The resolution and potential value of Australian seasonal rainfall forecasts based on the five phases of the Southern Oscillation Index
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
We assess the resolution of the Southern Oscillation Index (SOI) seasonal rainfall forecasting system and calculate the loss in potential value of the forecasting system using a cost/loss model. Forecasts of the probability of a ‘dry’ autumn, winter, spring, and summer were obtained for 226 towns across Australia, based on the 5 phases of the SOI. For every town the variance ratio, the observed forecast variance as a proportion of the variance of a perfect forecasting system, was calculated for each season. Value score curves, showing the expected value of the forecasts as a proportion of the expected value of perfect information, were calculated for every town for each season. Maps of variance ratio and maps of mean value scores across Australia were produced by ordinary kriging. In all seasons and regions the SOI forecasting system had a variance ratio of less than 0.20, indicating that resolution and skill were never high. Variance ratios greater than 0.10 only occurred in parts of south-eastern Australia and the Cape York region during spring and in the Townsville region during summer. The variance ratio was less than 0.05 for the majority of Australia during autumn, winter, and summer. The mean value scores for actions that are only triggered by a large shift in the forecast from climatology were uniformly close to zero in all seasons and regions, indicating that little or no value can be derived in such cases. Actions triggered by a moderate shift of the forecast were also generally associated with low value scores. Mean value scores above 0.20 were limited to actions with a decision threshold close to climatology and only occurred in parts of south-eastern Australia and the Cape York region during spring and in the Townsville region during summer. We conclude that the imperfect resolution of the SOI forecasting system has a substantial effect on potential value. The forecasting system can potentially deliver value to users with actions that are triggered by a small shift in the forecast from climatology, especially in eastern Australia during spring, but not to users with actions that are only triggered by a large shift of the forecasts from climatology.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.002 |
| 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 it