Methods to achieve accurate projection of regional and global raster databases

Abstract

This research aims at building a decision support system (DSS) for selecting an optimum projection considering various factors, such as pixel size, areal extent, number of categories, spatial pattern of categories, resampling methods, and error correction methods.Specifically, this research will investigate three goals theoretically and empirically and, using the already developed empirical base of knowledge with these results, develop an expert system for map projection of raster data for regional and global database modeling.The three theoretical goals are as follows:(1) The development of a dynamic projection that adjusts projection formulas for latitude on the basis of raster cell size to maintain equal-sized cells.(2) The investigation of the relationships between the raster representation and the distortion of features, number of categories, and spatial pattern.(3) The development of an error correction and resampling procedure that is based on error analysis of raster projection. HYPOTHESESRegarding the first goal, we hypothesize that regional and global raster data can be accurately projected with appropriate equations that account for raster cell size and latitudinal position.For the second goal, we hypothesize that scale factors explain the impact of distortion on raster representation and that more categories and more complex spatial patterns cause more errors.Finally, we hypothesize for the third goal that error correction and resampling methods can be used for optimizing the projection accuracy of regional and global raster datasets.This proposed research potentially has impacts on all U.S. Geological Survey (USGS) programs involving the use of large regional and global raster data sets, such as Global Change Research and Place-Based Studies.A complete description of this research project is included in USGS Open-File Report 01-181 (Usery and others, 2001).This current report will document project accomplishments during the first year, include the workplan for the second year, and document publications accomplished or planned from the research.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.