Quantifying effects of tracking data bias on species distribution models

Abstract

Telemetry datasets are becoming increasingly large and covering a wider range of species using different technologies (GPS, Argos, light‐based geolocation). Together, such datasets hold tremendous potential to understand species' space use at broad spatial scale, through the development of species distribution or habitat suitability models (SDMs) to predict environmental dependencies of species across space and time. However, tracking datasets can be heavily biased and an assessment of how such biases affect SDM predictions, and therefore, our interpretation of animal distributions is lacking. We generated simulated tracks based on predetermined environmental values for a random predator and a central place forager, and then sampled positions from those tracks based on a combination of five common biases in tracking datasets: (a) tagging location; (b) tracking device; (c) data gaps within tracks; (d) premature tag detachment (or failure) and (e) different processing methods. We then used 240 combinations of the resulting biased simulated datasets to develop binomial generalised linear (GLM) and additive (GAM) models to estimate habitat suitability in different environmental sets (cool deep, cool coastal, warm deep and warm coastal environments). Our results show that tagging location and length of tracks have the largest effects in decreasing model performance, but that these biases can be overcome by adding a small percentage of additional, relatively less biased tracks to the dataset. In comparison, the effects from all other biases were almost negligible, including for low resolution tracking datasets for which sufficient tracks are available. We also highlight the need for a cautionary approach when using processing methods that can introduce other biases (e.g. interpolated locations). Similar trends were obtained for the random predator and the central place forager, but with relatively lower model performance for the latter. We provide evidence that even non‐GPS tracking datasets can be readily used to improve the knowledge of large‐scale space use by species without the need for detailed processing and tracking reconstruction. This is especially relevant in the current context of rapid increase in data acquisition and the urgent need to address the large spatial scale ecological consequences of global change.

Winter distribution of juvenile and sub-adult male Antarctic fur seals (Arctocephalus gazella) along the western Antarctic Peninsula

“…However, simulated tracks can generate replication at the same locations of the real track, hence leading to contradictory information in binomial models (i.e. same location and date defined as either presence and absence) and potentially reduce model performance 54 . To reduce the amount of pseudo-replication and prevent overlap between real and simulated tracks, we gridded all presence and pseudo-absence locations per individual at 0.1 degrees on a daily basis and filtered out pseudo-absences that were adjacent to any presence grid cell (i.e.…”

Where did they not go? Considerations for generating pseudo-absences for telemetry-based habitat models

“…Separation in environmental niche space may dominate any differences between pseudo-absence generation approaches. For example, [ 51 ] found CRWs were less successful than background sampling. However, the study used CRWs only within the species’ domain and background sampling from outside the species’ domain to understand habitat use.…”

A standardisation framework for bio‐logging data to advance ecological research and conservation

“…For this step, a common temporal resolution and grid-cell size should be defined aiming to have standardised Level IV products readily available. This common spatiotemporal resolution could be monthly at 1 degree x 1 degree grid-cell sizes to reduce data gaps in environmental data collected by satellites, such as chlorophyll-a (Scales et al, 2017), and following results from other recent literature (Amoroso et al, 2018;Kroodsma et al, 2018aKroodsma et al, , 2018bO'Toole et al, 2020). This gridding step should be applied to data Levels II and III to, respectively, produce Levels IVa (gridded curated data) and IVb (gridded interpolated data).…”