A MaxEnt predictive model for palaeontological sites in the Siwalik Hills: A case study from the Pinjor Formation of the Upper Siwalik Hills near Chandigarh, northern India

Abstract

This study presents application of the MaxEnt predictive model to identify potential palaeontological sites in the Siwalik Hills, focusing on the Pinjor Formation near Chandigarh, northern India. The Siwalik region holds rich vertebrate palaeontological records, yet lacks comprehensive site prediction models. This research employs MaxEnt (3.4.0) software, to predict areas suitable for fossil occurrences. Georeferenced presence data was derived from literature and field surveys, for training the model. Environmental predictors including Pinjor geological deposit boundary, Vegetation Index (NDVI), and Slope and Aspect were from a digital elevation model. Furthermore, model development involved parameter tuning, for developing a potentially useful model. Field validation of the model through pedestrian surveys identified new fossil localities, demonstrating the model's practical applicability. This research emphasizes the importance of MaxEnt in developing site predictive models, offering a valuable tool for optimizing palaeontological field surveys. While highlighting the model's success, the study also recognizes its limitations, especially concerning landscape and vegetation changes over time. Overall, this work establishes a foundation for further research in predictive modelling for palaeontological exploration in the Siwalik region and emphasizes the need for multidisciplinary efforts in salvage palaeontology to mitigate anthropogenic threats.