thumbnail

The accuracy of selected land use and land cover maps at scales of 1:250,000 and 1:100,000

Circular 829
By:
https://doi.org/10.3133/cir829

Links

Abstract

Land use and land cover maps produced by the U.S. Geological Survey are found to meet or exceed the established standard of accuracy. When analyzed using a point sampling technique and binomial probability theory, several maps, illustrative of those produced for different parts of the country, were found to meet or exceed accuracies of 85 percent. Those maps tested were Tampa, Fla., Portland, Me., Charleston, W. Va., and Greeley, Colo., published at a scale of 1:250,000, and Atlanta, Ga., and Seattle and Tacoma, Wash., published at a scale of 1:100,000. For each map, the values were determined by calculating the ratio of the total number of points correctly interpreted to the total number of points sampled. Six of the seven maps tested have accuracies of 85 percent or better at the 95-percent lower confidence limit. When the sample data for predominant categories (those sampled with a significant number of points) were grouped together for all maps, accuracies of those predominant categories met the 85-percent accuracy criterion, with one exception. One category, Residential, had less than 85-percent accuracy at the 95-percent lower confidence limit. Nearly all residential land sampled was mapped correctly, but some areas of other land uses were mapped incorrectly as Residential.
Publication type Report
Publication Subtype USGS Numbered Series
Title The accuracy of selected land use and land cover maps at scales of 1:250,000 and 1:100,000
Series title Circular
Series number 829
DOI 10.3133/cir829
Edition -
Year Published 1980
Language ENGLISH
Publisher U.S. Govt. Print. Off.,
Description 24 p. : diagrs. ; 26 cm.
Google Analytic Metrics Metrics page
Additional publication details