Parameter-state ensemble thinning for short-term hydrological prediction
Bibliographic record
Abstract
The main sources of uncertainty in hydrological modelling can be summarized as structural errors, parameter errors, and data errors. Operational modellers are generally more concerned with predictive ability than model errors, and this paper presents a new, simple method to improve predictive ability. The method is called parameter-state ensemble thinning (P-SET). P-SET takes a large ensemble of continuous model runs and applies screening criteria to reduce the size of the ensemble. The goal is to find the most promising parameter-state combinations for analysis during the prediction period. Each prediction period begins with the same large ensemble, but the screening criteria are free to select a different sub-set of simulations for each separate prediction period. The case study is from June to October 2014 for a small (1324 km 2 ) watershed just north of Lake Superior in Ontario, Canada, using a Canadian semi-distributed hydrologic land-surface scheme. The study examines how well the approach works given various levels of certainty in the data, beginning with certainty in the streamflow and precipitation, followed by uncertainty in the streamflow and certainty in the precipitation, and finally uncertainty in both the streamflow and precipitation. The approach is found to work in this case when streamflow and precipitation are fairly certain, while being more challenging to implement in a forecasting scenario where future streamflow and precipitation are much less certain. The main challenge is determined to be related to parametric uncertainty and ideas for overcoming this challenge are discussed. The approach also highlights model structural errors, which are also discussed.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".