Florian Hartig

← Publications

Available and missing data to model impact of climate change on European forests

Paloma Ruiz-Benito, Giorgio Vacchiano, Emily R. Lines, Christopher P. O. Reyer, Sophia Ratcliffe, Xavier Morin, Florian Hartig, Annikki Mäkelä, Rasoul Yousefpour, Jimena E. Chaves, Alicia Palacios-Orueta, Marta Benito-Garzón, Cesar Morales-Molino, J. Julio Camarero, Alistair S. Jump, Jens Kattge, Aleksi Lehtonen, Andreas Ibrom, Harry J. F. Owen & Miguel A. Zavala

Ecological Modelling, 416, 108870 (2020)
Cite this
@article{ruizbenito2020available,
  author = {Paloma Ruiz-Benito and Giorgio Vacchiano and Emily R. Lines and Christopher P. O. Reyer and Sophia Ratcliffe and Xavier Morin and Florian Hartig and Annikki Mäkelä and Rasoul Yousefpour and Jimena E. Chaves and Alicia Palacios-Orueta and Marta Benito-Garzón and Cesar Morales-Molino and J. Julio Camarero and Alistair S. Jump and Jens Kattge and Aleksi Lehtonen and Andreas Ibrom and Harry J. F. Owen and Miguel A. Zavala},
  title = {Available and missing data to model impact of climate change on European forests},
  journal = {Ecological Modelling},
  volume = {416},
  pages = {108870},
  year = {2020},
  doi = {10.1016/j.ecolmodel.2019.108870},
}

DOI: 10.1016/j.ecolmodel.2019.108870
Citation statistics: 110 (Google Scholar) · 89 (OpenAlex), as of 29 October 2026

Google Scholar citations per year, as of 22 October 2026 (yearly counts are fetched separately from the total)

Funding

This study was supported through funding by European Cooperation in Science and Technology (COST) (FP1304 PROFOUND); European Union Seventh Framework Programme (FP7/2007-2013), Plant Fellows (PCOFUND-GA-2010-267243); Ministerio de Ciencia, Innovación y Universidades, Spain (CGL2015-69186-C2-2-R FUNDIVER; RTI2018-096884-B-C32 DARE); Comunidad de Madrid (S2013/MAE-2719 Remedinal-3; 2016-T2/AMB-1665); German Ministry of Science and Education (BMBF) (01LS1711A; 01LS1201A2); Investments for the Future program IdEx Bordeaux (ANR-10-IDEX-03-02); Bavarian Climate Research Network, research network BLIZ.

View article (DOI)

Abstract

Climate change is expected to cause major changes in forest ecosystems during the 21st century and beyond. To assess forest impacts from climate change, the existing empirical information must be structured, harmonised and assimilated into a form suitable to develop and test state-of-the-art forest and ecosystem models. The combination of empirical data collected at large spatial and long temporal scales with suitable modelling approaches is key to understand forest dynamics under climate change. To facilitate data and model integration, we identified major climate change impacts observed on European forest functioning and summarised the data available for monitoring and predicting such impacts. Our analysis of c. 120 forest-related databases (including information from remote sensing, vegetation inventories, dendroecology, palaeoecology, eddy-flux sites, common garden experiments and genetic techniques) and 50 databases of environmental drivers highlights a substantial degree of data availability and accessibility. However, some critical variables relevant to predicting European forest responses to climate change are only available at relatively short time frames (up to 10-20 years), including intra-specific trait variability, defoliation patterns, tree mortality and recruitment. Moreover, we identified data gaps or lack of data integration particularly in variables related to local adaptation and phenotypic plasticity, dispersal capabilities and physiological responses. Overall, we conclude that forest data availability across Europe is improving, but further efforts are needed to integrate, harmonise and interpret this data (i.e. making data useable for non-experts). Continuation of existing monitoring and networks schemes together with the establishments of new networks to address data gaps is crucial to rigorously predict climate change impacts on European forests.

What the paper shows and why it matters

(AI-generated)

Predicting how forests respond to climate change depends on combining empirical data gathered over large areas and long periods with models, and one challenge is to parameterise and test those models at large spatial scales (Hartig et al. 2012). Many kinds of data exist, from genetics, flux towers (which record local productivity continuously) and inventories to tree rings, palaeoecological records, remote sensing and environmental drivers such as climate and soils. They differ, however, in resolution, accessibility and harmonisation. The paper therefore asks which of these data are available and accessible at the European scale, and where the gaps lie that models of climate-change impacts need filled.

The paper argues that data on European forests and their environment are largely available and accessible, but that several variables critical for predicting climate responses are scarce, short or not integrated. It rests on a review of about 120 forest-related databases and 50 databases of environmental drivers, built on a literature review of the types of forest response and on the authors’ expert knowledge of each data type. Long records exist for tree growth and forest cover, but they are often localised. For intraspecific trait variability (differences within a species), defoliation, tree mortality and recruitment, data are available only over relatively short time frames (up to 10–20 years). The paper also finds gaps or poor integration for local adaptation and phenotypic plasticity (how far differences between populations are genetic or flexible), dispersal and physiological responses. For drivers, it names soil information and long-term data on disturbances and management as the main limitations. The paper adds that inventory data under-represent large trees and extreme responses, which can bias predictions.

Figure 2. Relationships between (a) the data that can be used to detect and inform on (b) the biological levels at which forests may respond as a result of climate change. From Ruiz-Benito et al. (2020), Ecological Modelling, licensed under CC BY 4.0. Click the image to enlarge.

The paper concludes that the main gap lies in the scarcity of high-quality, freely available data at high spatial and temporal resolution on the biological processes that climate change affects, such as dispersal, physiology and adaptation, while availability overall is improving. It sets four priorities: maintaining existing monitoring and adding targeted new monitoring for the gaps; harmonised, freely available databases; closer exchange between those who collect data and those who model; and methods that integrate data from different sources.

References