Drivers of Spatial and Temporal Variability in Vegetation Productivity on the Yamal Peninsula, Siberia, Russia

Abstract:

Arctic primary productivity has been impacted by climate change over recent decades. Despite an overall increase in remotely sensed primary productivity since 1982, the direction and magnitude of these trends exhibit spatiotemporal heterogeneity across multiple scales (local to continental, inter-annual to decadal). While the effects of climatic, geologic, biological, and anthropogenic drivers on Arctic vegetation productivity have been previously analyzed, there are uncertainties in determining their relative importance across regions with heterogeneous climatological and landscape conditions, and vegetation responses. This research aims to disentangle the effects of environmental drivers on tundra productivity across the Yamal Peninsula, a region that is characterized by steep climate and geological gradients, ice-rich continuous permafrost, diverse Arctic vegetation, indigenous reindeer herding, and gas extraction. These conditions allow for the Yamal Peninsula to serve as a case study for the determination of the critical drivers of productivity in Arctic regions. This analysis used multiannual time series of satellite-derived peak growing season (Max) NDVI and climatic, geologic, biological, and anthropogenic variables to determine the relative influence of each driver on the magnitude and direction of the Yamal Peninsula Max NDVI trend between 2001 and 2018, as well as the spatial distribution of Max NDVI. Max NDVI increased across a majority (55%) of the Yamal Peninsula despite decreases throughout the surrounding region (northwestern Siberia) since 2011. Areas of decreased Max NDVI were frequently associated with gas fields and their adjacent settlements. Climatic drivers (Summer Warmth Index, mean growing season precipitation, and snow-free period onset date) were weakly spatially correlated with Max NDVI and explained only 26.4% of the variance in Max NDVI change over the study period. However, older landscapes (>25,000 years), circumneutral substrates, and graminoid-dominated physiognomic vegetation units were significantly associated with an increase in Max NDVI. These results emphasize how the effects of climatic drivers on Max NDVI can potentially be enhanced, limited, or reversed due to interactions with geologic, biological, and anthropogenic drivers.