ 352

2.2. Soil characteristics .............................................................................................................. 353

2.3. Water resources ................................................................................................................. 354

2.4. Vegetation and crops parameters ..................................................................................... 355

2.5. Irrigation scheduling of crops ............................................................................................ 356

2.6. Transport domain, initial and boundary conditions ......................................................... 357

2.7. Calibration and validation of the model ............................................................................ 358

2.8. Scenario analysis ................................................................................................................ 359

2.9. Model evaluation ............................................................................................................... 359

3. Results and Discussion ........................................................................................................... 360

3.1. Calibration and validation of HYDRUS-2D ......................................................................... 360

3.2. Irrigation and annual water balance for crops .................................................................. 362

3.3. Hydraulic exchange at the river-buffer interface .............................................................. 362

3.4. Impact of a buffer width on solute dynamics ................................................................... 365

4. Implications of the findings ................................................................................................... 370

5. Conclusions ............................................................................................................................. 370

References ...................................................................................................................................... 371


Executive Summary

Riparian zones are essential to preserve water quality of rivers adjacent to large areas of

irrigated agriculture. We used HYDRUS (2D/3D) to quantify the long-term (1st July, 2009 to

30th June, 2017) influence of crops (almonds, wine grapes and potato-carrot) irrigated with

DAFF/recycled water (RCW) on water and solute exchange at the Gawler River interface in

relation to vegetation buffer widths from 10-110 m. The model was calibrated and validated

for groundwater table dynamics near the river considering all hydrologic fluxes occurring in

the domain such as rainfall, irrigation, evapotranspiration by crops and the vegetation buffer,

as well as groundwater fluctuations and daily water level changes in the river. The major

findings includes:

• The hydraulic exchange at river interface for different irrigated crops was found to be

sensitive to the buffer widths.

• The likely average annual water flow from the almond and annual horticulture irrigated

area to the river was nearly twice as much (2.1 and 1.8, respectively) that under wine

grapes.

• For wine grapes, almonds and annual horticulture, the average annual hydraulic balance

reached an equilibrium at 20, 65 and 55 m buffer widths, respectively.

• The average annual load of salts became negligible for wine grapes with a 20 m buffer

width.

• This study shows that buffer widths of 20, 60, and 40 m for irrigated wine grapes,

almond, and annual horticulture, respectively, are needed to restrict the migration of

salts to the river.

The optimised widths in this study differs from the existing guidelines in Australia. It is

suggested that there is a strong need to revise the existing riparian width guidelines for

maintaining good water quality in surface water bodies near RCW irrigated crops. Further

refinements are possible by incorporating the influence of preferential flow paths, improved

water stress response functions, and addressing the data limitations for calibration of the model

for solute dynamics.


1. Introduction

Crop production, especially in arid and semi-arid regions of the world where rainfall is not able

to meet the evapotranspiration needs of the crops, depends on supplemental irrigation. Irrigated

agriculture contributes 40% of the world food production from 20% of the cropped area, thus

makes a major contribution to the global food security (FAO, 2016). However, the use of low

water quality (e.g. recycled water, RCW) for irrigation can induce numerous adverse impacts.

Irrigated agriculture with RCW water may become unsustainable due to its contribution to soil

degradation, salinization, waterlogging, and environmental pollution. Global water security

warrants beneficial reuse of recycled water, such as irrigation, but with minimal potential

harmful impacts on ecosystems. Ecosystem impairment, particularly reduced soil quality,

biodiversity loss, and harm to amenity and cultural heritage values, is a growing global problem

(FAO, 2011). Therefore, future irrigation schemes must address trade-offs, particularly with

respect to inter-sectoral water allocations and environmental impacts.

Aquatic ecosystems adjacent to irrigated agriculture are most at risk due to the transport of

irrigation induced chemicals such as soluble salts, nitrates, and pesticides (Zhang et al., 2010).

The fate of these chemicals in the soils and their migration to receiving environments depend

on a number of factors including the vegetation, topography, climate, soil, irrigation,

groundwater level, and flow conditions in the stream (Klatt et al., 2017; Schilling et al., 2018).

Riparian vegetation can moderate the movement of water and solutes to water bodies by

interception and attenuation of chemicals moving through the buffer zone (King et al., 2016).

The Australian Water Act 1989 designated the riparian zone as 20 m on either side of a

waterway and does so for the 'declared' delivery of stability, conservation, or functioning of

the waterway. However, the Assessment of River Condition (ARC) found that more than 80 %

of Australian rivers and riparian lengths are affected by human-generated catchment

disturbance (Norris, 2001). Rassam et al. (2006) reported that the vegetative buffers along

many perennial and ephemeral streams varied from 15 to 20 m. It is of some concern that the

Water Act 2007 no longer makes specific references to riparian zones. Therefore, some

Australian states devised their own buffer zone guidelines, depending on the priority areas and

water quality protection needs. For example, in Western Australia (Department of Water,

Government of Western Australia, 2006), the minimum buffer width guidelines for protection

of margins of water supply sources, drains, and banks of the first and second order ephemeral

streams and wetlands is 30 m. On the other hand, the New South Wales (NSW) guidelines


(Department of Primary Industries, NSW, 2012) suggest the buffer widths (one side of a

watercourse) depend on the type of watercourse, i.e., 1st order (10 m + half channel width), 2nd

order (20 m + half channel width), 3rd order (30 m + half channel width), and 4th order including

estuaries and wetlands (40 m + half channel width). Similarly, the guidelines in Tasmania

suggest a 30 m width of native vegetation as a riparian boundary for wetlands and waterways.

Riparian widths for Queensland irrigators for maintaining good water quality is 30 m

(Department of Environment and Resource management, 2011). In Victoria, the guidelines

depend on the land use intensity, and for the high intensity land use such as irrigated crops, the

buffer width is 60 m for maintaining water quality in surface water bodies. Essentially, the

Victorian guidelines are more comprehensive and clearly define the management objective for

riparian widths (Hansen et al., 2010). To our knowledge, South Australia does not have any

guidelines for maintaining an appropriate riparian width for preventing subsurface migration

of salts and contaminants from irrigated agriculture to surface water bodies’ especially seasonal

rivers.

Several investigations have examined the functions of buffer zones for stream ecosystems (e.g.,

Mayer et al., 2007; Hansen et al., 2010). However, these have mostly dealt with the overland

movement of solutes via surface runoff and sediment transport. Only, a limited number of

modelling or case studies have evaluated the role played by buffer zones in reducing the

migration of irrigation induced soluble salts/contaminants via subsurface flow to streams (e.g.,

Naiman et al., 2005; Allaire et al., 2015). Subsurface flow paths can exhibit wide variations

depending on specific local conditions (Naiman et al., 2005) including subsurface lithology

and stratigraphy (DeVito et al., 2000; Hill et al., 2004). To our knowledge, no information is

currently available in the literature on the role played by buffer zones in dealing with the

irrigation induced solute interception or influencing its migration to water bodies.

Field experiments for assessing the role of a buffer zone on the subsurface water and salts

movement from irrigated cropping system to an adjoining river is both a complex and

expensive exercise. Therefore, numerical models are increasingly being used (e.g., Flipo et al.,

2014; Xian et al., 2017) for such assessment. Hydraulic exchange across the stream-aquifer has

been modelled with buffer zones (e.g., Phogat et al., 2017a) or without (e.g., Baratelli et al.,

2016). Similarly, Kidmose et al. (2010) employed a conceptual groundwater flow and reactive

transfer model to establish a relationship between flow paths and the fate of a pesticide in a

riparian wetland. Alaghmand et al. (2013) used a numerical model (Hydrogeosphere) to


evaluate the interaction between a river and a saline floodplain in relation to groundwater

fluctuations, incorporating evapotranspiration losses by riparian vegetation. Klatt et al. (2017)

explored the capability of a coupled hydro-biogeochemical model to evaluate the effectiveness

of buffer strips to reduce nitrogen loads into aquatic systems. However, most of these

modelling studies have been either conceptual and/or only partially calibrated for site-specific

flow and/or solute dynamics. The complicated nature of water and solute transport processes

(Sophocleous, 2010) and the inherent uncertainty of input data are some of the challenges in

simulating water flow and solute transport with physically based models. Nevertheless, such

models are valuable in understanding water flow and solute transport/reaction processes

involved in complex bio-geological environments.

This study uses a two-dimensional finite element numerical model HYDRUS (2D/3D (referred

to below as HYDRUS; Šimůnek et al., 2016) to quantify the extent of water and solute

exchange across a stream-buffer interface. The study involves complex heterogeneous

geological formations involving real-time climatic, vegetative (crop and buffer), and stream

flow conditions. The key objectives of this investigation were: i) to calibrate and validate a

numerical model (HYDRUS) for water table dynamics in an area adjacent to a seasonal river

(Gawler River) by incorporating daily water level fluctuations in the river, groundwater

dynamics, crop evapotranspiration, riparian zone vegetation evapotranspiration, and soil

heterogeneities; ii) to estimate the impact of different buffer zone widths on the flux exchange

at the river-buffer interface under different cropping systems, iii) to optimise the riparian width

to control the irrigation-induced solute movement to the river for different irrigated crops; and

iv) to estimate the residence time of the solute tracer migrating to the adjoining water body

through the subsurface under shallow water table conditions.

2. Materials and Methods

2.1. Experimental site

The study was carried out at the Virginia Park (34°38ʹ22.6ʺS and 138°32ʹ27.6ʺE) gauging

station at Gawler River which is situated at 12 m above the Australian Height Datum (AHD).

The Gawler River only flows during the rainy season (July to October). However, stagnant

water (about 30-100 cm)/base flow conditions prevail at other times at the gauging station. The

adjacent area, being a part of the vast Northern Adelaide Plains (NAP), has a relatively flat


topography with a gentle slope to the west. All relevant features of the study site are shown in

Figure. 1. The NAP experiences a Mediterranean climate, which is characterised by hot, dry

summers and cool to cold winters. Long-term (1900-2016) average rainfall in the region

amounts to 475 mm (Department of Environment, Water and Natural Resources, 2016) and

annual evapotranspiration amounts to 1308 mm, resulting in the irrigation demand for crop

production. Water table fluctuations in the area adjacent to the river were monitored in the

shallow wells. Location of these wells is shown in Figure. 1.

Figure 1. Map of the study site showing the Gawler River, the gauging station, shallow wells

(yellow circles) and adjacent cropped area.

2.2. Soil characteristics

The soils of the NAP are highly heterogeneous with depth. There is commonly a shallow clay

layer at a variable depth, which determines the root growth and crops to be grown. Broader soil

groups and geology of the site were obtained from the stratigraphic information of the site and

well logs within the vicinity of the site. There are in general 6 major geological layers, which

include red friable sandy loam soil, light brown silty topsoil, sandy clay, sandy non-calcareous

clay, non-calcareous fine sandy clay, and sand. The soil particle size distributions and bulk

densities of these soil groups were obtained from the previous soil analysis reported in ASRIS

(ASRIS, 2011) and the APSIM (Holzworth et al., 2014) data base. The particle size and bulk

density data were used to estimate soil hydraulic parameters using the ROSETTA module

embedded in the HYDRUS software environment. The saturated hydraulic conductivity (Ks),

and the α and n parameters were further adjusted during the calibration process and their final

optimised values are presented in Table 1.

NGawler River

Virginia Park

Wells


Table 1. Optimised soil hydraulic parameters used in numerical simulations.

Sr No Textural class Depth

(m)

θr

(cm3cm-3)

θs

(cm3cm-3)

α

(cm-1)

n Ks

(cm d-1)

l

1 Sandy loam 0-1 0.07 0.44 0.024 1.45 56.1 0.5

2 Silty loam 1-2 0.06 0.45 0.03 1.5 69.8 0.5

3 Clay loam 2-5.5 0.07 0.43 0.0227 1.414 39.9 0.5

4 Silty clay loam 3.5-6 0.08 0.44 0.023 1.303 25.0 0.5

5 Sandy clay 6-8 0.09 0.44 0.0234 1.268 18.8 0.5

6 Sand 8-12 0.05 0.41 0.124 2.28 350.0 0.5

2.3. Water resources

There are three main water sources used for irrigation in the NAP: surface water (the Gawler

River), groundwater from tertiary aquifers, and recycled water from the Bolivar sewage

treatment plant. Water resources in the NAP Prescribed Wells Area (PWA) are managed

through a water allocation plan. The largest reserves of fresh groundwater in the NAP are

contained in the Tertiary aquifers, i.e., T1 (the average thickness 60-70 m) and T2 aquifers (the

average thickness 80-120 m), below the 60 m depth (Gerges, 1999). Groundwater extractions

primarily occur from the T2 aquifer, which provides 73% of the total groundwater volume,

while the T1 aquifer is less developed (25%). The Quaternary aquifers provide the remaining

2% (Department of Environment, Water and Natural Resources, 2017a, b). The shallow

Quaternary (Q1 to Q4) aquifers have only a limited water source and usually high salinity

(2000 mg/L to 15,000 mg/L), which restrict them from being used for agricultural

consumption. The groundwater flow direction in all Quaternary aquifers is generally from the

Mount Lofty Ranges (MLR) in the east toward the Gulf of St. Vincent in the west (Lamontagne

et al., 2015). Groundwater recharge to the tertiary aquifers is thought to occur by lateral inflow

from the adjacent fractured rock aquifers of the MLR as direct recharge from rainfall in the

NAP is low (Bresciani et al., 2017). The groundwater level observation network for the NAP

PWA consists of 153 observation wells monitored at quarterly intervals (Department for Water,

South Australia, 2010). Sixty-one wells are within Quaternary aquifers and 82 are within

Tertiary aquifers. The groundwater salinity observation network for the NAP PWA consists of

40 observation wells monitored on an approximately annual basis (Department for Water,

South Australia, 2010).


The Gawler River downstream of the junction of the North and South Para Rivers, is a

Prescribed Watercourse in the Western MLR Prescribed Water Resources Area. Extraction

limits and minimum and maximum threshold flows rates have been set through the Western

MLR WAP (Department of Environment, Water and Natural Resources, 2016). Specifically,

10 GL per annum is the extraction limit defined in the WAP, with water allowed to be extracted

at flows rates of between 500 to 690 L/s.

Water resources for irrigated agriculture in the NAP have been supplemented by 19.5GL from

the Bolivar tertiary-treated wastewater plant since 1999. This recycled water supplies 400

irrigators (The Goyder Institute of Water Research, 2016). An additional 12 GL per year of

recycled water will soon be available to increase the area under irrigation. The recycled water

is a Class-A treated sewage water but contains on an average 1200 mg/L soluble solutes

(Stevens et al., 2003). Managed aquifer recharge (MAR) schemes have been operational in the

NAP for some decades, with a number of new wells established (e.g., Bolivar). The total current

injection rate is estimated to be 8.5 GL yr-1 (Bresciani et al., 2017).

2.4. Vegetation and crops parameters

The vegetation buffer at the study site is dominated by river red gums (Eucalyptus spp.) but its

width is highly variable along the longitudinal distance of the river, ranging from a couple to

hundreds of metres. The area adjoining the riparian buffer is used for intensive cropping such

as almonds, wine grapes, potato, carrot, and onion all along the river. On the southern side of

the river where the modelling domain was established, the land has been used for the wine

grape cultivation.

Daily crop evapotranspiration (ETC) for river red gum (Eucalyptus spp.) in the buffer zone and

irrigated crops (wine grape, almond, carrot, potato) grown in the adjacent river corridor were

estimated from daily reference crop evapotranspiration (ET0) data and local crop coefficients

(Kc). The ET0 (Allen et al., 1998) data was obtained from the nearby weather station (Edinburg

Raaf). The daily ETC values were divided into evaporation (Es) and transpiration (Tp)

components based on the leaf area index (LAI) as follows (Ritchie, 1972): - x LAI . e grK

s cE ET= (1)

p c sT ET E= −


Here, Kgr is the light extinction coefficient for global solar radiation and its value was set to 0.5

(Aubin et al., 2000; Phogat et al., 2017a) for all vegetations. The LAI data for wine grapes and

almonds was obtained from other studies (Phogat et al., 2017b, 2018b) and for the annual

horticulture (carrot and potato) crops from the literature (Reid and English, 2000; Deshi et al.,

2015). Canopy interception by the buffer vegetation was assumed to be 15% of precipitation

(Xiao et al., 2000). Estimated daily Es and Tp values and daily rainfall were used as input in

HYDRUS simulations.

The roots of the buffer strip vegetation were assumed to be distributed linearly from the soil

surface to a depth of 200 cm. Although roots of Eucalyptus can grow to a depth of 6-7 m

(Phogat et al., 2017a), however, due to shallow water table conditions at the site, roots generally

did not grow far below a water table due to the lack of the oxygen supply (Baker et al., 2001).

Similarly, the rooting depths of 100, 200, and 60 cm for wine grape, almond, and annual

horticulture (carrot and potato), respectively, were used in the modelling study based on

relevant studies from the region (Phogat et al., 2017b, 2018b). The root water uptake

parameters for almond and wine grape were also taken from these studies and the HYDRUS

database (potato and carrot). Since HYDRUS does not allow using different parameters for

stress response functions for the crop and buffer zones, hence the same Feddes’ parameters

(Feddes et al., 1978) were used for both parts of the domain. It must be noted that root water

uptake in HYDRUS depends on the availability of water in the soil, the root spatial distribution,

and differential transpiration fluxes in the crop and buffer zones. The root water uptake was

assumed to be linearly distributed with depth, with the maximum at the soil surface and zero

at the bottom of the rooting zone.

2.5. Irrigation scheduling of crops

The “trigger irrigation” option available in HYDRUS (Dabach et al., 2013) was used to

generate irrigation schedules for all crops (wine grape, almond, annual horticulture). Irrigation

is triggered when the predefined suction level in the soil profile is reached at a specified

location. The timing of such an event depends on the daily climate conditions, plant water

requirements, soil texture, and water availability in the soil profile. The trigger pressure used

for wine grapes, almond, and carrot-potato were -60, -25, and -15 kPa, respectively. Similar

trigger pressures have been used for these crops in other studies (Green, 2010; Phogat et al.,

2018a, b). Similar irrigation scheduling using tensiometers is widely used for different crops.


The trigger point in this study was located at a depth of 30 cm in the middle of the irrigated

crops area. Triggered irrigations had a solute concentration of 1200 mg/L (Stevens et al., 2003),

which is equal to the average quantity of total dissolved solids in the Class-A treated recycled

water from the Bolivar treatment plant, which is used for irrigation in the NAP region.

2.6. Transport domain, initial and boundary conditions

The transport domain represents a 400 m cross section from the middle of the river (Figure. 2).

The vertical dimension represents the distance from the Australian Height Datum (AHD) to

the soil surface (12 m) at the experimental site. The top width of the river was 10 m, the bottom

width 4 m, and the depth 4 m at the study site. The width of the buffer zone is 30 m from the

river bank. Therefore, the lateral width of the riparian zone at the Virginia Park gauging station

from the middle of the river is approximately 35 m, which also includes an unsealed road which

runs along the river. The finite element discretization resulted in 10,000 2D elements in a

standard rectangular 2D domain.

On the upper left side of the domain (Figure. 2), the atmospheric boundary was considered

through which the infiltrative influx or the evapotranspirative efflux occurs. A time-variable

flux boundary condition (treated similarly as an atmospheric boundary condition) was imposed

on the upper right side of the domain to represent the buffer zone, which had different fluxes

than the irrigated surface. The flux at this boundary was given by the difference between daily

rainfall and daily potential evaporation (Es). A special HYDRUS boundary condition (BC) was

specified in the river. This special BC assigns the hydrostatic pressure head on the boundary

below the water level in the river and a seepage face BC on the boundary above the water level.

The specified water levels in the river are linearly interpolated in time in order to smooth the

impact of daily fluctuations of water levels in the river (Phogat et al., 2017a). Measured values

of water table depths in the well near the left boundary of the domain (PTA100) were used to

define initial and time-variable pressure head boundary conditions. No flow was assumed as

the boundary. The initial pressure head condition in the domain was specified by interpolating

measured mean water table depths in the shallow wells (Figure. 1) in the adjacent area while

considering hydrostatic equilibrium conditions in the vertical direction. Daily rainfall in excess

of the soil infiltration capacity is accounted for as run off by HYDRUS. The longitudinal

dispersivity was assumed as one tenth of the modeling domain (with the transverse dispersivity


being one tenth of the longitudinal dispersivity) (Cote et al. 2003; Phogat et al., 2014) and the

molecular diffusion coefficient in water equal to 1.66 cm2/day (Phogat et al., 2018b).

Figure 2. Schematic representation of the flow domain showing the material distribution, the

river, the buffer strip, irrigated crops, and imposed boundary conditions.

2.7. Calibration and validation of the model

Measured water table depths (average of four quarterly measurements in a year) in the shallow

wells (PTA101, PTG080 and PTG087, Figure. 1) near the study site were used for the

calibration and validation of the model. Simulations were carried out for 1461 days (1st July

2009 to 30th June 2013) to calibrate the model for water table depths at the middle of the domain

(X = 200 m). For most sensitive model parameters including the saturated hydraulic

conductivity (Ks), and the coefficients α and n of different soil layers were varied manually and

no automated parameter optimisation procedure was used to calibrate the model. In addition to

a visual comparison of observed and simulated water table depths, a quantitative evaluation of

the model performance was undertaken using goodness-of-fit measures (see section 2.8)

similar to other studies (e.g., Alaghmand et al., 2013, 2014). The calibrated model was

validated for 1461 days (1st July 2013 to 30th June 2017) by comparing the measured and

simulated water table depths. The calibrated and validated model was then used to assess the

impact of other irrigated crops (almond and annual horticulture crops such as carrot and potato)

and the buffer zone widths on the migration of water and solutes to the river. More details on

different scenarios are given below in the Scenario Analysis section.

To understand the movement of irrigation-induced solutes/ agrochemical tracers to the river

water, we considered Total Dissolved Solids (TDS) as representative of all soluble solutes

which is consistent with numerous studies (e.g., Ramos et al., 2011; Phogat et al., 2014, 2018b).

The initial soil conditions in the domain were assumed to be solute free. The average quantity


of TDS (1200 mg/L) of Class-A treated (recycled) water from the Bolivar treatment plant

(Stevens et al., 2003) was applied during all triggered irrigations at the atmospheric boundary

where crop is being grown. However, the model calibration for solute dynamics could not be

conducted due to the non-availability of site-specific data for solute transport processes.

2.8. Scenario analysis

The calibrated and validated model was then used to simulate the dynamics of the hydrological

fluxes and solute movement for different buffer widths and for various irrigated crops (wine

grape, almond, and carrot-potato rotation). The simulations were executed for 8 years (1st July

2009 to 30th June 2017) plus further 8 years if needed (if solute did not reach the river) for all

3 irrigated crops for varying buffer zone widths (10 - 110 m) from the centre of the river. These

simulations were established to evaluate the appropriate width of the riparian zone to control

the lateral movement of solutes to the river. In the simulations extended in future climate (8

years), median climate change data for the Edinburg RAAF station were used (Charles et al.,

2015). Initial conditions and daily water level fluctuations for such future simulations were

imported from the previous simulations (8 years). To estimate the extent of leaching, the

annual water balance for different crops was computed using inputs such as rainfall, irrigation,

and model-simulated evaporation and transpiration and assuming similar initial pumping well

and river exchange conditions in the domain.

2.9. Model evaluation

The model’s performance was evaluated by comparing measured (M) and HYDRUS simulated

(S) water table depths. Correlation coefficients were estimated to understand the relationship

between measured and simulated values of water table depths during the calibration and

validation periods. The statistical error estimates [mean error (ME), mean absolute error

(MAE), and root mean square error (RMSE)] between the measured and simulated water table

depths were estimated as:

( )∑=

−=N

iii SM

NME

1

1 (2)

∑=

−=N

iii SM

NMAE

1

1 (3)


( )∑=

−=N

iii SM

NRMSE

1

21 (4)

Many studies (e.g., Coffey et al., 2004; Alaghmand et al., 2013) have used similar goodness of

fit measures (correlation coefficients and RMSE) as above.

We also evaluated the test of significance between the measured and simulated values of water

table depths using the paired t-test (tcal) as below:

𝑡𝑡𝑐𝑐𝑐𝑐𝑐𝑐 = 𝑑𝑑

𝑆𝑆𝑆𝑆𝑚𝑚�1𝑛𝑛

𝑑𝑑 = 𝑀𝑀� − 𝑆𝑆̅ and 𝑆𝑆𝑆𝑆𝑚𝑚 = �𝑛𝑛𝑠𝑠12+𝑛𝑛𝑠𝑠22

2𝑛𝑛−2 (5)

Here, n is the number of comparable paired points, s1 and s2 are the standard deviations of

measured and simulated data, respectively, d is the different between measured (𝑀𝑀�) and

simulated (𝑆𝑆̅) means values, SDm is the standard deviation of the mean, and tcal is the calculated

paired t-test value. The null hypothesis tests that there is no significant difference between the

mean values of measured and simulated water table depths.

Nash and Sutcliffe (1970) model efficiency (E) is a normalized statistic that expresses the

relative magnitude of the residual variance compared to the variance of the measured data

during the period under investigation, as given below:

( )

( )∑

∑

=

=

−

−−= N

ii

N

iii

MM

SME

1

2

1

2

1 (6)

The range of E lies between − ∞ and 1.0 (a perfect fit). An efficiency value between 0 and 1

is generally viewed as an acceptable level of performance. Efficiency <0 indicates that the

mean value of the observed time series would be a better predictor than the model and denotes

unacceptable performance (Moriasi et al., 2007; Legates and McCabe, 1999).

3. Results and Discussion

3.1. Calibration and validation of HYDRUS-2D

The measured buffer zone width (35 m from the middle of the Gawler River) and actual crop

grown (wine grape) adjoining the study site were considered for the calibration and validation

simulations executed from 1st July, 2009 to 30th June, 2013 and 1st July, 2013 to 30th June,

2017, respectively. The data in Figure. 3 demonstrates a consistent performance of the model

(i.e. R2 of 0.66 and 0.64, and E = 0.34 and 0.34, respectively) during calibration (2009-2013)


and validation (2013-2017) period. These values fell within the R2 values (0.35-0.84) reported

in other modelling studies (e.g., Coffey et al., 2004; Phogat et al., 2016). Other statistical

estimates (ME, MAE, RMSE and SD; see Figure. 3) during the calibration and validation period

were similar but slightly higher than previously observed values (e.g. Alaghmand et al., 2013).

This is because of wide fluctuation within the input data. The paired t-test indicated that there

was no significant difference (p= 0.05) between the measured and simulated mean depths of

water table. Overall, all these statistics confirm an adequate representation of groundwater

fluctuations by the model.

Figure 3. Relationship between measured and simulated water table depths, statistical error

estimates (ME, MAE and RMSE), standard deviation (SD) and model efficiency (E) values

during the calibration (2009-13) and validation (2013-17) periods.

Calibration of the model for solute dynamics was not possible due to the lack of site specific

data. However, numerous studies have shown that sufficiently calibrated and validated model

for complex hydraulic fluxes in a heterogeneous domain can offer valuable practical

understanding of bio-geochemical processes in the soil. For example, Alaghmand et al. (2013,

2014) indicated that the complexity associated with the quantification of the solute transport

parameters restricted them from validating the model on the observed concentration pattern.

However, their study unravelled the impacts of groundwater pumping on salinization risks of

a flood plain. Similarly, Carr et al. (2018) calibrated and validated a 2D model on gauged

groundwater elevations and hypothesized that the accurate representation of flow dynamics can

inform environmental management involving transport of sediments, nutrients, and heavy

metals. Other studies also have used similar approaches (e.g., Rousseau et al., 2012).

y = 0.58x + 105.17R² = 0.66

0

50

100

150

200

250

300

350

200 250 300 350 400

Mea

sure

d w

ater

tabl

e de

pths

(cm

)

Simulated water table depths (cm)

y = 0.40x + 148.99R² = 0.64

0

50

100

150

200

250

300

350

200 250 300 350 400

Simulated water table depths (cm)

ME =18cmMAE =21cmRMSE = 26cmSD = 34.1 cmE = 0.34

ME =16cmMAE =20cmRMSE = 28cmSD = 24.2 cmE = 0.34

Calibration Validation


3.2. Irrigation and annual water balance for crops

The extent of average annual irrigation among all crops during 2009-2017 (Figure. 4) was the

lowest in wine grape (242 and 320 mm), followed by almond (760 and 920 mm) and the highest

for annual horticulture (951 and 1226 mm) reflecting their specific evapotranspiration

requirements (Phogat et al., 2018b). Correspondingly, the leaching fraction/recharge flux

under almond (87-298 mm) and annual horticulture (100-252 mm) was 3- 3.8 times higher than

under wine grapes. Besides for wine grapes, a negative annual flux balance was recorded in

some years but, the overall average balance was positive over 8 years. These observations are

consistent with other studies (Green, 2010; Reynolds, 2010; Phogat et al., 2018b). It is well

understood that the contribution of leaching fraction/irrigation return flow from irrigated crops

can be a critical driver for the river-buffer hydraulic exchange (e.g., Berens et al., 2009).

Figure 4. Annual irrigation (mm) and recharge/discharge (mm) in the domain under a) wine

grape, b) almond, and 3) annual horticulture (carrot-potato) crops. Positive fluxes are recharge

and negative fluxes are discharge from the domain.

3.3. Hydraulic exchange at the river-buffer interface

Hydraulic exchange at this boundary depends on river flow dynamics, rainfall, water uptake

by the buffer zone vegetation and irrigated crops, soil evaporation, and shallow groundwater

fluctuations near the river. The Gawler River is seasonal and flow depends on the extent of

rainfall in the catchment during the winter. Numerous high rainfall/flood events during 2009-

2017 resulted in large amounts of recharge as indicated by large peaks in Figure. 5. During

summer, a large amount of irrigation is applied to crops in the adjoining area, which reverses

the flow gradient across the buffer-river boundary. Under such dynamic conditions, aquifer

-100-50

050

100150200250300350

Irri

gatio

n, R

echa

rge

(mm

/yr)

0

200

400

600

800

1000

1200

1400

Recharge/DischargeIrrigation

0

200

400

600

800

1000

a) wine grape b) almond c) annual horticulture


recharge or discharge may occur depending on the river level (Ghazavi et al., 2012), leading to

either a gaining or losing river (Rassam, 2011; Phogat et al., 2017a). However, irrigation-

induced flow to the river depends on the water uptake/evapotranspiration pattern of buffer zone

vegetation (Phogat et al., 2017a) and on the extent of return flow from the irrigated areas. As

shown earlier, the extent of irrigation-induced flow (recharge) is higher for almond and annual

horticulture as compared to wine grapes (Figure. 4). As a result, the average annual amount of

flux exchange at the river interface was nearly twice (1.8- 2.1 times) as much under almond

and annual horticulture than under wine grapes (Figure. 5). Additionally, the high evaporation

demand of the buffer zone vegetation may further limit the net discharges to the river. This was

probably not met by the low drainage flux under wine grapes. Overall, irrigation induced

drainage, groundwater discharge to the stream, evapotranspiration by the buffer zone

vegetation and irrigated crops play a key role in defining the extent of exchange between the

buffer zone and the river.


Figure 5. Daily flux exchange at the river-buffer interface for almond, wine grape, and annual

horticulture (carrot-potato) grown near the Gawler River during 2009-2017.

The impact of different buffer widths on the average annual water exchange at the river

interface for different irrigated crops is shown in Figure. 6. For wine grapes, the average annual

hydraulic balance was negative for the 10-20 m buffers during the simulation period (8 years,

2009-2017), indicating the dominance of flow from the irrigated area to the river system.

However, the reverse was observed for buffer widths > 20 m as the evapotranspiration demand

of the buffer vegetation governed the water exchange at the river-buffer interface. In the case

of almond, however, the overall water balance remained negative (discharge to the river) for a

buffer zone up to 65 m due to its 3 times higher irrigation than for wine grapes. Similarly,

under annual horticulture (carrot and potato) crops, the overall hydraulic balance was similar

a) Wine grapeDa

ily fl

ux (c

m2 /d

ay)

-2000

-1000

0

1000

10000

20000

30000Da

ily fl

ux (c

m2 /d

ay)

-2000

-1000

0

1000

10000

20000

30000

Jan-2010 Jan-2011 Jan-2012 Jan-2013 Jan-2014 Jan-2015 Jan-2016 Jan-2017

Daily

flux

(cm

2 /day

)

-2000

-1000

0

1000

10000

20000

30000

b) Almond

c) Annual horticulture


to almonds and the threshold buffer zone width for equilibrium flow conditions reached at 55

m (Figure. 6). Based on irrigation regime for irrigated crops, different buffer zone widths are

required for equilibrium flow conditions at the river-buffer interface.

Figure 6. An average balance of water exchange across the stream-aquifer interface for

different buffer widths under a) wine grapes, b) almond, and c) annual horticultural crops.

3.4. Impact of a buffer width on solute dynamics

It is well known that irrigation introduces large amount of solutes into the soil as the recycled

water contains appreciable amounts of soluble salts (approximately 1200 mg/L; Laurenson et

al., 2010). Temporal dynamics of irrigation induced salts in the soil profiles with 30 m buffer

widths for wine grapes, almonds, and annual horticulture is shown in Figure. 7. Irrigation

induced salts gradually moved downwards as well as laterally, but, the vertical movement was

faster. Initially, salts continued to build up in the soil and then migrated to the shallow

groundwater (water table at 4 m). It was noted that the irrigation induced salts entered the

shallow groundwater within 2 years of irrigation of almond and annual horticulture (Figure. 7),

but it took longer for wine grapes. It is noteworthy that after 8 years of simulation, the salts

were distributed in variable concentrations throughout the entire domain, including along the

river boundary, especially for almond and annual horticulture. The presence of salts along the

river boundary indicates that they may have already entered into the river system in low

concentrations (< 0.0011 mg/cm3) against the concentration gradient under these crops. In

contrast, for wine grapes, salts travelled a far shorter distance than for the other two crops and

-400000

-300000

-200000

-100000

0

100000

200000

0 10 20 30 40 50 60 70 80 90 100

Ann

ual f

lux

(cm

2 )

Buffer width (m)

wine grapealmondAnnual horticulture


the salt plume remained far from the river front boundary (Figure. 7). Nonetheless, the extent

and timing of salts migration into the river system was not clearly visible.

Excessive concentrations of soluble salts as shown on the right hand corner of Figure. 7 for

almond and annual horticulture can have an adverse impact on plant growth and transpiration

losses. However, this situation was obtained for a 30 m buffer width (Figure. 8), which is an

insufficient width to control salts mobilization to the river for these crops. This shows that in

the absence of an adequate buffer width, there is a chance of rapid secondary salinization close

to the river. In such situations, adopting native vegetation such as river red gums (with a salt

tolerance of 30 dS m−1) or black box trees (salt tolerance of 55 dS m−1) as buffer (Overton and

Jolly, 2004), could help maintain appropriate transpiration services in such situations

(Alaghmand et al., 2013, 2014).

Figure 7. Temporal dynamics of irrigation induced salts in the soil profile with a 30-m buffer

width under a) wine grapes, b) almonds, and c) annual horticulture (carrot-potato) at the

indicated times during 2009-2017. The river is situated at the top right corner of the domain.

The frequency, timing, and load of salt pulses entering into the river during 8 years (2009-

2017) of simulation under different crops as influenced by buffer zone widths is shown in

Figure. 8. It is apparent that for wine grapes, the salts/tracer pulse appeared in the river only

for buffer widths < 20 m. Further widening the buffer (30 - 60 m) and extending the simulations

to 16 years (2009-2025; data not shown) did not produce any appreciable amount of salts in

the river. In contrast, a significant salts/tracer pulse continued to appear in the river for 50 and

60 m buffer for annual horticulture and almond irrigated crops, respectively, during 8 years

(2009-2017) of simulation. Further extending the buffer (up to 100 and 110 m) and simulation

c) Annual horticultureb) Almondsa) Wine grapes

Ele

vatio

n fr

om A

ustr

alia

n H

eigh

t Dat

um (m

)

Distance from the centre of River (m)

1st July 2009

30th June 2011

30th June 2013

30th June 2015

30th June 2017


time (2009-2025) resulted in a very small amount of tracer salt to appear in the river (10-22 to

10-6 mg) for annual horticulture and almond. Therefore, it is concluded that a buffer width of

20, 50 and 60 m for wine grapes, annual horticulture (carrot-potato) and almonds, respectively,

is needed to restrict the subsurface migration of irrigation induced salts/tracer chemicals to the

river.

In terms of salt load the same buffer width for different crops released varying pulses in the

river (Figure. 8). For example, for a buffer width of 10 m for almond during the 8 years of

simulation was approximately twice the salts load than that for annual horticulture, which in

turn had 10 times higher loads than wine grapes. However, with increasing buffer width (e.g.

20 m), salts loads were drastically decreased (especially in almonds and annual horticulture),

but, the timing of occurrences of salts/tracer pulses in the river are similar irrespective of

irrigated crops. Notably, a large solute peak occurred in the river water after approximately 7

years of simulations (August to December, 2016), which transported different amounts of salt

to the river at different buffer widths. This time corresponds to the aftermath of a flood event

when the water level in the Gawler River reached the soil surface and completely saturated the

adjacent riparian zone. Subsequent receding water levels in the river created a steep gradient

from the buffer zone to the river, which conveyed a large amount of salts from the saturated

zone to the stream. Such observations were also reported in other studies (Bryan et al., 1998;

McKergow et al., 2003). Hence, the salts transported to the river may not be associated only

with irrigation, but also with the generation of hydraulic gradients, which push the salts

laterally into the river. However, the impact of hydraulic gradients gradually dissipates as the

buffer width increases, since the buffer zone acts as a barrier in transmitting the hydraulic

response between the river and the irrigated area. Overall, it appears that all components of

soil, water, crop and climate, play a crucial role and have a different influence on the salt

transport to the river and its water quality.


Figure 8. The effect of buffer zone widths (10, 20, 30, 40, 50 and 60 m) on the lateral migration

of irrigation induced salts into the river under almond (a to f), annual horticulture (g to k), and

wine grapes (l to m) during 2009-2017. The solute flux scale at the vertical axes is different in

different figures.

Salts load transported to the river and the residence time of solutes in the soil for different

buffer widths and crops are shown in Figure. 9. The amount of salts for the 10 m buffer was

very similar for almond and annual horticulture and about 40 times higher than for wine grapes.

Meanwhile, the salts transported to the river for the 20 m buffer were higher for annual

horticulture than for almond. When the buffer width was increased to 60 m, only a small

-1000-800-600-400-200

01/

7/09

1/7/

10

1/7/

11

1/7/

12

1/7/

13

1/7/

14

1/7/

15

1/7/

16

-40

-30

-20

-10

0

-5-4-3-2-10

-0.0006

-0.0004

-0.0002

0

-0.6

-0.4

-0.2

0

-0.03

-0.02

-0.01

0

b)20m

a)10m

c)30m

d)40m

e)50m

f)60m

Solu

te F

lux

(mg/

day)

-500-400-300-200-100

0

1/7/

09

1/7/

10

1/7/

11

1/7/

12

1/7/

13

1/7/

14

1/7/

15

1/7/

16

-75-60-45-30-15

0

-0.001-0.0008-0.0006-0.0004-0.0002

0

-1.2

-0.9

-0.6

-0.3

0

-0.05-0.04-0.03-0.02-0.01

0

g)10m

h)20m

i)30m

k)50m

j)40m

-36

-27

-18

-9

0

1/7/

09

1/7/

10

1/7/

11

1/7/

12

1/7/

13

1/7/

14

1/7/

15

1/7/

16

l)10m

-0.032

-0.024

-0.016

-0.008

0

m)20m

Almond Annual horticulture Wine grapes


additional reduction in the salt load was observed (99.9%). Similarly, for annual horticulture,

the average annual load of salts was reduced by 92.2% (to 1566 mg) for the 20 m buffer width

as compared to the 10 m buffer width. For a 99.9% reduction in the salt load, a 40 m buffer

width is needed. Therefore, it is established that maintaining a 20, 60, and 40 m buffer widths

for wine grapes, almonds, and annual horticulture can effectively reduce irrigation induced

salts/tracers transport to the river by 99.9%.

Figure 9. The amount of salts (line graph) and their residence times (bars) for different irrigated

crops as a function of a buffer width during 2009-2017.

The residence time was twice as long for wine grapes as compared to almonds and annual

horticulture for buffers <40 m wide. For example, the residence time for a 10 m buffer was 2.5

months for almonds and annual horticulture, and 5 months for wine grapes (Figure. 9). The

residence time for the solute movement to the stream starts increasing exponentially for larger

buffer widths (>40 m) and wine grapes. It took about 7 years for irrigation induced solutes,

though concentrations were very small, to appear in the stream when the buffer width was 60

m. The residence time for almonds and annual horticulture was very small (<100 days) for

buffer widths <30 m, then gradually increased to 450 days for a 70 m, and to much longer

values for larger buffer widths. A longer residence time for wine grapes is associated with a

relatively low recharge volume, which was unable to generate a sufficient hydraulic gradient

to push the salts into the river. Therefore, the extent of irrigation return flow is crucial for the

longevity of solutes in the soil system. The knowledge of the time required for the migration

of salts to the river helps in devising guidelines for maintaining the river water quality.

-25000

-20000

-15000

-10000

-5000

0

0

500

1000

1500

2000

2500

3000

10 20 30 40 50 60 70 80 90 100

Ann

ual s

alt l

oad

(mg)

Res

iden

ce ti

me

(day

s)

Buffer width (m)

Grape AlmondAnn Hort GrapeAlmond Ann Hort


4. Implications of the findings

This study demonstrated that for wine grapes, a 20 m buffer width could reduce the load of

irrigation salts discharged to the Gawler River down to a negligible level (a 99.9% reduction),

whereas for annual horticulture, for a similar salt load reduction, a buffer width of 40 m was

needed. On the other hand, for almonds that have much higher irrigation requirement, a buffer

zone of 60 m would be needed for a similar reduction. Therefore, the buffer width needs to

consider the nature of the production system adjacent to the river system being protected. The

decision about an appropriate width of the buffer requires a careful consideration of economic

consequences and not just ecological requirements.

This study highlights that considerations of local hydraulic, climate, and soil conditions, as

well as local geological heterogeneity, can have a marked impact on the requirement of

adequate vegetative buffer along the rivers. The buffer zone guidelines adopted in most of the

states (NSW = 40 m, Western Australia = 30 m, Tasmania = 30 m, and Victoria = 60 m) in

Australia for maintaining the river water quality are mainly based on overseas studies (Hansen

et al., 2010). Therefore, the consideration of the type of irrigated crops grown along the rivers

and surface water bodies could have varied impact on the maintenance of the buffer zone and

hence the riparian zone guidelines.

5. Conclusions

This study was carried out to understand the extent, frequency and nature of hydraulic and

solute dynamics relationship between river and irrigated crops interspersed with a buffer. The

calibrated and validated model (HYDRUS-2D) was used to evaluate long-term (8 years)

scenarios involving different irrigated crops (wine grapes, almond, and annual horticulture)

and varied buffer widths (10 to 110 cm) aimed at quantifying the hydraulic connection between

the river and the crop grown and optimising buffer width for controlling solute movement to

the river.

Statistical evaluation (correlation coefficient, ME, MAE, RMSE, t-test and model efficiency) of

the model for water movement showed a fairly good matching between the measured and the

simulated water table depths. The results obtained from 8-years of simulations showed that

three irrigated crops (wine grapes, almond, and annual horticulture) requiring different annual


irrigation applications have markedly different influence on the overall water and solute

movement within the crop-buffer-river ecosystem. Consequently, the average annual flow from

the soil to the river was 2.1 and 1.8 times higher under almond and annual horticulture,

respectively, compared to wine grapes. For average annual water balance to reach equilibrium,

buffer (Eucalyptus spp.) widths of 20, 65, and 55 m were needed for wine grapes, almond, and

annual horticultural crops (carrot-potato), respectively.

From the standpoint of migration of irrigation induced soluble salts into the river, the buffer

widths of 20, 60, and 40 m were able to reduce the salt load by >99% under wine grapes,

almond, and annual horticulture crops, respectively. Notably, existing guidelines do not

differentiate between crop types when specifying optimum buffer widths. Further work is

required to assess how the seasonal variability in irrigation water quality influences these

results, particularly in areas where recycled water is used for irrigation. The modelling

challenges and data limitations included the absence of preferential flow, the assumption of

similar water stress response functions for the buffer vegetation and irrigated crops, and lack

of data for calibration of the model against solute dynamics in the river. Further refinements in

the above findings may be achieved by addressing these gaps.

References

Alaghmand, S., Beecham, S. and Hassanli, A., 2013. Impacts of groundwater extraction on

salinization risk in a semi-arid floodplain. Nat. Hazards Earth Syst. Sci. 13, 3405–3418.

Alaghmand, S., Beecham, S. Hassanli, A., 2014. Impacts of vegetation cover on surface-

groundwater flows and solute interactions in a semi-arid saline floodplain: A case study

of the Lower Murray River, Australia. Environ. Process. 1, 59-71.

Allaire, S.E., Sylvain, C., Lange, S.F., Thériault, G. Lafrance, P., 2015. Potential efficiency of

riparian vegetated buffer strips in intercepting soluble compounds in the presence of

subsurface preferential flows. PLoS ONE 10(7), 1-21, e0131840.

https://doi.org/10.1371/journal.pone.0131840.

Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration guidelines for


Rome, Italy.

https://doi.org/10.1371/journal.pone.0131840


ASRIS, 2011. Australian Soil Resource Information System. http://www.asris.com.au.

Accessed on April 10, 2018.

Aubin, I., Beaudet, M., Messier, C., 2000. Light extinction coefficients specific to the

understory vegetation of the southern boreal forest, Quebec. Can. J. For. Res. 30, 168–

177.

Baker, T.T., Conner, W.H., Lockaby, B.G., Stanturf, J.A., Burke, M.K., 2001. Fine root

productivity and dynamics on a forested floodplain in South Carolina. Soil Sci. Soc. Am.

J. 65(2), 545-556.

Baratelli, F., Flipo, N., Moatar, F., 2016. Estimation of stream-aquifer exchanges at regional

scale using a distributed model: Sensitivity to in-stream water level fluctuations, riverbed

elevation and roughness. J. Hydrol. 542, 686-703.

Berens, V., White, M.G., Souter, N.J., 2009. Injection of fresh river water into a saline

floodplain aquifer in an attempt to improve the condition of river red gum (Eucalyptus

camaldulensis Dehnh.). Hydrol. Process. 23, 3464–3473.

Bresciani, E., Cranswick, R.H., Banks, E.W., Batlle-Aguilar, J., Cook, P.G., Batelaan, O.,

2017.Using hydraulic head, chloride and electrical conductivity data to distinguish

between mountain-front and mountain-block recharge to basin aquifers. Hydrol. Earth

Syst. Sci. Discuss. https://doi.org/10.5194/hess-2017-328.

Bryan, R.B., Hawke, R.M., Rockwell, D.L., 1998. The influence of subsurface moisture on hill

system evolution. Earth Surf. Process. Landf. 23, 773–789.

Carr, K.J., Tongbi, T., Ercan, A., Levent Kavvas, M., 2018. Evaluating the applicability of a

two-dimensional flow model of a highly heterogeneous domain to flow and

environmental management. J. Am. Water Resour. Assoc., 24, 184-197.

Charles, S.P., Fu, G., 2015. Statistically Downscaled Projections for South Australia. Goyder

Institute for Water Research, Technical Report No. 15/1, Adelaide, South Australia.

http://www.goyderinstitute.org/_r91/media/system/attrib/file/82/CC%20Task%203%20

CSIRO%20Final%20Report_web.pdf.

Coffey, M.E., Workman, S.R., Taraba, J.L., Fogle, A.W., 2004. Statistical procedures for

evaluating daily and monthly hydrologic model predictions. Trans. ASAE 47, 59-68.

Cote, C.M., Bristow, K.L., Charlesworth, P.B., Cook, F.J., Thorburn, P.J., 2003. Analysis of

soil wetting and solute transport in subsurface trickle irrigation. Irrig. Sci. 22, 143–156.

Dabach, S., Lazarovitch, N., Šimůnek, J., Shani, U. 2013. Numerical investigation of irrigation

scheduling based on soil water status. Irrigation Science, 31(1), 27-36.

http://www.asris.com.au/

https://doi.org/10.5194/hess-2017-328

http://www.goyderinstitute.org/_r91/media/system/attrib/file/82/CC%20Task%203%20CSIRO%20Final%20Report_web.pdf

http://www.goyderinstitute.org/_r91/media/system/attrib/file/82/CC%20Task%203%20CSIRO%20Final%20Report_web.pdf


Department of Environment and Resource Management, 2011. Queensland Wetland Buffer

Planning Guideline, Queensland Wetlands Program, Brisbane, Queensland.

Department of Environment and Water, 2018. Real-time ground water data. Department of

Environment and Water, Government of South Australia.

https://www.waterconnect.sa.gov.au/Systems/RTWD/Pages/Default.aspx, assessed

during 2017-18.

Department of Environment, Water and Natural Resources, 2016. Non-prescribed Surface

Water Resources Assessment – Adelaide and Mount Lofty Ranges Natural Resources

Management Region, DEWNR Technical report 2016/34, Department of Environment,

Water and Natural Resources, Government of South Australia, Adelaide.

Department of Environment, Water and Natural Resources, 2017a. Northern Adelaide Plains

Prescribed Water Area T1 Aquifer, Groundwater Level and Salinity Status Report 2016,

Department of Environment, Water and Natural Resources, Government of South

Australia, Adelaide.

Department of Environment, Water and Natural Resources, 2017b. Northern Adelaide Plains

Prescribed Water Area T2 aquifer, Groundwater Level and Salinity Status Report 2016,

Department of Environment, Water and Natural Resources, Government of South

Australia, Adelaide.

Department of Primary Industries, NSW, 2012. Guidelines for Riparian Corridors on

Waterfront Lands, Controlled Activities on Waterfront Lands, Government of New

South Wales, Australia, https://www.water.nsw.gov.au/data/assets/pdf

file/0004/547222/licensing approvals controlled activities riparian corridors.pdf.

Accessed on 24th April 2018.

Department of Water, 2010. Northern Adelaide Plain Prescribed Water Area (NAP PWA)

Groundwater Level and Salinity Status Report 2009-10. Department of Water,

Government of South Australia.

Department of Water, Government of Western Australia, 2006. Vegetation Buffers to Sensitive

Water Resources, Water Quality Protection Notes, WQPN 6, Perth, Western Australia.

Deshi, K. E., Obasi, M.O., Odiaka, N.I., Kalu, B.A., Ifenkwe, O.P., 2015. Leaf area index

values of potato (Solanum tuberosum L.) stored for different periods in different kinds of

stores. IOSR J. Agric. Vet. Sci. 8, 9-19.

DeVito, K.J., Fitzgerald, D., Hill, A.R., Aravena, R., 2000. Nitrate dynamics in relation to

lithology and hydrologic flow path in a river riparian zone. J. Environ. Qual. 29, 1075–

1084.

https://www.waterconnect.sa.gov.au/Systems/RTWD/Pages/Default.aspx

https://www.water.nsw.gov.au/data/assets/pdf%20file/0004/547222/licensing%20approvals%20controlled%20activities%20riparian%20corridors.pdf

https://www.water.nsw.gov.au/data/assets/pdf%20file/0004/547222/licensing%20approvals%20controlled%20activities%20riparian%20corridors.pdf


FAO, 2011. The state of the world’s land and water resources for food and agriculture

(SOLAW) - Managing systems at risk. Food and Agriculture Organization of the United

Nations, Rome and Earthscan, London.

http://www.fao.org/docrep/017/i1688e/i1688e.pdf.

FAO, 2016. AQUASTAT website. Food and Agriculture Organisation of the United Nations

(FAO). http://www.fao.org/nr/water/aquastat/didyouknow/index3.stm. Accessed on 24th

April 2018.

Feddes, R.A., Kowalik, P.J., Zaradny, H., 1978. Simulation of Field Water Use and Crop Yield.

Simulation Monographs, Pudoc, Wageningen, The Netherlands.

Flipo, N., Mouhri, A., Labarthe, B., Biancamaria, S., Rivière, A., Weill, P., 2014. Continental

hydrosystem modelling: the concept of nested stream–aquifer interfaces Hydrol. Earth

Syst. Sci. 18, 3121–3149.

Gerges, N.Z., 1999. The Geology and Hydrogeology of the Adelaide Metropolitan Area

Volume 1 & Volume 2. PhD thesis, Flinders University.

Ghazavi, R., Vali, A.B., Eslamian, S., 2012. Impact of flood spreading on groundwater level

variation and groundwater quality in an arid environment. Water Resour. Manage. 26(6),

1651-1663.

Green, G.P., 2010. Point and regional scale modelling of vadose zone water and salt fluxes in

an area of intensive horticulture. Thesis submitted to Flinders University, South

Australia. https://flex.flinders.edu.au/file/27320d77-7088-4d13-8bc5-

9c2b1fc4a071/1/Thesis-Green-2010.pdf.

Hansen, B., Reich, P., Sam Lake, P., Cavagnaro, T., 2010. Minimum Width Requirements for

Riparian Zones to Protect Flowing Waters and to Conserve Biodiversity: A Review and

Recommendations, Report to the Office of Water, Department of Sustainability and

Environment, Victoria.

Hill, A.R., Vidon, P.G.F., Langat, J., 2004. Denitrification potential in relation to lithology in

five headwater riparian zones. J. Environ. Qual., 33, 911–919.

Holzworth, D.P., Huth, N.I., deVoil, P.G., Zurcher, E.J., Herrmann, N.I., McLean, G., Chenu,

K., van Oosterom, E. J., Snow, V., Murphy, C., Moore, A.D., Brown, H., Whish, J.P.M.,

Verrall, S., Fainges, J., Bell, L.W., Peake, A.S., Poulton, P.L., Keating, B.A., 2014.

APSIM- evolution towards a new generation of agricultural systems simulation. Environ.

Model. Softw. 62, 327–350. https://doi.org/10.1016/j.envsoft.2014.07.009.

http://www.fao.org/docrep/017/i1688e/i1688e.pdf

http://www.fao.org/nr/water/aquastat/didyouknow/index3.stm

https://flex.flinders.edu.au/file/27320d77-7088-4d13-8bc5-9c2b1fc4a071/1/Thesis-Green-2010.pdf

https://flex.flinders.edu.au/file/27320d77-7088-4d13-8bc5-9c2b1fc4a071/1/Thesis-Green-2010.pdf

https://doi.org/10.1016/j.envsoft.2014.07.009


Kidmose, J., Dahl, M., Engesgaard, P., Nilsson, B., Christensen, B.S.B., Andersen, S.,

Hoffmann, C.C., 2010. Experimental and numerical study of the relation between flow

paths and fate of a pesticide in a riparian wetland. J. Hydrol. 386, 67–79.

King, S.E., Osmond, D.L., Smith, J., Burchell, M.R., Dukes, M., Evans, R.O., Knies, S.,

Kunickis, S., 2016. Effects of riparian buffer vegetation and width: A 12-year

longitudinal study. J. Environ. Qual. 45, 1243-1251.

Klatt, S., Kraus, D., Kraft, P. Breuer, L., Wlotzka, M., Heuveline, V., Haas, E., Kiese, R.,

Butterbach-Bahl, K., 2017. Exploring impacts of vegetated buffer strips on nitrogen

cycling using a spatially explicit hydro-biogeochemical modeling approach. Environ.

Model. Softw. 90, 55-67.

Lamontagne, S., Le Gal La Salle, C., Simmons, C., James-Smith, J., Harrington, N., Love, A.,

Smith, A., Hancock, G., Fallowfield, H., 2015. Estimation of Groundwater and

Groundwater N Discharge to the Adelaide Coastal Waters Study area. ACWS Technical

Report No. 4, Adelaide Coastal Waters Study Steering Committee, Flinders Centre for

Coastal and Catchment Environments, Fliders University, Adelaide, South Australia.

Laurenson, S., Kunhikrishnan, A., Bolan, N.S., Naidu, R., McKay, J., Keremane, G., 2010.

Management of recycled water for sustainable production and environmental protection:

A case study with Northern Adelaide Plains recycling scheme, Intern. J. Environ. Sci.

Dev. 1(2), 176-180.

Legates, D.R., McCabe, G.J., 1999. Evaluating the use of goodness-of-fit measures in

hydrologic and hydroclimatic model validation. Water Resour. Res. 35, 233-241.

Mayer, P.M., Reynolds, S.K., McCutchen, M.D., Canfield, T.J., 2007. Meta-analysis of

nitrogen removal in riparian buffers. J. Environ. Qual. 36, 1172-1180.

McKergow, L.A., Weaver, D.M., Prosser, I.P., Grayson, R.B., Reed, A.E., 2003. Before and

after riparian management: sediment and nutrient exports from a small agricultural

catchment, Western Australia. J. Hydrol. 270 (3), 253-272.

Naiman, R.J., Décamps, H., Mcclain, M.E., 2005. Riparia, Ecology, Conservation, and

Management of Streamside Communities, Elsevier Academic Press, Burlington, Ma.

Norris, R.H., Prosser, I.P., Young, B., Liston, P., Bauer, N., Davies, N., Dyer, F., Linke, S.,

Thoms, M., 2001. The Assessment of River Conditions (ARC). An Audit of the

Ecological Condition of Australian Rivers, CSIRO, National Land and Water Resources

Audit Office, Canberra.

Overton, I., Jolly, I., 2004. Integrated studies of floodplain vegetation health, saline

groundwater and flooding on the Chowilla floodplain, South Australia. CSIRO Land and


Water Technical Report 20/04.

http://www.clw.csiro.au/publications/technical2004/tr20-04.pdf.

Phogat, V., Cox, J.W., Šimůnek, J., 2018a. Identifying the future water and salinity risks to

irrigated viticulture in the Murray-Darling Basin, South Australia. Agric. Water Manage.

201, 107-117.

Phogat, V., Pitt, T., Cox, J.W., Šimůnek, J., Skewes, M.A., 2018b. Soil water and salinity

dynamics under sprinkler irrigated almond exposed to a varied salinity stress at different

growth stages. Agric. Water Manage. 201, 70-82.

Phogat, V., Potter, N.J., Cox, J.W., Šimůnek, J., 2017a. Long-term quantification of stream-

aquifer exchange in a variably saturated heterogeneous environment. Water Resour.

Manage. 31, 4353-4366.

Phogat, V., Skewes, M.A., Cox, J.W., Sanderson, J., Alam, J., Šimůnek, J., 2014. Seasonal

simulation of water, salinity and nitrate dynamics under drip irrigated mandarin (Citrus

reticulata) and assessing management options for drainage and nitrate leaching. J.

Hydrol. 513, 504-516.

Phogat V., Skewes, M.A., Cox, J.W., Šimůnek, J., 2016. Statistical assessment of a numerical

model simulating agro hydro-chemical processes in soil under drip fertigated mandarin

tree. Irrigat. Drainage Sys. Eng. 5(1) 1-9. http://dx.doi.org/10.4172/2168-9768.1000155.

Phogat, V., Skewes, M.A., McCarthy, M.G., Cox, J.W., Šimůnek, J., Petrie, P.R., 2017b.

Evaluation of crop coefficients, water productivity, and water balance components for

wine grapes irrigated at different deficit levels by a sub-surface drip. Agric. Water

Manage. 180, 22-34.

Rassam, D.W., 2011. A conceptual framework for incorporating surface–groundwater

interactions into a river operation–planning model. Environ. Model. Softw. 26(12), 1554-

1567.

Rassam, D.W., Fellows, C.S., De Hayr, R., Hunter, H., Bloesch, P., 2006. The hydrology of

riparian buffer zones; two case studies in an ephemeral and a perennial stream. J. Hydrol.

325, 308-324.

Reid, J.B., English, J.M., 2000. Potential yield in carrots (Daucus carota L.): Theory, Test, and

an Application. Ann. Bot. 85, 593-605.

Reynolds, A.G., 2010. Viticultural and vineyard management practices and their effects on

grape and wine quality. In: Managing wine quality volume 1: viticulture and wine

quality. Ed. A.G. Reynolds (Woodhead Publishing: Cambridge) pp. 365–444.

http://www.clw.csiro.au/publications/technical2004/tr20-04.pdf

http://dx.doi.org/10.4172/2168-9768.1000155


Ritchie, J.T., 1972. Model for predicting evaporation from a row crop with incomplete cover.

Water. Resour. Res. 8, 1204–1213.

Schilling, K.E., Jacobson, P.J. Wolter, C.F., 2018. Using riparian Zone scaling to optimize

buffer placement and effectiveness. Landscape Ecol. 33, 141–156.

Šimůnek, J., van Genuchten, M. Th., Šejna, M., 2016. Recent developments and applications

of the HYDRUS computer software packages. Vadose Zone J. 15(7), 1-25.

https://doi.org/10.2136/vzj2016.04.0033.

Stevens, D.P., McLaughlin, M.J., Smart, M.K., 2003. Effects of long-term irrigation with

reclaimed water on soils of the Northern Adelaide Plains, South Australia. Aust. J. Soil

Res. 41, 933-948.

The Goyder Institute for Water Research, 2016. Northern Adelaide Plains Water Stocktake

Technical Report, The Goyder Institute for Water Research, Adelaide, South Australia.

Xian, Y., Jin, M., Liu, Y., Si, A., 2017. Impact of lateral flow on the transition from connected

to disconnected stream–aquifer systems. J. Hydrol. 548, 353-367.

Xiao, Q., McPherson, E.G., Ustin, S.L., Grismer, M.E., Simpson, J.R., 2000. Winter rainfall

interception by two mature open-grown trees in Davis, California. Hydrol. Process. 14,

763–784.

Zhang, X., Liu, X., Zhang, M., Dahlgren, R.A., Eitzel, M., 2010. Vegetated buffers and a meta-

analysis of their mitigation efficacy in reducing non-point source pollution. J. Environ.

Qual. 39, 76–84.

https://doi.org/10.2136/vzj2016.04.0033

Appendix 11 Boron risks associated with recycled water irrigation | 378

Appendix 11 Boron risks associated with recycled water irrigation

Table of Contents

1 Introduction ................................................................................................................... 379

2 Methods and materials ................................................................................................... 380

2.1 Soil chemical and physical properties ....................................................................... 380

2.2 Boron analyses .......................................................................................................... 380

2.3 Boron sorption values ............................................................................................... 381

2.4 Irrigation water quality ............................................................................................. 382

2.5 1D solute transport model ........................................................................................ 382 2.5.1 Equilibrium sorption .................................................................................... 382 2.5.2 Non-equilibrium sorption ............................................................................ 383

3 Results and discussion .................................................................................................... 384

3.1 Boron speciation ....................................................................................................... 384

3.2 Native boron concentration in NAP soils .................................................................. 385 3.2.1 Factors affecting boron availability ............................................................. 385 3.2.2 Soil solution boron ....................................................................................... 385 3.2.3 Adsorbed soil boron ..................................................................................... 388 3.2.4 Total soil boron ............................................................................................ 390 3.2.5 Boron desorption ......................................................................................... 391 3.2.6 Boron sorption ............................................................................................. 393 3.2.7 Simulated boron behaviour ......................................................................... 398

4 Conclusion ...................................................................................................................... 406

5 References ...................................................................................................................... 407


1 Introduction

Boron (B) is a micronutrient that is required by plants in small quantities (<500 g/ha) (Shorrocks, 1997) and since B is relatively immobile in plants, once utilized in actively growing tissues, it is hardly re-translocated to other parts. Therefore, it is necessary to have a rather continuous source of B available to the plant throughout its growth cycle. Boron is required for the formation of new tissues but not the maintenance of older tissues, so actively growing plants require larger amounts of B than slowly growing or mature plants (Adriano, 2001).

Boron deficiency commonly occurs in sandy soils which have low CEC and OM content, where leaching and heavy cropping have diminished the soil B reserves (Adriano, 2001). Boron toxicity usually is seen in soil of marine sediment, in those soils derived from parent material rich in B, and in arid and semi-arid soils (Adriano, 2001).

Of all plant nutrient elements, the range between deficient and toxic levels of available concentration is smallest for boron (Goldberg, 1997). Small increases due to fertilizer application or via boron in irrigation water and natural variations in boron concentration with soil depth may result in a soil transitioning from deficient to toxic levels or vice versa.

An optimal soil test would measure B capacity, that is, all pools of plant available B including: soluble, organic, and adsorbed. Determination of the total B pool potentially available to plants is necessary for arid and semiarid land soils to evaluate the extent of leaching necessary for reducing B to below toxic levels.

The main factors affecting boron adsorption and plant available boron are soil pH, texture, water content, and temperature.


2 Methods and materials

2.1 Soil chemical and physical properties

Following the soil sampling, soil pH was measured in water (pHW) and in 0.01 M CaCl2 (pHCa). The latter is usually preferred as it is less affected by soil electrolyte concentration and provides more consistent measurements (Minasny et al., 2011). Further details of the analysis can be found in the Task 1 Report (Oliver et al., 2019).

2.2 Boron analyses

The boron concentration in soil solution, and therefore the boron available for plant uptake, is governed by boron adsorption rather than by the solubility of boron containing minerals (Goldberg, 1997). The fraction of B in soil that is available for plant uptake is termed the phytoavailable fraction. Total B is an unreliable measure of the bioavailable fraction in soils and often an extractant, such as water or CaCl2, is used as an index of the phytoavailable fraction (Adriano, 2001). Aitken and McCallum (1988) however, showed that the relationship between B concentrations measured in sunflowers and the hot 0.01M CaCl2 extractable fraction was dependent upon the soil texture. They found a better relationships with the soil solution fraction measured at maximum water holding capacity (MWHC) (termed B in soil solution).

Plants have varying degrees of tolerance to B in soil solution and Adriano (2001) suggests B concentrations <0.5 mg/L in soil solution are probably safe for most plants but many plants may be adversely affected when B levels are in the range of 0.50 to 5.0 mg/L. Threshold concentration ranges for B concentration in soil solution based on Leyshon and Jame (1993) are given in Table 1.

Table 1 Threshold boron concentration in soil solution for various crops and percentage of total samples (0-10, 10-30 and 30-60 cm) within each threshold range. Boron in soil solution determined at maximum water holding capacity. Threshold concentrations from Leyshon and Jame (1993). * data from Gupta et al. (1985).

Threshold concentration range for B in soil solution (field capacity basis)1

(mg/L B)

Percentage (number) of soil. Total soils n=82.

Crop within threshold concentration range

Very sensitive <0.5 67% (55) Lemon*, Grapefruit*, Avocado*, Orange*

Sensitive 0.5-1.0 18% (15) Fig, Grape, Walnut, Onion, Garlic

Moderately sensitive

1.0-2.0

5% (4) Broccoli, Red pepper, Carrot, Potato, Cucumber

Moderately tolerant

2.0-4.0

6% (5) Lettuce

Tolerant 4.0-6.0 4% (4) Tomato


Boron in soil solution was determined at maximum water holding capacity (MWHC) based on McLaughlin et al. (1997). To better represent the soil solution following irrigation, the B in soil solution was measured using a solution with a Cl concentration equivalent to that found in one of the primary irrigation water sources being considered, namely recycled waste water. The Cl concentration in the recycled waste water was 550 mg/L, compared with 4 mg/L in rain water. There was almost a 1:1 relationship between B in soil solution as determined using the high (550 mg/L) and low (4 mg/L) Cl solution (Task 1 Report, Oliver et al., 2019) so only data for the high Cl solutions are discussed below.

A subset of soils were sequentially extracted in duplicate with a high Cl (550 mg/LCl) solution in a 1:5 ratio. The high Cl solution was selected to represent the highest concentration in the reclaimed water from the DAFF treatment plant. The soils selected had been found to have relatively high B in soil solution and the sequential extractions were performed to determine whether B would continue to come into solution with each successive extraction. The soil solution was shaken on an end-over-end shaker for 24h, then centrifuged at 3000 rpm for 15 mins. The extract was decanted off and a fresh solution was added and the extraction repeated four times. The B extracted was determined by ICP-MS (Task 1 Report, Oliver et al., 2019).

Hot water extraction of boron is a commonly used method for obtaining an index of plant available boron in soils (Cox and Kamprath, 1972; Bingham, 1982). The hot water-soluble procedure measures B capacity in that it extracts B from the organic, adsorbed, and soluble pools of the soil (Offiah and Axley, 1993).

2.3 Boron sorption values

Sorption coefficients (Kd values) were determined for boron (B) using the OECD 106 standard protocol for the adsorption – desorption of chemicals using a batch equilibrium method (OECD/OCDE, 2000). Briefly, the procedure involved weighing 2g of soil into plastic vials in triplicate. There was an initial pre-equilibration for 24 h with 9.5 mL of 0.01M CaCl2, after which a known volume (0.5 mL) of a spiking solution of B was added to given an initial concentration (starting concentration) of 0.5 mg B/L. The solution was shaken again for 24h on an end-over-end shaker and the concentration remaining in solution was measured using an inductively coupled plasma mass spectrometer (ICP-MS). The soils selected for determining B Kd values were chosen to cover a range of soil textures, soil pH and native B concentrations.

The sorption coefficients (Kd) values were calculated as:

𝐾𝐾𝑑𝑑 =𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐶𝐶𝑜𝑜 𝑠𝑠𝐶𝐶𝐶𝐶𝑠𝑠𝐶𝐶𝑑𝑑 𝐶𝐶ℎ𝐶𝐶𝑒𝑒𝐶𝐶𝐶𝐶𝐶𝐶𝑒𝑒 (𝑚𝑚𝑚𝑚

𝑘𝑘𝑚𝑚𝑠𝑠𝐶𝐶𝐶𝐶𝑒𝑒)

𝐸𝐸𝐸𝐸𝐸𝐸𝐶𝐶𝑒𝑒𝐶𝐶𝑠𝑠𝐶𝐶𝐶𝐶𝐸𝐸𝑒𝑒 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶 𝑠𝑠𝐶𝐶𝐶𝐶𝑒𝑒 𝑠𝑠𝐶𝐶𝑒𝑒𝐸𝐸𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 (𝑚𝑚𝑚𝑚𝐿𝐿 )

= 𝐶𝐶𝑆𝑆𝐶𝐶𝐿𝐿

(1)

The concentration of the chemical sorbed by the soil was determined as:

𝐶𝐶𝐿𝐿 = [𝐼𝐼𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝑒𝑒 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 (𝑒𝑒𝑚𝑚/𝐿𝐿)×𝑣𝑣𝐶𝐶𝑒𝑒𝐸𝐸𝑒𝑒𝐶𝐶(𝐿𝐿)]−[𝐸𝐸𝐸𝐸𝐸𝐸𝐶𝐶𝑒𝑒𝐶𝐶𝑠𝑠𝐶𝐶𝐶𝐶𝐸𝐸𝑒𝑒 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶(𝑒𝑒𝑚𝑚/𝐿𝐿)×𝑣𝑣𝐶𝐶𝑒𝑒𝐸𝐸𝑒𝑒𝐶𝐶(𝐿𝐿)]𝑀𝑀𝐶𝐶𝑠𝑠𝑠𝑠 𝐶𝐶𝑜𝑜 𝑠𝑠𝐶𝐶𝐶𝐶𝑒𝑒 (𝑘𝑘𝑚𝑚)

(2)

The equilibrium mass was corrected for native B that came into solution when blank soils were shaken in 1:5 ratio in 0.01M CaCl2 (see Task 1 Report for details, Oliver et al., 2019).


2.4 Irrigation water quality

Boron concentration in recycled DAFF water were obtained from the Task 3 Report (Awad et al., 2019). A summary of the data is provided in Table 2.

Table 2 Boron concentration (mg/L) in Bolivar DAFF irrigation water. Period No. of

observations 5th percentile 50th percentile 95th percentile

2001-2011 122 0.204 0.327 0.529

2012-2016 57 0.197 0.334 0.529

2.5 1D solute transport model

2.5.1 Equilibrium sorption

Boron concentration in soil solution is considered to be determined mainly by adsorption-desorption reactions (Goldberg, 1997). Boron containing minerals such as tourmaline or hydrated boron minerals are either too insoluble or too soluble to control boron concentration. Main sorption sites for boron in soil are clays, aluminuim and iron oxides, magnesium hydroxite, calcite, and organic carbon (Goldberg, 1997). For clay minerals, the order of B adsorption per gram is: kaolinite < montmorillonite < illite (Keren and Mezuman, 1981). Calcium carbonate (CaCO3) acts as an important B adsorbing surface in calcareous soils. Also, addition of CaCO3 increases B fixation by soils because it increases the soil pH (Goldberg and Forster, 1991). Modelling of equilibrium boron adsorption on oxides, clay minerals, organic matter and other soil solid phases have typically been described using Langmuir and Freundlich adsorption isotherms (Goldberg, 1997; Marzadori et al., 1991). Equilibruim sorption assumes that the adsorption-desorption process is fast relative to fluid movement in soil and thus not time-dependent. The mathematical expression for the Langmuir equilibrium isotherm is as follows (Zhang and Selim, 2005): 𝑆𝑆 = 𝑆𝑆𝑒𝑒𝐶𝐶𝑚𝑚

𝐾𝐾𝐿𝐿×𝐶𝐶1+𝐾𝐾𝐿𝐿𝐶𝐶

(3)

where S is the total amount of adsorbed boron (µg/g), Smax represents the sorption maximum that can be related to soil properties (µg/g), KL is the Langmuir coefficient which is related to the binding strength (L/mg), and C is the boron concentration in solution (mg/L). The implementation of the Langmuir sorption isotherm in HYDRUS-1D has a slightly different notation (Šimůnek et al., 2008; Mallants et al., 2011): 𝑆𝑆 = 𝐾𝐾𝐷𝐷

𝐶𝐶1+𝜂𝜂×𝐶𝐶

(4)

where KD (L/kg) and ƞ (L/mg) are empirical or quasi-empirical constants. The parameter ƞ in Eq. (4) is equivalent to parameter KL in Eq (3) while KD in Eq. (4) is equivalent to Smax × KL. The Freundlich isotherm is defined as (Šimůnek et al., 2008; Mallants et al., 2011):


𝑆𝑆 = 𝐾𝐾𝐹𝐹𝐶𝐶𝑁𝑁 (5) where KF and N are the Freundlich parameters.

2.5.2 Non-equilibrium sorption

When the adsorption-desorption reaction is partially time dependent, the two-site sorption concept can be implemented (Selim et al., 1977; van Genuchten and Wagenet, 1989). In this case the sorption sites are split into equilibrium sites where sorption is instantaneous (type-1 sites with fast exchange between solid and liquid phase: SEq (µg/g) and sites where sorption is kinetically controlled (type-2 sites with time-dependent sorption: SKin (µg/g)). The mathematical expression for the mass balance of the two-site chemical non-equilibrium is (Šimůnek et al., 2008; Mallants et al., 2011):

𝑆𝑆𝑇𝑇𝐶𝐶𝐶𝐶 = 𝑆𝑆𝐸𝐸𝐸𝐸 + 𝑆𝑆𝐾𝐾𝐶𝐶𝐶𝐶 (6)

𝑆𝑆𝐸𝐸𝐸𝐸 = 𝑓𝑓𝑆𝑆𝑇𝑇𝐶𝐶𝐶𝐶 (7)

𝜕𝜕𝑆𝑆𝐾𝐾𝐾𝐾𝐾𝐾𝜕𝜕𝐶𝐶

= 𝛼𝛼 �(1 − 𝑓𝑓)𝐾𝐾𝐷𝐷𝐶𝐶

1+𝜂𝜂×𝐶𝐶− 𝑆𝑆𝐾𝐾𝐶𝐶𝐶𝐶� (8)

where STot represent the total sorbed phase concentration (µg/g), f (-) is the fraction of equilibrium sorption sites (defined by Eq. (7)), α is a first-order rate constant (h-1), and KD and ƞ are as defined previously. In Eq. (8) we have assumed that equilibrium sorption is represented by a Langmuir type isotherm, although linear and Freundlich isotherms can also be invoked.


3 Results and discussion

3.1 Boron speciation

Boron chemistry in soil is very simple: it does not display redox reactions and is not volatile. Dissolved boron in soil pore water is present only in the +3 valence state (Figure 1). The dominant boron species in soils at low pH values is the neutral and weak boric acid, B(OH)3. As the soil pH increases, boric acid forms the borate anion by accepting a hydroxyl ion: as a result, the proportion of the borate anion, B(OH)4

-, increases (Figure 1b). The ratio of these two boron species depends on the first dissociation constant of boric acid, Ka. The fresh water Ka value at temperature 25 °C is equal to 5.8×10−10 mol/L. The corresponding pKa value = 9.24 (pKa = - log(Ka)). As a result, at pH < 9.2, dissolved boron predominantly exists in the form of an uncharged oxyanion (B(OH)3) while at pH > 9.2, it is mostly present as B(OH)4

− (Figure 1).

The boron pKa value is known to be dependent on salinity and temperature. A closed-form equation was developed by Dickson (1990) to quantify the salinity and temperature dependency. We implemented this equation to calculate the pKa values for i) a temperature range from 5 to 45°C and, ii) a low (2042 mg/L, based on a maximum measured soil EC was 3712 μS/cm) and high salinity (5000 mg/L). Note that the lowest salinity level S in the Dickson model was 5000 mg/L, hence effects of the low salinity will be immaterial. Based on these calculated pKa values, the speciation model PhreeqC was run to test the effect on the B(OH)3 and B(OH)4

− species (Figure 1b). Temperature effects on the speciation are noticeable, but small. We conclude that the standard pKa value at 25°C is suitable for the NAP conditions.

Figure 1 Eh–pH predominance diagram of boron at 25 °C, 1.013 bars and activity of B = 10−6 calculated with The Geochemist’s WorkbenchTM (Bethke et al., 2019) (left). Distribution of aqueous boron species versus pH in soil water at

low EC (ionic strength I = 2.714×10-2 mol/kgw between pH 7-9) and high EC (5 g/L salinity) (right). Boron speciation calculations based on PhreeqC (Parkhurst and Appelo, 2013).

Boron adsorption in soil is predominantly on oxides, clay minerals, calcite, and organic matter (Goldberg et al., 1993). The adsorption of boron in US soils was shown to be pH dependent with a maximum sorption at pH 6 to 8 for soil rich in Al-oxides and between pH 7 to 9 for soils with predominantly Fe-oxides (Goldberg and Glaubig, 1985; Goldberg et al., 1993) (Figure 2). At low pH, B sorption is low and dominated by the neutral species B(OH)3. As the pH increases, the borate ion,

0 2 4 6 8 10 12 14pH

0

0.2

0.4

0.6

0.8

1

Boro

n Fr

actio

n

B(OH)4- at 45 °C, 5g/L

B(OH)3 at 45 °C, 5 g/L

B(OH)4- at 5 °C, 5 g/L

B(OH)3 at 5 °C, 5 g/L

B(OH)4- at 25 °C, low EC

B(OH)3 at 25 °C, low EC


B(OH)4-, becomes the most abundant species characterized by an increase in sorption (Goldberg and

Glaubig, 1986; Adriano, 2001). However, further increases in pH (i.e. pH > 9) result in increased OH- concentration relative to B(OH)4

- causing B adsorption to decrease due to competition with OH- for sorption sites (Goldberg and Glaubig, 1986; Adriano, 2001).

Boron adsorption for soils from the Northern Adelaide Plains will be discussed in the next section.

Figure 2 Adsorption of boron versus pH for a variety of soil (from Goldberg et al., 2000).

3.2 Native boron concentration in NAP soils

3.2.1 Factors affecting boron availability

Boron availability in soils is mainly affected by pH: as pH increases, boron adsorption increases and therefore its availability to plants decreases. As shown in Figure 2, boron adsorption increases as function of soil pH in the range 3 to 9, and decreases in the pH range 10-11.5.

3.2.2 Soil solution boron

Of all the soils assessed (surface layers 0-10, 10-30 and 30-60 cm), boron in soil solution was low (<0.5 mg/L B) for 72% of soils, while 13% of soils had >1 mg/L and 8% had >2 mg/L boron in soil solution (Task 1 Report, Oliver et al., 2019). Comparison of these boron levels with threshold values above which crops become sensitive to B indicates that the native B in some of the soils may already be at concentrations that are limiting or toxic to crop growth (Table 1).


The distribution of B in soil solution at the three sampling depths for the soils sampled across the NAP region is shown in Figure 3 (and in Figure 4 expressed per kg soil) and listed in Table 3. Hard red brown soils display the overall highest concentrations, where for the remaining three soils boron concentrations are similar. Also, hard red brown soils have the highest relative variability in boron as expressed through the coefficient of variation, CV (Table 3).

Boron concentration for all soils increases with depth. This may be due to several reasons, including higher organic matter in the top soil layer providing significant sorption capacity, higher concentrations of boron minerals at greater soil depth (inferred from total B versus depth, Figure 4), and leaching of boron under natural rainfall conditions. The mean boron concentration versus depth data will be used as initial solute conditions for the simulations with HYDRUS-1D; as the model domain is 2 m deep, the measured concentration at the 0.45 m depth is used throughout the remaining soil depth as initial concentration.

Figure 3 Boron in soil solution for soils of the Northern Adelaide Plains (data from Task 1).

0 4 8 12Boron in soil solution (mg/L)

50

40

30

20

10

0

Soil

dept

h (c

m)

Hard red brown soil

0 0.4 0.8 1.2 1.6Boron in soil solution (mg/L)

50

40

30

20

10

0

Soil

dept

h (c

m)

Deep uniform to gradational soil

0 0.1 0.2 0.3 0.4 0.5 0.6Boron in soil solution (mg/L)

50

40

30

20

10

0

Soil

dept

h (c

m)

Sand over Clay soil

0 0.2 0.4 0.6 0.8 1Boron in soil solution (mg/L)

50

40

30

20

10

0

Soil

dept

h (c

m)

Calcareous soil

maximum

minimum

mean


Table 3 Boron in soil solution as extracted using 550 mg/L chloride solution. Stdev = standard deviation, CV = coefficient of variation (=100x(Stdev/mean)) , N = number of observations. Data expressed as mg/kg are available in Figure 4 (data from Task 2).

Soil depth (cm)

Min (mg/L)

Max (mg/L)

Mean (mg/L)

Median (mg/L)

Stdev (mg/L)

CV (%) N

Hard red brown soil

0-10 0.12 0.77 0.29 0.23 0.21 70.7 11

10-30 0.12 9.1 1.3 0.35 2.63 201.8 12

30-60 0.06 11 2.07 0.80 2.99 142.2 16


0-10 0.12 0.29 0.21 0.21 0.12 56.9 2

10-30 0.19 0.60 0.34 0.27 0.18 54.1 4

30-60 0.23 1.21 0.63 0.65 0.39 61.8 5

Sand over clay soil

0-10 0.09 0.11 0.10 0.11 0.01 9.8 5

10-30 0.16 0.26 0.21 0.22 0.05 22.2 4

30-60 0.33 0.56 0.44 0.44 0.13 29.3 4

Calcareous soil

0-10 0.14 0.31 0.24 0.27 0.08 31.8 5

10-30 0.21 0.38 0.3 0.3 0.07 24.4 6

30-60 0.26 2.86 1.24 0.75 1.11 90.1 6


Figure 4 Total boron in soils (mg/kg soils) of the Northern Adelaide Plains (data from Task 1).

3.2.3 Adsorbed soil boron

The concentration of adsorbed boron was determined in two ways. The first method uses the hot water extraction of boron in 0.01M CaCl2, commonly used to measure B extracted from the adsorbed pools (organic, clays) and soluble pools of the soil (Offiah and Axley, 1993). The second method uses the batch method to determine Kd, in which the absorbed concentration is one of the measured parameters (see higher).

Adsorbed boron determined with the hot water extraction is shown in Figure 5 for hard red brown and deep uniform to gradational soil. Hard red brown soil have a maximum of adsorbed boron at about 40 cm depth, whereas the deep uniform to gradational soil display increasing adsorbed boron concentrations with depth. Boron behaviour in hard red brown soils is likely related to the depth distribution of the clay fraction; a typical clay fraction is 14 % (0-10 cm), 36% (10-30 cm), 55% (30-60 cm), 37% (60-90 cm), and 29% (90-120 cm). In the deep uniform to gradational soil, clay content

0 2 4 6 8Total boron in soil solution (mg/kg)

50

40

30

20

10

0

Soil

dept

h (c

m)

Hard red brown soil

0 0.2 0.4 0.6 0.8Total boron in soil solution (mg/kg)

50

40

30

20

10

0

Soil

dept

h (c

m)


0 0.05 0.1 0.15 0.2 0.25Total boron in soil solution (mg/kg)

50

40

30

20

10

0

Soil

dept

h (c

m)

Sand over Clay soil

0 0.1 0.2 0.3 0.4Total boron in soil solution (mg/kg)

50

40

30

20

10

0

Soil

dept

h (c

m)

Calcareous soil

maximum

minimum

mean


more or less continues to increases with depth; a typical profile is 12% (0-10 cm), 4% (10-30 cm), 16% (30-60 cm), 27% (60-90 cm), and 27% (90-120 cm).

Figure 5 Adsorbed boron for soils of the Northern Adelaide Plains based on hot water extraction (data from Task 1). Hard red brown based on data from soil profile NAP3, NAP4, NAP6, and NAP7. Deep uniform to gradational from soil

profile NAP1, NP2, NAP5.

Table 4 Adsorbed boron using hot water extract. Stdev = standard deviation, CV = coefficient of variation (=100x(Stdev/mean)) , N = number of observations.

Soil depth (cm)

Min (mg/kg)

Max (mg/kg)

Mean (mg/kg)

N

Hard red brown soil

0-10 1.04 2.88 1.74 4

10-30 1.00 39.82 12.88 4

30-60 2.10 28.46 14.25 4

60-100 1.78 27.06 10.57 6

100-200 2.98 20.47 11.36 4


0-10 0.72 2.09 1.30 3

10-30 0.76 2.48 1.45 3

30-60 1.4 7.33 2.96 5

60-100 2.1 17.23 7.17 3

100-200 3.67 24.6 12.08 5

0 20 40 60 80 100Adsorbed boron in soil (mg/kg)

160

120

80

40

0

Soil

dept

h (c

m)

Hard red brown soil

0 5 10 15 20 25Adsorbed boron in soil (mg/kg)

120

100

80

60

40

20

0

Soil

dept

h (c

m)


maximum

minimummean

Number of soil profiles = 4 Number of soil profiles = 3


3.2.4 Total soil boron

The aqua regia digestion method (US-EPA 3050 (1996) or ISO standard 11466 (1995)) is considered effective for measuring “total” trace element in soils and is usually used to give an estimate of the maximum element availability to plants. Depth distribution of total soil boron is shown in Figure 6 for all four soils. With hard red browns showing the overall highest concentrations, these soils could potentially cause suboptimal crop production for boron sensitive crops. Note that the total soil boron determined with the aqua regia digestion is sometimes lower than that determined with the hot water extraction shown in Figure 5. This is because measurements for those two measurements were based on samples from slightly different depths; given the vertical variability in clay, organic matter, mineralogy and other properties, vertical variability in total boron is expected to be considerable. It is therefore best to consider both the aqua regia digestion and the hot water extract together to estimate the total boron concentration in this study.

Figure 6 Total boron in soils of the Northern Adelaide Plains (microwave soil digest using reverse aqua regia) (data from Task 1).

0 20 40 60Total boron in soil (mg/kg)

80

60

40

20

0

Soil

dept

h (c

m)

Hard red brown soil

2 4 6 8 10Total boron in soil (mg/kg)

120

100

80

60

40

20

0

Soil

dept

h (c

m)


0 5 10 15 20 25Total boron in soil (mg/kg)

160

120

80

40

0

Soil

dept

h (c

m)

Sand over Clay soil

0 5 10 15 20 25Total boron in soil (mg/kg)

80

60

40

20

0

Soil

dept

h (c

m)

Calcareous soil

maximum

minimummean

(N=2)

(N=3)

(N=2)

(N=4) (N=1)

(N=1)

(N=1)

(N=1)

(N=4)

(N=1)

(N=1)

(N=1)

(N=1)

(N=1)

(N=2)

(N=2)


Total boron concentration in hard red brown was shown to be correlated with clay% and pH (Figure 7); for other soils such correlation was not evident (likely due to the limited number of data points).

Figure 7 Total boron as function of calcite, clay, and pH (data from Task 1).

3.2.5 Boron desorption

Sequential leaching tests over a period of 96 hours provided data on time-dependent boron desorption which were used to derive the first-order rate coefficient α (h-1). This analysis was undertaken only for hard red brown soil, as this is the soil with the highest concentrations and thus presents the highest risk. Figure 8a and b show the desorption curves for hard red brown soil depicted as liquid phase concentration (mg/L) versus time from which the cumulative desorbed boron concentration (mg/kg) was derived. In order to derive the first-order mass transfer coefficient α, the cumulative desorbed boron data were re-arranged based on the following form of the first-order kinetic expression (Pavlatou and Polyzopoulos, 1988):

0 20 40 60 80CaCO3 (%)

0

10

20

30

40

50

60

B - m

g/kg

Hard red brown soilCalcareous soilSand over clay soil


Y = 9.018 * ln(X) + 23.29 (r = 0.77)

0 20 40 60

Clay %

0

10

20

30

40

50

60

B - m

g/kg



7.6 7.8 8 8.2 8.4 8.6 8.8pH

0

10

20

30

40

50

60

B - m

g/kg



Y = 48.7 * X - 365.04 (r=0.79)


𝐿𝐿𝐿𝐿(𝑆𝑆𝑒𝑒𝐶𝐶𝑚𝑚 − 𝑆𝑆) = 𝐿𝐿𝐿𝐿 𝑆𝑆𝑒𝑒𝐶𝐶𝑚𝑚 − 𝛼𝛼 𝑡𝑡 (9)

where Smax is the total sorption capacity of the soil (mg/kg), S is the sorbed amount (mg/kg) at time t (h). Values for Smax were put equal to the total boron concentration obtained by microwave soil digest using reverse aqua regia. Least squares fitting of Eq. (9) with Smax fixed at independently measured values yielded the first-order rate coefficient α (Figure 8b and d). Table 5Table 5 provides a summary of parameters of Eq. (9).

Figure 8 Desorption data (a, c) and fitted first-order kinetic model (b, d). The desorption includes instantaneous (left axis) and cumulative values (right axis). Fitted parameters for models in (d) are given in Table 5.

The fraction of equilibrium sites, f, was calculated according to Eq. (7) from values of Smax and the total desorbed boron at the end of the desorption tests, considered to be equivalent to SEq. This assumes that the desorbed boron in the first 96 hrs is readily available for desorption and that it

0 20 40 60 80 100Time (hr)

0

1

2

3

4

5

Liqu

id p

hase

bor

on c

once

ntra

tion

(mg/

L)

0

10

20

30

40

50

Cum

ulat

ive

deso

rbed

bor

on (m

g/kg

)

CL014 20 cmCL014 40 cmNAP06 20 cm

Hard red brown

0 20 40 60 80 100Time (hr)

2

2.5

3

3.5

4

4.5

Amou

nt o

f bor

on d

esor

bed

(mg/

kg)

CL014 20 cmCL014 40 cmNAP06 20 cm

Hard red brown

ln(Smax - S) = 4.177 - 0.00152 t (r=0.962)

ln(Smax - S) = 4.177 - 0.0070 t (r=0.958)

ln(Smax - S) = 4.177 - 0.01059 t (r=0.924)

0 20 40 60 80 100Time (hr)

0

5

10

15

20

Liqu

id p

hase

bor

on c

once

ntra

tion

(mg/

L)

0

40

80

120

Cum

ulat

ive

deso

rbed

bor

on (m

g/kg

)

NAP06 45 cmNAP07 45 cmNAP13 45 cmNAP15 45 cmNAP20 45 cm

Hard red brown

0 20 40 60 80 100Time (hr)

2

3

4

5

6

Amou

nt o

f bor

on d

esor

bed

(mg/

kg)

NAP06 45 cmNAP07 45 cmNAP13 45 cmNAP15 45 cmNAP20 45 cm

Hard red brown


provides a reasonable estimation of the boron on the equilibrium sites. Although the desorption curves have not yet reached a steady-state (Figure 8a, c), considering the very long simulation times (several tens of years) considered here, the desorption data represented here can be considered to only represent relatively rapidly released boron. The second approach assumes that the total desorbed concentration obtained through the hot water extraction is a good estimator of SEq. Both approaches to estimate f are included in Table 5.

The inherent limitation in the data is that the short time scale involved cannot provide accurate reaction parameters for the long-term release of boron. For this reason the simulated boron releases are considered a sensitivity analysis only, and are not aimed at accurately predicting the boron behaviour in the soil profile under intense irrigation with recycled water.

Table 5 Fitted first-order kinetic parameter Smax for data shown in Figure 8. Hard red brown soils (desorbed after 96 hrs). Smax = total boron from microwave soil digest; f = S1/Smax, where S1 is cumulative desorbed boron at t = 96 hrs or total desorption from hot water extraction; r = correlation coefficient for linear regression.

Soil depth Reaction rate constant α (h-1)

Ln(Smax) Correlation coefficient r

Fraction of S1 sites, f (-)

[desorbed]

Fraction of S1 sites, f (-) [hot

water extraction]

CL014 20 cm 0.0070 4.177 0.958 0.467 0.0266

CL014 40 cm 0.01059 4.177 0.924 0.743 0.197

NAP06 20 cm 0.00152 4.177 0.962 0.605 0.610

NAP06 45 cm 0.0105 5.289 0.938 0.605 0.457

NAP07 45 cm 0.0129 3.663 0.977 0.692 0.366

NAP13 45 cm 0.0128 3.200 0.987 0.692 0.580

NAP15 45 cm 0.0164 3.610 0.998 0.797 0.385

NAP20 45 cm 0.00937 3.761 0.966 0.569 0.331

3.2.6 Boron sorption

Three types of sorption isotherm were derived for boron sorption, based on hard red brown soil: linear, Freundlich, and Langmuir.

The Langmuir parameters Smax and KL were obtained by fitting Eq. (3) to the isotherm data from Figure 9. The following data points were taken into consideration: equilibrium boron concentrations in liquid and solid phase from batch tests, and total desorbed boron concentration using the hot water extraction with 0.01 M CaCl2 solution. The latter is considered to provide an estimate of the boron adsorbed on clay minerals and organic carbon and boron in the soluble pools (Offiah and Axley, 1993). In estimating parameters of Eq. (3) we assumed the hot water extraction data represents Smax, which was then fixed for fitting the remaining parameter, i.e. with KL the only fitting parameter. Fitting by eye was done with results shown in Table 6; because only very few data points were available, fitting by eye is not worse than automated error minimization.


Results from the parameter estimation must be treated with care. Indeed, because the use of adsorption isotherms is basically a curve fitting excercise, the fitting parameters are only valid for the conditions under which the experiment was conducted. Therefore, prediction of B adsorption for conditions beyond those of the experiment will be highly unreliable, especially if this involves changes of soil solution B concentration, pH, and ionic strength (Goldberg, 1997).

Figure 9 Langmuir sorption isotherms. For hard red brown soils four data sets were considered for estimation of the Langmuir isotherms. The data originates from four soil profiles (NAP4, NAP6, NAP7, and NAP13) at three depths (0-10, 10-30, and 30-60 cm). The liquid phase/solid phase data were grouped according to the main soil sorbing material, i.e. clay and calcite. The four data groups also have a distinctly different pH. The first group (0-10 cm for NAP 4 and 10-30 cm for NAP7) has a pH between 5.9 and 6.3 with an equilibrium boron pore-water concentration of 0.54-0.57 mg/L (Figure 9). Both soil layers have a

0 0.2 0.4 0.6 0.8 1 1.2 1.4Liquid phase B (mg/L)

0

0.4

0.8

1.2

1.6

2

2.4

2.8

3.2

Adso

rbed

B (µ

g/g)

0-10 & 10-30 cm10-30 cm30-60 cm

Calcareous soil

0 1 2 3 4 5 6Liquid phase B (mg/L)

0

2

4

6

8

10

12

14

16

Adso

rbed

B (µ

g/g)

0-10 & 10-30 cm L-Ca30-60 cm L-Ca

30-60 cm H-Ca

10-30 cm H-Ca

Langmuir 1

Langmuir 2

Langmuir 3

Langmuir 4Freundlich 1

Freundlich 2

Freundlich 3

Freundlich 4

Hard red brown soil

0 0.2 0.4 0.6 0.8 1 1.2 1.4Liquid phase B (mg/L)

0

0.4

0.8

1.2

1.6

2

2.4

2.8

3.2

3.6

Adso

rbed

B (µ

g/g)

0-10 cm0-10 & 10-30 cm30-60 cm


0 0.1 0.2 0.3 0.4 0.5 0.6Liquid phase B (mg/L)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Adso

rbed

B (µ

g/g)

0-10 & 10-30 cm30-60 cm10-10 & 10-30 & 30-60 cm

Sand over clay soil


relatively low clay percentage (8-14%) and low levels of calcite (0.1%). The fitted Langmuir isotherm (“Langmuir 1”) is shown in Figure 9, with parameters listed in Table 6.

The second data group (30-60 cm from NAP13) has a pH of 8.4 and equilibrium boron concentrations of 1 mg/L. This soil layer has high clay content (44%) but low calcite (01%). Parameters for the “Langmuir 2” model are provided in Table 6 while Figure 9 shows data and fitted isotherm.

The third data group (30-60 cm from NAP7) has a pH of 8.9, the boron concentration in solution is 3.1 mg/L, and the percentage clay and calcite are 33 and 1.1%, respectively. Estimated parameters for the “Langmuir 3” model are given in Table 6 while data and model are shown in Figure 9.

The fourth data group (10-30 cm from NAP6) has a pH of 8.9, pore-water boron concentration of 5.8 mg/L, 27% clay and 3.2% calcite. This group has the overall highest value for Smax (Table 6) with corresponding isotherm “Langmuir4” shown in Figure 9.

Isotherm parameters were not determined for the other soil groups mainly for two reasons: i) there is no desorption data for those soils hence there are no first-order kinetic parameter values, ii) hard red brown soil has the highest risk of boron toxicity, hence we focus on this soil group.

Table 6 Parameters of the Langmuir sorption isotherms for boron (hard red brown soil). Soil depth (cm) Soil description Langmuir parameters Model name

Smax (µg/g) KL (L/mg)

0-10 (NAP4)

10-30 (NAP7)

14% clay; 0.1% calcite


1.52 1.5 Langmuir 1

30-60 (NAP13) 44% clay; 0.1% calcite

12.14 0.35 Langmuir 2

30-60 (NAP7) 33% clay; 1.1 % calcite

21.37 0.18 Langmuir 3

10-30 (NAP6) 27% clay; 3.2% calcite

65.23 0.053 Langmuir 4

The sorption maximum Smax is clearly related to the soil properties % clay and % calcite: the lowest % of clay and calcite produced the lowest value for Smax (1.52, seeTable 6), while the soil with the highest % calcite (3.2) and high % clay (27) had the highest Smax (65.23, see Table 6). Intermediate values for Smax exist for the soils with intermediate % calcite (1.1%) and high % clay (33%) or high % clay (44) but low calcite (0.1%). High calcite seems to dominate over high clay content; this is explained by the higher pH (8.9) when calcite is high (1.1-3.2%). As was shown in the boron speciation diagram (Figure 1), the boron species B(OH)3 and B(OH)4

-are present in about equal fractions with boron sorption being close to or at its maximum (also see Figure 2).

For the purpose of simulating boron behaviour in soil two hypothetical soil profiles were composed, each consisting of a set of the previously defined Langmuir models (Table 7). Type-1 profile is build around Langmuir model 1 and 3 has small Smax values from 0 to 30 cm, then Smax increases to medium values for the remainder of the profile. Type-2 profile is based on Langmuir model 1, 2, and


4 and has a maximum Smax from 15-30 cm, representing the effect of a higher clay percentage. For both profiles the Langmuir parameters have also been recalculated according to Eq. (4), which is the default model used in HYDRUS-1D (Table 7).

Table 7 HYDRUS-1D input parameters for Langmuir sorption isotherms (hard red brown soil). Soil type Soil

depth (cm)

Soil material number

Langmuir parameters (Eq. 3)

Langmuir parameters Eq. (4)

Model name

Smax (µg/g) KL (L/mg) Ƞ (L/mg) Kd (L/kg)

Hard red brown Type-1

0-15 1 1.52 1.5 1.5 2.27 Langmuir 1

15-30 2 1.52 1.5 1.5 2.27 Langmuir 1

30-60 3 21.37 0.18 0.18 3.85 Langmuir 3

60-100 4 21.37 0.18 0.18 3.85 Langmuir 3

100-200 5 21.37 0.18 0.18 3.85 Langmuir 3

Hard red brown Type-2

0-15 1 1.52 1.5 1.5 2.27 Langmuir 1

15-30 2 65.23 0.053 0.053 3.46 Langmuir 4

30-60 3 12.14 0.35 0.35 4.25 Langmuir 2

60-100 4 12.14 0.35 0.35 4.25 Langmuir 2

100-200 5 12.14 0.35 0.35 4.25 Langmuir 2

Fitted parameters for the Freundlich isotherm (Eq. 5) for hard red brown soil are listed in Table 8. The same data groups were used as with the Langmuir isotherms.

Table 8 Parameters of the Freundlich sorption isotherms for boron (hard red brown soil). Soil depth (cm) Soil

description: Clay/calcite

Freundlich parameters Eq. (5) Model name

N (µg/g) KF (L/kg)

0-10 (NAP4)

10-30 (NAP7)



0.40 1.1 Freundlich 1

30-60 (NAP13) 44% clay; 0.1% calcite

0.60 3.1 Freundlich 2

30-60 (NAP7) 33% clay; 1.1 % calcite

0.57 4 Freundlich 3

10-30 (NAP6) 27% clay; 3.2% calcite

0.64 5 Freundlich 4


Finally, also a linear isotherm was assumed with Kd values list in Table 9. These values were derived based on the methodology described in section 4.7.2.3.

Table 9 Parameters of the linear sorption isotherms for boron (hard red brown soil). # number of observations Soil depth (cm) Soil description: Clay/calcite Kd (L/kg) Model name

0-10 (NAP4)

10-30 (NAP7)



1.281 (#6) Linear 1

30-60 (NAP13) 44% clay; 0.1% calcite 3.231 (#3) Linear 2

30-60 (NAP7) 33% clay; 1.1 % calcite 2.490 (#3) Linear 3

10-30 (NAP6) 27% clay; 3.2% calcite 2.614 (#3) Linear 4

The non-equilibrium parameters α and f as used in HYDRUS-1D for the two-site sorption model are listed in Table 9.Two cases are considered, which differ only in the values of the fraction of S1 sites, f. The first method derived f from Smax based on the amount of boron desorbed after 96 hrs; while the second method derived f from Smax based on the hot water extract (Section 4.7.2.2). As Smax values for the second method are larger than those of the first method, its f values are slightly smaller.

Table 10 Two-site sorption model parameters used in Hydrus-1D model. Fraction of S1 sites, f, calculated as f = S1/Smax, where S1 is cumulative desorbed boron at t = 96 hrs (method 1) or desorbed from hot water extract (method 2).

Soil type Soil depth Reaction rate constant α (h-1)

Reaction rate constant α (d-1)

Fraction of S1 sites, f (-)

Hard red brown – method 1 (S1 desorption after 96 hrs)

0-15 cm 0.00426 0.102 0.536

15-30 cm 0.00426 0.102 0.743

30-60 cm 0.0129 0.309 0.727

60-100 cm 0.0129 0.309 0.727

100-200 cm 0.0129 0.309 0.727

Hard red brown – method 2 (desorbed from hot water extract)

0-15 cm 0.00426 0.102 0.318

15-30 cm 0.00426 0.102 0.318

30-60 cm 0.0129 0.309 0.415

60-100 cm 0.0129 0.309 0.415

100-200 cm 0.0129 0.309 0.415


A final step in setting up the boron sorption model is defining the initial boron concentrations in the soil profile. In HYDRUS-1D, both the liquid phase concentration (mass of solute per volume of water) and the adsorbed concentration (mass of solute per mass of soil) can be defined. Internally in the code, the total boron concentration, ST (µg/cm3 soil), is then defined as:

𝑆𝑆𝑇𝑇 = 𝜃𝜃𝐶𝐶 + 𝜌𝜌𝑠𝑠𝑆𝑆 (10)

where ϴ is soil water content (cm3/cm3), C is liquid phase concentration (mg/L or µg/cm3), ρb is solid density (g/cm3), and S is sorbed concentration (µg/g). Liquid phase concentration was measured by extraction at maximum water holding capacity, where the latter is defined as the water content at -5 kPa (McLaughin et al., 1997) (Task 1 Report, Oliver et al., 2019). These water content values were obtained from the water retention measurements (Section 3.2). The S parameter was based on the hot water extraction method, while bulk density was separately measured as part of a comprehensive set of soil physical measurements (Section 3.1). The relevant values for C and S are given in Table 11.

Table 11 Initial boron concentration used in Hydrus-1D model. * based on profile NAP7; ** mean value from Table 17; *** mean value from Table 18; # based on profile NAP7 hot water extraction

Soil type Soil depth (cm)


Initial liquid-phase

concentration (µg/cm3)

Initial sorbed concentration

(µg/g)

Hard red brown – Type 1

0-15 1 0.29** 1.74***

15-30 2 1.3** 12.88***

30-60 3 2.07** 14.25***

60-100 4 2.07** 10.57***

100-200 5 2.07** 11.36***

Hard red brown – Type 2

Soil depth (cm)


Initial liquid-phase

concentration (µg/cm3)*

Initial sorbed concentration

(µg/g)#

0-15 1 0.77 1.50

15-30 2 0.90 14.28

30-60 3 3.8 28.46

60-100 4 3.8 19.24

100-200 5 3.8 20.47

3.2.7 Simulated boron behaviour

Simulation of boron behaviour in soil was undertaken in a similar way as for the salinity and sodicity simulations, i.e. starting with a warming up period to initialise the soil (from 1970-2017), followed by


the 32 year of irrigation with recycled water (2018-2050). During the warming up period, the only boron added to the soil was via the rainwater, with an average boron concentration of 0.05 mg/L (Crosbie et al., 2012). From 2018 onwards, irrigation water with an average boron concentration of 0.33 mg/L is added to the soil. Pasture is as crop, with water requirements as previously estimated in Appendix 4.

The distribution of materials (soil horizons), initial solid phase solute concentration and the depth locations of observations points are shown in Figure 10.

Figure 10 HYDRUS-1D input data: distribution of soil materials (soil horizons 1-5), initial boron concentration on the solid phase (mg/kg), and depth location of observation points for outputting liquid phase concentrations.

The two-site non-equilibrium sorption model is selected with parameter estimated in the previous sections. Sorption on the equilibrium site is described by means of the Langmuir isotherm.

Given the uncertainty around the kinetic mass exchange parameter α, the simulations of boron behaviour following irrigation with recycled water are to be considered primarily as sensitivity analysis. Initial simulations with the derived α resulted in an unrealistic boron behaviour, with most of the adsorbed boron being released in the first few years after the start of the simulations. As mentioned in Section 4.7.3.2.5, the duration of the leaching test was too short to be useful for deriving reaction parameters representative of long-term, slow release processes. Therefore, the α was arbitrarily decreased by a 10-4 and 10-5 to mimic very slow release of boron from the kinetically controlled sorption sites. The boron behaviour is simulated first considering the adsorbed boron concentrations of profile NAP7 (Table 11). Subsequent sensitivity analysis will be carried out to analyse the effect of the initial sorbed boron concentration on the leaching behaviour.

Figure 11 and Figure 12 show boron concentration at five depths in the soil profile, i.e. 10, 30, 60, 100, and 200 cm (based on profile NAP7). When α is 10-4 times smaller than its measured value (Figure 11), boron concentrations are released from the sorption sites into the pore water which explains the initial increase in concentration, especially at the 100 and 200 cm depth. The latter two

Soil materials Initial soluteconcentration

Observation nodes

1

2

3

4

5

200 cm

100 cm

60 cm

30 cm

10 cm1.5014.28

28.46

19.24

20.47


depths have initially the largest adsorbed concentration, hence the release into the pore water is largest. Their concentrations drop towards the end of the warming up period. At the shallowest two depths, concentration remain rather constant during the warming up period. Their initial adsorbed concentration is the lowest, with the release into the pore water being much more modest compared to the deeper depths. Once the irrigation starts, boron concentrations increase throughout the entire profile, as boron is added to profile with each irrigation event. The largest increase is observed at the shallowest depths (10 and 30 cm), up to a factor 3. At the largest depths the increase is less than a factor 2. These calculated concentrations are somewhat lower than the measured pore water concentrations, as discussed in section 4.7.3.2.2 and Figure 3: average concentrations were 5.6 mg/L for the 0-10 cm interval, 1.3 mg/L for the 10-30 cm interval, and 2.07 mg/L for the 30-60 mg/L interval. While there is considerable spatial variability in the chemical parameters and in the boron concentrations, something that was not entirely captured by the sampling and analysis protocol, other reasons contribute to the underestimation of the boron concentrations. These include insufficient data to derive adequate parameters that describe the complex processes of boron adsorption-desorption, including lack of parameters that describe the longer-term exchange processes between solid and liquid phase. The current simulations are therefore preliminary results that need further corroboration, especially to get more representative sorption parameters.

In the second set of calculations the exchange parameter α was further reduced to 10-5 times smaller than its measured value (Figure 12). The boron behaviour is similar to that of the previous simulations, in that the warming period produces now produces nearly constant values until start of the irrigation. As soon as the irrigation commences, boron concentration increases, up to about a factor 4 for the shallowest depths. The highest concentrations for the shallowest depths are again similar to those of the previous case, around 0.4-0.5 mg/L. In other words, while the warming up period is sensitive to the α value, the irrigation period is much less sensitive with boron mainly determined by the amount added and the sorption parameters of the Langmuir isotherm.

As part of the sensitivity analysis we tested the effect of initial sorbed boron concentration on boron leaching. Three scenarios are considered: minimum, mean, and maximum values of adsorbed boron concentration in hard red brown soil as reported in Table 4. Simulations are carried out with the α value set to 10-5 × its base value; this small values was shown to give the most realistic boron behaviour in the warming up period (Figure 12). Simulated boron concentrations at four depths (30, 60, 100, and 200 cm) for the 80-year simulation period are displayed in Figure 13. While the effect of variability in initial sorbed boron is clearly noticeable in the pore-water concentrations, the overall variability in simulated concentrations is rather small. At the shallow depth (30 cm) concentrations range from 0.3 to 0.5 mg/L B, whereas at 200 cm depth the concentrations range from 0.4 – 0.5 mg/L B. Note that the slightly larger variation at the shallow depth is influenced by the temporal variability in the water flux; deeper in the soil profile such variations typically become smaller (e.g. Mallants et al., 2017). Overall the boron concentration as a result of irrigation increases by about a factor of four compared to the warming up period (where small amounts of boron were added from rainfall, i.e. 0.05 mg/L). Note that through irrigation 0.33 mg/L B is added to the soil profile. This accounts for the total increase in B concentration, i.e. from about 0.1 mg/L to about 4.5 mg/L.

The variability in boron concentration in the irrigation water was also tested. Three scenarios are considered, with the following boron concentrations: 0.2 mg/L (minimum), 0.33 mg/L (mean), and 0.53 mg/L B (maximum), as per the statistical parameters from Table 2. As expected, simulated boron concentrations in the soil profile react in a linear way to the increased concentration in the irrigation water. For example, the maximum B concentration (0.53) in irrigation water is about 2.5 times larger than the minimum concentration (0.2 mg/L). As a result, the boron concentration in the soil for the former conditions is also about 2.5 times larger than for the latter, i.e. about 0.7 mg/L versus 0.3 mg/L (Figure 14). This is true for all depths. Interestingly, the variation in B concentration


as a result of variation in irrigation water quality is larger than the variation in B concentration due to variability in adsorbed boron. This illustrates that one of the factors to manage B in soil is through managing the B concentrations in the irrigation water.

In developing sorption models for boron different Langmuir isotherm parameters were derived that resulted in two types of sorption profiles, i.e. Type-1 and Type-2 (Table 7). The sensitivity of boron leaching towards these sorption models was tested by running two scenarios, one with the Type-1 and the other with the Type-2 data. As can be seen in Figure 15, boron leaching is not sensitive to the variation in these parameters (at least not for the variability considered here). Indeed, a nearly identical boron behaviour is observed for both scenarios at all soil depths.

By testing the sensitivity of boron leaching towards several key sources of variability (and thus uncertainty), we demonstrated that the leaching model is least sensitive to the natural soil variability (sorbed boron concentration and sorption models) and most sensitive to the variation in boron concentration in the irrigation water. Considering the mean boron concentration in irrigation water, simulations showed that boron in soil would increase by about a factor of four as a result of long-term irrigation. As the B concentration time series show, a quasi-steady state condition is achieved across all depths illustrating that there does not seem to be a long-term accumulation of B in the soil profile. Over time, an equilibrium is established between the boron added and that leaving the soil profile by drainage.

Figure 11 Boron concentration in pore water of hard red brown soils (Type-2 sorption parameters, Table 7). Two-site non-equilibrium model (Type-2) based on data in Table 11 and Table 10. Initial concentration Type-2 from Table 11. The

α value was set to 10-4 × its base value.

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)

10 cm30 cm60 cm100 cm200 cm


Figure 12 Boron concentration in pore water of hard red brown soils (Type-2 sorption parameters, Table 7). Two-site non-equilibrium model (Type-2) based on data in Table 11 and Table 10. Initial concentration Type-2 from Table 11. The

α value was set to 10-5 × its base value. As can be expected, the final steady-state boron concentration in the soil pore water will also depend on the initial boron concentration prior to adding boron via the irrigation water. As the initial boron concentration (i.e., prior to adding boron containing irrigation water) in the pore water and on the solid phase increases, the effect on the long-term boron concentration will diminish. This means that soils with an already high boron concentration will be at a lesser risk relative to soils with a much lower boron concentration. For instance, initial pore water boron concentrations in the range 0.2-03 mg/L may see an increase of a factor 2 at most. For initial boron concentrations of about 1 mg/L, the increase is not more than 40% (results not shown).

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7Bo

ron

conc

entr

atio

n in

por

e w

ater

(mg/

L)

10 cm30 cm60 cm100 cm200 cm


Figure 13 Sensitivity analysis of boron leaching in hard red brown soil (Type-2 sorption parameters, Table 7). Effects of using different initial sorbed B concentration (minimum, mean, maximum from Table 4). The α value was set to 10-5 × its

base value.

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6Bo

ron

conc

entr

atio

n in

por

e w

ater

(mg/

L)

30 cm T2-4 (mean)30 cm T2-5 (max)30 cm T2-6 (min)

Observation node at 30 cm depth

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)




Figure 14 Sensitivity analysis of boron leaching in hard red brown soil (Type-2 sorption parameters, Table 7). Effects of using different B concentration in irrigation water. Mean = 0.33 mg/L B, Min = 0.2 mg/L B, Max = 0.53 mg/L B. The α

value was set to 10-5 × its base value.

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)

30 cm T2-4 (mean)30 cm T2-7 (min)30 cm T2-8 (max)


0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)




Figure 15 Sensitivity analysis of boron leaching in hard red brown soil. Effects of using different Langmuir sorption isotherm parameters (Type-1 and Type-2 sorption parameters, Table 7). The α value was set to 10-5 × its base value.

0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)

30 cm T2-4 (Type-2 Langmuir)30 cm T2-8 (Type-1 Langmuir)


0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)



0 10 20 30 40 50 60 70 80Time (year)

0

0.1

0.2

0.3

0.4

0.5

0.6

Boro

n co

ncen

trat

ion

in p

ore

wat

er (m

g/L)




4 Conclusion

Modelling long-term boron transport in soil is complicated by a number of factors, including rates of mineral dissolution, adsorption-desorption processes, including whether or not sorption is linear or non-linear, instantaneous or kinetically controlled. Based on best available data on boron in soils from the NAP region, the boron adsorption processes were derived by considering the following conceptual model:

• Boron adsorption/desorption is governed by a two-site model, where boron is distributed across equilibrium sites (S1: sorption is instantaneous) and kinetically controlled sites (S2: sorption is time-dependent). Best-estimate parameter values were derived for the fraction f of equilibrium sites S1 and the kinetic mass exchange parameter α for S2 sites. Parameter f was found to vary between 0.32 (most sites are kinetically controlled, i.e. S2 > S1) and 0.74 (most sites display instantaneous sorption, i.e. S1> S2). The α parameter had values between 0.1 and 0.31 day-1. However, because α was derived from short-term desorption tests (96 hrs), the values are thought not to be representative for calculating long-term boron behaviour in soil. For this reason this parameter was arbitrarily decreased until realistic boron behaviour was simulated. Future work should address this uncertainty by developing longer-term desorption tests that produce kinetic parameters for long-term simulations.

• Boron isotherms were derived as either linear (the Kd model) or non-linear (Langmuir or Freundlich model). The best estimate parameters from the Langmuir model were used in the simulations, as these provide greatest flexibility in describing the complex sorption process.

Several modelling scenarios were undertaken and mainly served as a sensitivity analysis, given the uncertainty around key parameters such as the α parameter. By testing the sensitivity of boron leaching towards several key sources of variability (and thus uncertainty), we demonstrated that the leaching model is least sensitive to the natural soil variability (sorbed boron concentration and sorption models) and most sensitive to the variation in boron concentration in the irrigation water. Considering the mean boron concentration in irrigation water, simulations showed that boron in soil could increase by about a factor of four as a result of long-term irrigation. As the B concentration time series show, a quasi-steady state condition is achieved across all depths illustrating that there does not seem to be a long-term accumulation of B in the soil profile. Over time, an equilibrium is established between the boron added and that leaving the soil profile by drainage.

Interestingly, the variation in B concentration as a result of variation in irrigation water quality is larger than the variation in B concentration due to variability in adsorbed boron. This illustrates that one of the key factors to manage B in soil is through managing the B concentrations in the irrigation water. The current simulations are preliminary results that need further corroboration, especially to get more representative sorption parameters.

Importantly, as the initial boron concentration (i.e., prior to adding boron containing irrigation water) in the pore water and on the solid phase increases, the effect on the long-term boron concentration will diminish. This means that soils with an already high boron concentration will be at a lesser risk relative to soils with a much lower boron concentration.


5 References

Adriano, D. 2001. Trace elements in terrestrial environments: biogeochemistry, bioavailability, and risks of metals. 2nd Edn

Aitken, R.L. and McCalum, L.E. 1988. Boron toxicity in soil solution. Aust. J. Soil Res. 26, 605-610.

Awad, J., Vanderzalm, J., Pezzaniti, D., Olubukoa Esu, O.-O., van Leeuwen, J., 2019. Sustainable Expansion of Irrigated Agriculture and Horticulture in Northern Adelaide Corridor: Source water options/water availability, quality and storage consideration. Goyder Institute for Water Research Technical Report Series No.xx/18, Adelaide, South Australia.

Bethke, C.M., Farrell, B., Yeakel, S. 2019. The Geochemist’s Workbench® Release 12. GWB Reference Manual. Aqueous Solutions, LLC Champaign, Illinois.

Bingham, F. T. 1982. Boron. In 'Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties'. Ed. A. L. Page. pp. 431-47. Am. Soc. Agron.: Madison, Wisc.

Cox, R. F., and Kamprath, E. J. 1972. Micronutrient soil tests. In 'Micronutrients in Agriculture'. Eds J. J. Mordvedt, P. M. Giordano and W. L. Lindsay. pp. 289-317. Soil Sci. Soc. Am.: Madison, Wisc.

Crosbie, R.S., Morrow, D., Cresswell, R.G., Leaney, F.W., Lamontagne, S. and Lefournour, M. 2012. New insights into the chemical and isotopic composition of rainfall across Australia. CSIRO Water for a Healthy Country Flagship, Australia.

Dickson, A.G. 1990. Thermodynamics of the dissociation of boric acid in synthetic seawater from 273.15 to 318.15 K. Deep Sea Research 5: 755-766.

Goldberg, S. and Glaubig, R.A. 1985. Boron adsorption on aluminium and iron-oxide minerals. Soil Science Society of America Journal 49: 1374-1379.

Goldberg, S. and Forster, H.S. 1991. Boron sorption on calcareous soils and reference calcites. Soil Sci. 152: 304–310.

Goldberg S., Forster H.S., and Heick E.L., 1993. Boron adsorption mechanisms on oxides, clay minerals, and soils inferred from ionic strength effects. Soil Sci. Soc. Am. J. 57: 704-708.

Goldberg, S. 1997. Reactions of boron with soils. Plant and Soil 193: 35–48.

Goldberg S., Lesch S.M., and Suarez D.L., 2000. Predicting boron adsorption by soils using soil chemical parameters in the constant capacitance model. Soil Sci. Soc. Am. J. 64: 1356-1363.

Gupta U.C., Jame, Y.M., Campbell, C.A., Leyshon, A.J., and Nicholaichuk, W. 1985. Boron toxicity and deficiency: A review. Canadian Journal of Soil Science 65(3): 381-409.

ISO 11466. 1995. Soil quality: extraction of trace elements soluble in aqua regia. ISO, Geneva, Switzerland.

Keren, R., and Mezuman, U. 1981. Boron adsorption y clay minerals using a phenomenological equation. Clays and Clay Minerals 29: 198-204.


Leyshon, A.J. and Jame, Y-W. 1993. Boron toxicity and irrigation management In: Eds. Gupta U.C. Boron and its role in crop production, pp 207 – 226.

Mallants, D., van Genuchten, M.Th., Šimůnek, J., Jacques, D., Seetharam, S. 2011. Leaching of contaminants to groundwater – Chapter 18. In (F.Swartjens, Editor) Dealing with Contaminated Sites. From Theory towards Practical Application, ISBN 978-90-481-9756-9, Springer Verlag, Dordrecht, The Netherlands, 2011, pp. 787-850. (doi: 10.1007/978-90-481-9757-6_1.)

Marzadori, C., Antisari, L.V., Ciavatta, C., and Sequi, P. 1991. Soil Organic Matter Influence on Adsorption and Desorption of Boron. Soil Sci. Soc. Am. J. 55:1582-1585.

McLaughlin, M. J., Tiller, K. G. and Smart, M. K. 1997. Speciation of cadmium in the soil solution of saline/sodic soils and relationship with cadmium concentrations in potato tubers (Solanum tubersum L.). Aust. J. Soil Res. 35: 183-98.

Minasny, B., Mcbratney, A.B., Brough, D.M., and Jacquier, D. 2011. Models relating soil pH measurements in water and calcium chloride that incorporate electrolyte concentration. European Journal of Soil Science, 10.1111/j.1365-2389.2011.01386.x.

OECD/OCDE. 2000. OECD Text No. 106: Adsorption – Desorption using a batch equilibrium method. http://www.oecd-ilibrary.org/environment/test-no-106-adsorption-desorption-using-a-batch-equilibrium-method_9789264069602-en.

Offiah, O. O., and J. H. Axley. 1993. Soil testing for boron on acid soils. In Boron and Its Role in Crop Production. U.C. Gupta (ed.). CRC Press, Boca Raton, FL, pp. 105–123.

Oliver, D.P., Fruzangohar, M., Johnston, C., Ouzman, J. and Barry, K. 2018, Sustainable expansion of irrigated agriculture and horticulture in northern Adelaide corridor. Task 1: Development and optimisation of modelling domain and impact assessment of irrigation expansion on the receiving environment, Goyder Institute for Water Research Technical Report Series No. 18/xx, Adelaide, South Australia.

Parkhurst, D.L., and Appelo, C.A.J., 2013. Description of input and examples for PHREEQC version 3—A computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations: U.S. Geological Survey Techniques and Methods, book 6, chap. A43, 497 p., available only at https://pubs.usgs.gov/tm/06/a43/.

Pavlatou, A., Polyzopoulos, N.A., 1988. The role of diffusion in the kinetics of phosphate desorption: The relevance of the Elovich equation. J. Soil Sci. 39: 425-436.

Selim, H.M., Davidson, J.M., and RAO, P.S.C. 1977. Transport of reactive solutes through multilayered soils. Soil Sci. Soc. Am. J. 41: 1-10.

Shorrocks, V. 1997. The occurrence and correction of boron deficiency. Plant and Soil 193, 121-148.

Šimůnek, J., van Genuchten, M.Th., and Šejna, M., 2008. Development and applications of the HYDRUS and STANMOD software packages and related codes. Vadose Zone Journal, Special Issue “Vadose Zone Modeling” 7, 587-600. doi:10.2136/VZJ2007.0077.

US-EPA 3050B. 1996. Acid digestion of sediments, sludges and soils. Environmental Protection Agency, Washington, DC.

http://www.oecd-ilibrary.org/environment/test-no-106-adsorption-desorption-using-a-batch-equilibrium-method_9789264069602-en

http://www.oecd-ilibrary.org/environment/test-no-106-adsorption-desorption-using-a-batch-equilibrium-method_9789264069602-en

https://pubs.usgs.gov/tm/06/a43/


van Genuchten, M. Th., and Wagenet, R.J. 1989.Two-site/two-region models for pesticide transport and degradation: Theoretical development and analytical solutions. Soil Sci. Soc. Am. J. 53: 1303-1310.

Zhang, H. and Selim, M. 2005. Kinetics of Arsenate Adsorption-Desorption in Soils. Env. Sci. Technol., 39, 6101-6108.

Appendix 12 Climate extremes and their impact on crop growth | 410

Appendix 12: Climate extremes and their impacts on

crop growth and irrigation requirement

Contents

Executive summary .............................................................................................................................. 411

1. Introduction ...................................................................................................................................... 413

2. Literature review .............................................................................................................................. 414

3. Northern Adelaide Plains (NAP) study area .................................................................................... 417

4. Selection of climate indices ............................................................................................................. 418

5. Application of climate indices to the NAP region ........................................................................... 419

5.1. Daily temperature, number of hot days and hot spells ............................................................... 419

5.2. Number of dry days and dry spells ............................................................................................... 421

5.3. Number of cold days ..................................................................................................................... 423

6. Crop specific impacts and mitigations ............................................................................................. 425

6.1. Horticultural crops (potatoes, carrots, onions) ............................................................................ 425

6.2. Perennial horticulture (vines, almonds, pistachios) .................................................................... 427

6.3. Broad acre crops (lucerne, pasture) ............................................................................................. 433

7. Irrigation requirement estimated with FAO-56 .............................................................................. 435

8. Climate change effect on plant water stress ................................................................................... 442

9. Conclusions ....................................................................................................................................... 448

References ............................................................................................................................................ 450


Executive summary

Agricultural industries such as viticulture, perennial and annual horticultures can be impacted by

climate extremes. In this study, the effect of climate extremes on crop growth and irrigation

requirement in the Northern Adelaide Plains (NAP), South Australia (SA), was investigated. The

climate indices for historic (1985-2017) and future (2018-2050) climate were analyzed to

determine the frequency of extreme climate events, and its impact on the growth of horticultural

crops (potatoes, carrots, and onions), fruit trees (vines, almonds, and pistachios), and broad acre

crops (lucerne and pasture).

Using greater daily temperature indices and frequency of hot days relative to historic climate,

results showed that the NAP region will be subjected to significant warming in the near future.

There will be also a higher frequency for dry days (with rainfall less than 1 mm) and dry spells

compared with historic data. The NAP region, following global climate change predictions, will

be subjected to milder winters with smaller frequencies of frost and chill days.

The following consequences are anticipated for the crops in the NAP region, as the result of these

climatic shifts:

• Potatoes will likely see a yield decrease and a higher risk to being invaded by pests;

• The growth and yield in carrots will be stimulated due to increased frequency of hot and

dry days. However, extreme heat events may reduce the quality of carrots and mid-season

drought stress can depress the yield in carrots.

• Warmer climate may reduce the duration of crop growth, yield, and seed production for

onions; whereas, the low rainfall condition can reduce the risk of infection by pests (i.e.,

leaf blight).

• The projected drought and extremely hot weather in the NAP region can negatively impact

vines, which may result in poor budburst, leaf loss, bunch damage, and consequently low

yield and production or even crop loss. Warmer summers may lead to adopting warmer

climate grape varieties.

• For almonds and pistachios, it is expected that their yield and production may be impacted

by drought in the NAP region if not properly managed. While the current NAP climate

accommodates chill requirements for these fruits, the projected climate shows there would


be some years that this requirement cannot be met at all, which presents challenges to

growers with changes in management practices and perhaps even new cultivars needed to

accommodate new growing conditions.

• As the NAP region will be subjected to an increased number of hot days, a decline in

pasture production is anticipated.

In order to understand the effect of weather extremes on irrigation practices, the irrigation

requirement for abovementioned crops was calculated using the FAO-56 dual crop coefficient

method (this report, Annex 7). The irrigation requirements for these crops were compared under

historic and future climate scenarios to provide insights of future climate impact. Results showed

that crops will require a higher amount of annual irrigation under future climate, depending on the

soil textures and crop stress tolerance. Regardless of the crop type, sand over clay soils require the

highest irrigation (714 – 956 mm) while deep uniform to gradational soils need the lowest

irrigation (643 – 910 mm). It was concluded that annual horticultural crops could face more

irrigation related risks in the future climate compared to deep rooted perennial horticultural crops.

The results of water balance simulation showed that under future climate, pastures on deep uniform

to gradational soils will experience 243 extra water stress days (over a 32 year period) compared

to historic climate, or 7.6 days per year. Over the 32 years period, hard red brown soils will

experience 105 extra water stress days or 3.3 per year under future climate. This implies that

pasture’s yield and production will be reduced if not irrigated.


1. Introduction

The observed global climate trends show an increase in air temperature, decrease in oceanic

waters’ pH, and sea level rise over the past three decades, providing strong evidence of a global

climate change (Jones and Moberg, 2003; Church and White, 2006, 2011; Steffen, 2006; Fisher,

2007; Vose et al., 2012; IPCC, 2013; Marcott et al. 2013; Allen et al., 2018). The consequences

associated with climate change include, but are not limited to, hotter and drier summers, milder

winters, reduced rainfall events, water resources’ scarcity, increased risk of flooding and coastal

erosion due to sea level rise, risk to marine calcifying organisms such as oyster, mussel, coral and

plankton varieties as a result of decrease in oceanic waters’ pH, decrease in food security due to

impacts to agricultural industries, as well as increase in frequency of extreme weather events such

as strong winds, lightning, hail storms, drought and heat waves (Fisher, 2007; Australian Bureau

of Meteorology, 2008, 2009a, 2009b, 2010, 2011; Wheeler and von Braun, 2013, Thomas et al.,

2016).

Agricultural industries such as viticulture, perennial and annual horticultures, and annual cropping

are key industries which can be impacted by climate extremes (White et al., 2006; Thomas et al.,

2016; Ziska et al., 2016). Extreme climate conditions such as drought, heatwaves, extreme heat

and cold (or frost), expose a great risk to wine grape production (White et al., 2006; Shellie and

Glenn, 2008; Wittwer, 2008; Australian Bureau of Meteorology, 2008, 2009a, 2009b, 2011; Jones

et al., 2010; Crimp et al., 2012, 2015; Araujo et al., 2016), and crop plants’ growth and yield

(Garrett et al., 2006; McMichael et al., 2007; Lobell and Field, 2007; Thornton, 2012; Rehman et

al., 2015; Zhu and Troy, 2018). To mitigate the vulnerability to climate change, it is required to

understand the impacts of near-term climate change on the supply of key food commodities

through analyzing past climate trends (Lobell et al., 2011). Through the proper planning,

management, and irrigation practices, negative impacts of climate extremes on the crop growth

and production can be minimized efficiently. Moreover, identifying which particular crops and

regions have been most impacted by extreme climate conditions would assist us to measure and

analyze their impacts and adapt accordingly (Lobell et al., 2011). The expanded rainwater

harvesting, water storage and conservation, efficient irrigation, adjusting planting dates and crop

variety, and relocating crops (IPCC, 2013; Haverkort and Verhagen, 2008) are among the

adaptation techniques which can be used to minimize these negative impacts.


This research was aimed to investigate the effect of climate extremes on crop growth and irrigation

requirement in the Northern Adelaide Plains (NAP), South Australia (SA). The climate indices for

historic (1985-2017) and future (2018-2050) climate were analyzed to determine the frequency of

extreme climate events and its impact on the growth of horticultural crops (potato, carrot, and

onion), fruit trees (vines, almonds, and pistachio), and broad acre crops (lucerne and pasture). First,

the irrigation requirement was calculated by means of the FAO-56 dual crop coefficient method

for the abovementioned crops to provide insights of future climate on irrigation requirement and

impacts of weather extremes on these crops. Next, the soil water availability was calculated with

the mechanistic soil water balance simulator HYDRUS-1D for the reference crop pasture. For the

calculation and comparison of climate statistics for historic and future climate, two time series of

identical length (i.e. 32 years) were used.

2. Literature review

Due to the importance of agricultural crops, a broad range of studies has been conducted to

investigate the impact of climate change on crop growth and yield. In general, these studies have

emphasized on negative impacts of weather extremes (e.g., high temperature, low rainfall) on the

production of crops, especially during the growing period (Reilly et al., 2002; Fischer et al., 2005;

Lobell and Field, 2007; Thornton, 2012; Haile et al., 2017). However, climate change can lead to

changes not only in crop yields (Adams et al. 1990; Reilly et al. 2002; Peng et al., 2004; Deschenes

and Greenstone, 2007) but also in crop yield variability (Chen et al., 2004; McCarl et al., 2008).

The negative consequences associated with high temperatures include reduced critical growth

periods of crops (Haile et al., 2017), more frequent crop diseases (Coakley et al., 1999; Garrett et

al., 2006), and increased vulnerability of crops to insect pests (CCSP, 2008; Jones and Yosef,

2015; Haile et al., 2017), damage to reproductive tissues of plants, and increased pollen sterility

(Roberts and Schlenker, 2009; Thornton, 2012). Drought (low rainfall) or excessive rainfall not

only affect the crop production, but also yield and acreage effects (McCarl et al., 2008; Haile et

al., 2017).

Previous studies (Rosenzweig and Parry, 1994; Lobell and Field, 2007, 2011; Haile et al., 2017)

have been mostly concentrated on the impact of the climate change on the world’s key staple crops


(e.g., wheat, rice, maize, soybeans, barley and sorghum), due to their importance as a major source

of food around the world. The first global assessment of the potential climate change impacts on

staple crops was conducted by Rosenzweig and Parry (1994). Using numerical models for wheat,

rice, maize, and soybeans, they anticipated positive yield changes for countries that are positioned

in middle and high latitudes and negative yield changes for countries that are located in low

latitudes as a result of climate change. Lobell and Field (2007) used empirical/statistical models to

understand the relationship between climate and crop yield at a global scale, and concluded that

recent climate trends have had a negative impact on the global production of major crops such as

wheat, rice, maize, soybean, barley, and sorghum. A regression model was used by Lobell and

Field (2011) to understand the effect of climate on yield outcomes for staple crops. They concluded

that, in most situations, crop yields are more sensitive to temperature extremes than precipitation

extremes. Their study, in agreement with the previous study by Rosenzweig and Parry (1994),

showed that a 1°C rise tended to decrease yields up to 10% except in high latitude countries, where

in particular rice gains from warming. Increases in precipitation, on the other hand, increases yields

for nearly all crops and countries (Rosenzweig and Parry, 1994; Lobell and Field, 2011). More

recently, Haile et al. (2017) developed an empirical model to investigate the impact of climate

change and weather extremes on crop production, and predicted that climate change could reduce

the production of staple crops by 9% in the 2030s and by 23% in the 2050s, globally. Their study

found that increasing mean growing season temperature may not pose a major problem to crop

production. Unlike mean temperature, increasing temperatures at the two extremes (e.g., higher

minimum temperature for rice and higher maximum temperature for maize) does negatively

impact the crop production. A similar situation exists for wheat and soybeans, where a higher

average temperature becomes problematic after a certain critical level. A drawback of all these

models is that they fail to simulate future yield responses when cropping areas shift, as evidenced

by the recent expansion of soybean area in Brazil (FAO, 2006), or when the range of future

temperatures exceeds those for which the models were calibrated (Lobell and Field, 2007).

In comparison, fewer studies have been conducted to investigate the impact of climate extremes

on the growth of horticultural crops (potato, carrot, and onion), fruit trees (vines, almonds, and

pistachio), and broad acre crops (lucerne and pasture). Reidsma et al. (2015), using an integrated

assessment (Rotmans and Van Asselt, 1998; Van Ittersum et al., 2008) and farming systems


analysis (Janssen and van Ittersum, 2007), showed that while impacts of climate change on potato

yields at a farm level (Dutch province Flevoland) are positive, the increased frequency of extreme

events (e.g., heat waves) can turn positive impacts into negative ones.

However, adaptation techniques such as drip irrigation and planting in wider ridges can reduce

negative impacts. Furthermore, recent studies (e.g., Pulatov et al., 2015) suggested that climate

change may shift the planting/harvesting time for potato, depending on the geographic location of

the country. For example, planting of potato at an earlier date could be possible in the warm climate

regions. This is desirable as it can reduce the risk of summer drought (Supit et al., 2012), attacks

by Colorado potato beetle (Jönsson et al., 2013; Pulatov et al., 2014), and infestation by late blight

(Haverkort and Verhagen, 2008; Wiik, 2014). However, in cold climate regions such as north

Europe, earlier potato harvesting may expose a high risk of frost damage (Haverkort, 1990; Pulatov

et al., 2015).

For determinate crops such as onions, warming can reduce the duration of crop growth and hence

yield (Idso, 1990; Wheeler et al., 1994, 1996; Harrison et al., 1995; Daymond et al., 1997; Wurr

et al., 1998), while it may stimulate growth and yield in indeterminate crops such as carrots. For

onions, warmer future climate scenarios may lead to yield decrease and cooler future climate

scenarios may result in yield increase (Harrison et al., 1995; Maracchi et al., 2005). The situation

is reverse for carrots that show an increased yield in warmer climate conditions and a decreased

yield in cooler conditions (Wurr et al., 1998; Lal et al., 2018).

Extreme climate conditions can mean significant concerns for vines, if management strategies are

not employed. Previous studies (Hayman et al., 2012) showed that vines flowering during extreme

events are poor in fruit production and consequently low in yields. Serious negative consequences

have been reported from vineyards in South Australia and Victoria as a result of heatwaves,

including poor budburst, leaf loss, bunch damage, and even crop losses (The Australian Wine

Research Institute, 2018). A simulation study by Araujo et al. (2016) suggested that grape yields

are more sensitive to spring and summer droughts in the absence of irrigation. There are also

additional studies showing that warmer and drier weather conditions may shift the produced wine

grapes to warmer climate grape varieties (White et al., 2006; Thomas et al., 2016). Indeed,


heatwaves can result in sunburnt fruit along with rapid sugar level increases. Higher than expected

sugar levels will lead to higher than desirable alcohol levels which can impact on yeast’s ability to

complete fermentation, therefore affect wine style and potentially lead to loss of wine quality.

Also, increased temperatures and periods of hot dry weather will lead to faster ripening and shifting

forward harvest dates (The Australian Wine Research Institute, 2015).

Climate change is likely to also affect chilling and plant dormancy, which are vital for fruit trees

including almonds and pistachios (Schwartz, 1999; Baldocchi and Wong, 2008; Luedeling and

Brown, 2011; Luedeling et al., 2011; Luedeling, 2012). Previous research has shown dramatic

losses in number of chilling hours for warm growing regions such as South Australia (Luedeling,

2012).

Soil water availability, water flux, root water uptake, and evapotranspiration (evaporation and

transpiration) are intimately coupled processes influencing crop growth and yield. Previous studies

have generally not taken into account the effect of climate extremes on these processes. Therefore,

the present work attempts to fill this gap in knowledge and will first assess the impact of climate

change on irrigation requirement for horticultural crops (potato, carrot, and onion), fruit trees

(vines, almonds, and pistachio), and broad acre crops (lucerne and pasture) in the NAP region.

Next, the soil water availability is compared between historic and future climate and conclusions

are drawn about consequences for crop growth.

3. Northern Adelaide Plains (NAP) study area

The Northern Adelaide Plains (NAP) represents a highly diverse region with active viticulture,

horticulture, annual cropping, livestock, and dairy industries (Thomas et al., 2016). This region

benefits from a variety of fertile soils, including well-drained sandy soils and heavier clay soils;

high quality groundwater is supplied by confined sand and limestone aquifers (Pitt et al., 2013).

In addition, the region has been recently (since 1999) equipped with the Bolivar wastewater

treatment plant which provides tertiary-treated wastewater via a pipeline network as a

supplementary source of water for crop irrigation (Pitt et al., 2013; The Goyder Institute for Water

Research, 2016; The Australian Water Association, 2019). These features along with


Mediterranean-type climate (with hot, dry summers and mild, wet winters) make the NAP a

suitable region for cultivation of horticultural crops, fruits, and broad acre crops. However, the

NAP region, similar to many regions in the world, is vulnerable to climate change and proper

management and irrigation practices are required to mitigate the negative impacts of climate

change on crop production.

4. Selection of climate indices

A total of ten climate indices were considered in this study and include i) mean of daily average

temperature, ii) mean of daily maximum temperature, iii) mean of daily minimum temperature, iv)

mean difference between the daily maximum and minimum temperature, v) the number of days

when daily maximum temperature exceeds 35°C (extremely hot days), vi) the number of hot

spell1s, vii) the number of days when daily precipitation drops below 1 mm (dry days), viii) the

number of dry spells2, ix) the number of days when daily minimum temperature goes below 0°C

(extremely cold or frost days), and x) the number of days when daily minimum temperature in

winter is below 7.2°C (chilling days). These indices were calculated using climate parameters such

as temperature, precipitation, and humidity for historic (1985-2017) and future (2018-2050)

climate data. Note that to have the same number of data points in historic and future time series,

both time series were limited to 32 years each.

The climate data used is based on downscaled climate change projections for South Australia

(Charles and Fu, 2014; The Goyder Institute for Water Research, 2019a), generated from a subset

of global climate models (GCMs). These models were chosen because they were found to perform

better at representing climate drivers that are particularly influential on rainfall in South Australia,

such as the El Nino Southern Oscillation (ENSO) and Indian Ocean Dipole. The projected climate

data are the local scale climate projections, which includes rainfall, temperature, solar radiation,

vapor pressure deficit and evapotranspiration data for South Australia (The Goyder Institute for

Water Research, 2019b).

1 Three to five consecutive days above 35 °C; six or more days above 35 °C are counted as two hot spells. 2 Three or more consecutive days with less than 1 mm of rain

http://www.goyderinstitute.org/news/2019/international-watershed-management-research-centre-and-joint-research-initiatives-emerge-from-south-australian-water-mission-to-china/




5. Application of climate indices to the NAP region

In this section, climate indices were analyzed to determine the frequency of extreme climate events

and their spells for historic and future climate in the Northern Adelaide Plains (NAP), SA.

5.1. Daily temperature, number of hot days and hot spells

In order to compare the historic and future temperature regimes, daily temperature indices

including annual mean of daily average temperature, annual mean of daily maximum temperature,

annual mean of daily minimum temperature, and annual mean difference between the daily

maximum and minimum temperature (Table 1) were calculated for historic and future climate.

Future temperature indices are all greater than historic ones, providing an evidence that the NAP

region will be subjected to warming in the near future.

Previous studies (e.g., IPCC, 2018) have already warned about negative consequences, potential

impacts and associated risks of warming, which include increases in mean temperature in most

land and ocean regions, hot extremes in most inhabited regions, heavy precipitation in several

regions, and the probability of drought and precipitation deficits in some regions (IPCC, 2018). It

is anticipated that the NAP region will experience some of these negative consequences, if not all,

as it will be subjected to warming in the future.

Table 1 Daily temperature indices for historic (1985-2017) and future (2018-2050) climate. a, b: mean values are significantly different at the 95% level based on t-test.

1985-2017

2018-2050

1985-2017 2018-2050 1985-2017 2018-2050 1985-2017 2018-2050

Annual mean of daily average

temperature (°C)

Annual mean of daily maximum temperature

(°C)

Annual mean of daily minimum temperature

(°C)

Annual mean difference between the daily

maximum and minimum temperature

(°C) 16.57a 17.29b 22.18a

23.08b 10.97a

11.50b

11.21a

11.58b

Figure 1 shows the number of days with maximum daily temperature at or above 35°C (hot day)

for historic and future climate. For only one statistic (95th percentile) does the historic climate

produce larger number of days compared to the future climate. This is because in the year 2004

there was a record number of hot days (95th percentile = 29) that exceeded those of any future year


(95th percentile = 21). In comparison with historic climate, future climate showed a significantly

greater number of days (449 versus 379) with maximum temperature at or above 35°C (Table 1).

Figure 1 Number of days with a maximum temperature above 35°C for historic (1985-2017) and future (2018-2050) climate.

In addition to single days above the threshold, hot spells (3 or more consecutive days > 35°C) were

also calculated. Although spells at or above 35°C for future data (21 spells) is slightly less than

that of the historic data (24 spells), there is now also the possibility of hot spells occurring at or

above 40°C. The cumulative distribution function of the number of days in any year when

temperature exceeds 35°C (Figure 2) shows that there is approximately an equal shift towards a

greater number of days > 35°C across the full distribution function (for nearly all percentiles). As

mentioned above, an exception exists at the 95th and higher percentile because of an exceptionally

hot 2004.

35 40 45 50Maximum temperature (°C)

0

5

10

15

20

25

Num

ber o

f day

s

5th percentile - historic climateMedian - historic climate95th percentile - historic climate5th percentile - future climateMedian - future climate95th percentile - future climate


Table 1 Number of hot days and hot spells for historic (1985-2017) and future (2018-2050) climate.

1985-2017 2018-2050 1985-2017 2018-2050 Days at or above 35°C Spells at or above 35°C 379 449 24 21

Days at or above 40°C Spells at or above 40°C 46 68 0 1

Figure 2 Cumulative distribution function for number of days in a year when temperature exceeds 35 °C.

5.2. Number of dry days and dry spells

A dry day is defined when daily rainfall is less than 1 mm. To determine the frequency of dry days,

the cumulative distribution function (CDF) was plotted (Figure 3) for both historic and future

climate data. Figure 3 indicates a greater occurrence of dry days in the future compared with the

historic data, with generally an equal shift towards more dry days across all percentiles. For

example, on average (50th percentile) the number of dry days increases from 290 to 296. This is

an increase by about 2.1 %. Additionally, spells of three or more dry days (dry spells) with rainfall

0 10 20 30Number of days with T >= 35

0

20

40

60

80

100

Cum

ulat

ive

dist

ribut

ion

func

tion

(%)

Historic climateFuture climate


less than 1 mm for historic and future climate were also calculated (Refer to Table 2 and 4). On

average (50th percentile) the number of dry spells increases from 81 to 86. This is an increase by

about 6.2 %.

Figure 3 Cumulative distribution function for number of days (per year) with a rainfall less than 1 mm for historic (1985-2017) and future (2018-2050) climate data.

Table 2 Total number of dry days and dry spells for historic (1985-2017) and future (2018-2050) climate.

1985-2017 2018-2050 1985-2017 2018-2050

Dry days Dry spells

9603 9805 2681 2808

260 280 300 320Number of days with a rainfall < 1

0

20

40

60

80

100

Cum

ulat

ive

dist

ribut

ion

func

tion

(%)

Historic climate Future climate


Figure 4 Cumulative distribution function for number of spells (3 days or more with rain < 1 mm) per year with for historic (1985-2017) and future (2018-2050) climate data.

5.3. Number of cold days

A frost day is defined when daily minimum temperature goes below 0°C. Based on the CDF of

days with temperature less than 0°C for both historic and future climate data, Figure 5 shows that

the number of frost days in the future will be less than the historic data, confirming that the NAP

region, following a global climate change, will be subjected to milder winters. It should be noted,

however, that the dryer climate could increase the severity of frost despite their reduced frequency.

This could have serious consequences for crops whose phenology has been advanced or

compressed due to the milder winters/springs. As an example, changes to almond phenological

timing could impact the synchronization of pollinator varieties with main commercial varieties.

60 70 80 90 100Number of spells with rainfall < 1

0

20

40

60

80

100

Cum

ulat

ive

dist

ribut

ion

func

tion

(%)

Historic climateFuture climate


Figure 5 Cumulative distribution function for annual number of frost days (temperature less than 0°C) for historic (1985-2017) and future (2018-2050) climate data.

Fruit trees such as almonds and pistachios require a minimum period of cold winter weather after

which a fruit-bearing tree will blossom. The chilling requirement can be calculated as chilling

hours (or chilling days), which is the sum of total amount of time in a winter spent at certain

temperatures (Lockwood and Coston, 2005; Texas A&M University Agrilife Research &

Extension, 2019). The adequate amount of winter chilling results in homogeneous and

simultaneous flowering (Luedeling et al., 2009a; PGAI, 2019). In SA, almonds (Pitt et al., 2013)

require a winter chilling period of 400–900 hours (≈ 16.5–37.5 days) and pistachios require winter

a chilling period of 600–1050 hours (≈ 25–43.75 days) below 7.2 °C (Küden et al., 1994) to initiate

flowering. Figure 6 shows that the annual number of chilling days in the future will decrease on

average by 4 days (or about 7.7 %). Table 3 shows that the total number of frost days is only half

that of the future climate, while the total number of chilling days decreased by almost 150 days.

0 1 2 3 4Number of frost days

0

20

40

60

80

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%

) Historic climateFuture climate

https://en.wikipedia.org/wiki/Weather

https://en.wikipedia.org/wiki/Fruit-bearing_tree

https://en.wikipedia.org/wiki/Blossom


Figure 6 Cumulative distribution function for number of annual chilling winter days (with a temperature less than 7.2°C) for historic (1985-2017) and future (2018-2050) climate data.

Table 3 Total number of frost and chilling days for historic (1985-2017) and future (2018-2050) climate.

1985-2017 2018-2050 Frost days

20 9

Chilling days 1749 1607

6. Crop specific impacts and mitigations

6.1. Horticultural crops (potatoes, carrots, onions)

Potato is considered as the fourth most important food crop after rice, wheat and maize (Pulatov

et al., 2015). In SA, approximately 11,900 ha of land has been allocated for potato growth

providing an average annual production of over 300,000 tonnes (ABS, 2012; Pitt et al., 2013). The

NAP region contributes 14% to SA potato industry (Pitt et al., 2013), which highlights its critical

role in supplying this crop.

30 40 50 60 70Number of chilling days

0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%

) Historic climate

Future climate


Potatoes can be grown in many months of the year, provided it has 60-90 days of frost-free

conditions to be successfully harvested. The best planting times is March-April, as the soil is warm,

growth is rapid and there are generally less pests. The NAP region has two main harvest periods,

i.e. from June to August and from November to February.

With an increased frequency of hot and dry days (and dry spells) in the future (Figure 1 and Figure

2), negative impacts are anticipated for potato including a yield decrease and a higher risk to being

invaded by pests such as Colorado potato beetle and late blight which can reduce the quality of

produced potatoes. Research (Haverkort & Verhagen, 2008) has shown that potatoes’ growth

slows down significantly when the daily average temperature falls below 5 °C or rises above 21

°C, and their photosynthetic capacity halts completely at temperatures below 2 °C and above 35

°C. Projected temperature patterns (Figure 1) in the NAP region show that potatoes may suffer

from lack of photosynthetic capacity (and hence low yield) due to frequent hot days with

temperatures above 35 °C, which mainly occurs throughout January and February. In the NAP

region, the potatoes that are harvested from November to February, are most at risk and may

require earlier harvest to avoid the effects of hot days.

In order to mitigate the negative impacts of the shortage in rainfall, more efficient irrigation

techniques, such as drip irrigation and planting in wider ridges can be used. To reduce the negative

impacts associated with high temperatures, potatoes can be planted/harvested at an earlier date, as

suggested by recent studies (Haverkort and Verhagen, 2008; Supit et al., 2012; Wiik, 2014;

Jönsson et al., 2013; Pulatov et al., 2014, 2015).

More than 30,000 tonnes of carrots are produced in NAP region yearly, which makes the carrot

one of the most economically important crops in the region (ABS, 2012; Pitt et al., 2013).

Packaging and processing facilities located in the Adelaide Plains process nearly half of the SA

carrots (Pitt et al., 2013). Indeterminate crops such as carrots are less sensitive to heat stress

compared to determinate crops such as onions. From the projected future temperature (Figure 1),

it is expected that warmer temperature may stimulate the growth and yield in carrots (Harrison et

al., 1995; Wheeler et al., 1996; Wurr et al., 1998; Olesen and Bindi, 2002; Maracchi et al., 2005).


However, extreme heat events may reduce the quality of carrots as a result of acrid flavor

development (Simon, 1985) and textural deterioration (Pitt et al., 2013). Warmer climate may

expose a further disadvantage for carrots, as it then becomes their tendency to invest energy into

foliage rather than tuber growth (Wurr et al., 1998). Carrots do not need irrigation during early

growth, and dry soil conditions during this stage even increases the yield (Dragland, 1978).

However, mid-season drought stress can depress the yield in carrots. Dry soil conditions during

early growth may also influence their chemical composition (Dragland, 1978; Sørensen et al.,

1997) and increase the risk of infections by common scab (Schoneveld, 1991). Advanced

management and irrigation techniques are required to handle negative consequences of heat and

drought stresses for vegetable crops such as carrots, grown in broad acre areas. Use of center pivot

(instead of fixed overhead sprinkler) is an example of an efficient method, which has been recently

used for irrigation of carrots. This method successfully reduced the disease buildup, and enhanced

the end-product quality (Government of South Australia, 2006).

In the NAP region, more than 13,000 tonnes of onions are produced annually which are planted

on 206 ha (ABS, 2012). Onion is considered a cool season vegetable crop, which can be more

adversely affected by temperature extremes than some warm season crops (McKoewn et al., 2004;

Hazara and Som, 2009; Lal et al., 2018). It is anticipated that warmer climate reduces the duration

of crop growth, yield, and seed production for onions (Idso, 1990; Wheeler et al., 1994, 1996;

Harrison et al., 1995; Daymond et al., 1997; Wurr et al., 1998; Maracchi et al., 2005; Lal et al.,

2018). It should be noted that low rainfall (or dry) condition can reduce the risk of infection by

leaf blight (Msuya et al., 2005), which is a common disease in onions. To alleviate the negative

impacts of the climate extremes on onions, it might be a useful strategy to cultivate a variety of

onions that is more tolerant to weather extremes. For example, the dry onion (Allium cepa L.)

might be a good choice as it can be grown in a wide range of climatic conditions from relatively

hot and dry areas to fairly cool and humid zones (Tindall, 1983; Messiaen, 1992).

6.2. Perennial horticulture (vines, almonds, pistachios)

In 2018, approximately 3000 tonnes of wine grapes were produced in the NAP region (Wine

Australia, 2018) planted on about 600 ha (ABS, 2012). A large percentage of the produced wine

grapes are red variety comprising largely Shiraz (Wine Australia, 2018). The projected extreme


climate conditions such as drought and extremely hot weather in the NAP region (Figure 1 and

Figure 2) can negatively impact vines. This may include poor budburst, leaf loss, bunch damage,

and consequently low yield and production or even crop loss (Hayman et al., 2012; Araujo et al.,

2016; The Australian Wine Research Institute, 2018). In addition, temperatures greater than 35°C

can reduce the amount of skin anthocyanin, the principal component in grapes responsible for wine

color (Spayd et al., 2002; Shellie and Glenn, 2008). Warmer summers may lead to adopting

warmer climate grape varieties (e.g. Marsanne or Viognier from Southern France, Tempranillo

from Spain or Fiano from Southern Italy) to avoid poorer quality wines (White et al., 2006; Thomas

et al., 2016), as they shift the grape ripening from milder fall and late summer to hotter mid-

summer.

Efficient management and irrigation strategies should be employed to mitigate negative

consequences that arise from extreme climate conditions. Alternative irrigation methods such as

deficit irrigation can be used to manage vegetative and reproductive growth of wine grapes (and

other perennial fruit crops), and enhance the product’s quality or to increase water use efficiency

(Shellie and Glenn, 2008). The application of exogenous compounds such as kaolin particle film

(Al2Si2O5(OH)4) to grapevines is suggested during abiotic stress periods, as it results in lower

canopy temperatures and stomatal conductance, higher protection of photosystem structure and

function in leaves exposed to excessive heat and solar radiation (Glenn et al., 2010; Dinis et al.,

2016a, b, 2017; Conde et al., 2018). The other advantages of applying kaolin particle film is that

it can promote total soluble solids and anthocyanin amount in grape berries, improve the molecular

mechanisms involved in phenolics’ synthesis (Shellie and Glenn, 2008; Ou et al., 2010; Song et

al., 2012; Shellie, 2015; Conde et al., 2016, 2018), enhance the sucrose synthesis and its transport

capacity in leaves when the temperature exceeds 30 °C (Conde et al., 2018), and substantially

improve the grape leaf metabolome that can tolerate the constraints of excessive summer stress

(Conde et al., 2018) such as water deficit and high temperature with a better capacity.

Australia is considered the second major producer of the almonds (7%) in the world (ABA, 2017)

after the United States (80%). Almonds are commercially cultivated in three states (Victoria, South

Australia and New South Wales) of Australia (The Almond Board of Australia, 2012). They are

best suited to areas with cool to cold winters and hot, dry summers, such as southern Australia. In

the NAP region 724 ha of land has been used for almond cultivation (ABA, 2017), accounting for


2% of the Australian almond industry (Pitt et al., 2013). Although almond is considered as a

drought resistant species (De Herralde et al., 2003; Rouhi et al., 2007) which is able to withstand

frequent periods of low soil moisture accompanied by high evaporation rate and temperature

during the growing season, its production and survival can be affected by the occurrence of cyclic

and severe droughts (Ghrab et al., 2008). Under projected shortage of rainfall for 2018-2050, it is

expected that yield and production of almonds could be impacted by drought in the NAP region if

not properly managed.

Almonds require winter a chilling period of 400–900 hours (≈ 16.5–37.5 days) below 7.2 °C (Pitt

et al., 2013) to initiate flowering. The analysis of historic and future climate data (Figure 6) shows

that the number of chilling days is expected to decrease in the future. The current NAP climate

still accommodates this chill requirement (Figure 7A). However, projections (Figure 7B) for 2018-

2050 show that there would be some years (e.g., 2037) that chilling requirement cannot be met for

the almonds and warmer winters will bring chill accumulation closer to the lower end of almond

requirements. Previous research (Pitt et al., 2013) in the NAP region has shown that once flowering

commenced, the warmer spring is likely to be advantageous for pollination and nut development,

as well as reduced pressure from disease. One of the problems associated with increased frequency

of extreme heat events (Figure 1) in summer is bud initiation during the post-harvest development

phase, which may affect the following year’s crop (Pitt et al., 2013). Increasing irrigation is

suggested to reduce the negative impacts of high summer temperatures.

The use of the number of chilling hours below 7.2 °C was found to overstate the loss of chill under

warmer conditions, especially in warmer sites (Leudeling and Brown 2011). An alternative

approach uses a dynamic model (Zhang and Taylor, 2011) to estimate chill portions; for almonds

the minimum required chill portions depend on variety, ranging from 22 to 32 (Thomas et al.,

2010). Thomas et al. (2010) calculated that for a +1°C temperature change, 10% less rainfall and

3% higher evapotranspiration, the mean chill portions on the NAP would be 40. This is still higher

than the required value (32). These results are not materially different from those obtained based

on the number of chilling hours below 7.2 °C: for all but one year is the chilling condition satisfied

(Figure 7).


Figure 7 Number of chilling days for almonds. (A) During historic climate (1985-2017); (B) During future climate (2018-2050).

1990 2000 2010 2020Year

10

20

30

40

50

60

70N

umbe

r of c

hilli

ng d

ays

(A)

Almonds chilling requirement - higher limit

Almonds chilling requirement - lower limit

2020 2030 2040 2050Year

10

20

30

40

50

60

70

Num

ber o

f chi

lling

day

s

(B)

Almonds chilling requirement higher limit

Almonds chilling requirement lower limit


Pistachio trees were first planted in south-eastern Australia in the early 1980s (NPA, 2019; PGAI,

2019). In 1982, CSIRO introduced a new variety of pistachio tree, Sirora, which has a high yield

and is well suited for Australian weather conditions. This variety includes special features such as

an excellent flavour, bright green kernel colour, high percentage of wide splits (easy to open), as

well as white shell (NPA, 2019). Pistachios are considered drought tolerant species (Elloumi et al.,

2013); however, projected heat or drought stress (Figure 1 and Figure 3), may reduce the yield and

lead to development of a hull cracking known as “early split” (Doster and Michailides, 1995;

Hadavi, 2005; Cotty and Jaime-Garcia, 2007).

Similar to almonds, flowering and nut yield of pistachio trees is a function of chill accumulation

(Elloumi et al., 2013). Previous studies have shown that warm winter temperatures with lack of

chilling induce flower bud drop, underdevelopment of the pistil (Rodrigo, 2000; Rodrigo and

Herrero, 2002), floral and leaf bud bursting delay (Arora et al., 2003), poor fruit set and low quality

(George et al., 2002; Arora et al., 2003), irregular flowering and rosette formation (George et al.,

2002), smaller terminal shoot extension and abnormal leaves (Breeuwer et al., 2008; Javanshah,

2010). Insufficient chill may also reduce pollination and crop yields (Rattigan and Hill, 1986;

Gradziel et al., 2007; Baldocchi and Wong, 2008). Pistachios require winter a chilling period of

600–1050 hours (≈ 25–43.75 days) below 7.2 °C (Küden et al., 1994). While the current NAP

climate (Figure 8A) hardly accommodates this chill requirement, projections for 2018-2050 show

there would be some years that chilling requirement cannot be met for pistachios (Figure 8B). This

may present challenges to growers with changes in management practices and perhaps even

cultivars needed to accommodate new growing conditions (Luedeling and Brown, 2011; Luedeling

et al., 2009b; Zhang and Taylor, 2011). It should be noted that development of a dynamic model

(Fishman et al., 1987a, b; Zhang and Taylor, 2011) may assist farmers in determining the accurate

amount of chill required for ‘Sirora’ pistachio grown in Australia, especially in the areas where

the minimum winter temperature would be greater than 7.2 °C (Zhang and Taylor, 2011). The

dynamic model, which incorporates non-stationary and time-inhomogeneous processes into

chilling models, may better describe the bud break and yield responses of pistachio to chill in

southeast Australia (Zhang and Taylor, 2011). This model (Fishman et al., 1987a, b; Zhang and

Taylor, 2011) assumes that winter chill is accumulated in a two-step process. Initially, cold

temperatures result in the formation of an intermediate chilling product which can be destroyed by

high temperatures. Once a critical amount of this chilling product has accumulated, it converts to


a chill portion, which cannot be destroyed. A certain chill portion accumulation indicates

fulfilment of chilling requirement.

Figure 8 Number of chilling days for pistachios. (A) During historic climate (1985-2017); (B) During future climate (2018-2050).

To overcome the problems of lack of winter chilling for pistachios, the application of hydrogen

cyanamide is suggested, which has been demonstrated to advance and synchronize flowering of

1990 2000 2010 2020Year

20

30

40

50

60

70

Num

ber o

f chi

lling

day

s

(A)

Pistachios chilling requirement - higher limit

Pistachios chilling requirement - lower limit

2020 2030 2040 2050Year

20

30

40

50

60

70

Num

ber o

f chi

lling

day

s

(B)

Pistachios chilling requirement higher limit

Pistachios chilling requirement lower limit


male and female cultivars (Pontikis, 1989; Rahemi and Asghari, 2004). Previous research (Barone

et al., 2005) has shown that applications of hydrogen cyanamide and urea improve the pistachio

vegetative growth and maximum assimilation rate. In Australia, winter oil has been commonly

used due to its major role in pistachio production during low-chill years (Zhang and Taylor, 2011).

However, the time when the oil should be applied is a matter of debate since it can influence bloom

date and yield in pistachios (Zhang and Taylor, 2011).

6.3. Broad acre crops (lucerne, pasture)

Lucerne (syn. alfalfa, Medicago sativa) is considered as the most important and widely grown

fodder plant in the world since it provides an excellent source of protein and energy for livestock

(AgriFutures Australia, 2019). As an herbaceous perennial legume, Lucerne is capable of fixing

the atmospheric nitrogen and adding it to the organic nitrogen content of the soil which can remain

in the soil for several years and contribute to the nutrition of subsequent crops or pastures

(AgriFutures Australia, 2019). A deep tap root of up to 15m gives this plant several agronomic

benefits, including survival in dry periods of summer due to its ability to use water at depth in the

soil profile. The deep tap root has another benefit; that is, the plant can use water to alleviate rising

water tables and improve soil structure (AgriFutures Australia, 2019). Lucerne is considered as a

summer growing plant; however, plant breeding has extended the growing season, with many

winter varieties available for cultivation (AgriFutures Australia, 2019). It is usually planted in

autumn (irrigated Lucerne may also be planted in spring) when there is sufficient soil moisture and

soil temperatures are warm enough (AgriFutures Australia, 2019).

Lucerne was introduced to Australia over 200 years ago. In Australia, it is currently cultivated

across 3.2 million ha (Humphries, 2013) with more than 50 varieties available to Australian

growers (AgriFutures Australia, 2019). Although this plant suits a wide range of soil types and

environments from subtropical regions in Queensland to the cool climate of Tasmania, it grows

best in Southern Australia (AgriFutures Australia, 2019). It can be grown as a short-term pasture

in a cropping rotation to improve soil condition between annual crops and provide valuable forage,

and to produce the fodder (high quality hay, silage or chaff) and seed.


The temperature ranges of 15–25°C for daytime and 10–20°C during the night, are considered as

optimum temperatures for dry matter production (AgriFutures Australia, 2019). However, highly

active winter cultivars can tolerate lower temperatures. As the NAP region will be subjected to an

increased number of hot days, a decline in pasture production is anticipated for this region. This is

in agreement with previous studies (Cullen et al. 2009; Moore and Ghahramani 2013; Thomas,

2016) which predicted that climate change will lead to a decline in pasture production in South

Australia.

Winter frost can damage the new growth, but this may not influence the production of Lucerne

due to its ability to reshoot from its crown throughout the season (AgriFutures Australia, 2019).

Lucerne is considered as a relatively drought tolerant plant (AgriFutures Australia, 2019), but its

yield is a function of supplied water volume. It can survive with a minimum of 300–400 mm of

annual rainfall in the environments ranging from Tasmania to northern New South Wales

(AgriFutures Australia, 2019). A fodder production enterprise will have an irrigation water

requirement of 7–13ML/ha, depending on location. Seed production enterprises operate

successfully in south east Australia in an annual rainfall zone of 400–600mm but many enterprises

will budget to apply 4–8ML/ha of irrigation water (AgriFutures Australia, 2019). An efficient

water management should be practiced to maximise water use efficiency and to avoid waterlogging

(Lucerne is intolerant of even short periods of waterlogging) in irrigated Lucerne. Depending on

the end use of the crop, the irrigation strategy will be designed to provide adequate moisture at

critical development points of the crop hence monitoring of the soil moisture (e.g., using

tensiometers) is recommended (AgriFutures Australia, 2019).

Depending on varieties, Lucerne requires a period of dormancy over winter to build up its store of

energy for growth in spring and summer. Adequate dormancy is important for the persistence or

longevity of the crop, as well as protecting the plant from winter cold (AgriFutures Australia,

2019). Dormant and semi-dormant varieties grow very little during winter, while winter-active

varieties grow in winter. It should be noted that all Lucerne varieties grow well during spring and

summer, but their growth’ pattern depends on the availability of soil moisture. In general, dormant


and semi-dormant varieties persist longer than highly winter-active ones under stressful conditions

(NSW Department of Primary Industries, 2013).

7. Irrigation requirement estimated with FAO-56

Development of efficient irrigation guidelines requires detailed understanding of the irrigation

requirement for different crops, considering the effect of the climate regime (Phogat et al., 2019).

In the present study, the FAO-56 dual crop coefficient method (Allen et al., 1998, 2005) has been

used to estimate the long-term irrigation requirements of different crops (potato, carrot, onion,

wine grapes, almond, pistachio, pasture) for current and future climate in different soils

[Calcareous (Cal), hard red brown (HRB), sand over clay (SoC) and deep uniform to gradational

(DuG)] in the NAP region. Details of the methodology and resulting IR for all soil-crop

combinations were discussed in Annex 7. We limited the discussion here to a high-level summary

of the findings in terms of cumulative distribution functions (CDF) of annual irrigation

requirements (note that this statistical analysis complements the initial study).

Figure 9 to Figure 15 show the CDF of annual irrigation requirement (IR) under current and future

climate for different crops (potato, carrot, onion, wine grapes, almond, pistachio, pasture),

cultivated on various soils in the NAP region. Results show that crops will require a higher amount

of annual irrigation under future climate, depending on the soil textures and crop stress tolerance.

Regardless of the crop type, sand over clay soils require the highest irrigation (714.3 – 955.7 mm)

while deep uniform to gradational soils need the lowest irrigation (642.8 – 910.0 mm). The average

percent increase in IR in different soils ranges from:

• 3.5-5.8% for grapes;

• 6.0-7.2% for almonds;

• 3.0-4.5% for pistachios;

• 7.0-8.4% for pasture;

• 6.2-7.4% for carrot;

• 9.2-10.3% for onion;

• 8.8-11.0% for potato.


This highlights the greater dependency of these crops on irrigation. It should be noted that the

salinity of the region’s water resources presents a challenge to some irrigators (not only to

irrigators of perennial crops). Some NAP producers of potatoes and almonds currently face

significant salinity pressure (particularly those accessing saline groundwater). Increasing water

application will also increase the mass of salt imported into the growing systems. To mitigate this

problem, savvy irrigation management techniques including aligning leaching events with rain and

making use of less saline water sources (recycled water) are recommended.

Figure 9 CDF of annual irrigation requirement for winegrapes under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.

100 200 300 400 500Annual irrigation requirement (IR, mm)

0

25

50

75

100

Cum

ulat

ive

dist

ribut

ion

func

tion

(%)

Cal soil - historic climateCal soil - future climateHRB soil - historic climateHRB soil - future climateSoC soil - historic climateSoC soil - future climateDuG soil - historic climateDuG soil - future climate


Figure 10 CDF of annual irrigation requirement for almonds under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.


0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%

)Cal soil - historic climateCal soil - future climateHRB soil - historic climateHRB soil - future climateSoC soil - historic climateSoC soil - future climateDuG soil - historic climateDuG soil - future climate


Figure 11 CDF of annual irrigation requirement for pistachios under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.


0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%



Figure 12 CDF of annual irrigation requirement for pasture under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.


0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%



Figure 13 CDF of annual irrigation requirement for carrots under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.


0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%



Figure 14 CDF of annual irrigation requirement for onions under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.


0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%



Figure 15 CDF of annual irrigation requirement for potatoes under historic and future climate. Cal soil = calcareous soil; HRB soil = hard red brown soil; SoC soil = sand over clay soil; DuG = deep uniform to gradational soil.

8. Climate change effect on plant water stress

Traditionally readily available water (RAW) is defined as the fraction of total available soil water

(TAW) which a crop can extract from the root zone without suffering water stress. This situation

occurs when the pressure head is at optimal condition for root water extraction. TAW (mm) is

defined as

( ) rwpfc Z TAW θθ −= 1000 (1)

where θfc is the water content at field capacity (cm3/cm3), θwp represents the water content at wilting

point (cm3/cm3) and Zr is the rooting depth (m). The fraction of TAW which a crop can extract

from the root zone without suffering water stress is the RAW, given as:

200 400 600 800Annual irrigation requirement (IR, mm)

0

25

50

75

100Cu

mul

ativ

e di

strib

utio

n fu

nctio

n (%



TAW pRAW = (2)

where p is the average fraction of TAW that can be depleted from the root zone before water stress

occurs (0-1). Because both RAW and TAW are static parameters that give no indication of the

actual soil water available to crops, an alternative approach was adopted to compare conditions

under which a crop experiences crop stress under historic and future climate.

In the present work, the Feddes (1978) model was used to define water stress conditions in soils

and its effects on root water uptake, using root water uptake parameters (Feddes parameters) from

Table 4. Following Rassam et al. (2018), the Feddes model assumes water uptake to be zero when

the soil is close to its saturation state (i.e., wetter than some arbitrary “anaerobiosis point” P0), and

in the situations when the pressure head is less (more negative) than the wilting point (P3). For

pressure heads between P2 and P3 (or P0 and Popt), the relationship between water uptake with

pressure head is linear. The root water uptake is optimal between pressure heads Popt and P2. Root

water uptake can be optimal even at full saturation, when both P0 and Popt are equal to zero.

Table 4 Root water uptake parameters for pasture (Wesseling, J. G., 1991).

P0 (cm) -10 Popt (cm) -25 P2H (cm) -200 P2L (cm) -800 P3 (cm) -8000 r2H(cm/day) 0.5 r2L (cm/day) 0.1

For pasture, the pressure head (P2L) of 800 cm (or equivalent matric suction of -800 cm) was

considered to be the highest suction beyond which root water extraction becomes suboptimal (i.e.

stressed). We calculated the number of total days the pasture’s root zone would experience water

stress, i.e. number of days with a matric suction was less than -800 cm. This was calculated for

both historic and future climate (Table 5). Under future climate, pastures on deep uniform to

gradational soils will experience 243 extra water stress days than under historic climate, or 7.6

days per year. Over the 32 years period, hard red brown soils will experience only 105 extra water


stress days or 3.3 per year under future climate. Note that the considered water stress is only a

practical water stress, as plants can still extract water but it will be under sub-optimal conditions.

Full water stress (plant water uptake ceases completely) occurs for pasture at a matric suction of -

8000 cm (Table 4).

Table 5 Number of days over the complete simulation period (32 years or 11869 days) that the soil matric suction over the root zone was less than -800 cm. Difference between historic and future climate and average days per year.

Soil group Historic climate Future climate Difference Days/year

Calcareous 8274 8487 213 6.7

Hard red brown 6397 6502 105 3.3

Sand over clay 7314 7501 187 5.8

Deep uniform to

gradational

8597 8840 243 7.6

The full data set of simulated root zone pressure heads for pasture for 32 years of simulation (11869

days with each day a data point) is shown as cumulative distribution function (probability of non-

exceedance) in Figure 16 to Figure 19. Each graph shows the percentage of the simulation time the

soil matric suction was at a certain value. Two threshold values are also shown: the matric pressure

P2 (see Table 4) beyond which the water uptake is no longer optimal (here -800 cm), and the matric

pressure P3 beyond which the root water uptake ceases completely (here -8000 cm). This allows

to estimate the percentage of time the crop has optimal root water uptake, the percentage of time

the crop experiences partial water stress (between P2 and P3), and the percentage of time the crop

does not take up water (beyond P3: full water stress). The largest difference in soil matric pressure

between historic and future climate is observed during the partial water stress period. This is the

period during which irrigation becomes most beneficial.


Figure 16 CDF (probability of non-exceedance) of matric suction for pasture subject to a natural condition (only rainfall, no irrigation) under historic and future climate (calcareous soil). Data covers 32 years of simulation.

10 100 1000 10000

Soil matric suction (cm)

0

10

20

30

40

50

60

70

80

90

100Pr

obab

ility

of n

on-e

xcee

danc

e (%

)

Historic climateFuture climate-800 cm: water stress threshold

-8000 cm: crop wilting point

Optimal water uptake Partial water stress

Full water stress

Calcareous soil


Figure 17 CDF (probability of non-exceedance) of matric suction for pasture subject to a natural condition (only rainfall, no irrigation) under historic and future climate (hard red brown soil). Data covers 32 years of simulation.

10 100 1000 10000


0

10

20

30

40

50

60

70

80

90

100Pr

obab

ility

of n

on-e

xcee

danc

e (%

)




Full water stress

Hard red brown soil


Figure 18 CDF (probability of non-exceedance) of matric suction for pasture subject to a natural condition (only rainfall, no irrigation) under historic and future climate (sand over clay soil). Data covers 32 years of simulation.

10 100 1000 10000


0

10

20

30

40

50

60

70

80

90

100Pr

obab

ility

of n

on-e

xcee

danc

e (%

)




Full water stress

Sand over clay soil


Figure 19 CDF (probability of non-exceedance) of matric suction for pasture subject to a natural condition (only rainfall, no irrigation) under historic and future climate (deep uniform to gradational soil). Data covers 32 years of simulation.

9. Conclusions

The effect of climate extremes on crop growth and irrigation requirement in the Northern Adelaide

Plains (NAP), South Australia (SA) was investigated using two time series of identical length as

representatives of historic (1985-2017) and future (2018-2050) climate. The climate indices for

historic and future climate were analyzed to determine the frequency of extreme climate events,

and its impact on the growth of horticultural crops (potatoes, carrots, and onions), fruit trees (vines,

almonds, and pistachios), and broad acre crops (Lucerne and pasture).

Results show that the NAP region will be subjected to more hot (> 35°C) and dry days (< 1 mm

rain), and milder winters. It was predicted that different crops in the region will be impacted

10 100 1000 10000


0

10

20

30

40

50

60

70

80

90

100Pr

obab

ility

of n

on-e

xcee

danc

e (%

)




Full water stress



differently by these climatic shifts. Potatoes will be subjected to a yield decrease and a higher risk

to being invaded by pests, while the growth and yield in carrots will be stimulated due to increased

frequency of hot and dry days (and dry spells). However, extreme heat events may reduce the

quality of carrots and mid-season drought stress depress the yield in carrots. The warmer climate

may reduce the duration of crop growth, yield, and seed production for onions, whereas, the low

rainfall condition can reduce the risk of infection by pests (i.e., leaf blight). The projected drought

and extremely hot weather in the NAP region can negatively impact vines, which may results in

poor budburst, leaf loss, bunch damage, and consequently low yield and production or even crop

loss. Warmer summers may lead to adopting warmer climate grape varieties to avoid loss of quality

of wines. For almonds and pistachios, it is expected that their yield and production will be impacted

by drought in the NAP region if not properly managed. While the current NAP climate hardly

accommodates chill requirements for these fruits, the projected climate show there would be some

years that the chilling requirement cannot be met.

As the NAP region will be subjected to an increased number of hot and dry days, higher irrigation

requirements are needed. Irrigation requirements for abovementioned crops, simulated via the

FAO-56 dual crop coefficient method, revealed that generally crops will require a higher amount

of annual irrigation in the future, depending on the soil textures and crop stress tolerance.

Simulation results illustrated that sand over clay soils require the highest irrigation (from 714 to

956 mm) while deep uniform to gradational soils need the lowest irrigation (from 643 to 910 mm)

among other soils, regardless of the crop type. It is anticipated that annual horticultural crops face

more irrigation related risks under the future climate compared to deep rooted perennial

horticultural crops.

The results of water balance simulation showed that under future climate, pastures on deep uniform

to gradational soils will experience 243 extra water stress days (over a 32 year period) compared

to historic climate, or 7.6 days per year. Over the 32 years period, hard red brown soils will

experience only 105 extra water stress days or 3.3 per year under future climate. This highlights

the fact that under future climate, pasture’s production will be reduced in the absence of irrigation.

As the NAP region will be subjected to increased number of hot days, a decline in pasture

production is anticipated for this region. This is in agreement with previous studies (Cullen et al.


2009; Moore and Ghahramani 2013; Thomas, 2016) which predicted that climate change will lead

to decline in pasture production in South Australia.

References

ABA, 2017. The Almond Board of Australia, 2017. Almond Insights 2017-2018. Berri, South

Australia: The Almond Board of Australia.

ABS, 2012. Agricultural Commodities: Small Area Data, Australia, 2010/11, Australian Bureau

of Statistics Cat. No. 7503.0, Canberra.

Adams, R.M., Rosenzweig, C., Peart, R.M., Ritchie, J.T., McCarl, B.A., Glyer, J.D., Curry, R.B.,

Jones, J.W., Boote, K.J. and Allen Jr, L.H., 1990. Global climate change and US agriculture.

Nature, 345(6272), p.219.

AgriFutures Australia, 2019. https://www.agrifutures.com.au/farm-diversity/lucerne/.

Allen, M. R., Dube, O. P., Solecki, W., Aragón–Durand, F., Cramer, W., Humphreys, S., Kainuma,

M., Kala, J., Mahowald, N., Mulugetta, Y., Perez, R., Wairiu, M., Zickfeld, K., 2018.

Framing and Context. In: Global warming of 1.5°C. An IPCC Special Report on the impacts

of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas

emission pathways, in the context of strengthening the global response to the threat of

climate change, sustainable development, and efforts to eradicate poverty [V. Masson-

Delmotte, P. Zhai, H. O. Pörtner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani, W. Moufouma-

Okia, C. Péan, R. Pidcock, S. Connors, J. B. R. Matthews, Y. Chen, X. Zhou, M. I. Gomis,

E. Lonnoy, T. Maycock, M. Tignor, T. Waterfield (eds.)]. In Press.

Allen, R.G., 2000. Using the FAO-56 dual crop coefficient method over an irrigated region as part

of an evapotranspiration inter-comparison study. J. Hydrol. 229 (1–2), 27–41.

Allen, R.G., Pereira L.S., 2009. Estimating crop coefficients from fraction of ground cover and

height. Irrig. Sci. 28, 17–34.

Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: Guidelines for


Rome, Italy.

https://www.agrifutures.com.au/farm-diversity/lucerne/


Allen, R.G., Pereira, L.S., Smith, M., Raes, D., Wright, J.L., 2005. FAO-56 dual crop coefficient

method for estimating evaporation from soil and application extensions. J. Irrig. Drain. Eng.

131, 2–13.

Araujo, J.A., Abiodun, B.J. and Crespo, O., 2016. Impacts of drought on grape yields in Western

Cape, South Africa. Theoretical and applied climatology, 123(1-2), pp.117-130.

Arora, R., Rowland, L.J. and Tanino, K., 2003. Induction and release of bud dormancy in woody

perennials: a science comes of age. HortScience, 38(5), pp.911-921.

Australian Bureau of Meteorology, 2008. An exceptional and prolonged heatwave in Southern

Australia. Special Climate Statement 15.

Australian Bureau of Meteorology, 2009a. The exceptional January-February 2009 heatwave in

south-eastern Australia. Special Climate Statement 17.

Australian Bureau of Meteorology, 2009b. A prolonged spring heatwave in central and south-

eastern Australia. Special Climate Statement 19.

Australian Bureau of Meteorology, 2010. "Extreme Heat Services for South Australia".

Retrieved 06 February 2019.

Australian Bureau of Meteorology, 2011. An exceptional summer heatwave in greater Sydney and

the Hunter Valley. Special Climate Statement 27.

Baldocchi, D. and Wong, S., 2008. Accumulated winter chill is decreasing in the fruit growing

regions of California. Climatic Change, 87(1), pp.153-166.

Barone, E., La Mantia, M., Marra, F.P., Motisi, A. and Sottile, F., 2005. Manipulation of the

vegetative and reproductive cycle of pistachio (Pistacia vera L.). Options

Mediterraneennes, 63, pp.355-364.

Bouton, J.H., 2012. Breeding lucerne for persistence. Crop and Pasture Science, 63(2), pp.95-106.

Breeuwer, A., Heijmans, M.M., Robroek, B.J. and Berendse, F., 2008. The effect of temperature

on growth and competition between Sphagnum species. Oecologia, 156(1), pp.155-167.

CCSP (Climate Change Science Program), 2008. Synthesis and Assessment Product 4.3 (SAP

4.3): the effects of climate change on agriculture, land resources, water resources, and bio-

diversity in the United States. Washington, DC: CCSP.

Chen, C.C., McCarl, B.A. and Schimmelpfennig, D.E., 2004. Yield variability as influenced by

climate: A statistical investigation. Climatic Change, 66(1-2), pp.239-261.

http://www.bom.gov.au/announcements/media_releases/sa/20100115_First_Heatwave_SA_Jan.shtml


Charles, S.P. and Fu, G., 2014. Statistically Downscaled Projections for South Australia-Task 3

CSIRO Final Report. Goyder Institute for Water Research, Technical Report Series, (15/1).

Church, J.A. and White, N.J., 2006. A 20th century acceleration in global sea‐level

rise. Geophysical research letters, 33(1).

Church, J.A. and White, N.J., 2011. Sea-level rise from the late 19th to the early 21st

century. Surveys in geophysics, 32(4-5), pp.585-602.

Coakley, S.M., Scherm, H. and Chakraborty, S., 1999. Climate change and plant disease

management. Annual review of phytopathology, 37(1), pp.399-426.

Conde, A., Neves, A., Breia, R., Pimentel, D., Dinis, L.T., Bernardo, S., Correia, C.M., Cunha, A.,

Gerós, H. and Moutinho-Pereira, J., 2018. Kaolin particle film application stimulates

photoassimilate synthesis and modifies the primary metabolome of grape leaves. Journal of

plant physiology, 223, pp.47-56.

Conde, A., Pimentel, D., Neves, A., Dinis, L.T., Bernardo, S., Correia, C.M., Gerós, H. and

Moutinho-Pereira, J., 2016. Kaolin foliar application has a stimulatory effect on

phenylpropanoid and flavonoid pathways in grape berries. Frontiers in plant science, 7,

p.1150.

Cotty, P.J. and Jaime-Garcia, R., 2007. Influences of climate on aflatoxin producing fungi and

aflatoxin contamination. International journal of food microbiology, 119(1-2), pp.109-115.

Crimp, S., Bakar, K.S., Kokic, P., Jin, H., Nicholls, N. and Howden, M., 2015. Bayesian space–

time model to analyse frost risk for agriculture in Southeast Australia. International Journal

of Climatology, 35(8), pp.2092-2108.

Crimp, S., Gobbet, D., Thomas, D., Bakar, S., Nidumolu, U., Hayman, P. and Hopwood, G., 2012.

Understanding frost risk in a variable and changing climate. Grains Research and

Development Corporation, Project code MCV00010.

Cullen, B.R., Johnson, I.R., Eckard, R.J., Lodge, G.M., Walker, R.G., Rawnsley, R.P. and

McCaskill, M.R., 2009. Climate change effects on pasture systems in south-eastern

Australia. Crop and pasture science, 60(10), pp.933-942.

Daymond, A.J., Wheeler, T.R., Hadley, P., Ellis, R.H. and Morison, J.I.L., 1997. The growth,

development and yield of onion (Allium cepa L.) in response to temperature and

CO2. Journal of Horticultural Science, 72(1), pp.135-145.


De Herralde, F., Biel, C. and Save, R., 2003. Leaf photosynthesis in eight almond tree

cultivars. Biologia Plantarum, 46(4), pp.557-561.

Deschênes, O. and Greenstone, M., 2007. The economic impacts of climate change: evidence from

agricultural output and random fluctuations in weather. American Economic Review, 97(1),

pp.354-385.

Dinis, L.T., Bernardo, S., Conde, A., Pimentel, D., Ferreira, H., Félix, L., Gerós, H., Correia, C.M.

and Moutinho-Pereira, J., 2016a. Kaolin exogenous application boosts antioxidant capacity

and phenolic content in berries and leaves of grapevine under summer stress. Journal of plant

physiology, 191, pp.45-53.

Dinis, L.T., Ferreira, H., Pinto, G., Bernardo, S., Correia, C.M. and Moutinho-Pereira, J., 2016b.

Kaolin-based, foliar reflective film protects photosystem II structure and function in

grapevine leaves exposed to heat and high solar radiation. Photosynthetica, 54(1), pp.47-55.

Dinis, L.T., Malheiro, A.C., Luzio, A., Fraga, H., Ferreira, H., Gonçalves, I., Pinto, G., Correia,

C.M. and Moutinho-Pereira, J., 2017. Improvement of grapevine physiology and yield under

summer stress by kaolin-foliar application: water relations, photosynthesis and oxidative

damage. Photosynthetica, pp.1-12.

Doster, M.A. and Michailides, T.J., 1995. The relationship between date of hull splitting and decay

of pistachio nuts by Aspergillus species. Plant disease (USA).

Dragland, S., 1978. Nitrogen and water requirements for carrots. Forsk. Fors. L. andbruket, 29,

pp.139-59.

Du, Y.L., Wang, Z.Y., Fan, J.W., Turner, N.C., He, J., Wang, T. and Li, F.M., 2013. Exogenous

abscisic acid reduces water loss and improves antioxidant defence, desiccation tolerance and

transpiration efficiency in two spring wheat cultivars subjected to a soil water

deficit. Functional Plant Biology, 40(5), pp.494-506.

Elloumi, O., Ghrab, M., Kessentini, H. and Mimoun, M.B., 2013. Chilling accumulation effects

on performance of pistachio trees cv. Mateur in dry and warm area climate. Scientia

Horticulturae, 159, pp.80-87.

EPA, 2016. United States Environmental Protection Agency (EPA). Climate Change Indicators:

U.S. and Global Precipitation. https://www.epa.gov/climate-indicators/climate-change-

indicators-us-and-global-precipitation. Accessed December 14, 2018.

https://www.epa.gov/climate-indicators/climate-change-indicators-us-and-global-precipitation

https://www.epa.gov/climate-indicators/climate-change-indicators-us-and-global-precipitation


FAO (Food and Agriculture Organization of the United Nations), 2006. FAO Statistical Databases

Available from: http://faostat.fao.org.

Feddes, R. A., P. J. Kowalik, and H. Zaradny, Simulation of Field Water Use and Crop Yield, John

Wiley & Sons, New York, NY, 1978.

Fischer, G., Shah, M., Tubiello, F.N. and Van Velhuizen, H., 2005. Socio-economic and climate

change impacts on agriculture: an integrated assessment, 1990–2080. Philosophical

Transactions of the Royal Society of London B: Biological Sciences, 360(1463), pp.2067-

2083.

Fisher, P., 2007. Planning for a climate change, a case study, city of Fort Phillip. A NATCLIM

initiative.

Fishman, S., Erez, A. and Couvillon, G.A., 1987a. The temperature dependence of dormancy

breaking in plants: mathematical analysis of a two-step model involving a cooperative

transition. Journal of Theoretical Biology, 124(4), pp.473-483.

Fishman, S., Erez, A. and Couvillon, G.A., 1987b. The temperature dependence of dormancy

breaking in plants: computer simulation of processes studied under controlled

temperatures. Journal of Theoretical Biology, 126(3), pp.309-321.

Gale, P., Drew, T., Phipps, L.P., David, G. and Wooldridge, M., 2009. The effect of climate change

on the occurrence and prevalence of livestock diseases in Great Britain: a review. Journal of

applied microbiology, 106(5), pp.1409-1423.

Garrett, K.A., Dendy, S.P., Frank, E.E., Rouse, M.N. and Travers, S.E., 2006. Climate change

effects on plant disease: genomes to ecosystems. Annu. Rev. Phytopathol., 44, pp.489-509.

Gattuso, J.P., Hendriks, I.E. and Brewer, P.G., 2014. Free-ocean CO2 enrichment (FOCE)

systems: present status and future developments.

George, A.P., Broadley, R.H., Nissen, R.J. and Ward, G., 2000, November. Effects of new rest-

breaking chemicals on flowering, shoot production and yield of subtropical tree crops.

In International Symposium on Tropical and Subtropical Fruits 575 (pp. 835-840).

Ghrab, M., Gargouri, K. and Ben Mimoun, M., 2008. Long-term effect of dry conditions and

drought on fruit trees yield in dryland areas of Tunisia. Options méditerranéennes, Series

AN, 80.

http://faostat.fao.org/


Glenn, D.M., Cooley, N., Walker, R., Clingeleffer, P. and Shellie, K., 2010. Impact of kaolin

particle film and water deficit on wine grape water use efficiency and plant water

relations. HortScience, 45(8), pp.1178-1187.

Government of South Austraulia, 2006. Maximising returns from water in the Australian vegetable

industry: South Australia. www.ruralsolutions.sa.gov.au.

Gradziel, T.M., Lampinen, B., Connell, J.H. and Viveros, M., 2007. Winters’ Almond: An Early-

blooming, Productive, and High-quality Pollenizer for ‘Nonpareil. HortScience, 42(7),

pp.1725-1727.

Gruber, N., Hauri, C., Lachkar, Z., Loher, D., Frölicher, T.L. and Plattner, G.K., 2012. Rapid

progression of ocean acidification in the California Current System. science, 337(6091),

pp.220-223.

Hadavi, E., 2005. Several physical properties of aflatoxin-contaminated pistachio nuts:

Application of BGY fluorescence for separation of aflatoxin-contaminated nuts. Food

additives and contaminants, 22(11), pp.1144-1153.

Haile, M.G., Wossen, T., Tesfaye, K. and von Braun, J., 2017. Impact of climate change, weather

extremes, and price risk on global food supply. Economics of Disasters and Climate Change,

1(1), pp.55-75.

Harrison, P.A., Butterfield, R.E. and Gawith, M.J., 1995. Effects on winter wheat, sunflower,

onion and grassland in Europe, in P. A. Harrison, R. Butterfield and T. E. Downing (eds.),

Climate Change and Agriculture in Europe: Assessment of Impacts and Adaptations.

Research Report No.9, Environmental Change Unit, University of Oxford, UK, pp. 330–

388.

Haverkort, A.J., 1990. Ecology of potato cropping systems in relation to latitude and

altitude. Agricultural Systems, 32(3), pp.251-272.

Haverkort, A.J. and Verhagen, A., 2008. Climate change and its repercussions for the potato supply

chain. Potato research, 51(3-4), p.223.

Hayman, P., Longbottom, M., McCarthy, M. and Thomas, D., 2012. Managing grapevines during

heatwaves. GWRDC, Australia.

Hazara, P. and Som, M.G. 2009. Vegetable Seed Production and Hybrid Technology. Kalayani

Publisher, Ludhiana. 408p.

http://www.ruralsolutions.sa.gov.au/


Hose, E., Steudle, E. and Hartung, W., 2000. Abscisic acid and hydraulic conductivity of maize

roots: a study using cell-and root-pressure probes. Planta, 211(6), pp.874-882.

Houghton, J.T., Meiro Filho, L.G., Callander, B.A., Harris, N., Kattenburg, A. and Maskell, K.,

1996. Climate change 1995: the science of climate change. Climatic Change, p.584.

Howe, C., Jones R.N., Maheepala, S. and Rhodes, B. 2005. Implications of Potential Climate

Change for Melbourne’s Water Resources. CSIRO Atmospheric Research and Melbourne

Water, Aspendale and Melbourne, Vic, Australia.

www.melbournewater.com.au/content/library/news/whats_new/climate_change_study.pdf.

Humphries, A.W., 2013. Future applications of lucerne for efficient livestock production in

southern Australia. Crop and Pasture Science, 63(9), pp.909-917.

Idso, S.B., 1990. Interactive Effects of Carbon Dioxide and Climate Variables on Plant Growth

1. Impact of Carbon Dioxide, Trace Gases, and Climate Change on Global Agriculture,

(impactofcarbond), pp.61-69.

IPCC. 2013. Summary for the policy makers, In: Climate Change 2013. Physical Science Basis:

Contribution of Working Group I to the fifth assessment Report of the Intergovernmental

Panel on Climate Change. Cambridge University Press, Cambridge, UK and New York,

USA.

IPCC, 2018. Summary for Policymakers. In: Global warming of 1.5°C. An IPCC Special Report

on the impacts of global warming of 1.5°C above pre-industrial levels and related global

greenhouse gas emission pathways, in the context of strengthening the global response to

the threat of climate change, sustainable development, and efforts to eradicate poverty [V.

Masson-Delmotte, P. Zhai, H. O. Pörtner, D. Roberts, J. Skea, P. R. Shukla, A. Pirani, W.

Moufouma-Okia, C. Péan, R. Pidcock, S. Connors, J. B. R. Matthews, Y. Chen, X. Zhou,

M. I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, T. Waterfield (eds.)]. World

Meteorological Organization, Geneva, Switzerland, 32 pp.

Javanshah, A., 2010. Global warming has been affecting some morphological characters of

pistachio trees (Pistacia vera L.). African Journal of Agricultural Research, 5(24), pp.3394-

3401.

Janssen, S. and Van Ittersum, M.K., 2007. Assessing farm innovations and responses to policies:

a review of bio-economic farm models. Agricultural systems, 94(3), pp.622-636.

http://www.melbournewater.com.au/content/library/news/whats_new/climate_change_study.pdf


Jones, A.D. and Yosef, S., 2015. The implications of a changing climate on global nutrition

security. The fight against hunger and malnutrition. Oxford University Press, Oxford,

pp.432-466.

Jones, J.E., Wilson, S.J., Lee, G. and Smith, A.M., 2010. Effect of frost damage and pruning on

current crop and return crop of Pinot Noir. New Zealand Journal of Crop and Horticultural

Science, 38(3), pp.209-216.

Jones, P.D. and Moberg, A., 2003. Hemispheric and large-scale surface air temperature variations:

An extensive revision and an update to 2001. Journal of climate, 16(2), pp.206-223.

Jönsson, A.M., Pulatov, B., Linderson, M.L. and Hall, K., 2013. Modelling as a tool for analysing

the temperature‐dependent future of the C olorado potato beetle in E urope. Global change

biology, 19(4), pp.1043-1055.

Küden, A.B., Kaska, N., Tanriver, E., Tekin, H. and Ak, B.E., 1994, September. Determining the

chilling requirements and growing degree hours of some pistachio nut cultivars and regions.

In I International Symposium on Pistachio 419 (pp. 85-90).

Lal, M., Kanwar, H.S., Verma, S.C., Chand, H. and Sharma, A., 2018. Impact of Climate

Variability on Seed Yield of Onion in Mid Hills of Himachal Pradesh, India. Int. J. Curr.

Microbiol. App. Sci, 7(2), pp.273-283.

Leuliette, E., 2014. The budget of recent global sea level rise: 1995-2013. Published by the

National Oceanic and Atmospheric Administration. [online

pdf] http://www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise/documents/NOAA_NESD....

Accessed December 14, 2018.

Li, C., Yin, C. and Liu, S., 2004. Different responses of two contrasting Populus davidiana

populations to exogenous abscisic acid application. Environmental and Experimental

Botany, 51(3), pp.237-246.

Lindsey, R., 2017. Climate Change: Global Sea Level. National Oceanic and Atmospheric

Administration. https://www.climate.gov/news-features/understanding-climate/climate-

change-global-sea-level.

Lobell, D.B. and Field, C.B., 2007. Global scale climate–crop yield relationships and the impacts

of recent warming. Environmental research letters, 2(1), p.014002.

Lobell, D.B. and Field, C.B., 2011. California perennial crops in a changing climate. Climatic

Change, 109(1), pp.317-333.

http://www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise/documents/NOAA_NESDIS_Sea_Level_Rise_Budget_Report_2014.pdf

https://www.climate.gov/news-features/understanding-climate/climate-change-global-sea-level

https://www.climate.gov/news-features/understanding-climate/climate-change-global-sea-level


Lobell, D.B., Schlenker, W. and Costa-Roberts, J., 2011. Climate trends and global crop

production since 1980. Science, 333(6042), pp.616-620.

Lockwood, D.W. and Coston, D.C., 2005. Peach tree physiology. Southeastern Peach Growers

Handbook, Cooperative Extension Service, University of Georgia. Athens, Georgia, US,

pp.5-7.

Luedeling, E., 2012. Climate change impacts on winter chill for temperate fruit and nut production:

a review. Scientia Horticulturae, 144, pp.218-229.

Luedeling, E. and Brown, P.H., 2011. A global analysis of the comparability of winter chill models

for fruit and nut trees. International Journal of Biometeorology, 55(3), pp.411-421.

Luedeling, E., Girvetz, E.H., Semenov, M.A. and Brown, P.H., 2011. Climate change affects

winter chill for temperate fruit and nut trees. PloS one, 6(5), p.e20155.

Luedeling, E., Zhang, M. and Girvetz, E.H., 2009a. Climatic changes lead to declining winter chill

for fruit and nut trees in California during 1950–2099. PloS one, 4(7), p.e6166.

Luedeling, E., Zhang, M., Luedeling, V. and Girvetz, E.H., 2009b. Sensitivity of winter chill

models for fruit and nut trees to climatic changes expected in California's Central

Valley. Agriculture, ecosystems & environment, 133(1-2), pp.23-31.

Küden, A.B., Küden, A., Nikpeyma, Y. and Kaska, N., 1994. September. Effects of chemicals on

bud break of pistachios under mild climate conditions. In I International Symposium on

Pistachio 419 (pp. 91-96).

Malhotra, S.K., 2017. Horticultural Crops and Climate Change-A Review. Indian J. agric.

Science, 87(1), pp.12-22.

Maracchi, G., Sirotenko, O. and Bindi, M., 2005. Impacts of present and future climate variability

on agriculture and forestry in the temperate regions: Europe. Climatic Change, 70(1-2),

pp.117-135.

Marcott, S.A., Shakun, J.D., Clark, P.U. and Mix, A.C., 2013. A reconstruction of regional and

global temperature for the past 11,300 years. Science, 339(6124), pp.1198-1201.

McCarl, B.A., Villavicencio, X. and Wu, X., 2008. Climate change and future analysis: is

stationarity dying?. American Journal of Agricultural Economics, 90(5), pp.1241-1247.

McKeown, N.M., Meigs, J.B., Liu, S., Saltzman, E., Wilson, P.W. and Jacques, P.F., 2004.

Carbohydrate nutrition, insulin resistance, and the prevalence of the metabolic syndrome in

the Framingham Offspring Cohort. Diabetes care, 27(2), pp.538-546.


McMichael, A.J., Powles, J.W., Butler, C.D. and Uauy, R., 2007. Food, livestock production,

energy, climate change, and health. The lancet, 370(9594), pp.1253-1263.

Messiaen, C.M., 1992. The tropical vegetable garden. Principles for improvement and increased

production with application to the main vegetable types. Macmillan Press Ltd.

Min, S.K., Zhang, X., Zwiers, F.W. and Hegerl, G.C., 2011. Human contribution to more-intense

precipitation extremes. Nature, 470(7334), p.378.

Moore, A.D. and Ghahramani, A., 2013. Climate change and broadacre livestock production

across southern A ustralia. 1. Impacts of climate change on pasture and livestock

productivity, and on sustainable levels of profitability. Global Change Biology, 19(5),

pp.1440-1455.

Msuya, D.G., Reuben, S.O.W.M., Mbilinyi, L.B., Maerere, A.P., Msogoya, T., Mulungu, L.S. and

Misangu, R.N., 2005. Evaluation of field performance and storage of some tropical short-

day onion (Allium cepa L.) cultivars. West African Journal of Applied Ecology, 8(1).

NOAA (National Oceanic and Atmospheric Administration), 2018. National Centers for

Environmental Information. https://www.noaa.gov/resource-collections/ocean-

acidification.

NPA (Nut Producers of Australia), 2019. https://nutproducers.com.au/pistachio-products/,

[Accessed 01-Febraury-2019].

NSW Department of Primary Industries, 2013.

http://www.dpi.nsw.gov.au/__data/assets/pdf_file/0007/469960/lucerne-varieties.pdf

[accessed 22-Feb-19].

Olesen, J.E. and Bindi, M., 2002. Consequences of climate change for European agricultural

productivity, land use and policy. European journal of agronomy, 16(4), pp.239-262.

Olesen, J.E., Trnka, M., Kersebaum, K.C., Skjelvåg, A.O., Seguin, B., Peltonen-Sainio, P., Rossi,

F., Kozyra, J. and Micale, F., 2011. Impacts and adaptation of European crop production

systems to climate change. European Journal of Agronomy, 34(2), pp.96-112.

Ou, C., Du, X., Shellie, K., Ross, C. and Qian, M.C., 2010. Volatile compounds and sensory

attributes of wine from cv. Merlot (Vitis vinifera L.) grown under differential levels of water

deficit with or without a kaolin-based, foliar reflectant particle film. Journal of agricultural

and food chemistry, 58(24), pp.12890-12898.

https://www.noaa.gov/resource-collections/ocean-acidification

https://www.noaa.gov/resource-collections/ocean-acidification

https://nutproducers.com.au/pistachio-products/

http://www.dpi.nsw.gov.au/__data/assets/pdf_file/0007/469960/lucerne-varieties.pdf


Parris, A., P. Bromirski, V. Burkett, D. Cayan, M. Culver, J. Hall, R. Horton, K. Knuuti, R. Moss,

J. Obeysekera, A. Sallenger, and J. Weiss., 2012. Global Sea Level Rise Scenarios for the

US National Climate Assessment. NOAA Tech Memo OAR CPO-1. 37 pp.

[online] http://cpo.noaa.gov/sites/cpo/Reports/2012/NOAA_SLR_r3.pdf. Accessed

December 14, 2018.

Pendergrass, A.G. and Hartmann, D.L., 2014. Changes in the distribution of rain frequency and

intensity in response to global warming. Journal of Climate, 27(22), pp.8372-8383.

Pendergrass, A.G., Lehner, F., Sanderson, B.M. and Xu, Y., 2015. Does extreme precipitation

intensity depend on the emissions scenario?. Geophysical Research Letters, 42(20),

pp.8767-8774.

Phogat, V., Mallants, D., and Cox, J., 2019. Estimation of long-term irrigation requirement of

crops for NAP region, South Australia. Unpublished Report.

Pistachio Growers Association Industry (PGAI), 2019. https://www.pgai.com.au/

Pitt, T., Osti, A., Alcoe, D., and Green, G., 2013. Climate change in the Northern Adelaide Plains

and implications for horticulture. DEWNR Technical Note.

https://www.waterconnect.sa.gov.au/Content/Publications/DEW/DEWNR_TN_2013_09.pdf.

Pontikis, C.A., 1989. Effects of hydrogen cyanamide on bloom advancement in female pistachio

(Pistacia vera L.). Fruit varieties journal (USA).

Pulatov, B., Linderson, M.L., Hall, K. and Jönsson, A.M., 2015. Modeling climate change impact

on potato crop phenology, and risk of frost damage and heat stress in northern

Europe. Agricultural and forest meteorology, 214, pp.281-292.

Pulatov, B., Hall, K., Linderson, M.L. and Jönsson, A.M., 2014. Effect of climate change on the

potential spread of the Colorado potato beetle in Scandinavia: an ensemble

approach. Climate Research, 62(1), pp.15-24.

Peng, S., Huang, J., Sheehy, J.E., Laza, R.C., Visperas, R.M., Zhong, X., Centeno, G.S., Khush,

G.S. and Cassman, K.G., 2004. Rice yields decline with higher night temperature from

global warming. Proceedings of the National Academy of Sciences, 101(27), pp.9971-9975.

Pelto, M.S., 2015. Global climate. Alpine glaciers [in “State of the Climate in 2014”]. Bulletin of

the American Meteorological Society, 96(12), pp.S19-S20.

Rassam, D., Šimůnek, J., Mallants, D., and Van Genuchten M.T., 2018. The HYDRUS-1D

Software Package for Simulating the Movement of Water, Heat, and Multiple Solutes in

http://cpo.noaa.gov/sites/cpo/Reports/2012/NOAA_SLR_r3.pdf

https://www.waterconnect.sa.gov.au/Content/Publications/DEW/DEWNR_TN_2013_09.pdf


Variably Saturated Media: Tutorial, Version 1.00. CSIRO Land and Water, Australia, pp.

183. – ONLINE, ISBN: 978-1-4863-1001-2.

Rattigan, K. and Hill, S.J., 1986. Relationship between temperature and flowering in

almond. Australian Journal of Experimental Agriculture, 26(3), pp.399-404.

Rahemi, M. and Asghari, H., 2004. Effect of hydrogen cyanamide (dormex), volk oil and

potassium nitrate on budbreak, yield and nut characteristics of pistachio (Pistacia vera

L.). The Journal of Horticultural Science and Biotechnology, 79(5), pp.823-827.

Rehman, M.U., Rather, G.H., Gull, Y., Mir, M.R., Mir, M.M., Waida, U.I. and Hakeem, K.R.,

2015. Effect of Climate Change on Horticultural Crops. In Crop Production and Global

Environmental Issues (pp. 211-239). Springer, Cham.

Reilly, J.M., J. Graham, D.G. Abler, R. Darwin, S. Hollinger, C. Izaurralde, S. Jagtap, J. Jones, J.

Kimble, B. McCarl, L. Mearns, D. Ojima, E.A. Paul, K. Paustian, S. Riha, N. Rosemberg,

C. Rosenzweig, and F. Tubiello, 2002. Changing Climate and Changing Agriculture: Report

of the Agricultural Sector Assessment Team. Cambridge: Cambridge University Press.

Reidsma, P., Wolf, J., Kanellopoulos, A., Schaap, B.F., Mandryk, M., Verhagen, J. and van

Ittersum, M.K., 2015. Climate change impact and adaptation research requires integrated

assessment and farming systems analysis: a case study in the Netherlands. Environmental

Research Letters, 10(4), p.045004.

Rodrigo, J., 2000. Spring frosts in deciduous fruit trees—morphological damage and flower

hardiness. Scientia Horticulturae, 85(3), pp.155-173.

Rodrigo, J. and Herrero, M., 2002. Effects of pre-blossom temperatures on flower development

and fruit set in apricot. Scientia Horticulturae, 92(2), pp.125-135.

Rotmans, J. and Van Asselt, M., 1996. Integrated assessment: a growing child on its way to

maturity. Climatic Change, 34(3-4), pp.327-336.

Rosenzweig, C. and Parry, M.L., 1994. Potential impact of climate change on world food

supply. Nature, 367(6459), p.133.

Rouhi, V., Samson, R., Lemeur, R. and Van Damme, P., 2007. Photosynthetic gas exchange

characteristics in three different almond species during drought stress and subsequent

recovery. Environmental and Experimental Botany, 59(2), pp.117-129.


Sørensen, J.N., Jørgensen, U. and Kühn, B.F., 1997. Drought effects on the marketable and

nutritional quality of carrots. Journal of the Science of Food and Agriculture, 74(3), pp.379-

391.

Schoneveld, J.A., 1991. Op tijd beregenen voorkomt schurft. Groenten en

Fruit/Vollegrondsgroenten, 1, p.25.

Schwartz, M.D., 1999. Advancing to full bloom: planning phenological research for the 21st

century. International Journal of Biometeorology, 42(3), pp.113-118.

Shellie, K., 2015. Foliar reflective film and water deficit increase anthocyanin to soluble solids

ratio during berry ripening in Merlot. American Journal of Enology and Viticulture, pp.ajev-

2015.

Shellie, K. and Glenn, D.M., 2008. Wine grape response to foliar particle film under differing

levels of preveraison water stress. HortScience, 43(5), pp.1392-1397.

Seckin, B., Sekmen, A.H. and Türkan, I., 2009. An enhancing effect of exogenous mannitol on the

antioxidant enzyme activities in roots of wheat under salt stress. Journal of Plant Growth

Regulation, 28(1), p.12.

Simon, P.W., 1985. Carrot flavor: effects of genotype, growing conditions, storage, and

processing. In Evaluation of quality of fruits and vegetables (pp. 315-328). Springer,

Dordrecht.

Šimůnek, J., van Genuchten M.T., Šejna, M. 2016. Recent developments and applications of the

HYDRUS computer software packages. Vadose Zone J. 15, 25.

Song, J., Shellie, K.C., Wang, H. and Qian, M.C., 2012. Influence of deficit irrigation and kaolin

particle film on grape composition and volatile compounds in Merlot grape (Vitis vinifera

L.). Food chemistry, 134(2), pp.841-850.

Steffen, W., 2006. Stronger evidence but new challenges: climate change science 2001-2005.

Department of the Environment and Heritage & Australian Greenhouse Office.

Supit, I., Van Diepen, C.A., De Wit, A.J.W., Wolf, J., Kabat, P., Baruth, B. and Ludwig, F., 2012.

Assessing climate change effects on European crop yields using the Crop Growth

Monitoring System and a weather generator. Agricultural and Forest Meteorology, 164,

pp.96-111.

Texas A&M University Agrilife Research & Extension, 2019. "Chilling hours".

https://etweather.tamu.edu/chill/ [accessed 07-Febraury-2019].

http://etweather.tamu.edu/chill/

https://etweather.tamu.edu/chill/


The Australian Water Association, 2019. http://www.awa.asn.au.

The Pistachio Growers Association Incorporated (PGAI), 2019. http://www.pgai.com.au/chilling-

requirements [accessed 4-Feb-19].

The Goyder Institute for Water Research, 2016. Northern Adelaide Plains Water Stocktake

Technical Report, The Goyder Institute for Water Research, Adelaide, South Australia.

The Goyder Institute for Water Research, 2019a.

http://www.goyderinstitute.org/research/foundation-research/climate-change/[accessed 11-

Feb-19].

The Goyder Institute for Water Research, 2019b. SA Climate Ready Regional Summary – AMLR, http://www.goyderinstitute.org/_r210/media/system/attrib/file/201/SA%20Climate%20Ready%20

Regional%20Summary%20-%20AMLR.pdf [accessed 11-Feb-19].

The Australian Wine Research Institute, 2015. Winemaking in a changing climate. Factsheet.

The Australian Wine Research Institute, 2018. Heatwave and effects on the vines:

https://www.awri.com.au/industry_support/climate-environment/heatwave-and-effects-on-the-

vines/.

Thomas, D., Hayman, P. and Bachar, Z., 2016. A Comparative Climate Vulnerability Assessment

for Key Agricultural Industries in the Adelaide and Mount Lofty Ranges Natural Resource

Management Region. Adelaide and Mount Lofty Ranges Natural Resources Management

Board, Adelaide.

Thompson, P.R., Merrifield, M.A., Leuliette, E., Sweet, W., Chambers, D.P., Hamlington, B.D.,

Jevrejeva, S., Marra, J.J., MItchum, G.T., Nerem, R.S., Widlansky, M.J., 2018. Sea level

variability and change [in State of the Climate in 2017]. Bulletin of the American

Meteorological Society, 99(8), S84–S87.

Thornton, P.K., 2012. Impacts of climate change on the agricultural and aquatic systems and

natural resources within the CGIAR’s mandate.Tindall H. D., 1983. Vegetables in the

tropics. New Delhi, India. CBS Publishers and Distributors, 533 pp.

Tindall, H.D., 1983. Vegetables in the Tropics. AVI Technical Books.

Van Ittersum, M.K., Ewert, F., Heckelei, T., Wery, J., Olsson, J.A., Andersen, E., Bezlepkina, I.,

Brouwer, F., Donatelli, M., Flichman, G. and Olsson, L., 2008. Integrated assessment of

agricultural systems–A component-based framework for the European Union

(SEAMLESS). Agricultural systems, 96(1-3), pp.150-165.

http://www.awa.asn.au/

http://www.pgai.com.au/chilling-requirements

http://www.pgai.com.au/chilling-requirements


http://www.goyderinstitute.org/research/foundation-research/climate-change/


http://www.goyderinstitute.org/_r210/media/system/attrib/file/201/SA%20Climate%20Ready%20Regional%20Summary%20-%20AMLR.pdf

http://www.goyderinstitute.org/_r210/media/system/attrib/file/201/SA%20Climate%20Ready%20Regional%20Summary%20-%20AMLR.pdf

https://www.awri.com.au/industry_support/climate-environment/heatwave-and-effects-on-the-vines/

https://www.awri.com.au/industry_support/climate-environment/heatwave-and-effects-on-the-vines/


Vose, R.S., Arndt, D., Banzon, V.F., Easterling, D.R., Gleason, B., Huang, B., Kearns, E.,

Lawrimore, J.H., Menne, M.J., Peterson, T.C. and Reynolds, R.W., 2012. NOAA's merged

land–ocean surface temperature analysis. Bulletin of the American Meteorological

Society, 93(11), pp.1677-1685.

Westra, S., Alexander, L.V. and Zwiers, F.W., 2013. Global increasing trends in annual maximum

daily precipitation. Journal of Climate, 26(11), pp.3904-3918.

Wesseling, J. G., 1991. Meerjarige simulaties van gronwateronttrekking voor verschillende

bodmprofielen, grondwatertrappen en gewassen met het model SWATRE. Report 152,

Winand Staring Centre, Wageningen, the Netherlands.

Wheeler, T. and Von Braun, J., 2013. Climate change impacts on global food

security. Science, 341(6145), pp.508-513.

Wheeler, T.R., Morison, J.I.L., Ellis, R.H. and Hadley, P., 1994. The effects of CO2, temperature

and their interaction on the growth and yield of carrot (Daucus carota L.). Plant, Cell &

Environment, 17(12), pp.1275-1284.

Wheeler, T.R., Ellis, R.H., Hadley, P., Morison, J.I.L., Batts, G.R. and Daymond, A.J., 1996.

Assessing the effects of climate change on field crop production. Aspects of Applied Biology

(United Kingdom).

White, M.A., Diffenbaugh, N.S., Jones, G.V., Pal, J.S. and Giorgi, F., 2006. Extreme heat reduces

and shifts United States premium wine production in the 21st century. Proceedings of the

National Academy of Sciences, 103(30), pp.11217-11222.

Wiik, L., 2014. Potato late blight and tuber yield: results from 30 years of field trials. Potato

research, 57(1), pp.77-98.

Wine Australia, 2018. 2018 SA Winegrape Crush Survey.

https://www.wineaustralia.com/getmedia/e38a0014-504e-4b2f-8f33-078716eb885c/SA-

full-report-2018 [Accessed 01 February 2019].

Wittwer, G., 2008. Will drought erode the competitiveness of Australia's wine industry?. Monash

University, Centre of Policy Studies and the Impact Project.

Wurr, D.C.E., Hand, D.W., Edmondson, R.N., Fellows, J.R., Hannah, M.A. and Cribb, D.M.,

1998. Climate change: a response surface study of the effects of CO 2 and temperature on

the growth of beetroot, carrots and onions. The Journal of Agricultural Science, 131(2),

pp.125-133.

https://www.wineaustralia.com/getmedia/e38a0014-504e-4b2f-8f33-078716eb885c/SA-full-report-2018

https://www.wineaustralia.com/getmedia/e38a0014-504e-4b2f-8f33-078716eb885c/SA-full-report-2018


Zhang, J. and Taylor, C., 2011. The dynamic model provides the best description of the chill

process on ‘Sirora’pistachio trees in Australia. HortScience, 46(3), pp.420-425.

Zhou, L., Xu, H., Mischke, S., Meinhardt, L.W., Zhang, D., Zhu, X., Li, X. and Fang, W., 2014.

Exogenous abscisic acid significantly affects proteome in tea plant (Camellia sinensis)

exposed to drought stress. Horticulture research, 1, p.14029.

Zhu, X. and Troy, T.J., 2018. Agriculturally Relevant Climate Extremes and Their Trends in the

World's Major Growing Regions. Earth's Future, 6(4), pp.656-672.

Ziska, L., Crimmins, A., Auclair, A., DeGrasse, S., Garofalo, J.F., Khan, A.S., Loladze, I., de

León, A.P., Showler, A., Thurston, J. and Walls, I., 2016. Ch. 7: Food safety, nutrition, and

distribution. The impacts of climate change on human health in the United States: A

scientific assessment, pp.189-216.

The Goyder Institute for Water Research is a partnership between the South Australian Government through the

Department for Environment and Water, CSIRO, Flinders University, the University of Adelaide, the University of South

Australia, and the International Centre of Excellence in Water Resource Management.

sustainable expansion of irrigated agriculture and ... nap task2...tomato, capsicum and eggplant was...

Documents