Watershed Characteristics

Watershed characteristics associated with the degree of urbanization in each of the study areas were determined for both physical (land cover, infrastructure, hydrologic modifications, soils, topography, and climate) and socioeconomic conditions. Additionally, several composite measures of the intensity of urban development were determined by using a multimetric approach (McMahon and Cuffney, 2000; Cuffney and Falcone, 2009). Watershed characteristics data tables are summarized in table 2.

Land Cover

All land-cover data were based on the National Land Cover Data 2001 (NLCD01) dataset classification scheme and protocols (U.S. Geological Survey, 2005) and were compiled using geographic information system (GIS) software. For this investigation, the data included land cover for the entire watershed (table 2-A, 78 KB), land cover for the entire watershed that is “distance weighted” (land-cover areas close to the sampling site have a higher weighting than areas farther away; weighting is the inverse of the distance of any area in the watershed to the sampling location (table 2-B, 70 KB); and land cover within an approximately 200-meter (m)-wide riparian corridor for all stream segments within each study watershed (table 2-C, 72 KB). Measures of the degree of land-cover fragmentation also were calculated (tables 2-D to 2-K). Information about the derivation of these data is contained in Falcone and others (2007).

Table 2-D. Fragmentation statistics – Mean euclidean distance (79 KB)

Table 2-E. Fragmentation statistics – Largest patch index (78 KB)

Table 2-F. Fragmentation statistics – Number of patches (54 KB)

Table 2-G. Fragmentation statistics – Mean patch area (70 KB)

Table 2-H. Fragmentation statistics – Patch density (72 KB)

Table 2-I. Fragmentation statistics – Mean proximity index (71 KB)

Table 2-J. Fragmentation statistics – Proportion of like adjacencies (72 KB)

Table 2-K. Fragmentation statistics – Mean shape index (72 KB)

Socioeconomic Characteristics

Data from the 2000 U.S. Census (GeoLytics, Inc., 2001) were used to develop socioeconomic characteristics for each of the study watersheds. These characteristics include energy use (table 2-L, 65 KB), ethnicity (table 2-M, 86 KB), housing (table 2-N, 77 KB, and table 2-O, 120 KB), income (table 2-P, 94 KB), and population (table 2-Q, 101 KB). Further information about the derivation of these data is contained in Falcone and others (2007).

Infrastructure

Information about roads, point-source dischargers, toxic-release sites, and dams was used to characterize infrastructure in the study watersheds. Road data were obtained from the Census 2000 Topologically Integrated Geographic Encoding and Referencing (TIGER) database (GeoLytics, Inc., 2004). Point-source discharger locations were derived from the U.S. Environmental Protection Agency (USEPA) National Pollutant Discharge Elimination System (NPDES) database (U.S. Environmental Protection Agency, 2005a). Toxic-release locations were derived from the USEPA Toxic Release Inventory (TRI) database (U.S. Environmental Protection Agency, 2005b). Data on dam locations were obtained primarily from the National Inventory of Dams (U.S. Army Corps of Engineers, 1996; table 2-R, 59 KB).

Urban Intensity Index

Three different multimetric indices of urban intensity (UII) were derived for each of the nine urban studies based on a set of census, land-cover, and infrastructure variables that were correlated with population density (Cuffney and Falcone, 2009; table 1, 108 KB). The UII characterizes important types of disturbance associated with urbanization and is assumed to relate to changes in biological assemblages better than any single urban characteristic.

The first set of indices (metropolitan area urban intensity index, MA-UII) was developed uniquely for each study area on the basis of an assessment of factors deemed important in characterizing urbanization in each region. The number of variables included in these indices ranged from as few as 5 to as many as 40. The other two multimetric indices were developed by using a common set of three variables—housing-unit density, percentage of developed land cover, and road density. These variables serve as a proxy for the comprehensive changes associated with urbanization and were selected on the basis of high positive correlations (rho ≥ 0.7) to population density in all of the study areas. One set of index values, representing a gradient of urban intensity, was developed separately for each individual study area (metropolitan-area scaled national urban intensity index, MA-NUII), with values for each study ranging from 0 to 100. A second set of index values was developed to account for differences among the study areas in the rate of change in the MA-NUII relative to changes in population density. This index was scaled to range from 0 to 100 across the nine study areas (nationally scaled urban intensity index, NUII). The range of values for this second index in any of the EUSE study areas can be less than 100, which reflects a limited range of population density in the study area.

Hydrologic Modifications

In-channel structures and other hydrologic modifications can affect the timing, magnitude, duration, and variability of streamflow. Geographic information system analysis was used to identify hydrologic modifications in EUSE study watersheds, which resulted in defining such variables as the number and density of dams, dam storage area, and the percentage of waterbodies associated with canals and other human-developed conveyances (table 2-S, 110 KB).

Soils, Topography, and Climate

Soils, topography, and climate can exert important influences on stream ecosystems at multiple scales. Data on soils and topography (table 2-T, 185 KB) and climate (table 2-U, 153 KB) were prepared using methods described in Falcone and others (2007).