Skip to main content

Modelling Tree Response to Aridity Increase with Traits

Periodic Reporting for period 1 - M-TRAIT (Modelling Tree Response to Aridity Increase with Traits)

Reporting period: 2018-02-01 to 2020-01-31

Even if we succeed in decarbonizing the world’s economy, the targeted global temperature increase of 1.5 to 2 K by 2100 will still produce significant changes in the water cycle . One predicted change is an increase in the frequency and severity of droughts in Europe . For forests, the consequences of these droughts are expected to range from productivity decline to abrupt mortality . By 2100 these effects are projected to translate into economic losses estimated to range between 14 and 50% in the forestry sector . Forest stakeholders and policymakers are well aware of the need to adapt to future climate; yet, despite decades of research our predictions of the effects of droughts on forest ecosystems are still highly uncertain – a lack of knowledge that calls for novel approaches and knowledge transfer.
Land surface models (LSMs) simulate the biophysical and biogeochemical processes of the interaction between the terrestrial biosphere and the atmosphere. They are a key component of the so-called ‘Earth systems models’ used for recommendations by the intergovernmental panel for climate change, and where they are expected to tackle large-scale questions related to land-atmosphere interactions. The predictive power of LSMs is hampered by their over-parameterization , that implies that a good result can be obtained for the wrong reasons. Indeed, most physiological processes are represented by semi-empirical equations, with parameters calibrated for specific sites and conditions that do not always have ecological meaning and that might not necessarily hold under future climate. For LSMs to be able to simulate forest response to droughts outside their calibration range requires the implementation of the main physiological characteristics the so-called ‘traits’ that vary with environmental conditions , and drive trees’ response to environmental changes.
This requirement leads to the overall challenge of this fellowship, to advance the representation of drought- related physiological processes in LSMs using the cutting-edge multidisciplinary approach of trait modelling. The aim is to enhance society’s understanding of future impacts of drought on European forests.
Regarding the quantification of the main determinants of short-term tree response to drought via traits and their variability, the ER engaged in a large team effort gathering 15 re-searchers to build a database of hydraulic plasticity of plant traits. To build the database, 278 publications were examined by the group, of which 221 contained exploitable data that allowed the extraction of 4686 double data points quantifying the plasticity of one of 254 species. The data analysis is suffering delay due to the Covid crisis so the outcome usable for modelling will only available in a few months.
Regarding the introduction of a traits-based approach into a Land Surface Model, a model configuration was setup to constrain the model parameters with 4 measured physiological traits identified as the most likely to be implemented in models for the availability of the data and their physiological role. Simulations were run over 18 sites in Catalunya for which traits measurements were available. The simulations had to stop due to a change in the model version after flaws were identified in the previous version. This work is to be continued with the most recent and up-to-date version of the model.
Due to the delay in objective 2, the objective to simulate the physiological response of forests to more frequent droughts in the future was reformulated as to evaluate the forest management that would be most efficient in mitigating climate change. Simulations were run over all of Europe with a range of forest management scenarios and their climate impact evaluated.
Finally, due to the changes operated in objectives 2 and 3,objective 4 was also updated as an exploration of the determinants of the substitution potential of the wood sector. A group of researcher from different disciplines was gathered by the ER and the knowledge gaps were evidenced and listed; This work resulted in the writing of funding proposals and a publication that is still to be finalized.
The identification of physiological tree traits crucial to the forests' drought response which results are still pending will be of use for the whole ecophysiology and vegetation modelling community. Their application in a Land Surface Model as non-static parameters will be an important evaluation of the role traits must play in land surface modelling in thr future.