The repository here contains supporting documents for the study "Long-term changes in aerosol loading over the ‘BIHAR’ State of India using nineteen years (2001-2019) of high-resolution satellite data (1 x 1 km2)" published in Atmospheric Pollution Research Journal. The article can be accessed here https://www.sciencedirect.com/science/article/pii/S1309104221003214
The study is carried out to analyse long term trend on Aerosol Optical Depth (AOD) retrieved using satellite data for the state of Bihar, India. Indo-Gangetic Plain (IGP) in the Indian sub-continent faces massive aerosol loading, for which it is regarded as a global air pollution hotspot. This study examined the spatial variation in columnar aerosol loading (2001–2019) over the eastern state of Bihar and its 38 administrative districts affected by transboundary transport and local emissions. In the current study, aerosol optical depth (AOD0.55μm) retrieved at 1 × 1 Km2 resolution by the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor using the Multi-Angle Implementation of Atmospheric Correction algorithm (MCD19A2, Collection 6) was analyzed. A significant increase in AOD from 0.49 to 0.68, with an inter-annual variability of 12.0%, was observed. The highest seasonal aerosol loading of the range 0.51–0.75 was observed during the post-monsoon (OND) and winter (JF) seasons due to long-range transport of pollutants from upper and central IGP regions and aerosols emitted locally, which further was aggravated by poor metrological conditions. AOD was not found to vary significantly with the land use pattern (e.g., urban, rural, and background), implying the substantial influence of regional transport. A significant increase in the annual AOD trend (0.0106 year−1) and seasonal trends (0.0072–0.0182 year−1) was observed for the analysis period. Districts located along the Ganga river stretch exhibited the highest annual AOD rate with an overall percentage increase of 40–50%. Low to moderate model performance (R2:0.25–0.65; MAPE: 15.2–26.7%) was exhibited by Auto-Regressive Integrated Moving Average (ARIMA) time series model over the study area for the successive two years (2020 and 2021), emphasizing the districts with a potential for high aerosol loading that requires immediate addressing under Business as usual (BAU) scenarios. The overall study augments policymakers with decision-making support to instigate air quality measures at the state epicenters. The study also recommends inter-state coordination to develop an integrated airshed management approach for holistic improvement in state air quality.
Essential libraries/datasets for performing the analysis
- Satellite AOD dataset (Used google earth engine to retreive the data)
- Global Human settlement Layer dataset
- Python IDE
- Rasterio library for analysing the satellite data
- Gdal library for processing raster data
- geopandas library for geopsatial data
- Matplotlib or seborn for data visulaisation
- PIL library for GIF visualisation
- Pandas and Numpy
- R IDE
- tidyverse, dplyr for basic data cleaning
- Forecast library
- Openair library for Mann-Kendall and HYSPLIT analysis.
- factoextra for clustering analysis
- ggplot2 for visualisation
Graphical visualisation summarising the overall study
