Skip to main content
Dryad logo

Data and code from: Fruit bat migration matches green wave in seasonal landscapes

Citation

Hurme, Edward et al. (2022), Data and code from: Fruit bat migration matches green wave in seasonal landscapes, Dryad, Dataset, https://doi.org/10.5061/dryad.z08kprrdb

Abstract

Migrating grazers and carnivores respond to seasonal changes in the environment and often match peaks in resource abundance. However, it is unclear if and how frugivorous animals use phenological events to time migration, especially in the tropics. The straw-colored fruit bat (Eidolon helvum), Africa’s most gregarious fruit bat, forms large seasonal colonies throughout much of sub-Saharan Africa. We hypothesized that aggregations of E. helvum match the timing of their migration with phenologies of plant growth or precipitation. Using monthly colony counts from across much of the species’ range, we matched peak colony size to landscape phenologies and explored the variation among colonies matching the overall closest phenological event. Peak colony size was closest to the peak instantaneous rate of green-up, and sites with closer temporal matching were associated with higher maximum greenness, short growing season, and larger peak colony size. Eidolon helvum seem to time their migrations to move into highly seasonal landscapes to exploit short-lived explosions of food and may benefit from collective sensing to time migrations. The link between rapid changes in colony size and phenological match may also imply potential collective sensing of the environment. Overall decreasing bat numbers along with various threats might cause this property of large colonies to be lost. Remote sensing data, although, indirectly linked to fruiting events, can potentially be used to globally describe and predict the migration of frugivorous species in a changing world.

Methods

Colonies of Eidolon helvum bats were counted by members of the Eidolon Monitoring Network (www.eidolonmonitoring.com) at seventeen sites across Africa. Most colonies were counted by observing the bats at their roost site and multiplying the average number of bats in a cluster, the average number of clusters on a branch, the average number of occupied branches in a tree, and the number of occupied trees at the roost site (Hayman et al. 2012; Fahr et al. 2015). For roosts that were inaccessible without disturbing the colony, observers would count numbers of passing bats during evening emergence and extrapolate their numbers by the area monitored (Sørensen & Halberg 2001).

The table includes the longitude and latitude coordinates of the roost, geeID (a unique ID for each location use to extract remote sensing values from Google Earth Engine), country, location (city or park), date of the count, count estimate, and observer names.

We calculated the ratio of the colony size as the fraction of the maximum colony size for that site across a moving window of all counts within a year, before and after each count.

Usage Notes

See the attached ReadMe file and see https://github.com/ehurme/EidolonGreenWave and www.eidolonmonitoring.com for the latest updates to the code and counts.