Related Experiment Video
Updated: Jul 6, 2026
05:45
Composition and Distribution Analysis of Bioaerosols Under Different Environmental Conditions
Published on: January 7, 2019
10.6K
Particulate matter estimation using satellite datasets: a machine learning approach
Sunita Verma1, Ajay Sharma1,2, Swagata Payra3
1Institute of Environment and Sustainable Development, Banaras Hindu University, Varanasi, 221105, Uttar Pradesh, India.
Environmental Science and Pollution Research International
|December 3, 2024
Summary
This study developed an interpretable machine learning model to estimate Particulate Matter 10 (PM10) concentrations in India using satellite Aerosol Optical Depth (AOD). The model achieved a strong R-squared value of 0.78, showing potential for accurate air quality monitoring.
Area of Science:
- Environmental Science
- Atmospheric Science
- Machine Learning Applications
Background:
- Accurate estimation of ground-level Particulate Matter 10 (PM10) concentrations is crucial for air quality assessment in India.
- Satellite-derived Aerosol Optical Depth (AOD) offers a potential data source for PM10 estimation, but requires careful validation and calibration.
- Existing methods may lack interpretability or struggle with regional variations in aerosol properties.
Purpose of the Study:
- To develop and validate an interpretable machine learning model for estimating PM10 concentrations across India.
- To utilize AOD data from INSAT-3D and Moderate Resolution Imaging Spectroradiometer (MODIS) satellites, alongside ground-based measurements.
- To assess the performance of different satellite AOD products and their correlation with ground-level data.
Main Methods:
- Development of an interpretable Random Forest machine learning model for PM10 estimation.
- Utilized 7 years (2014-2020) of satellite AOD data (INSAT-3D, MODIS) and ground PM10 data from Central Pollution Control Board (CPCB).
- Validated satellite AOD against Aerosol Robotic Network (AERONET) data and applied filtering techniques to INSAT-3D AOD for improved correlation.
Main Results:
- MODIS AOD showed good correlation with AERONET AOD; INSAT-3D AOD improved significantly after filtering, achieving correlations of 0.66 (Jaipur) and 0.57 (Kanpur).
- The trained Random Forest model achieved a high R-squared (R²) of 0.78 for PM10 estimation against observed concentrations in 2020.
- The model demonstrated effective training, minimizing overestimation and underestimation, though some instances persisted, suggesting a need for larger datasets.
Conclusions:
- An interpretable machine learning model using calibrated satellite AOD is optimal for PM10 estimation over India.
- The study highlights the importance of AOD data quality control and calibration for accurate air quality modeling.
- Further refinement of the model with expanded datasets can enhance the accuracy of PM10 concentration estimates.
Related Concept Videos
01:29Application of Linearization and Approximation
A drone flying through complex terrain often relies on more than one sensing method to estimate small changes in altitude. Along with direct measurements, air pressure provides a useful indirect indicator of vertical movement. Atmospheric pressure decreases as altitude increases, and this relationship is commonly described using an exponential model. Although accurate, converting pressure measurements into altitude values requires calculations that are too complex to perform repeatedly during...
01:23Linear Approximations
For a differentiable function of two variables, linear approximation estimates values near a known point by replacing the curved surface with its tangent plane. Consider the function\begin{equation*}f(x,y)=x^2+3y^2\end{equation*}near the point (2, 1). The exact value at this point is f(2, 1) = 22 + 3(1)2 = 4 + 3 = 7.The linear approximation of f(x, y)) near (a, b) is\begin{equation*}L(x,y)=f(a,b)+f_x(a,b)(x-a)+f_y(a,b)(y-b)\end{equation*}First, compute the partial derivatives: fx(x, y) = 2x and...