change layer. The 2001 NOAA data are based on 22 land use classes 
(http://www.csc.noaa.gov/crs/lca/oldscheme.html), which are not exactly the same as those used 
in the NLCD. In January 2007, the Multi-Resolution Land Characteristics Consortium (MRLC, 
http://www.mrlc.gov ) released a new national land cover data, NLCD 2001. The areas classified 
by the NOAA C-Cap program were incorporated into the 2001 release. Procedures used in the 
development of the 2001 land cover data layer are presented in Homer et al (2004). The land 
cover in NLCD 2001 is based on 30-meter resolution data derived from Landsat imagery and 
uses 21 classes that are a modified version of the land classes used in the 1992 NCLD analysis 
(see http://www.eDa.gov/mrlc/classification.htnii for a crosswalk of the two schemes). 
Land Use Patterns and Watershed Characteristics 
Land cover data from the 1992 and 2001 MRLC NLCD data and from the NOAA 1995 and 2001 
data were used to calculate the area and percentage of the watershed for each of the 21 (NLCD) 
or 22 (NOAA) land use classes. Accuracy of the 1992 NLCD data by EPA region is presented at 
http://landcover.usgs.gov/accuracv/index.php . Based on this analysis, users were cautioned about 
applying the data to highly localized studies, such as over a small a watershed. Accuracy of the 
NLCD data of the MRLC zone that contains the Yaquina watershed is estimated to be 86.1%. 
These data sets were used to generate estimates of the area of impervious surfaces in each 
watershed using default coefficients from the Analytical Tools Interface for Landscape 
Assessments (ATtILA) software (U.S. EPA, 2004b). The MRLC 2001 impervious surface layer, 
which represents an estimate of developed impervious surface per pixel by percent 
imperviousness, was clipped and summarized for the watershed. Overall accuracies for the 
impervious surfaces from the 2001 MRLC data range from 83 to 91 percent (Homer et al., 2004; 
Yang et al. 2003), and represent a higher resolution estimate of impervious surfaces than 
available from ATtiLA. Estimates of nitrogen and phosphorus loadings from land use were 
calculated from the watershed land cover data using coefficients from the ATtiLA program. 
Estimates of slope were calculated for each watershed from slope surfaces generated 
from 10 meter (Oregon) DEMS (digital elevation models). Mean slope by percent and by 
degrees for land surfaces above the mean high water level were calculated and all slope values 
were exported to an Access database. The 30 meter DEMS were obtained from the National 
Elevation Dataset (NED, http://ned.usgs.gov) , a seamless mosaic of the best elevation data. The 
149 
n 
