Intensive soil survey for the Andrews Watersheds 1, 2 & 3, Andrews Experimental Forest

Available on-line

See data access policy at www.fsl.orst.edu/lter (especially the data use
            policy)

The accuracy of the aerial photo interpretation was assessed through comparison of
            the percent cover maps against the plot cover measurements. Some differences were
            expected between the mapped cover percentages and the plot measurement, because of the
            difference in the resolution of the mapped areas and the 2x2m plot size. The mapped
            areas were drawn with an expectation that there was significant heterogeneity in the
            vegetation cover within each map unit. These results are presented in Appendix B of
            thesis. The 1972 herbaceous cover map agreed with the plot data for 45 plots, and in
            general overestimated herbaceous cover. The 1972 conifer and evergreen broadleaf cover
            map agreed with the plot data for 31 plots, and in general overestimated conifer and
            evergreen broadleaf cover. The 1972 deciduous broadleaf cover map agreed with the plot
            data for 58 plots, and in general underestimated deciduous cover. The 1979 conifer and
            evergreen broadleaf cover map agreed with the plot data for 35 plots, and in general
            overestimated conifer and evergreen broadleaf cover. The 1979 deciduous broadleaf cover
            map agreed with the plot data for 44 plots, and in general underestimated deciduous
            cover. The 1990 herbaceous cover map agreed with the plot data for 27 plots, and in
            general overestimated herbaceous cover. The 1990 evergreen broadleaf cover map agreed
            with the plot data for 18 plots, and in general overestimated evergreen broadleaf cover.
            The 1990 conifer cover map agreed with the plot data for 28 plots, and in general
            overestimated conifer cover. The 1990 deciduous broadleaf cover map agreed with the plot
            data for 34 plots. These map assessments were not used to adjust the cover map for two
            reasons. The resolution of the map units was quite a bit larger than the plot size, and
            there was no discernible spatial pattern in the plot data that could have been used to
            adjust the cover maps.

While substantial efforts are made to ensure the accuracy of data and
            documentation, complete accuracy of data sets cannot be guaranteed. All data are made
            available "as is". The Andrews LTER shall not be liable for damages resulting from any
            use or misinterpretation of data sets.