HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Feng, G. L. Articles by Letey, J. Search for Related Content PubMed Articles by Feng, G. L. Articles by Letey, J. Agricola Articles by Feng, G. L. Articles by Letey, J. Related Collections Plant and Soil Interactions Other Models Plant and Environment Interactions Irrigation

DIVISION S-1—SOIL PHYSICS

Evaluation of a Model for Irrigation Management Under Saline Conditions

I. Effects on Plant Growth

^a Soil and Water Science Unit, Univ. of California, Riverside, CA 92521
^b Inst. of Soils, Water, and Environmental Sci., Volcani Center, ARO, P.O. Box 6, Bet Dagan, Israel

* Corresponding Author (john.letey{at}ucr.edu)

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL DESCRIPTION AND... RESULTS REFERENCES

 
Sustained irrigation agriculture is critical for food and fiber production to support the growing human population. Increasing salinity is a significant factor in many irrigated lands. Knowledge on the proper time and amount of water of various salinities to be applied for optimum yield of a given crop is important. Because of the numerous variables, computer simulation models would be valuable to partially replace expensive field experiments. Simulated relative yields of corn from the ENVIRO-GRO model were compared with experimentally measured yields which had irrigation water electrical conductivities (ECs) of 1.7, 4.0, 5.0, 8.0, and 10.2 dS m^-1 and average irrigation intervals of 3.5-, 7-, 14-, and 21-d treatments. The simulations were run for 5 yr although the field experiment was conducted only 1 yr. The simulated root-weighted average matric and osmotic heads prior to irrigation were evaluated during the year. Depending on treatment, yields were depressed because of either osmotic or matric effects or a combination of the two. The agreement between simulated and measured yields was good for all treatments indicating that the model appropriately accounted for both osmotic and matric effects. On the basis of these results, the ENVIRO-GRO model can be used with confidence in simulating the consequences of irrigation management options under saline conditions. The simulated yields were lower the second and subsequent years compared with the first year, emphasizing the fact that results from 1-yr experiments cannot be used reliably for long-term predictions because the results are highly dependent on the soil conditions at the beginning of the growing season.

Abbreviations: EC, electrical conductivity • V-H, van Genuchten-Hanks

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL DESCRIPTION AND... RESULTS REFERENCES

 
SUSTAINED IRRIGATION AGRICULTURE is critical for food and fiber production to support the growing human population. More than one-third of the total harvest has been produced on irrigated land which amounts to only 17% of the world's cropland (Hillel, 1991, p. 227). Salinity is a significant factor in many of the irrigated lands and must be considered in developing optimum irrigation strategies. The time and amount of irrigation water application of various salinities along with crop selection based on salinity tolerance are some of the management variables. The effects of these management variables on crop yield and the amount and salinity of the water percolating below the root zone are necessary in establishing optimal management practices. Obtaining this information through field research is difficult and expensive because of the number of variables to consider.

The advent of high-speed computers has facilitated the evaluation of multifactor interactions through computer simulation models. However, the utility of this approach requires that the model adequately depict the real situation. Some current models dealing with plant growth, water flow, and agricultural chemical movement such as CERES-MAIZE (Jones and Kiniry, 1986) and GLEAMS (Leonard et al., 1987) do not have a provision for irrigation water salinity and therefore have limited utility where salinity is a factor.

A plant water uptake term is required in models to link a plant to the soil system. Models such as LEACHM (Hutson and Wagenet, 1992) and RZWQ (Great Plains Systems Research, 1992) do include salinity. However, they use a resistor-type plant water uptake term proposed by Nimah and Hanks (1973). The resistor-type water uptake term describes the microscale physics of water flow from the soil to and through the plant roots. However, this type of water uptake function was shown to be insensitive to salinity and generally inadequate to properly evaluate plant water uptake under saline conditions (Cardon and Letey, 1992a).

Cardon and Letey (1992b) developed the modified van Genuchten-Hanks (V-H) model to simulate crop production under various irrigation management regimes including saline conditions. The model used a plant water uptake term which uses empirical functions that relate observed plant water uptake to water potential (Feddes et al., 1976).

The V-H model used the general soil water flow equation as described by Richards including a plant water uptake term. The empirical plant water uptake function (van Genuchten, 1987) is defined as

[1]

where T_p(t) is the potential transpiration rate, (z,t) is the plant root distribution function, (h,) is a crop matric potential-salinity stress function, h is the soil matric head, and is the soil osmotic head defined by salt concentration.

The crop matric-salinity stress function is given by van Genuchten (1987)

[2]

where ß accounts for the differential response of the crop to matric and osmotic influences, and is equal to the ratio ₅₀:h₅₀, where h₅₀ and ₅₀ are the soil matric head and osmotic head at which maximum transpiration is reduced by 50%.

Modifications to the V-H model were introduced by Pang and Letey (1998). The uptake function (Eq. [1]) does not allow the plant to compensate for water stress by removing extra water from the root zone where the water is not deficient. Furthermore, it can be noted from Eq. [2] that the value of is less than unity except where and h = 0 (no threshold value). Maas and Hoffman (1977) reported that empirical relationships between salinity and plant growth could be characterized by a threshold salinity value beyond which yields declined. Thus, adjustments were made to the V-H model to produce an effective threshold value for the matric-salinity stress function and to allow additional water uptake from zones where did not exceed the threshold to compensate for zones where water stress exceeded the threshold.

The values of ₅₀ and the threshold of can be determined from the Maas and Hoffman (1977) coefficients. Most experiments conducted to determine the relationship between yield and matric suction allow the plant to extract water to predetermined suction values before irrigating. This type of experiment allows estimates of the threshold value of h, but it is almost impossible to experimentally determine the value of h₅₀. On the basis of these limitations, Pang and Letey (1998) proposed the following procedures.

First, the values of the threshold osmotic potential (_t) and ₅₀ were determined for the crop under consideration using the Maas and Hoffman (1977) coefficients. Assuming no matric stress (h = 0), the values of _t and ₅₀ were inserted into Eq. [2] to calculate the threshold value of the matric potential-salinity stress function (₁). This threshold value of ₁ will be used for the entire root zone. Next, the user must select a threshold value for matric potential (h_t). Then the selected threshold values of h_t and ₅₀ and calculated with the value of being set equal to zero are inserted into Eq. [2] to calculate ß. This ß value is used to calculate the value of h₅₀ using the equation of h₅₀ = ₅₀/ß. The value of ß was selected so that _t represents values for both _t and h_t.

Pang and Letey (1998) added modules for N and pesticide factors and referred to the model as the ENVIRO-GRO model. The focus of the present manuscript is on salinity and it is assumed that N and pesticides are not factors. Cardon and Letey (1992b) compared simulated results from the V-H model to experimental values from an experiment on corn conducted in Israel that contained several levels of water salinity and irrigation time intervals (Cardon and Letey, 1992b). The model as used did not allow plants to compensate for water stress by removing extra water from the root zone where water was not deficient, nor did it allow a threshold value for and h. Furthermore, without a rationale for selecting a value for h₅₀, the computations were done assuming ₅₀ = h₅₀. One purpose for this paper is to compare simulated results which account for compensation and threshold values as well as a rationale for selecting a more appropriate value for h₅₀ to the measured field experiments. One additional factor is that the previous comparison was limited to the information published on the corn field experiment (Shalhevet et al., 1986), whereas the present study utilized the detailed experimental conditions. A second purpose for this manuscript is to detail the temporal behavior of the matric and osmotic heads as well as plant response, which has not been previously done. A comparison of the simulated and measured salt distribution in the soil profile as well as the effect of different rooting patterns on crop response are presented elsewhere (Feng et al., 2003).

	EXPERIMENTAL DESCRIPTION AND SIMULATION PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL DESCRIPTION AND... RESULTS REFERENCES

 
The model was tested using data from a study with sweet corn (Zea mays ‘Jubilee’) conducted at the Gilat Agricultural Experimental Station in the northern Negev of Israel (Shalhevet et al., 1986). The experiment was a split-plot design with five levels of EC of the irrigation water (EC_i equal to 1.7, 4.0, 5.0, 8.0, and 10.2 dS m^-1) as main treatments and four irrigation intervals (3.5, 7, 14, and 21 d) as subtreatments in three replicates. The 3.5-d treatment is the average interval between irrigations which varied slightly during the course of the experiment. Also, the irrigation water salinity varied slightly from the average values that are reported. The input data to the model were the specific amounts and dates of each irrigation.

Corn was sown on April 18 and harvested 80 d later. Prior to the initiation of the salinity and irrigation interval treatments, all plots received three uniform establishing irrigations that totaled 140 mm of fresh water (EC_i = 1.0 dS m^-1) until 37 d after planting. The sprinkler irrigation treatments received depths equal to potential evapotranspiration for the 3.5-d interval and the depths required to replenish the soil water to field capacity to the 90-cm depth as measured by neutron scattering in the longer irrigation interval treatments. These procedures resulted in different total amounts of water being applied for the different irrigation intervals.

The soil was a silt loam (Calcic Haploxeralfs) with a bulk density of 1.48 Mg m^-3, saturated volumetric water content was 0.44, and the saturated hydraulic conductivity was 0.84 cm h^-1. The parameters used in the Hutson and Cass (1987) hydraulic function of the model for the soil were: water content at the inflection point (_i) was 0.44 m³ m^-3, the matric potential at the inflection point (h_i) was -16.23 cm, the air entry matric potential (a) was -16.23 cm, the exponent (b) of the equation relating matric potential to water content was 7.66, and the exponent (bhb) for the equation relating hydraulic conductivity to water content was set at 18.31.

The values for parameters in the plant water uptake function (in units of kPa) were _t = -122, ₅₀ = -138, h_t = -50, and h₅₀ = -425. The values were obtained by multiplying the EC_e (average root zone electrical conductivity) values in Maas and Hoffman tolerance table by 72. The matric potential values were selected by a process described above.

The lower boundary was set at 2m with 5-cm increments of soil depth for computation. The bottom boundary condition was set as free drainage. The upper boundary condition was set as flux control condition with infiltration of irrigation according to the input rate.

The root distribution was not measured in the experiment. The final root distribution chosen for the simulation was that reported by Bar-Yosef (1999) from an experiment on corn in Israel. It was assumed that the corn reached maximum rooting depth and final distribution on the 20th day after planting. The rooting depth as a function of time was programmed to reach the final distribution after 20 d as depicted in Fig. 1 .

View larger version (27K): [in this window] [in a new window]   Fig. 1. The root distribution as a function of time after planting used in the simulations.

The measured pan evaporation and crop coefficients for corn determined in the area were used to calculate the potential evapotranspiration during the course of the experiment. The model does include a feedback mechanism by which the crop coefficient is adjusted for reduced growth caused by water stress.

Even though the experiment was only 1 yr, the simulations were conducted for a 5-yr period. The assumption was that the water and salt distributions at the end of one growing season represented the initial conditions for the succeeding years. These simulations would represent the long-term consequences of imposing the given treatments. The simulations were conducted assuming no rainfall between seasons. However, the model is capable of simulating rainfall if desired.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL DESCRIPTION AND... RESULTS REFERENCES

 
The simulated relative yield is presented as a function of the irrigation water salinity for the different irrigation time intervals during the first 2 yr in Fig. 2 . The simulated yield decreased with increasing irrigation water salinity and increasing time interval between irrigations. Thus, the results suggest that both the matric and osmotic components contribute to the decreased yield.

View larger version (21K): [in this window] [in a new window]   Fig. 2. The simulated relative yield for the various irrigation intervals and irrigation water salinities for two years. The water salinities in dS m-1 were S1 (1.7), S2 (4.0), S3 (5.3), S4 (7.9) and S5 (10.2).

The model simulates the yield relative to a nonstressed crop. The experimental data were yield for a given treatment. In this case, even the lowest salinity of the irrigation water used in the experiment is considered to be relatively high for salt-sensitive corn. Therefore, the measured yield for a nonstressed crop is not accurately known. Thus, the comparison between measured and simulated results has a level of uncertainty.

A plot of simulated relative yield compared with measured relative yield is presented in Fig. 3 for conditions that the unstressed corn yield was 3.00 or 3.10 Mg ha^-1. Shalhevet et al. (1986) stated that a maximum yield of 3.1 Mg ha^-1 was obtained in an adjacent N fertilization experiment where nonsaline water was applied by drip irrigation. The mean simulated relative yield was 0.70 and the observed mean relative yields were 0.68 (unstressed yield equal to 3.0) and 0.66 (unstressed yield equal to 3.1). The Willmott's index of agreement (Willmott, 1981) was 0.96 (unstressed yield equal to 3.0) and 0.93 (unstressed yield equal to 3.1). The agreement between measured and simulated results is best for the case where the unstressed corn yield was 3.00 Mg ha^-1. As significant as the absolute fit, because of uncertainty on unstressed yield, is that the different treatments affected the simulated yields in the same manner as they affected the measured yields.

View larger version (18K): [in this window] [in a new window]   Fig. 3. Comparison of measured and simulated relative yields assuming unstressed yield equal to 3.0 and 3.1 Mg ha-1.

View larger version (22K): [in this window] [in a new window]   Fig. 4. The root-weighted average matric and osmotic heads prior to irrigation as a function of time for the 3.5-d irrigation interval and irrigation water salinities of 1.7 dS m-1 (S1) and 10.2 dS m-1 (S5).

View larger version (20K): [in this window] [in a new window]   Fig. 5. The root-weighted average matric and osmotic heads prior to irrigation as a function of time for the 14-d irrigation interval and irrigation water salinities of 1.7 dS m-1 (S1) and 10.2 dS m-1 (S5).

The model allows tracking the simulated relative ET on a periodic basis as well as the computed cumulative relative yield during the season. These results are depicted in Fig. 6 and 7 for the same treatments that were illustrated in Fig. 4 and 5 for comparative purposes. Considering first the results for the 3.5-d intervals, note that the crop was not stressed during much of the season (Fig. 6); however, there were periods of simulated stress which coincide with the dates of the lowest h values as depicted in Fig. 4. These were also at times when the interval between irrigation was the longest. The fluctuations in short term stress were dampened out by the cumulative effect.

View larger version (25K): [in this window] [in a new window]   Fig. 6. The periodic and cumulative simulated relative ET for the 3.5-d irrigation interval and irrigation water salinities of 1.7 dS m-1 (S1) and 10.2 dS m-1 (S5).

View larger version (26K): [in this window] [in a new window]   Fig. 7. The periodic and cumulative simulated relative ET for the 14-d irrigation interval and irrigation water salinity of 1.7 dS m-1 (S1) and 10.2 dS m-1 (S5).

Although the model was designed as a research tool to simulate the consequences of various irrigation management options under saline conditions, the results presented in Fig. 4 through 7 suggest that the model could be used to track and guide irrigation through the growing season. The reliability of the model to simulate salt distribution in the profile along with the sensitivity of the model to root distribution inputs are reported in a separate paper (Feng et al., 2003). The model does simulate the amounts and salinity of water leaching below the root zone, but no experimental data were available for comparison.

In conclusion, the agreement between the simulated and measured results on crop yield strongly suggest that the model can be used with confidence in simulating irrigation management options under saline conditions. Significantly, the experiment imposed treatments that caused osmotic stress, matric stress, and a combination of the two. The agreement between simulated and measured results was equal regardless of the source of the stress. Of importance is that the model does not require calibration and there were no curve-fitting parameter adjustments used. Reliable input data, however, are required. The initial conditions are important for one-season simulations.

Received for publication March 5, 2002.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL DESCRIPTION AND... RESULTS REFERENCES

 

• Bar-Yosef, B. 1999. Advances in fertigation. p. 1–77. In D.L. Sparks (ed.) Advances in Agronomy. Vol. 65. Academic Press, New York.
• Cardon, G.E., and J. Letey. 1992a. Plant water uptake terms evaluated for soil water and solute movement models. Soil Sci. Soc. Am. J. 32:1876–1880.
• Cardon, G.E., and J. Letey. 1992b. Soil-based irrigation and salinity management model: I. Plant water uptake calculations. Soil Sci. Soc. Am. J. 56:1881–1887.[Abstract/Free Full Text]
• Feddes, R.A., P.J. Kowalik, K.K. Malinka, and H. Zaradny. 1976. Simulation of field water uptake by plants using a soil water dependent root extraction function. J. Hydrol. (Amsterdam) 31:13–26.
• Feng, G.L., A. Meiri, and J. Letey. 2003. Evaluation of a model for irrigation management under saline conditions: II. Salt distribution and rooting pattern effects. Soil Sci. Soc. Am. J. 67:77–80 (this issue).[Abstract/Free Full Text]
• Great Plains Systems Research. 1992. Rootzone water quality model. Version 1.0. GPSR Tech. Rep. No. 2. USDA-ARS-GPSR, Fort Collins, CO.
• Hillel, D. 1991. Out of the Earth. The Free Press, New York.
• Hutson, J.L., and A. Cass. 1987. A retentivity function for use in soil-water simulation models. J. Soil Sci. 38:105–113.
• Hutson, J.L., and R.J. Wagenet. 1992. LEACHM. Leaching estimation and chemistry model: A process based model of water and solute movement, transformations, plant uptake and chemical reactions in saturated zone. Version 3. Dep. of Agronomy, Cornell Univ., Ithaca, NY.
• Jones, C.A., and J.R. Kiniry. 1986. The CERES-Maize: A simulation model of maize growth and development. Texas A&M Univ. Press, College Station, TX.
• Leonard, R.A., W.G. Knisel, and D.A. Still. 1987. GLEAMS: Groundwater loading effects of agricultural management systems. Trans. ASAE 30:1403–1418.
• Maas, E.V., and G.J. Hoffman. 1977. Crop salt tolerance—Current assessment. J. Irrig. Drain. Div. Am. Soc. Civ. Eng. 103(IR2):115–134.
• Nimah, M.N., and R.J. Hanks. 1973. Model for estimating soil, water, plant, and atmospheric interactions: I. Description and sensitivity. Soil Sci. Soc. Am. Proc. 37:522–527.
• Pang, X.P., and J. Letey. 1998. Development and evaluation of ENVIRO-GRO, an integrated water, salinity, and nitrogen model. Soil Sci. Soc. Am. J. 62(5):1418–1427.[Abstract/Free Full Text]
• Shalhevet, J., A. Vinten, and A. Meiri. 1986. Irrigation interval as a factor in sweet corn response to salinity. Agron. J. 78:539–545.[Abstract/Free Full Text]
• van Genuchten, M.Th. 1987. A numerical model for water and solute movement in and below the root zone. Res. Rep. 121. USDA-ARS, U.S. Salinity Lab., Riverside, CA.
• Willmott, C.J. 1981. On the validation of models. Phys. Geogr. 2:184–194.

This article has been cited by other articles:

				H. Steppuhn, M. Th. van Genuchten, and C. M. Grieve Root-Zone Salinity: I. Selecting a Product-Yield Index and Response Function for Crop Tolerance Crop Sci., January 1, 2005; 45(1): 209 - 220. [Abstract] [Full Text] [PDF]

				G. L. Feng, A. Meiri, and J. Letey Evaluation of a Model for Irrigation Management Under Saline Conditions: II. Salt Distribution and Rooting Pattern Effects Soil Sci. Soc. Am. J., January 1, 2003; 67(1): 77 - 80. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Feng, G. L. Articles by Letey, J. Search for Related Content PubMed Articles by Feng, G. L. Articles by Letey, J. Agricola Articles by Feng, G. L. Articles by Letey, J. Related Collections Plant and Soil Interactions Other Models Plant and Environment Interactions Irrigation