Modelling of lateritic iron-nickel ore deposits with geographically weighted regression method

It is estimated that in the next 20 years, the global and Turkish demand for nickel will double due to advancements in steel production, electric vehicles, and energy storage. The world's and Turkey's largest nickel resources are found in nickel laterite deposits, which form during the chemical weathering of ultramafic rocks and contain iron, cobalt, manganese, and other metallic elements. Therefore, it is essential to intensify resource exploration in polymetallic lateritic nickel deposits and discover new, economically viable nickel sources. At this stage, selecting the most suitable method for resource estimation of polymetallic lateritic nickel deposits is of great importance. In this study, the aim was to investigate whether the Geographically Weighted Regression (GWR) method, which is one of the geographic information system (GIS) modeling techniques widely used in the spatial modeling of surface soil analysis data, can also be applied to near-surface lateritic iron-nickel deposits. For this purpose, the resource modeling of the Adatepe lateritic iron-nickel deposit, located in the Sivrihisar-Mihalıççık district of Eskişehir, was conducted using GWR methods, and the applicability of the GWR method was evaluated by comparing its performance with the Ordinary Kriging (OK) and Co-Kriging (CK) methods.As a result of the performance comparisons of block estimates, it was found that the OK and CK methods tended to overestimate low-grade regions and underestimate high-grade regions, while the GWR method provided estimates closer to the actual grades in low-grade regions and slightly underestimated high-grade regions. It was concluded that more realistic grade estimates could be achieved in lateritic iron-nickel deposits, which also contain polymetallic materials, using the GWR method.