Mass-balance model inputs for the Kaskawulsh River headwaters
Bibliographic record
Abstract
This dataset contains the input files necessary to run the mass-balance model (see Young et al. (2021), Robinson et al. (2024)) for the Kaskawulsh River headwaters from 1980–2022. Downscaled/bias corrected temperature/precipitation inputs will be uploaded separately due to the large file sizes. Downscaling > CoarseNARR_KRH This folder contains the coarse North American Regional Reanalysis (NARR) data that were downscaled to obtain the temperature/precipitation inputs used to drive the Kaskawulsh River headwaters mass-balance model (see Robinson, 2024). All files contain an array of 6 x 6 gridcells, centered on the Kaskawulsh Glacier, Yukon, Canada. The bottom left corner of the grid has coordinates 502592 E, 6624579 N and is located over Yakutat Bay, Alaska. Variables: air: NARR 3-hourly multi-level air temperatures hgt: NARR 3-hourly multi-level geopotential height apcp: NARR daily total surface precipitation KRH_CE.nc: Time-invariant elevation of NARR gricells Input_geometry > KRH This folder contains the input geometries needed to run the model for the Kaskawulsh River headwaters. KRH_Xgrid.txt / KRH_Ygrid.txt - contains the Easting/Northing (UTM 7N) of each gridcell in the model domain. Zgrids - a folder containing the annual surface elevation grids (m a.s.l.) for 1979–2022 (see Robinson et al. (2024) for details on how these were derived. KRH_SfcType.txt - a value of 0 indicates the gridcell is glacierized terrain, a value of 1 indicates that the gridcell is non-glacierized terrain. KRH_Tributaries.txt - gridcell value indicates which portion of the Kaskawulsh Glacier the gridcell belongs to (1 = South Arm, 2 = Stairway Glacier, 3 = Central Arm, 4 = North Arm, 5 = Trunk). Solar Contains one file per year with the 3-hourly potential direct clear-sky solar radiation, calculated using the shading program from the Hock (1999) Distributed Enhanced Temperature-Index Model (DETIM). For details see Hock (1999). Debris ‘KRH_debrismap.txt’ is a text file with dimensions x,y = 230 gridcells x 329 grid cells. Gridcells are 200 m x 200 m. Each gridcell contains either a value representing the estimated debris thickness (in meters), or NaN if the gridcell is not debris covered. Debris thicknesses are from the Rounce et al. (2021) global debris thickness dataset, interpolated to the Kaskawulsh River headwaters 200 m domain by Robinson (2024). For details on the debris thickness estimate, see Rounce et al. (2021). For details on the interpolation of the original debris thickness estimate to the gridcell size of the model, see Robinson (2024). Tuning > parameters This directory contains a folder ‘initial_params’ with all the original values of a_ice, a_snow, and MF used to tune all the mass-balance models in Robinson et al. (2024) (in groups of 10,000 param combinations). Each additional directory contains the final 100 parameter combinations tuned for each model presented in Robinson et al. (2024), including: The reference model. The debris-free model. The model with sub-debris melt-scaling from Rounce et al. (2021). The model with uncorrected accumulation. The model with accumulation bias corrected with precipitation gauge data. Tuning > snowlines Contains 53 .npy files with the rasterized versions of observed snow cover on the Kaskawulsh Glacier delineated from satellite images (see Robinson, 2024). These rasters are used in the second stage of tuning to compare modelled and observed snowlines.
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 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 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) | 0.098 | 0.037 |
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".