Accuracy, realism and general applicability of European forest models
Abstract
Forest models are instrumental for understanding and projecting the impact of climate change on forests. A considerable number of forest models have been developed in the last decades. However, few systematic and comprehensive model comparisons have been performed in Europe that combine an evaluation of modelled carbon and water fluxes and forest structure. We evaluate 13 widely used, state‐of‐the‐art, stand‐scale forest models against field measurements of forest structure and eddy‐covariance data of carbon and water fluxes over multiple decades across an environmental gradient at nine typical European forest stands. We test the models' performance in three dimensions: accuracy of local predictions (agreement of modelled and observed annual data), realism of environmental responses (agreement of modelled and observed responses of daily gross primary productivity to temperature, radiation and vapour pressure deficit) and general applicability (proportion of European tree species covered). We find that multiple models are available that excel according to our three dimensions of model performance. For the accuracy of local predictions, variables related to forest structure have lower random and systematic errors than annual carbon and water flux variables. Moreover, the multi‐model ensemble mean provided overall more realistic daily productivity responses to environmental drivers across all sites than any single individual model. The general applicability of the models is high, as almost all models are currently able to cover Europe's common tree species. We show that forest models complement each other in their response to environmental drivers and that there are several cases in which individual models outperform the model ensemble. Our framework provides a first step to capturing essential differences between forest models that go beyond the most commonly used accuracy of predictions. Overall, this study provides a point of reference for future model work aimed at predicting climate impacts and supporting climate mitigation and adaptation measures in forests.
What the paper shows and why it matters
(AI-generated)
Forest models are used to project how climate change will affect forests, but they differ in structure and so in their predictions. Judging them needs three things: fit to observations (accuracy), correct responses to environmental drivers (realism) and coverage of many species (generality). The paper builds on a classic distinction between these dimensions, which are hard to maximise together, as further specified by Kramer et al. (2002). It notes that comparisons of many complex models on common data are missing in Europe. The problem it takes on is how structurally different models perform on one benchmark, and whether their structure produces trade-offs among the three.
The study finds that no single European forest model is best on accuracy, realism and generality together, although several excel on each, and that the ensemble mean scores highest overall, although its accuracy is only intermediate and its species coverage is high by construction. It found no systematic trade-off among the three dimensions, a result it says may partly reflect how realism was measured. Thirteen models were run without site-specific calibration and compared with multi-decade measurements of forest structure and of carbon and water fluxes (from eddy-covariance towers) at nine boreal and temperate stands. Structure variables such as basal area were generally predicted with smaller errors than annual fluxes, and simpler models were not necessarily less accurate than complex ones. For daily productivity, the ensemble mean reproduced the observed responses to environmental drivers more realistically than any single model, which the study attributes to over- and underestimates cancelling out; single models nevertheless beat it in several cases. Almost all models covered Europe’s dominant tree species.
The study reads its results as showing there is no “silver bullet”: the best model depends on site and purpose. Rarer tree species need several models, and several output variables should be evaluated. It presents its framework as a first attempt to put these three dimensions into operation for complex climate-impact models and expects the study to serve as a benchmark for new models. Its stated limits include that realism was tested only through daily productivity and that eight of nine stands were fairly uniform single-species stands. Junttila et al. (2023), sharing three co-authors with the paper, cite the ensemble result in support of using model ensembles for policy planning.
References
- Kramer et al. (2002). Evaluation of six process‐based forest growth models using eddy‐covariance measurements of CO2 and H2O fluxes at six forest sites in Europe. Global Change Biology 8, 213–230. doi:10.1046/j.1365-2486.2002.00471.x
- Junttila et al. (2023). Quantification of forest carbon flux and stock uncertainties under climate change and their use in regionally explicit decision making: Case study in Finland. Ambio 52, 1716–1733. doi:10.1007/s13280-023-01906-4