Abstract
Spatial variation in the landscape factors climate, geomorphology, and lithology cause significant differences in water quality issues even when land use pressures are similar. The Physiographic Environment Classification (PEC) classifies landscapes based on their susceptibility to the loss of water quality contaminants. The classification is informed by a conceptual model of the landscape factors that control the hydrochemical maturity of water discharged to streams. In New Zealand, a case study using climatic, topographic, and geological data classified the country into six, 36, and 320 classes at Levels 1 (Climate), 1–2 (Climate + Geomorphology), and 1–3 (Climate + Geomorphology + Lithology), respectively. Variance partitioning analysis applied to New Zealand’s national surface water monitoring network ( n = 810 stations) assessed the contributions of PEC classes and land use on the spatial variation of water quality contaminants. Compared to land use, PEC explained 0.6× the variation in Nitrate Nitrite Nitrogen (NNN), 1.0× in Total Kjeldahl Nitrogen (TKN), 1.8× in Dissolved Reactive Phosphorus (DRP), 2.3× in Particulate Phosphorus (PP), 2.6× in E. coli , and 4.3× in Turbidity (TURB). Land use explained more variation in riverine NNN, while landscape factors explained more variation in DRP, PP, E. coli , and TURB. Overall, PEC accounted for 2.1× more variation in riverine contaminant concentrations than land use. The differences in contaminant concentrations between PEC classes ( p < 0.05), after adjusting for land use, were consistent with the conceptual model of hydrochemical maturation. PEC elucidates underlying causes of contaminant loss susceptibility and can inform targeted land management across multiple scales.