Prediction of topsoil stoniness using soil type information and airborne gamma-ray data

27-38

Artikel aus einer Zeitschrift

200209907

Canadian Journal of Forest Research 2022, 52(1) ; Jrg. 52 ; 1 (2022) , 27-38

The stoniness of topsoil can have a significant impact on the cost-effectiveness and quality of work in mechanized
forest operations. The operations and their models should be selected on a stand-specific basis, while the physical properties of the soil, including stoniness, to achieve maximum efficiency and to minimize the damage caused by heavy forest machinery. The aim of this study was to examine whether the stoniness of the topsoil can be predicted using the gamma-ray values available from geophysical data collected at low altitude and using soil type information. Stoniness was measured at several sites with various soil types, which were then divided into stoniness index classes (SICs) for further analysis by ordinal regression analysis using gamma-ray and soil type data. The predictions associated with SIC classification were 52% accurate and 79% acceptable (61 class from the correct class), with kappa values of 0.55 and 0.72, respectively. The SIC prediction results were promising and showed the potential of gamma-ray and soil type data for estimating topsoil stoniness.
Key words: forest trafficability, gamma-ray radiation, stoniness index, ordinal regression, kappa.