Hybrid Inversion of Airborne TEM Data - Mixed Precision Responses for Tunable Performance and Accuracy

Modern airborne TEM surveys often produce extremely large datasets that can be very time consuming to invert using a full system forward model description. Often, approximate forward models are used providing very fast results at the expense of accuracy. We show how forward models of different precision can be combined in a versatile hybrid inversion, providing tunable accuracy and performance. Our study includes inversions of actual field data using different degrees of approximation. We show how this influences the resulting earth resistivity model, discuss the implications and provide performance metrics.