Development of an integrated modeling approach for identifying multilevel non-point-source priority management areas at the watershed scale

Abstract

The identification of priority management areas (PMAs) at the large-basin scale is notably complex because of the random nature of watershed processes, which complicates the application of traditional deterministic PMAs. In this study, a multilevel PMA (ML-PMA) framework is designed as a new tool to pinpoint these sensitive areas, within a basin, that contribute the most to water quality deterioration. The main advantage of the ML-PMA framework is the wide availability of its supplementary tools and its complete framework, which integrates both watershed and river processes in addressing PMAs at the watershed scale. The watershed model, stream model, and a Markov chain approach are integrated to depict the dynamics of watershed processes and various water quality statutes. Based on the results of this study, the river migration process is vital for water quality degradation in the river network and significantly influenced the final PMA map. In addition, the proposed ML-PMA framework considers the impact of climatic conditions and hydrological properties and allows for a more cost-effective allocation of PMAs among different years. In the authors' view, the connectivity of PMAs in terms of flux distribution and propagation downstream on which the ML-PMA is based makes the ML-PMA framework particularly interesting for watershed non-point-source pollution control.

Targeting priority management areas for multiple pollutants from non-point sources

“…Previously, we have showed that the land use change in the TGRR would have a limited impact on runoff and N, while sediment and P were sensitively biased by land use data [32]. Besides, we have also demonstrated that climatic forces have a limited impact on the spatial distributions of these high-and lowlevel PMAs, but they have an enhanced impact on medium-level PMAs [52]. In this sense, the designs of MP-PMAs at watershed scale should meet both current water quality targets and also take into consideration anticipated changes in future precipitation pattern and land use.…”

Structural uncertainty in watershed phosphorus modeling: Toward a stochastic framework

“…3, all specific simulated NPS-P data in the cumulative probability curve represent one or more behavioral statistics of model performance over a large range of P cycle algorithms/equations. For example, the cumulative probability of simulated NPS-P in 2003 (a typical wet year (Chen et al, 2014)) was given in Fig. 3.…”

Climate Change Impact on Critical Source Area Identification in a Maryland Watershed

“…Following the nomenclature of Chen et al (2014), the method used in this study would be classified as an HRU load per unit area (HRULA) approach. The HRULA approach in our study is coupled with a quantile method in which HRUs exporting the highest concentrations of surface runoff (SurQ), sediments (TSS), nitrogen (TN), and phosphorus (TP) are separated from the rest of the watershed area and considered to be hotspots (CSAs).…”

“…As a result, various strategies for identifying such areas have been explored in many research studies (Chen et al, 2014;Chu et al, 2004;Huaifeng et al, 2010;Huang et al, 2015;Niraula et al, 2013;Sexton et al, 2010;Shang et al, 2012;White et al, 2009;Winchell et al, 2014). These strategies include the subwatershed load approach (SLA), river/reach load approach (RLA), river/reach concentration approach (RCA), subwatershed load per area approach (SLAA), and HRU load approach (HRULA), with the choice of method typically selected based on scope, purpose, target pollutants, and tools used in the study (Chen et al, 2014). The use of spatially distributed, physically based hydrologic models, in conjunction with geographic information systems (GIS), is currently the preferred approach for CSA identification, but simpler indices and loading functions are still applied where advanced tools are unavailable or training is missing (Niraula et al, 2013).…”