ALBERT

Description: Land models, which have been developed by the modeling community in the past two decades to predict future states of ecosystems and climate, have to be critically evaluated for their performance skills of simulating ecosystem responses and feedback to climate change. Benchmarking is an emerging procedure to measure and evaluate performance of models against a set of defined standards. This paper proposes a benchmarking framework for evaluation of land models. The framework includes (1) targeted aspects of model performance to be evaluated; (2) a set of benchmarks as defined references to test model performance; (3) metrics to measure and compare performance skills among models so as to identify model strengths and deficiencies; and (4) model improvement. Component 4 may or may not be involved in a benchmark analysis but is an ultimate goal of general modeling research. Land models are required to simulate exchange of water, energy, carbon and sometimes other trace gases between the atmosphere and the land-surface, and should be evaluated for their simulations of biophysical processes, biogeochemical cycles, and vegetation dynamics across timescales in response to both weather and climate change. Benchmarks that are used to evaluate models generally consist of direct observations, data-model products, and data-derived patterns and relationships. Metrics of measuring mismatches between models and benchmarks may include (1) a priori thresholds of acceptable model performance and (2) a scoring system to combine data-model mismatches for various processes at different temporal and spatial scales. The benchmark analyses should identify clues of weak model performance for future improvement. Iterations between model evaluation and improvement via benchmarking shall demonstrate progress of land modeling and help establish confidence in land models for their predictions of future states of ecosystems and climate.

Description: Existing dynamic global vegetation models (DGVMs) have a~limited ability in reproducing phenology and decadal dynamics of vegetation greenness as observed by satellites. These limitations in reproducing observations reflect a poor understanding and description of the environmental controls on phenology, which strongly influence the ability to simulate longer term vegetation dynamics, e.g. carbon allocation. Combining DGVMs with observational data sets can potentially help to revise current modelling approaches and thus to enhance the understanding of processes that control seasonal to long-term vegetation greenness dynamics. Here we implemented a~new phenology model within the LPJmL (Lund Potsdam Jena managed lands) DGVM and integrated several observational data sets to improve the ability of the model in reproducing satellite-derived time series of vegetation greenness. Specifically, we optimized LPJmL parameters against observational time series of the fraction of absorbed photosynthetic active radiation (FAPAR), albedo and gross primary production to identify the main environmental controls for seasonal vegetation greenness dynamics. We demonstrated that LPJmL with new phenology and optimized parameters better reproduces seasonality, inter-annual variability and trends of vegetation greenness. Our results indicate that soil water availability is an important control on vegetation phenology not only in water-limited biomes but also in boreal forests and the arctic tundra. Whereas water availability controls phenology in water-limited ecosystems during the entire growing season, water availability co-modulates jointly with temperature the beginning of the growing season in boreal and arctic regions. Additionally, water availability contributes to better explain decadal greening trends in the Sahel and browning trends in boreal forests. These results emphasize the importance of considering water availability in a new generation of phenology modules in DGVMs in order to correctly reproduce observed seasonal to decadal dynamics of vegetation greenness.

Description: The net ecosystem exchange of CO2 (NEE) between the atmosphere and a beech forest (Sorø, Denmark) showed significant interannual variation (IAV) over 13 years (1997–2009) of observations. The forest sequestered, on average, 157 g C m−2 yr−1, ranging from a source of 32 to a sink of 344 g C m−2 yr−1 in 1998 and 2008, respectively. The objectives of this study were to evaluate to what extent and at which temporal scale, climatic variability (through direct response) and changes in ecosystem functional properties (through biotic response) regulated the IAV in the ecosystem carbon balance. To address this question, we performed correlation analysis between the carbon fluxes and climate variables at different time scales. The response of CO2 exchange to climatic variability was significantly higher at short time scales and the limiting factors changed intra-annually. Combinations of climate anomalies in different periods of the year either intensified or attenuated the aggregated ecosystem responses, implying that the changing distribution of climate anomalies, in addition to the average climate change, could have stronger impacts on the ecosystem carbon balance in the future. A semi empirical model was used to estimate a set of parameter time series for each of the 13 years, which was considered to represent the functional properties of the ecosystem. The climate and parameter time series were applied factorially by year to quantify their relative importance for the IAV in carbon flux. At an annual time scale, as much as 77 % of the IAV in NEE could be attributed to the variation in both photosynthesis and respiration related model parameters, indicating a strong influence of functional change. The possible causes for the observed functional change could not be addressed with the available dataset. This demonstrates the need for more targeted experiments, such as long-term measurements of leaf nitrogen content. Our approach incorporated seasonal variation in the ecosystem status and demonstrated a significant role of biotic factors on the carbon dynamics in a typical temperate deciduous forest. The method can be applied at other sites to explore ecosystem behaviour across different plant functional types and climate gradients. Further, this approach showed how important it is to incorporate functional change in process based models, which could guide model development and consequently reduce the uncertainties in long-term projection of global ecosystem carbon balance.

Description: There is a growing consensus that land surface models (LSMs) that simulate terrestrial biosphere exchanges of matter and energy must be better constrained with data to quantify and address their uncertainties. FLUXNET, an international network of sites that measure the land surface exchanges of carbon, water and energy using the eddy covariance technique, is a prime source of data for model improvement. Here we outline a multi-stage process for fusing LSMs with FLUXNET data to generate better models with quantifiable uncertainty. First, we describe FLUXNET data availability, and its random and systematic biases. We then introduce methods for assessing LSM model runs against FLUXNET observations in temporal and spatial domains. These assessments are a prelude to more formal model-data fusion (MDF). MDF links model to data, based on error weightings. In theory, MDF produces optimal analyses of the modelled system, but there are practical problems. We first discuss how to set model errors and initial conditions. In both cases incorrect assumptions will affect the outcome of the MDF. We then review the problem of equifinality, whereby multiple combinations of parameters can produce similar model output. Fusing multiple independent data provides a means to limit equifinality. We then show how parameter probability density functions (PDFs) from MDF can be used to interpret model process validity, and to propagate errors into model outputs. Posterior parameter distributions are a useful way to assess the success of MDF, combined with a determination of whether model residuals are Gaussian. If the MDF scheme provides evidence for temporal variation in parameters, then that is indicative of a critical missing dynamic process. A comparison of parameter PDFs generated with the same model from multiple FLUXNET sites can provide insights into the concept and validity of plant functional types (PFT) – we would expect similar parameter estimates among sites sharing a single PFT. We conclude by identifying five major model-data fusion challenges for the FLUXNET and LSM communities: 1) to determine appropriate use of current data and to explore the information gained in using longer time series; 2) to avoid confounding effects of missing process representation on parameter estimation; 3) to assimilate more data types, including those from earth observation; 4) to fully quantify uncertainties arising from data bias, model structure, and initial conditions problems; and 5) to carefully test current model concepts (e.g. PFTs) and guide development of new concepts.

Description: Quantification of ecosystem carbon pools is a fundamental requirement for estimating carbon fluxes and for addressing the dynamics and responses of the terrestrial carbon cycle to environmental drivers. The initial estimates of carbon pools in terrestrial carbon cycle models often rely on the ecosystem steady state assumption, leading to initial equilibrium conditions. In this study, we investigate how trends and inter-annual variability of net ecosystem fluxes are affected by initial non-steady state conditions. Further, we examine how modeled ecosystem responses induced exclusively by the model drivers can be separated from the initial conditions. For this, the Carnegie-Ames-Stanford Approach (CASA) model is optimized at set of European eddy covariance sites, which support the parameterization of regional simulations of ecosystem fluxes for the Iberian Peninsula, between 1982 and 2006. The presented analysis stands on a credible model performance for a set of sites, that well represent the plant functional types and selected descriptors of climate and phenology present in the Iberian region – except for a limited northwestern area. The effects of initial conditions on inter-annual variability and on trends, results mostly from the recovery of pools to equilibrium conditions; which control most of the inter-annual variability (IAV) and both the magnitude and sign of most of the trends. However, by removing the time series of pure model recovery from the time series of the overall fluxes, we are able to retrieve estimates of inter-annual variability and trends in net ecosystem fluxes that are quasi-independent from the initial conditions. This approach reduced the sensitivity of the net fluxes to initial conditions from 47% and 174% to −3% and 7%, for strong initial sink and source conditions, respectively. With the aim to identify and improve understanding of the component fluxes that drive the observed trends, the net ecosystem production (NEP) trends are decomposed into net primary production (NPP) and heterotrophic respiration (RH) trends. The majority (~97%) of the positive trends in NEP is observed in regions where both NPP and RH fluxes show significant increases, although the magnitude of NPP trends is higher. Analogously, ~83% of the negative trends in NEP are also associated with negative trends in NPP. The spatial patterns of NPP trends are mainly explained by the trends in fAPAR (r=0.79) and are only marginally explained by trends in temperature and water stress scalars (r=0.10 and r=0.25, respectively). Further, we observe the significant role of substrate availability (r=0.25) and temperature (r=0.23) in explaining the spatial patterns of trends in RH. These results highlight the role of primary production in driving ecosystem fluxes. Overall, our study illustrates an approach for removing the confounding effects of initial conditions and emphasizes the need to decompose the ecosystem fluxes into its components and drivers for more mechanistic interpretations of modeling results. We expect that our results are not only specific for the CASA model since it incorporates concepts of ecosystem functioning and modeling assumptions common to biogeochemical models. A direct implication of these results is the ability of this approach to detect climate and phenology induced trends regardless of the initial conditions.

Description: We present a Simple Diagnostic Photosynthesis and Respiration Model (SDPRM) that has been developed based on pre-existing formulations. The photosynthesis model is based on the light use efficiency logic, suggested by Monteith1977, for calculating the Gross Primary Production (GPP) while the ecosystem respiration (Reco) model is based on the formulations introduced by Lloyd1994 and modified by Reichstein2003. SDPRM is driven by satellite-derived fAPAR (fraction of Absorbed Photosynthetically Active Radiation) and climate data from NCEP/NCAR. The model estimates 3-hourly values of GPP for seven major biomes and daily Reco. The motivation is to provide a-priori fields of surface CO2 fluxes with fine temporal and spatial scales, and their derivatives with respect to adjustable model parameters, for atmospheric CO2 inversions. The estimated fluxes from SDPRM showed that the model is capable of producing flux estimates consistent with the ones inferred from atmospheric CO2 inversion or simulated from process-based models. In this Technical Note, different analyses were carried out to test the sensitivity of the estimated fluxes of GPP and Reco to their driving forces. The spatial patterns of the climatic controls (temperature, precipitation, water) on the interannual variability of GPP are consistent with previous studies even though SDPRM has a very simple structure and few adjustable parameters, and hence it is much easier to modify than more sophisticated process-based models used in these previous studies. According to SDPRM, the results show that temperature is a limiting factor for the interannual variability of Reco over the cold boreal forest, while precipitation is the main limiting factor of Reco over the tropics and the southern hemisphere, consistent with previous regional studies.

Description: The flux of carbon dioxide from the soil to the atmosphere (soil respiration) is one of the major fluxes in the global carbon cycle. At present, the accumulated field observation data cover a wide range of geographical locations and climate conditions. However, there are still large uncertainties in the magnitude and spatiotemporal variation of global soil respiration. Using a global soil respiration dataset, we developed a climate-driven model of soil respiration by modifying and updating Raich's model, and the global spatiotemporal distribution of soil respiration was examined using this model. The model was applied at a spatial resolution of 0.5° and a monthly time step. Soil respiration was divided into the heterotrophic and autotrophic components of respiration using an empirical model. The estimated mean annual global soil respiration was 91 Pg C yr-1 (between 1965 and 2012; Monte Carlo 95% confidence interval: 87–95 Pg C yr-1) and increased at the rate of 0.09 Pg C yr-2. The contribution of soil respiration from boreal regions to the total increase in global soil respiration was on the same order of magnitude as that of tropical and temperate regions, despite a lower absolute magnitude of soil respiration in boreal regions. The estimated annual global heterotrophic respiration and global autotrophic respiration were 51 and 40 Pg C yr-1, respectively. The global soil respiration responded to the increase in air temperature at the rate of 3.3 Pg C yr-1 °C−1, and Q10 = 1.4. Our study scaled up observed soil respiration values from field measurements to estimate global soil respiration and provide a data-oriented estimate of global soil respiration. Our results, including the modeled spatiotemporal distribution of global soil respiration, are based on a semi-empirical model parameterized with over one thousand data points. We expect that these spatiotemporal estimates will provide a benchmark for future studies and also help to constrain process-oriented models.