Air quality forecasting in a large city

Abstract

We describe the different air pollution statistical forecasting models that have been used in Santiago, Chile during the fall/winter period for the last ten years. Effort has been concentrated on particulate matter PM10 for which a standard of 150 µg/m 3 for the 24 h average is currently established. Inputs to the models are concentrations measured at several monitoring stations distributed throughout the city and meteorological information in the region. Outputs are the expected maxima concentrations for the following day at the site of the same monitoring stations. Forecast values using neural network models are compared with the results obtained with linear models and persistence. Recently, a clustering algorithm has appeared as a potentially useful tool to detect high concentration episodes in advance.

Air quality warnings and temporary driving bans: Evidence from air pollution, car trips, and mass-transit ridership in Santiago

“…The issuance of environmental episodes is based on a pollution forecasting system of Santiago's particulate matter concentrations used by the Ministry of Environment (Ministerio de Medio Ambiente -MMA) since 2000. This daily forecast, based on a prediction model called the Cassmassi model, takes place every winter day at 8pm, generating a 24-hour moving average prediction of PM 10 and PM 2.5 concentrations for the following day (Perez, 2008;Salini, 2009). 14 These expected concentrations correlate with two indexes of Air Quality from Particulates (Índice de Calidad del Aire Referido a Partículas -ICAP) which, inspired by the former US EPA Air Quality Index, were created to easily correlate different levels of PM 10 and PM 2.5 on the same scale.…”

“…where y t+1 is the expected 24-hour moving average of PM 10 on day t+1; ν t is the forecasted atmospheric stability on day t taking discrete values from 1 to 5; y t is the 24-hour moving average of PM 10 measured on day t at 10:00am (local time); x t is the temperature ( • C) of the 925 hP a level registered in the weather station Santo Domingo (located at 80km west of Santiago) at 12:00pm UTC on day t, and h t is 24-hour change in height measured at 500 level registered in Santo Domingo on day t at 12:00pm UTC. The inclusion of x t and h t as part of the equations in the Cassmassi model are intended to control for the strength of thermal inversions in Santiago (Perez, 2008).…”