INVERSE MODELING OF MULTIFREQUENCY INDUCTION ELECTROMAGNETIC DATA

THE RECENT ADVENT OF A DIGITAL, BROADHAND, HAND-HELD ELECTROMAGNETIC (EM) SENSOR PROVIDES AN OPPURTUNITY FOR A SIGNIFICANT ADVANCE IN RAPID AND MEANINGFUL SITE CHARACTERIZATION FOR DEPTH RANGES AND CONDITIONS WHERE GROUND PENETRATING RADAR (GPR) IS INEFFECTIVE. THIS PROPOSAL PRESENTS A FEASIBILITY STUDY TO INVERT WIDEBAND INDUCTION EM DATA TO IMAGE SUBSURFACE FEATURES. ULIKE OTHER TECHNIQUES SUCH AS GPR, LOW FREQUENCY EM TECHNIQUES ARE SENSITIVE TO DEEPER STRUCTURES DUE TO THE GREATER SKIN DEPTH OF THE LONGER WAVELENGTH SIGNALS. THE ISSUE THIS PROPOSAL ADDRESSES IS TO INVERT THE EM DATA TO CONSTRUCT A TWO-DIMENSIONAL ELECTRICAL CONDUCTIVITY VS. DEPTH MODEL. DATA COLLECTION COULD BE ACCOMPLISHED BY A SINGLE OPERATOR, WITHOUT THE INSTRUMENT TOUCHING THE GROUND, AND SIGNAL PROCESSING COULD BE COMPLETED AUTOMATICALLY IN REAL-TIME. ELECTRICAL CONDUCTIVITY ANOMALIES, SUCH AS DUE TO FLUID-FILLED PORES OR FRACTURES, COULD THEN BE EASILY INTERPRETED IN THE FIELD.