Use of spatial surrogates to assess the potential for non-point source pollution in large watersheds Heidi L. N. Moltz, 1 *† Walter Rast, 1 Vicente L. Lopes 1 and Stephen J. Ventura 2 1 Aquatic Resources Program, Department of Biology, Texas State University, San Marcos, TX, and 2 Department of Soil Science, University of Wisconsin, Madison, WI, USA Abstract Sediment represents a major non-point source pollutant throughout the world. In addition to reduced agricultural produc- tivity as the result of the loss of fertile soil, soil erosion also can have significant water-quality impacts in downstream waterbodies, reducing water transparency, degrading aquatic habitats and reducing the operational life and water storage capacity of reservoirs producing hydroelectric power. Various other pollutants also can absorb to sediment particles, creating additional downstream water-quality concerns for humans and the natural environment. In view of its human and environmental significance, two indices (an erosion index and runoff index) were developed to identify areas within the US portion of the Rio Grande Basin exhibiting physical characteristics conducive to producing significant non-point source pollution loads, focusing on land erosion as a sediment source. The erosion index is an adaptation of the Univer- sal Soil Loss Equation, being the product of rainfall erosivity (R factor), soil erodibility (K factor), and a topographic factor (LS factor). The erosion index correlated well with measurements of sediment yields from runoff plots. The Curve Number was used as the runoff index. In conjunction with identified pollutant-generating land uses, or source landscapes, these indices were used to identify sub-watersheds within the US portion of the Rio Grande Basin that merit further investigation for non-point source pollution prevention and control via the use of hydrologic modelling techniques. Key words Curve Number, erosion hazard, non-point source pollution, reservoir sedimentation, Rio Grande, risk assessment. INTRODUCTION Experience around the world indicates that pollution generated from non-point sources is a major cause of water-quality degradation. Although contaminants from non-point sources are not necessarily significantly differ- ent from point sources, rainfall and snowmelt events are the primary drivers of the former. Accordingly, the gener- ation and movement of non-point source pollutants (NPSPs) is more erratic and less frequent, and the dura- tion shorter, than the typically more continuous dis- charges from point sources. A comparison of the defining characteristics of point and NPSPs (Thornton et al. 1999) is provided in Table 1. A critical element to the success and sustainability of any non-point source pollution prevention programme is the ability to utilize limited time and funds most effec- tively to address priority areas, particularly in large water- sheds. In fact, a range of sophisticated assessment tools currently exists to assist decision-makers and managers to identify key areas for the implementation and appli- cation of non-point source pollution prevention pro- grammes. These tools include hydrologic models that provide detailed assessments of hydrologic processes and calculate pollutant loads, as well as providing information on the efficacy of alternative control practices and land- use changes applied at the field, farm, and sub-watershed scale. Process-based models provide an ideal tool for facilitating water resources management at small scales. However, the accuracy and reliability of hydrologic mod- els of diffuse (non-point) pollution decreases with the increasing complexity and size of the system being mod- elled (Novotny & Olem 1994). Very large watersheds, an *Corresponding author. Email: hmoltz@icprb.org †Present address: Interstate Commission on the Potomac River Basin, Rockville, MD, USA. Accepted for publication 12 September 2010. Ó 2011 The Authors Doi: 10.1111/j.1440-1770.2011.00460.x Journal compilation Ó 2011 Blackwell Publishing Asia Pty Ltd Lakes & Reservoirs: Research and Management 2011 16: 3–13