Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India
Abstract Airborne particles finer than 2.5 microns (PM2.5) constitute a major public health risk in India. Therefore, extensive scientific studies must be conducted to assess the PM2.5 exposures of Indians and determine the “exposure-response function” specific to India. While Peninsular India inclu...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2022-05-01
|
| Series: | Aerosol and Air Quality Research |
| Subjects: | |
| Online Access: | https://doi.org/10.4209/aaqr.220110 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825197615974711296 |
|---|---|
| author | V. P. Lavanyaa S. Varshini Souvik Sankar Mitra Kiran M. Hungund Rudrodip Majumdar R. Srikanth |
| author_facet | V. P. Lavanyaa S. Varshini Souvik Sankar Mitra Kiran M. Hungund Rudrodip Majumdar R. Srikanth |
| author_sort | V. P. Lavanyaa |
| collection | DOAJ |
| description | Abstract Airborne particles finer than 2.5 microns (PM2.5) constitute a major public health risk in India. Therefore, extensive scientific studies must be conducted to assess the PM2.5 exposures of Indians and determine the “exposure-response function” specific to India. While Peninsular India includes three megacities with populations exceeding 10 million each, there are very few studies on air quality modelling in this region compared to North India. In this paper, the authors describe a Linear Mixed Effects (LME) model to estimate monthly-average PM2.5 concentrations at a spatial resolution of 1 km2 between 2016 and 2019 in the megacities of Bengaluru and Hyderabad with a total population of 23 million. This model is based on covariates such as aerosol optical depth (AOD), meteorological parameters, and Land-use-Land-cover (LULC) variables and is validated with extensive datasets from continuous and manual air quality monitoring stations through a 10-fold cross-validation process. The final LME model can explain more than 60 percent of the variation in the PM2.5 concentrations in Bengaluru and Hyderabad. This model is then used to predict the monthly-average grid-wise PM2.5 concentrations in more than 800 grids in each of these two cities to study the spatial and temporal patterns in PM2.5 concentrations between 2016 and 2019. These spatiotemporal maps of PM2.5 concentration are critical to overcoming the misclassification of exposure and will form a crucial input to much-needed PM exposure-response studies in these two megacities. This paper can serve as a useful framework for similar studies by showing the way to bridge the gaps in the current air quality monitoring network in Peninsular India. |
| format | Article |
| id | doaj-art-6d08151c9c854bb98ef4d1319cd002a7 |
| institution | Kabale University |
| issn | 1680-8584 2071-1409 |
| language | English |
| publishDate | 2022-05-01 |
| publisher | Springer |
| record_format | Article |
| series | Aerosol and Air Quality Research |
| spelling | doaj-art-6d08151c9c854bb98ef4d1319cd002a72025-02-09T12:17:42ZengSpringerAerosol and Air Quality Research1680-85842071-14092022-05-0122712010.4209/aaqr.220110Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular IndiaV. P. Lavanyaa0S. Varshini1Souvik Sankar Mitra2Kiran M. Hungund3Rudrodip Majumdar4R. Srikanth5National Institute of Advanced Studies, Indian Institute of Science CampusNational Institute of Advanced Studies, Indian Institute of Science CampusNational Institute of Advanced Studies, Indian Institute of Science CampusNational Institute of Advanced Studies, Indian Institute of Science CampusNational Institute of Advanced Studies, Indian Institute of Science CampusNational Institute of Advanced Studies, Indian Institute of Science CampusAbstract Airborne particles finer than 2.5 microns (PM2.5) constitute a major public health risk in India. Therefore, extensive scientific studies must be conducted to assess the PM2.5 exposures of Indians and determine the “exposure-response function” specific to India. While Peninsular India includes three megacities with populations exceeding 10 million each, there are very few studies on air quality modelling in this region compared to North India. In this paper, the authors describe a Linear Mixed Effects (LME) model to estimate monthly-average PM2.5 concentrations at a spatial resolution of 1 km2 between 2016 and 2019 in the megacities of Bengaluru and Hyderabad with a total population of 23 million. This model is based on covariates such as aerosol optical depth (AOD), meteorological parameters, and Land-use-Land-cover (LULC) variables and is validated with extensive datasets from continuous and manual air quality monitoring stations through a 10-fold cross-validation process. The final LME model can explain more than 60 percent of the variation in the PM2.5 concentrations in Bengaluru and Hyderabad. This model is then used to predict the monthly-average grid-wise PM2.5 concentrations in more than 800 grids in each of these two cities to study the spatial and temporal patterns in PM2.5 concentrations between 2016 and 2019. These spatiotemporal maps of PM2.5 concentration are critical to overcoming the misclassification of exposure and will form a crucial input to much-needed PM exposure-response studies in these two megacities. This paper can serve as a useful framework for similar studies by showing the way to bridge the gaps in the current air quality monitoring network in Peninsular India.https://doi.org/10.4209/aaqr.220110Aerosol Optical Depth (AOD)Linear Mixed Effects (LME) modelLULC classificationExposure-response functionSpatiotemporal maps |
| spellingShingle | V. P. Lavanyaa S. Varshini Souvik Sankar Mitra Kiran M. Hungund Rudrodip Majumdar R. Srikanth Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India Aerosol and Air Quality Research Aerosol Optical Depth (AOD) Linear Mixed Effects (LME) model LULC classification Exposure-response function Spatiotemporal maps |
| title | Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India |
| title_full | Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India |
| title_fullStr | Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India |
| title_full_unstemmed | Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India |
| title_short | Geospatial Modelling for Estimation of PM2.5 Concentrations in Two Megacities in Peninsular India |
| title_sort | geospatial modelling for estimation of pm2 5 concentrations in two megacities in peninsular india |
| topic | Aerosol Optical Depth (AOD) Linear Mixed Effects (LME) model LULC classification Exposure-response function Spatiotemporal maps |
| url | https://doi.org/10.4209/aaqr.220110 |
| work_keys_str_mv | AT vplavanyaa geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia AT svarshini geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia AT souviksankarmitra geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia AT kiranmhungund geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia AT rudrodipmajumdar geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia AT rsrikanth geospatialmodellingforestimationofpm25concentrationsintwomegacitiesinpeninsularindia |