Towards an ensemble-based evaluation of land surface models in light of uncertain forcings and observations
Bibliographic record
Abstract
Abstract. Quantification of uncertainty in fluxes of energy, water, and CO2 simulated by land surface models (LSMs) remains a challenge. LSMs are typically driven with, and tuned for, a specified meteorological forcing data set and a specified set of geophysical fields. Here, using two data sets each for meteorological forcing and land cover representation (in which the increase in crop area over the historical period is implemented in the same way), as well as two model structures (with and without coupling of carbon and nitrogen cycles), the uncertainty in simulated results over the historical period is quantified for the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC) model. The resulting eight (2×2×2) model simulations are evaluated using an in-house model evaluation framework that uses multiple observation-based data sets for a range of quantities. The simulated area burned, fire CO2 emissions, soil carbon mass, vegetation carbon mass, runoff, heterotrophic respiration, gross primary productivity, and sensible heat flux show the largest spread across the eight simulations relative to their global ensemble mean values. Simulated net atmosphere–land CO2 flux, a critical determinant of the performance of LSMs, is found to be largely independent of the simulated pre-industrial vegetation and soil carbon mass, although our framework represents the historical increase in crop area in the same way in both land cover representations. This indicates that models can provide reliable estimates of the strength of the land carbon sink despite some biases in carbon stocks. Results show that evaluating an ensemble of model results against multiple observations disentangles model deficiencies from uncertainties in model inputs, observation-based data, and model configuration.
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.
How this classification was reachedexpand
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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".