HR: 0800h
AN: H41D-0449 [Abstracts]
TI: Characteristics and Impact of Imperviousness From a GIS-based Hydrological Perspective
AU: Moglen, G E
EM: moglen@umd.edu
AF: Department of Civil and Environmental Engineering,
University of Maryland, 1173 Glenn L. Martin Hall, College Park, MD 20742
United States
AU: * Kim, S
EM: sunghee@wam.umd.edu
AF: Department of Civil and Environmental Engineering,
University of Maryland, 1173 Glenn L. Martin Hall, College Park, MD 20742
United States
AB:
With the concern that imperviousness can be differently quantified depending on data sources and methods, this study assessed
imperviousness estimates using two different data sources: land use and land cover. Year 2000 land use developed by the
Maryland Department of Planning was utilized to estimate imperviousness by assigning imperviousness coefficients to unique
land use categories. These estimates were compared with imperviousness estimates based on satellite-derived land cover from
the 2001 National Land Cover Dataset. Our study developed the relationships between these two estimates in the form of
regression equations to convert imperviousness derived from one data source to the other. The regression equations are
considered reliable, based on goodness-of-fit measures.
Furthermore, this study examined how quantitatively different imperviousness estimates affect the prediction of hydrological
response both in the flow regime and in the thermal regime. We assessed the relationships between indicators of hydrological
response and imperviousness-descriptors. As indicators of flow variability, coefficient of variance, lag-one
autocorrelation, and mean daily flow change were calculated based on measured mean daily stream flow from the water year 1997
to 2003. For thermal variability, indicators such as percent-days of surge, degree-day, and mean daily temperature
difference were calculated base on measured stream temperature over several basins in Maryland. To describe imperviousness
through the hydrological process, GIS-based spatially distributed hydrological models were developed based on a water-balance
method and the SCS-CN method. Imperviousness estimates from land use and land cover were used as predictors in these models
to examine the effect of imperviousness using different data sources on the prediction of hydrological response. Indicators
of hydrological response were also regressed on aggregate imperviousness. This allowed for identifying if hydrological
response is more sensitive to spatially distributed imperviousness or aggregate (lumped) imperviousness. The regressions
between indicators of hydrological response and imperviousness-descriptors were evaluated by examining goodness-of-fit
measures such as explained variance or relative standard error.
The results show that imperviousness estimates using land use are better predictors of flow variability and thermal
variability than imperviousness estimates using land cover. Also, this study reveals that flow variability is more sensitive
to spatially distributed models than lumped models, while thermal variability is equally responsive to both models. The
findings from this study can be further examined from a policy perspective with regard to policies that are based on a
threshold concept for imperviousness impacts on the ecological and hydrological system.
DE: 1803 Anthropogenic effects (4802, 4902)
DE: 1819 Geographic Information Systems (GIS)
SC: Hydrology [H]
MN: Fall Meeting 2005