A spatial decision support system framework for optimization of cropping pattern and water resources allocation at pasargard plains, fars province, Iran
In recent years, optimization models have been extensively used to address water management issues by means of proposing appropriate cropping patterns and water allocation rules. However, in spite of their significance these models often consider the land as an entirety and fail to account for the v...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2014
|
Subjects: | |
Online Access: | http://psasir.upm.edu.my/id/eprint/64724/1/FK%202014%20152IR.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | In recent years, optimization models have been extensively used to address water management issues by means of proposing appropriate cropping patterns and water allocation rules. However, in spite of their significance these models often consider the land as an entirety and fail to account for the variation of many contributing factors, including the ownership and fragmentation of lands and resources, or other unique characteristics that each farm may have. Moreover, the applicability of the present models can be limited due to the absence of inclusive frameworks in which the decision-maker could input his/her knowledge of the problem, properly model the hydrological and economic processes, test various scenarios, and arrive at judicious decisions. In the present study, a Spatial Decision Support System (SDSS) framework was developed to tackle the above-mentioned problems and was used to propose optimal decisions for agricultural lands of the Pasargad plain in central Iran as a case study. In this modular SDSS, the multi-purpose cadaster data help to form a spatial farm database in which the various characteristics of each farm are included. A unit response matrix groundwater model was coupled with a modified version of Genetic Algorithm (GA) in order to optimize cropping patterns and water allocation decisions. The proposed GA-based optimization model—namely Piece-Wise Genetic Algorithm (PWGA)—was capable of proposing optimal cropping patterns, deficit irrigation rules, and conjunctive use decisions for each farm, and to tackle the large number of decision variables involved. Furthermore, the Policy Analysis Matrix (PAM) was used as a module to limit the number of possible cropping decisions based on social and economic analyses. Meteorological data were also incorporated into the framework using an evapotranspiration module. Various scenarios were defined with regards to allowable water drawdown, water prices, and climatic conditions, and the associated optimal results for the case study are obtained from the SDSS. In this regards, the results acquired when considering the study area as a plain entirety are compared with those obtained from farm-based decision support. The results indicated that the use of lump models is problematic due to its tendency to overestimate the resulting net benefits, whereas the incorporation of cadastre-driven data into the optimization framework contributes to providing decisions that are more realistic. More importantly, the obtained decisions from SDSS—in the forms of maps and written reports—are detailed enough to be directly advised to farm-owners. Furthermore, the consistency of the PAM results with those of the optimization model confirms the suitability of employing PAM-based codes for refinement of the choice of crops. The use of a modular decision support framework in which each of the modules could work independently has also enabled the users of SDSS to employ the system selectively and adapt it according to their needs. |
---|