Premium
Predictors of urban variable source area: a cross‐sectional analysis of urbanized catchments in the United States
Author(s) -
Lim Theodore Chao
Publication year - 2016
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.10943
Subject(s) - impervious surface , environmental science , watershed , drainage basin , hydrology (agriculture) , logistic regression , surface runoff , urbanization , land use , stormwater , catchment area , geography , cartography , statistics , ecology , geology , mathematics , computer science , machine learning , biology , geotechnical engineering
Many studies have empirically confirmed the relationship between urbanization and changes to the hydrologic cycle and degraded aquatic habitats. While much of the literature focuses on extent and configuration of impervious area as a causal determinant of degradation, in this article, I do not attribute causes of decreased watershed storage on impervious area a priori. Rather, adapting the concept of variable source area (VSA) and its relationship to incremental storage to the particular conditions of urbanized catchments, I develop a statistically robust linear regression‐based methodology to detect evidence of VSA‐dominant response. Using the physical and meteorological characteristics of the catchments as explanatory variables, I then use logistic regression to statistically analyze significant predictors of the VSA classification. I find that the strongest predictor of VSA‐type response is the percent of undeveloped area in the catchment. Characteristics of developed areas, including total impervious area, percent‐developed open space and the type of drainage infrastructure, do not add to the explanatory power of undeveloped land in predicting VSA‐type response. Within only developed areas, I find that total impervious area and percent‐developed open space both decrease the odds of a catchment exhibiting evidence of VSA‐type response and the effect of developed open space is more similar to that of total impervious area than undeveloped land in predicting VSA response. Different types of stormwater management infrastructure, including combined sewer systems and infiltration, retention and detention infrastructure are not found to have strong statistically significant effects on probability of VSA‐type response. VSA‐type response is also found to be stronger during the growing season than the dormant season. These findings are consistent across a national cross‐section of urbanized watersheds, a higher resolution dataset of Baltimore Metropolitan Area watersheds and a subsample of watersheds confirmed not to be served by (combined sewer systems). Copyright © 2016 John Wiley & Sons, Ltd.