Critical evaluation of diameter increment modelling in the Great Lakes region

Simulations of forest stand dynamics in a modelling framework including Forest Vegetation Simulator (FVS) are diameter driven, thus the diameter or basal area increment model needs a special attention. This dissertation critically evaluates diameter or basal area increment models and modelling approaches in the context of the Great Lakes region of the United States and Canada. A set of related studies are presented that critically evaluate the sub-model for change in individual tree basal diameter used in the Forest Vegetation Simulator (FVS), a dominant forestry model in the Great Lakes region. Various historical implementations of the STEMS (Stand and Tree Evaluation and Modeling System) family of diameter increment models, including the current public release of the Lake States variant of FVS (LS-FVS), were tested for the 30 most common tree species using data from the Michigan Forest Inventory and Analysis (FIA) program. The results showed that current public release of the LS-FVS diameter increment model over-predicts 10-year diameter increment by 17% on average. Also the study affirms that a simple adjustment factor as a function of a single predictor, dbh (diameter at breast height) used in the past versions, provides an inadequate correction of model prediction bias. In order to re-engineer the basal diameter increment model, the historical, conceptual and philosophical differences among the individual tree increment model families and their modelling approaches were analyzed and discussed. Two underlying conceptual approaches toward diameter or basal area increment modelling have been often used: the potential-modifier (POTMOD) and composite (COMP) approaches, which are exemplified by the STEMS/TWIGS and Prognosis models, respectively. It is argued that both approaches essentially use a similar base function and neither is conceptually different from a biological perspective, even though they look different in their model forms. No matter what modelling approach is used, the base function is the foundation of an increment model. Two base functions – gamma and Box-Lucas – were identified as candidate base functions for forestry applications. The results of a comparative analysis of empirical fits showed that quality of fit is essentially similar, and both are sufficiently detailed and flexible for forestry applications. The choice of either base function in order to model diameter or basal area increment is dependent upon personal preference; however, the gamma base function may be preferred over the Box-Lucas, as it fits the periodic increment data in both a linear and nonlinear composite model form. Finally, the utility of site index as a predictor variable has been criticized, as it has been widely used in models for complex, mixed species forest stands though not well suited for this purpose. An alternative to site index in an increment model was explored, using site index and a combination of climate variables and Forest Ecosystem Classification (FEC) ecosites and data from the Province of Ontario, Canada. The results showed that a combination of climate and FEC ecosites variables can replace site index in the diameter increment model.

Towards greater accuracy in individual-tree mortality regression

Background mortality is an essential component of any forest growth and yield model. Forecasts of mortality contribute largely to the variability and accuracy of model predictions at the tree, stand and forest level. In the present study, I implement and evaluate state-of-the-art techniques to increase the accuracy of individual tree mortality models, similar to those used in many of the current variants of the Forest Vegetation Simulator, using data from North Idaho and Montana. The first technique addresses methods to correct for bias induced by measurement error typically present in competition variables. The second implements survival regression and evaluates its performance against the traditional logistic regression approach. I selected the regression calibration (RC) algorithm as a good candidate for addressing the measurement error problem. Two logistic regression models for each species were fitted, one ignoring the measurement error, which is the “naïve” approach, and the other applying RC. The models fitted with RC outperformed the naïve models in terms of discrimination when the competition variable was found to be statistically significant. The effect of RC was more obvious where measurement error variance was large and for more shade-intolerant species. The process of model fitting and variable selection revealed that past emphasis on DBH as a predictor variable for mortality, while producing models with strong metrics of fit, may make models less generalizable. The evaluation of the error variance estimator developed by Stage and Wykoff (1998), and core to the implementation of RC, in different spatial patterns and diameter distributions, revealed that the Stage and Wykoff estimate notably overestimated the true variance in all simulated stands, but those that are clustered. Results show a systematic bias even when all the assumptions made by the authors are guaranteed. I argue that this is the result of the Poisson-based estimate ignoring the overlapping area of potential plots around a tree. Effects, especially in the application phase, of the variance estimate justify suggested future efforts of improving the accuracy of the variance estimate. The second technique implemented and evaluated is a survival regression model that accounts for the time dependent nature of variables, such as diameter and competition variables, and the interval-censored nature of data collected from remeasured plots. The performance of the model is compared with the traditional logistic regression model as a tool to predict individual tree mortality. Validation of both approaches shows that the survival regression approach discriminates better between dead and alive trees for all species. In conclusion, I showed that the proposed techniques do increase the accuracy of individual tree mortality models, and are a promising first step towards the next generation of background mortality models. I have also identified the next steps to undertake in order to advance mortality models further.