On the scale-dependence of the predictability of precipitation patterns by numerical weather prediction models
Bibliographic record
Abstract
Despite the increased realism of convective-allowing Numerical Weather Prediction (NWP) forecasts, their accuracy has not improved as expected, especially for quantitative precipitation forecasting (QPF). Many factors can account for the poor QPF skill, from intrinsic predictability limitations to shortcomings of the evaluation method. Given the complexity of precipitation processes, more effort is needed to understand their predictability.In this thesis, data from NOAA’s Hazardous Weather Testbed (HWT) forecasting Spring Experiments of 2008-2013 are analyzed to study precipitation predictability at the mesoscale. This unique data set, consisting of high-resolution (4-km grid spacing) ensemble precipitation forecasts covering most of the Continental US, with different types of perturbation methodologies and mesoscale data assimilation, allows the investigation of the QPF sensitivity to different forecast uncertainties.Several questions are addressed. First, a methodology is developed to quantify the loss of precipitation predictability with spatial scale and forecast time. Then, this methodology is applied to the 2008 data to characterize the scale dependence of the predictability of precipitation by an ensemble that has both large-scale initial and lateral boundary condition (IC/LBC) perturbations and varied model physics. The predictability by the ensemble is found to be very short-lived (2 hours for meso-γ scales, and 10 hours for meso-β scales). When compared to radar derived precipitation maps, the ensemble forecasts fully lose skill at meso-β scales after the first 3 hours.The case-dependence of precipitation predictability is also explored. Statistical relationships between predictability and the effect of large-scale forcing are usually weak. However, two different types of precipitation systems can be identified: widespread systems associated with mid-latitude cyclones, and mesoscale systems whose evolution is modulated by the diurnal cycle of solar heating. While the overall predictability differs between these two types of cases, the rate at which predictability is lost with forecast time and spatial scale does not.By applying the methodology to the entire data set of 2008-2013, the impact of different types of ensemble perturbations can be assessed. The results show that all the types of perturbations analyzed here fail to generate sufficient dispersion, but that large-scale IC/LBC uncertainties account for most of the forecast error.The gain in QPF skill achieved through radar-data-assimilation, and the advantage of using these sophisticated NWP models rather than simpler statistical forecasting methods for very-short term forecasting 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 | 0.001 | 0.015 |
| 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 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| 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".