An application of plot-scale NDVI in predicting carbon dioxide exchange and leaf area index in heterogeneous subarctic tundra

Abstract

Accurate predictions of carbon dioxide (CO2) exchange in tundra regions are critical because of the impacts of climate change on carbon stored in permafrost. Examining the relationships between the normalized difference vegetation index (NDVI) and vegetation and CO2 exchange can contribute to understanding how small-scale variation in vegetation community contributes to remotely sensed estimates of landscape characteristics. CO2 fluxes were measured with a portable chamber in a range of tundra vegetation communities. Biomass and leaf area were measured with destructive harvest, and NDVI was derived using a hand-held infrared camera. NDVI was weakly correlated with leaf area index in only some vegetation communities and significantly correlated with biomass, including mosses. NDVI was significantly related to photosynthetic activity and net CO2 uptake in all vegetation groups, but to ecosystem respiration only in wet sedge. These results suggest that at a plot scale, NDVI shows variable success in predicting ecosystem properties and CO2 exchange in heterogeneous tundra vegetation.