Evaluating forest stress factors using various forest growth modeling approaches

This paper examines in detail the relative merits of several modeling approaches to evaluating forest site, physical, chemical and climatic stresses. The modeling approaches considered are (i) empirical growth and yield (management) models (STEMS and FOREST); (ii) successional models (JABOWA/FORET gap-phase simulation models and FORSKA); (iii) forest process models (FOREST-BGC). The models were analyzed in terms of their assumptions, formulation, and predictive ability. The analysis incorporated intensive measurements of tree growth and weather conditions taken over 7 years in two stands of northern hardwoods located in Upper Michigan (USA). The process model was very detailed and required many input variables. All models performed poorly when pridictions were compared with the observed data making accurate site-specific predictions impossible without local calibrations.