Visualizing migration patterns in 3D helps inform causal mechanisms

Background/Question/Methods

Despite centuries of research on the subject we still don’t have a consensus view of the mechanisms dictating animal migration.  Classic experiments have shown that animals use sunlight, moonlight, day length, resource availability, and biological interactions as both cues and guides for long-distance and repeated navigation.  Contemporary works have focused on experimentally isolating these mechanisms and assigning relative importance to each. A fixed observational frame of reference is central to a complete understanding of observed ecological patterns and I think it can help the field distinguish between current competing hypotheses. As a tool to test this, I have developed a 3-dimensional GIS and visualization tool to study global-scale processes through time and in relation to the sun.  Using the R programming language, I’ve collapsed latitude, longitude, and time into a single variable by plotting them all as spatial locations in the earths orbit.  This open source tool provides a robust method for calculating the interrelatedness of daily activity, phenology, and spatial location by mapping data to this arena for direct comparison between space and time.

Results/Conclusions

Built using NIST published standards and achieving 1-second precision, my prototype is calibrated to a Shearwater GPS dataset and layered with NOAA and NASA data for wind, NPP, temperature, precipitation, solar radiation, and light location (moon and sun).  Seeing these layers interact in 3-dimensions presents a vastly different picture than you see in 2-dimensional projections. The most-clear example of a new perspective developed through this improved visualization is the seemingly strong spatial dependence many species have on the solar terminator.  This is the line separating the light and dark half of the earth, which most terrestrial locations cross twice a day at dusk and dawn.  My shearwater prototype indicates that these migratory birds may be responding more strongly to the terminator than they are to the other factors described above.  This is a result that can only be tested in a framework that allows both controlling for space and time and visualizing data in a way that is simple and informative.  Here I present a tool that is capable of standardizing ecological parameters using a rigorous mathematical framework, while simultaneously providing high-quality data visualizations.