A wide range of environmental variables are periodically quantified in gridded format through remote sensing. They include attributes of the biosphere, the cryosphere, and meteorology, and range in spatial resolution from sub-meter to kilometers and in temporal resolution from hours to weeks with usually a strong trade-off between both. Today, remote sensing data are the prime source of information to describe landscape dynamics in space as well as time, generating new possibilities to analyze the links between the environment and animal behavior.

We used data from the space-borne MODerate resolution Imaging Spectroradiometer (MODIS) and Tropical Rainfall Measuring Mission (TTRM) to quantify spatiotemporal changes in a variety of seasonal ecosystems that host mammal migrations. By combining these data with animal locations acquired through radio and GPS collars we show how they can be used to identify patterns and drivers in migratory behavior.

Results/Conclusions

Vegetation indices derived from MODIS suggest that the altitudinal migration Giant Pandas (Ailuropoda melanoleuca) make in spring in the Qinling Mountains in China is timed to coincide with the sequential development of shoots in the bamboo species they forage on. Golden Takin (Budorcas taxicolor), the other large grazer in the ecosystem, time their spring migration earlier than the Pandas in correspondence with the earlier spring phenology of their forage species, and as indicated by a distinct pattern in the remotely sensed vegetation index.

Elephants (Loxodonta africana) in the Marsabit area in Kenya also make a seasonal altitudinal migration. Here, the remote sensing data indicate a strong spatial segregation between the availability of water and the quality of forage. During the dry seasons, the elephants are confined to a small high-altitude forest with year-round water but they descend to the lower-elevation scrublands as soon as rain provides water there and causes vegetation to flush, providing nutrient-rich forage.

A similar pattern emerges in the migration of Zebras (Equus quagga) in Southern Africa, with animal movement very closely matching the rain-driven changes in vegetation, as captured by MODIS and TTRM. We show how the seasonal landscape changes that drive the very long migrations of the Zebras, (over 300 km, but also those behind the shorter migrations of the Elephants (20-90 km) and Giant Panda’s (3 km) can be accurately captured by the analysis of remote sensing time series. As such, it proves very valuable to visualize and measure the agreement between environmental and migratory dynamics, and consequently the vulnerability of the latter to environmental change.