Levelling of helicopter-borne frequency-domain electromagnetic data Bernhard Siemon ⁎ Federal Institute for Geosciences and Natural Resources (BGR), Stilleweg 2, D-30655 Hannover, Germany Received 30 January 2007; accepted 12 November 2007 Abstract For about three decades helicopter-borne electromagnetic (HEM) measurements have been used to reveal the resistivity distribution of the upper one hundred metres of the earth's subsurface. HEM systems record secondary fields, which are 3–6 orders of magnitude smaller than the transmitted primary fields. As both the primary fields and the secondary fields are present at the receivers, well-designed bucking coils are often used to reduce the primary fields at the receivers to a minimum. Remaining parts of the primary fields, the zero levels, are generally corrected by subtracting field values recorded at high altitudes (standard zero levelling) or estimated from resistivities of neighbouring lines or from resistivity maps (advanced zero levelling). These zero-levelling techniques enable the correction for long-term, quasi-linear instrumental drift. Short-term variations caused by temperature changes due to altitude variations, however, cannot be completely corrected by this procedure resulting in stripe patterns on thematic maps. Statistical methods and/or 2-D filter techniques called statistical levelling (tie-line levelling) and empirical levelling (microlevelling), respectively, used to correct stripe patterns in airborne geophysical data sets are, in general, not directly applicable to HEM data. Because HEM data levelling faces the problem that the parameter affected by zero-level errors, the secondary field, differs from the parameter generally levelled, the apparent resistivity. Furthermore, the dependency of the secondary field on both the resistivity of the subsurface and the sensor altitude is strongly nonlinear. A reasonable compromise is to microlevel both half-space parameters: apparent resistivity and apparent depth, followed by a recalculation of the secondary field components based on the half-space parameters levelled. Advantages and disadvantages of the diverse levelling techniques are discussed using a HEM data set obtained in a hilly region along the Saale River between the cities of Saalfeld and Jena in central Germany. It turns out from a comparison of apparent resistivity and apparent depth maps derived from levelled HEM data that manually advanced zero levelling of major level errors and automatic microlevelling of remaining minor level errors yield the best results. © 2007 Elsevier B.V. All rights reserved. Keywords: Airborne geophysics; Helicopter-borne electromagnetics; HEM data; Levelling; Map production; Apparent resistivity 1. Introduction Helicopter-borne frequency-domain electromagnetics (HEM) has proved to be a successful tool for geological and hydro- geological mapping, mineral and groundwater exploration and a number of environmental investigations during the last thirty years. As the requirements for accurate resistivity models increase more and more, both improved airborne systems and sophisti- cated processing and interpretations techniques are needed. Most of the modern HEM systems are multi-frequency devices operating at 4–6 frequencies ranging from 200 Hz to 200,000 Hz. Data processing comprises the preparation of secondary field values including calibration, data correction, levelling and filtering (e.g. Valleau, 2000), the inversion to homogeneous and/or layered half-space resistivity models (e.g. Sengpiel and Siemon, 2000), and the presentation of the resistivity models as resistivity maps (e.g. Fraser, 1978) and cross-sections (e.g. Sengpiel, 1990; Sengpiel and Siemon, 1998). Airborne surveys are generally flown on parallel lines (also called survey lines, profile lines, traverse lines or acquisition Available online at www.sciencedirect.com Journal of Applied Geophysics 67 (2009) 206 – 218 www.elsevier.com/locate/jappgeo ⁎ Tel.: +49 511 643 3488; fax: +49 511 643 3663. E-mail address: bernhard.siemon@bgr.de. URL: http://www.bgr.bund.de/. 0926-9851/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jappgeo.2007.11.001