Estimating time series of aerosol particle number concentrations in the five HEAPSS cities on the basis of measured air pollution and meteorological variables

详细信息    查看全文

• 作者：Paatero ; Pentti ; Aalto ; Pasi ; Picciotto ; Sally ; Bellander ; Tom ; Casta??o ; Gemma ; Cattani ; Giorgio ; et. al.
• 关键词：Particulate matter ; Retrospective assessment ; Ultrafine particles ; Regularized linear model
• 刊名：Atmospheric Environment
• 出版年：2005
• 出版时间：April, 2005
• 年：2005
• 卷：39
• 期：12
• 页码：2261-2273
• 全文大小：390 K

文摘

Background: Ultrafine particles in urban air represent a potentially important health risk, and are not well characterized by mass concentrations like PM₁₀ or PM_2.5. The aerosol particle number concentration (PNC) is dominated by ultrafine particles, but urban PNC measurement campaigns have only recently started in many cities and missing data impedes much research. Hence, reliable estimation techniques are needed. Past estimations of ambient concentrations of particulate matter have focused on mass concentrations. This project developed city-specific models for estimating PNC using available data on other air pollutants and meteorological variables during a period when PNC was measured, and applied them retrospectively to predict daily PNC levels during the Health Effects of Air Pollution on Susceptible Subpopulations (HEAPSS) study period in order to enable epidemiological analyses.

Methods: Monitoring of PNC began in April 2001 using condensation particle counters (3022A, TSI) in Augsburg, Barcelona, Helsinki, Rome, and Stockholm. Concurrent measurements of air pollutants and weather were used, as well as selected interactions between the two, to fit a regularized linear model (also called ridge regression). This technique is robust with respect to inclusion of irrelevant explanatory variables and can be modified to be highly tolerant of missing data, two highly beneficial features when there are many explanatory variables.

Results: The most important predictor variables were the nitrogen oxides. The models appear to fit PNC data relatively well, with R² of 0.77, 0.80, 0.58, 0.84, 0.81 respectively for the five cities. Split-halves analysis (modelling on half of the data with validation on the other half) indicates that the modelling process was fairly reliable.

Conclusion: A statistical model can be applied to existing data on traffic-related air pollutants and weather variables in order to predict PNC levels. The retrospective prediction of PNC levels appears to be sufficiently reliable for use in epidemiological research.