A cross‐checked global monthly weather station database for precipitation covering the period 1901–2010
Bibliographic record
Abstract
Abstract Comprehensive monthly weather station databases are the foundation for many gridded climate data products, and they are widely used to characterize regional climate conditions, track climate change and research the impact of climate on natural and managed ecosystems. However, weather station databases are often regional in coverage, and they can have extensive gaps in station coverage over time. They may also contain errors in climate records, station coordinates or elevation. Here, we assemble a comprehensive monthly weather station database for precipitation from multiple reputable data sources. We use digital elevation models and nearby stations to search for inconsistencies in reported station locations and recorded precipitation values. We also estimated missing values in weather station time series using a linear model approach based on interpolated anomaly surfaces. The resulting station records were ranked into ten classes, according to the completeness of records, the reliability of missing value estimations and other criteria. We corrected incomplete or erroneous location and elevation information for 12% of all available station records. A total of 23% of monthly records that had missing values could be estimated with high or moderate confidence. We sub‐sampled our global database of more than 80,000 stations with various spatial filters, so that only the highest quality station for a given area was retained. Our contribution significantly enhances global data coverage compared to individual databases currently available. Even when accepting only the stations within the top two quality ranks in our combined database, and applying the coarsest spatial filter of one station per approximately 1,600 km2, the remaining station count of more than 20,000 stations exceeds the largest alternative database (without a spatial filter applied) by more than 50%.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.005 | 0.009 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.005 | 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".