Corrected ERA5 Precipitation by Machine Learning Significantly Improved Flow Simulations for the Third Pole Basins
Bibliographic record
Abstract
Abstract Precipitation is one of the most important atmospheric inputs to hydrological models. However, existing precipitation datasets for the Third Pole (TP) basins show large discrepancies in precipitation magnitudes and spatiotemporal patterns, which poses a great challenge to hydrological simulations in the TP basins. In this study, a gridded (10 km × 10 km) daily precipitation dataset is constructed through a random-forest-based machine learning algorithm (RF algorithm) correction of the ERA5 precipitation estimates based on 940 gauges in 11 upper basins of TP for 1951–2020. The dataset is evaluated by gauge observations at point scale and is inversely evaluated by the Variable Infiltration Capacity (VIC) hydrological model linked with a glacier melt algorithm (VIC-Glacier). The corrected ERA5 (ERA5_cor) agrees well with gauge observations after eliminating the severe overestimation in the original ERA5 precipitation. The corrections greatly reduce the original ERA5 precipitation estimates by 10%–50% in 11 basins of the TP and present more details on precipitation spatial variability. The inverse hydrological model evaluation demonstrates the accuracy and rationality, and we provide an updated estimate of runoff components contribution to total runoff in seven upper basins in the TP based on the VIC-Glacier model simulations with the ERA5_cor precipitation. This study provides good precipitation estimates with high spatiotemporal resolution for 11 upper basins in the TP, which are expected to facilitate the hydrological modeling and prediction studies in this high mountainous region. Significance Statement The Third Pole (TP) is the source of water to the people living in the areas downstream. Precipitation is the key driver of the terrestrial hydrological cycle and the most important atmospheric input to land surface hydrological models. However, none of the current precipitation data are equally good for all the TP basins because of high variabilities in their magnitudes and spatiotemporal patterns, posing a great challenge to the hydrological simulation. Therefore, in this study, a gridded daily precipitation dataset (10 km × 10 km) is reconstructed through a random-forest-based machine learning algorithm correction of ERA5 precipitation estimates based on 940 gauges in 11 TP basins for 1951–2020. The data eliminate the severe overestimation of original ERA5 precipitation estimates and present more reasonable spatial variability, and also exhibit a high potential for hydrological application in the TP basins. This study provides long-term precipitation data for climate and hydrological studies and a reference for deriving precipitation in high mountainous regions with complex terrain and limited observations.
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 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 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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 teacher head, 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".