Forecasting Fine-Grained Air Quality Based on Big Data

TLDR

The study forecasts 48‑hour air‑quality readings at a monitoring station using meteorological, forecast, and neighboring station data. The model integrates a linear‑regression temporal predictor, a neural‑network spatial predictor, a dynamic aggregator that weights them by meteorological data, and an inflection predictor for sudden changes, and is deployed on Bing Maps and Azure. The approach outperforms baselines on data from 43 Chinese cities, has been deployed by the Ministry of Environmental Protection to deliver hourly 48‑hour forecasts for four major cities, and is applicable worldwide.

Abstract

In this paper, we forecast the reading of an air quality monitoring station over the next 48 hours, using a data-driven method that considers current meteorological data, weather forecasts, and air quality data of the station and that of other stations within a few hundred kilometers. Our predictive model is comprised of four major components: 1) a linear regression-based temporal predictor to model the local factors of air quality, 2) a neural network-based spatial predictor to model global factors, 3) a dynamic aggregator combining the predictions of the spatial and temporal predictors according to meteorological data, and 4) an inflection predictor to capture sudden changes in air quality. We evaluate our model with data from 43 cities in China, surpassing the results of multiple baseline methods. We have deployed a system with the Chinese Ministry of Environmental Protection, providing 48-hour fine-grained air quality forecasts for four major Chinese cities every hour. The forecast function is also enabled on Microsoft Bing Map and MS cloud platform Azure. Our technology is general and can be applied globally for other cities.