Gaining local accuracy while not losing generality — extending the range of gap model applications

Abstract

For the study of long-term processes in forests, gap models generally sacrifice accuracy (i.e., simulating system behavior in a quantitatively accurate manner) for generality (i.e., representing a broad range of systems’ behaviors with the same model). We selected the gap model ForClim to evaluate whether the local accuracy of forest succession models can be increased based on a parsimonious modeling approach that avoids the additional complexity of a 3D crown model, thus keeping parameter requirements low. We improved the representation of tree crowns by introducing feedbacks between (i) light availability and leaf area per tree and (ii) leaf area per tree and diameter growth rate to account for the self-pruning in real stands. The local accuracy of the new model, ForClim v2.9.5, was considerably improved in simulations at three long-term forest research sites in the Swiss Alps, while its generality was maintained as shown in simulations of potential natural vegetation along a broad environmental gradient in Central Europe. We conclude that the predictive ability of a model does not depend on its complexity, but on the reproduction of patterns. Most importantly, model complexity should be consistent with the objectives of the study and the level of system understanding.

Ungulate herbivory modifies the effects of climate change on mountain forests

“…Trees are established as saplings with a diameter at breast height (dbh) of 1.27 cm, whereby the number of new trees is a function of species-specific responses to winter temperature, light availability, growing degree-days, and browsing pressure (Bugmann 1994). ForClim performed well in several studies comparing simulated forests with local, site specific forest data (Didion et al 2009b, c;Risch et al 2005), and also with potential natural vegetation along large temperature and precipitation gradients within and between various mountain regions (Didion et al 2009c;Bugmann and Solomon 2000;Shao et al 2001). As input the model requires monthly means and standard deviations of temperature and precipitation as well as their cross-correlation (Bugmann 1994).…”

“…1, Fig. 1), which represents a moderate intensity that approximates an average browsing intensity in Swiss forests (Didion et al 2009c). This spin-up simulation was used for all of the following scenarios in order to have identical starting conditions for the browsing and climate change scenarios.…”

“…However, the role of ungulate herbivores as a driver of forest succession in response to a changing climate has, to our knowledge, not been addressed in modeling studies. We aimed to fill this gap by using the forest gap model ForClim (Didion et al 2009c) to simulate forest development under climate change and different ungulate browsing pressure.…”

Models for supporting forest management in a changing environment

“…The issue of hybrid modeling has also been debated in the specific case of forest gap models (Reynolds et al, 2001), developed for investigating spatial dynamics and succession in forest ecosystems, and thus of interest for assessing at least the relative importance of species in forest resources. Recent developments have shown that they can also be used to simulate managed stands (Didion et al, 2009). The incorporation of process components has traditionally been limited in these models, where the influence of environmental factors mostly relies on empirical relationships (Bugmann 2001).…”

“…A single «super PBM» to be used in all countries for all species and for all situations could not be identified and is unlikely to ever exist, because modeling is a deliberate simplification of reality, and the simplification will always include and induce siteor at least region-specific aspects. A model cannot at the same time be completely general in its scope and applicability while providing locally highly accurate results (Levins, 1966), as it observed in a concrete case study (Didion et al, 2009). Furthermore, there are other challenges to be met, such as: -Data availability.…”