Influence of synoptic and local atmospheric patterns on PM10 air pollution levels: a model application to Naples (Italy)

TL;DR

This study analyzes how synoptic and local atmospheric patterns influence PM10 pollution in Naples, Italy, and develops a predictive model based on meteorological conditions to forecast pollution crises with high accuracy.

Contribution

It introduces a new empirical model that predicts PM10 levels using local meteorological data, tailored to Naples' topography and weather patterns.

Findings

Severe pollution correlates with high geopotential heights at 850 hPa.

Wind stress between 1 and 2 m/s and thermal inversion of at least 3°C/200m predict pollution peaks.

The model achieves a correlation coefficient of 0.80 with observed PM10 data.

Abstract

We investigate the relationship between synoptic/local meteorological patterns and PM10 air pollution levels in the metropolitan area of Naples, Italy. We found that severe air pollution crises occurred when the 850 and 500 hpa geopotential heights and their relative temperatures present maximum values above the city. The most relevant synoptic parameter was the 850 hPa geopotential height, which is located about 1500 m of altitude. We compared local meteorological conditions (specifically wind stress, rain amount and thermal inversion) against the urban air pollution levels from 2009 to 2013. We found several empirical criteria for forecasting high daily PM10 air pollution levels in Naples. Pollution crises occurred when (a) the wind stress was between 1 and 2 m/s, (b) the thermal inversion between two strategic locations was at least 3{\deg}C/200m and (c) it did not significantly rain for at least 7 days. Beside these meteorological conditions, severe pollution crises occurred also during festivals when fireworks and bonfires are lighted, and during anomalous breeze conditions and severe fire accidents. Finally, we propose a basic model to predict PM10 concentration levels from local meteorological conditions that can be easily forecast a few days in advance. The synthetic PM10 record predicted by the model was found to correlate with the PM10 observations with a correlation coefficient close to 0.80 with a confidence greater than 99%. The proposed model is expected to provide reliable information to city officials to carry out practical strategies to mitigate air pollution effects. Although the proposed model equation is calibrated on the topographical and meteorological conditions of Naples, it should be easily adaptable to alternative locations.