Modeling and monitoring changes in the terrestrial carbon cycle

Each year the terrestrial biosphere absorbs about one third of anthropogenic CO2 emissions and acts as a net sink of carbon from the atmosphere. The net land sink is considered an ecosystem services because removing and storing CO2 from the atmosphere mitigates global climate change. However, there is great uncertainty in the future of the net land sink under changing climate conditions. Therefore, the need to gain a better understanding of processes that control the rate and spatial distribution of land-atmosphere carbon exchange is evident. Since there are no spatially continuous observations of carbon dynamics at the global-scale, we use models as tools to help us address these needs. There are a variety of modeling approaches, each making different assumptions about processes that control carbon dynamics and how those processes are represented in a model. As a result, large spread in model estimates exists. There are no direct observations of carbon dynamics comparable to model estimates (e.g., 0.5° by 0.5° globally gridded and sub-daily to monthly), however atmospheric CO2 measurements provide a potentially powerful observational constraint because they offer an integrated view of surface sources and sinks of carbon. I assess a group of models by comparing them against satellite observations of CO2. I show that models are consistent with the general patterns of satellite-based CO2 measurements, indicating that models can simulate the key drivers of terrestrial carbon exchange. Additionally, I show that models that explicitly include fire disturbance and nitrogen cycling perform better than models that do not include either of these processes, suggesting that these processes are important in controlling land-atmosphere carbon dynamics and need to be included in models.;Modeling can give us a better idea of the processes controlling the net land sink, however humans also play a role in terrestrial carbon dynamics. Emissions from anthropogenic land-use change partially offset the land sink and can also decrease the strength of the land sink over time through degradation. Planned management activities have recognized the importance of the net land sink in mitigating global climate and thus develop payments to preserve the ecosystem service it provides. However, planned management activities require cost effective and accurate monitoring at a variety of scales in order to be effective. I review current approaches of monitoring carbon dynamics including ground-based professional and community monitoring, remote sensing, and modeling. Each method has strengths and weaknesses, which vary based on implementation cost, resolution, and the data they provide. However, none of these measuring techniques alone can satisfy all the needs of planned management activities (cost effective, local-scale to national-scale monitoring). This has highlighted the need for a comprehensive monitoring system that can fill the gap currently existing in monitoring planned management activities. Accurate monitoring becomes especially critical when considering the ambiguity in future human behaviors (land-use change and fossil fuel emissions) and how this may affect the net land sink and the ecosystem service it provides.