1 file | 492.87 KB

# Optimizing Hydrus 1D for irrigation management purposes in sandy grassland

Meisam Rezaei (UGent) , Stijn Van Hoey (UGent) , Piet Seuntjens (UGent) and Wim Cornelis (UGent)
Author
Organization
Abstract
To improve irrigation strategies in terms of higher efficiency and sustainability, computer simulation models can play a key role. In this study, the Hydrus 1D model was optimized for an irrigated grassland on sandy soil. Manual and inverse stepwise calibration was applied in which both heterogeneous and homogeneous conceptualizations of the soil profile were considered. Local sensitivity analysis showed that the model was most sensitive for saturated hydraulic conductivity Ks and the van Genuchten parameters n and α. Results show that Hydrus 1D closely simulated soil water content at five depths as compared to water content measurements from soil moisture probes taken in 2011. The optimized model performed well when validating against water content data taken in the 2012 growing season. The model errors, expressed by deviations between observed and modeled soil water content were, however, different for each individual depth. The smallest differences between the observed value and soil-water content were attained when using an automated inverse optimization method, in which the initial parameter values were optimized using a stepwise approach. Correlation between estimated parameters suggested that the solution was nonunique, indicating an overparameterization of the model. Using inappropriate initial estimates of soil hydraulic parameters in combination with multiple local minima and uncertainties in the data, inhibits the identification of the global minimum. Our results show that optimization is highly influenced by the number of parameters to be estimated and the initial values of soil hydraulic parameters.
Keywords
sensitivity analysis, water flow, Hydrus 1D, model optimization and validation, soil hydraulic properties

• (...).pdf
• full text
• |
• UGent only
• |
• PDF
• |
• 492.87 KB

## Citation

Please use this url to cite or link to this publication:

Chicago
Rezaei, Meisam, Ingeborg Joris, Wesley Boënne, Stijn Van Hoey, Piet Seuntjens, and Wim Cornelis. 2013. “Optimizing Hydrus 1D for Irrigation Management Purposes in Sandy Grassland.” In Water Technology and Management : Proceedings of the 2nd European Symposium, ed. Leen Bastiaens, 122–126. Mol, Belgium: VITO.
APA
Rezaei, M., Joris, I., Boënne, W., Van Hoey, S., Seuntjens, P., & Cornelis, W. (2013). Optimizing Hydrus 1D for irrigation management purposes in sandy grassland. In L. Bastiaens (Ed.), Water technology and management : proceedings of the 2nd European symposium (pp. 122–126). Presented at the 2nd European symposium on Water Technology and Management, Mol, Belgium: VITO.
Vancouver
1.
Rezaei M, Joris I, Boënne W, Van Hoey S, Seuntjens P, Cornelis W. Optimizing Hydrus 1D for irrigation management purposes in sandy grassland. In: Bastiaens L, editor. Water technology and management : proceedings of the 2nd European symposium. Mol, Belgium: VITO; 2013. p. 122–6.
MLA
Rezaei, Meisam, Ingeborg Joris, Wesley Boënne, et al. “Optimizing Hydrus 1D for Irrigation Management Purposes in Sandy Grassland.” Water Technology and Management : Proceedings of the 2nd European Symposium. Ed. Leen Bastiaens. Mol, Belgium: VITO, 2013. 122–126. Print.
@inproceedings{4267021,
abstract     = {To improve irrigation strategies in terms of higher efficiency and sustainability, computer simulation models can play a key role. In this study, the Hydrus 1D model was optimized for an irrigated grassland on sandy soil. Manual and inverse stepwise calibration was applied in which both heterogeneous and homogeneous conceptualizations of the soil profile were considered. Local sensitivity analysis showed that the model was most sensitive for saturated hydraulic conductivity Ks and the van Genuchten parameters n and \ensuremath{\alpha}. Results show that Hydrus 1D closely simulated soil water content at five depths as compared to water content measurements from soil moisture probes taken in 2011. The optimized model performed well when validating against water content data taken in the 2012 growing season. The model errors, expressed by deviations between observed and modeled soil water content were, however, different for each individual depth. The smallest differences between the observed value and soil-water content were attained when using an automated inverse optimization method, in which the initial parameter values were optimized using a stepwise approach. Correlation between estimated parameters suggested that the solution was nonunique, indicating an overparameterization of the model. Using inappropriate initial estimates of soil hydraulic parameters in combination with multiple local minima and uncertainties in the data, inhibits the identification of the global minimum. Our results show that optimization is highly influenced by the number of parameters to be estimated and the initial values of soil hydraulic parameters.},
author       = {Rezaei, Meisam and Joris, Ingeborg and Bo{\"e}nne, Wesley and Van Hoey, Stijn and Seuntjens, Piet and Cornelis, Wim},
booktitle    = {Water technology and management : proceedings of the 2nd European symposium},
editor       = {Bastiaens, Leen},
isbn         = {9789058570109},
language     = {eng},
location     = {Leuven, Belgium},
pages        = {122--126},
publisher    = {VITO},
title        = {Optimizing Hydrus 1D for irrigation management purposes in sandy grassland},
year         = {2013},
}