Coupling between carbon cycling and climate in a high-elevation, subalpine forest: a model-data fusion analysis

Fundamental questions exist about the effects of climate on terrestrial net ecosystem CO2 exchange (NEE), despite a rapidly growing body of flux observations. One strategy to clarify ecosystem climate-carbon interactions is to partition NEE into its component fluxes, gross ecosystem CO2 exchange (GEE) and ecosystem respiration (R-E), and evaluate the responses to climate of each component flux. We separated observed NEE into optimized estimates of GEE and R-E using an ecosystem process model combined with 6 years of continuous flux data from the Niwot Ridge AmeriFlux site. In order to gain further insight into the processes underlying NEE, we partitioned R-E into its components: heterotrophic (R-H) and autotrophic (R-A) respiration. We were successful in separating GEE and R (E), but less successful in accurately partitioning R-E into R-A and R-H. Our failure in the latter was due to a lack of adequate contrasts in the assimilated data set to distinguish between R-A and R-H. We performed most model runs at a twice-daily time step. Optimizing on daily-aggregated data severely degraded the model's ability to separate GEE and R-E. However, we gained little benefit from using a half-hourly time step. The model-data fusion showed that most of the interannual variability in NEE was due to variability in GEE, and not R-E. In contrast to several previous studies in other ecosystems, we found that longer growing seasons at Niwot Ridge were correlated with less net CO2 uptake, due to a decrease of available snow-melt water during the late springtime photosynthetic period. Warmer springtime temperatures resulted in increased net CO2 uptake only if adequate moisture was available; when warmer springtime conditions led into mid-summer drought, the annual net uptake declined.