Estimating plant root water uptake using a neural network approach

Water uptake by plant roots is an important process in the hydrological cycle, not only for plant growth but also for the role it plays in shaping microbial community and bringing in physical and biochemical changes to soils. The ability of roots to extract water is determined by combined soil and plant characteristics, and how to model it has been of interest for many years. Most macroscopic models for water uptake operate at soil profile scale under the assumption that the uptake rate depends on root density and soil moisture. Whilst proved appropriate, these models need spatio-temporal root density distributions, which is tedious to measure in situ and prone to uncertainty because of the complexity of root architecture hidden in the opaque soils. As a result, developing alternative methods that do not explicitly need the root density to estimate the root water uptake is practically useful but has not yet been addressed. This paper presents and tests such an approach. The method is based on a neural network model, estimating the water uptake using different types of data that are easy to measure in the field. Sunflower grown in a sandy loam subjected to water stress and salinity was taken as a demonstrating example. The inputs to the neural network model included soil moisture, electrical conductivity of the soil solution, height and diameter of plant shoot, potential evapotranspiration, atmospheric humidity and air temperature. The outputs were the root water uptake rate at different depths in the soil profile. To train and test the model, the root water uptake rate was directly measured based on mass balance and Darcy's law assessed from the measured soil moisture content and soil water matric potential in profiles from the soil surface to a depth of 100 cm. The ‘measured’ root water uptake agreed well with that predicted by the neural network model. The successful performance of the model provides an alternative and more practical way to estimate the root water uptake at field scale.     

Root system parameters determining water uptake of field crops

Summary The distribution of a crop rooting system can be defined by root length density (RD), root length (RL) per soil layer of depth z, sum of root length (SRL) in the soil profile (total root length) or rooting depth (z _r . The combined influence of these root system parameters on water uptake is not well understood. In the present study, field data are evaluated and an attempt is made to relate a daily maximum water uptake rate (WU_max) per unit soil volume as measured in different soil layers of the profile to relevant parameters of the root system. We hypothesize that local uptake rate is at its maximum when neither soil nor root characteristics limit water flow to, and uptake by, roots. Leaf area index and the potential evapotranspiration rate (ET _p ) are also important in determining WU_max, since these quantities influence transpiration and hence total crop water uptake rate. Field studies in Germany and in Western Australia showed that WU_max depends on RD. In general, there was a strong correlation between the maximum water uptake rate of a soil layer (LWU_max) normalized by ET _p and RL normalized by SRL. The quantity LWU_max · ET _p ^-1 was linearly related to (RL/SRL)^1/2. The data show that the single root model will not predict the influence of RD on WU_max correctly under field conditions when water-extracting neighboring roots may cause non-steady-state conditions within the time span of sequential observations. Since the rooting depth z _r was linearly related to (SRL)^1/2, the relation: LWU_max · ET _p ^-1 = f (RL^1/2/z _r ) holds. Furthermore it was found that the maximum specific uptake rate per cm root length UR_max was inversely related to RD^1/2 and to SRL^1/2 or z _r of the profile. Observed high specific uptake rates of shallow rooted crops might be explained not only by their lower RD-values but also by the additional effect of a low z _r . The relations found in this paper are helpful for realistically describing the sink term of dynamic water uptake models.Growing plants extract water from the soil to meet transpiration needs. Rates of transpiration and of water uptake are set by evaporative demand and by plant and soil factors which influence capacity to meet that demand. These factors include crop canopy size and leaf characteristics, root system characteristics and hydraulic properties of the soil and the soil-root interface. Soil and root system properties vary with depth and all factors vary in time, so that parameters related to them require constant updating over a crop season.Dynamic simulation models describe water uptake by root systems under field conditions as a function of soil depth and time. Many of these simulation approaches are based on Gardner's (1960) single root model (Feddes 1981). These simulation procedures follow the assumption that water uptake is proportional to a difference in water potential between the bulk soil and the root surface or the plant interior, to the hydraulic conductivity of the soil-plant system and to the effectiveness of competing roots in water uptake. The effectiveness factor accounts more or less empirically for the influence of various root system parameters on water uptake such as percentage of active roots absorbing water, root surface permeability, root length density determining the distance between neighbouring roots, or total root length and depth of the root system. Such models however, will not always reflect correctly the influence of root system characteristics on water uptake since these assumptions have rarely been tested under field conditions. In many instances, there is better agreement between simulated and measured total water use of plants than between predicted and observed water depletion by roots within individual layers of the soil profile (Alaerts et al. 1985).Water uptake by an expanding root system as a function of depth and time has been studied under field conditions for several crops (listed in Herkelrath et al. 1977a; Feddes 1981; Hamblin 1985). They show that the dynamics of water uptake depend on root length density and the availability of soil water. However, the combined influence of root length density, total root length and rooting depth on the water uptake pattern has not been assessed. An evaluation of root system parameters with respect to soil water extraction should aid our understanding of how roots perform under field conditions and may assist our efforts to formulate the water uptake function of roots in dynamic simulation studies more realistically.The aim of the present investigation is to develop an approach that relates measured water uptake rates to relevant parameters of the root systems. This approach will be confined to situations where water uptake in a soil layer is not restricted by unfavorable soil conditions, such as in wet soil, by insufficient aeration and, in dry soil, by reduced water flow towards roots or by increased contact resistance (Herkelrath et al. 1977b). We will define a maximum water uptake rate WU_max that is neither soil-limited nor appreciably limited by the decreasing permeability of aging roots. This WU_max will be related to relevant root system parameters as they exist when WU_max is observed. Hence, water uptake by roots in a very wet, as well as in a dry soil, has been excluded from consideration.     

Evaluation of the RZWQM-CERES-Maize hybrid model for maize production

The root zone water quality model (RZWQM) was developed primarily for water quality research with a generic plant growth module primarily serving as a sink for plant nitrogen and water uptake. In this study, we coupled the CERES-Maize Version 3.5 crop growth model with RZWQM to provide RZWQM users with the option for selecting a more comprehensive plant growth model. In the hybrid model, RZWQM supplied CERES with daily soil water and nitrogen contents, soil temperature, and potential evapotranspiration, in addition to daily weather data. CERES-Maize supplied RZWQM with daily water and nitrogen uptake, and other plant growth variables (e.g., root distribution and leaf area index). The RZWQM-CERES hybrid model was evaluated with two well-documented experimental datasets distributed with DSSAT (Decision Support System for Agrotechnology Transfer) Version 3.5, which had various nitrogen and irrigation treatments. Simulation results were compared to the original DSSAT-CERES-Maize model. Both models used the same plant cultivar coefficients and the same soil parameters as distributed with DSSAT Version 3.5. The hybrid model provided similar maize prediction in terms of yield, biomass and leaf area index, as the DSSAT-CERES model when the same soil and crop parameters were used. No overall differences were found between the two models based on the paired t test, suggesting successful coupling of the two models. The hybrid model offers RZWQM users access to a rigorous new plant growth model and provides CERES-Maize users with a tool to address soil and water quality issues under different cropping systems.     

Effect of irrigation methods on root development and profile soil water uptake in winter wheat

The 2-year field experiments were carried out to research the effect of different irrigation methods, namely border irrigation, sprinkler irrigation, and surface drip irrigation, on root development and profile water uptake in winter wheat. Results showed that the main root distribution zone moved upward under sprinkler and surface drip irrigation when compared to the traditional border irrigation. Profile root distribution pattern changed with irrigation methods. Soil profile water uptake was correlated to the root system and soil water dynamics. Due to the appropriate soil water and higher root density in the surface soil layer under sprinkler and surface drip irrigation, the main water uptake zone was concentrated in the upper layer. Because of the water deficit in the surface layer under border irrigation, water uptake in 50–100 cm depth was stimulated, which caused the main uptake zone downward. The amount and pattern of root water uptake varied with irrigation methods. This may provide valuable information on the aspect of agricultural management.     

质地和根系深度对水分探头埋设的仿真模拟

利用Hydrus-1D模型模拟不同植物根系深度和不同土壤质地条件下的土壤水分动态与平衡,研究了根系分布深度和质地对控制灌溉土壤水分探头埋设深度的影响,并利用试验进行了验证. 土壤质地和植物根系分布深度对探头埋设深度有显著影响,砂壤土和壤土分别采用高频低灌量和低频率高灌量的方法.浅根系植物（10 cm）在砂壤土条件下探头埋设5 cm深度最佳,但是根系深度增大到30 cm,探头应该埋设到20 cm深度.对壤土而言,利用位于根系1/2至1/3处的探头控制灌溉. 太浅的埋设深度会导致灌溉频率增大,太深的埋设可能造成植物缺水.黏土条件下,结果较为复杂,探头的埋设深度需要田间试验研究. 研究结果表明：针对具体植物,因其需水规律和生理特征的不同,根据植物需水规律来调整探头的控制范围达到高效节水目的.     

Macroscopic scale soil moisture dynamics model for a wheat crop

Summary A numerical soil moisture dynamics model was developed for; wheat crop using either observed or generated root length densities with root sink incorporating diminishing rate of water uptake by plant roots due to decreasing soil moisture in drying cycles and loss of absorptive power of roots due to ageing. The simulated soil moisture contents were overestimated by 6.0 and 9.6% on an overall basis by the model when observed and generated root length densities were used, respectively, in comparison to observed moisture contents. The model using generated root length densities simulated less water uptake in comparison with the model which utilized observed root length densities.     

A simulation model of water dynamics in winter wheat field and its application in a semiarid region

Many models for water flow in cropped soil contain parameters such as rooting density, root permeability, and root water potential. Usually these parameters are chosen by trial-and-error method and direct measurements are difficult and impractical in some cases. This study presents a simulation model capable of analyzing water transport dynamics in a soil–plant–atmosphere continuum (SPAC). This model is developed by combining an existing mathematical model for soil water flow, a modified transpiration model taking into account of the air pressure and diurnal changes of the extinction coefficient of crop canopies, and a new simple model for root water uptake. Using data from lysimeters in a field experiment carried out on a wheat crop, we also developed two new empirical equations for the estimation of total canopy resistance and soil evaporation.We then applied the model for 2 years (1990–1991, 1991–1992) on winter wheat in a semiarid area of northwest China. Required parameters, particularly soil hydraulic and crop parameters, were determined by field and laboratory tests. Outputs from the simulation were in good agreement with the independent field measurements of seasonal changes in soil water content, canopy transpiration, surface evaporation, and root water uptake along the soil profile. In addition, this simulation agreed well with the actual measurements of seasonal crop water consumption and soil water balance among the treatments for different irrigation amounts.     

Effective irrigation uniformity as related to root zone depth

Summary In models used for relating the yield to irrigation uniformity it has been assumed that the spatial distribution of irrigation water, as measured at the soil surface, is indeed the water distribution at any depth throughout the root zone. In the present paper the distribution of infiltrated water within the soil bulk, as determined by an analytic solution of the two-dimensional unsaturated flow equation, did not conform to this assumption. A new alternative definition of irrigation uniformity is proposed under the assumption that water uptake by roots does not affect the flux distribution within the soil profile. In this analysis the spatial distribution of irrigation water flux at the soil surface, which is the upper boundary condition of the flow equation, is assumed to be a sine function. The solution to this problem indicates that there is a damping effect, which increases with soil depth, on the surface flux fluctuations. Furthermore, the actual irrigation uniformity at a given depth below the soil surface depends upon the initial uniformity at the surface and the distance between adjacent water sources. The closer the water sources are to each other, the shallower is the depth needed to damp the oscillations down to a certain level. This may explain why the actual uniformity of drip irrigation is high while the detailed distribution is very nonuniform and on the other hand, why the actual uniformity of sprinkler guns is low while the detailed actual distribution is close to uniform. Two uniformity coefficients are derived in this study: 1. A depth dependent coefficient which is made up of the damping factor that multiplies the flux fluctuations at the soil surface; 2. An effective uniformity coefficient, which is an average of the depth dependent coefficient over a part or the entire root zone. Different degrees of uniformity are expected when water is applied by different irrigation systems having similar uniformity coefficients at the soil surface, but dissimilar distances between the emitters. Assuming that crop yield depends to some extent on the uniformity of water depth actually available to the roots, the yields associated with such irrigation systems will probably also vary.     

Patterns of water withdrawal beneath an irrigated peach orchard on a red-brown earth

Summary Water withdrawal from the soil beneath an irrigated peach orchard is described over depth and time after irrigation for a red-brown earth where the hydraulic properties vary with depth. Relationships between water uptake by roots, root concentration and soil-water suction were explored over protracted drying cycles. In the early stages of drying water uptake by roots was well correlated with root concentration over the profile but, over time, water uptake was redistributed over the root system. Theoretical analysis suggests that poor utilization of water from depth on this soil was associated mainly with low root concentrations and low root (radial) conductance. Practical considerations for improved water management in the root zone of peach orchards on shallow soils are discussed.     

Sugarcane transpiration with shallow water-table: sap flow measurements and modelling

The most common sugarcane variety in the Gharb plain of Morocco (CP 66-345 variety) was grown in a lysimeter in the laboratory. It developed during 6 months with a water-table at 0.7 m below the soil surface. The water-table was then successively maintained with a Mariotte bottle at 0.45, 0.2 and 0.05 m from the soil surface for 21, 31 and 24 days, respectively. Transpiration was measured by Dynamax sap flow sensors. Soil water pressure heads were measured at six different depths; soil hydraulic properties and root density profile were also determined. No transpiration reduction was observed with soil waterlogging. Two different models were used to predict the pattern of root water uptake (RWU) with water-table at 0.45 m below the soil surface. These two models are based on a RWU function used as sink term in the Richards equation. The first model, HYDRUS-2D (Simunek et al., 1996), is based on the α-model RWU (Feddes et al., 1978a) which depends on a reduction function varying according to the soil water pressure head and on the root density. The second model, SIC (Breitkopf and Touzot, 1992) is based on the h_r-model RWU (Whisler and Millington, 1968, Feddes et al., 1974). It is proportional to the difference between soil and root pressure heads, to unsaturated hydraulic conductivity and to root density. Calculated soil water flows from pressure head measurements are compared to predicted pressure heads by the two models. These predictions compare well with the measured values and show that sugarcane roots mainly absorbed water in the water-table. However, while goods predictions were obtained using the actual root density profile with the h_r-model, it was necessary to modify this profile to obtain proper results using the α-model.     

Wheat root growth,grain yield and water uptake as influenced by soil water regime and depth of nitrogen placement in a loamy sand soil

Root growth, grain yield and water uptake by wheat in relation to soil water regime and depth of nitrogen (N) placement were studied in metallic cylinders filled with loamy sand soil. Root-length and -weight densities were greater under irrigated than under unirrigated conditions and they increased with deep placement as compared to surface mixing of fertilizer N. The differences were relatively larger in the deeper than in the upper soil layers and increased during later stages of plant growth. Under non-irrigated conditions, constant water table at 100 cm depth produced maximum root growth in the top 30 cm soil. Water uptake rate increased with increase in root density depending on root age and soil water status. Dry matter accumulation at different stages of plant growth and grain yield varied significantly with moisture regime and depth of N placement. Deep placement of fertilizer N under shallow water table and non-irrigated conditions caused greater root growth, better water utilization and a higher production.     

Crop root system response to irrigation

Summary In the field, root systems develop in response to both endogenous plant design and soil environment. Downward penetration of root systems results primarily from the growth of monocot axes or of dicot taproots; root proliferation at a given depth results from the growth of laterals at that depth. Root length densities generally decline exponentially with depth under well-watered conditions. Root growth rates are partially controlled by soil conditions. Under irrigation, the most critical soil properties for root growth are oxygen diffusion rate, water content and soil strength and all of these properties are inter-related. Under excess irrigation, especially in heavy soils, root growth may be limited by oxygen diffusion rate. Under limited irrigation, root growth may be limited by lack of water or high soil strength. When irrigation maintains wet surface soils, most of the root system is found in the upper part of the profile where the majority of the water is also taken up.     

作物根系吸水率模型的试验研究

以土壤水动力学理论为基础 ,通过取土水洗法测定了有效根量 ,考虑了目前大多用一维根系吸水率模型而忽视的一些因素 ,建立了有效根量密度分布函数。由理论分析和计算 ,得到了较为切合实际的二维根系吸水率模型。     

Comparison of dynamic and static APRI-models to simulate soil water dynamics in a vineyard over the growing season under alternate partial root-zone drip irrigation

In this paper, a two-dimensional (2D) dynamic model of root water uptake was proposed based on soil water dynamic and root dynamic distribution of grapevine, and a function of soil evaporation related to soil water content was defined under alternate partial root-zone drip irrigation (APDI). Then the soil water dynamic model of APDI (dynamic APRI-model) was developed on the basis of the 2D dynamic model of root water uptake and soil evaporation function over the growing season. Soil water dynamic in APDI was respectively simulated by dynamic and static APRI-models. The simulated soil water contents by two models were compared with the measured value. Results showed that values of root-mean-square-error (RMSE) for dynamic APRI-model were less than that of the static APRI-model either in the east side or the west side of grapevine. The average relative error between the simulated and measured value was less than 5% for dynamic APRI-model, indicating that the dynamic APRI-model is better than the static APRI-model in simulating the soil moisture dynamic throughout the growing season under the APDI.     

Simulation of soil moisture profiles for scheduling of irrigations

A mathematical model for simulating soil water content in the root zone was developed by taking into consideration soil physical properties, crop and climatic parameters. The governing differential equation for unsaturated flow of water in the soil was solved numerically using the Crank-Nicholson finite difference technique. The water uptake by plants was simulated by using two different sink functions. The model predictions were in good agreement with field data and thus it is possible to schedule irrigations.     

基于PEST的RZWQM2模型参数优化与验证

根据糯玉米-冬小麦田间喷灌试验不同处理结果,利用独立的自动参数估计软件PEST对RZWQM2模型进行参数优化,并分析了24个模型参数的综合敏感度。通过控制不同观测变量(土壤含水率、土壤氮素含量、作物叶面积指数、产量)模拟差异函数值在目标函数中的比重,优化目标方程,确定模型参数,并用田间试验数据对模型进行验证。结果表明,在不同观测变量的模拟差异函数值最接近条件下,冬小麦出叶间隔特性参数、冬小麦春化作用敏感特性参数及糯玉米出叶间隔特性参数等3个参数对模型整体模拟效果影响最大。相比试错法而言,基于PEST优化的RZWQM2模型能够更准确地模拟糯玉米-冬小麦轮作系统中水分、氮素及作物生长情况。     

Estimating field-measured, plant extractable water from soil properties: beyond statistical models

For effective irrigation management we need to know the water storage capacity of the soil reservoir. Though plant extractable water is best measured in the field, sometimes it is useful to estimate it. Laboratory-derived retention curves do not necessarily reflect field conditions. Statistical models to estimate plant extractable water from other soil properties are restricted by assumptions that are difficult to check, and they can look very complicated. We propose to test a physical-based model that exploits the similarity between the particle size distribution curve and the soil water retention curve. A large data set of soil properties from the USA was used. Detailed particle size fraction data enabled the construction of simulated soil water retention curves for 388 samples. The physical-based model was compared against a statistical model that was derived from a subset of the data base. The statistical model fit the data better than the physical-based model. On the other hand, the statistical model overpredicted the soil water limits of those soils that were not used in the derivation of the statistical model. The strength of the physical model is that it represents a cause and effect relationship between particle size distribution and soil water retention. Also, it is conceptually simple and requires few inputs. The physical model may be improved by considering soil structure and type of clay.     

冬小麦生长土壤中硝态氮动态的集中参数模型模拟

应用包含有限参数的集中参数模型(LPM)来模拟根区NO3-N的浓度。为率定和检验模型,在野外进行6个小区的水氮动态试验,并根据有关文献确定相关参数,然后借助于其中的2个小区实测资料对作物根系吸氮参数进行优化μ。模拟结果表明,模拟值与田间实测值基本吻合,说明所采用的模型及所确定的参数可用于模拟和预测根层土壤中NO3-N的变化规律。最后,对LPM进行了灵敏度分析,结果表明模拟结果对硝化参数k2、根系吸氮参数μ最为灵敏,并且随着尿素水解参数k1、硝化参数k2增大而增大,随着根系吸氮参数μ、反硝化参数k3的增大而减小。     

黄土丘陵区红枣经济林根系分布与土壤水分关系研究

为明确半干旱黄土丘陵区不同年龄无灌溉旱作矮化修剪密植枣林的根系分布范围与其土壤水分的空间关系,利用根钻法测定枣林株间不同深度的根系分布、枣树主干就近位置的根系量,并采用土钻取土和中子仪定位测定结合了解不同年龄的枣林10 m深度的土壤水分。结果表明:随着树龄增加,1、3、5、12 a枣树根系最大深度年平均增值在减小,12 a枣林垂直根系达520 cm。枣树株间100 cm处向下的根系深度较浅,枣林的垂直根系最大和最小值之差先增加后减小,12 a枣林垂直根系之差只有180 cm。研究区枣树株间水平根系在枣林3 a时开始交汇,枣树水平根系延伸无法确定,所得到的水平方向根系实际是枣林多株树汇集的根系。枣林垂直根系对土壤水分的垂直变化作用显著,但矮化修剪密植枣林株间根系深度差异并没有造成土壤水分因此而波动。随着枣树树龄的增加根系深度和土壤水分干层均增加,0~2 m土层的土壤水分年内变化幅度也增加,而且根层范围的土壤水分随着树龄增加在降低,但是土壤干层深度稍大于测得的根系深度。     

浅松覆盖条件下二维根系吸水率模型

运用土壤水动力学方程反求根系吸水率,分析玉米拔节期梯形剖面取根资料,建立有效根密度分布函数,结合土壤剖面水分分布,建立适合浅松覆盖条件下的二维根系吸水率模型,并应用田间实测数据进行验证。结果表明,模拟值与实测值平均相对误差3.75%～7.11%,满足土壤水分动态模拟精度要求。