GIBSI: an integrated modelling system for watershed management – sample applications and current developments

Abstract

Abstract. Hydrological and pollutant fate models have long been developed for research purposes. Today, they find an application in integrated watershed management, as decision support systems (DSS). GIBSI is such a DSS designed to assist stakeholders in watershed management. It includes a watershed database coupled to a GIS and accessible through a user-friendly interface, as well as modelling tools that simulate, on a daily time step, hydrological processes such as evapotranspiration, runoff, soil erosion, agricultural pollutant transport and surface water quality. Therefore, GIBSI can be used to assess a priori the effect of management scenarios (reservoirs, land use, waste water effluents, diffuse sources of pollution that is agricultural pollution) on surface hydrology and water quality. For illustration purposes, this paper presents several management-oriented applications using GIBSI on the 6680 km 2 Chaudière River watershed, located near Quebec City (Canada). They include impact assessments of: (i) municipal clean water program; (ii) agricultural nutrient management scenarios; (iii) past and future land use changes, as well as (iv) determination of achievable performance standards of pesticides management practices. Current and future developments of GIBSI are also presented as these will extend current uses of this tool and make it useable and applicable by stakeholders on other watersheds. Finally, the conclusion emphasizes some of the challenges that remain for a better use of DSS in integrated watershed management.

Study of the Climate Change Impacts on Water Quality in the Upstream Portion of the Cau River Basin, Vietnam

“…Since 1998, several applied studies have been conducted. Numerous upgrades to this integrated modeling system have been added since inception [17].…”

Analysis of parameter uncertainty in hydrological and sediment modeling using GLUE method: a case study of SWAT model applied to Three Gorges Reservoir Region, China

“…The conceptual parameters such as CN 2 in the SCS curve method are defined as the conceptualization of a non-quantifiable process, and determined by the process of model calibration. Conversely, physical parameters can be measured or estimated based on watershed characteristics when intensive data collection is possible (Vertessy et al, 1993;Nandakumar and Mein, 1997). Because of the unknown spatial heterogeneity of a studied area and the expensive experiments which may be involved, the physical parameters are usually determined by calibrating the model against the measured data (Raat et al, 2004).…”

“…It has been shown that parameter uncertainty is inevitable in hydrological modeling and a corresponding assessment should be conducted before model prediction in the decision making process. Studies of parameter uncertainty have been conducted in the area of integrated watershed management (Zacharias et al, 2005), peak flow forecasting (Jorgeson and Julien, 2005), soil loss prediction (Cochrane and Flanagan, 2005), nutrient flux analysis (Murdoch et al, 2005;Miller et al, 2006), assessment of the effect of land use change (Eckhardt et al, 2003;Shen et al, 2010;Xu et al, 2011) and climate change impact assessment (Kingston and Taylor, 2010), among many others. Nevertheless, parameter identification is a complex, non-linear problem and numerous possible solutions might be obtained by optimization algorithms (Nandakumar and Mein, 1997).…”

“…Watershed hydrology and river water quality models are important tools for watershed management for both operational and research programs (Quilbé and Rousseau, 2007; Van et al, 2008;Sudheer and Lakshmi, 2011). However, due to spatial variability in the processes, many of the physical models are highly complex and generally characterized by a multitude of parameters (Xuan et al, 2009).…”

“…Uncertainty estimates are routinely incorporated into Total Maximum Daily Load (TMDL) (Quilbé and Rousseau, 2007). Usually, the uncertainty in hydrological modeling is from model structures, input data and parameters (Lindenschmidt et al, 2007).…”