Lapse Rate Adjustments of Gridded Surface Temperature Normals in an Area of Complex Terrain: Atmospheric Reanalysis versus Statistical Up-Sampling
Bibliographic record
Abstract
The applicability of elevation-regression based interpolation methods for long-term temperature normals, for example the Parameter-elevation Regressions on Independent Slopes Model (PRISM), becomes increasingly limited in data sparse, complex terrain such as that found in mountainous British Columbia (BC), Canada.Recent methods to improve both the resolution and accuracy of interpolation models have focused on the development of "up-sampling" algorithms based on local lapse rate adjustments to the original interpolated surfaces.Lapse rates can be derived from statistical models (e.g., elevation-based polynomial regression equations) or dynamical models (e.g., vertical temperature profiles from numerical weather prediction (NWP) models).This study compares a widely used statistical up-sampling algorithm, ClimateBC, with two NWP reanalysis products, the National Centers for Environmental Prediction/National Corporation for Atmospheric Research, Reanalysis 1 (NCEP1) and the more modern European Centre for Medium-range Weather Forecasts (ECMWF) Reanalysis Interim (ERA-Interim).Thirty-year climate normals for maximum and minimum temperatures were calculated using statistical up-sampling and NWP lapse rate adjustments to existing PRISM-based climate normals at a subset of stations in BC.Specifically, up-sampling model evaluation was performed using 1951-80 climate normals from an independent set of 54 surface stations (1 m to 2347 m) which were not included in the PRISM interpolation or assimilated into the NWP reanalysis products.All models performed similarly for minimum temperature, which showed only a slight improvement over PRISM.For maximum temperature, ClimateBC, NCEP1 and ERA-Interim all performed significantly better than PRISM, in particular during spring and summer.The ERA-Interim reanalysis outperformed NCEP1 in almost all months.The results suggest that lapse rate adjustment algorithms based on reanalysis products will have greater potential as progress continues on developing NWP components.RSUM [Traduit par la rdaction] L'application des techniques d'interpolation par rgression en fonction de l'altitude pour les normales de temprature long terme, comme le Parameter-elevation Regressions on Independent Slopes Model (PRISM), devient trs difficile dans les rgions accidentes pour lesquelles on dispose de donnes insuffisantes, par exemple les secteurs montagneux de la Colombie-Britannique (C.-B.) au Canada.Les toutes dernires mthodes destines augmenter le degr de rsolution des modles d'interpolation et leur prcision reposent sur la conception d'algorithmes d'chantillonnage vertical fonds sur l'ajustement des surfaces interpoles originales au moyen du gradient vertical local.Nous pouvons tablir les gradients verticaux partir de modles statistiques (p.ex., des quations de rgression polynomiales en fonction de l'altitude) ou de modles dynamiques (p.ex., des profils verticaux de temprature partir de modles de prvision numrique du temps (PNT)).Dans la prsente tude, nous comparons un algorithme d'chantillonnage vertical statistique communment utilis, le programme ClimateBC, deux produits de ranalyse de PNT, celle des National Centres for Environmental Prediction/National Corporation for Atmospheric Research Reanalysis 1 (NCEP1), et la ranalyse provisoire (ERA-Interim) du Centre europen pour les prvisions mtorologiques moyen terme (ECMWF).Les normales climatiques de trente ans pour les tempratures maximums et minimums ont t calcules en appliquant la mthode d'chantillonnage vertical statistique et l'ajustement du gradient obtenu par PNT aux normales climatiques tablies partir du PRISM pour un sous-ensemble de stations en Colombie-Britannique. Plus particulirement, nous avons procd l'valuation du modle d'chantillonnage vertical en nous servant des normales climatiques , pour un ensemble de 54 stations d'observation en surface indpendantes (1 m 2347 m), exclues du modle d'interpolation PRISM et des produits de ranalyse de PNT.Pour tous les modles, nous avons obtenu des rsultats comparables pour la temprature minimum, soit une lgre amlioration seulement par rapport au PRISM.Pour la temprature maximum, nous avons obtenu avec ClimateBC, NCEP1 et ERA-Interim, des rsultats nettement plus probants qu'avec PRISM, notamment au
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) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 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) | 0.004 | 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".