Biologia, Bratislava, 62/5: 557—561, 2007Section BotanyDOI: 10.2478/s11756-007-0110-8

Root effects on soil water and hydraulic properties

Horst H. Gerke1 & Rolf O. Kuchenbuch2

1Soil Landscape Research Institute, Leibniz-Centre for Agricultural Landscape Research (ZALF) Eberswalder Str. 84,D-15374 Muncheberg, Germany; e-mail: hgerke@zalf.de2University of Rostock, Justus-von-Liebig Weg 6, D-18059 Rostock, Germany; e-mail: rolf.kuchenbuch@uni-rostock.de

Abstract: Plants can affect soil moisture and the soil hydraulic properties both directly by root water uptake and indi-rectly by modifying the soil structure. Furthermore, water in plant roots is mostly neglected when studying soil hydraulicproperties. In this contribution, we analyze effects of the moisture content inside roots as compared to bulk soil moisturecontents and speculate on implications of non-capillary-bound root water for determination of soil moisture and calibrationof soil hydraulic properties.In a field crop of maize (Zea mays) of 75 cm row spacing, we sampled the total soil volumes of 0.7 m × 0.4 m and 0.3 m deepplots at the time of tasseling. For each of the 84 soil cubes of 10 cm edge length, root mass and length as well as moisturecontent and soil bulk density were determined. Roots were separated in 3 size classes for which a mean root porosity of0.82 was obtained from the relation between root dry mass density and root bulk density using pycnometers. The spatiallydistributed fractions of root water contents were compared with those of the water in capillary pores of the soil matrix.Water inside roots was mostly below 2–5% of total soil water content; however, locally near the plant rows it was up to20%. The results suggest that soil moisture in roots should be separately considered. Upon drying, the relation betweenthe soil and root water may change towards water remaining in roots. Relations depend especially on soil water retentionproperties, growth stages, and root distributions. Gravimetric soil water content measurement could be misleading andTDR probes providing an integrated signal are difficult to interpret. Root effects should be more intensively studied forimproved field soil water balance calculations.

Key words: maize roots; root volume; root water content; spatial distribution; soil moisture; hydraulic properties

Introduction

Plants can affect soil moisture distributions and the soilhydraulic properties either directly by root water up-take (e.g., Feddes et al. 1988; Zhuang et al. 2001) andaccumulating water inside the root biomass, or moreindirectly by modifying the soil pore structure throughthe growing root system (e.g., Angers & Caron 1998;Rasse et al. 2000; Kodesova et al. 2006). In modelsdescribing water movement in plants (e.g., Thornley1996), water in soil, root, and shot is separately consid-ered. In soil water flow models (e.g., Hillel 1998) basedon Richards equation, water movement is describedonly in capillary pores and plant effects on soil waterare mostly considered in form of a sink term accountingfor root water uptake in single or dual porosity systems(e.g., Roose & Fowler 2004; Zumr et al. 2006). In addi-tion to specific plant water stress function coefficients,soil hydraulic properties, root distributions, and fieldsoil moisture data are required for model development.However, complex feedback reactions and impacts ofroot systems on spatial and temporal variability of thestate variables (i.e., soil moisture, suction head) (e.g.,

Hupet & Vanclooster 2005) and hydraulic parameterssuch as increased hydraulic conductivity and preferen-tial flow in stable biopores of plant roots (e.g., Mitchellet al. 1995) are mostly neglected. The methods to de-termine soil water contents are affected by water insideroots, for instance time-domain reflectometry (TDR)is misinterpreted mostly pronounced at low water con-tents (Mojid & Cho 2004).The objective of this study was to separately deter-

mine moisture content in the roots as compared to thebulk soil moisture content and to speculate on the rel-evance for soil moisture determination techniques andinterpretation of soil hydraulic properties.

Material and methods

For the experiments, we selected a field site near the villageof Paulinenaue located in the lowlands of the river Havel,about 50 km northwest of the city of Berlin, Germany. Soiltype was Mollic Gleysol (FAO, 2006) with 30–cm deep cul-tivated organic-rich sandy topsoil originating from a formerdegraded shallow fen. In a field crop of maize (Zea maysL.) of 75 cm row spacing, the total soil volume of a 0.7 mlong, 0.4 m wide plot, covering the stand of about 6–7 maize

* Presented at the International Conference on Bioclimatology and Natural Hazards, Poľana nad Detvou, Slovakia, 17–20September 2007.

c©2007 Institute of Botany, Slovak Academy of Sciences

558 H. H. Gerke & R. O. Kuchenbuch

Fig. 1. Soil block sampling of 1 L volumes from Gleysol top-soil horizon at Paulinenaue. Cubic soil blocks are of 10 cm edgelength; seven blocks were sampled perpendicular to 75 cm Maizeplant rows; four blocks in direction of rows; three depths.

plants, was sampled down to 0.3 m depth in form of cube-shaped soil blocks of 10 cm edge length (Fig. 1) at the timeof tasseling (i.e., flowering stage, on July 10, 2003).

Immediately after sampling, the field-moist mass, M sf ,

was determined for each of the 84 soil samples of 10 cm edgelength before they were air-dried and stored. Soil dry mass,M sd, was obtained from sub-samples that were oven-dried

at 105◦C for 24 h. After submerging the remaining soil inwater for 24 h, re-hydrated roots were washed out and ‘fresh’root mass, M r

f , was determined for the radii-classes ‘coarse’(>0.5 mm), ‘medium’ (0.2–0.5 mm), and ‘fine’ (<0.2 mm).Root samples were stored in alcohol until total root lengthwas determined using the line-intersect method (Tennant1975; Hamblin & Tennant 1987).

The (dry) bulk density of roots, ρrd, was determinedusing a pycnometer called GeoPyc (Micromeritics Ltd.,Monchengladbach) for samples from each of the three sizeclasses (Fig. 2). The automated GeoPyc device measuresthe displaced volume occupied by the roots including theporosity of the roots in a calibrated cylinder filled with fine-textured spheres that behave as a liquid. Soil bulk density,ρsb, was determined from 104

◦C oven-dried root-free massof each 1 L sample. Solid densities of root biomass, ρrs, andsoil particles, ρss, were determined again using pycnometerand Xylol as wetting liquid, while the wet (fresh) densitiesof root biomass, ρrf , were measured using water. In analogyto soil properties and neglecting effects of swelling, a meanroot porosity of n = 0.82 for the larger roots (Table 1) wasobtained from the relation

n = 1− ρrd/ρrs (1)

For each 1 L volume, V s, of the soil blocks, the volumesof roots, V rf , and water θsr were calculated as

V rf =M rf /ρrf (2a)

Table 1. Root density (g cm−3) values for re-hydrated moist roots (wet) ρrf , air-dried root biomass (dry) ρrs, and air-dried intact roots,ρrd, (bulk volume); mean and standard deviation (SD) from three replicate samples and twelve measurements. The mean root watercontent of θr = 0.812, corresponds with the porosity of coarser roots.

wet dry bulk root density

Mean radius mean SD mean SD mean SD

mm g cm−3

Fine <0.2 1.17 0.067 0.88 0.021 0.094 0.021Medium 0.2–0.5 1.13 0.034 0.79 0.028 0.178 0.028Coarse >0.5 1.06 0.031 1.33 0.065 0.241 0.065

coarse medium fine>0.5 mm 0.2–0.5 mm <0.2 mm

Fig. 2. Typical dry maize roots of three mean diameter size classes. The rule bars indicates the scale in centimeters.

Root effects on soil water 559

Table 2. Soil water contents inside Maize root biomass, θsr , related to soil volume for each of the 84 soil blocks (%) calculated fromroot biomass distribution.

In direction of plant rowsSoil depth Soil blocks

1 2 3 4cm Perpendicular

to plant rows Vol. water content, %

0–10 1 0.04 0.11 0.08 0.182 1.32 1.97 1.05 1.523 0.36 0.09 0.18 0.444 0.00 0.12 0.08 0.105 0.14 0.10 0.14 0.086 0.07 0.06 0.06 0.077 0.08 0.05 0.03 0.10

Mean 0.34 0.36 0.23 0.36

10–20 1 0.18 0.09 0.34 0.252 0.27 0.14 0.24 0.603 0.11 0.09 0.13 0.274 0.09 0.09 0.08 0.155 0.08 0.07 0.06 0.106 0.05 0.00 0.05 0.007 0.07 0.08 0.11 0.15

Mean 0.12 0.09 0.15 0.25

20–30 1 0.13 0.10 0.19 0.162 0.14 0.19 0.13 0.153 0.17 0.17 0.05 0.124 0.09 0.13 0.07 0.215 0.27 0.12 0.15 0.116 0.13 0.18 0.10 0.107 0.19 0.17 0.12 0.18

Mean 0.16 0.15 0.12 0.15

θsr =V rfV s

θr (2b)

using the wet root densities, ρrf (Table 1) for each size class;and the volumetric soil water contents inside roots, θsr , usinga single mean value of the root-volume related water contentof the wet (fresh) roots of θr = 0.812 obtained from relationsbetween wet and dry root densities (Table 2). The spatiallydistributed soil water contents inside roots were comparedwith soil water contents in capillary pores of the soil matrixas

θrrel = θsr/(θs − θsr)× 100. (3)

where θs is the total water content obtained by (M sf −M s

d)ρw

using unit water density.Soil water retention curves were obtained for 300 cm3

undisturbed cores using the evaporation method; near-saturated values and soil hydraulic conductivity with thedouble-membrane through-flow method.

Results and discussion

The mean soil bulk densities, ρsb, were ranging betweenvalues of about 1.40 g cm−3 (0–10 cm depth), 1.47 gcm−3 (10–20 cm), and 1.56 g cm−3 (20–30 cm), with atendency towards smaller values close to the plant row.Mean soil water contents, θs, were 0.137 cm3 cm−3 (0–10 cm depth), 0.09 cm3 cm−3 (10–20 cm), and 0.12 cm3

cm−3 (20–30 cm). Densities of wet roots, (re-hydrated)ρrf , were larger for finer (1.17 g cm

−3) as compared withcoarser roots (1.06 g cm−3) while the dry root mass

density, ρrs, was largest for the coarser roots; and thedry root bulk density, ρrd, was increasing with the sizeof the roots (Table 1). Here, standard deviations wererather high indicating that probably unknown contri-butions of mineral particles, which could not completelybe removed because they were strongly attached mainlyto larger roots (Fig. 2), may have affected the individ-ual results. The estimated mean root water content ofθr = 0.812, corresponded with the porosity calculatedfor coarser roots. This value is, of course, a simplifica-tion since the root water content and densities are vari-able and depending on the osmotic state of the cells(Hamza et al. 2007). Note that the fraction of water inroots is maximal here because roots were fully saturatedwith water here while under water stress in the field theroot water fraction could be smaller. For the purpose ofthis study it might be sufficient to compare soil and rootwater contents since no other information is available toutilize the spatially distributed root mass distributions.For both the root biomass and the soil and root

water contents, we found relatively large gradients per-pendicular to Maize rows and in the vertical direction.For the individual soil blocks, the volumetric soil wa-ter contents in roots, θsr , were mostly below 0.2 % butranging between 1 and 2 % close to the Maize plants(Table 2, row #2). Mean soil water contents in rootswas decreasing with depths from 0.23–0.36 % in 0–10cm to 0.12–0.16 % in 20–30 cm.

560 H. H. Gerke & R. O. Kuchenbuch

0

5

10

15

20

25

1 2 3 4 5 6 7103 cm3 blocks perpendicular to Maize row

%

0-10 cm10-20 cm20-30 cm

Fig. 3a. Mean fractions of water in Maize roots relative to thesoil water, θrrel, in three soil depth intervals for the 10-cm cubedblocks perpendicular to plant rows (75 cm rows), plants were herelocated above blocks #2.

0

5

10

15

20

25

1 2 3 4 5 6 7103 cm3 blocks perpendicular to Maize row

%

0-10 cm10-20 cm20-30 cm

Fig. 3b. Volumetric soil water contents including root water(100 θs) for the same blocks as in Fig 3a.

The relative proportion of soil water inside roots(Fig. 3a) was mostly below 5% of the total soil wa-ter content; however, the local values were up to 20%depending on the root biomass and the soil moisture

distributions (Fig. 3b). For the values perpendicular toplant rows and averaged (4 values) in row direction,the contribution of root water to soil water showed aclear trend towards the location of the Maize row thatresulted from a combination of relatively low soil wa-ter content and intense root biomass. Similar effects ofMaize row layout on soil water were reported (e.g., Am-ato & Ritchie 2002; Hupet & Vanclooster 2005).The soil hydraulic properties (Fig. 4) are assumed

to represent the water in mostly capillary soil pores;water retention and unsaturated hydraulic conductiv-ity curves are in standard lab experiments usually de-termined for soil samples excluding root effects. Here,these curves for the Gleysol of the Paulinenaue site al-low calculating largely different matric potential val-ues if water contents with and without root water areused. The example (Fig. 4) indicates that 20-% rootwater content suggests a shift in matric potentials froma value of –10,000 to –1,000 cm and a similar increasein hydraulic conductivity by a factor of 10. The resultssuggest that soil moisture in intensively rooted horizonsshould be separated in water inside living biomass andwater inside the capillary porous soil matrix. Upon dry-ing, the relation between the two soil water fractionsmay change dramatically towards water remaining inbiomass. This is especially true for the more sandy soilswith lower water retention properties and depending ongrowth stage and root distribution. Root effects on soilmoisture measurements are different depending on themethod used for water content determination. Not onlythe location of water content sensors is critical as indi-cated by Hupet & Vanclooster (2005) but also the rootwater effects.Gravimetric soil water content measurement will

be misleading if roots are not excluded and effectsare not considered, and water retention and flow ratesbased on these data will be questionable. Other in-struments such as TDR probes will provide some still

0.1

1

10

100

1000

10000

100000

0 0.1 0.2 0.3 0.4

Volumetric Water Content, θ , cm3 cm-3

-Mat

ric P

oten

tial,

h, c

m

0.0001

0.001

0.01

0.1

1

10

100

Hyd

raul

ic C

ondu

ctiv

ity, K

,cm

/d

1 - 7 cm12-18 cm22-28 cm

h (θ )

no root water

K (θ )

roots included

Fig. 4. Hydraulic soil properties of the Gleysol at Paulinenaue, determined using 300 cm3 cylindrical core samples. Soil water retention,h(θ), and unsaturated hydraulic conductivity, K(θ), as a function of the volumetric soil water content, θ, obtained with the ‘doublemembrane’ (symbols) and the ‘evaporation’ (curves) method are shown for three depths intervals.

Root effects on soil water 561

unknown integrated result (Mojid & Cho 2004) prob-ably not comparable to the capillary matric poten-tial heads determined with tensiometers. Consequently,Mojid & Cho (2004) evaluated a 4-phase dielectric mix-ture model that includes the volume of the roots sep-arately. Water content measurements are critical forvalidation of soil water and root water uptake models(e.g., Gong et al. 2006) especially when spatially dis-tributed flow rates and root water uptake are describedor inverse modelling is applied for parameter estimation(e.g., Sonnleitner et al. 2003).In most cases, the soil water available for root up-

take will be overestimated if root water is not consid-ered, and irrigation demand and plant stresses are in-correctly interpreted. Root effects should be more in-tensively studied before field soil water balance calcu-lations could be improved. Furthermore, the analysis ofthe total soil volume suggests that the root effects onsoil moisture are spatially variable and temporally rel-evant only at different stages in the vegetation period,a topic that requires further quantification.We conclude that soil hydraulic properties cali-

brated or fitted using field moisture data may be mis-leading if the root effects are not considered. Combined3D models of root architecture and flow patterns (e.g.,Darrah et al. 2006) may provide a challenging alterna-tive to include the spatial distribution of root water aswell as for improved testing soil moisture instrumentsin the root zone.

Acknowledgements

We thank Mrs. J. Busse and E. Kruger from the ZALF-Institute of Landscape Matter Dynamics for excellent as-sistance in soil and root analyses, Mr. B. Lobitz from theLVL, Paulinenaue, for field work and sampling assistant,and Mrs. R. Hypscher, ZALF-Institute of Soil LandscapeResearch, Muncheberg, for her skilled assistance in densitymeasurements. Financial support was provided by the Fed-eral Ministry of Food, Agriculture, and Consumer Protec-tion (BMVEL) and by the Ministry for Rural Development,Environment and Consumer Protection (MLUV) of Bran-denburg.

References

Angers D.A. & Caron J. 1998. Plant-induced changes in soil struc-ture: Processes and feedbacks. Biogeochem. 42: 55–72.

Amato M. & Ritchie J.T. 2002. Spatial distribution of roots andwater uptake of maize (Zea mays L.) as affected by soil struc-ture. Soil Sci. Soc. Am. J. 42: 773–780.

Darrah P.R., Jones D.L., Kirk G.J.D. & Roose T. 2006. Modellingthe rhizosphere: A review of methods for ‘upscaling’ to thewhole-plant scale. Eur. J. Soil Sci. 57: 13–25.

FAO, 2006. World reference base for soil resources 2006. A frame-work for international classification, correlation and commu-nication. World Soil Resources Report No. 103, 2006 edition,Rome.

Feddes R.A., Kabat P., van Bakel P.J.T., Bronswijk J.J.B. &Halbertsma J. 1988. Modelling soil water dynamics in theunsaturated zone—state of the art. J. Hydrol. 100: 69–111.

Gong D.Z., Kang S.Z., Zhang L., Du T.S. & Yao L.M. 2006. Atwo-dimensional model of root water uptake for single appletrees and its verification with sap flow and soil water contentmeasurements. Agr. Water Management 83(1–2): 119–129.

Hamblin A. & Tennant D. 1987. Root length density and wa-ter uptake in cereals and grain legumes: How well are theycorrelated? Aust. J. Agric. Res. 38: 513.

Hamza M.A., Anderson S.H. & Aylmore L.A.G. 2007. Computedtomographic evaluation of osmotica on shrinkage and recov-ery of lupin (Lupinus angustifolius L.) and radish (Raphanussativus L.) roots. Environ. Exp. Bot. 59(4): 334–339.

Hillel D. 1998. Environmental Soil Physics. Academic Press, SanDiego.

Hupet F. & Vanclooster A. 2005. Micro-variability of hydrologicalprocesses at the maize row scale: implications for soil watercontent measurements and evapotranspiration estimates. J.Hydrol. 303: 247–270.

Kodesova R., Kodes V., Zigova A. & Simunek J. 2006. Impact ofplant roots and soil organisms on soil micromorphology andhydraulic properties. Biologia 61(Suppl. 19): S339–S343.

Mitchell A.R., Ellsworth T.R. & Meek B.D. 1995. Effects of rootsystems on preferential flow in swelling soil. Commun. SoilSci. Plant Analys. 26(15–16): 2655–2666.

Mojid M.A. & Cho H. 2004. Evaluation of the time-domain re-flectometry (TDR)-measured composite dielectric constant ofroot-mixed soils for estimating soil-water content and rootdensity. J. Hydrol. 295: 263–275.

Rasse D.P., Smucker A.J.M. & Santos D. 2000. Alfalfa root andshoot mulching effects on soil hydraulic properties and aggre-gation. Soil Sci. Soc. Am. J. 64(2): 725–731.

Roose T. & Fowler A.C. 2004. A model for water uptake by plantroots. J. Theoret. Biol. 228(2): 155–171.

Sonnleitner M.A., Abbaspour K.C. & Schulin R. 2003. Hydraulicand transport properties of the plant-soil system estimatedby inverse modelling. Eur. J. Soil Sci. 54(1): 127–138.

Tennant D. 1975. A test of a modified line-intersect method ofestimating root length. J. Ecol. 63(4): 995–1001.

Thornley J.H.M. 1996. Modelling water in crops and plant ecosys-tems. Ann. Bot. 77(4): 261–275.

Zhuang J., Nakayama K., Yu G.R. & Urushisaki T. 2001. Es-timation of root water uptake of maize: an ecophysiologicalperspective. Field Crops Res. 69(4): 201–213.

Zumr D., Dohnal M., Hrncir M., Cislerova M., Vogel T. & DolezalF. 2006. Simulation of soil water dynamics in structuredheavy clay soils with respect to root water uptake. Biologia61(Suppl. 19): S320–S323.

Received April 30, 2007Accepted June 27, 2007