A global spatial-temporal land use regression model for nitrogen dioxide air pollution
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Introduction: The World Health Organization (WHO) recently revised its health guidelines for Nitrogen dioxide (NO 2 ) air pollution, reducing the annual mean NO 2 level to 10 μg/m 3 (5.3 ppb) and the 24-h mean to 25 μg/m 3 (13.3 ppb). NO 2 is a pollutant of global concern, but it is also a criteria air pollutant that varies spatiotemporally at fine resolutions due to its relatively short lifetime (~hours). Current models have limited ability to capture both temporal and spatial NO 2 variation and none are available with global coverage. Land use regression (LUR) models that incorporate timevarying predictors (e.g., meteorology and satellite NO 2 measures) and land use characteristics (e.g., road density, emission sources) have significant potential to address this need. Methods: We created a daily Land use regression model with 50 × 50 m 2 spatial resolution using 5.7 million daily air monitor averages collected from 8,250 monitor locations. Results: In cross-validation, the model captured 47%, 59%, and 63% of daily, monthly, and annual global NO 2 variation. Daily, monthly, and annual root mean square error were 6.8, 5.0, and 4.4 ppb and absolute bias were 46%, 30%, and 21%, respectively. The final model has 11 variables, including road density and built environments with fine (30 m or less) spatial resolution and meteorological and satellite data with daily temporal resolution. Major roads and satellite-based estimates of NO 2 were consistently the strongest predictors of NO 2 measurements in all regions. Discussion: Daily model estimates from 2005–2019 are available and can be used for global risk assessments and health studies, particularly in countries without NO 2 monitoring.
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.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it