Global sensitivity analysis for large-scale socio-hydrological models using Hadoop

摘要: A multi-agent system (MAS) model is coupled with a physically-based groundwater to understand the declining water table in heavily irrigated Republican River basin. Each agent MAS associated five behavioral parameters, and we estimate their influences on models using Global Sensitivity Analysis (GSA). This paper utilizes Hadoop-based Cloud Computing techniques Polynomial Chaos Expansion (PCE) based variance decomposition approach for improvement of GSA large-scale socio-hydrological models. With techniques, running 1000 scenarios can be completed within two hours Hadoop clusters, substantial over 42 days required run these sequentially desktop machine. Based results, conducted surrogate derived from PCE measure impacts spatio-temporal variations parameters crop profits table, identifying influential parameters. Develop framework combine cloud computing GSA.Reduce computation time simulations hours.Identify via sensitivity analysis.