soil & water management & conservation evaluation of soil ......soil science society of...

Soil Science Society of America Journal

Evaluation of Soil Water and Nitrogen Distribution by Site-Specific Irrigation Scheduling Method in Tomato

Crop Grown on Sandy Soil

Soil & Water Management & Conservation

Irrigation scheduling based on a real-time and location-specific (rT-Ls) model increases irrigation water savings and yield. The rT-Ls irrigation scheduling models have been developed as smartphone applications and have been used for crop-specific irrigation requirements. Although many rT-Ls irrigation mod-els have been tested and used in several agronomic and horticultural crops in Florida, none of these irrigation-scheduling models has been tested for their impacts on nutrient distribution in Florida’s sandy soils. A two-season (fall 2015 and spring 2016) study was conducted to determine the effects of an rT-Ls–based irrigation scheduling on soil water, NO3

-–N, and NH4+–N distri-

butions during a tomato cropping season. In both seasons, an rT-Ls model for tomato was evaluated at three irrigation application rates (66, 100, and 150% rT-Ls–suggested amounts) and compared with a historic evapotranspiration (eT)-based irrigation schedule (Historic eT) currently recommended in Florida. This study suggests that the rT-Ls model improves water savings by 20 and 17% for the fall and spring seasons, respectively, compared with the Historic eT irrigation scheduling method. No specific pattern was observed for soil NH4

+–N concentration between scheduling methods, but the rT-Ls model maintained a higher soil NO3

-–N concentration within the crop root zone and hence could reduce NO3

-–N leaching potential. In each season, compared with the Historic eT irrigation method, the rT-Ls improved both nitrogen recovery and irrigation water use efficiency in the open-field fresh-market tomato production system. results obtained in this study clearly demonstrate that irrigation applications using the rT-Ls irrigation scheduler improved irri-gation scheduling accuracy by maintaining nutrients within the tomato root zone and hence could reduce nutrient leaching potential in sandy soil.

Abbreviations: DAT, days after transplanting; ET, evapotranspiration; ET0, reference evapotranspiration; iWUE, irrigation water use efficiency; REC-N, nitrogen recovery; RT-LS, real-time and location-specific irrigation scheduling; SI, SmartIrrigation; TMY, total marketable yield.

Tomato production in the United States is dominated by the states of California and Florida (Costa and Heuvelink, 2005). Tomato is a major vegetable crop in Florida, with a total harvested area of 11,000 ha, with a

value of US$ 382 million in 2016 (USDA, 2017). This represents 40% of the total production value in the United States for the same year (FDACS, 2017). In 2015, tomato ranked fourth in economic value among all agricultural crop commodities in Florida, next to orange, sugarcane, and ornamental plants (USDA, 2017).

Nutrients, particularly nitrogen (N), are a major part of tomato produc-tion. Nitrogen is critical for obtaining high yield and fruit quality (Hochmuth and Hanlon, 2014). The current recommended N rate for tomato in Florida is 224 kg ha–1 (Liu et al., 2016), but vegetable growers in Florida often apply fertil-izer at rates higher than recommended (Hochmuth and Hanlon, 2014), up to an

Ibukun T. Ayankojo* kelly T. Morgan kamal Mahmoud

Institute of Food and Agricultural Science South West Florida Research and Education Center Univ. of Florida Immokalee, FL 34142

Core Ideas

•smartIrrigation is a smartphone-based crop irrigation scheduling method using real-time and location-specific (rT-Ls) weather information.

•The rT-Ls irrigation scheduling method improved irrigation water savings and increased crop yield.

•The rT-Ls model can reduce nutrient leaching by maintaining water and nutrient within crop root zone.

•efficient irrigation scheduling can improve crop performance and increase production efficiencies.

Soil Sci. Soc. Am. J. 83:761–771 doi:10.2136/sssaj2018.07.0268 Received 20 July 2018. Accepted 11 Feb. 2019. *Corresponding author ([email protected]). © Soil Science Society of America. This is an open access article distributed under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Published June 13, 2019

762 Soil Science Society of America Journal

average of 417 kg ha–1 N for a production season (Sanjay et al., 2014). An increase in N application can reduce uptake efficiency (Liu et al., 2014), increase post-harvest soil N residue (Sainju et al., 1999; Zhang et al., 2011), and increase N leaching (Simonne and Ozores-Hampton, 2010b). These problems become even more challenging under inappropriate irrigation management, especially on sandy soils with a shallow water table. Therefore, efficient irrigation practice is critical to maintaining nutrients in the crop root zone (Zotarelli et al., 2007; Zotarelli, 2009a).

The introduction of drip irrigation system and plastic mulch, which has become a common cultural practice in Florida vegeta-ble production, can conserve soil moisture by reducing surface evaporation losses and increasing crop water (i.e., irrigation wa-ter use efficiency [iWUE]) and nutrient use efficiency (Zotarelli et al., 2008a). Although drip irrigation could potentially be more environmentally friendly, crop iWUE and nutrient use ef-ficiency using a drip irrigation system are dependent on the type of irrigation scheduling method and irrigation application time. This is because inappropriate irrigation scheduling method can increase the risk of nutrient leaching (Zotarelli et al., 2009a) and potentially reduce yield of drip-irrigated tomato crop.

A commonly adopted irrigation scheduling method for vegetable crops is the use of evapotranspiration (ET)-based soil balance models with long-term average daily ET or historic ET (Simonne and Ozores-Hampton 2010a). Other frequently used scheduling methods include soil moisture content (soil moisture sensor–based irrigation) (Zotarelli et al., 2008b); fixed time-based irrigation (Migliaccio et al., 2010); and, most recently, irri-gation scheduling based on real-time and location-specific (RT-LS) ET data (Migliaccio et al., 2014).

The RT-LS irrigation scheduling method, commonly known as SmartIrrigation (SI), is an ET-based soil water balance model designed as smartphone applications (SI App) using reference evapotranspiration (ET0) from the FAO Penman–Monteith procedure and crop coefficient to determine crop water require-ments (Migliaccio et al., 2016). Many SI applications have been tested and used in several crops, such as citrus (Citrus sinensis L.), cotton (Gossypium L.) (Vellidis et al., 2016), strawberry (Fragaria spp.), avocado (Persea Americana Mill.) (Migliaccio et al., 2014), tomato (Solanum lycopersicum L.) (Ayankojo et al., 2018; Miller et al., 2018b), and watermelon [Citrullus lanatus (Thunb.) Matsun. & Nakai] (Miller et al., 2018a).

In this study, the RT-LS method was compared with the grow-er’s standard irrigation scheduling method (irrigation scheduling based on historic ET averages [HI]) as recommended by Zotarelli et al. (2015). This study hypothesized that RT-LS improves irriga-

tion scheduling accuracy (in open-field tomato production with plastic mulch) and maintains higher nutrient concentration with-in the tomato root zone, thereby reducing N leaching potential and increasing N recovery (REC-N) compared with HI.

Although many of these SI Apps have been tested and proven effective for saving water (Miller et al., 2018b) and increasing crop yield (Ayankojo et al., 2018; Vellidis et al., 2014), none of these SmartIrrigation schedulers has been evaluated for soil water and nutrient dynamics in a cropping system. Therefore, this study was conducted (i) to evaluate the effects of RT-LS irrigation sched-uling model on soil moisture distribution in open-field tomato production with plastic mulch, (ii) to evaluate the effect of soil water distribution on soil ammonium-N (NH4

+–N) and nitrate-N (NO3

––N) distributions along the tomato growing season, and (iii) to determine the effects of irrigation scheduling on iWUE and REC-N in open-field fresh-market tomato production.

MATerIALs AND MeTHODssite Description and experimental Design

A two-season study was conducted during the fall 2015 and spring 2016 seasons on drip-irrigated fresh market tomato. The planting and final harvesting dates were 14 Sept. and 13 Dec. 2015 and 3 Feb. and 31 May 2016 for the fall and spring seasons, respectively. These studies were conducted at the Southwest Florida Research and Education Center in Immokalee (26°27´44˝ N and 81°26´36˝W).

The elevation at the study site was 10.4 m asl. The average air temperature and precipitation at the experimental location range from 16 to 27°C and from 965 to 1727 mm, respectively (USDA–NRCS, 2015). The soil at the study site was classified as Immokalee fine sand (Arenic, Alaquods, Sandy, Siliceous, Hyperthermic), with a nearly flat slope (0–2%), low runoff class, and poor natural drainage (USDA–NRCS, 2015). Selected phys-ical and chemical properties (Table 1) of the soil at the experimen-tal site were adapted from a previous study by Kadyampakeni et al. (2014b) on the same soil type adjacent to the experimental site.

During pre-planting operations, the soil at the study location was wet. A seepage irrigation system was used to maintain soil water at an optimum condition for bed formation. During bed (height, 20 cm; width, 81 cm) formation, a fumigant Pic Clor 60 (Agrian, Fresno, CA) (a.i chloropicrin and 1,3-dichloropropene at 59.6 and 39%, respectively) was applied at the rate of 223 kg ha–1, after which the beds were immediately covered with plastic mulch (Berry Plastics, Ecansville, IL). Two thin-wall drip lines (5 mil streamline Plus 630 series; Netafim, Fresno, CA) with an emitter discharge rate of 0.9 L h–1 located at every 0.61 m on the drip line

Table 1. soil physical and chemical properties at the experimental site for the Immokalee fine sand.

soil depth pH†

Organic matter

Cation exchange capacity

Bulk density

saturated hydraulic conductivity

Hydraulic conductivity at field capacity

saturated moisture content

Moisture content at field capacity

residual moisture content

cm % cmolc kg–1 g cm–3 – cm h–1 – – m3 m–3 –

0–15 5.6 0.61 7.63 1.62 15.82 0.37 0.34 0.09 0.013

15–30 5.2 0.41 0.74 1.62 13.97 0.13 0.36 0.10 0.013

30–45 5.8 0.49 0.33 1.59 13.22 0.63 0.39 0.10 0.013† Soil/water ratio = 1:2 (w/v). Table adapted from Kadyampakeni et al. (2014b).

www.soils.org/publications/sssaj 763

were placed on the bed (under the plastic mulch) for irrigation and fertigation. Before planting, treatment plots were arranged in a randomized complete block design with four replicates per treatment. Each plot consisted of three adjacent rows that were 18 m long and 1.8 m apart. At 21 d after fumigation, tomato seedlings (variety Charger; Sakata, Morgan Hill, CA) were transplanted at 0.61 m planting distance (90 plants per plot), resulting in 8966 plants ha–1.

In this study, four treatments were evaluated from two irrigations scheduling methods (the HI schedule and an irrigation schedule based on RT-LS weather information). The HI irrigation schedule is a grower-standard irrigation scheduling method recommended by Zotarelli et al. (2015) and was used as the control or standard to which the RT-LS scheduling method was evaluated. The daily amount of irrigation water application for the HI treatment (T1) was based on monthly ET0 data (Table 2) recommended for southwestern region of Florida (Zotarelli et al., 2015) using the following equation:

0 cET *ETc

0.95K =

[1]

where ET0 is the corresponding monthly reference evapotrans-piration value for Southwest Florida (Table 2), Kc is the crop coefficient value corresponding to a specific crop growth stage (Table 3), and 0.95 is the system efficiency used for the drip ir-rigation system (Zotarelli et al., 2015).

Irrigation schedules from the RT-LS treatments (SI) were ap-plied at 66% (T2), 100% (T3), and 150% (T4) of the SI Vegetable App recommended rate. Both T2 and T4 are within the irrigation application rates for low and high irrigation levels for tomato crop commonly found in the literature (Monte et al., 2013; Nangare et al., 2016). At the time of scheduling, the SI App automatically con-nects to the Florida Automated Weather Network and University of Georgia Weather Network stations for specific scheduling locations in Florida and Georgia, respectively. In this study, the RT-LS schedul-ing method used meteorological data of the previous 5 d prior to the scheduling time from the Florida Automated Weather Network sta-tion located within 0.5 km of the study site to cal-culate ET. Crop coefficient values were determined based on the time from planting and scheduling dates (Migliaccio et al., 2016).

Daily total irrigation time was divided into two or three (depending on the irrigation vol-ume) daily irrigation application frequencies. At every application, the amount of irrigation water applied was controlled by a hose-end ir-rigation timer (model IZEHTMR; Rain Bird, Azusa, CA). The amount of irrigation water applied for each treatment was measured and recorded by a flow meter (M 3.81 cm size; Netafim). A 15 PSI water pressure regulator (Senninger Irrigation Inc., Orlando, FL) was

installed per treatment to obtain a precise flow rate and to main-tain constant pressure along the drip lines.

Fertilizer application was the same for all treatments. Fertilizer was applied according to University of Florida/Institute of Food and Agricultural Sciences recommendations for tomato (Liu et al., 2016). Total season nutrient application (224 kg ha–1 N, 4 kg ha–1 P, and 224 kg ha–1 K) was applied as 25% pre-plant N and K using 18–0–18 N–P2O5–K2O fertilizer containing ammonium-nitrate (Griffin Fertilizer Co., Frostproof, FL). The remaining 75% each of N and K and 100% of P were applied through fertigation using N–P2O5–K2O ammonium-nitrate fertilizer 20–2–20 (Peters Inc., Allentown, PA). Fertigation was conducted twice a week for all treatments. During fertigation, the required amount of fertilizer was dissolved in 19 L of water for each treatment and injected into the drip lines using a pres-sure pump (12 VDC, 1.8 GPM; SHURflo, Cypress, CA).

soil sampling, Analytic Methods, and soil Moisture Distribution

The effect of irrigation scheduling method and rates on soil nutrient distribution were evaluated at 0 to 15 cm (D1), 15 to 30 cm (D2), and 30 to 45 cm (D3) soil depths. Because 85 to 95% of tomato root is within the first 30 cm of soil depth (Zotarelli

Table 2. Historical penman method reference evapotranspiration for six Florida regions (adapted from zotarelli et al., 2015).

Month Northwest Northeast Central Central West southwest southeast

–––––––––––––––––––––––––– m3 ha–1 d–1 ––––––––––––––––––––––––––

Jan. 15.23 17.77 17.77 17.77 20.31 20.31

Feb. 17.77 20.31 25.39 25.39 27.93 27.93

Mar. 25.39 25.39 30.46 33.01 33.01 33.01

Apr. 33.01 35.54 40.62 40.62 43.16 43.16

May 40.62 40.62 45.69 45.69 45.69 45.69

June 43.16 40.62 45.69 45.69 45.69 43.16

July 43.16 40.62 43.16 43.16 45.69 45.69

Aug. 38.08 38.08 43.16 40.62 43.16 40.62

Sept. 33.01 33.01 35.54 35.54 38.08 35.54

Oct. 22.85 25.39 27.93 27.93 30.46 30.46

Nov. 17.77 17.77 20.31 20.31 22.85 22.85

Dec. 12.70 15.23 15.23 15.23 17.77 17.77

Table 3. Tomato growth stages, crop coefficient (Kc) values, and duration for open-field tomato production during the fall 2015 and the spring 2016 production sea-sons in Immokalee, FL.

Crop growth stage

Kc values†

Kc Duration Kc application period for T1

T1‡ T2, T3, T4‡ Fall 2015 Spring 2016

–––––––––– DAT§ ––––––––––

1 0.40 0–14 0–18 14–28 Sept. 3–24 Feb.

2 0.75 15–35 19–37 29 Sept.–12 Oct. 25 Feb.–9 Mar.

3 1.00 36–84 38–83 13 Oct.–7 Dec. 10 Mar.–27 Apr.

4 1.00 85–96 84–95 8–19 Dec. 28 Apr.–11 May

5 0.85 100–last harvest 96–last harvest 20–23 Dec. 12–31 May† Values obtained from Zotarelli et al. (2015).‡ TI, irrigation scheduling based on historic evapotranspiration average weather

information; T2, T3, and T4, irrigation scheduling based on real-time and location-specific evapotranspiration-based scheduler at 66, 100, and 150%, respectively.

§ Days after transplanting.


et al., 2009b), soil samples were collected at one sampling depth below root zone (30–45 cm soil depth) to detect nutrient leach-ing. At each sampling depth, soil samples were collected from three locations between plants at the center of the bed with a soil core sampler (2.8 cm internal diameter) at 30, 60, and 90 d after transplanting (DAT) during both seasons. For each sample, anal-yses for soil NO3

––N and NH4+–N content were conducted

with 2 M KCl extracting solution and processed using standard methods (Hanlon et al., 1997). Soil NO3

––N and NH4+–N

(mg L–1) analyses were conducted with a Flow Injection Analyzer (QuikChem 8500, Lachat Co., Milwaukee, WI) at 660 and 520 nm, respectively. The reported values for soil NO3

––N and NH4

+–N content were based on oven-dry soil using Eq. [2]:

( ) ( )( )

1

1

Soil nutrient content (mg kg )

mg L 0.004 L

Mass of ovendry soil used kgR

−

−

=

× [2]

where R is the NO3––N and NH4

+–N readings in mg L–1 from the raw output, and 0.004 is the volume (L) of 2 M KCl extract-ing solution used.

Soil moisture sensors (SDI-12 Drill and Drop Probe; Sentek, Stepney, South Australia) were used throughout the season to monitor the soil movement pattern. These sensors consist of multiple sensing units located at different spots along the length (91 cm) of the probe that allows for multiple readings across soil depth. Each probe was installed vertically into the soil and connected to a radio that powers the probe and stores the data. The radio sends the data to the cloud every 15 min; there-fore, soil moisture condition can be monitored in real-time. In both seasons, two soil moisture sensors were installed per treat-ment midway between two consecutive plants, with moisture readings taken every 15 min at soil depths of 5, 10, 25, 35, and 45 cm. These soil depths were considered appropriate for this study because about 85 to 95% of tomato roots are concentrated within the first 30 cm of soil depth (Zotarelli et al., 2008b). For the purpose of this study, the sensors were not calibrated to the soil condition because the sensors were used only to monitor the pattern of soil water movement (wetting front) across soil depths, and therefore calibration was not necessary.

Crop Biomass estimation and Nutrient AccumulationTissue samplings consisted of both above-ground (leaves,

stems, and fruits) and below-ground (roots) biomass at 30, 60, and 90 DAT except for fruit sampling at 60 and 90 DAT. At every sampling date, a representative plant per plot was selected (Hartz and Bottoms, 2009) and separated into leaves, stems, root, and fruits. The collected root biomass samples considered total root mass by digging out the root of the selected plant on each plot (Ehdaie et al., 2003; Rens et al., 2015). The collected root samples were washed to remove soil particles before drying. Dry biomass weights were obtained by placing the sample in the tissue-drying oven at 65°C for 5 to 7 d.

Total dry biomass from each plot was determined as the sum of the weight for dry biomass samples (leaves, stem, fruits, and roots) for each replicate of each treatment. Dry samples were weighed to determine total dry matter accumulation and then ground to determine total N content. All biomass samples were analyzed for N content using the NA2500 C/N analyzer (Thermoquest CE Instruments) as described by Kadyampakeni (2012). Total N uptake or accumulation was determined as the sum of the product between dry biomass samples (leaves, stem, fruits, and roots) and their respective N concentration.

Tomato Harvest, Yield estimation, and efficienciesTomato harvests of matured fruits were conducted three

times per season. Tomato yield estimation was based on harvested fruits from 15 plants in the middle row of three rows per plot at the mature green stage. Harvested fruits were graded according to the USDA standards as medium (5.72–6.43 cm fruit diameter), large (6.35–7.00 cm), and extra-large (>7.00 cm). Fruits with un-marketable quality were also recorded (USDA, 1997). Marketable yield estimation was based on the fresh weight of fruits with mar-ketable qualities (USDA, 1997). Irrigation water use efficiency (kg m–3) was determined using Eq. [3] as the ratio of the market-able yield (kg ha–1) to the total volume of irrigation water applied (m3 ha–1) for each treatment (Zotarelli et al., 2008b):

( )( )3

Total marketable yield kgiWUE

Total volume of irrigation water applied m=

[3]

Nitrogen recovery was determined using Eq. [4]:

NL+NS+NF+NR REC-N

TNA+NI=

[4]

where NI is the initial soil N content before planting; NL, NS, NF, and NR are the total N uptake accumulated in tomato leaf, stem, fruits, and root, respectively; and TNA is the total N ap-plied for each season.

Data AnalysisThe study used the GLM procedure of SAS Version 9.3

(SAS Institute Inc., Cary, NC) to compare treatments for sig-nificant differences for each variable evaluated. A one-factorial model was developed using a randomized complete block design with irrigation rate as the main effect for soil water and nutrient distributions, total plant biomass, nutrient uptake, yield, and effi-ciencies. Statistical analysis for all variables presented in this study considered all four replicates of each treatment unless otherwise stated. All variables were analyzed separately for each year due to a significant interaction between seasons and DAT. Duncan’s multiple range test was used (a = 0.05) for mean separation when significant differences were observed among treatments.

resULTs AND DIsCUssIONsWeather Condition and Irrigation Water Application

The weather conditions were different during the two sea-sons. Daily average temperatures were higher during the fall sea-


son than the spring season (Fig. 1). Early in the seasons (at the first 30 DAT), daily average temperatures were 26°C during the fall season and 17°C during the spring season. Because ambient temperature is a major factor for plant growth and development (Neild and Seeley, 1977), the lower temperature during the early growth stages of the spring season delayed plant growth by 2 wk and extended the total season length to 17 wk, compared with 15 wk in a warmer fall season.

The total amount of rainfall received during the fall season (280 mm) was greater compared with the spring season (107 mm) (Fig. 2). During the fall season, rainfall was more frequent, espe-cially early in the season. Approximately 180 mm of rainfall (65% of the season total) during the fall season were received at the first 10 DAT; therefore, during the fall season, no irrigation was ap-plied until 5 wk after transplanting. The influence of rainfall on soil moisture level was negligible during the spring season because of the low rain intensity and frequency; hence, irrigation started at transplanting and continued throughout the season.

Compared with the fall season, total irrigation water ap-plied and total number of irrigation events for each treatment were greater during spring because of a longer and drier produc-tion season (Table 4). However, the pattern of irrigation water applied in both seasons was similar for each treatment (Fig. 3). Total irrigation depths increased in the order of T2 < T3 < T1 < T4 for both seasons. Total water savings for T3 were 20 and 17% greater compared with T1 for the fall and spring seasons, respec-tively. Water savings from the SI-based irrigation schedule could be greater than the reported values if the model considered pre-cipitation. The RT-LS model could be designed to skip irrigation (via actionable information or notification to the user), especially after major rainfall event(s). The current version of the SI App-Tomato model (used in this study) does not have this function; however, work is in progress on this and other series of updates to further improve model accuracy and users’ experience.

Although both scheduling methods (HI and SI used for T1 and T3, respectively) are ET based, the difference in the amount of irrigation water scheduled by both methods and the relatively

greater water savings for T3 were attributed to the differences in the assumptions behind the scheduling methods. The HI sched-ule is based on monthly single-value ET data for a Florida region and was derived from historical average ET data. Therefore, the variation in actual water requirement for vegetable crops in Florida could be up to 25% higher or lower than the HI-based irrigation schedule (Zotarelli et al., 2015). Therefore, irrigation scheduling from HI could be in excess of or less than actual plant water re-quirements. Irrigation scheduling using SI was based on RT-LS weather data (Migliaccio et al., 2014); therefore, this scheduling method could improve scheduling accuracy to actual crop water requirements. These results were similar to those reported by Vellidis et al. (2014), where a location-specific irrigation scheduler significantly reduced irrigation water in cotton (Gossypium spp).

sensor-Based soil Moisture DistributionThe soil moisture characteristics of the sandy soil at the

study site indicate limited soil water holding capacity (Table 1). The soil moisture patterns as recorded by the installed soil mois-

Fig. 1. Average daily temperature during fall (2015) and spring (2016) production seasons at the experimental site. Data obtained from Florida Automated Weather Network.

Fig. 2. Cumulative rainfall during fall (2015) and spring (2016) production seasons at the experimental site. Data obtained from the Florida Automated Weather Network.

Table 4. Total seasonal irrigation application time and volume for drip-irrigated tomato production during fall 2015 and spring 2016 seasons in Immokalee, FL.

Treatment† Total applications Total irrigation Difference‡

–––––––––– mm ––––––––––

Fall 2015

T1 231 242.31 40.18

T2 231 134.75 –67.38

T3 231 202.13 0.00

T4 231 302.34 100.21

Spring 2016

T1 301 449.69 66.16

T2 301 255.69 -127.84

T3 301 383.53 0.00

T4 301 575.30 191.77† TI, irrigation scheduling based on historic evapotranspiration

average weather information; T2, T3, and T4, irrigation scheduling based on real-time and location-specifi c evapotranspiration-based scheduler at 66, 100, and 150%, respectively.

‡ Difference in total season irrigation applied with respect to T3.


ture sensors indicated ample soil moisture content throughout the season during the fall (Fig. 4) and the spring (Fig. 5).

Regardless of soil depth and treatments, the pattern of soil moisture levels earlier during the fall season (0–20 DAT) were higher than the observed pattern during the spring season for the same period after transplanting. The high soil moisture con-tent between 0 and 20 DAT during the fall season was attrib-uted to the unusually high amount of rainfall received early in the fall season. At 20 DAT, field capacity was assumed to be the soil moisture content at the end of a rapid soil moisture loss after an irrigation event. Soil moisture levels were similar and close to field capacity across all treatments at the 0- to 15-cm soil depth during the fall and spring seasons. Although soil moisture levels for T2 and T3 dropped below field capacity toward the end of the spring season (Fig. 5), soil moisture for these treatments dur-ing this period was sufficient to maintain plant physiological ac-tivities. These results suggest that a lower irrigation application rate (up to 66% ETc) than the current crop requirement (T2) for tomato may be sufficient for meeting plant water needs on a sandy soil at this critical soil depth (0–15 cm), where about 70 to 75% of tomato roots are concentrated (Zotarelli et al., 2008b).

During the spring season, average daily soil moisture levels were higher for T1 and T4 compared with T2 and T3 at both the 15- to 35-cm and the 35- to 45-cm soil depths. Similar re-sults were observed during the fall season at the 35- to 45-cm soil depth, where average daily soil moisture content was greater for T1 and T4 compared with T2 and T3. This indicates a greater possibility of water percolation for T1 and T4 beyond the toma-to rooting depth. The general trend of soil moisture level from both seasons suggests that irrigation scheduling from T1 and T4 could be excessively higher than actual water requirement by tomato crop grown on a sandy soil and therefore may present a higher risk of nutrient leaching. This is because excessive water application may result in increased nutrient leaching and con-tamination potentials as well as reduced production efficiencies (Alhammadi and Al-Shrouf, 2013).

effects of soil Water Distribution on Ammonium-N Movement

Ammonium-nitrogen movement pattern in the soil was not consistent in both seasons and across irrigation rates and scheduling methods. During the fall season, there were no differences in soil NH4

+–N concentration across all treat-ments at all sampling dates and depths ex-cept at 90 DAT, when NH4

+–N concen-trations were lower (P = 0.01) for T1 and T4 at D1 and higher (P < 0.0001) at D3 (Table 5). At 30 DAT during the spring sea-son, soil NH4

+–N level at D3 was signifi-cantly greater (P = 0.02) for T1 compared with other treatments (Table 5). However, these differences could not be attributed to treatment effects or leaching due to low

irrigation application rate (required by young growing tomato plants early in the seasons) at this period in the production season. Therefore, these differences could be attributed to soil sampling close to soluble fertilizer (applied as pre-plant) concentrations. At 60 DAT during the spring season, soil NH4

+–N concentrations were lowest for T1 and T2 at D1 and D2 (P = 0.03 and 0.04, re-spectively), but no differences were observed at 90 DAT among treatments across all sampling depths. The observed differences in soil NH4

+–N concentrations during the fall season could have been influenced by higher soil moisture content from high rain-fall because soil NH4

+–N distributions were not different among treatments across all depths at 90 DAT during a drier, more water-demanding spring season.

The observed movement pattern for soil NH4+–N was not

consistent because its movement in the soil could be restricted or limited. Plausible restriction to NH4

+–N movement as ob-served in this study could be due to adsorption and retardation effects and insufficient soil moisture content (from a low-volume drip irrigation system) to cause NH4

+–N displacement in the soil (Kadyampakeni et al., 2014a). Similarly, because soil mois-ture levels were close to field capacity during a major part of the production seasons, it could be expected that most NH4

+–N was adsorbed to soil particles or transformed into NO3

––N via nitrification (Khakural and Alva, 1996).

However, these results contradict those reported by Sato et al. (2009), who found significant movement of NH4

+–N in a similar soil type with tomato production. However, the Sato study used a seepage irrigation system often characterized by a high volume of applied irrigation water (Smajstrla and Locascio, 1996) with repeated raising and lowering of the water table. As a result, the reported soil moisture content by these authors was up to three times (30% volumetric water content) greater than field capacity (10% volumetric water content). Therefore, these contradictions could be due to the differences in irrigation and nutrient management practices associated with seepage and drip irrigation systems. Compared with drip irrigation, irrigation ap-

Fig. 3. Cumulative irrigation depth for open-fi eld fresh-market tomato production during fall 2015 (A) and spring 2016 (B). T1, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-time and location-specifi c evapotranspiration-based scheduler at 66, 100, and 150%, respectively.


plication efficiency in seepage system is low; hence, large volumes of water are required to maintain water near the crop root zone (Smajstrla and Locascio, 1996). Additionally, in seepage irrigation with plastic mulch, total season crop nutrient requirement is ap-plied as a one-time application at pre-plant, compared with a drip system that allows for a more precise and much smaller amount of nutrient application according to crop growth. Therefore, the high volume of irrigation water and/or high rate of fertilizer ap-plications (typical of seepage system) may increase water perco-lation and nutrient displacement in sandy soil compared with a low-volume drip system (Hartz and Hochmuth, 1996; Simonne and Ozores-Hampton 2010b; Zotarelli et al., 2009a).

effects of soil Water Distribution on Nitrate-N Movement

No significant differences were observed among treatments for soil NO3

––N distribution at 30 DAT for all sampling depths in both seasons except at D1 during the spring season where soil NO3

––N content was highest (P = 0.01) for T3 but not signifi-cantly different for T3 and T2 (Table 6). Similar to NH4

+–N distribution at 30 DAT, these differences could not be attributed to treatment effects due to low irrigation water applications at this early growth stage, suggesting either the presence of fertilizer grains in the soil samples taken or a nonuniform distribution of fertilizer in the bed during pre-plant application.

Fig. 4. soil moisture distribution at different soil depths during fall season. error bars represent 1 se of two replications. Horizontal lines represent soil moisture at fi eld capacity. T1, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-time and location-specifi c evapotranspiration-based scheduler at 66, 100, and 150%, respectively.

Fig. 5. soil moisture distribution at different soil depths during spring season. error bars represent 1 se of two replications. Horizontal lines represent soil moisture at fi eld capacity. T1, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-time and location-specifi c evapotranspiration-based scheduler at 66, 100, and 150%, respectively.


At 60 and 90 DAT, soil NO3––N content varied signifi-

cantly among treatments at all sampling depths in both seasons (Table 6). During the fall season, NO3

––N concentration at 60 DAT at D1 was highest (P < 0.0001) for T2 (43.79 kg ha–1) and lowest for T1 and T4 (2.03 and 4.82 kg ha–1, respectively). Similar results were observed at D1 during the spring season where soil NO3

––N concentrations were higher for both T2 and T3 (51.13 and 43.49 kg ha–1, respectively; P = 0.007) com-pared with T1 and T4 (13.66 and 4.03 kg ha–1, respectively).

Soil NO3––N distributions followed a similar pattern at D2

during both seasons, where NO3––N levels were significantly

greater for treatments with lower irrigation rates (T2 and T3) compared with higher rates (T1 and T4).

Although the volume of irrigation water applied in both sea-sons was higher for T4 compared with T1, soil NO3

––N levels at 60 DAT for D3 (below the root zone) was greater for T1 (32.8 and 21.03 kg ha–1 for the fall and spring seasons, respectively) compared with T4 (4.74 and 11.14 kg ha–1 for the fall and

Table 5. effects of irrigation rates and scheduling methods on soil NH4+–N distribution.

DAT† Treatment‡

NH4+–N distribution, fall 2015 NH4

+–N distribution, spring 2016

0–15 cm 15–30 cm 30–45 cm 0–15 cm 15–30 cm 30–45 cm

––––––––––––––––––––––––––––––––––––––––––– kg ha–1 –––––––––––––––––––––––––––––––––––––––––––

30

T1 5.10 3.93 3.45 39.36 36.30 39.50a§

T2 9.50 3.24 6.64 37.91 20.30 18.67b

T3 6.04 2.85 5.94 43.97 32.22 26.50ab

T4 4.80 2.80 4.53 37.91 28.70 19.94b

P value 0.77 0.49 0.20 0.98 0.39 0.02

60

T1 5.58 5.58 5.40 2.55b 6.21a 5.54

T2 5.41 5.26 5.28 6.81a 4.62ab 5.71

T3 6.55 5.58 5.22 5.31ab 5.78a 8.88

T4 5.41 9.88 5.57 2.01b 3.21b 5.91

P value 0.71 0.32 0.96 0.03 0.04 0.49

90

T1 1.95b 1.68b 3.81b 3.73 4.68 5.50

T2 6.81a 2.22b 1.57c 9.17 4.91 3.75

T3 6.87a 4.04a 1.77c 9.31 4.37 5.80

T4 2.53b 3.96a 6.06a 2.16 4.33 7.50

P value 0.01 0.007 <0.0001 0.18 0.99 0.64† Days after transplanting.‡ TI, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-

time and location-specific evapotranspiration-based scheduler at 66, 100, and 150%, respectively.§ Means followed by different letters are significantly different at α = 0.05. Data were analyzed by DAT and depth for each season. Duncan’s

multiple range test analysis was used as mean separation when significant differences were observed.

Table 6. effects of irrigation rates and scheduling methods on soil NO3-–N distribution.

DAT† Treatment‡

NO3––N distribution, fall 2015 NO3

––N distribution, spring 2016

1–15 cm 15–30 cm 30–45 cm 1–15 cm 15–30 cm 30–45 cm

–––––––––––––––––––– kg ha–1 ––––––––––––––––––––

30

T1 49.8 16.50 18.94 29.07ab§ 27.40 7.23

T2 66.63 18.90 26.25 36.35ab 12.66 5.85

T3 67.70 19.63 22.59 53.06a 28.37 5.78

T4 54.71 27.99 30.39 12.89b 9.71 10.68

P value 0.64 0.19 0.30 0.01 0.10 0.73

60

T1 6.93b 8.13b 32.8a 4.56b 22.81a 21.03a

T2 26.91a 20.00a 5.61b 27.33a 8.58b 3.71c

T3 14.68ab 25.86a 4.38b 28.19a 11.81b 2.44c

T4 6.37b 4.73b 4.74b 1.25b 4.34b 11.14b

P value 0.01 0.03 0.0004 0.03 0.002 <0.0001

90

T1 2.03c 2.36c 8.70b 13.66b 13.18ab 30.73a

T2 43.79a 39.98a 3.59c 51.13a 25.82a 17.07b

T3 28.24b 18.01b 2.69c 43.49a 25.52a 15.67b

T4 4.82c 2.17c 16.71a 4.03b 6.27b 37.25a

P value <0.0001 0.0004 0.0003 0.007 0.02 0.005† Days after transplanting.‡ TI, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-

time and location-specific evapotranspiration -based scheduler at 66, 100, and 150%, respectively.§ Means followed by different letters are significantly different at α = 0.05. Data were analyzed by DAT and depth for each season. Duncan’s

multiple range test analysis was used as mean separation when significant differences were observed.


spring seasons, respectively). These results suggest that greater irrigation application volume as scheduled by T4 (compared with T1) could further increase NO3

––N leaching into a deeper soil layer even beyond D3. This is because in soils, particularly sandy soils, NO3

––N moves with the soil wetting front, and the greater the irrigation water application the further soil NO3

––N tends to move away from the root zone (Li et al., 2003).

The general pattern of soil NO3––N distribution at 90

DAT was consistent with that of 60 DAT in both seasons where NO3

––N levels were greater in the root zone (0- to 30-cm soil depth [Zotarelli et al., 2008b]) for lower irrigation application volumes (T2 and T3) compared with higher irrigation volumes (T1 and T4). The observed NO3

––N movement pattern at 60 and 90 DAT suggests an increase in leaching potential for T1 and T4 in both seasons, whereas irrigation application from T2 and T3 maintained NO3

––N concentrations within tomato rooting depths. Compared with T3, average soil NO3

––N levels (60 and 90 DAT average) at the root zone were 79 and 80% lower for T1 and T4, respectively, during the fall and 48 and 85%, respectively, during the spring. The results from this study agree with those reported in the literature where NO3

––N leaching increased with an increase in irrigation volume or soils moisture content in tomato (Wang and Xing, 2016; Zotarelli et al., 2007) and other vegetable crops (Guimera et al., 1995; Zotarelli et al., 2007) with a drip irrigation system. However, other than the direct effects of irrigation application volume on NO3

––N leaching, low organic matter in the topsoil (typical of the experimental location) could have contributed to the increase in downward movement of soil NO3

––N as irrigation volume increased (Zotarelli et al., 2007). Generally, the presented soil NO3

––N values are relatively high-er than most values reported in the literature for drip-irrigated

tomato. This could be due to the nitrification of the NH4–based N source (42% of the total N applied) from the applied fertilizer. However, the presented values are similar to those presented by Brewer et al. (2018) at the same study location and under similar irrigation and nutrient rates during the spring season of 2014.

N recovery and Irrigation Water-Use efficienciesAlthough in both seasons there were no differences among

treatments in tomato total biomass accumulation (Table 7), to-tal marketable yield (TMY) was significantly affected (fall: P = 0.03; spring: P = 0.01) by irrigation rates (Table 8). In both sea-sons, TMY was highest for T3 compared with other treatments except during the fall season, where TMY was similar for T2 and T3 (Ayankojo et al., 2018). Although water applications for T2 were lower than actual crop requirements, a wetter fall season could have influenced T2 crop performance. Compared with T3, the lower yield for T1 and T4 was attributed to NO3

––N leach-ing that resulted from higher irrigation water applications.

Nitrogen recovery ranged from 65 to 86% during the fall season and from 37 to 61% during the spring season (Table 6). Generally, REC-N was lower during the spring season compared with fall season due to lower total N accumulation (Table 7) and yield (Table 8). Nitrogen recovery was significantly greater for T3 compared with other treatments in both seasons (fall: P = 0.01; spring: P = 0.04), except in the spring season where T2 and T3 were similar at 56 and 61%, respectively. Lower REC-N for T1 and T4 in both seasons was attributed to NO3

––N leaching below the root zone due to a higher daily volume of irrigation water applications.

Irrigation water use efficiency for tomato production in both seasons was affected by irrigation application rates (Table 8).

Table 7. effects of irrigation rates and scheduling methods on tomato biomass and nitrogen accumulation and recovery (reC-N).

Treatment†

Fall 2015 spring 2016

Total biomass Total N accumulation reC-N Total biomass Total N accumulation reC-N

–––––––––– kg ha–1 –––––––––– % –––––––––– kg ha–1 –––––––––– %

T1 2778.64 146.31b‡ 65.32b 2618.36 83.82c 37.42b

T2 3619.74 149.54b 66.76b 2806.01 125.94ab 56.22ab

T3 3300.33 193.45a 86.36a 3079.66 136.41a 60.90a

T4 2816.68 155.71b 69.51b 2642.90 96.92bc 43.27b

P value 0.15 0.02 0.01 0.52 0.02 0.04† TI, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-

time and location-specific evapotranspiration-based scheduler at 66, 100, and 150%, respectively.‡ Means followed by different letters are significantly different at α = 0.05.

Table 8. effects of irrigation rates and scheduling methods on tomato yield and irrigation water-use efficiency (iWUe).

Treatment†

Fall 2015 spring 2016

Total marketable yield Total irrigation water applied iWUe Total marketable yield Total irrigation water applied iWUe

Mg ha–1 m3 ha–1 kg m–3 Mg ha–1 m3 ha–1 kg m–3

T1 70.16b‡ 2364.64 29.67c 65.03b 3742.37 17.38b

T2 87.96ab 1222.88 71.93a 57.22b 2090.30 27.37a

T3 95.08a 1852.85 51.32b 81.56a 3135.45 26.01a

T4 70.99b 2779.27 25.54c 66.22b 4703.17 14.08b

P value 0.03 – <0.001 0.01 – <0.001† TI, irrigation scheduling based on historic evapotranspiration average weather information; T2, T3, and T4, irrigation scheduling based on real-

time and location-specific evapotranspiration-based scheduler at 66, 100, and 150%, respectively.‡ Means followed by different letters are significantly different at α = 0.05.


Irrigation water use efficiency increased in the order of T2 > T3 > T1 = T4 at 72, 51, 30, and 26%, respectively, during fall (P < 0.001) and T2 = T3 > T1 = T4 at 27, 26, 17, and 14%, respec-tively, during spring (P < 0.001). Lower total irrigation water ap-plied during the fall season (due to a more frequent rainfall early in the season) and higher yield increased iWUE compared with the spring season. These results were similar to those reported by Miller et al. (2018b), where iWUE was greater for an RT-LS scheduler compared with soil moisture sensor–based irrigation schedule in tomato. Reported iWUE rates in this study were sim-ilar to those reported for tomato in the literature (Djidonou et al., 2013; Zotarelli et al., 2009b). Generally, iWUE and REC-N were greater for an irrigation schedule based on real-time and location-specific weather information compared with the corre-sponding schedule based on historic weather averages. Therefore, these results suggest that irrigation scheduling using real-time and location-specific ET data can improve iWUE and REC-N com-pared with a schedule based on historical ET averages.

CONCLUsIONsThis study showed that, regardless of season, irrigation

scheduling from the RT-LS scheduler increased water savings by at least 17% compared with scheduling based on historic ET averages. Greater REC-N and iWUE efficiencies for the for-mer suggest higher levels of irrigation scheduling accuracy with greater water-saving potential in tomato production. Because ir-rigation schedules from historic ET data are fairly similar (for a particular period of the year; e.g., a month) over several years, this scheduling method tends to be easily managed, and hence growers may adopt to use their experience over the years. The RT-LS model, on the other hand, allows for actionable notifica-tion function based on the actual weather information. In addi-tion to increasing water savings, this would eliminate the need for a daily check of model input and output data and thereby could improve the user’s experience and ease scheduling.

According to the observed results in this study, the effects of irrigation treatments on soil NH4

+–N distributions were not consistent across soil depths and DAT; therefore, no specific pattern of movement could be established. However, the effects on soil NO3

––N levels were more evident across all sampling depths. At a 100% irrigation application rate, soil NO3

––N levels were greater within the tomato root zone (upper 30 cm soil depth) for the RT-LS–based irrigation schedule compared with a schedule based on historic ET data. Therefore, irrigation scheduling based on RT-LS weather information could enhance scheduling accuracy and reduce potential nutrient leaching in tomato production on sandy soils.

reFereNCesAlhammadi, M.S., and A.M. Al-Shrouf. 2013. Irrigation of sandy soils, basics

and scheduling. In: A. Goyal and M. Asif, editors, Crop production. InTechOpen, London, UK. p. 49–67.

Ayankojo, I.T., K.T. Morgan, M. Ozores-Hampton, and K.W. Migliaccio. 2018. Effects of real-time and location-specific drip irrigation scheduling on water use, plant growth, nutrient accumulation and yield of Florida Fresh-market tomato. HortScience 53(9):1372–1378. doi:10.21273/

HORTSCI13183-18Brewer, M.T., K.T. Morgan, L. Zotarelli, C.D. Stanley, and D. Kadyampakeni.

2018. Effect of drip irrigation and nitrogen, phosphorus and potassium application rates on tomato biomass accumulation, nutrient content, yield and soil nutrient status. Hortic. J. 5. doi:10.4172/2376-0354.1000227

Costa, J.M., and E. Heuvelink. 2005. The tomato crop and industry. In: E. Heuvelink, editor, Tomatoes. Centre for Agriculture and Bioscience International, Wallingford, UK. p. 1–20.

Djidonou, D., X. Zhao, E.H. Simonne, K.E. Koch, and J.E. Erickson. 2013. Yield, water-, and nitrogen-use efficiency in field-grown grafted tomatoes. HortScience 48(4):485–492. doi:10.21273/HORTSCI.48.4.485

Ehdaie, B., R.W. Whitkus, and J.G. Waines. 2003. Root biomass, water-use efficiency, and performance of wheat-rye translocations of chromosomes 1 and 2 in spring bread wheat ‘Pavon’. Crop Sci. 43:710–717. doi:10.2135/cropsci2003.0710

FDACS. 2017. Florida agriculture overview and statistics. www.freshfromflorida.com/Agriculture-Industr y/Florida-Agriculture- Over view-and-Statistics (accessed 24 May 2019).

Guimera, J., O. Marfa, L. Candela, and L. Serrano. 1995. Nitrate leaching and strawberry production under drip irrigation management. Agric. Ecosyst. Environ. 56:121–135. doi:10.1016/0167-8809(95)00620-6

Hanlon, E.A., J.S. Gonzalez, and J.M. Battos. 1997. Institute of Food and agricultural Sciences (IFAS) Extension Soil Testing laboratory (ESTL) and Analytical Research Laboratory (ARL) chemical procedures and training manual. Circ. 812. Univ. of Florida, Gainesville.

Hartz, T.K., and T.G. Bottoms. 2009. Nitrogen requirements for drip irrigated processing tomatoes. HortScience 44(7):1988–1993. doi:10.21273/HORTSCI.44.7.1988

Hartz, T.K., and G.J. Hochmuth. 1996. Fertility management of drip irrigation vegetables. HortTech. 6:168–172. doi:10.21273/HORTTECH.6.3.168

Hochmuth, G., and E. Hanlon. 2014. A summary of N, P, and K research with tomato in Florida. Fla. Coop. Ext. Serv. SL 355. Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville.

Kadyampakeni, D.M. 2012. Citrus advanced production system: Understanding water and NPK uptake and leaching in Florida flatwoods and ridge soils. Ph.D. diss. Univ. of Florida, Gainesville.

Kadyampakeni, D.M., K.T. Morgan, A.W. Schumann, P. Nkedi-Kizza, and K. Mahmoud. 2014a. Ammonium and nitrate distribution in soil using drip and micro sprinkler irrigation for citrus production. Soil Sci. Soc. Am. J. 78:645–654. doi:10.2136/sssaj2013.07.0319

Kadyampakeni, D.M., K.T. Morgan, A.W. Schumann, P. Nkedi-Kizza, and T.A. Obreza. 2014b. Water use in drip- and microsprinkler-irrigated citrus trees. Soil Sci. Soc. Am. J. 78:1351–1361. doi:10.2136/sssaj2014.02.0054

Khakural, B.R., and A.K. Alva. 1996. Transformation of urea and ammonium nitrate in an entisol and a spodosol under citrus production. Commun. Soil Sci. Plant Anal. 27:3045–3057. doi:10.1080/00103629609369760

Li, J.S., J.J. Zhang, and L. Ren. 2003. Water and nitrogen distribution as affected by fertigation of ammonium nitrate from a point source. Irrig. Sci. 22:19–30.

Liu, G., E.R. Simonne, K.T. Morgan, G.J. Hochmuth, M. Ozores-Hampton, and S. Agehara. 2016. Fertilizer management for vegetable production in Florida. In: J.S. Freeman, G.E. Vallad, and P.J. Dittmar, editors, Vegetable production handbook for Florida. Fla. Inst. of Food and Agric. Sci. Univ. of Florida, Gainesville.

Liu, G., L. Zotarelli, Y. Li, D. Dinkins, Q. Wang, and M. Ozores-Hampton. 2014. Controlled-released and slow-released fertilizers as nutrient management tools. Fla. Coop. Ext. Serv. HS 1255. Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville.

Migliaccio, K.W., C. Fraisse, G. Vellidis, K.T. Morgan, D. Rowland, L. Zotarelli, and J.H. Debastiani Andreis. 2014. Smart Irrigation Apps. http://en.envirocitiesmag.com/articles/issue-8/Art3.pdf (accessed 3 Jan. 2018).

Migliaccio, K.W., K.T. Morgan, G. Vellidis, L. Zotarelli, C. Fraisse, B.A. Zurweller, et al. 2016. Smartphone apps for irrigation scheduling. ASABE 59(1):291–301. doi:10.13031/trans.59.11158

Migliaccio, K.W., B. Schaffer, J.H. Crane, and F. Davies. 2010. Evapotranspiration and soil water sensor irrigation scheduling methods for papaya production in south Florida. Agric. Water Manage. 97:1452–1460. doi:10.1016/j.agwat.2010.04.012

Miller L., G. Vellidis, and T. Coolong. 2018a. Comparing a smartphone irrigation scheduling application with water balance and soil moisture-based irrigation methods: Part II. Plasticulture-grown watermelon.


HortTech. 28(3).Miller, L., G. Vellidis, O. Mohawesh, and T. Coolong. 2018b. Comparing a

smartphone irrigation scheduling application with water balance and soil moisture-based irrigation methods: Part I-Plasticulture-grown tomato. Horttechnology 28(3).

Monte, J.A., D.F. de Carvalho, L.O. Medici, L.D.B. da Silva, and C. Pimentel. 2013. Growth analysis and yield of tomato crop under different irrigation depths. Rev. Bras. Eng. Agric. Ambient. 17(9):926–931. doi:10.1590/S1415-43662013000900003

Nangare, D.D., Y. Singh, P.S. Mumar, and P.S. Minhas. 2016. Growth, fruit yield and quality of tomato (Lycopersicon esculentum Mill.) as affected by deficit irrigation on phenological basis. Agric. Water Manage. 171:73–79. doi:10.1016/j.agwat.2016.03.016

Neild, R.E., and W.M. Seeley. 1977. Growing degree days predictions for corn and sorghum development and some applications to crop production in Nebraska. Res. Bull. 280. Nebraska Agric. Exp. Stn., Lincoln.

Rens, L.R., L. Zotarelli, D.J. Cantliffe, P.J. Stoffella, D. Gergela, and D. Fourman. 2015. Biomass accumulation, marketable yield and quality of Atlantic potato in response to nitrogen. Agron. J. 107:931–942. doi:10.2134/agronj14.0408

Sainju, U.M., B.P. Singh, S. Rahman, and V.R. Reddy. 1999. Soil nitrate-nitrogen under tomato following tillage, cover cropping, and nitrogen fertilization. J. Environ. Qual. 28:1837–1844. doi:10.2134/jeq1999.00472425002800060021x

Sanjay S., G.S. Hendricks, F.M. Roka, and T.O. Obreza. 2014. Water and nitrogen BMPs for tomato and watermelon: Water quality and economics. Fla. Coop. Ext. Serv. AE 503. Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville.

Sato, S., K.T. Morgan, M. Ozores-Hampton, and E.H. Simonne. 2009. Spatial and temporal distribution in sandy soils with seepage irrigation: I. Ammonium and nitrate. Soil Sci. Soc. Am. J. 73:1044–1052. doi:10.2136/sssaj2008.0206

Simonne, E.H., and M. Ozores-Hampton. 2010a. Water management for tomato. In: M. Ozores-Hampton and C. Snodgrass, editors, Proc. Florida Tomato Institute, Naples FL. 8 Sept. 2010. Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville. p. 34–37.

Simonne, E.H., and M. Ozores-Hampton. 2010b. Fertilizer and nutrient management for tomato. In: M. Ozores-Hampton and C. Snodgrass, editors, Proc. Florida Tomato Institute, Naples, FL. 8 Sept. 2010. Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville. p. 37–40.

Smajstrla, A.G., and S.J. Locascio. 1996. Automated drip irrigation scheduling of tomato using tensiometers. In: C.R. Camp, E.J. Sadler, and R.E. Yoder, editors, Evapotranspiration and irrigation scheduling. Am. Soc. Agr. Eng., San Antonio, TX. p. 845–850.

USDA. 1997. Standard for grade of fresh tomatoes. USDA Agricultural Marketing Service, Washington, DC.

USDA. 2017. 2016 State agriculture overview Florida. USDA, National

Agricultural Statistics Service, Washington, DC. https://www.nass.usda.gov/Quick_Stats/Ag_Overview/stateOverview.php?state=FLORIDA (accessed 11 Apr. 2017).

USDA–NRCS. 2015. Web soil survey. USDA, Washington, DC. https://websoilsurvey.sc.egov.usda.gov/App/WebSoilSurvey.aspx (accessed 3 Jan. 2016).

Vellidis, G., V. Liakos, J.H. Andreis, C.D. Perry, W.M. Porter, E.M. Barnes, et al. 2016. Developing and assessment of a smartphone application for irrigation scheduling in cotton. Comput. Electron. Agric. 127:249–259. doi:10.1016/j.compag.2016.06.021

Vellidis, G., V. Liakos, C. Perry, M. Tucker, G. Collins, J. Snider, et al. 2014. A smartphone app for scheduling irrigation on cotton. In: S. Boyd, M. Huffman, and B. Robertson, editors, Proc. of the 2014 Beltwide Cotton Conference, New Orleans, LA. 5–9 Jan. 2014. National Cotton Council, Memphis, TN. p 1–11.

Wang, X., and Y. Xing. 2016. Effects of irrigation and nitrogen fertilizer input levels on soil NO3

--N content and vertical distribution in greenhouse tomato (Lycopersicum esculentus Mill). Scientifica (Cairo) 2016:59710915.

Zhang, T.Q., K. Liu, C.S. Tan, J. Warner, and Y.T. Wang. 2011. Processing tomato nitrogen and soil residual nitrogen as influenced by nitrogen and phosphorus addition with crip-fertigation. Soil Sci. Soc. Am. J. 75:738–745. doi:10.2136/sssaj2009.0365

Zotarelli, L., M.D. Dukes, E.H. Liu, G.D. Simonne, and S. Agehara. 2015. Principles and practices of irrigation management for vegetable. In: J.H. Freeman, G.E. Vallad, and P.J. Dittmar, editors, Vegetable production handbook of Florida. Univ. of Florida Inst. of Food and Agri. Sci., Gainesville. p. 11–18.

Zotarelli, L., M.D. Dukes, and R. Munoz-Carpena. 2009a. Soil water distribution and nitrate leaching of drip irrigation controlled by soil moisture sensors. Proc. Estudio en la Zona no Saturada del Suelo, Vol. IX. Univ. of Florida Dep. of Agricultural and Biological Engineering, Gainesville.

Zotarelli, L., M.D. Dukes, J.M.S. Scholberg, T. Hanselman, K. Femminella, and R. Muñoz-Carpena. 2008a. Nitrogen and water use efficiency of squash for a plastic mulch bed system on a sandy soil, as affected by nitrogen rate and irrigation method. Sci. Hortic. (Amsterdam) 116:8–16. doi:10.1016/j.scienta.2007.10.029

Zotarelli, L., M.D. Dukes, J.M.S. Scholberg, E. Muñoz-Carpena, and J. Icerman. 2009b. Tomato nitrogen accumulation and fertilizer use efficiency on a sandy soil, as affected by nitrogen rate and irrigation scheduling. Agric. Water Manage. 96:1247–1258. doi:10.1016/j.agwat.2009.03.019

Zotarelli, L., M. Johannes, M. Scholberg, M.D. Dukes, and E. Muñoz-Carpena. 2007. Monitoring of nitrate leaching in sandy soils: Comparison of three methods. J. Environ. Qual. 36:953–962. doi:10.2134/jeq2006.0292

Zotarelli, L., J.M. Scholberg, M.D. Dukes, R. Muñoz-Carpena, and J. Icerman. 2008b. Tomato yield, biomass accumulation, root distribution and irrigation water use efficiency on a sandy soil, as affected by nitrogen rate and irrigation scheduling. Agric. Water Manage. 96:23–34. doi:10.1016/j.agwat.2008.06.007

soil & water management & conservation evaluation of soil ......soil science society of...

Documents