PLACE II relies on raster-based data for all calculations, whereas PLACE I used a combination of raster and vector inputs. PLACE II eliminates some of the inherent spatial error involved when analyzing raster and vector data within a common geospatial workspace by converting and processing all data to a common global grid format, resolution and coordinate system. While this process actually adds a measure of error (since raster depictions at 2.5 arc-minutes cannot represent small and\or linear features, such as islands or shorelines, quite as precisely and accurately as vector data), at the same time it standardizes the distribution of the cross-tabulated error, since all variables will have very nearly the same number of input units (pixels) and their placement and cross-tabular analyses, will be uniformly assessed.  Additionally, PLACE I is for the year 1990 only while PLACE II contains data for the years 1990 and 2000.