Rationale: WG3 will review model comparison and multi-model assessment techniques as well as statistical and computational methods for model parameterisation, calibration, and evaluation (Milestone W7). It will develop a common database and simulation protocol for calibration and model comparisons (Milestone W8). The exact protocols will be developed in the course of the meetings, but it is planned to make use of Bayesian statistics both for calibration and model comparisons. All protocols will be made freely available under a creative commons license to maximize its utilization even beyond this CA. Based on this, WG3 will examine uncertainties within and between forest models, compare models in their predictive ability, and identify data that are particularly useful for further constraining model uncertainty (connection with WG1 and WG2) (Milestone W9). WG3 will also collaborate with WG4 on the facilitation of multi-model climate change impact projections to quantify predictive uncertainties, which will be based on the protocols developed in WG3.
TG 13: Review of model parameterization methods (Florian Hartig / Björn Reineking)
TG 14: MCMC algorithm comparison (Francesco Minunno)
TG 15: Statistical problems in fitting models to heterogeneous data (David Cameron / Michael Dietze)
TG 18: The merits of multi constraints data sets for model parameter estimation (Andreas Ibrom)
TG 19: Model-comparison Nitrogen (Federico Magnani)
Participation: If you are interested in participating in one of the task groups, please contact the task group coordinators directly. If you are interested in the working group in general, please contact the working group leaders.