Process-based/mechanistic forest models

Process-based or mechanistic models were developed to model key growth process(es) and fundamental causes of productivity such as: photosynthesis and respiration, carbon allocation, nutrient cycles and climate effects. They are mathematical representations of biological systems that incorporate our understanding of physiological and ecological mechanisms into predictive algorithms. They take into account at the physiological level plant responses to site factors either if they are manipulated by humans directly, such as fertility, or indirectly, such as atmospheric carbon dioxide concentrations, For example, the effect of change in atmospheric carbon dioxide concentrations is manifested through the process of photosynthesis, at the scale of individual leaves. This primary effect changes growth rates, alters investment in growth above and below ground, affects nutrient acquisition and concentrations in tissues, water relations, competitive interactions, rates of decomposition and microbial populations, insect feeding habits, and energy, nutrient, and water flow through the forest ecosystem. Hence, the ability to accurately estimate how changes in carbon dioxide concentrations affect those fundamental processes will determine the ability of the model to predict how the increase of carbon dioxide concentrations will affect the forest. Overall these models have several limitations:

• This type of model requires high quantity of detailed data which is rarely available at national, regional or even lower scales.  This is due to the fact that they were originally designed and used for research purposes only, although these models have been more recently developed for use in practical forest management. However, the models designed to provide predictions for management require simpler and more readily available data than those designed for research.
• The level of processes are lower than those that need to be estimated at stand level. Scaling these model subcomponents over space and time results in error propagation, typically an unknown quantity. Moreover physiological or molecular processes are not necessarily always relevant to explain problems at hand. For example these processes do not tell us about the connection between stem taper and stand density, difference in the root/shoot ratio between shaded and exposed trees, effect of density on height of trees with identical diameter, self-thinning and succession, or about many other processes of practical importance. The tree or the stand are more than the sum of its finite parts and this approach alone is not capable of describing growth of a stand per si.
• These models operate in a level which is considered too refine for the current state of knowledge. Since the actual number of these processes is too large to grasp and some of them are still unknown, the models are inherently incomplete.
• Validation is difficult to do since key physiological observations are difficult to measure.
• These models do not make use of the readily available inventory data and the long-term observations on permanent plots. Instead these models require sophisticated and rarely available information. There are not regularly measured radiation absorption, transpiration, rate of senescence, and annual retranslocation of nutrients on traditional forest sample plots.