Global River Ice Dataset - validation dataset
Bibliographic record
Abstract
Documentation for nws_breakup_nogeo.csv and nws_freezeup_nogeo.csv Alaskan river ice records from National Weather Service (NWS), including nws_breakup_nogeo.csv containing location (description) and dates of ice breakup and related conditions and nws_freezeup_nogeo.csv containing location (description) and dates of ice freeze-up and related conditions. Note that both dataset do not contain exact geolocations of the observation. We thank Dr. Scott Lindsey at the Alaska-Pacific River Forecast Center for providing these datasets. Documentation for landsat_river_ice_validation.csv This file contains 20,687 same-day river ice condition from Landsat and from in situ, and consists of the following associated properties for each comparison: date: The date on which both the Landsat river ice (length) fraction and in situ river ice condition were observed (data type: string; format: "YYYY-MM-DD"). ice_in_situ: The ice condition on rivers observed in situ. For records from NWS, we assumed river has been ice covered between the date of "first_ice" to the date of "breakup" in the following year and ice-free between the date of "breakup" and the following "first_ice" date. For records from Water Survey of Canada, river was treated as ice-covered whenever the daily "Flow" data were flagged with "B"–meaning backwater effect (data type: integer; range: 0 (ice-free) or 1 (ice-covered)). ice_landsat: The river ice length fraction derived from Landsat image (data type: float; range: [0, 1]). cloud_landsat: The cloud fraction derived from Landsat image (data type: float; range: [0, 0.25]). LANDSAT_SCENE_ID: The unique Landsat TOA image identifier (data type: string). site_id: The ID of the site in its original dataset. dat_source: The source of the in situ river ice record (data type: string; values: ("National Weather Service (Alaska)", "Water Survey of Canada"). longitude: The longitude of the site (data type: float, format: decimal degree). latitude: The latitude of the site (data type: float, format: decimal degree). A subset (N = 18,930) of this dataset was used in the evaluation of the river ice classification. This subset was calculated by applying the following two constraints on the full dataset in the landsat_river_ice_validation.csv: \(cloud\_landsat ≤ 0.05\) \(site\_id \neq 10BE013\) & \(site\_id \neq 08KE016\) The second constraint exclude two Canadian sites from the evaluation as via manual inspection, we found that the Landsat-derived ice fraction for this two sites came from river reaches that were different from where the in situ records were observed.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.023 | 0.039 |
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".