| Summary: | It is well established that the terrestrial biosphere is currently a net carbon sink, driven by increases in productivity, but the mechanisms involved remain uncertain. Observations indicating large-scale increases in productivity across northern terrestrial ecosystems include aircraft data that show an increase in the amplitude of the seasonal cycle of CO2 (ASC) between 45-90N of 56±9.8% over the last 50 years (1958-61 - 2009-11), and satellite observed greening trends of 0.25-0.5%yr 1 between 30-90N since the 1980s. I analyse output from 13 terrestrial biosphere models taking part in the Multi-scale Synthesis and Terrestrial Model Intercomparison Project (MsTMIP) against these observations, which together provide a powerful benchmarks for models. I also validate a simple, data-driven model for photosynthesis, the P-model, which I use to further examine changes in productivity. New approaches are also required for analysing how model output relates to real-world observations. Thus, I additionally use the metric of ecosystem light-use e fficiency (eLUE), the ratio of ecosystem productivity to the light absorbed, to distinguish structural (i.e. greening) from physiological (i.e. rate of carbon fixation) vegetation changes in models on ecosystem scales. The MsTMIP models have good skill at representing observed greening trends and thus capture structural changes in vegetation, but they underestimate the observed increase in ASC, indicating that they underestimate changes in eLUE and productivity. The P-model, which is driven by observed greening, has a similar trend in productivity to the MsTMIP models and eLUE changes contribute only 34% to this trend. Although both the P-model and MsTMIP models show that productivity changes were driven primarily by CO2 fertilisation, this analysis suggests that models require better representation of CO2 fertilisation and other processes that can increase eLUE. Overall these results indicate that eLUE trends, larger than current models allow, are driving increases in productivity across high northern latitudes.
|
|---|
