Combined Effects of Numerical Method Type and Time Step on Water Stressed Actual Crop ET

آب و خاک

Article 1, Volume 29, Issue 4, September 2015, Pages 776-786   

Document Type: Research Article

DOI: 10.22067/jsw.v0i0.20585

Author

B. Ghahraman^*

Ferdowsi University of Mashhad

Abstract

Introduction: Actual crop evapotranspiration (Eta) is important in hydrologic modeling and irrigation water management issues. Actual ET depends on an estimation of a water stress index and average soil water at crop root zone, and so depends on a chosen numerical method and adapted time step. During periods with no rainfall and/or irrigation, actual ET can be computed analytically or by using different numerical methods. Overal, there are many factors that influence actual evapotranspiration. These factors are crop potential evapotranspiration, available root zone water content, time step, crop sensitivity, and soil. In this paper different numerical methods are compared for different soil textures and different crops sensitivities.
Materials and Methods: During a specific time step with no rainfall or irrigation, change in soil water content would be equal to evapotranspiration, ET. In this approach, however, deep percolation is generally ignored due to deep water table and negligible unsaturated hydraulic conductivity below rooting depth. This differential equation may be solved analytically or numerically considering different algorithms. We adapted four different numerical methods, as explicit, implicit, and modified Euler, midpoint method, and 3-rd order Heun method to approximate the differential equation. Three general soil types of sand, silt, and clay, and three different crop types of sensitive, moderate, and resistant under Nishaboor plain were used. Standard soil fraction depletion (corresponding to ETc=5 mm.d-1), pstd, below which crop faces water stress is adopted for crop sensitivity. Three values for pstd were considered in this study to cover the common crops in the area, including winter wheat and barley, cotton, alfalfa, sugar beet, saffron, among the others. Based on this parameter, three classes for crop sensitivity was considered, sensitive crops with pstd=0.2, moderate crops with pstd=0.5, and resistive crops with pstd=0.7. Therefore, nine different classes were formed by combination of three crop types and three soil class types. Then, the results of numerical methods were compared to the analytical solution of the soil moisture differential equation as a datum. Three factors (time step, initial soil water content, and maximum evaporation, ETc) were considered as influencing variables.
Results and Discussion: It was clearly shown that as the crops becomes more sensitive, the dependency of Eta to ETc increases. The same is true as the soil becomes fine textured. The results showed that as water stress progress during the time step, relative errors of computed ET by different numerical methods did not depend on initial soil moisture. On overall and irrespective to soil tpe, crop type, and numerical method, relative error increased by increasing time step and/or increasing ETc. On overall, the absolute errors were negative for implicit Euler and third order Heun, while for other methods were positive. There was a systematic trend for relative error, as it increased by sandier soil and/or crop sensitivity. Absolute errors of ET computations decreased with consecutive time steps, which ensures the stability of water balance predictions. It was not possible to prescribe a unique numerical method for considering all variables. For comparing the numerical methods, however, we took the largest relative error corresponding to 10-day time step and ETc equal to 12 mm.d-1, while considered soil and crop types as variable. Explicit Euler was unstable and varied between 40% and 150%. Implicit Euler was robust and its relative error was around 20% for all combinations of soil and crop types. Unstable pattern was governed for modified Euler. The relative error was as low as 10% only for two cases while on overall it ranged between 20% and 100%. Although the relative errors of third order Heun were the smallest among the all methods, its robustness was not as good as implicit Euler method. Excluding one large error of 50%, the average relative errors in this method was less than 10%. However, the ETc is time-dependent and varies from one day to another. So, averaging ETc over a larger time step brings about more error in computations. Accumulated relative error in Eta (ETp=5 mm.d-1, W0=Wj, t=1 d) under medium soil and crop type was decreased as the number of time steps increased, irrespective of the numerical method.
Conclusions: Based on practical considerations, we propose implicit Euler for its robustness, and 3-rd order Heun for its low maximum relative error for all combinations of soil and crop types.

Keywords

Soil moisture depletion; Soil water balance; Stability in numerical methods; Water stress index

References

1- Allen R.G., Pereira L.S., Raes D., and Smith M. 1998. Crop evapotranspiration –Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper 56. Food and Agriculture Organization of the United Nations: Rome, Italy. ISBN 92-5-104219-5. Online at http://www.fao.org/docrep/X0490E/X0490E00.htm

2- Annandale J.G., Benade N., Jovanovic N.Z., Steyn J.M., Du Sautoy N., and Marais D. 1999. Facilitating Irrigation Scheduling by Means of the Soil Water Balance Model, Water Research Commission Report No. 753/1/99, Pretoria, South Africa.

3- Cesanelli A., and Guarracino L. 2009. Estimation of actual evapotranspiration by numerical modeling of water flow in the unsaturated zone: a case study in Buenos Aires, Argentina. Hydrogeology Journal, 17: 299-306.

4- Cesanelli A., and Guarracino L. 2011. Numerical modeling of actual evapotranspiration of a coffee crop. Scientica Agricola (Piracicaba, Brazil), 68(4): 395-399.

5- Chapra S.S., and Canale R.P. 2005. Numerical Methods for Engineers. Tata McGrawHill Book Company.

6- Higham N.J. 1996. Accuracy and stability of numerical algorithms, Society of Industrial and Applied Mathematics, Philadelphia, 1996. ISBN 0-89871-355-2

7- Jensen M.E., Robb D.C.N., and Fanzoy C.E. 1970. Scheduling irrigation using climate- crop- soil data. Journal of the Irrigation and Drainage Division, Proceedings of the American Society of Civil Engineers 96: 25-28.

8- Jovanovic N., and Israel S. 2012. Critical Review of Methods for the Estimation of Actual Evapotranspiration in Hydrological Models, Evapotranspiration - Remote Sensing and Modeling, Ayse Irmak (Ed.), ISBN: 978-953-307-808-3, InTech, Available from:

9- http://www.intechopen.com/books/evapotranspiration-remote-sensing-and-modeling/critical-review-of-methods-for-the-estimation-of-actual-evapotranspiration-in-hydrological-models

10- Moazenzadeh R. 2013. Monitoring Hydrologic System of Neishabour Watershed using Remote Sensing Technique. PhD Dissertation. Water Engineering Department, College of Agriculture, Ferdowsi University of Mashhad (in Persian with English abstract)

11- Muralidharan D., and Knapp K.C. 2009. Spatial dynamics of water management in irrigated agriculture. Water Resources Research, 45: W05411. DOI: 10.1029/2007WR006756.

12- Poulovassilis A., Anadranistaki M., Liakatas A., Alexandris S., and Kerkides P. 2001. Semi- empirical approach for estimating actual evapotranspiration in Greece. Agricultural Water Manegement, 51: 143-152.

13- Rao N.H., Sarma P.B.S., and Chander, S. 1988. Irrigation scheduling under a limited water supply. Agricultural Water Management, 15: 165-175.

14- Shafiei M. 2009. Modeling of surface water balance by using SWAT model and GIS tool in Neyshabour watershed. MSc Thesis. Water Engineering Department, College of Agriculture, Ferdowsi University of Mashhad (in Persian with English abstract).

15- Shang S. 2012. Calculating actual crop evapotranspiration under soil water stress conditions with appropriate numerical methods and time step. Hydrological Processes, 26: 3338–3343. DOI: 10.1002/hyp.8405

16- Shang S.H., and Mao X.M. 2006. Application of a simulation based optimization model for winter wheat irrigation scheduling in Nort China. Agricultural Water Management, 85: 314-322.

17- Zhang L., Dawes W.R., and Walker G.R. 2001. The Response of Mean Annual Evapotranspiration to Vegetation Changes at Catchment Scale. Water Resources Research, 37: 701-708.