This blog post is provided by Felipe A. Toro-Cardona, Juan L. Parra and Octavio R. Rojas-Soto and tells the #StoryBehindthePaper for the paper “Predicting daily activity time through ecological niche modeling and microclimatic data”, which was recently published in Journal of Animal Ecology. In their paper they explore how the Gila monster’s daily activity is impacted by microclimate, and find that daily activity varies between seasons, so traditional methods are not effective models for this species, but using different temporal scale data can help predict daily activity.

This blog post is provided by Felipe A. Toro-Cardona, Juan L. Parra and Octavio R. Rojas-Soto and tells the #StoryBehindthePaper for the paper “Predicting daily activity time through ecological niche modeling and microclimatic data”, which was recently published in Journal of Animal Ecology. In their paper they explore how the Gila monster’s daily activity is impacted by microclimate, and find that daily activity varies between seasons, so traditional methods are not effective models for this species, but using different temporal scale data can help predict daily activity.

All species show a time preference for feeding and breeding; migratory birds usually breed during the spring, some butterflies only feed during their caterpillar phase, and public knowledge tells us that the likelihood of fishing increases at dawn or dusk. Two features that explain these preferences are species’ physiological limits and microclimatic conditions. The activity of the species varies at different temporal scales; but most research has concentrated on annual or seasonal variation.

Temporal patterns of activity have been usually studied between seasons through radio tracked animals, camera traps, or using ecological niche models with monthly resolution. These models relate climatic variables with georeferenced observations of the species to generate a prediction of the suitable environmental conditions for each species. These models are used to study biodiversity in the context of climate change, conservation, and invasions. However, activity patterns are rarely considered.

Our study began using standard methodologies to model the niche of the Gila Monster (Heloderma suspectum) recognizing its seasonal activity pattern. These methods use the month or season in which the species is most active as a reference, assuming that the climate in that moment (year, month, or season) is the most suitable for the species. In this way, a prediction is obtained for each month or season. We made a first approximation for the Gila monster using May as the time of breeding activity. However, this first approximation did not show adequate results since predictions were not accurate for other months (e.g. August).

After a bibliographic review of the ecology of the Gila monster, we found that the species presents hourly activity patterns that vary between seasons. In spring, the species presents a peak of activity in the morning between 9 and noon. In summer, due to the high temperatures, the species altered its activity in two parts, a diurnal one from 8 a.m. to 11 a.m. and a crepuscular/nocturnal one from 4:00 p.m. to 9:00 p.m. Finally, when winter began, its activity decreased considerably, limiting itself to a few hours between 11 am and 1pm. This temporal variation in climate made it difficult to make predictions using previous methods.

One possible solution to adequately model the niche of the Gila monster, was to consider its daily seasonal pattern of activity. With this purpose in mind, we first obtained all the records of the species. We then divided our data into a training set that included only museum specimens without the time of capture, and a testing set consisting of citizen science observations with the time of each record. For each record, we simulated its microclimate (minimum and maximum temperature and relative humidity at ground level) for each hour of the day throughout a year. Once the microclimate values ​​for each record were obtained, we used the training set to generate the niche models using minimum volume ellipsoids. These ellipsoids represent the microclimatic space used by the Gila monster and may be interpreted as an approximation of its fundamental niche.

The model generated with the microclimatic data from all activity hours of the different months included 95% of the citizen science records. In contrast, the model with microclimatic data only from May only included 13%. Further, the best model allowed us to recover the spring and winter unimodal pattern of daily activity and a bimodal pattern during the summer. We observed greater microclimatic suitability in the hours of greatest activity reported in previous studies. In addition, the peaks of activity and inactivity are consistent with and may be regulated through the species’ physiological tolerance to high temperatures. In other words, if the weather gets too hot (> 37°C on the ground at a particular time of the day), Gila monsters crawl into their burrows.

We conclude that it is possible to rescue the daily activity patterns at an hourly level with niche models and that the estimation of the niche breadth can vary considerably depending on the temporal resolution of environmental variables used. This has interesting implications and applications for studies of climate change, conservation proposals, and even sampling designs anywhere on the planet.

Read the paper

Read the full paper here: Toro-Cardona, F.A., Parra, J.L. and Rojas-Soto, O.R. (2023), Predicting daily activity time through ecological niche modeling and microclimatic data. J Anim Ecol. Accepted Author Manuscript. https://doi.org/10.1111/1365-2656.13895