The climate dependence of the terrestrial carbon cycle, including parameter and structural uncertainties

Abstract

Abstract.The feedback between climate and the terrestrial carbon cycle will be a key determinant of the dynamics of the Earth System (the thin layer that contains and supports life) over the coming decades and centuries. However, Earth System Model projections of the terrestrial carbon-balance vary widely over these timescales. This is largely due to differences in their terrestrial carbon cycle models. A major goal in biogeosciences is therefore to improve understanding of the terrestrial carbon cycle to enable better constrained projections. Utilising empirical data to constrain and assess component processes in terrestrial carbon cycle models will be essential to achieving this goal. We used a new model construction method to data-constrain all parameters of all component processes within a global terrestrial carbon model, employing as data constraints a collection of 12 empirical data sets characterising global patterns of carbon stocks and flows. Our goals were to assess the climate dependencies inferred for all component processes, assess whether these were consistent with current knowledge and understanding, assess the importance of different data sets and the model structure for inferring those dependencies, assess the predictive accuracy of the model and ultimately to identify a methodology by which alternative component models could be compared within the same framework in the future. Although formulated as differential equations describing carbon fluxes through plant and soil pools, the model was fitted assuming the carbon pools were in states of dynamic equilibrium (input rates equal output rates). Thus, the parameterised model is of the equilibrium terrestrial carbon cycle. All but 2 of the 12 component processes to the model were inferred to have strong climate dependencies, although it was not possible to data-constrain all parameters, indicating some potentially redundant details. Similar climate dependencies were obtained for most processes, whether inferred individually from their corresponding data sets or using the full terrestrial carbon model and all available data sets, indicating a strong overall consistency in the information provided by different data sets under the assumed model formulation. A notable exception was plant mortality, in which qualitatively different climate dependencies were inferred depending on the model formulation and data sets used, highlighting this component as the major structural uncertainty in the model. All but two component processes predicted empirical data better than a null model in which no climate dependency was assumed. Equilibrium plant carbon was predicted especially well (explaining around 70 % of the variation in the withheld evaluation data). We discuss the advantages of our approach in relation to advancing our understanding of the carbon cycle and enabling Earth System Models to make better constrained projections.

Data-based modelling and environmental sensitivity of vegetation in China

“…With huge advances in the availability of relevant observations to constrain models, and in the size of problems that can now be tackled using statistical methods, there is considerable scope to develop relatively simple models, which are informed by process understanding but also firmly based on observations (see e.g. Smith et al, 2013).…”

Changing How Earth System Modeling is Done to Provide More Useful Information for Decision Making, Science, and Society

“…For example, increasing the spatial resolution of a specific atmospheric physics model is justifiable in order to predict atmospheric dynamics more precisely (Shaffrey et al 2009;Palmer 2012) but if its computational requirements restrict the inclusion of other details then the model may be less accurate than had an alternative atmospheric model formulation been adopted to allow other component processes to be represented more accurately. The adequacy of a model structure, including the level of detail, can be assessed by the degree to which predictions can recapture the known (and relevant) dynamics of interest, although such assessments are not in widespread use (Judd et al 2008;Le Bauer et al 2013;Smith et al 2013).…”

“…Such methods will be important for propagating uncertainty into projections and enabling assessments of the value of alternative model formulations in terms of precision, accuracy, and overall confidence in how well the model captures reality (Vrught et al 2005;Berliner and Wikle 2007;Scholze et al 2007;Sexton and Murphy 2012;Le Bauer et al 2013;Smith et al 2013). New studies examining the tradeoffs between the level of model detail and the ability to quantify uncertainty would be informative in relation to this (Smith 2002;Ferro et al 2012;Palmer 2012).…”

“…To facilitate the exchange of model components, future components could be developed in such a way that facilitates their use within alternative model structures. This strategy was recently employed by Smith et al (2013) to develop a global terrestrial carbon model with the intention to facilitate investigations into the costs and benefits of alternative model components and formulations for predicting global terrestrial carbon. In that study the modeling framework included code libraries that enabled model components to obtain the data they require to make outputs from online databases, from local computers, or from other model components, depending on the structural information specified.…”

Patch-Based Forest Change Detection from Landsat Time Series

“…Because disturbances like fire, disease, and insect outbreaks create variation in forest mortality [12,13], they can lead to poorly understood variation in carbon flux over large spatial scales [14,15]. Consequently, forest mortality and disturbance are some of the least represented parts of terrestrial ecosystem models [16,17]. Accurate methods that map historical trends in forest change at the landscape scale are necessary for monitoring, establishing baseline levels of forest change, and parameterizing models of ecosystem dynamics.…”