Error propagation in forest growth models in the context of regional forecasts
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
In forestry, tree-level growth models provide predictions of forest dynamics andthereby, they support decision-making. Although they are widely used, the uncertainty of their predictions is rarely assessed. Understanding the sources of uncertainty and estimating their impact is an essential step forward in a period where large-scale forecasts are becoming more popular. This thesis addresses the issue of uncertainty estimation in regional growth forecasts. The effects of large-scale disturbances were also studied.The growth model ARTEMIS-2009, which applies to most forest types in Quebec,Canada, was taken as a case study. A bootstrap hybrid estimator was used toestimate the model- and the sampling-related variances. The total variance wasthen decomposed to determine which model component induced the greatest shareof variance in the forecasts. Then, the survival analysis approach was used to develop a harvest model based on plot and regional variables. This model was integrated into ARTEMIS so that harvesting combined with spruce budworm (SBW) outbreaks were accounted in the simulations. Then, their contributions in terms of uncertainty were estimated. The results revealed that the sampling accounted for most of the variance in short-term forecasts. In long-term forecasts, the model contribution turned out to be as important as that of the sampling. The variance decomposition per model component indicated that the mortality sub-model induced the highest variability in the forecasts. A great deal of uncertainty was induced by the natural disturbances when they were accounted for in the projections. In particular, SBW showed to be the most important source of uncertainty compared to harvest activities and sampling. In the light of these results, our recommendations are that the effort to reduce uncertainty should focus on the sampling in short-term forecasts, and on the mortality sub-model and SBW occurrence in mid- and long-term forecasts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it