The PROFOUND Database for evaluating vegetation models and simulating climate impacts on European forests
View article (DOI) Preprint Software: ProfoundData
Abstract
Process-based vegetation models are widely used to predict local and global ecosystem dynamics and climate change impacts. Due to their complexity, they require careful parameterization and evaluation to ensure that projections are accurate and reliable. The PROFOUND Database (PROFOUND DB) provides a wide range of empirical data on European forests to calibrate and evaluate vegetation models that simulate climate impacts at the forest stand scale. A particular advantage of this database is its wide coverage of multiple data sources at different hierarchical and temporal scales, together with environmental driving data as well as the latest climate scenarios. Specifically, the PROFOUND DB provides general site descriptions, soil, climate, CO₂, nitrogen deposition, tree and forest stand level, and remote sensing data for nine contrasting forest stands distributed across Europe. Moreover, for a subset of five sites, time series of carbon fluxes, atmospheric heat conduction and soil water are also available. The climate and nitrogen deposition data contain several datasets for the historic period and a wide range of future climate change scenarios following the Representative Concentration Pathways (RCP2.6, RCP4.5, RCP6.0, RCP8.5). We also provide pre-industrial climate simulations that allow for model runs aimed at disentangling the contribution of climate change to observed forest productivity changes. The PROFOUND DB is available freely as a “SQLite” relational database or “ASCII” flat file version (at https://doi.org/10.5880/PIK.2020.006; Reyer et al., 2020). The data policies of the individual contributing datasets are provided in the metadata of each data file. The PROFOUND DB can also be accessed via the ProfoundData R package (https://CRAN.R-project.org/package=ProfoundData; Silveyra Gonzalez et al., 2020), which provides basic functions to explore, plot and extract the data for model set-up, calibration and evaluation.
What the paper shows and why it matters
(AI-generated)
Process-based vegetation models simulate how forests respond to climate, and their projections inform policy and management. This calls for quality control: because the models are complex, they must be calibrated against observations and compared with each other, and the formal statistical methods used for this need systematic reference data. Such data are scattered over many sources and formats, so model comparisons have often been limited by what data were available and comparable. Earlier benchmarking systems (Kelley et al. 2013; Collier et al. 2018) brought important progress, but many initiatives of this kind work at the global scale with global products, which lack long-term, detailed measurements of forest stand structure. The problem the paper takes on is how to assess models rigorously at the scale of single forest stands without harmonised data.
The study presents the PROFOUND Database, a freely available, harmonised collection of evaluation and driving data for nine contrasting European forest stands, built so that both stand-scale models and global vegetation models can be run and compared with it. For each stand, the database holds site and soil descriptions, climate, atmospheric CO₂ and nitrogen deposition, tree and stand measurements and satellite-derived (MODIS) products; for five stands it also holds time series of carbon fluxes, heat conduction and soil water. Climate comes both bias-corrected with local data and as 0.5° grid cells, the form global models use. Future climate follows four Representative Concentration Pathways (RCPs, scenarios of greenhouse-gas concentrations), and pre-industrial climate simulations allow model runs aimed at separating the contribution of climate change to observed changes in forest productivity. An R package, ProfoundData, provides functions to explore and extract the data.
The paper concludes that the database facilitates model evaluation, calibration, uncertainty analysis and model intercomparison, and that it highlights the value of long-term environmental monitoring data for robust inferences about causal processes and future dynamics of forests. It is designed so that uncertainties can be told apart: structural ones through comparing models on common data, input ones through the range of climate datasets and scenarios, and parametric ones through inverse calibration, that is, fitting model parameters to observations. Lasch-Born et al. (2020) used data from the database to evaluate the forest model 4C at four European sites.
References
- Kelley et al. (2013). A comprehensive benchmarking system for evaluating global vegetation models. Biogeosciences 10, 3313–3340. doi:10.5194/bg-10-3313-2013
- Collier et al. (2018). The International Land Model Benchmarking (ILAMB) System: Design, Theory, and Implementation. Journal of Advances in Modeling Earth Systems 10, 2731–2754. doi:10.1029/2018ms001354
- Lasch-Born et al. (2020). Description and evaluation of the process-based forest model 4C v2.2 at four European forest sites. Geoscientific Model Development 13, 5311–5343. doi:10.5194/gmd-13-5311-2020