DIFFERENTIATING SOIL PHYSICAL-PROPERTIES FROM MULTIPLE BAND DOQ DATA

The purpose of this study was to investigate the use of color infarared-digital orthophoto quadrangle (CIR-DOQ) data in differentiating among surface soils of similar properties and to identify the CIR-DOQ spectral band most useful in estimating selected soil properties, The CIR-DOQ data were generated by scanning four 9 x 9 in CIR aerial photographs at a uniform pixel sampling grid of 25 microns, resulting in 2 m of ground sample resolution, Correlation, regression, and discriminant analyses were used in analyzing the data, The overall accuracy of the CIR-DOQ data in differentiating the surface soils was 62.8%. AU three of the CIR-DOQ bands were important in differentiating the surface soils, Significant correlations were found among the CIR-DOQ data and soils variables studied; however, the amount of variance explained was quite low, The prediction equations for estimating soil properties accounted for less than 35% of the variability in the data, The low percentages can be attributed to the season of the year and climatic conditions that existed when the CIR photographs were taken, These conditions had a profound impact on the spectral quality of the photographs and, therefore, the quality of data recorded by the scanner. It was concluded that even though the spatial resolution of the CIR-DOQ data is superior to most digital spectral data used today, the spectral quality is inadequate for differentiating among surface soils and for generating mathematical equations for predicting soil properties.