Enhancing Prediction of Satellite-Based NO2 Concentration in ASEAN Using Random Forest Models with Spatial Coordinates and Temporal Lag Features

Nabil Naufal; Anik Djuraidah; Yenni Angraini

doi:10.59395/ijadis.v7i2.1602

Authors

Nabil Naufal Study Program of Statistics and Data Science, Bogor Agricultural University, Indonesia https://orcid.org/0009-0000-5957-5923
Anik Djuraidah Study Program of Statistics and Data Science, Bogor Agricultural University, Indonesia
Yenni Angraini Study Program of Statistics and Data Science, Bogor Agricultural University, Indonesia https://orcid.org/0000-0003-3186-2378

DOI:

https://doi.org/10.59395/ijadis.v7i2.1602

Keywords:

Random Forest, NO2 Concentration, Satellite Remote Sensing, Spatio-temporal Features, ASEAN

Abstract

Nitrogen dioxide (NO₂) is a critical indicator of anthropogenic emissions, making its monitoring essential for environmental management in the rapidly urbanizing ASEAN region. While satellite imagery provides the necessary high spatial coverage to overcome the limitations of sparse ground-based stations, traditional machine learning models applied to these datasets often overlook the inherent spatial heterogeneity and temporal persistence of air pollutants. This study aims to quantify the marginal contributions of geographic coordinates and temporal lag features to the prediction accuracy of satellite-derived NO₂ concentrations using Random Forest (RF) models. By evaluating four progressive RF configurations across 1,500 locations from July 2018 to December 2024, the study identifies the optimal feature combination for regional air quality modeling. The results demonstrated that the RF model incorporating both geographic coordinates and a 12-month lag variable achieved the best performance, yielding an of 0.832 and an RMSE of 5.42 . Feature importance analysis revealed that the 12-month lag of NO₂, nighttime lights, and location were the most influential predictors, highlighting the strong annual seasonality and the impact of economic activities on pollution levels. These findings provide a robust, data-driven framework for regional air quality monitoring and policy formulation in developing tropical regions.

337 166

Downloads

Download data is not yet available.

References

[1] M. Li et al., MIX: a mosaic Asian anthropogenic emission inventory under the international collaboration framework of the MICS-Asia and HTAP, Atmos. Chem. Phys., vol. 17, no. 2, pp. 935963, Jan. 2017, doi: 10.5194/acp-17-935-2017. DOI: https://doi.org/10.5194/acp-17-935-2017

[2] K. D. Kanniah, N. A. F. Kamarul Zaman, D. G. Kaskaoutis, and M. T. Latif, COVID-19s impact on the atmospheric environment in the Southeast Asia region, Science of The Total Environment, vol. 736, p. 139658, Sep. 2020, doi: 10.1016/j.scitotenv.2020.139658. DOI: https://doi.org/10.1016/j.scitotenv.2020.139658

[3] K. Vohra et al., Rapid rise in premature mortality due to anthropogenic air pollution in fast-growing tropical cities from 2005 to 2018, Sci. Adv., vol. 8, no. 14, Apr. 2022, doi: 10.1126/sciadv.abm4435. DOI: https://doi.org/10.1126/sciadv.abm4435

[4] M. Rendana, M. H. Dahlan, F. Alia, D. Bahrin, R. Pebrianto, and P. K. Wardhani, Impact of vegetation cover and land surface temperature on the seasonal tropospheric NO2 level variation from satellite observation, Advances in Space Research, vol. 74, no. 10, pp. 47604772, Nov. 2024, doi: 10.1016/j.asr.2024.07.073. DOI: https://doi.org/10.1016/j.asr.2024.07.073

[5] P. Fan et al., Urbanization, economic development, and environmental changes in transitional economies in the global south: a case of Yangon, Ecol. Process., vol. 11, no. 1, p. 65, Nov. 2022, doi: 10.1186/s13717-022-00409-6. DOI: https://doi.org/10.1186/s13717-022-00409-6

[6] T. Holloway et al., Satellite Monitoring for Air Quality and Health, Annu. Rev. Biomed. Data Sci., vol. 4, no. 1, pp. 417447, Jul. 2021, doi: 10.1146/annurev-biodatasci-110920-093120. DOI: https://doi.org/10.1146/annurev-biodatasci-110920-093120

[7] F. Tack et al., Assessment of the TROPOMI tropospheric NO 2 product based on airborne APEX observations, Jun. 29, 2020. doi: 10.5194/amt-2020-148. DOI: https://doi.org/10.5194/egusphere-egu2020-19261

[8] Z. A. Hanami et al., SpatialTemporal Changes in Air Pollutants in Four Provinces of Sumatra Island, Indonesia: Insights from Sentinel-5P Satellite Imagery, Urban Science, vol. 9, no. 2, p. 42, Feb. 2025, doi: 10.3390/urbansci9020042. DOI: https://doi.org/10.3390/urbansci9020042

[9] D. Stratoulias, N. Nuthammachot, R. Dejchanchaiwong, P. Tekasakul, and G. R. Carmichael, Recent Developments in Satellite Remote Sensing for Air Pollution Surveillance in Support of Sustainable Development Goals, Remote Sens. (Basel)., vol. 16, no. 16, p. 2932, Aug. 2024, doi: 10.3390/rs16162932. DOI: https://doi.org/10.3390/rs16162932

[10] M. Belgiu and L. Drgu, Random forest in remote sensing: A review of applications and future directions, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 114, pp. 2431, Apr. 2016, doi: 10.1016/j.isprsjprs.2016.01.011. DOI: https://doi.org/10.1016/j.isprsjprs.2016.01.011

[11] T. Hengl, M. Nussbaum, M. N. Wright, G. B. M. Heuvelink, and B. Grler, Random forest as a generic framework for predictive modeling of spatial and spatio-temporal variables, PeerJ, vol. 6, p. e5518, Aug. 2018, doi: 10.7717/peerj.5518. DOI: https://doi.org/10.7717/peerj.5518

[12] A. N. Pradana, A. Djuraidah, and A. M. Soleh, Land Use Change Modelling Using Logistic Regression, Random Forest and Additive Logistic Regression in Kubu Raya Regency, West Kalimantan, Forum Geografi, vol. 37, no. 2, Dec. 2023, doi: 10.23917/forgeo.v37i2.23270. DOI: https://doi.org/10.23917/forgeo.v37i2.23270

[13] V. van Zoest, F. B. Osei, G. Hoek, and A. Stein, Spatio-temporal regression kriging for modelling urban NO 2 concentrations, International Journal of Geographical Information Science, vol. 34, no. 5, pp. 851865, May 2020, doi: 10.1080/13658816.2019.1667501. DOI: https://doi.org/10.1080/13658816.2019.1667501

[14] A. Srivastava, A. K. Thakur, A. Rai, R. N. Yadava, R. D. Garg, and P. K. Garg, Temporal and Spatial Variations of Nitrogen Dioxide Concentrations in Kolkata, India, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. X-G-2025, pp. 855862, Jul. 2025, doi: 10.5194/isprs-annals-X-G-2025-855-2025. DOI: https://doi.org/10.5194/isprs-annals-X-G-2025-855-2025

[15] A. Zhalehdoost and M. Taleai, Unravelling the importance of spatial and temporal resolutions in modeling urban air pollution using a machine learning approach, Sci. Rep., vol. 15, no. 1, p. 27708, Jul. 2025, doi: 10.1038/s41598-025-13639-3. DOI: https://doi.org/10.1038/s41598-025-13639-3

[16] J. Fan, P. Zhang, J. Chen, B. Li, L. Han, and Y. Zhou, Quantitative Estimation of Missing Value Interpolation Methods for Suomi-NPP VIIRS/DNB Nighttime Light Monthly Composite Images, IEEE Access, vol. 8, pp. 199266199288, 2020, doi: 10.1109/ACCESS.2020.3035408. DOI: https://doi.org/10.1109/ACCESS.2020.3035408

[17] S. Cheng, P. Peng, and F. Lu, A lightweight ensemble spatiotemporal interpolation model for geospatial data, International Journal of Geographical Information Science, vol. 34, no. 9, pp. 18491872, Sep. 2020, doi: 10.1080/13658816.2020.1725016. DOI: https://doi.org/10.1080/13658816.2020.1725016

[18] A. S. Fotheringham, C. Brunsdon, and M. Charlton, Geographically weighted regression: The analysis of spatially varying relationships. Wiley, 2002.

[19] N. S. Maharani et al., APLIKASI MODEL ARIMA GARCH DALAM PERAMALAN DATA NILAI TUKAR RUPIAH TERHADAP DOLAR TAHUN 2017-2022, Jurnal Matematika Sains dan Teknologi, vol. 24, no. 1, pp. 3750, May 2023, doi: 10.33830/jmst.v24i1.4875.2023. DOI: https://doi.org/10.33830/jmst.v24i1.4875.2023

[20] L. Breiman, Random Forests, Mach. Learn., vol. 45, no. 1, pp. 532, Oct. 2001, doi: 10.1023/A:1010933404324. DOI: https://doi.org/10.1023/A:1010933404324

[21] A. Vamsikrishna and E. V. Gijo, New Techniques to Perform Cross-Validation for Time Series Models, Operations Research Forum, vol. 5, no. 2, p. 51, Jun. 2024, doi: 10.1007/s43069-024-00334-8. DOI: https://doi.org/10.1007/s43069-024-00334-8

[22] T. O. Hodson, Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not, Geosci. Model Dev., vol. 15, no. 14, pp. 54815487, Jul. 2022, doi: 10.5194/gmd-15-5481-2022. DOI: https://doi.org/10.5194/gmd-15-5481-2022

[23] D. C. Montgomery, E. A. Peck, and G. G. Vining, Introduction to Linear Regression Analysis Fifth Edition, Fifth. New Jersey: John Wiley & Sons, Inc., 2012.

[24] A. de Myttenaere, B. Golden, B. Le Grand, and F. Rossi, Mean Absolute Percentage Error for regression models, Neurocomputing, vol. 192, pp. 3848, Jun. 2016, doi: 10.1016/j.neucom.2015.12.114. DOI: https://doi.org/10.1016/j.neucom.2015.12.114

[25] C. Tofallis, A better measure of relative prediction accuracy for model selection and model estimation, Journal of the Operational Research Society, vol. 66, no. 8, pp. 13521362, Aug. 2015, doi: 10.1057/jors.2014.103. DOI: https://doi.org/10.1057/jors.2014.103

[26] S. Georganos et al., Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling, Geocarto Int., vol. 36, no. 2, pp. 121136, Jan. 2021, doi: 10.1080/10106049.2019.1595177. DOI: https://doi.org/10.1080/10106049.2019.1595177

[27] M. A. Suprayogi, B. Sartono, and K. A. Notodiputro, GEOGRAPHICALLY WEIGHTED MACHINE LEARNING MODEL FOR ADDRESSING SPATIAL HETEROGENEITY OF PUBLIC HEALTH DEVELOPMENT INDEX IN JAVA ISLAND, BAREKENG: Jurnal Ilmu Matematika dan Terapan, vol. 18, no. 4, pp. 25772588, Oct. 2024, doi: 10.30598/barekengvol18iss4pp2577-2588. DOI: https://doi.org/10.30598/barekengvol18iss4pp2577-2588