An artificial neural network-based approach to identifying mammalian fossil localities in the Great Divide Basin, Wyoming

Abstract

Successful identification of fossil-bearing sedimentary deposits in the field typically requires expert knowledge in geology and anatomy and some degree of luck. One way to reduce the role of serendipity is to develop an empirical model that increases the likelihood of locating productive fossil-bearing deposits by identifying combinations of geological, geospatial and spectral features that are common to productive localities. In this example, a neural network classifier successfully identified Eocene mammalian fossil localities in the Great Divide Basin, Wyoming. This approach has broad implications for many other types of anthropological field research that also involve unique geospatially distributed phenomena.