By: Stef Haesen and Koenraad Van Meerbeek
Ecological research continues to rely on coarse-gridded macroclimatic data, which depend on standardized meteorological data collected by weather stations. These data collection protocols aim to minimize the influence of microclimatic factors. Therefore, conventional climate data fail to capture the substantial temperature differences between the interior of forests and their surroundings. Indeed, on scorching summer days, the forest environment can feel significantly cooler than the surrounding areas. Yet, quantifying the extent of this cooling effect in specific forest patches has proven to be a challenging task. In response, the sGlobe research group from the Forest, Nature, and Landscape division at KU Leuven has developed a model capable of predicting the air-conditioning function of European forests at high spatial resolutions.
Utilizing a unique dataset comprising over 1200 mini-weather stations scattered throughout Europe's forests, researchers assessed the temperature difference between the forest understory and the existing weather station network. This network of sensors situated on the forest floor is exceptional in its ability to accurately record the true temperature beneath the forest canopy. Their findings reveal that during the summer, maximum temperatures within European forests can be up to 8 degrees lower than those outside the forested areas. Similarly, in winter, minimum temperatures within forests can be up to 6 degrees higher than outside.
“The VSC Computation Centre played a crucial role in this research by providing the computational resources required to generate predictions at exceptionally high spatial resolutions.”
Trees, with their foliage and branches forming a canopy, create a thermal insulating layer above the forest floor. Furthermore, the transpiration process during photosynthesis removes heat from the air. Consequently, summer maximum temperatures within forests are significantly lower compared to their surroundings. This natural cooling effect provided by forest canopies plays a crucial role in mitigating the impact of summer heat waves, making it increasingly important in the context of a warming climate.
Figure 1: Novel high-resolution predictions of temperature patterns within European forests unveil the substantial buffering effect of the forest canopy. Here, we combined topographical, vegetational and macroclimatic predictors in a machine-learning model to quantify the buffering capacity of European forests.
This study marks the first instance where researchers have produced maps quantifying the temperature-buffering capacity of forests at such high-resolution and over such an extensive geographical range. The Flemish Supercomputer Centre (VSC) played a crucial role in this research by providing the computational resources required to generate predictions at exceptionally high spatial resolutions. Employing the developed geospatial modelling pipeline, predictions were made across approximately 3 billion pixels, replicating the process 360 times. One can imagine the significant computational challenges this presented. The VSC's High-Performance Computing (HPC) infrastructure, particularly its efficient multi-core CPU parallel computing capabilities, enabled us to accomplish our mission.
Read the full publication of this article in Global Change Biology.