A Cosmic View of 'Tundra Gardens': Satellite Imagery Provides a Landscape-Scale Perspective of Arctic Fox Ecosystem Engineering.

Most animal ecology studies using remote sensing data have assessed how environmental characteristics shape animal abundance, distribution, or behavior. But the increasing availability of high-resolution data offers new opportunities to study how animals, in turn, shape ecosystems at diverse scales. We evaluate the efficacy of using Sentinel-2 satellite imagery to quantify the effects of Arctic fox (Vulpes lagopus) denning activity (nutrient accumulation, bioturbation) on vegetation. Using an imagery-derived metric (NDVI), we compared maximum plant productivity and plant phenology patterns on 84 Arctic fox dens vs. reference sites, i.e., points generated within preferred denning habitat areas (predicted from a habitat selection analysis). We show that high-resolution imagery can be used to measure the effects of Arctic fox denning activity on vegetation. Plant productivity and the rate of green up were both greater on fox dens compared to reference (preferred-habitat) sites. Productivity on reference sites was lower than average productivity on the tundra (i.e., random sites), indicating foxes primarily establish dens in low-productivity areas. Plant productivity on dens was also unrelated to recent occupancy patterns, indicating fox denning activity has long-term legacy effects on plants that last beyond the lifetime of foxes. Our findings support Arctic foxes being classified as ecosystem engineers in low-Arctic tundra ecosystems by converting low-productivity sites into relatively high-productivity sites through their denning activity. We demonstrate the efficacy of using remote sensing technologies to study how predators increase landscape heterogeneity and influence ecosystem dynamics through patch-scale mechanisms, and ultimately advance our understanding of animal functional roles. [ABSTRACT FROM AUTHOR]