Deep Learning for Forecasting Runoffs over China under Climate Changes
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
The temporal and spatial distribution of water resources over China has changed and may continue changing in the future under ongoing global warming. Scientific water resources management requires reliable forecasting of the change. Meanwhile, the performance of deep learning in achieving it has not been comprehensively explored. To fill this gap, deep learning, i.e., multilayer perceptron (MLP) in this study, is used to study the change of streamflow over China under climate changes. MLP is compared with other machine learning methods for investigating its strengths, and three river basins (i.e., Xiangxi, Jinghe and Zhongzhou) in central, northwestern and southeastern China, respectively are selected to represent hydrologic regimes over China. Four regional climate models are used to drive MLP for forecasting streamflow from 2021 to 2050 under two greenhouse-gas emission scenarios (i.e., RCPs 4.5 and 8.5). Modeling results show that MLP is more accurate than the other methods, especially in terms of peak streamflow volumes. Annual average temperature in the three basins will increase, while precipitation shows different changing trends. The simulation accuracies among the regional climate models (RCMs) are slightly different. Correspondingly, streamflow will increase, and the increments decrease from Jinghe, through Xiangxi, to Zhongzhou River Basins. Due to climate changes, flooding will become more frequent in Jinghe and Xiangxi River Basins, Jinghe River Basin will experience no runoff in winter, and the timing of peak runoffs in Zhongzhou River Basin will move forward. Compared with the RCP 4.5 scenario, the above trends are more obvious under the RCP 8.5 scenario.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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