Novel methods for estimating extreme design rainfalls at gauged and ungauged locations in a changing climate
Bibliographic record
Abstract
Information on the variability of extreme rainfalls in time and in space is of critical importance for many types of extreme hydrologic studies related to the estimation of runoff characteristics for planning, design, and management of various water resources systems.In particular, for urban watersheds that are generally characterized by a fast response, the design of different urban infrastructures (such as small dams, culverts, storm sewers, detention basins, and so on) require hence an accurate and robust estimation of extreme design rainfalls for very high temporal resolutions (ranging from a few minutes to one day) in order to provide an accurate and reliable description of runoff properties for urban inundation management.In addition, in recent years, climate change has been recognized as having a profound impact on the hydrologic cycle at different temporal and spatial scales.Consequently, the intensity and frequency of extreme storm events in most regions will be likely increased in the future.The present study is therefore was carried out to develop appropriate methods for improving the accuracy of design rainfall estimation at gauged and ungauged locations in the current climate as well as in the context of climate variability and climate change.This study can be divided into five primary parts.The first part presents a general procedure for assessing systematically the performance of different commonly used probability distributions in extreme rainfall frequency analyses based on their descriptive as well as predictive abilities.This assessment procedure relies on an extensive set of graphical and numerical performance criteria to identify the most suitable models that could provide the most accurate and most robust extreme rainfall estimates.The proposed systematic iv assessment approach has been shown to be more efficient and more robust than the traditional model selection method based on only limited goodness-of-fit criteria.To test the feasibility of the proposed procedure, an illustrative application was carried out using 5-minute, 1-hour, and 24hour annual maximum rainfall data from a network of 21 raingages located in the Ontario region in Canada.Results have indicated that the Generalized Extreme Values (GEV), Generalized Normal (GNO), and Pearson Type 3 (PE3) models were the best models for describing the distribution of daily and sub-daily annual maximum rainfalls in this region.The GEV distribution, however, was preferred to the GNO and PE3 because it was based on a more solid theoretical basis for representing the distribution of extreme random variables.The second part introduces a new probability-weighted-moment-based scaling Generalized Extreme Value (GEV/PWM) distribution model for modeling rainfall extremes across a wide range of time scales (e.g., from several minutes to one day).The GEV distribution has been recommended in the national guidelines of many countries.The mathematical framework and the scaling properties of the proposed GEV/PWM model were derived.The relations between the GEV/PWM model and three existing scaling models such as the non-central-moment-based GEV (GEV/NCM) and the NCM-and PWM-based Gumbel models (GUM/NCM and GUM/PWM) were described.A comparative study was then carried out to asses the performance of these models using the available extreme rainfall data from a network of 74 raingages located across Canada.The scaling behaviours of extreme rainfall processes were also analyzed using both NCM and PWM estimation methods.Results of this comparative study have indicated the superior performance of the proposed GEV/PWM model as compared to the existing GEV/NCM, GUM/NCM, and GUM/PWM based on an extensive set of graphical and numerical comparison Comparing & Assessing results Considering other criteria Climate Scenarios Regression models • Real space: 2 or 3 parameters • Log space: 1st to 6th order polynm Rainfall Frequency Atlas/Maps Scale-invariance models Design Storm Ch.2 Ch.4 Ch.6 -1.24 -0.96 -0.96 -0.82 -0.35 -0.21 -0.80 -0.89 -1.36 38 -0.09 -0.22 -0.55 -1.15 -1.58 -1.58 -2.11 -2.45 -1.99 2 -0.21 -0.47 -0.82 -0.67 -0.11 -0.09 -0.06 -0.74 -1.27 39 -0.43 -0.80 -0.84 -1.49-1.05 -0.48 -0.44 -0.71 -1.12 3 -0.60 -0.66 -0.13 -0.63 -1.54 -1.38 -0.84 -0.16 -0.09 40 -0.14 -0.43 -0.47 -1.07 -1.13 -0.32 -1.13 -0.87 -0.03 4 -1.29 -1.39 -1.29 -0.57-0.65 -1.39 -1.85 -0.79 -1.29 41 -0.77 -0.57-0.59 -1.02 -0.54 -0.24 -0.29 -0.44 -0.44 5 -0.24 -0.27 -0.22 -0.11 -0.57-1.14 -1.19 -1.03 -0.08 42 -1.34-1.22 -1.53 -0.39 -0.29 -0.16 -0.23 -0.02 -0.72 6 -0.59 -0.32 -0.30 -0.05 -0.82 -2.11 -2.41 -2.88 -0.77 43 -1.17 -1.36 -1.43 -1.64 -1.30 -0.57-0.18 -0.29 -0.63 7 -0.19 -0.02 -0.49-0.36 -0.45 -0.62 -0.83 -0.96 -1.40 44 -0.60 -0.25 -0.08 -0.16 -0.07 -0.08 -1.31 -1.02 -0.56 8 -0.63 -0.34 -0.10 -0.18 -0.13 -0.42 -1.36 -0.52 -0.63 45 -1.41 -1.20 -1.27 -0.98 -1.48 -0.80 -0.58 -0.99 -2.26 9 -1.39 -1.22 -0.91 -0.58 -1.44 -1.31 -1.08 -0.58 -0.35 46 -0.80 -0.62 -0.42 -0.34 -0.94 -1.28 -0.80 -0.60 -0.72 10 -0.89 -0.62 -0.81 -1.00 -0.65 -0.43 -1.16 -1.54 -1.46 47 -1.38 -1.27 -1.00 -0.92 -1.08 -0.08 -0.03 -0.24 -0.81 11 -1.58 -1.13 -0.95 -1.47 -1.29 -1.95 -0.01 -0.65 -1.78 48 -0.36 -1.27 -1.83 -1.70 -1.85 -1.50 -0.26 -0.01 -0.30 12 -2.98 -1.53 -0.92 -1.03 -1.69 -1.90 -1.27 -2.11 -2.79 49 -0.05 -0.88 -1.00 -0.46 -0.41 -0.47 -0.44 -0.73 -0.48 13 -2.10 -1.67 -1.85 -1.70 -1.22 -0.49-1.12 -1.45 -2.25 50 -2.29 -2.76 -1.67 -1.74 -2.06 -1.69 -1.83 -1.89 -0.33 14 -1.88 -2.02 -2.11 -2.58 -2.32 -1.95 -1.31 -1.74 -2.04 51 -0.62 -1.36 -1.36 -1.81 -2.05 -1.24 -0.21 -0.87 -1.60 15 -0.73 -0.17 -0.13 -0.21 -0.55 -0.60 -0.06 0.00 -0.40 52 -0.25 -0.98 -0.89 -0.87 -0.42 -0.39 -0.22 -0.33 -0.01 16 -0.26-0.38 -0.58 -1.00 -1.27 -0.88 -0.25 -0.83 -1.14 53 -0.40 -0.49-0.40 -0.40 -0.66 -1.04 -0.15 -0.02 -0.08 17 -1.53-1.24 -0.95 -0.70 -0.70 -0.66 -0.54 -0.31 -0.19 54 -0.89 -0.35 -0.64 -0.46 -0.85 -1.63 -2.16 -2.35 -2.5718 -1.49-1.23 -0.59 -0.17 -0.36 -1.15 -0.47 -1.11 -0.89 55 -0.71 -0.44 -0.77 -1.03 -0.55 -0.77 -1.34 -1.02 -0.59 19 -0.02 -0.19 -0.15 -0.32 -0.11 -0.19 -0.06 -0.28 -0.36 56 -1.99 -1.66 -1.13 -0.34 -1.44 -1.46 -0.91 -1.50 -1.68 20 -1.30 -0.70 -0.10 -0.37 -0.60 -0.47 -0.63 -0.65 -0.44 57 -0.30 -1.26 -1.66 -1.53 -0.03 -0.08 -0.63 -0.32 -0.43 21 -0.35 -0.16 -0.12 -0.23 -0.06 -0.06 -0.74 -0.87 -0.99 58 -0.36 -1.61 -2.52 -2.88 -2.13 -1.14 -0.86 -0.14 -1.36 22 -0.60 -0.31 -0.13 -0.16 -0.21 -0.68 -0.39 -1.17 -1.12 59 -1.70 -1.75 -1.49-2.43 -2.48 -2.06 -2.45 -1.72 -1.56 23 -0.59 -0.11 -0.18 -0.33 -0.15 -0.13 -0.38 -0.68 -1.19 60 -2.19 -0.89 -0.75 -0.27 -0.61 -0.38 -1.17 -0.17 -0.84 24 -1.30-1.14 -1.62 -1.38 -1.20 -1.42 -0.48 -0.92 -1.44 61 -1.34 -0.75 -0.27 -0.59 -1.09 -0.48 -0.52 -0.77 -1.50 25 -0.04 -0.60 -0.26 -0.66 -0.64 -0.51 -0.66 -0.95 -0.5762 -0.41 -0.18 -0.82 -1.02 -0.79 -0.25 -1.34 -1.50 -0.68 26 -0.68 -1.10 -1.48 -1.46 -1.12 -1.26 -1.84 -1.20 -1.14 63 -1.55 -1.96 -1.75 -1.26 -0.85 -0.75 -0.17 -0.19 -0.06 27 -1.22 -1.08 -1.30 -1.19 -1.30 -0.87 -0.89 -0.84 -0.78 64 -2.43 -1.43 -1.13 -0.70 -0.32 -0.28 -0.60 -0.47 -0.02 28 -0.18 -0.16 0.00 -0.08 -0.10 -0.39 -0.42 -0.26 -0.10 65 -0.44 -0.58 -0.43 -0.97 -1.38 -0.77 -1.08 -1.40 -1.94 29 -1.41 -1.62 -1.43 -
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 | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 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) | 0.001 | 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".