Highlights:• Availability of fast and precise ground and drone-borne magnetic surveys.• Extraction of information on the sources through potential field transforms.

No abstract

<p>Since the 70’s, ship-mounted three-component magnetometers are used for marine geophysical mapping, with the benefits of being able to be operated permanently with a minimum of technical maintenance. However, to obtain accuracies similar to those of ship-towed absolute scalar magnetometers, the intense interfering magnetic fields generated by the hull and steel parts of the ship have to be removed. The most common correction method, called “vector compensation”, uses high precision inertial navigation systems in order to correct the measured data for the ship’s magnetic field and calculate the vector of the compensated magnetic field in the Earth coordinated system.</p><p>This work alternatively uses the “scalar compensation” method applied in airborne magnetism since the 60’s. The aim is to compute the intensity of the compensated magnetic field without measurements of the attitude of the vector and using linear least-square regression analysis. This correction method is applied to shipboard three-component magnetometer data acquired on different vessels during different surveys. Results are compared to those obtained with ship-towed absolute scalar magnetic measurements.</p><p><strong>Keywords</strong>: shipboard three-component magnetic measurements; magnetic compensation; marine magnetics.</p>

<p><span>Ground magnetic surveys are commonly used for imaging near-surface structures in archaeological studies. Usually, surveys are conducted using vertical component gradiometers or scalar gradiometers to produce a vertical pseudo-gradient map. Scalar magnetometers can also be used, albeit less frequently, to produce maps of the total magnetic anomaly. In all these cases, the equipment is pushed or pulled by an operator or carried behind a vehicle. Here we present a third approach made available by the use of three-component fluxgate magnetometers: fast surveys over large areas using a compact lightweight drone flying automatically 1 to 2 m above the ground and high precision surveys acquired by an operator 0,2 to 1 m above the ground. A case study on the gallo-roman site of Oedenburg, </span><span>located</span><span> along the Rhine River in its upper valley, illustrates the results that can be obtained with the approach. A comparison with previously acquired pseudo-gradient surveys show</span><span>s</span><span> that the presented method allows a faster coverage, a greater resolution for the imaging of short wavelength structures (such as walls) and a better capacity of imaging large wavelength structures (such as pathways, palaeochannels or soil composition variations). As the site is crossed by a high voltage electric power line, a method to suppress the high-amplitude 50 Hz frequency magnetic field is presented.</span></p>

<p>We compare marine magnetic measurements simultaneously acquired with absolute and three-component fluxgate sensors to evaluate their respective benefits for marine geophysical mapping and detection surveys.</p><p>Shom collected the data in shallow waters, in the Bay of Brest (France) and in the Iroise Sea, during two cruises in the Fall 2021. As per standard practice, an absolute Overhauser magnetometer was towed 180 m behind the 60 m-long Laplace and Lapérouse hydrographic vessels. In addition, two vector magnetometers were temporarily installed at the top of the ship’s mast and on the roof of a 10 m-long launch. Scalar data were processed following Shom’s standards: shift to sensor position, layback adjustments, removal of gyrations and spikes, filtering and calculation of magnetic anomalies by removing the IGRF model (Alken et al., 2021) and reducing external variations measured at a local reference station. Vector data were corrected for the strong magnetic fields generated by the hull and other steel components of the ship by the application of a “scalar compensation” using a least-squares regression analysis (Leliak, 1961) on data from figures of merit. The compensated vector data then need to be low-pass filtered to remove uncorrected variations of attitude and heading. Magnetic anomalies were finally computed by removing the median value for each profile and reducing external variations from the same local reference station.</p><p>Our first results show that maps of total-field anomalies derived from vector data acquired on the ship are very close to those of the absolute data upward-continued to the altitude of the mast. This similarity suggests that it is possible to perform good-quality magnetic surveys without the constraint of having to tow an instrument. The different processing steps however raise the detection threshold for anthropogenic objects lying on the seafloor or partially buried. Vector data acquired on smaller launchs are much more complicated to compensate as ranges of pitch, roll and heading variations are greater than for a large ship and potentially imperfectly sampled by the figures of merit.</p>