A Comparative Study of Spatial Aggregation Methodologies under the BioEarth Framework

Bhagyam Chandrasekharan1, Kirti Rajagopalan2, Keyvan Malek1, Claudio Stockle3, Jennifer C Adam4 and Michael Brady5, (1)Washington State University, Pullman, WA, United States, (2)Washington State University, Biological Systems Engineering, Pullman, WA, United States, (3)Washington State University, Department of Biological Systems Engineering, Pullman, WA, United States, (4)Washington State University, Civil & Environmental Engineering, Pullman, United States, (5)Washington State University, School of Economic Sciences, Pullman, United States
Abstract:
The increasing probability of water resource scarcity due to climate change has highlighted the need for adopting an economic focus in modelling water resource uses. Hydro-economic models, developed by integrating economic optimization with biophysical crop models, are driven by the economic value of water, revealing it’s most efficient uses and helping policymakers evaluate different water management strategies. One of the challenges in integrating biophysical models with economic models is the difference in the spatial scales in which they operate. Biophysical models that provide crop production functions typically run at smaller scale than economic models, and substantial spatial aggregation is required. However, any aggregation introduces a bias, i.e., a discrepancy between the functional value at the higher spatial scale and the value at the spatial scale of the aggregated units.

The objective of this work is to study the sensitivity of net economic benefits in the Yakima River basin (YRB) to different spatial aggregation methods for crop production functions. The spatial aggregation methodologies that we compare involve agro-ecological zones (AEZs) and aggregation levels that reflect water management regimes (e.g. irrigation districts).

Aggregation bias can distort the underlying data and result in extreme solutions. In order to avoid this we use an economic optimization model that incorporates the synthetic and historical crop mixes approach (Onal & Chen, 2012). This restricts the solutions between the weighted averages of historical and simulated feasible planting decisions, with the weights associated with crop mixes being treated as endogenous variables.

This study is focused on 5 major irrigation districts of the YRB in the Pacific Northwest US. The biophysical modeling framework we use, BioEarth, includes the coupled hydrology and crop growth model, VIC-Cropsyst and an economic optimization model. Preliminary findings indicate that the standard approach of developing AEZs does not perform well when overlaid with irrigation districts. Moreover, net economic benefits were significantly different between the two aggregation methodologies. Therefore, while developing hydro-economic models, significant consideration should be placed on the aggregation methodology.

Back to: From Glacierized Watersheds to Global River Basins: Advances in Water Resources Management, Modeling, and Monitoring II Posters