A novel technique for establishing soil topographic index thresholds in defining hydrologically sensitive areas in landscapes.

Hydrologically sensitive areas (HSAs) are useful for analyzing a watershed. However, correct delineation of HSAs is critical for conducting such analysis. Scientifically defensible methods for delineating HSAs are lacking. The objectives of this study are to identify threshold soil topographic indices (STIs) based on the relationship between the observed soil moisture and the calculated STI in two sites using a trellis plot approach, and validate the identified threshold STIs in delineating HSAs in 15 watersheds in North Central New Jersey based on a linear mixed modeling of the relationship between land use and water quality at the watershed and HSA scales. Field soil moisture data are collected during April, May, June, July, August, and September for three years at Christie Hoffman Park and Fairview Farm in Central New Jersey. The linear mixed models assess the relationship between land use metrics in terms of percentages of land uses and three water quality indicators including total suspended solids, total nitrogen and total phosphorus in streams at both watershed and HSA scales. Trellis plot analyses based on a polynomial regression model of order of two to four identify the threshold STIs ranging from nine to 15 for delineating HSAs. The linear mixed modeling results indicate that the relationships between land use and three water quality indicators at the HSA scales are similar to their relationships at the watershed scale. The predictive powers of these HSA and watershed scale models are very similar. These results suggest that it is appropriate for policymakers and watershed managers to use HSAs rather than entire watersheds to characterize watersheds and devise management strategies to optimize resource uses. The novel technique developed in this study can be used in other parts of the world.