VNIR Spectroscopy Estimation of Soil Quality Indicators

Knowledge of within-field spatial variability in soil quality indicators is important to assess the impact of site-specific management on the soil. Standard methods for measuring these properties require considerable time and expense, so sensor-based approaches would be useful. The purposes of this research were (1) to evaluate changes in soil quality indicators after initiation of a precision conservation system on a typical claypan-soil research field in northeast Missouri and (2) to evaluate the ability of visible and near infrared (VNIR) spectroscopy to estimate those soil quality indicators. Soil samples were obtained to a 15-cm depth on a 30-m grid spacing, plus at a number of random sampling locations, before and three years after conversion from a corn-soybean rotation to a no-till soybean-wheat-hay rotation. Laboratory analyses were conducted for potential indicators of soil quality, including chemical constituents, labile and total organic carbon, and particulate organic matter carbon (POM-C) and nitrogen (POM-N). VNIR reflectance of dried and sieved samples was obtained in the laboratory using a spectrometer with a wavelength range from 350 to 2500 nm. Calibrations of VNIR reflectance to soil quality parameters were accomplished using partial least squares regression. Laboratory analysis showed a significant increase in soil quality indicator values over the study period, particularly POM-C and POM-N. These increases were greatest in the wetter, lower-landscape portions of the field, perhaps due to higher biological activity. VNIR spectroscopy was not able to accurately estimate the soil variables studied, with the exception of pH. This may have been due to their relatively low variability across the study area. Further research will evaluate VNIR spectroscopy for estimating soil quality indicators across multiple management systems where variability is expected to be greater.