Scholarlink

<p>Small island developing states face air-quality management challenges from both local emissions and recurrent transboundary Saharan dust. This study presents a WHO-2021 Compliance and Dust-Day Screening Framework for Trinidad and Tobago that integrates PM₂.₅ / PM₁₀ monitoring with routine meteorology to identify guideline exceedance days and estimate whether they were dust-influenced. Valid days were defined using a ≥ 75% daily completeness rule, with sensitivity checks at alternate capture thresholds. Dust screening combined a transparent heuristic candidate screen with an adjudicated labeling workflow using local records and external corroboration. The probabilistic Dust Attribution Score (DAS) model used logistic regression with physically interpretable predictors, including RH anomaly, easterly transport, rainfall absence and regional coherence. Validation against a temporally independent 2024 holdout showed strong discrimination (AUC ≈ 0.92) and good calibration (Brier ≈ 0.12) relative to the adjudicated operational labels, with high-confidence DAS values showing high specificity for dust-influenced days. Because the candidate screen and DAS model share the same four physical signal domains (RH anomaly, transport, rainfall and regional coherence), these estimates are best interpreted as validation of an operationally consistent workflow rather than of a fully independent reference standard. Applied to Trinidad and Tobago during 2022–2024, the framework indicated annual PM₂.₅ levels of roughly 10–18&#xa0;µg m⁻³ at monitored sites and frequent short-term guideline exceedances, many clustered during major Saharan dust episodes. Illustrative counterfactual scenario analyses suggest that some non-extreme exceedance days may be sensitive to reductions in locally generated PM under simplified fixed-assumption conditions, whereas the largest dust episodes would likely still require preparedness and risk communication. These scenarios are intended for policy screening rather than as dynamic forecasts of intervention effects. The framework supports a prototype Dust Watch / Dust Warning workflow for Trinidad and Tobago and may provide a structured template for other dust-affected settings, subject to local calibration, prospective testing and external validation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

WHO-2021 compliance and dust-day screening framework for small-island air quality management: development and application in Trinidad & Tobago

  • 期刊论文

  • 发表日期：

    2026年5月28日

查看全文
  • Treina Dinoo Ramlochan,
  • Koffka Khan,
  • Wayne Rajkumar,
  • Shanta Ramnath,
  • Steven Rampersad,
  • Winston Elibox

摘要

Small island developing states face air-quality management challenges from both local emissions and recurrent transboundary Saharan dust. This study presents a WHO-2021 Compliance and Dust-Day Screening Framework for Trinidad and Tobago that integrates PM₂.₅ / PM₁₀ monitoring with routine meteorology to identify guideline exceedance days and estimate whether they were dust-influenced. Valid days were defined using a ≥ 75% daily completeness rule, with sensitivity checks at alternate capture thresholds. Dust screening combined a transparent heuristic candidate screen with an adjudicated labeling workflow using local records and external corroboration. The probabilistic Dust Attribution Score (DAS) model used logistic regression with physically interpretable predictors, including RH anomaly, easterly transport, rainfall absence and regional coherence. Validation against a temporally independent 2024 holdout showed strong discrimination (AUC ≈ 0.92) and good calibration (Brier ≈ 0.12) relative to the adjudicated operational labels, with high-confidence DAS values showing high specificity for dust-influenced days. Because the candidate screen and DAS model share the same four physical signal domains (RH anomaly, transport, rainfall and regional coherence), these estimates are best interpreted as validation of an operationally consistent workflow rather than of a fully independent reference standard. Applied to Trinidad and Tobago during 2022–2024, the framework indicated annual PM₂.₅ levels of roughly 10–18 µg m⁻³ at monitored sites and frequent short-term guideline exceedances, many clustered during major Saharan dust episodes. Illustrative counterfactual scenario analyses suggest that some non-extreme exceedance days may be sensitive to reductions in locally generated PM under simplified fixed-assumption conditions, whereas the largest dust episodes would likely still require preparedness and risk communication. These scenarios are intended for policy screening rather than as dynamic forecasts of intervention effects. The framework supports a prototype Dust Watch / Dust Warning workflow for Trinidad and Tobago and may provide a structured template for other dust-affected settings, subject to local calibration, prospective testing and external validation.