Evaluating conditional density estimation networks as a probabilistic \ndownscaling tool: application to precipitation in Newfoundland
Bibliographic record
Abstract
The objective of this research is to better quantify the distribution of extreme precipitation \nwithin Newfoundland, for the current climate conditions. The province of \nNewfoundland is interested in this information as guidance for climate adaptation \nand longterm infrastructure planning purposes. Extreme analyses are commonly limited \nby short periods of observation (often only 30 years or less), resulting in large \nuncertainty regarding rare events. These limitations can be addressed by increasing \nthe period of observation, or partially addressed using synthetic time series generated \nby stochastic weather generators. These weather generators attempt to replicate \nrelevant statistics of the observed climate, using various statistical modelling tools. \nHere, a probabilistic neural network-based downscaling method is used to predict the \nprobability distribution of precipitation at a study site conditional upon the synoptic \nstate of the atmosphere; that is, features of a given day’s large-scale atmospheric \nstate are used to estimate the likelihood of all possible precipitation amounts for that \nday. My results show that CDEN-based probabilistic downscaling generally agrees \nbetter with observations than uncorrected reanalysis-based precipitation estimates. \nHowever, it gives significantly higher estimates for extreme events at low frequency \n(e.g., 100 year return periods). This approach is demonstrated for St. John’s airport. \nComparison with nearby stations, Windsor Lake, suggests that these higher estimates \nmay be reasonable; however, further work and a longer observational record is needed to determine whether these estimates represent added value to the raw observational \nrecord or a shortcoming of our CDEN-based methodology.
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 | 0.005 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".