Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model
Nitrogen dioxide (NO<sub>2</sub>) is an important pollutant related to human activities, which has short-term and long-term effects on human health. An ensemble learning model was constructed and applied to estimate daily NO<sub>2</sub> concentrations in the Beijing–Tianjin–H...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-02-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/13/4/758 |
_version_ | 1797396065373126656 |
---|---|
author | Yanding Pan Chen Zhao Zhaorong Liu |
author_facet | Yanding Pan Chen Zhao Zhaorong Liu |
author_sort | Yanding Pan |
collection | DOAJ |
description | Nitrogen dioxide (NO<sub>2</sub>) is an important pollutant related to human activities, which has short-term and long-term effects on human health. An ensemble learning model was constructed and applied to estimate daily NO<sub>2</sub> concentrations in the Beijing–Tianjin–Hebei region between 2010 and 2016. A variety of predictive variables included satellite-based troposphere NO<sub>2</sub> vertical column concentration, meteorology, elevation, gross domestic product (GDP), population, land-use variables, and road network. The ensemble learning model achieved two things: a 0.01° × 0.01° grid resolution and the estimation of historical data for the years 2010–2013. The ensemble model showed good performance, whereby the <i>R</i><sup>2</sup> of tenfold cross-validation was 0.72 and the <i>R</i><sup>2</sup> of test validation was 0.71. Meteorological hysteretic effects were incorporated into the model, where the one-day lagged boundary layer height contributed the most. The annual NO<sub>2</sub> estimation showed little change from 2010 to 2016. The seasonal NO<sub>2</sub> estimation from highest to lowest occurred in winter, autumn, spring, and summer. In the annual maps and seasonal maps, the NO<sub>2</sub> estimations in the northwest region were lower than those in the southeast region, and there was a heavily polluted band in the south of the Taihang Mountains. In coastal areas, the annual NO<sub>2</sub> estimations were higher than the NO<sub>2</sub> monitored values. The drawback of the model is underestimation at high values and overestimation at low values. This study indicates that the ensemble learning model has excellent performance in the simulation of NO<sub>2</sub> with high spatial and temporal resolution. Furthermore, the research framework in this study can be a generally applied for drawing implications for other regions, especially for other cities in China. |
first_indexed | 2024-03-09T00:44:55Z |
format | Article |
id | doaj.art-a4abfb401e834910a7ade03807823257 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T00:44:55Z |
publishDate | 2021-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-a4abfb401e834910a7ade038078232572023-12-11T17:34:14ZengMDPI AGRemote Sensing2072-42922021-02-0113475810.3390/rs13040758Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning ModelYanding Pan0Chen Zhao1Zhaorong Liu2College of Environmental Sciences and Engineering, Peking University, Beijing 100871, ChinaInternational Economic & Technical Cooperation and Exchange Center, Ministry of Water Resources, Beijing 100053, ChinaCollege of Environmental Sciences and Engineering, Peking University, Beijing 100871, ChinaNitrogen dioxide (NO<sub>2</sub>) is an important pollutant related to human activities, which has short-term and long-term effects on human health. An ensemble learning model was constructed and applied to estimate daily NO<sub>2</sub> concentrations in the Beijing–Tianjin–Hebei region between 2010 and 2016. A variety of predictive variables included satellite-based troposphere NO<sub>2</sub> vertical column concentration, meteorology, elevation, gross domestic product (GDP), population, land-use variables, and road network. The ensemble learning model achieved two things: a 0.01° × 0.01° grid resolution and the estimation of historical data for the years 2010–2013. The ensemble model showed good performance, whereby the <i>R</i><sup>2</sup> of tenfold cross-validation was 0.72 and the <i>R</i><sup>2</sup> of test validation was 0.71. Meteorological hysteretic effects were incorporated into the model, where the one-day lagged boundary layer height contributed the most. The annual NO<sub>2</sub> estimation showed little change from 2010 to 2016. The seasonal NO<sub>2</sub> estimation from highest to lowest occurred in winter, autumn, spring, and summer. In the annual maps and seasonal maps, the NO<sub>2</sub> estimations in the northwest region were lower than those in the southeast region, and there was a heavily polluted band in the south of the Taihang Mountains. In coastal areas, the annual NO<sub>2</sub> estimations were higher than the NO<sub>2</sub> monitored values. The drawback of the model is underestimation at high values and overestimation at low values. This study indicates that the ensemble learning model has excellent performance in the simulation of NO<sub>2</sub> with high spatial and temporal resolution. Furthermore, the research framework in this study can be a generally applied for drawing implications for other regions, especially for other cities in China.https://www.mdpi.com/2072-4292/13/4/758NO<sub>2</sub> estimationhigh spatiotemporal resolutionensemble learning modeltypical polluted area |
spellingShingle | Yanding Pan Chen Zhao Zhaorong Liu Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model Remote Sensing NO<sub>2</sub> estimation high spatiotemporal resolution ensemble learning model typical polluted area |
title | Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model |
title_full | Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model |
title_fullStr | Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model |
title_full_unstemmed | Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model |
title_short | Estimating the Daily NO<sub>2</sub> Concentration with High Spatial Resolution in the Beijing–Tianjin–Hebei Region Using an Ensemble Learning Model |
title_sort | estimating the daily no sub 2 sub concentration with high spatial resolution in the beijing tianjin hebei region using an ensemble learning model |
topic | NO<sub>2</sub> estimation high spatiotemporal resolution ensemble learning model typical polluted area |
url | https://www.mdpi.com/2072-4292/13/4/758 |
work_keys_str_mv | AT yandingpan estimatingthedailynosub2subconcentrationwithhighspatialresolutioninthebeijingtianjinhebeiregionusinganensemblelearningmodel AT chenzhao estimatingthedailynosub2subconcentrationwithhighspatialresolutioninthebeijingtianjinhebeiregionusinganensemblelearningmodel AT zhaorongliu estimatingthedailynosub2subconcentrationwithhighspatialresolutioninthebeijingtianjinhebeiregionusinganensemblelearningmodel |