COS 276-3 - CANCELLED - Scaling-up vegetation properties with remote sensing to improve predictions of carbon fluxes in montane ecosystems

Abstract: Terrestrial ecosystems are experiencing rapid changes in vegetation cover related to climate changes, land use, and drought, but the consequences of these changes on the water and carbon cycle are still unclear. Montane meadows are ecosystems where soils are carbon-rich, and show sharp transitions in ecosystem structure and composition, due to the high extreme climatic conditions that the biota experiences such as snowmelt date, temperature seasonality, and length of the growing season. Alpine plant communities are strongly associated with high heterogeneity in microclimatic conditions and topographic variability, but the consequences of this variation for ecosystem fluxes remains unclear. Although, vegetation properties such as plant traits, species diversity, functional diversity are usually sampled at local scales, remote sensing methods can facilitate scaling up vegetation properties to regional scales. We quantified above- and below-ground carbon fluxes along an elevation gradient (2700-3500 m) in 38 plots (1.3 m²) in the East River Watershed in Colorado at the peak of the growing season. We measured reflectance of plant communities at the leaf-level, canopy and from the airplane. We then evaluate whether spectral diversity, plant traits, and vegetation indices were able to predict rates of carbon fluxes. Specifically, we quantified net ecosystem exchange (NEE), gross primary productivity (GPP), ecosystem (Reco) and soil respiration (SoilR). We found that at the leaf and canopy level, hyperspectral-derived traits such as specific leaf area and lead dry matter content explained the most variation of above-ground fluxes (NEE and GPP), while at the scale of the airplane (1 m² pixel size) vegetation indices were the strongest predictors and explained over 70% of the variation in the rates of NEE and GPP. Overall, changes in below-ground fluxes were poorly explained by either trait spectra or vegetation indices at the leaf and or using reflectance from the airplane, but the reflectance from the canopy alone explained more than 50% of the variation in the rates of Reco and SoilR, which suggest a potential to predict below-ground fluxes using proximal remote sensing. Our findings highlight the potential of vegetation indices to predict carbon fluxes in alpine meadows, and improve our understanding on how changes in the spectral resolution affect our ability to predict carbon fluxes in alpine communities. Future analyses will evaluate the effect of trait spectra and vegetation indices on water fluxes and the potential of scaling-up models to predict carbon and water cycling.