Starting my current post doc, I found it was unintuitive to understand how parameters of the Earth’s magnetic field are spread upon the Earth’s surface. I struggled to illicit a clear understanding from reading the literature and so, to get to grips with it, I approached it as a practical problem to be solved. I built a visualisation tool to help me see for myself where different combinations of magnetic parameters exist, and this unexpectedly revealed potential issues in a commonly used method. This gave rise to our paper published in Communications Biology.
My current work with Prof. Richard Holland and Dr Oliver Lindecke explores how bats may sense and use the Earth’s magnetic field. I started this post doc position with little knowledge of either bats or magnetic fields, and so had some catching up to do. I have a background in electronic engineering which tends to push me towards practical rather than theoretical approaches to learning. To get to grips with how magnetic parameters arrange themselves on the Earth’s surface, I built a tool in MATLAB that allowed me to input magnetic field values to see which locations on Earth these values exist. Researchers investigating animal behaviour and magnetic fields tend to be interested in three parameters: magnetic total intensity, magnetic inclination, and magnetic declination. These parameters in combination (or alone) may (or may not) be utilised by animals to help them navigate long (or perhaps short) distances. Very generally speaking, magnetic total intensity and inclination both change as you move North and South – a latitudinal gradient of change. Magnetic declination on the other hand has shifting gradients of change depending on where on Earth you are. If you stay in Western Europe, declination (mostly) changes as you move East and West – a longitudinal gradient of change.
In an idealised scenario, a bi-coordinate system might consist of one parameter indicating latitude, and the other longitude. In Western Europe then, paring a value of either total intensity or inclination with a value of declination would do a good job of localising: it would give a location that would be intuitive and unique. However, declination is not a straightforward magnetic parameter – an animal can only calculate it by comparing the angle between the sun (or some other celestial cue) and magnetic North. Very few studies of animal magnetoreception consider declination or include celestial cues. This means that declination cannot be determined. Instead, the majority of studies use only total intensity along with inclination.
In order to be able to conduct our own experiments, I wanted to understand how to use only these two parameters in a magnetic map given that their gradients of change are so similar. Using the MATLAB tool, I plotted values used in virtual displacements from published studies. It is no revelation that there are areas where the isolines of inclination and intensity are near parallel, rendering bi-coordinate navigation unlikely (Boström et al.,s 2012; Lohmann and Lohmann, 1994). Despite this, it appeared to be ubiquitous that there were many possible locations for virtual displacements that were not shown in the methods or discussed in the results. With support from my co-authors (also including Dr Florian Packmor), I went ahead and investigated every study we could find. This turned into a manuscript, with our main finding that alternative possible locations were always present when only inclination and total intensity were used in virtual displacements.
The presence of other possible locations is not necessarily a problem, often these locations are some distance away from the intended location, and therefore it is unlikely that they would be confused with each other. This was not always the case though. We hope that this paper alongside our MATLAB tool can help researchers to both design their virtual magnetic displacement experiments, and to fully interpret their results in the light of all the possible locations. The tool is available here: ViMDAL - File Exchange - MATLAB Central (mathworks.com)