Airborne Radiometric Surveys and Machine Learning Algorithms for Revealing Soil Texture

Soil texture is key information in agriculture for improving soil knowledge and crop performance, so the accurate mapping of this crucial feature imperative rationally planning cultivations targeting interventions. We studied relationship between radioelements Mezzano Lowland (Italy), a 189 km2 agricultural plain investigated through dedicated airborne gamma-ray spectroscopy survey. The K Th abundances were used to retrieve clay sand content by means multi-approach method. Linear (simple multiple) non-linear (machine learning algorithms with deep neural networks) predictive models trained tested adopting 1:50,000 scale map. comparison these approaches highlighted that model introduces significant improvements prediction fractions. predicted maps compared regional maps. Although macro-structures equally present, data permits us shedding light on finer features. Map areas higher coincident paleo-channels crossing Etruscan Roman periods, confirmed hydrographic setting historical geo-morphological features study area.