Assessing land use and land cover influence on surface water quality using a parametric weighted distance function

Abstract

Stream water quality is directly influenced by land use and human practices in the surrounding environment. Understanding such effects and the spatial extent of impacts is essential to generate reliable information for ecosystem-based management of water resources. We identified sources of impact on water quality and characterized indicator-specific landscape influence on samples collected during base flow along the Chubut River (43ºS, 69ºW). We modeled Total Nitrogen (TN), Total Phosphorous (TP), Soluble Reactive Phosphorous (SRP) concentrations and 15N of particulate organic matter along the river, as a function of effective contribution areas (AEC) of Land Use/Land Cover (LULC). AECs were calculated by assuming that landscape influence decays exponentially with the Euclidean distance between a given LULC parcel and the sampling point. We calibrated the model to the observations by estimating an indicator-specific decay rate. Agriculture and barren lands were the main sources of phosphate nutrients whereas urban areas were the main source of TN. Radius of landscape influence for SRP (100-180 km) was larger than for TP (10-25 km), reflecting different patterns of mobilization and delivery in the catchment. 15N variation was explained by vegetation cover but the influence rapidly decreased (1-4 km) reflecting a mostly autochthonous source of organic matter.