Open AccessArticle Remote Sensing of El Niño–Southern Oscillation Impact on Methane Flux Potential from Rice Cultivation in Thailand by Warisara Tundam Warisara Tundam SciProfiles Scilit Preprints.org Google Scholar 1, Parkin Maskulrath Parkin Maskulrath SciProfiles Scilit Preprints.org Google Scholar 1,*, Kittichai Duangmal Kittichai Duangmal SciProfiles Scilit Preprints.org Google Scholar 1, Satreethai Poommai Satreethai Poommai SciProfiles Scilit Preprints.org Google Scholar 1, Onanong Phewnil Onanong Phewnil SciProfiles Scilit Preprints.org Google Scholar 1, Yibo Liu Yibo Liu SciProfiles Scilit Preprints.org Google Scholar Yibo Liu has been serving as an Associate Professor and Master's Supervisor in the Department of at [...] Read more 2, Siqing Zhang Siqing Zhang SciProfiles Scilit Preprints.org Google Scholar 3, Wladyslaw Witold Szymanski Wladyslaw Witold Szymanski SciProfiles Scilit Preprints.org Google Scholar 1,4, Piyanuch Jaikaew Piyanuch Jaikaew SciProfiles Scilit Preprints.org Google Scholar 5, Tasuku Kato Tasuku Kato SciProfiles Scilit Preprints.org Google Scholar 6 Juntariga Boonphue Juntariga Boonphue SciProfiles Scilit Preprints.org Google Scholar 1 1 Department of Environmental Science, Faculty of Environment, Kasetsart University, Bangkok 10900, Thailand 2 Yale-NUIST Center on Atmospheric Environment, State Key Laboratory of Climate System Prediction and Risk Management, Nanjing University of Information Science and Technology, Nanjing 211544, China 3 Key Laboratory of Ecosystem Carbon Souse and Sink, China Meteorological Disasters (CIC-FEMD), Nanjing University of Information Science and Technology, Nanjing 211544, China 4 Aerosol and Environmental Physics Division, Faculty of Physics, University of Vienna, 1090 Vienna, Austria 5 Department of Civil and Environmental Engineering, Faculty of Engineering, Srinakharinwirot University, Nakon Nayok 26120, Thailand 6 United Graduate School of Agricultural Science, Tokyo University of Agriculture and Technology, Tokyo 183-8538, Japan * Author to whom correspondence should be addressed. Environments 2026, 13(6), 320; https://doi.org/10.3390/environments13060320 (registering DOI) Submission received: 7 April 2026 / Revised: 26 May 2026 / Accepted: 3 June 2026 / Published: 7 June 2026 Rice cultivation commonly employs the continuous flooding (CF) method, which depends heavily on water availability creating anaerobic conditions for methane (CH 4) emissions. Rainfed rice areas rely on precipitation for irrigation, making the system sensitive to climatic variability. This study examines associations between ENSO phases and satellite-observed atmospheric XCH 4 variability over Thailand using GOSAT as the primary long-term dataset from 2012 to 2022, with Sentinel-5P/TROPOMI used as a supporting dataset for recent spatial patterns. The analysis conducted covers three cropping seasons: (1) January–April, (2) May–August, and (3) September–December. The results indicate comparable average atmospheric methane concentrations of 1787.94 ± 11.50 XCH 4 (ppb) during El Niño, 1788.8 ± 11.22 XCH 4 (ppb) in neutral conditions, and 1793.45 ± 10.93 XCH 4 (ppb) during La Niña. The obtained data indicate a seasonal variability, with the highest satellite-observed XCH 4 values found during September–December, corresponding to the main growing period of wet-season rice. The results suggest that climate change amplifies these anomalies through altered precipitation patterns and water availability. Current rice cultivation practices warrant reconsideration, in particular the alternate wetting and drying (AWD) method, offering reduced CH 4 emissions while conserving water resources. This underscores the importance of water management strategies for sustainable rice production and resilience to climate variability. 1. Introduction Methane (CH 4) is one of the most significant greenhouse gases contributing to global warming, with a heat-capturing potential approximately 25–28 times greater than that of carbon dioxide (CO 2) over a 100-year period [ 1, 2]. A major source of CH 4 emissions is anthropogenic activity, especially rice cultivation, which cannot be overlooked given that rice serves as a basic food for a large and growing global population [ 3]. Variations in rainfall patterns directly influence rice cultivation practices, affecting both crop yield and growing conditions. In particular, water levels in rice paddies create anaerobic soil environments that play a crucial role in CH 4 production and its subsequent emission into the atmosphere [ 8]. Previous studies have demonstrated that ENSO variability significantly affects agricultural productivity altering water availability [ 9, 10]. At the same time, research has consistently shown that water management in rice fields is a key factor governing CH 4 production processes [ 11, 12]. Furthermore, findings by the authors of [ 13, 14] reveal that the increase in atmospheric CH 4 concentrations between 2014 and 2017 was positively correlated with temperature and, in particular, precipitation. During this period, global CH 4 concentrations increased above baseline levels by 12.7 ± 0.5 ppb in 2014, 10.1 ± 0.7 ppb in 2015, 7.0 ± 0.7 ppb in 2016, and 7.7 ± 0.7 ppb in 2017. Despite these advancements, the relationship between water levels in rice paddies and CH 4 emissions has not been fully understood, particularly at broader spatial scales. Existing studies examining the link between ENSO variability and CH 4 production in tropical regions have largely relied on ground-based measurements, such as plot-scale chamber experiments [ 15], which may not adequately capture regional variability. In recent years, remote sensing technologies have demonstrated a strong potential for monitoring greenhouse gas concentrations at a higher spatial resolution, enabling analyses at regional, national, and global scales [ 16]. This study utilizes satellite-derived XCH 4 variability data from GOSAT and Sentinel-5P to analyze national-scale CH 4 variability in Thailand across different ENSO phases from 2012 to 2022, separating temporal scales into rice-growing and non-growing seasons. While the XCH 4 represents a column-averaged atmospheric quantity, the analysis is interpreted as atmospheric CH 4 variability not rigorously linked to rice paddy CH 4 emissions. The study also addresses the applicability of satellite-observed column-averaged XCH 4 over major rice-growing regions of Thailand varying during ENSO phases, accounting for the long-term atmospheric CH 4 trend. Consequently, it can be postulated that the La Niña-associated periods show higher satellite-observed XCH 4 over rice-growing regions than during the El Niño periods as the wetter conditions and higher water availability provide greater rice cultivation activities and increasing flooding duration for traditional growing practices, thus enhancing anaerobic conditions in rice cultivation areas. In addition, the introduction of the alternate wetting and drying (AWD) rice cultivation technique, which involves periodic transitions between wet and dry soil conditions, has been shown to reduce the duration of waterlogging and limit soil anaerobicity [ 17]. This method can reduce CH 4 emissions by approximately 35% compared to continuous flooding, without compromising rice yields [ 2, 18]. 2. Materials and Data 2.1. Study Areas This study utilizes remotely sensed data covering Thailand (5°37′–20°27′ N and 97°22′–105°37′ E), encompassing a total area of approximately 513,120 km 2 ( Figure 1) within the tropical climate zone. According to [ 19], Thailand’s agricultural sector emitted 58,486.02 Gg CO 2 equivalent in 2017, with methane identified as the dominant greenhouse gas contributor. This finding is further supported by data from [ 20]. 2.2. El Niño and La Niña Events in Thailand The ENSO variability over the study period was characterized by applying the Oceanic Niño Index (ONI) for the Niño 3.4 region (5°N–5°S, 170°W–120°W), calculated from the 3-month moving average of persistent sea surface temperature (SST) anomalies over at least five overlapping seasons. In the selection of the ENSO classification, a positive ONI value indicates El Niño conditions, while a negative value indicates La Niña conditions. Under the standard classification criteria, El Niño is defined as ONI ≥ +0.5 °C, La Niña as ONI below −0.5 °C, while values between −0.4 °C and +0.4 °C are classified as neutral [ 21, 22]. 2.3. Satellite Data This study utilizes remotely sensed observations for the atmospheric CH 4 column concentrations derived from two satellite platforms: (1) the Greenhouse Gases Observing Satellite (GOSAT) using the Fourier Transform Spectrometer (FTS) Mitsubishi Electric Corporation (MELCO) in Tokyo, Japan., Level 2 CH 4 column amount (SWIR) with a spatial resolution of approximately 10.5 km diameter resolution; and (2) the Sentinel-5 Precursor (Sentinel-5P) equipped with the TROPOspheric Monitoring Instrument (TROPOMI) Institute for Space Research (SRON), Netherlands providing data with a spatial resolution of 5.5/7 × 7 km 2. Individual satellite pixels were first aggregated into monthly or seasonal regional means for each of the rice-growing areas. These regional monthly and seasonal means were used as the statistical sampling units to reduce pseudo-replication caused by spatial autocorrelation. Differences among the ENSO phases and rice-growing periods were evaluated using pairwise comparisons with bootstrap confidence intervals because of the limited number of ENSO years. To ensure data quality and geographically relevant data for Thailand ( Figure 1), only pixels with validity > 50 were used. The analysis employed off-line (OFFL) Level 2 Product (L2) and applied the Discrete Cosine Transform (DCT) method following the approach in [ 26]. Data from GOSAT and Sentinel-5P were combined to enhance the spatial and temporal continuity of CH 4 data, and subsequently resampled on a 0.25° × 0.25° resolution grid covering Thailand [ 27, 28, 29]. It is important to note that satellite-derived CH 4 concentrations (XCH 4) represent column-averaged dry-air mole fractions, which may differ from ground-based measurements. Ground observations typically report higher concentrations due to the absence of vertical atmospheric mixing effects. Satellite measurements, by contrast, integrate CH 4 concentrations throughout the atmospheric column, resulting in comparatively lower values, with typical differences of ±15–20 ppb depending on location and latitude. Nevertheless, satellite-derived XCH 4 shows a strong agreement with ground-based observations, with correlation coefficients ranging from 0.66 to 0.80 [ 30, 31]. Atmospheric CH 4 was analyzed using satellite-derived XCH 4, defined as the column-averaged on the dry-air mole fraction of CH 4. The GOSAT data was used as the primary long-term satellite dataset for the 2012–2022 analysis as it provides observations throughout the full study period. Sentinel-5P/TROPOMI was used only from late 2017/2018 onward as a supporting high-resolution dataset as the overlapping periods were used as a supporting dataset to examine recent high-resolution spatial patterns. Therefore, the 2012–2017 period was interpreted from GOSAT. The data from GOSAT and Sentinel-5P/TROPOMI were processed ( Figure 2) then aggregated to common seasonal time scales ( Table 2). After quality filtering following the HYSPLIT analysis, the valid XCH 4 retrievals were resampled to a common 0.25° × 0.25° grid. The seasonal means were then calculated from valid observations within each grid cell, and seasonal averages were then calculated according to the rice-growing season. 2.4. Mapping Satellite-Observed XCH 4 Variability Atmospheric CH 4 observations were processed and analyzed using Jupyter Notebook. The methodological workflow for generating CH 4 emission maps across Thailand is presented in Figure 2. Satellite-derived CH 4 datasets were categorized according to ENSO phases in the decade from 2012 to 2022. To account for long-term increasing trends in atmospheric CH 4 concentrations, a linear detrending approach was applied [ 32, 33]. Seasonal analysis was structured based on Thailand’s rice cultivation calendar modified from [ 34, 35]and divided into three periods: Period 1 (January to April), Period 2 (May to August), and Period 3 (September to December) ( Table 2). Period 3 accounts for approximately 82% of the total national rice production, corresponding to the main growing season with planting occurring from May to June, crop growth from July to October, and harvesting from November to December. Periods 1 and 2 represent the dry-season rice production (accounting for 18% of total production), with planting from January to February, harvesting from March to June [ 34, 36]. As the atmospheric CH 4 has a multi-year lifetime and continues to show a strong long-term global increasing trend, raw XCH 4 values include both background atmospheric growth and shorter-term regional variability. Therefore, this study applied a detrending and anomaly-based framework to reduce the influence of long-term atmospheric CH 4 accumulation before comparing ENSO phases [ 2]. Average atmospheric column methane concentrations (XCH 4) reflect both long-term CH 4 accumulation and short-term interannual or seasonal variability [ 37, 38]. Therefore, this study applied the following processing steps: (1) monthly grid cell XCH 4 values were calculated, (2) the annual or linear trend was removed, (3) monthly XCH 4 anomalies were calculated, and (4) the anomalies were aggregated by rice-growing season and ENSO phase. In the study of the annual mean XCH 4 values from 2012 to 2022 were used to estimate the long-term CH 4 growth trend using linear regression. XCH 4, trend = ax + b (1) where XCH 4, trend is the estimated long-term methane trend; a is the slope of the regression line; b is the intercept; x is the study year. The resulting annual trend was then subtracted from each XCH 4 satellite observation for each pixel for each study period, with the monthly anomalies recalculated relative to the long-term monthly mean to reduce seasonal bias. XCH 4, detrended, pixel = XCH 4, pixel − Trend year (2) where XCH 4, detrended,pixel is the detrended methane concentration for each pixel; XCH 4, pixel is the original satellite-observed methane concentration; Trend year is the annual methane trend correction derived from the regional annual mean XCH 4 time series. This trend correction process reduces the influence of long-term atmospheric methane accumulation while preserving the spatial variability and intergenerational fluctuations associated with regional CH 4 dynamics and ENSO-related climate variability. The satellite-derived XCH 4, defined as the column-averaged dry-air mole fraction of CH 4, was applied. GOSAT was used as the primary long-term satellite dataset for the 2012–2022 analysis because it provides observations throughout the study period. Sentinel-5P/TROPOMI was used only for the overlapping period after its availability and was applied as a supporting dataset to examine recent high-resolution spatial patterns. Therefore, the 2012–2017 period was interpreted from GOSAT only, while the later period was analyzed using GOSAT with an additional comparison against Sentinel-5P/TROPOMI. The two satellite datasets were not directly merged as an uncorrected single time series. Instead, GOSAT and Sentinel-5P/TROPOMI were processed separately using their respective quality-screening procedures, and then aggregated to common monthly and seasonal time scales. After quality filtering, valid XCH 4 retrievals were resampled to a common 0.25° × 0.25° grid to support spatial comparison. Monthly means were calculated from valid observations within each grid cell, and seasonal averages were then calculated according to the rice-growing and monsoon periods used in this study. During the overlapping period, GOSAT and Sentinel-5P/TROPOMI were compared to assess differences in spatial pattern and XCH 4 magnitude. The GOSAT record was used as the main dataset for long-term ENSO-related temporal analysis, while Sentinel-5P/TROPOMI was used to support the interpretation of spatial variability during the recent period. This approach avoids treating Sentinel-5P/TROPOMI as if it were available for the full 2012–2022 period and reduces the possibility that apparent temporal differences are caused by changes in satellite sampling or sensor composition [ 39]. To improve ENSO-related interpretation, the statistical analysis was conducted using monthly and seasonal XCH 4 variability. Monthly XCH 4 anomalies were calculated after removing the long-term linear trend and subtracting the long-term monthly climatological mean. This approach reduced the influence of both long-term global atmospheric CH 4 growth and the seasonal cycle. ENSO classification was based on the Oceanic Niño Index (ONI) at the monthly scale ( Table 1). Because ENSO conditions can change within a year, particularly during transition periods, the XCH 4 variability were grouped according to El Niño, neutral, and La Niña conditions. Differences in XCH 4 anomalies among ENSO phases and rice-growing periods were evaluated using bootstrap confidence intervals based on the limited number of ENSO events. In addition, correlation analysis was conducted between ONI and detrended XCH 4 anomalies. Lagged correlations using 1- to 3-month lags were also tested to account for delayed hydrological and agricultural responses to ENSO-related rainfall variability and data. 2.5. XCH 4 Variability Under Air Mass Transport (HYSPLIT Model) The backward trajectories were used to characterize recent air mass transport pathways reaching the selected rice-growing areas in support of the interpretation of spatial and temporal variability in satellite-observed XCH 4. However, the trajectory analysis was not used as a definitive source attribution because XCH 4 represents the column-integrated CH 4 concentration that can reflect contributions from multiple CH 4 sources, including rice paddies, wetlands, livestock, waste, biomass burning, fossil fuel activities, and regional background CH 4 [ 40]. Therefore, the HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) model results are interpreted only as a supporting transport context for understanding possible air mass movement patterns associated with satellite-observed XCH 4 variability. In this study, 48 h air mass backward trajectories were screened and verified with the HYSPLIT model with GDAS 1-degree global meteorological data available from 2006 to the present [ 41, 42]. This also accounted for the additional land use/land cover data from the ESRI Living Atlas Land Cover Explorer(Esri, Redlands, CA, USA using the Sentinel-2 10 m Annual Global Land Cover dataset) to examine the transport pathways of air masses in Thailand during the study period. Trajectories were initialized at three elevations: 100, 300, and 500 m above ground level, representing the lower atmospheric layers where CH 4 accumulation and transport may influence near-surface and column-averaged CH 4 observations [ 43, 44]. The HYSPLIT model was used to screen and interpret possible air mass transport influence, but it was not used to definitively remove all non-rice CH 4 sources. In the selection data filtering, the analysis focused on five representative rice-growing regions in Thailand, selected to represent important rice cultivation areas ( Figure 3): Nakhon Sawan (15°39′10.1412″ N, 100°31′39.9″ E), Suphan Buri (14°25′35.058″ N, 100°03′10.4292″ E), Phetchaburi (13°02′44.4768″ N, 100°01′19.8336″ E), Ubon Ratchathani (14°52′26.4324″ N, 105°10′19.6248″ E), and Phitsanulok (17°05′36.0528″ N, 100°47′56.9508″ E). The backward trajectories were used to characterize recent air mass transport pathways reaching the selected rice-growing areas in support of the interpretation of spatial and temporal variability in satellite-observed XCH 4. However, the trajectory analysis was not used as a definitive source attribution because XCH 4 represents a column-integrated CH 4 concentration that can reflect contributions from multiple CH 4 sources, including rice paddies, wetlands, livestock, waste, biomass burning, fossil fuel activities, and regional background CH 4 [ 40]. Therefore, the HYSPLIT results are interpreted only as a supporting transport context for understanding possible air mass movement patterns associated with satellite-observed XCH 4 variability. 3. Results 3.1. Seasonal Climate Impacts on Rice Cultivation in Thailand The relationship between annual rainfall, rice cultivation areas, and rice yield in Thailand reveals that the El Niño years (2015, 2016, and 2019) experienced higher temperatures (27.9–28.1 °C), together with the lowest rainfall (1344–1718 mm), and showed a significant decline in rice cultivation area (<10,000 kha in 2015 and 2019). This reduction corresponded with lower average yields (2.53–2.70 t/ha), with 2015—classified as a strong El Niño year—recording the lowest yield (2.53 t/ha). In contrast, La Niña years (2020, 2021, and 2022) experienced significantly higher rainfall (1529–2012 mm) and slightly lower temperatures (27.4–28.0 °C), In 2022, both rainfall and rice yield reached their highest values, with yields of 2.86 t/ha. Neutral years (2012, 2013, and 2017) exhibited intermediate conditions, with rainfall ranging from 1682 to 2017 mm, temperatures between 27.4 and 27.7 °C, cultivation areas of 10,756–11,100 kha, and relatively high yields (2.83–2.90 t/ha). Correlation analysis further supports these findings. The relationship between the ONI, rainfall, and rice cultivation in Thailand revealed a significant negative correlation (r = −0.70) indicating reduced rainfall during El Niño events. Similarly, rice cultivation area showed a significant negative correlation (r = −0.72) with ONI. These results were statistically significant ( p < 0.05), indicating that the reduced water availability during El Niño leads to a decrease in rice cultivation areas. Water availability plays a critical role in determining the rice cultivation extent in Thai agriculture. Water management is also a key factor influencing CH 4 variability. Continuous flooding (CF), which remains the most commonly practiced irrigation method among Thai farmers, creates prolonged anaerobic soil conditions that promote methanogenesis and enhance CH 4 production [ 10, 45]. Considering the combined effects of current rice cultivation practices, it is evident that, although the main growing season in Thailand begins in the latter part of Period 2, peak CH 4 emissions occur during Period 3. This pattern aligns with key rice growth stages, particularly the tillering stage (approximately 40–60 days after planting) and the heading stage (approximately 70–90 days after planting). During these phases, increased physiological activity, along with the development of stem and root systems, enhances the capacity for gas transport and facilitates CH 4 emission into the atmosphere [ 45, 47]. 3.2. Variability in Atmospheric CH 4 in Thailand CH 4 has an atmospheric lifetime of approximately 9–12 years [ 2], meaning that observed concentrations reflect both past and present emissions. As illustrated in Figure 4, the measured annual column-averaged methane (XCH 4, ppb) consists of two main components: (1) CH 4 accumulated in the atmosphere from emissions in previous years, and (2) CH 4 newly emitted within the current year. Consequently, observed XCH 4 values do not directly represent year-to-year changes in CH 4 emissions. To address this issue, a linear detrending approach was applied to isolate the long-term accumulation data from annual emission variability ( Figure 5) [ 44]. This method enables a clearer assessment of CH 4 fluctuations associated with ENSO phases. The results indicate that the average CH 4 concentration during El Niño years was 1787.94 ± 11.50 ppb, compared to 1793.45 ± 10.93 ppb during La Niña years and 1788.80 ± 11.22 ppb during neutral years. Seasonal variations in CH 4 concentrations across the three periods of the year are presented in Figure 6. The overall (2012–2022) average CH 4 concentration during Period 1 (January–April) was 1791.19 ± 0.75 XCH 4 (ppb), while Period 2 (May–August) had the lowest average at 1764.07 ± 1.64 XCH 4 (ppb), and Period 3 (September–December) had the highest average at 1815.01 ± 1.69 XCH 4 (ppb). The differences are statistically significant ( p < 0.05). These changes in the CH 4 variability can be taken into overall seasonal trend but it should not be interpreted as direct evidence of rice paddy CH 4 emissions. The results indicate that satellite-observed XCH 4 varied seasonally and showed highest values during the main wet-season rice-growing period. This pattern is consistent with the timing of rice cultivation, especially during the late rainy season and the growing season (Period 3). This period involves prolonged waterlogging and is most conducive to anaerobic conditions within rice fields, where the concentration areas are within rice field land cover. Period 2 shows the lowest CH 4 values, which is consistent with the early rainy season. This can be explained by the rice youth phase, during which CH 4 production has not yet reached its peak. During the seedling stage of rice growth, the plants are still very small; thus, the accumulation of organic substrates within the rice fields is still very limited, resulting in fewer methanogenesis substrates and consequently lower CH 4 release [ 18, 48]. However, during Period 3, there is accumulation in the rice panicle initiation, heading, and maturation stages where, in traditional practices, the fields are flooded promoting strong anaerobic digestion leading to the highest CH 4 production. This result showed an evident correlation between the CH 4 variability, paddy field water dynamics, and rice growth stages. This was consistent with rice calendar timing, paddy water dynamics, and the known methane production mechanisms in flooded rice systems, particularly during the late rainy season and growing season (Period 3), corresponding to the heading–ripening stage of rice. During this period, rice plants have a fully developed root system and aerenchyma, which can efficiently transport CH 4 from the soil to the atmosphere [ 45]. Furthermore, prolonged waterlogging during this time results in continuous anaerobic conditions, promoting the activity of methanogens and methanogenesis. Conversely, Period 2, representing the early rainy season and the seeding–tilling stage of rice, shows the lowest CH 4 release. This is because the rice plants’ root systems and aerenchyma are not yet fully developed, resulting in low CH 4 transport efficiency. Additionally, high water flow and drainage in the paddy field increase soil oxygenation and reduce anaerobic conditions. This results in a decrease in methanogenic microbial activity [ 18, 45, 47]. 4. Discussion Atmospheric CH 4 Variability Across ENSO Phases The satellite-observed XCH 4 showed spatial and seasonal differences across Thailand during the study period. The observed XCH 4 values overlapped with the country’s major rice-growing areas during the La Nina seasonal conditions. However, it has to be mentioned that these patterns should be interpreted as regional atmospheric methane variability rather than direct rice paddy CH 4 emissions. The XCH 4 represents the column-averaged methane and may be affected by background atmospheric CH 4, transport pathways, vertical mixing, retrieval sensitivity, and multiple methane source sectors, including wetlands, livestock, waste, biomass burning, fossil fuel activities, and rice paddies, which can reflect the association of rice phenology on atmospheric CH 4 dynamics [ 49]. Even so, across ENSO phases, XCH 4 values were generally higher during La Niña periods and lower during El Niño periods. This pattern was consistent with ENSO-related rainfall variability and its potential influence on rice cultivation practices which control flooding duration leading to anaerobic soil [ 50]. Nevertheless, in the statistical association between ENSO phase and XCH 4 should not be interpreted as directly. Conversely, the results indicate that ENSO-related hydroclimatic variability may contribute to regional atmospheric methane variability over rice-growing landcover. The applied HYSPLIT model analysis provides additional context on air mass pathways reaching the selected main rice-growing regions in Thailand. However, the trajectories do not quantify source contributions and cannot separate rice paddy CH 4 from other regional sources. Therefore, the HYSPLIT results are used only to support interpretation of possible transport influences, not to establish direct source attribution. During El Niño years, satellite-observed XCH 4 values were lower than the 10-year average at 1787.94 ± 11.50 XCH 4 (ppb), while the highest XCH 4 values were observed during La Niña years at 1793.45 ± 10.93 XCH 4 (ppb). Neutral years recorded 1788.8 ± 11.22 XCH 4 (ppb). As mentioned earlier, these differences should be interpreted as atmospheric methane variability, not direct CH 4 emission changes ( Figure 6). The results are consistent with ENSO-associated differences in atmospheric XCH 4 under varying hydroclimatic conditions; with statistical analysis confirming significant differences in mean values among the three periods ( p < 0.05). However, they do not to establish direct rice paddy source attribution. The study suggests that La Niña years are associated with higher-than-normal rainfall and increased rice paddy flooding. This plays a crucial role in creating an anaerobic environment suitable for CH 4 production and release. During neutral periods, when there are no severe climatic anomalies, water levels in the paddy fields fluctuate less, resulting in XCH 4 values close to the 10-year average. During El Niño periods, rainfall and water resources are reduced, corresponding to lower satellite-observed XCH 4 values [ 1, 51]. Consequently, the ENSO variability may significantly influence rice cultivation dynamics in Thailand through changes in rainfall and water availability [ 52]. The El Niño years experience approximately 17.2% less rainfall and an 8.7% decrease in rice yield, reflecting the impact of drought on Thailand’s rice production system. There is a significant negative correlation between the ONI, rainfall (r = −0.70), and rice-cultivated area (r = −0.72) as the water availability limits agricultural activities. Currently, Thai rice cultivation is dependent on seasonal rainfall and irrigation; prolonged drought during the El Niño periods leads to a reduced cultivated area and yield. Furthermore, the reduced water availability during El Niño may inhibit CH 4 production in rice paddies by decreasing the flooding duration and the anaerobic soil conditions suitable for CH 4 synthesis compared to normal conditions as the atmospheric CH 4 concentration during El Niño decreased slightly by 0.31% compared to the neutral years. Although the decrease is relatively small, the study suggests that climate variability affects not only agricultural yield but also other aspects of rice production ( Figure 6). Therefore, the lower XCH 4 variability during El Niño may be associated with reduced flooding and rice cultivation intensity. Conversely, during the La Niña years, the rainfall increased by 15.6% and rice yields by 6.7%. The increased water volume supports prolonged flooding in rice paddies, resulting in increased arable area and rice growth. These conditions favor CH 4 production, as flooded soil creates an anaerobic environment suitable for methanogenesis [ 12]. Consequently, satellite-observed XCH 4 during La Niña was only slightly higher (~0.05%) compared to El Niño. This study highlights the challenges in interpreting satellite-observed CH 4 variability under ENSO-driven changes in rainfall, cultivated area, and rice productivity. The CH 4 emission and XCH 4 variability in agricultural areas is mostly biogenic, typically concentrated in the lowest part of the atmosphere before being uplifted and transported the local boundary layer [ 53]. CH 4 concentration varies over different periods of the day; with better atmospheric uplift in the daytime, CH 4 rises to higher altitudes and disperses rapidly, while at night, due to atmospheric stability, CH 4 accumulates near the surface [ 53, 54]. In relation to the rice paddies, most CH 4 is transported through the rice stems (plant-mediated transport) and emitted into the atmosphere near the ground at an altitude of approximately 0.5–3 m [ 55]. The measurements using closed chamber and eddy covariance methods confirm that CH 4 exchange is significant in the range of approximately 0–1.8 m above ground level (m AGL) [ 56]. It was seen that distribution at the geographic level is controlled by turbulent mixing processes and near-ground winds. The agricultural areas show that CH 4 can be lifted into the lower atmosphere to an altitude of approximately 50–500 m AGL. In this application of the satellite data, the CH 4 emitted from the surface accumulates in air masses moving in the direction of the wind, resulting in an increasing concentration over distance in the form of plumes with the heights of 100, 300, and 500 m AGL selected to represent lower-boundary-layer transport conditions. These heights do not imply that satellite XCH 4 measures CH 4 at those specific altitudes, because XCH 4 is a column-averaged mole fraction [ 44, 53]. The analysis using the HYSPLIT model revealed that the primary air mass trajectories affecting representative rice-growing areas in Thailand mainly originate from the northeast (NE), reflecting the influence of air masses from southern China and the Indochina region. Phitsanulok province had the highest proportion of northeasterly wind trajectories (58.93%), followed by Suphan Buri (48.21%), Nakhon Sawan (44.64%), Ubon Ratchathani (42.86%), and Phetchaburi (37.50%). Furthermore, southwesterly (SW) air mass trajectories were found in many areas, particularly in Suphan Buri (32.14%) and Phetchaburi (28.57%), reflecting the influence of monsoon circulation and air mass movement over the Gulf of Thailand and the surrounding coastal areas. Meanwhile, Ubon Ratchathani province had a relatively high proportion of northerly (N) wind trajectories (32.14%), indicating the influence of mainland winds from Southeast Asia, as shown in Table 3Figure 7. The analysis of air mass trajectories combined with land use/land cover (LULC) data reveals that numerous air masses move through cultivated areas, flooded vegetation areas, forests, and urban areas before reaching the cultivated areas of the representative provinces. The majority of these trajectories pass through agricultural areas and flooded vegetation areas in Thailand, Laos, Vietnam, and Southern China, which are major CH 4 emitters. As a result, the trajectories indicate that some air masses passed over land cover types that may contribute CH 4. However, this does not prove that rice paddies were the dominant source of the observed XCH 4 enhancement; therefore, the trajectories in which the air mass passes through the possible CH 4 source are flagged and interpreted cautiously. In comparing the results with other studies, particularly in major agricultural regions such as China, India, Vietnam, and Thailand, the findings suggest that changes in rainfall, temperature, and flooding patterns and duration within the rice paddies that are driven by climate variability influence CH 4 production and its atmospheric variability [ 49, 56]. The atmospheric CH 4 concentration observed as column-averaged XCH 4 in this study ranged from approximately 1788 to 1793 ppb, which is consistent with the reported XCH 4 range for the Monsoon Asia region of 1750–1900 ppb [ 48]. This shows that Thailand’s atmospheric CH 4 concentration falls within the regional areas of other Asian agricultural regions. Furthermore, the authors of [ 57] reported regional CH 4 emissions for Asia in the range of 45.20–64.35 Tg yr −1, demonstrating the region’s significant role as a global CH 4 source. Additionally, variations in the field-based CH 4 emissions from rice-paddy-dominated areas have been reported in Vietnam at 217–375 kg CH 4 ha −1 season −1 [ 58], China at 29.7–252.2 kg CH 4 ha −1 [ 59], Eastern India at 7–608 kg CH 4 ha −1 [ 60], and southern India at 2.2–59.3 kg CH 4 ha −1 season −1 [ 61], as shown in Table 4. Apparently, CH 4 variability is influenced by not only ENSO-related climate patterns but also local agricultural practices, cropping intensity, water management, flooding duration, soil conditions, and regional environmental factors [ 60]. Thus, this study should be recognized as an analysis of ENSO-associated satellite-observed XCH 4 variability over Thailand, while field-based and regional inventory studies provide a supporting context for potential CH 4 contributions from rice cultivation. The applicability of satellite data linked with ground-based observations is widely accepted practice in CH 4 studies for improving and verifying the reliability of CH 4 variability assessments at a local scale [ 61]. Furthermore, the results can be applied to examine the annual variation in CH 4 in the Asian monsoon region by using satellite data in comparison with ground-based data to improve the accuracy of atmospheric CH 4 change analysis and reduce uncertainty [ 57]. 5. Conclusions The findings of this study demonstrate that the ENSO phenomenon is one of the drivers associated with CH 4 variability in Thailand’s atmosphere. These atmospheric variations are linked to rice paddy conditions and the methanogenesis process, both of which are strongly influenced by water dynamics, including rainfall variability, water availability, and flooding regimes that promote anaerobic environments. Satellite data show long-term regional pattern changes over 12 years. This provides data for temporal interpretation of water management, which is an important factor associated with rice yield and XCH 4 variability. The satellite-derived column-averaged methane (XCH 4) represents the atmospheric concentration’s variability. Those XCH 4 values may be influenced by multiple methane sources, atmospheric transport, background methane trends, or vertical mixing. Therefore, the observed spatial and seasonal patterns indicate associations between ENSO-related climate variability, rice calendar timing, and atmospheric CH 4 variability. However, they do not establish rice paddies as the dominant source of the observed XCH 4 enhancement. The analysis of GOSAT as the primary long-term dataset, with Sentinel-5P/TROPOMI used as a supporting dataset for recent spatial patterns, organized according to Thailand’s rice-growing calendar, revealed distinct temporal patterns in satellite-observed XCH 4 variability. The HYSPLIT backward trajectories were used to provide air mass transport context and not decisive source attribution. Satellite-observed XCH 4 values were the highest during La Niña years, lowest during El Niño years, and intermediate during neutral conditions. Seasonal variation was also evident, with the highest observed variability occurring in Period 3 (September–December), likely due to prolonged flooding conditions. In contrast, Period 2 (May–August) exhibited the lowest variability, associated with early-season rainwater drainage, while Period 1 (January–April) showed moderate variability, partly influenced by delayed harvesting from the previous cycle. The limitation of this study can be seen in the sparse availability of ground-based observation of flux monitoring stations around the study area, which impedes the ability to validate satellite XCH 4 patterns enabling the separation of rice paddy emissions from other possible regional CH 4 sources. Future studies envision the integration of ground-based flux measurements, emission inventories, rice area cultivation data, water management information, and atmospheric transport modeling. That would allow an improved source attribution, strengthening satellite-based observations and modeling in efforts to reveal complexities of atmospheric methane origin and transport. The findings presented here emphasize the combined influence of interannual climate variability and seasonal agricultural dynamics on atmospheric CH 4 variability over rice-growing regions. The results may provide a basis for improving rice management practices, particularly the adoption of alternate wetting and drying (AWD), which can significantly reduce field-scale CH 4 emissions while maintaining productivity and enhancing climate resilience. Furthermore, the use of satellite-derived CH 4 data offers a potential for upscaling emission estimates to regional and national levels, supporting more effective monitoring and mitigation strategies. Author Contributions Conceptualization, W.T. and P.M.; methodology, K.D., O.P., Y.L. and S.Z.; software, Y.L. and S.Z.; validation, P.M.; formal analysis, S.P. and J.B.; writing—original draft preparation, W.T.; writing—review and editing, P.M. and W.W.S.; project administration, P.J. and T.K. All authors have read and agreed to the published version of the manuscript. Funding The research was financially supported by the National Research Council of Thailand through research core funding No. FY2026 CSA for Small Paddy Farmers to Reduce Methane Emissions and to Increase Yields in Terraced Paddy Areas. Data Availability Statement The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding authors. Acknowledgments The authors would like to express their sincere gratitude to the Faculty of Environment, Kasetsart University, and The King’s Royally Initiated Laem Phak Bia Environmental Research and Development Project Initiative for their support of this research. Conflicts of Interest The authors declare no conflicts of interest. Abbreviations The following abbreviations are used in this manuscript: CF continuous flooding ENSO El Niño–Southern Oscillation AWD alternate wetting and drying SST sea surface temperature ONI Oceanic Niño Index m AGL meters above ground level References Figure 1. Study areas: rice cultivation regions shown on the map of Thailand (Years 2022–2024). Figure 1. Study areas: rice cultivation regions shown on the map of Thailand (Years 2022–2024). Figure 2. Workflow for generating XCH 4 variability maps of Thailand using Jupyter Notebook version 7.2.2. Figure 2. Workflow for generating XCH 4 variability maps of Thailand using Jupyter Notebook version 7.2.2. Figure 3. Workflow for supporting interpretation of XCH 4 variability using air mass. Transport model. Figure 3. Workflow for supporting interpretation of XCH 4 variability using air mass. Transport model. Figure 4. Growing area, yields, rainfall, and mean annual temperature in Thailand rice cultivation (2012–2022). Figure 4. Growing area, yields, rainfall, and mean annual temperature in Thailand rice cultivation (2012–2022). Figure 5. Long-term trends for Thailand’s CH 4 emissions and annual CH 4 concentration variability. Figure 5. Long-term trends for Thailand’s CH 4 emissions and annual CH 4 concentration variability. Figure 6. Maps of satellite-observed XCH 4 variability in Thailand during El Niño, neutral, and La Niña periods. Panels represent the following: ( a) Period 1—El Niño; ( b) Period 1— neutral; ( c) Period 1—La Niña; ( d) Period 2—El Niño; ( e) Period 2—neutral; ( f) Period 2—La Niña; ( g) Period 3—El Niño; ( h) Period 3—neutral; ( i) Period 3—La Niña. Figure 6. Maps of satellite-observed XCH 4 variability in Thailand during El Niño, neutral, and La Niña periods. Panels represent the following: ( a) Period 1—El Niño; ( b) Period 1— neutral; ( c) Period 1—La Niña; ( d) Period 2—El Niño; ( e) Period 2—neutral; ( f) Period 2—La Niña; ( g) Period 3—El Niño; ( h) Period 3—neutral; ( i) Period 3—La Niña. Figure 7. Spatial distribution of backward air mass trajectories from different land use/land cover for Thailand: ( a) Nakhon Sawan; ( b) Suphan Buri; ( c) Phetchaburi; ( d) Ubon Ratchathani; ( e) Phitsanulok. Figure 7. Spatial distribution of backward air mass trajectories from different land use/land cover for Thailand: ( a) Nakhon Sawan; ( b) Suphan Buri; ( c) Phetchaburi; ( d) Ubon Ratchathani; ( e) Phitsanulok. Table 1. Seasonal Oceanic Niño Index (ONI) values for ENSO classification during 2012–2022. Table 1. Seasonal Oceanic Niño Index (ONI) values for ENSO classification during 2012–2022. Year DJF JFM FMA MAM AMJ MJJ JJA JAS ASO SON OND NDJ 2012 −0.9 −0.7 −0.6 −0.5 −0.3 0.0 0.2 0.4 0.4 0.3 0.1 −0.2 2013 −0.4 −0.4 −0.3 −0.3 −0.4 −0.4 −0.4 −0.3 −0.3 −0.2 −0.2 −0.3 2015 0.5 0.5 0.5 0.7 0.9 1.2 1.5 1.9 2.2 2.4 2.6 2.6 2016 2.5 2.1 1.6 0.9 0.4 −0.1 −0.4 −0.5 −0.6 −0.7 −0.7 −0.6 2017 −0.3 −0.2 0.1 0.2 0.3 0.3 0.1 −0.1 −0.4 −0.7 −0.8 −1.0 2019 0.7 0.7 0.7 0.7 0.5 0.5 0.3 0.1 0.2 0.3 0.5 0.5 2020 0.5 0.5 0.4 0.2 −0.1 −0.3 −0.4 −0.6 −0.9 −1.2 −1.3 −1.2 2021 −1.0 −0.9 −0.8 −0.7 −0.5 −0.4 −0.4 −0.5 −0.7 −0.8 −1.0 −1.0 2022 −1.0 −0.9 −1.0 −1.1 −1.0 −0.9 −0.8 −0.9 −1.0 −1.0 −0.9 −0.8 ONI, (−0.5)–(−1.5) °C La Niña, (−0.4)–0.4 °C neutral, 0.5–2.0 °C El Niño. DJF, December–January–February; JFM, January–February–March; FMA, February–March–April; MAM, March–April–May; AMJ, April–May–June; MJJ, May–June–July; JJA, June–July–August; JAS, July–August–September; ASO, August–September–October; SON, September–October–November; OND, October–November–December; NDJ, November–December–January. Source: [ 22]. Table 2. Three cropping periods aligned with the rice cultivation calendar. Table 2. Three cropping periods aligned with the rice cultivation calendar. Period Rice Cultivation Phases Period 1 (January–April) Planting and growing phase of dry-season rice. Period 2 (May–August) Planting and growing phase of wet-season rice and harvesting of dry-season rice. Period 3 (September–December) Growing phase and harvesting of wet-season rice. * * The dominant share of rice production in Thailand. Table 3. Directional frequency of backward air mass trajectories toward representative rice cultivation provinces. Table 3. Directional frequency of backward air mass trajectories toward representative rice cultivation provinces. N (%) NE (%) E (%) SE (%) S (%) SW (%) W (%) NW (%) Nakhon Sawan 3.57 44.64 1.79 1.79 7.14 23.21 7.14 10.71 Suphan Buri 3.58 48.21 1.79 1.78 3.56 32.14 5.36 3.57 Phetchaburi 17.86 37.50 5.36 3.56 1.79 28.57 3.57 1.78 Ubon Ratchathani 32.14 42.86 5.36 5.35 1.78 7.15 1.78 3.57 Phitsanulok 1.78 58.93 7.14 8.93 3.57 17.86 0 1.79 N, north; NE, northeast; E, east; SE, southeast; S, south; SW, southwest; W, west; NW, northwest. Table 4. Selected field-based CH 4 emission and satellite-observed atmospheric XCH 4 variability in Asian regions. Table 4. Selected field-based CH 4 emission and satellite-observed atmospheric XCH 4 variability in Asian regions. Location Methane Parameter Value Unit Reference Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. © 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. Share and Cite MDPI and ACS Style Tundam, W.; Maskulrath, P.; Duangmal, K.; Poommai, S.; Phewnil, O.; Liu, Y.; Zhang, S.; Szymanski, W.W.; Jaikaew, P.; Kato, T.; et al. Remote Sensing of El Niño–Southern Oscillation Impact on Methane Flux Potential from Rice Cultivation in Thailand. Environments 2026, 13, 320. https://doi.org/10.3390/environments13060320 AMA Style Tundam W, Maskulrath P, Duangmal K, Poommai S, Phewnil O, Liu Y, Zhang S, Szymanski WW, Jaikaew P, Kato T, et al. Remote Sensing of El Niño–Southern Oscillation Impact on Methane Flux Potential from Rice Cultivation in Thailand. Environments. 2026; 13(6):320. https://doi.org/10.3390/environments13060320 Chicago/Turabian Style Tundam, Warisara, Parkin Maskulrath, Kittichai Duangmal, Satreethai Poommai, Onanong Phewnil, Yibo Liu, Siqing Zhang, Wladyslaw Witold Szymanski, Piyanuch Jaikaew, Tasuku Kato, and et al. 2026. "Remote Sensing of El Niño–Southern Oscillation Impact on Methane Flux Potential from Rice Cultivation in Thailand" Environments 13, no. 6: 320. https://doi.org/10.3390/environments13060320 APA Style Tundam, W., Maskulrath, P., Duangmal, K., Poommai, S., Phewnil, O., Liu, Y., Zhang, S., Szymanski, W. W., Jaikaew, P., Kato, T., & Boonphue, J. (2026). Remote Sensing of El Niño–Southern Oscillation Impact on Methane Flux Potential from Rice Cultivation in Thailand. Environments, 13(6), 320. https://doi.org/10.3390/environments13060320 Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here. Article Metrics Article metric data becomes available approximately 24 hours after publication online.
