Spatiotemporal Variations of Smoke Haze in a Mountain–Valley Region of Northern Thailand Based on Microsensors and Day–Night Remote Sensing

Biomass-burning is a significant source of particulate matter, including PM 2.5 , in Southeast Asia, with a peak occurring each spring. This study aimed to understand how smoke from this burning disperses across the varied terrains of Chiang Mai province, Thailand, during the spring of 2019. To achieve this, the Aerobox, a microsensor system, was deployed at three distinct locations: a hilltop site at DAK, the Fang valley with a medium-sized town, and the Chiang Mai valley, which contains a large city. The study evaluated three correction methods for the Aerobox data: linear regression, multiple linear correlation, and the random forest machine learning algorithm. The random forest algorithm produced the best results, improving the correlation coefficient with standard instruments from 0.92 to 0.97. The daily average PM 2.5 at all three sites followed a similar pattern, correlating with the daily count of fires. The diurnal variation of ambient PM 2.5 is influenced by both the terrain and the development of the planetary boundary layer (PBL). Concentrations typically rise in the morning until 9:00 or 10:00, then decrease as the PBL height increases throughout the day. When the PBL begins to lower in the evening, between 17:00 and 18:00, concentrations recover overnight. The precise timing and scale of this variation differ by location. A comparison of daily averaged AERONET and satellite-measured Aerosol Optical Depth (AOD) showed a moderate to high correlation, with coefficients of 0.62, 0.86, and 0.71 for Chiang Mai, Fang, and DAK, respectively. The daily average surface PM 2.5 had a moderate to high correlation (0.45 to 0.76) with AERONET-derived AOD but only a low-to-moderate correlation (0.27 to 0.54) with satellite-retrieved AOD. The correlation between hourly averaged PM 2.5 and AERONET-derived AOD was lower (0.07 to 0.56) and influenced by the time of day. However, when the data was segmented by time of day, the correlation improved (0.19 to 0.69), highlighting the strong diurnal variation. This suggests that a single daily measurement from a satellite-borne instrument may not accurately represent the full-day aerosol loading due to the strong diurnal variation of both surface PM 2.5 and columnar aerosol loading. Therefore, a distributed network of microsensors can provide more reliable local air quality data, especially in complex terrains. Observations at three sites reveal the distribution and transport of biomass-burning pollutants over Chiang Mai’s complex terrain.

Authors

Institutions

Publication Details

Journal
Aerosol and Air Quality Research
Published
2026-11-20
DOI
https://doi.org/10.1007/s44408-026-00181-8
Primary Topic
Atmospheric chemistry and aerosols
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Spatiotemporal Variations of Smoke Haze in a Mountain–Valley Region of Northern Thailand Based on Microsensors and Day–Night Remote Sensing

Duangduean Thepnuan, Wan Wiriya, Shantanu Kumar Pani, Chiao-Ling Pan et al.
3 links
Aerosol and Air Quality Research
Atmospheric chemistry and aerosols
article

Spatiotemporal Variations of Smoke Haze in a Mountain–Valley Region of Northern Thailand Based on Microsensors and Day–Night Remote Sensing

Duangduean Thepnuan, Wan Wiriya, Shantanu Kumar Pani, Chiao-Ling Pan, Somporn Chantara, Sheng‐Hsiang Wang, Sumaman Buntoung, BRENT N. HOLBEN, Amit Bhujel, Kun-Hsu Wu, Pattarapanitchai Somjet
article en

Abstract

Biomass-burning is a significant source of particulate matter, including PM 2.5 , in Southeast Asia, with a peak occurring each spring. This study aimed to understand how smoke from this burning disperses across the varied terrains of Chiang Mai province, Thailand, during the spring of 2019. To achieve this, the Aerobox, a microsensor system, was deployed at three distinct locations: a hilltop site at DAK, the Fang valley with a medium-sized town, and the Chiang Mai valley, which contains a large city. The study evaluated three correction methods for the Aerobox data: linear regression, multiple linear correlation, and the random forest machine learning algorithm. The random forest algorithm produced the best results, improving the correlation coefficient with standard instruments from 0.92 to 0.97. The daily average PM 2.5 at all three sites followed a similar pattern, correlating with the daily count of fires. The diurnal variation of ambient PM 2.5 is influenced by both the terrain and the development of the planetary boundary layer (PBL). Concentrations typically rise in the morning until 9:00 or 10:00, then decrease as the PBL height increases throughout the day. When the PBL begins to lower in the evening, between 17:00 and 18:00, concentrations recover overnight. The precise timing and scale of this variation differ by location. A comparison of daily averaged AERONET and satellite-measured Aerosol Optical Depth (AOD) showed a moderate to high correlation, with coefficients of 0.62, 0.86, and 0.71 for Chiang Mai, Fang, and DAK, respectively. The daily average surface PM 2.5 had a moderate to high correlation (0.45 to 0.76) with AERONET-derived AOD but only a low-to-moderate correlation (0.27 to 0.54) with satellite-retrieved AOD. The correlation between hourly averaged PM 2.5 and AERONET-derived AOD was lower (0.07 to 0.56) and influenced by the time of day. However, when the data was segmented by time of day, the correlation improved (0.19 to 0.69), highlighting the strong diurnal variation. This suggests that a single daily measurement from a satellite-borne instrument may not accurately represent the full-day aerosol loading due to the strong diurnal variation of both surface PM 2.5 and columnar aerosol loading. Therefore, a distributed network of microsensors can provide more reliable local air quality data, especially in complex terrains. Observations at three sites reveal the distribution and transport of biomass-burning pollutants over Chiang Mai’s complex terrain.

Aerosol and Air Quality Research
Goddard Space Flight Center (US), National Central University (TW), Chiang Mai University (TH), Silpakorn University (TH), Chiang Mai Rajabhat University (TH)
Climate action
Atmospheric chemistry and aerosols
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon