Back to Search
Start Over
A Framework to Predict High-Resolution Spatiotemporal PM2.5 Distributions Using a Deep-Learning Model: A Case Study of Shijiazhuang, China.
- Source :
- Remote Sensing; sep2020, Vol. 12 Issue 17, p2825, 1p
- Publication Year :
- 2020
-
Abstract
- Air-borne particulate matter, PM<subscript>2.5</subscript> (PM having a diameter of less than 2.5 micrometers), has aroused widespread concern and is a core indicator of severe air pollution in many cities globally. In our study, we present a validated framework to predict the daily PM<subscript>2.5</subscript> distributions, exemplified by a use case of Shijiazhuang City, China, based on daily aerosol optical depth (AOD) datasets. The framework involves obtaining the high-resolution spatiotemporal AOD distributions, estimation of the spatial distributions of PM<subscript>2.5</subscript> and the prediction of these based on a convolutional long short-term memory (ConvLSTM) model. In the estimation part, the eXtreme gradient boosting (XGBoost) model has been determined as the estimation model with the lowest root mean square error (RMSE) of 32.86 µg/m<superscript>3</superscript> and the highest coefficient of determination regression score function (R<superscript>2</superscript>) of 0.71, compared to other common models used as a baseline for comparison (linear, ridge, least absolute shrinkage and selection operator (LASSO) and cubist). For the prediction part, after validation and comparison with a seasonal autoregressive integrated moving average (SARIMA), which is a traditional time-series prediction model, in both time and space, the ConvLSTM gives a more accurate performance for the prediction, with a total average prediction RMSE of 14.94 µg/m<superscript>3</superscript> compared to SARIMA's 17.41 µg/m<superscript>3</superscript>. Furthermore, ConvLSTM is more stable and with less fluctuations for the prediction of PM<subscript>2.5</subscript> in time, and it can also eliminate better the spatial predicted errors compared to SARIMA. [ABSTRACT FROM AUTHOR]
- Subjects :
- BOX-Jenkins forecasting
STANDARD deviations
FORECASTING
PARTICULATE matter
Subjects
Details
- Language :
- English
- ISSN :
- 20724292
- Volume :
- 12
- Issue :
- 17
- Database :
- Complementary Index
- Journal :
- Remote Sensing
- Publication Type :
- Academic Journal
- Accession number :
- 145987261
- Full Text :
- https://doi.org/10.3390/rs12172825