Csengeri Piroska - Tóth Arnold (szerk.): A Herman Ottó Múzeum évkönyve 54. (Miskolc, 2015)

Régészet - Tóth Krisztina–Szabó Ádám–Homoki Balázs: Archaeological research at the Twin-barrows near Onga (Northeast Hungary in 2015)

Archaeological research at the Twin-barrows near Onga (Northeast Hungary) in 2015 277

GEOMAGNETIC SURVEY

The geomagnetic survey was carried out to the north along the road, running west-east direction north from the mounds, this is due to the vegetation which only allowed us to work there (Map 8. A). The piles were surrounded by corn in August, 2015. We used MAGNETO® DLM 5 channel device with fluxgate sensors made by Sensys company.

Using a magnetometer we can detect up to 0.2 nT resolution of the fine changes of the earth’s magnetic field. The purpose of this method is to search for magnetic interference in the top layer of the soil, which can be connected in archaeological perspective to traces of human activities. It is important to note that the top layer of the soil has a greater magnetic effect than lower layers, which is due to biological (bacterial activity), physical (fragmentation) and chemical (transformation of rocks) processes (KVAMME 2006b, 214; BECKER 2009,134). Taking these characteristics of the soil into account we are able to detect places where they filled a deeper pit with soil with greater magnetic effects, or where they have taken material, clay for use. This means we can find ditches, foundations, kilns, graves and different kind of pits. However, the presence of fire, such as furnaces, burned traces of houses and fired building materials could be an important clue because they can be detected by their higher magnetic effects (KVAMME 2006b, 216—218). It is important to mention that iron objects can cause strong dipole anomalies (with two apparent poles, the positive peak appearing as white and the negative peak as black coloured). It is important to mention imported building materials as well, because sometimes their effect differ from that of the background and this could make them visible on the anomaly map, like the presence of volcanic rock for instance (KVAMME 2006b, 220).

For research we used 5 pieces of FGM-650 type fluxgate gradiometers. There are two fluxgate magnetometers in each probe, which are placed 650 mm from each other. The soul of a fluxgate magnetometer is a soft iron core, which is surrounded by two contrary coiled wires with the same number of passes. The instrument is sensitive to the magnetic field component which is parallel to the iron cores. During the survey the sensors are held vertically and the vertical component of the total magnetic field is measured. There are many benefits of using gradiometer configuration such as removing the time dependant natural changes in the magnetic field as well as total magnetic field and low frequency components — slowly changing effects like

large scale geological effects. As a result of using this configuration only the near surface anomalies will be visible on the anomaly map, essentially it works like a high pass filter.

The operation of the gradiometer configuration as follows. During the measuring the upper and the lower positioned fluxgate magnetometers in a sensor also measures the magnetic field. The theory of the falloff rule is used which says that the strength of the magnetic field is proportional to the inverse of the third power of the distance (KVAMME 2006b, 222). This means that the near surface anomalies cause stronger readings on the lower sensor than the higher one. The field computer connected to the sensors subtracts the higher positioned magnetometer’s data from the lower one. This step removes the total field, time dependant natural effects and slowly changing geological anomalies (low frequency components). The second step involves dividing the differences by the distance between the vertically separated magnetometers. This step transforms the anomaly map to a gradient map which shows the changes in the near surface magnetic field. The penetration depth depends on the depth of the source but it is generally about 0.75—1 meter. The sensors’ distance from the ground can affect the penetration depth. When we assembled the instrument we put the sensor into a plastic cartridge with 0.5 meter separation from each other. This meant that the track lines were 0.5 meter from each other. The sampling density along the track lines was set to 0.1 meter. The result of the survey is a colour coded anomaly map which shows the changes of the vertical magnetic component of the magnetic field near surface.

For processing the measurements we used Magneto®Arch, Golden Software Surfer, Snuffler and Gwyddion programs. Before beginning data processing, it must be emphasized that every filtering and processing step used to remove unwanted anomalies from the anomaly map will damage or may remove archaeologically interested anomalies as well. To prevent anomalies from being filtered, the income and outcome maps have to be compared at every processing step. As a first step of processing the data must be transferred from the memory of the instrument to the PC with Magneto®Arch. Other than data transfer, the program also has other important functions like filtering the heading error and the drift. However, sometimes the measured data ends up being staggered, which is an error that originates from the operator. This happens if the operator was not moved at constant pace so the tracks are slid — anomalies at the edges of the tracks