SWATGP data assimilation system(SWATGPDAS): introduction and validation

Abstract:

This paper presents the development of a high spatial revolution hydrological data assimilation system (SWATGPDAS) based on distributed hydrological model Gridded and Parallelized Soil and Water Assessment Tool (SWATGP). The purpose of this work is to obtain an improved estimation of hydrological states for watershed hydrological process prediction and water resource management by integrating multisource datum, hydrological model and data assimilation technology. SWATGPDAS is a software that integrates the use of remote sensing, such as MODIS LAI, Snow cover Fraction, Snow water equivalent, soil moisture and ground based observational data (e.g., watershed discharge, ground sensor networks), along with advanced hydrological model SWATGP and parallel data assimilation framework (PDAF) to accurately characterize watershed hydrological states and fluxes. SWATGPDAS employs high performance computing technologies to deal with the computational challenges of high resolution and/or large area hydrological modeling. Observational System Simulation Experiments (OSSEs) are designed to analysis and discuss the ability of hydrological simulation in SWATGPDAS. Both the temporal and spatial correspondences in trend/pattern and magnitudes between simulated and “true” states (soil moisture, snow water equivalent and discharge) were satisfactory with assimilation integration, suggesting the reliability of using SWATGPDAS for regional water balance studies.