A test of functional convergence in carbon fluxes from coupled C and N cycles in Arctic tundra

Abstract

Arctic ecosystems exhibit functional convergence in Net Ecosystem Exchange (NEE’s) of CO2 response to light, air temperature, the Normalized Difference Vegetation Index (NDVI), and Leaf Area Index (LAI), potentially simplifying predictions of climate change impacts on the arctic C cycle over space and time. This convergence is hypothesized to be induced by tightly coupled carbon and nitrogen cycles, but has never been explicitly tested. We used model-data fusion on a mass balance model (i.e. the Coupled Carbon and Nitrogen; CCaN) to determine whether functional convergence in NEE results from tightly coupled carbon (C) and nitrogen (N) cycles. CCaN captured a majority, but not all, of NEE and NDVI observations across eight growing seasons, and MODIS NDVI observations across a tundra latitudinal gradient in Alaska. The hypothesis of temporal functional convergence was challenged by model-data disagreements for NEE during shoulder seasons and low light. The hypothesis of spatial functional convergence was challenged by the underestimation of NDVI in warm tundra. CCaN structure and parameter uncertainty analyses revealed that factors controlling leaf litter rate, the proportion of N in leaves, and net N mineralization rate are critical knowledge gaps in predicting pan-arctic NEE and NDVI in a future warmer climate.