Identifying decadal trends in deweathered concentrations of criteria air pollutants in Canadian urban atmospheres with machine learning approaches
Bibliographic record
Abstract
This study investigates long-term trends of criteria air pollutants, including NO 2 , CO, SO 2 , O 3 and PM 2.5 , and O x (meaning NO 2 +O 3 ) measured in 10 Canadian cities during the last 2 to 3 decades. We also investigated associated driving forces in terms of emission reductions, perturbations due to varying weather conditions and large-scale wildfires, as well as changes in O 3 sources and sinks. Two machine learning methods, the random forest algorithm and boosted regression trees, were used to extract deweathered mixing ratios (or mass concentrations) of the pollutants. The Mann–Kendall trend test of the deweathered and original annual average concentrations of the pollutants showed that, on the timescale of 20 years or longer, perturbation due to varying weather conditions on the decadal trends of the pollutants are minimal (within ±2 % ) in about 70 % of the studied cases, although it might be larger (but at most 16 %) in the remaining cases. NO 2 , CO and SO 2 showed decreasing trends in the last 2 to 3 decades in all the cities except CO in Montréal. O 3 showed increasing trends in all the cities except Halifax, mainly due to weakened titration reaction between O 3 and NO. O x , however, showed decreasing trends in all the cities except Victoria, because the increase in O 3 is much less than the decrease in NO 2 . In three of the five eastern Canadian cities, emission reductions dominated the decreasing trends in PM 2.5 , but no significant trends in PM 2.5 were observed in the other two cites. In the five western Canadian cities, increasing or no significant trends in PM 2.5 were observed, likely due to unpredictable large-scale wildfires overwhelming or balancing the impacts of emission reductions on PM 2.5 . In addition, despite improving air quality during the last 2 decades in most cities, an air quality health index of above 10 (representing a very high risk condition) still occasionally occurred after 2010 in western Canadian cities because of the increased large-scale wildfires.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".