Critical Role of Secondary Organic Aerosol in Urban Atmospheric Visibility Improvement Identified by Machine Learning
Bibliographic record
Abstract
Understanding the relationship between atmospheric visibility and aerosol emission sources and identifying the key drivers of visibility have significant implications for the radiative forcing of aerosol. In this work, we combined the positive matrix factorization (PMF) model and machine learning (ML) models (the extreme gradient boosting model (XGBoost) and the Shapely additive explanations model (SHAP)) to identify the key drivers of visibility improvement based on long-term observations of visibility and PM 2.5 composition in Shenzhen, China. From 2014 to 2021, the annual average levels of visibility increased from 17.2 to 27.0 km, which is tightly associated with the decreasing year by year PM 2.5 concentrations. ML models, with distinct advantages in dealing with nonlinear relationships, revealed that secondary organic aerosol (SOA) is the major driver determining visibility, which is inconsistent with inorganic salts being the major driver identified by the widely used traditional linear method. Visibility improvement in Shenzhen was also found primarily driven by a decrease in SOA, highlighting that SOA in PM 2.5 plays a critical role in radiative balance. This is the first study to investigate source impacts on atmospheric visibility using novel ML models, reflecting the great potential of ML methods in air pollution data analysis.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| 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".