Development of a river breakup prediction model
Bibliographic record
Abstract
The dynamic breakup of river ice can initiate ice runs where large masses of ice floes accumulate as ice jams. These ice jams can cause severe inundation and infrastructure damage. Several Norwegian rivers are prone to ice run events, however there are currently no models available in Norway for predicting this specific hydrological phenomenon. Ice-related problems are often dealt with on a site-to-site basis and rely heavily on local knowledge. Other countries, such as Canada and Sweden, have implemented statistical, machine learning and process-based modelling approaches. Being able to accurately predict the timing and severity of ice run and ice jam events improves the ability to take suitable mitigation measures and limit negative consequences. The aim of this work is to develop a model to predict ice run events in two Norwegian rivers, the Beiarn River and the Stjørdal River, and thereby address the need for predicting this hydrological hazard.The work presented here is part of a master thesis study that will be completed by May 2023. Both Stjørdal and Beiarn River have been monitored by NVE in the latter half of the 20th century, and the timing and severity of historical ice run events are obtained from this data. The predictors are given by hydrometeorological and ice thickness data, both observed and modelled. The Distance Distribution Dynamics (DDD) model developed by NVE is used for simulating daily discharge, and a simple ice growth model from NVE is used for modelling ice thickness. The prediction model itself is a work in progress, initially taking a logistic regression approach. If time allows, other approaches within machine learning such as random forest will be attempted. The dataset is severely imbalanced given the rarity of ice run events and the limited length of the observed series. Different methods are evaluated in terms of their ability to deal with this issue. The ultimate objective of this project is to develop a model providing daily probabilistic forecasts of the likelihood of ice run events in the coming days.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
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".