Integration of remote sensing and geophysical data to enhance lithological mapping utilizing the Random Forest classifier: a case study from Komopa, Papua Province, Indonesia

Lithological information is important in mineral resource exploration, geological observations, mine planning or degradation vulnerability assessment. Currently, lithology mapping can be performed a fast, inexpensive, and easy way using remote sensing data machine learning. Remote techniques have become valuable promising tool for lithological units searching minerals. Typically, the integration of with geophysical provides better diagnosis to than single-source methodologies. Accordingly, this study used combination airborne utilizing Random Forest algorithm small training samples enhance Komopa, Papua Province, Indonesia. Geophysical consisting magnetic, electromagnetic, radiometric were added one by gradually data, which includes Sentinel 2A, ALOS PALSAR, DEM (digital elevation model) compare accuracy classification results from each dataset. The showed that model combined three types produced best classification, an overall 0.81, precision 0.66, recall 0.47, F1 score 0.52. This fused increase 8% accuracy, 6% precision, 11% recall, 13% when compared only data.