Application of Multi-Source Remote Sensing Fusion for Identifying Smoke Fugitive Channels in the Sulabulak Fire Area, Urumqi, China

Application of Multi-Source Remote Sensing Fusion for Identifying Smoke Fugitive Channels in the Sulabulak Fire Area, Urumqi, China | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Application of Multi-Source Remote Sensing Fusion for Identifying Smoke Fugitive Channels in the Sulabulak Fire Area, Urumqi, China Zhicheng Yang, Qiang Zeng This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4856299/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Underground coal fires are a pervasive global environmental issue, especially in coal-rich regions, causing significant environmental damage, safety hazards, and economic losses. These fires release smoke containing carbon dioxide and other harmful gases, exacerbating climate change. This study presents a novel comprehensive analysis method using multi-source remote sensing technology to detect smoke fugitive channels caused by coal fires. We utilized 29 Landsat-8 satellite images of the Sulabulak fire area in China to retrieve vegetation coverage (FVC) and land surface temperature (LST), identifying sparse vegetation and high-temperature anomaly areas. Additionally, 135 dual-polarized Sentinel-1A images were used to obtain surface deformation through SBAS-InSAR and PS-InSAR techniques. The integration of these datasets, validated by field survey data, revealed a high degree of overlap between the identified smoke fugitive channels and subsidence areas. Our results demonstrate an annual increase in sparse vegetation areas, high-temperature anomalies, and ground subsidence, indicating intensified coal fire combustion and expanding smoke fugitive channels. This method's effectiveness in identifying coal fire areas underscores its potential for enhancing coal fire monitoring and management, contributing to more accurate carbon emission estimates and improved mitigation strategies. Smoke fugitive channels identification Underground coal fires Vegetation coverage Land surface temperature InSAR Technique Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Introduction Coal field fires, as a long-term and complex environmental and safety issue worldwide, pose severe challenges, especially in coal-rich regions (Zeng, 2023). Underground coal fires not only result in significant economic losses but also have widespread impacts on surface ecosystems and the atmosphere (Biswal et al., 2019; Liang et al., 2023). The release of smoke from coal field fires, particularly carbon dioxide and other harmful gases, exacerbates the severity of global climate change (Carroll et al., 2022; Roy et al., 2019; H. Zhang et al., 2023). In China, carbon dioxide emissions from coal fires account for 0.32% of the global emissions, while global coal fire emissions contribute to 0.42% of the total global emissions (Minx et al., 2021). Therefore, accurate monitoring and assessment of the dynamic changes in coal fire areas and their smoke fugitive channels are crucial for implementing effective fire control measures and achieving China's "carbon peak" and "carbon neutrality" goals. Previous studies have employed various techniques for detecting and monitoring coal fires, including thermal infrared (TIR), multispectral, and hyperspectral remote sensing, as well as unmanned aerial vehicles (UAVs). (He et al., 2020; Yan et al., 2020). These methods, while providing valuable insights, have limitations in accurately identifying smoke fugitive channels due to issues like data acquisition range, cost, sensitivity to weather conditions, and spatial resolution constraints. (Wang et al., 2022). For example, UAVs offer high-resolution monitoring but are limited by their operational range and sensitivity to weather conditions (Kelly et al., 2019; McKenna et al., 2018). Slavecki RJ first applied thermal infrared remote sensing technology to coal fire detection (Biswal et al., 2019). This method is favored for its non-contact nature, low cost, time efficiency, and wide spatial coverage (Chen et al., 2022; Du et al., 2022; Kuenzer et al., 2007). It utilizes thermal infrared data to obtain land surface temperature (LST), then identifies thermal anomalies associated with coal fires using specific criteria. Although it can quickly identify thermal anomalies and is widely used, the results are relatively coarse and ineffective for detecting thermal anomalies in summer, particularly in deep underground coal fire areas (Martha et al., 2010; Mishra et al., 2011; Roy et al., 2015). This indicates that using a single remote sensing image is insufficient for accurately detecting thermal anomalies and does not meet the need for comprehensive identification of coal fires and their smoke fugitive channels. As coal field fires spread deeper underground, their concealment increases, posing challenges to traditional thermal anomaly detection methods (Yan et al., 2020). To more accurately identify coal fires and their smoke fugitive channels, integrated satellite remote sensing methods have become a research hotspot (Liu et al., 2019; Schroeder et al., 2016; Yu et al., 2022). InSAR technology can effectively detect land subsidence, providing a new perspective for identifying coal fire areas (Aditiya & Ito, 2023; Hooper et al., 2012). These techniques can reveal surface deformations caused by coal fire combustion, offering valuable information even when initial subsidence is not apparent. To date, researchers have successfully combined PS-InSAR and SBAS-InSAR methods with global positioning systems and field survey data, comprehensively verifying the critical role of these methods in identifying coal fires (Riyas et al., 2021; Wang et al., 2019). However, since surface subsidence can be caused by various factors, including mining activities and the coal fire itself, relying solely on subsidence data makes it challenging to accurately distinguish and identify coal fires. Given these limitations, there is a need for a more integrated approach that combines multiple remote sensing techniques to improve the accuracy and reliability of smoke fugitive channel detection. This study aims to fill this gap by proposing a novel comprehensive analysis method that integrates multi-source remote sensing data, including Landsat-8 and Sentinel-1A, to detect and monitor smoke fugitive channels caused by coal fires. By employing this method, we aim to more effectively monitor coal fire activities under different seasonal and climatic conditions, accurately locate smoke fugitive channels, and provide a scientific basis for formulating fire extinguishing measures and calculating carbon emissions in coal fire areas. The structure of this paper is as follows: Section 2 describes the study area and datasets. Section 3 outlines the research methodology. Section 4 presents the experimental results and analysis. Section 5 discusses the findings and validates the proposed method. Finally, Section 6 concludes the paper and suggests directions for future research. Materials and methods Study Area The Sulabulak fire area is located in the north-central region of Xinjiang, China, approximately 86 km south of Urumqi(Fig. 1 ). This area lies between latitudes 87°27′ and 87°32′ and longitudes 43°20′ and 43°23′. The terrain is characterized by mid-low mountain hills, with elevations ranging from 1995 to 2211 meters, and relative height differences of 50 to 110 meters. The region has a continental climate with annual precipitation between 170.4 and 201.1 mm, and predominantly experiences southwest winds. The geological structures of the Sulabulak fire area include the Jurassic Xishanyao Formation (J2x) and Quaternary sediments. The coal seams, which are highly flammable due to their low sulfur, low phosphorus, high calorific value, and high melting temperature, consist mainly of black, blocky, or layered coal with well-developed joints and fractures. Coal fires were discovered during field surveys in 2016, primarily caused by incomplete stripping and underground mining that exposed shallow coal seams to spontaneous combustion. These fires have progressively worsened, causing significant environmental damage and economic losses. Datasets This study utilized 29 Landsat Collection 2 Level 2 images from 2013 to 2023. Released by the US Geological Survey (USGS), this dataset comprises enhanced Earth observation products from the Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) on Landsat 8 and Landsat 9 satellites (Masek et al., 2020). These datasets, featuring surface reflectance and temperature products, are processed with advanced radiative transfer models and supplementary atmospheric data to enhance accuracy and usability (Claverie et al., 2018). Available in image formats of approximately 185 km by 180 km, the data utilize Universal Transverse Mercator (UTM) projection, or Polar Stereographic (PS) projection for polar regions (Dwyer et al., 2018). Stored as 16-bit unsigned integers, the datasets include metadata for converting digital numbers (DN) to radiance, reflectance, and brightness temperature. These datasets, available via the USGS Earth Explorer portal, USGS EROS Machine-to-Machine (M2M) API, and AWS Simple Storage Service (S3), offer direct access for detailed analysis (Crawford et al., 2023). This study utilizes Sentinel-1 data from the European Space Agency (ESA). Sentinel-1, a C-band satellite with a 12-day revisit cycle (Torres et al., 2012), provided 135 images acquired between January 13, 2018, and December 13, 2023. The standard acquisition mode was Interferometric Wide (IW) with dual-polarization (VV/VH). Precise orbit corrections used data from the ESA Sentinel-1 Quality Control website. Additionally, 30-meter spatial resolution SRTM data served as external reference digital elevation models (DEMs) for InSAR processing. For SBAS-InSAR processing, 428 interferograms were generated, with temporal baselines ranging from 12 to 60 days and the longest spatial baseline being 122.1 meters. The average baseline length for PS-InSAR was − 157.2 meters. Temporal and spatial baseline graphs from SBAS-InSAR and PS-InSAR processing are shown in Fig. 2 . Vegetation cover retrieve Accurate vegetation cover estimation is crucial for understanding the impact of coal fires on the local environment, particularly in identifying smoke fugitive channels. Vegetation cover inversion leverages the spectral absorption and reflection characteristics of vegetation across different spectral bands (Chen et al., 2024). This study employs the pixel dichotomy method to extract the Fractional Vegetation Cover (FVC) for the study area over the period from 2013 to 2023, reflecting the vegetation coverage status in the region. The Normalized Difference Vegetation Index (NDVI) is calculated using the near-infrared and red bands of remote sensing data. Based on this, the Fractional Vegetation Cover (FVC) is calculated using the pixel dichotomy model. The formulas are as follows: $$\:\begin{array}{c}NDVI=\frac{{R}_{NI}-R}{{R}_{NI}+R} \left(1\right)\end{array}$$ $$\:\begin{array}{c}FVC=\frac{NDVI-NDV{I}_{soil}}{NDV{I}_{veg}-NDV{I}_{soil}} \left(2\right)\end{array}$$ where NDVI is the Normalized Difference Vegetation Index, R NI is the spectral value of the near-infrared band, and R is the spectral value of the red band; NDVI soil represents the NDVI of bare soil; and NDVI veg represents the NDVI of fully vegetated areas. The values for NDVI soil and NDVI veg are selected from the 5%-95% confidence interval. After obtaining the FVC values, the vegetation cover for each period is reclassified using ArcGIS software. This reclassification is based on the field vegetation conditions in the area, ensuring that the remote sensing data aligns with ground truth observations. The number of pixels and the area for each interval are then statistically analyzed to understand the temporal dynamics of vegetation cover. This approach enables a detailed analysis of vegetation cover changes over time, providing insights into the environmental impact of coal fires and aiding in the identification of smoke fugitive channels. Temperature retrieve In this study, we used ENVI software to monitor land surface temperature (LST) with products from the United States Geological Survey (USGS) Landsat-8 Collection 2 Level 2. The LST data is captured by the Thermal Infrared Sensor (TIRS) on the Landsat 8 satellite and undergoes preprocessing, including atmospheric correction and surface emissivity correction (Galve et al., 2022). The LST calculation process follows these steps: $$\:\begin{array}{c}L\left(\lambda\:\right)={M}_{L}\times\:{Q}_{cal}+{A}_{L} \left(3\right)\end{array}$$ $$\:\begin{array}{c}T=\frac{{K}_{2}}{\text{ln}\left(\frac{{K}_{1}}{L\left(\lambda\:\right)}+1\right)} \left(4\right)\end{array}$$ $$\:\begin{array}{c}{T}_{s}=\frac{T}{1+\left(\lambda\:\times\:\frac{T}{\rho\:}\right)\times\:\text{ln}\left(ϵ\right)} \left(5\right)\end{array}$$ The digital number (DN) is first converted to spectral radiance L(λ). In formula 3, M L is the radiance multiplicative factor, Q cal is the calibrated pixel value (DN), and A L is the radiance additive factor. Using the Planck equation in formula 4, we calculate the brightness temperature. T is the absolute temperature (in Kelvin), and K1 and K2 are pre-launch calibration constants. Formula 5 uses a modified Planck equation to convert the radiance back to actual temperature, where T S is the land surface temperature (in Kelvin), λ is the wavelength (in meters), and ρ is defined as h*c/σ , with h being the Planck constant, c the speed of light, and σ the Stefan-Boltzmann constant; ϵ is the surface emissivity (Meghraj et al., 2023). InSAR deformation retrieval SBAS-InSAR is an advanced time-series Synthetic Aperture Radar (InSAR) analysis method that constructs short-time baseline differential interferograms by selecting and combining collected data. This approach efficiently overcomes spatial decorrelation by selecting target points based on spatial coherence coefficients and building a linear model. Using Singular Value Decomposition (SVD) or the Least Squares Method (LSM), linear deformation rates and elevation error values for the target points can be obtained, resulting in a comprehensive deformation time series over the entire observation period (Selvakumaran et al., 2018). SBAS-InSAR technology excels in processing long-term observational data and monitoring large-area deformation, making it especially suitable for detailed ground deformation studies. Persistent Scatterer Interferometry (PS-InSAR) is a technique used for monitoring ground subsidence and deformation across various regions. It identifies stable scatterers, known as Persistent Scatterers (PS), which exhibit consistent phase behavior over long periods. PS-InSAR combines high precision and efficiency, enabling continuous monitoring of ground deformation, particularly useful for detecting slow-moving deformations. This technique provides high temporal resolution deformation information, offering detailed measurements of ground subsidence and deformation (P. Zhang et al., 2023). When used together, PS-InSAR and SBAS-InSAR allow tracking of various subsidence rates, significantly enhancing monitoring density and accuracy (Ramzan et al., 2022). The technical workflows for SBAS-InSAR and PS-InSAR are illustrated in Fig. 3 . These complementary methods provide thorough and precise monitoring of ground deformation, essential for comprehending and managing the dynamics and effects of underground coal fires. Detailed processing methods for SBAS and PS techniques can be found in (Tian et al., 2023; Z. Zhang et al., 2023). These methods provide robust technical support for studying smoke fugitive channels and surface deformation in underground coal fire areas, essential for precise monitoring and developing effective mitigation measures. Integrated Smoke Channel Detection Previous research shows some overlap between temperature anomalies and surface deformation (Zhou et al., 2013; L. Jiang et al., 2011). However, temperature anomalies caused by underground coal fires exhibit temporal lag, non-linear characteristics, and significant seasonal variation (Song & Kuenzer, 2014). Directly overlaying these results can lead to data omissions and errors. To address this issue, this study integrates high-temperature threshold extraction and analysis of sparse to moderately sparse vegetation coverage, supplemented with surface deformation data, to verify and accurately locate smoke fugitive channels in coal fire areas. Figure 4 . Frequency superposition for time series images to extract coal fire pixels The high-temperature threshold ( Th ) is derived from temperature inversion data. It is defined as the sum of the mean surface temperature ( Tm ) and twice the standard deviation ( Tsd ), serving as the optimal threshold to distinguish between coal fire areas and non-coal fire areas (W. Jiang et al., 2011), as shown in formula 6. To obtain more accurate temperature threshold information for coal fire areas, we incorporated thermal interference data from non-coal fire regions collected through field surveys, along with the location and thermal information of actual coal fire boundaries. We conducted a comparative filtering experiment to extract fire area ranges at different time series. As illustrated in Fig. 4 , the high-temperature threshold attribute was set to 1, while other areas were set to 0, and then the data were superimposed. A pixel was identified as an abnormal temperature region when its frequency of appearance in the same location and within the set period reached the predefined high-frequency filtering threshold. $$\:\begin{array}{c}{T}_{h}={T}_{m}+2{T}_{Sd} \left(6\right)\end{array}$$ By employing this method, this study can more accurately identify areas affected by underground coal fires, thereby providing a more reliable scientific basis for the monitoring and management of coal fire disasters. Result Vegetation retrieve After conducting field surveys in the Sulabulak fire area and assessing the local vegetation growth, the derived vegetation coverage was classified into different levels as shown in Table 1 : sparse vegetation (0 ≤ FVC < 0.3), moderately sparse vegetation (0.3 ≤ FVC < 0.5), dense vegetation (0.5 ≤ FVC < 0.7), and very dense vegetation (0.7 ≤ FVC < 1). After processing with ArcGIS, part of the vegetation inversion results are shown in Fig. 4 . Table 1 Classification of FVC FVC classify FVC Value Ranges Sparse Vegetation 0 ≤ FVC<0.3 Moderately Sparse Vegetation 0.3 ≤ FVC<0.5 Dense Vegetation 0.5 ≤ FVC<0.7 Very Dense Vegetation 0.7 ≤ FVC<1 The processed results, depicted in Fig. 4 , indicate that low vegetation cover areas are mainly distributed in the southern and southeastern parts of the study area, while high vegetation cover areas are located in the western valleys and their sides. Over the study period from 2013 to 2023, there was a notable increase in the area of sparse and moderately sparse vegetation. Statistical analysis shows that the proportion of sparse vegetation cover decreased from 4.8–4.3% (a reduction of 0.11 km²), while moderately sparse vegetation cover increased from 15.9–26.1% (an area increase of 2.88 km²). Dense vegetation cover also increased from 34.5–37.1% (an area increase of 0.55 km²). Conversely, the proportion of very dense vegetation cover decreased from 44.9–32.6% (an area decrease of 2.60 km²). Temperature retrieve and thermal anomaly detection The heat from underground coal fires transfers to the surface, raising surface temperatures in coal fire areas above those of the surrounding environment. Surface temperature inversion of remote sensing images from 2013 to 2023 was performed, with some results displayed in Fig. 5 . Analysis of Fig. 5 reveals that high-temperature areas are distributed linearly from northeast to southwest, while low-temperature areas are primarily located in the southern and southwestern mountain valleys. During the study period from 2013 to 2023, the highest retrieved temperature was 36.68°C, and the lowest was − 20.13°C. Using the high-temperature anomaly threshold, retrieved temperatures for each period were reclassified in ArcGIS to extract areas exceeding the temperature threshold. The results indicate that from 2013 to 2023, the area of temperature anomalies increased from 0.5–3.5%, corresponding to an area increase of 0.65 km². Fugitive flue gas channel analysis The sparse vegetation cover areas and high-temperature anomaly areas were extracted using ArcGIS software, and their change curves are shown in Fig. 6 . The analysis indicates that the area of sparse vegetation cover increased from 1.49 km² to 2.77 km², while the area of high-temperature anomalies expanded from 0.10 km² to 0.75 km². Both areas show a consistent year-by-year increase. Based on the results of vegetation cover and surface temperature inversion, the areas with sparse vegetation cover and high-temperature anomalies were overlaid to identify the smoke fugitive channels of coal fire combustion, as illustrated in Fig. 6 . The figure demonstrates that from 2013 to 2023, the overlapping points of sparse vegetation cover and high-temperature anomaly areas exhibited a fluctuating pattern over time. These overlapping areas are predominantly located in the central and eastern parts of the study area. The number of overlapping points was at its lowest in 2017, with only four points, and reached its peak in 2023 with 66 points. This transition from isolated points to broader areas in the overlapping regions indicates an expanding trend of the smoke fugitive channels, suggesting an intensification of coal fire combustion. Deformation monitoring results and analysis Using data from 135 dual-polarization (VV/VH) Sentinel-1A images, covering 72 time periods from January 2018 to December 2023, deformation in the line-of-sight (LOS) was obtained using SBAS and PS methods. The cumulative surface subsidence data provides insights into the temporal evolution, magnitude, subsidence trends, and the distribution of major subsidence areas within the study region. The time series of cumulative surface subsidence for the study area is illustrated in Fig. 7 . From January 2018 to December 2023, the study area exhibited a maximum cumulative subsidence of -123.9 mm and a maximum cumulative uplift of 48.41 mm, with an increasing trend observed annually. As shown in Fig. 8 , the time series deformation of pixel points located in the subsidence area (P1 point in Fig. 8 ) using the two methods (SBAS and PS) indicates good consistency. The results of SBAS and PS were imported into ArcGIS, and 20,000 random points were generated for overlay analysis. After removing invalid points, 9,050 overlay points remained for correlation analysis, as shown in Fig. 9 (a). The correlation coefficient (R²) between the deformation rates obtained from the two monitoring results is 0.89, demonstrating a strong correlation. The histogram in Fig. 9 (b) shows the deformation rate differences between SBAS and PS for the same pixel points. Most differences fall within ± 4 mm/yr, with a standard deviation of 0.75 mm/yr and a mean value of -1.34 mm/yr. This suggests that, although there are slight discrepancies between the two methods, both provide reliable ground deformation monitoring results. Overall, the SBAS monitoring results for surface deformation are robust and can be effectively combined with temperature and vegetation coverage data to identify smoke fugitive channels in coal fire areas. This combined method improves the precision and dependability of coal fire detection and monitoring, offering essential insights for future research and mitigation efforts. Discussion To analyze the performance of this method in detecting smoke fugitive channels in different regions, this section overlays the deformation time series, LST, and FVC. Additionally, the impacts of coal fires, mining activities, and other factors were analyzed using LST, FVC, and deformation data from characteristic points in various regions. This analysis reveals the evolutionary characteristics of smoke fugitive channels in different areas. Figure 10 shows the identified areas where sparse vegetation coverage and high-temperature anomalies overlap, along with subsidence monitoring data. It can be seen that the subsidence points and the overlapping areas are largely consistent. Nine characteristic points were selected from A1 to D2. The time series data for subsidence, LST, and FVC were extracted and overlaid, as shown in Fig. 11 , with the time series of LST (red line), FVC (green line), and deformation (blue line). The distribution of the characteristic points and field survey data classify them into four categories. The first category includes A1, A2, and A3, located at smoke fugitive channel points in coal fire areas. The second category includes B1 and B2, which are in high-temperature areas with normal vegetation coverage. The third category includes C1 and C2, which are in low vegetation areas with normal temperatures. The fourth category includes D1 and D2, which are control points located far from the fire area. Characteristic points A1, A2, and A3 are located in coal fire areas. As shown in Fig. 11 , the ground subsidence at category A points is significant and increases over time, indicating that these areas are severely affected by coal fires. Additionally, the surface temperature (LST) is generally high, particularly outside of the summer months, but weak during summer. Most coal fire anomalies fall into this category, explaining why single thermal infrared imagery is ineffective for detecting coal fires in summer (Song & Kuenzer, 2014). The vegetation cover (FVC) is generally low and shows little variation over time, indicating that the vegetation is severely affected by coal fires and difficult to recover. A significant correlation exists between temperature, FVC, and ground subsidence. The data distribution for category A points shows a large median subsidence with a wide range, indicating significant and unstable ground subsidence in the area(Fig. 12 ); the LST median is high with large fluctuations, indicating drastic temperature changes; the FVC median is low with a small range, indicating minimal vegetation cover. According to field surveys, this area was affected by spontaneous combustion of residual coal in inadequately mined underground spaces. These areas can be marked as smoke fugitive channels. Based on field investigations, characteristic points B1 and B2 are covered with coal piles. As coal accumulates, it slowly oxidizes upon contact with air, releasing heat and causing thermal anomalies. However, these coal piles do not combust, resulting in relatively low heat release. The surrounding areas have high vegetation cover with significant variation, likely influenced by seasonal changes. Additionally, ground subsidence was observed at B1 and B2, but the magnitude is small, and the region remains relatively stable. Therefore, it can be concluded that these points are not smoke fugitive channels from coal fires. Characteristic point C1 is located in the Xinxing coal mine, and C2 is located in the Shunda coal mine. These areas experience significant human activities, such as mining operations and exploration projects. The low vegetation cover at these points may be due to these activities or other complex environmental factors. The ground subsidence in these areas is minimal, indicating stable surface conditions. The temperature variations are normal and not significantly affected by coal fires. Therefore, in the proposed method, these areas can be excluded from influencing the detection results of smoke fugitive channels caused by coal fires. Characteristic points D1 and D2 in category D exhibit the smallest median subsidence and the smallest range of variation, indicating extremely stable surface conditions unaffected by coal fires. The median LST is relatively low with a small range of temperature variation, showing that these areas have stable temperatures without significant high-temperature anomalies. The FVC median is the highest with a large range of variation, indicating good vegetation cover that is healthy and responsive to seasonal changes. According to field surveys, these areas have a stable ecological environment, unaffected by coal fires or other human activities, making them suitable as control points. The stability of category D points provides a baseline reference for understanding environmental changes in areas affected by coal fires. The analysis above indicates that the deformation and heat release processes in the study area are complex. The combustion of coal and mining activities alter the underground stress state, contributing to observed subsidence. High-temperature anomalies and sparse vegetation coverage areas include those caused by coal combustion and heat accumulation from other human activities or surface objects. Therefore, identifying smoke fugitive channels caused by coal fires cannot depend on a single data source, as done in previous studies (Deng et al., 2021; He et al., 2020; Li et al., 2018). Although the method proposed in this paper is suitable for both mining and non-mining areas, it still requires thorough analysis and field verification to ensure its accuracy and applicability. Conclusions This study proposes a novel comprehensive analysis method utilizing multi-source remote sensing technology to detect smoke fugitive channels caused by coal fires. The analysis employed twenty-nine Landsat-8 satellite images from the Sulabulak fire area in China to retrieve vegetation coverage (FVC) and land surface temperature (LST). Sparse vegetation coverage and high-temperature anomaly areas were extracted and overlaid to identify smoke fugitive channels. Additionally, 135 dual-polarized Sentinel-1A images were used to obtain ground deformation data via SBAS-InSAR and PS-InSAR techniques. Field survey data validated the identification of smoke fugitive channels, showing a high degree of agreement at nine characteristic points within the overlapping and subsidence areas. This confirms the method's accuracy in determining the spatial distribution of smoke fugitive channels. From the results of this study, we can draw the following conclusions: By analyzing Landsat-8 and Sentinel-1A data for the last decade, we found that the area of sparse vegetation cover in the Sulabulak fire area has increased in size to 2.77 km 2 , and the area of high-temperature anomalies has expanded to 0.75 km 2 , with a cumulative surface deposition of -123.9 mm. These data indicate an increase in the intensity of burning in the coal fires and an increase in the extent of smoke escape routes. Combining surface deformation data from SBAS-InSAR technology verified the effectiveness of remote sensing in detecting smoke fugitive channels, particularly in identifying ground subsidence related to coal fire activities. The analysis showed that ground subsidence is primarily concentrated in coal fire areas and smoke fugitive channels, with a high degree of overlap with high-temperature anomaly areas and sparse vegetation coverage. This suggests that coal fire combustion and the presence of smoke fugitive channels are major factors causing ground subsidence, highlighting the need to comprehensively consider various environmental and human activity factors when identifying and evaluating coal fire regions. While this method has been successful in the Sulabulak fire area, the complexity of coal field regions necessitates further field validation and detailed analysis to ensure the method's generalizability and accuracy. Future research will extend this method to different types and scales of coal fire areas, considering factors such as terrain, climatic conditions, and human activities. Additionally, based on the identified smoke fugitive channel areas, we will estimate carbon dioxide emissions using coal sample combustion characteristics, soil temperature, and moisture parameters, which will be a focus of our future research. Declarations Ethics approval All authors have read, understood, and have complied as applicable with the statement on “Ethical responsibilities of Authors” as found in the Instructions for Authors. Consent to participate Informed consent was obtained from all individual participants included in the study. Competing interests The authors declare no competing interests. Funding This research was financially supported by Innovation Leading Talent project (grant no.2023TSYCLJ0003) funded by the Xinjiang Department of Science and Technology of China. Author Contribution Zhicheng Yang：Writing – original draft, Conceptualization, Investigation, Methodology. Qiang Zeng: Supervision, Investigation, Funding acquisition, Writing – review and editing. Data Availability The datasets are available upon reasonable request from the corresponding author. References Aditiya, A., Ito, T., 2023. Present-day land subsidence over Semarang revealed by time series InSAR new small baseline subset technique. International Journal of Applied Earth Observation and Geoinformation 125, 103579. https://doi.org/10.1016/j.jag.2023.103579 Biswal, S.S., Raval, S., Gorai, A.K., 2019. Delineation and mapping of coal mine fire using remote sensing data - a review. Int. J. Remote Sens. 40, 6499–6529. https://doi.org/10.1080/01431161.2018.1547455 Carroll, M., Gao, C.X., Campbell, T.C.H., Smith, C.L., Dimitriadis, C., Berger, E., Maybery, D., Ikin, J., Abramson, M.J., Sim, M.R., McFarlane, A., Smith, K., Guo, Y., 2022. Impacts of coal mine fire-related PM2.5 on the utilisation of ambulance and hospital services for mental health conditions. Atmospheric Pollution Research 13, 101415. https://doi.org/10.1016/j.apr.2022.101415 Chen, M., Xue, Yayong, Xue, Yibo, Peng, J., Guo, J., Liang, H., 2024. Assessing the effects of climate and human activity on vegetation change in Northern China. Environmental Research 247, 118233. https://doi.org/10.1016/j.envres.2024.118233 Chen, X., Peng, J., Song, Z., Zheng, Y., Zhang, B., 2022. Monitoring Persistent Coal Fire Using Landsat Time Series Data From 1986 to 2020. IEEE Trans. Geosci. Remote Sensing 60, 5001616. https://doi.org/10.1109/TGRS.2022.3142350 Claverie, M., Ju, J., Masek, J.G., Dungan, J.L., Vermote, E.F., Roger, J.-C., Skakun, S.V., Justice, C., 2018. The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sens. Environ. 219, 145–161. https://doi.org/10.1016/j.rse.2018.09.002 Crawford, C.J., Roy, D.P., Arab, S., Barnes, C., Vermote, E., Hulley, G., Gerace, A., Choate, M., Engebretson, C., Micijevic, E., Schmidt, G., Anderson, C., Anderson, M., Bouchard, M., Cook, B., Dittmeier, R., Howard, D., Jenkerson, C., Kim, M., Kleyians, T., Maiersperger, T., Mueller, C., Neigh, C., Owen, L., Page, B., Pahlevan, N., Rengarajan, R., Roger, J.-C., Sayler, K., Scaramuzza, P., Skakun, S., Yan, L., Zhang, H.K., Zhu, Z., Zahn, S., 2023. The 50-year Landsat collection 2 archive. Science of Remote Sensing 8, 100103. https://doi.org/10.1016/j.srs.2023.100103 Deng, J., Ge, S., Qi, H., Zhou, F., Shi, B., 2021. Underground coal fire emission of spontaneous combustion, Sandaoba coalfield in Xinjiang, China: Investigation and analysis. Science of The Total Environment 777, 146080. https://doi.org/10.1016/j.scitotenv.2021.146080 Du, X., Sun, D., Li, F., Tong, J., 2022. A Study on the Propagation Trend of Underground Coal Fires Based on Night-Time Thermal Infrared Remote Sensing Technology. Sustainability 14, 14741. https://doi.org/10.3390/su142214741 Dwyer, J.L., Roy, D.P., Sauer, B., Jenkerson, C.B., Zhang, H.K., Lymburner, L., 2018. Analysis Ready Data: Enabling Analysis of the Landsat Archive. Remote Sens. 10, 1363. https://doi.org/10.3390/rs10091363 Galve, J.M., Sánchez, J.M., García-Santos, V., González-Piqueras, J., Calera, A., Villodre, J., 2022. Assessment of Land Surface Temperature Estimates from Landsat 8-TIRS in A High-Contrast Semiarid Agroecosystem. Algorithms Intercomparison. Remote Sensing 14, 1843. https://doi.org/10.3390/rs14081843 He, X., Yang, X., Luo, Z., Guan, T., 2020. Application of unmanned aerial vehicle (UAV) thermal infrared remote sensing to identify coal fires in the Huojitu coal mine in Shenmu city, China. Sci Rep 10, 13895. https://doi.org/10.1038/s41598-020-70964-5 Hooper, A., Bekaert, D., Spaans, K., Arıkan, M., 2012. Recent advances in SAR interferometry time series analysis for measuring crustal deformation. Tectonophysics 514–517, 1–13. https://doi.org/10.1016/j.tecto.2011.10.013 Jiang, L., Lin, H., Ma, J., Kong, B., Wang, Y., 2011. Potential of small-baseline SAR interferometry for monitoring land subsidence related to underground coal fires: Wuda (Northern China) case study. Remote Sensing of Environment 115, 257–268. https://doi.org/10.1016/j.rse.2010.08.008 Jiang, W., Wu, J., Gu, L., Yang, B., Chen, Q., Liu, X., 2011. Monitoring Method of Underground Coal Fire Based on Night Thermal Infrared Remote Sensing Technology. Spectrosc. Spectr. Anal. 31, 357–361. https://doi.org/10.3964/j.issn.1000-0593(2011)02-0357-05 Kelly, J., Kljun, N., Olsson, P.-O., Mihai, L., Liljeblad, B., Weslien, P., Klemedtsson, L., Eklundh, L., 2019. Challenges and Best Practices for Deriving Temperature Data from an Uncalibrated UAV Thermal Infrared Camera. Remote Sens. 11, 567. https://doi.org/10.3390/rs11050567 Kuenzer, C., Zhang, J., Tetzlaff, A., van Dijk, P., Voigt, S., Mehl, H., Wagner, W., 2007. Uncontrolled coal fires and their environmental impacts: Investigating two arid mining regions in north-central China. Appl. Geogr. 27, 42–62. https://doi.org/10.1016/j.apgeog.2006.09.007 Li, F., Yang, W., Liu, X., Sun, G., Liu, J., 2018. Using high-resolution UAV-borne thermal infrared imagery to detect coal fires in Majiliang mine, Datong coalfield, Northern China. Remote Sens. Lett. 9, 71–80. https://doi.org/10.1080/2150704X.2017.1392632 Liang, Y., Yang, Y., Guo, S., Tian, F., Wang, S., 2023. Combustion mechanism and control approaches of underground coal fires: a review. Int J Coal Sci Technol 10, 24. https://doi.org/10.1007/s40789-023-00581-w Liu, J., Wang, Y., Li, Y., Dang, L., Liu, X., Zhao, H., Yan, S., 2019. Underground Coal Fires Identification and Monitoring Using Time-Series InSAR With Persistent and Distributed Scatterers: A Case Study of Miquan Coal Fire Zone in Xinjiang, China. IEEE Access 7, 164492–164506. https://doi.org/10.1109/ACCESS.2019.2952363 Martha, T.R., Guha, A., Kumar, K.V., Kamaraju, M.V.V., Raju, E.V.R., 2010. Recent coal-fire and land-use status of Jharia Coalfield, India from satellite data. Int. J. Remote Sens. 31, 3243–3262. https://doi.org/10.1080/01431160903159340 Masek, J.G., Wulder, M.A., Markham, B., McCorkel, J., Crawford, C.J., Storey, J., Jenstrom, D.T., 2020. Landsat 9: Empowering open science and applications through continuity. Remote Sens. Environ. 248, 111968. https://doi.org/10.1016/j.rse.2020.111968 McKenna, P., Erskine, P.D., Lechner, A.M., Phinn, S., 2018. Measuring fire severity using UAV imagery in semi-arid Central Queensland, Australia (vol 38, pg 4244, 2017). Int. J. Remote Sens. 39, 4285–4285. https://doi.org/10.1080/01431161.2018.1443780 Meghraj, K.C., Leigh, L., Pinto, C.T., Kaewmanee, M., 2023. Method of Validating Satellite Surface Reflectance Product Using Empirical Line Method. Remote Sens. 15, 2240. https://doi.org/10.3390/rs15092240 Minx, J.C., Lamb, W.F., Andrew, R.M., Canadell, J.G., Crippa, M., Doebbeling, N., Forster, P.M., Guizzardi, D., Olivier, J., Peters, G.P., Pongratz, J., Reisinger, A., Rigby, M., Saunois, M., Smith, S.J., Solazzo, E., Tian, H., 2021. A comprehensive and synthetic dataset for global, regional, and national greenhouse gas emissions by sector 1970–2018 with an extension to 2019. Earth Syst. Sci. Data 13, 5213–5252. https://doi.org/10.5194/essd-13-5213-2021 Mishra, R.K., Bahuguna, P.P., Singh, V.K., 2011. Detection of coal mine fire in Jharia Coal Field using Landsat-7 ETM + data. Int. J. Coal Geol. 86, 73–78. https://doi.org/10.1016/j.coal.2010.12.010 Ramzan, U., Fan, H., Aeman, H., Ali, M., A. A. Al-qaness, M., 2022. Combined analysis of PS-InSAR and hypsometry integral (HI) for comparing seismic vulnerability and assessment of various regions of Pakistan. Sci Rep 12, 22423. https://doi.org/10.1038/s41598-022-26159-1 Riyas, M.J., Syed, T.H., Kumar, H., Kuenzer, C., 2021. Detecting and Analyzing the Evolution of Subsidence Due to Coal Fires in Jharia Coalfield, India Using Sentinel-1 SAR Data. Remote Sens. 13, 1521. https://doi.org/10.3390/rs13081521 Roy, D., Singh, G., Seo, Y.-C., 2019. Coal mine fire effects on carcinogenicity and non-carcinogenicity human health risks. Environmental Pollution 254, 113091. https://doi.org/10.1016/j.envpol.2019.113091 Roy, P., Guha, A., Kumar, K.V., 2015. An approach of surface coal fire detection from ASTER and Landsat-8 thermal data: Jharia coal field, India. Int. J. Appl. Earth Obs. Geoinf. 39, 120–127. https://doi.org/10.1016/j.jag.2015.03.009 Schroeder, W., Oliva, P., Giglio, L., Quayle, B., Lorenz, E., Morelli, F., 2016. Active fire detection using Landsat-8/OLI data. Remote Sens. Environ. 185, 210–220. https://doi.org/10.1016/j.rse.2015.08.032 Selvakumaran, S., Plank, S., Geiß, C., Rossi, C., Middleton, C., 2018. Remote monitoring to predict bridge scour failure using Interferometric Synthetic Aperture Radar (InSAR) stacking techniques. International Journal of Applied Earth Observation and Geoinformation 73, 463–470. https://doi.org/10.1016/j.jag.2018.07.004 Song, Z., Kuenzer, C., 2014. Coal fires in China over the last decade: A comprehensive review. International Journal of Coal Geology 133, 72–99. https://doi.org/10.1016/j.coal.2014.09.004 Tian, H., Tao, Y., Kou, P., Alonso, A., Luo, X., Gong, C., Fan, Y., Lei, C., Gou, Y., 2023. Monitoring and evaluation of gully erosion in China’s largest loess tableland based on SBAS-InSAR. Nat Hazards 117, 2435–2454. https://doi.org/10.1007/s11069-023-05950-x Torres, R., Snoeij, P., Geudtner, D., Bibby, D., Davidson, M., Attema, E., Potin, P., Rommen, B., Floury, N., Brown, M., Traver, I.N., Deghaye, P., Duesmann, B., Rosich, B., Miranda, N., Bruno, C., L’Abbate, M., Croci, R., Pietropaolo, A., Huchler, M., Rostan, F., 2012. GMES Sentinel-1 mission. Remote Sens. Environ. 120, 9–24. https://doi.org/10.1016/j.rse.2011.05.028 Wang, T., Shi, J., Ma, Y., Husi, L., Comyn-Platt, E., Ji, D., Zhao, T., Xiong, C., 2019. Recovering Land Surface Temperature Under Cloudy Skies Considering the Solar-Cloud-Satellite Geometry: Application to MODIS and Landsat-8 Data. J. Geophys. Res.-Atmos. 124, 3401–3416. https://doi.org/10.1029/2018JD028976 Wang, Z., Zhou, J., Liu, S., Li, M., Zhang, X., Huang, Z., Dong, W., Ma, J., Ai, L., 2022. A Land Surface Temperature Retrieval Method for UAV Broadband Thermal Imager Data. IEEE Geosci. Remote Sens. Lett. 19. https://doi.org/10.1109/LGRS.2021.3100586 Yan, S., Shi, K., Li, Y., Liu, J., Zhao, H., 2020. Integration of satellite remote sensing data in underground coal fire detection: A case study of the Fukang region, Xinjiang, China. Front. Earth Sci. 14, 1–12. https://doi.org/10.1007/s11707-019-0757-9 Yu, B., She, J., Liu, G., Ma, D., Zhang, R., Zhou, Z., Zhang, B., 2022. Coal fire identification and state assessment by integrating multitemporal thermal infrared and InSAR remote sensing data: A case study of Midong District, Urumqi, China. ISPRS Journal of Photogrammetry and Remote Sensing 190, 144–164. https://doi.org/10.1016/j.isprsjprs.2022.06.007 Zhang, H., Thanh, H.V., Han, F., Wang, Y., Zhang, Xun, Zhao, R., Sasaki, K., Zhang, Xiaoming, 2023. Development of an in-situ gel from CO2-captured complex solution and inhibiting coal spontaneous combustion: A case study in thermal engineering. Case Studies in Thermal Engineering 50, 103423. https://doi.org/10.1016/j.csite.2023.103423 Zhang, P., Qian, X., Guo, S., Wang, B., Xia, J., Zheng, X., 2023. A New Method for Continuous Track Monitoring in Regions of Differential Land Subsidence Rate Using the Integration of PS-InSAR and SBAS-InSAR. Remote Sensing 15, 3298. https://doi.org/10.3390/rs15133298 Zhang, Z., Hu, C., Wu, Z., Zhang, Zhen, Yang, S., Yang, W., 2023. Monitoring and analysis of ground subsidence in Shanghai based on PS-InSAR and SBAS-InSAR technologies. Sci Rep 13, 8031. https://doi.org/10.1038/s41598-023-35152-1 Zhou, L., Zhang, D., Wang, J., Huang, Z., Pan, D., 2013. Mapping Land Subsidence Related to Underground Coal Fires in the Wuda Coalfield (Northern China) Using a Small Stack of ALOS PALSAR Differential Interferograms. Remote Sensing 5, 1152–1176. https://doi.org/10.3390/rs5031152 Zeng, Q.: Causes, Monitoring, Extinction, and Eco-environmental Impacts of Underground Coal Fires: A Comprehensive Perspective, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-914, https://doi.org/10.5194/egusphere-egu23-914, 2023. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4856299","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":347804011,"identity":"f7a9eb18-9a77-4d73-bbe4-3f20ff7c9079","order_by":0,"name":"Zhicheng Yang","email":"","orcid":"","institution":"Xinjiang University","correspondingAuthor":false,"prefix":"","firstName":"Zhicheng","middleName":"","lastName":"Yang","suffix":""},{"id":347804012,"identity":"b7d1baa1-2133-475c-be27-93c85a151fd1","order_by":1,"name":"Qiang Zeng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYNACAyBmbwBTjA3Ea+E5QJIWEJBIAFOEtci39x5+8abALk8+8o1BMQ+DjeyGA8zPHuDTwthzLs1yjkFyseHttARjHoY04w0H2MwN8GlhlsgxM+YxYE7cODv5AFDL4cQNB3jYJPBpYZN/A9JSn7hx5sEGoJb/hLXwSPAYP+YxOJw4X4IZZMsBwlokeHLMGOcYHE/cwJOWYAj0lPHMw2xmeLXIt58x/vDmT3Xi/PYzZgZvKuxk+443P8OrBeQdCR4gaXCAgc0AHJnMBNSDlHwAaZFvYGB+QFjxKBgFo2AUjEQAACFQRPePg78rAAAAAElFTkSuQmCC","orcid":"","institution":"Xinjiang University","correspondingAuthor":true,"prefix":"","firstName":"Qiang","middleName":"","lastName":"Zeng","suffix":""}],"badges":[],"createdAt":"2024-08-04 09:53:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4856299/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4856299/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":64711317,"identity":"c011dc7a-1c9e-44cc-b7d9-632476b75c33","added_by":"auto","created_at":"2024-09-18 01:58:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2349836,"visible":true,"origin":"","legend":"\u003cp\u003eOverview map of the study area: (a) Geographic location map, (b) Elevation sketch map, and (c) Visible-light image of the study area. The red area in (c) indicates the extent of the fire zone.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/9af115fb13b88adcefd1e9ac.png"},{"id":64712689,"identity":"3e865bcc-60f4-4097-92ed-f27550ceed64","added_by":"auto","created_at":"2024-09-18 02:06:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":127145,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal and perpendicular baselines for SBAS and PS methods (2018-2024).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/8853265a32d898a35044342d.png"},{"id":64710401,"identity":"f8318a59-9d79-4268-a988-fc838813efa6","added_by":"auto","created_at":"2024-09-18 01:50:43","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":37266,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow of SBAS-InSAR and PS-InSAR\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/0b9e10b241c053256ad6c1cd.jpg"},{"id":64710406,"identity":"5e20d1f0-9e6a-4881-a72a-72667439daef","added_by":"auto","created_at":"2024-09-18 01:50:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":104186,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency superposition for time series images to extract coal fire pixels\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/dff31af88c5a4d9ba586fcb4.png"},{"id":64710403,"identity":"4fbdb4f1-1a41-4501-8164-d6ba18dc35b5","added_by":"auto","created_at":"2024-09-18 01:50:43","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1333126,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 4.\u003c/strong\u003e Fractional Vegetation Cover Inversion Results from 2013 to 2023 (a-f)\u003c/p\u003e","description":"","filename":"41.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/b5b36dcea835385b6f5850b1.png"},{"id":64710407,"identity":"10456f34-c9d4-43f7-a3fb-bf4a5fc03874","added_by":"auto","created_at":"2024-09-18 01:50:44","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1208232,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 5.\u003c/strong\u003e Temperature inversion results for 2013-2023 (a-f)\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/d0a6a62865d1760380922f9d.png"},{"id":64710405,"identity":"290c6793-674e-41b4-aec6-5c90009c8333","added_by":"auto","created_at":"2024-09-18 01:50:44","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":332814,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 6.\u003c/strong\u003e Change curves of sparse vegetation cover area and high-temperature anomaly area; Overlapping areas of sparse vegetation cover and high-temperature anomalies.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/db7bf52dd4e8034909475853.png"},{"id":64712700,"identity":"09f9a7ec-868e-4b58-9a36-3b9d856e8df1","added_by":"auto","created_at":"2024-09-18 02:06:44","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1073653,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 7.\u003c/strong\u003e Time Series of Cumulative Surface Subsidence.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/855508d5fcde3b1b184fe3d2.png"},{"id":64711319,"identity":"0288bd0c-2f7b-41e0-8b2b-7c821d060a92","added_by":"auto","created_at":"2024-09-18 01:58:44","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":495605,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 8.\u003c/strong\u003e Ground deformation monitoring results: (a) Deformation rates obtained using SBAS; (b) Deformation rates obtained using PS; (c) Time series deformation curves for selected pixels, showing consistency between SBAS and PS methods.\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/786023e3156b13b1ebe56190.png"},{"id":64712690,"identity":"e5eeb5af-6494-4285-b5e2-dbb4aa198f56","added_by":"auto","created_at":"2024-09-18 02:06:44","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":55301,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 9.\u003c/strong\u003e Correlation analysis between SBAS and PS (a); Histogram of deformation rate differences between SBAS and PS for the same pixel points (b).\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/5c42c84c043513743520f26c.png"},{"id":64710411,"identity":"8f7add3a-2ec1-4737-ad06-d0853ee8a2ef","added_by":"auto","created_at":"2024-09-18 01:50:44","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":1573133,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 10.\u003c/strong\u003e Illustration of subsidence and overlaid characteristic points in the study area.\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/3d52be696f5b711b3c2ef048.png"},{"id":64711321,"identity":"0d0fc533-2b5c-4fa1-97bf-cbb1c1377f6d","added_by":"auto","created_at":"2024-09-18 01:58:44","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":135237,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 11.\u003c/strong\u003e Overlaid time series of vertical cumulative deformation, LST, and FVC for the 9 characteristic points. Blue represents the vertical cumulative deformation rate, red represents LST, and green represents FVC.\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/efef88ad237e464c04e6edfd.png"},{"id":64710410,"identity":"ffc7892a-72a2-4f73-b62a-9f963a41dab8","added_by":"auto","created_at":"2024-09-18 01:50:44","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":40145,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 12.\u003c/strong\u003e Box plots of LST, FVC, and surface deformation for the 9 characteristic points.\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/ec5c0fee7b526436e8390d72.png"},{"id":65197779,"identity":"7afc6639-fa3d-4b27-a36a-ed5829969e51","added_by":"auto","created_at":"2024-09-24 15:47:15","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8441250,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4856299/v1/1e2bed0d-5cdf-46ee-a2b7-97af78ffc1c8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Application of Multi-Source Remote Sensing Fusion for Identifying Smoke Fugitive Channels in the Sulabulak Fire Area, Urumqi, China","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCoal field fires, as a long-term and complex environmental and safety issue worldwide, pose severe challenges, especially in coal-rich regions (Zeng, 2023). Underground coal fires not only result in significant economic losses but also have widespread impacts on surface ecosystems and the atmosphere (Biswal et al., 2019; Liang et al., 2023). The release of smoke from coal field fires, particularly carbon dioxide and other harmful gases, exacerbates the severity of global climate change (Carroll et al., 2022; Roy et al., 2019; H. Zhang et al., 2023). In China, carbon dioxide emissions from coal fires account for 0.32% of the global emissions, while global coal fire emissions contribute to 0.42% of the total global emissions (Minx et al., 2021). Therefore, accurate monitoring and assessment of the dynamic changes in coal fire areas and their smoke fugitive channels are crucial for implementing effective fire control measures and achieving China's \"carbon peak\" and \"carbon neutrality\" goals.\u003c/p\u003e \u003cp\u003ePrevious studies have employed various techniques for detecting and monitoring coal fires, including thermal infrared (TIR), multispectral, and hyperspectral remote sensing, as well as unmanned aerial vehicles (UAVs). (He et al., 2020; Yan et al., 2020). These methods, while providing valuable insights, have limitations in accurately identifying smoke fugitive channels due to issues like data acquisition range, cost, sensitivity to weather conditions, and spatial resolution constraints. (Wang et al., 2022). For example, UAVs offer high-resolution monitoring but are limited by their operational range and sensitivity to weather conditions (Kelly et al., 2019; McKenna et al., 2018). Slavecki RJ first applied thermal infrared remote sensing technology to coal fire detection (Biswal et al., 2019). This method is favored for its non-contact nature, low cost, time efficiency, and wide spatial coverage (Chen et al., 2022; Du et al., 2022; Kuenzer et al., 2007). It utilizes thermal infrared data to obtain land surface temperature (LST), then identifies thermal anomalies associated with coal fires using specific criteria. Although it can quickly identify thermal anomalies and is widely used, the results are relatively coarse and ineffective for detecting thermal anomalies in summer, particularly in deep underground coal fire areas (Martha et al., 2010; Mishra et al., 2011; Roy et al., 2015). This indicates that using a single remote sensing image is insufficient for accurately detecting thermal anomalies and does not meet the need for comprehensive identification of coal fires and their smoke fugitive channels.\u003c/p\u003e \u003cp\u003eAs coal field fires spread deeper underground, their concealment increases, posing challenges to traditional thermal anomaly detection methods (Yan et al., 2020). To more accurately identify coal fires and their smoke fugitive channels, integrated satellite remote sensing methods have become a research hotspot (Liu et al., 2019; Schroeder et al., 2016; Yu et al., 2022). InSAR technology can effectively detect land subsidence, providing a new perspective for identifying coal fire areas (Aditiya \u0026amp; Ito, 2023; Hooper et al., 2012). These techniques can reveal surface deformations caused by coal fire combustion, offering valuable information even when initial subsidence is not apparent. To date, researchers have successfully combined PS-InSAR and SBAS-InSAR methods with global positioning systems and field survey data, comprehensively verifying the critical role of these methods in identifying coal fires (Riyas et al., 2021; Wang et al., 2019).\u003c/p\u003e \u003cp\u003eHowever, since surface subsidence can be caused by various factors, including mining activities and the coal fire itself, relying solely on subsidence data makes it challenging to accurately distinguish and identify coal fires. Given these limitations, there is a need for a more integrated approach that combines multiple remote sensing techniques to improve the accuracy and reliability of smoke fugitive channel detection. This study aims to fill this gap by proposing a novel comprehensive analysis method that integrates multi-source remote sensing data, including Landsat-8 and Sentinel-1A, to detect and monitor smoke fugitive channels caused by coal fires. By employing this method, we aim to more effectively monitor coal fire activities under different seasonal and climatic conditions, accurately locate smoke fugitive channels, and provide a scientific basis for formulating fire extinguishing measures and calculating carbon emissions in coal fire areas.\u003c/p\u003e \u003cp\u003eThe structure of this paper is as follows: Section 2 describes the study area and datasets. Section 3 outlines the research methodology. Section 4 presents the experimental results and analysis. Section 5 discusses the findings and validates the proposed method. Finally, Section 6 concludes the paper and suggests directions for future research.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003eStudy Area\u003c/p\u003e\n\u003cp\u003eThe Sulabulak fire area is located in the north-central region of Xinjiang, China, approximately 86 km south of Urumqi(Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). This area lies between latitudes 87\u0026deg;27\u0026prime; and 87\u0026deg;32\u0026prime; and longitudes 43\u0026deg;20\u0026prime; and 43\u0026deg;23\u0026prime;. The terrain is characterized by mid-low mountain hills, with elevations ranging from 1995 to 2211 meters, and relative height differences of 50 to 110 meters. The region has a continental climate with annual precipitation between 170.4 and 201.1 mm, and predominantly experiences southwest winds. The geological structures of the Sulabulak fire area include the Jurassic Xishanyao Formation (J2x) and Quaternary sediments. The coal seams, which are highly flammable due to their low sulfur, low phosphorus, high calorific value, and high melting temperature, consist mainly of black, blocky, or layered coal with well-developed joints and fractures. Coal fires were discovered during field surveys in 2016, primarily caused by incomplete stripping and underground mining that exposed shallow coal seams to spontaneous combustion. These fires have progressively worsened, causing significant environmental damage and economic losses.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDatasets\u003c/p\u003e\n\u003cp\u003eThis study utilized 29 Landsat Collection 2 Level 2 images from 2013 to 2023. Released by the US Geological Survey (USGS), this dataset comprises enhanced Earth observation products from the Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) on Landsat 8 and Landsat 9 satellites (Masek et al., 2020). These datasets, featuring surface reflectance and temperature products, are processed with advanced radiative transfer models and supplementary atmospheric data to enhance accuracy and usability (Claverie et al., 2018). Available in image formats of approximately 185 km by 180 km, the data utilize Universal Transverse Mercator (UTM) projection, or Polar Stereographic (PS) projection for polar regions (Dwyer et al., 2018). Stored as 16-bit unsigned integers, the datasets include metadata for converting digital numbers (DN) to radiance, reflectance, and brightness temperature. These datasets, available via the USGS Earth Explorer portal, USGS EROS Machine-to-Machine (M2M) API, and AWS Simple Storage Service (S3), offer direct access for detailed analysis (Crawford et al., 2023).\u003c/p\u003e\n\u003cp\u003eThis study utilizes Sentinel-1 data from the European Space Agency (ESA). Sentinel-1, a C-band satellite with a 12-day revisit cycle (Torres et al., 2012), provided 135 images acquired between January 13, 2018, and December 13, 2023. The standard acquisition mode was Interferometric Wide (IW) with dual-polarization (VV/VH). Precise orbit corrections used data from the ESA Sentinel-1 Quality Control website. Additionally, 30-meter spatial resolution SRTM data served as external reference digital elevation models (DEMs) for InSAR processing. For SBAS-InSAR processing, 428 interferograms were generated, with temporal baselines ranging from 12 to 60 days and the longest spatial baseline being 122.1 meters. The average baseline length for PS-InSAR was \u0026minus;\u0026thinsp;157.2 meters. Temporal and spatial baseline graphs from SBAS-InSAR and PS-InSAR processing are shown in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003cp\u003eVegetation cover retrieve\u003c/p\u003e\n\u003cp\u003eAccurate vegetation cover estimation is crucial for understanding the impact of coal fires on the local environment, particularly in identifying smoke fugitive channels. Vegetation cover inversion leverages the spectral absorption and reflection characteristics of vegetation across different spectral bands (Chen et al., 2024). This study employs the pixel dichotomy method to extract the Fractional Vegetation Cover (FVC) for the study area over the period from 2013 to 2023, reflecting the vegetation coverage status in the region.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Normalized Difference Vegetation Index (NDVI) is calculated using the near-infrared and red bands of remote sensing data. Based on this, the Fractional Vegetation Cover (FVC) is calculated using the pixel dichotomy model. The formulas are as follows:\u003c/p\u003e\n\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equa\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}NDVI=\\frac{{R}_{NI}-R}{{R}_{NI}+R} \\left(1\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equb\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}FVC=\\frac{NDVI-NDV{I}_{soil}}{NDV{I}_{veg}-NDV{I}_{soil}} \\left(2\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003ewhere \u003cem\u003eNDVI\u003c/em\u003e is the Normalized Difference Vegetation Index, \u003cem\u003eR\u003c/em\u003e\u003csub\u003e\u003cem\u003eNI\u003c/em\u003e\u003c/sub\u003e is the spectral value of the near-infrared band, and \u003cem\u003eR\u003c/em\u003e is the spectral value of the red band; \u003cem\u003eNDVI\u003c/em\u003e\u003csub\u003e\u003cem\u003esoil\u003c/em\u003e\u003c/sub\u003e represents the \u003cem\u003eNDVI\u003c/em\u003e of bare soil; and \u003cem\u003eNDVI\u003c/em\u003e\u003csub\u003e\u003cem\u003eveg\u003c/em\u003e\u003c/sub\u003e represents the \u003cem\u003eNDVI\u003c/em\u003e of fully vegetated areas. The values for \u003cem\u003eNDVI\u003c/em\u003e\u003csub\u003e\u003cem\u003esoil\u003c/em\u003e\u003c/sub\u003e and \u003cem\u003eNDVI\u003c/em\u003e\u003csub\u003e\u003cem\u003eveg\u003c/em\u003e\u003c/sub\u003e are selected from the 5%-95% confidence interval.\u003c/p\u003e\n\u003cp\u003eAfter obtaining the FVC values, the vegetation cover for each period is reclassified using ArcGIS software. This reclassification is based on the field vegetation conditions in the area, ensuring that the remote sensing data aligns with ground truth observations. The number of pixels and the area for each interval are then statistically analyzed to understand the temporal dynamics of vegetation cover.\u003c/p\u003e\n\u003cp\u003eThis approach enables a detailed analysis of vegetation cover changes over time, providing insights into the environmental impact of coal fires and aiding in the identification of smoke fugitive channels.\u003c/p\u003e\n\u003cp\u003eTemperature retrieve\u003c/p\u003e\n\u003cp\u003eIn this study, we used ENVI software to monitor land surface temperature (LST) with products from the United States Geological Survey (USGS) Landsat-8 Collection 2 Level 2. The LST data is captured by the Thermal Infrared Sensor (TIRS) on the Landsat 8 satellite and undergoes preprocessing, including atmospheric correction and surface emissivity correction (Galve et al., 2022). The LST calculation process follows these steps:\u003c/p\u003e\n\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equc\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}L\\left(\\lambda\\:\\right)={M}_{L}\\times\\:{Q}_{cal}+{A}_{L} \\left(3\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equd\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}T=\\frac{{K}_{2}}{\\text{ln}\\left(\\frac{{K}_{1}}{L\\left(\\lambda\\:\\right)}+1\\right)} \\left(4\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Eque\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}{T}_{s}=\\frac{T}{1+\\left(\\lambda\\:\\times\\:\\frac{T}{\\rho\\:}\\right)\\times\\:\\text{ln}\\left(ϵ\\right)} \\left(5\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eThe digital number (DN) is first converted to spectral radiance L(\u0026lambda;). In formula 3, \u003cem\u003eM\u003c/em\u003e\u003csub\u003e\u003cem\u003eL\u003c/em\u003e\u003c/sub\u003e is the radiance multiplicative factor, \u003cem\u003eQ\u003c/em\u003e\u003csub\u003e\u003cem\u003ecal\u003c/em\u003e\u003c/sub\u003e is the calibrated pixel value (DN), and \u003cem\u003eA\u003c/em\u003e\u003csub\u003e\u003cem\u003eL\u003c/em\u003e\u003c/sub\u003e is the radiance additive factor. Using the Planck equation in formula 4, we calculate the brightness temperature. \u003cem\u003eT\u003c/em\u003e is the absolute temperature (in Kelvin), and \u003cem\u003eK1\u003c/em\u003e and \u003cem\u003eK2\u003c/em\u003e are pre-launch calibration constants. Formula 5 uses a modified Planck equation to convert the radiance back to actual temperature, where \u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003eS\u003c/em\u003e\u003c/sub\u003e is the land surface temperature (in Kelvin), \u003cem\u003e\u0026lambda;\u003c/em\u003e is the wavelength (in meters), and \u003cem\u003e\u0026rho;\u003c/em\u003e is defined as \u003cem\u003eh*c/\u0026sigma;\u003c/em\u003e, with \u003cem\u003eh\u003c/em\u003e being the Planck constant, \u003cem\u003ec\u003c/em\u003e the speed of light, and \u003cem\u003e\u0026sigma;\u003c/em\u003e the Stefan-Boltzmann constant; \u003cem\u003eϵ\u003c/em\u003e is the surface emissivity (Meghraj et al., 2023).\u003c/p\u003e\n\u003cp\u003eInSAR deformation retrieval\u003c/p\u003e\n\u003cp\u003eSBAS-InSAR is an advanced time-series Synthetic Aperture Radar (InSAR) analysis method that constructs short-time baseline differential interferograms by selecting and combining collected data. This approach efficiently overcomes spatial decorrelation by selecting target points based on spatial coherence coefficients and building a linear model. Using Singular Value Decomposition (SVD) or the Least Squares Method (LSM), linear deformation rates and elevation error values for the target points can be obtained, resulting in a comprehensive deformation time series over the entire observation period (Selvakumaran et al., 2018). SBAS-InSAR technology excels in processing long-term observational data and monitoring large-area deformation, making it especially suitable for detailed ground deformation studies.\u003c/p\u003e\n\u003cp\u003ePersistent Scatterer Interferometry (PS-InSAR) is a technique used for monitoring ground subsidence and deformation across various regions. It identifies stable scatterers, known as Persistent Scatterers (PS), which exhibit consistent phase behavior over long periods. PS-InSAR combines high precision and efficiency, enabling continuous monitoring of ground deformation, particularly useful for detecting slow-moving deformations. This technique provides high temporal resolution deformation information, offering detailed measurements of ground subsidence and deformation (P. Zhang et al., 2023).\u003c/p\u003e\n\u003cp\u003eWhen used together, PS-InSAR and SBAS-InSAR allow tracking of various subsidence rates, significantly enhancing monitoring density and accuracy (Ramzan et al., 2022). The technical workflows for SBAS-InSAR and PS-InSAR are illustrated in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e. These complementary methods provide thorough and precise monitoring of ground deformation, essential for comprehending and managing the dynamics and effects of underground coal fires.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDetailed processing methods for SBAS and PS techniques can be found in (Tian et al., 2023; Z. Zhang et al., 2023). These methods provide robust technical support for studying smoke fugitive channels and surface deformation in underground coal fire areas, essential for precise monitoring and developing effective mitigation measures.\u003c/p\u003e\n\u003cp\u003eIntegrated Smoke Channel Detection\u003c/p\u003e\n\u003cp\u003ePrevious research shows some overlap between temperature anomalies and surface deformation (Zhou et al., 2013; L. Jiang et al., 2011). However, temperature anomalies caused by underground coal fires exhibit temporal lag, non-linear characteristics, and significant seasonal variation (Song \u0026amp; Kuenzer, 2014). Directly overlaying these results can lead to data omissions and errors.\u003c/p\u003e\n\u003cp\u003eTo address this issue, this study integrates high-temperature threshold extraction and analysis of sparse to moderately sparse vegetation coverage, supplemented with surface deformation data, to verify and accurately locate smoke fugitive channels in coal fire areas.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFigure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e. Frequency superposition for time series images to extract coal fire pixels\u003c/p\u003e\n\u003cp\u003eThe high-temperature threshold (\u003cem\u003eTh\u003c/em\u003e) is derived from temperature inversion data. It is defined as the sum of the mean surface temperature (\u003cem\u003eTm\u003c/em\u003e) and twice the standard deviation (\u003cem\u003eTsd\u003c/em\u003e), serving as the optimal threshold to distinguish between coal fire areas and non-coal fire areas (W. Jiang et al., 2011), as shown in formula 6. To obtain more accurate temperature threshold information for coal fire areas, we incorporated thermal interference data from non-coal fire regions collected through field surveys, along with the location and thermal information of actual coal fire boundaries. We conducted a comparative filtering experiment to extract fire area ranges at different time series. As illustrated in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e, the high-temperature threshold attribute was set to 1, while other areas were set to 0, and then the data were superimposed. A pixel was identified as an abnormal temperature region when its frequency of appearance in the same location and within the set period reached the predefined high-frequency filtering threshold.\u003c/p\u003e\n\u003cdiv id=\"Equf\" class=\"Equation\"\u003e\n\u003cdiv id=\"FileID_Equf\" class=\"mathdisplay\"\u003e$$\\:\\begin{array}{c}{T}_{h}={T}_{m}+2{T}_{Sd} \\left(6\\right)\\end{array}$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eBy employing this method, this study can more accurately identify areas affected by underground coal fires, thereby providing a more reliable scientific basis for the monitoring and management of coal fire disasters.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Result","content":"\u003cp\u003eVegetation retrieve\u003c/p\u003e \u003cp\u003eAfter conducting field surveys in the Sulabulak fire area and assessing the local vegetation growth, the derived vegetation coverage was classified into different levels as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e: sparse vegetation (0\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026thinsp;\u0026lt;\u0026thinsp;0.3), moderately sparse vegetation (0.3\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026thinsp;\u0026lt;\u0026thinsp;0.5), dense vegetation (0.5\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026thinsp;\u0026lt;\u0026thinsp;0.7), and very dense vegetation (0.7\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026thinsp;\u0026lt;\u0026thinsp;1). After processing with ArcGIS, part of the vegetation inversion results are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eClassification of FVC\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFVC classify\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFVC Value Ranges\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSparse Vegetation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026lt;0.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerately Sparse Vegetation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026lt;0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDense Vegetation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.5\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026lt;0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVery Dense Vegetation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.7\u0026thinsp;\u0026le;\u0026thinsp;FVC\u0026lt;1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe processed results, depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, indicate that low vegetation cover areas are mainly distributed in the southern and southeastern parts of the study area, while high vegetation cover areas are located in the western valleys and their sides. Over the study period from 2013 to 2023, there was a notable increase in the area of sparse and moderately sparse vegetation. Statistical analysis shows that the proportion of sparse vegetation cover decreased from 4.8\u0026ndash;4.3% (a reduction of 0.11 km\u0026sup2;), while moderately sparse vegetation cover increased from 15.9\u0026ndash;26.1% (an area increase of 2.88 km\u0026sup2;). Dense vegetation cover also increased from 34.5\u0026ndash;37.1% (an area increase of 0.55 km\u0026sup2;). Conversely, the proportion of very dense vegetation cover decreased from 44.9\u0026ndash;32.6% (an area decrease of 2.60 km\u0026sup2;).\u003c/p\u003e \u003cp\u003eTemperature retrieve and thermal anomaly detection\u003c/p\u003e \u003cp\u003eThe heat from underground coal fires transfers to the surface, raising surface temperatures in coal fire areas above those of the surrounding environment. Surface temperature inversion of remote sensing images from 2013 to 2023 was performed, with some results displayed in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAnalysis of Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e reveals that high-temperature areas are distributed linearly from northeast to southwest, while low-temperature areas are primarily located in the southern and southwestern mountain valleys. During the study period from 2013 to 2023, the highest retrieved temperature was 36.68\u0026deg;C, and the lowest was \u0026minus;\u0026thinsp;20.13\u0026deg;C. Using the high-temperature anomaly threshold, retrieved temperatures for each period were reclassified in ArcGIS to extract areas exceeding the temperature threshold. The results indicate that from 2013 to 2023, the area of temperature anomalies increased from 0.5\u0026ndash;3.5%, corresponding to an area increase of 0.65 km\u0026sup2;.\u003c/p\u003e \u003cp\u003eFugitive flue gas channel analysis\u003c/p\u003e \u003cp\u003eThe sparse vegetation cover areas and high-temperature anomaly areas were extracted using ArcGIS software, and their change curves are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. The analysis indicates that the area of sparse vegetation cover increased from 1.49 km\u0026sup2; to 2.77 km\u0026sup2;, while the area of high-temperature anomalies expanded from 0.10 km\u0026sup2; to 0.75 km\u0026sup2;. Both areas show a consistent year-by-year increase.\u003c/p\u003e \u003cp\u003eBased on the results of vegetation cover and surface temperature inversion, the areas with sparse vegetation cover and high-temperature anomalies were overlaid to identify the smoke fugitive channels of coal fire combustion, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. The figure demonstrates that from 2013 to 2023, the overlapping points of sparse vegetation cover and high-temperature anomaly areas exhibited a fluctuating pattern over time. These overlapping areas are predominantly located in the central and eastern parts of the study area. The number of overlapping points was at its lowest in 2017, with only four points, and reached its peak in 2023 with 66 points. This transition from isolated points to broader areas in the overlapping regions indicates an expanding trend of the smoke fugitive channels, suggesting an intensification of coal fire combustion.\u003c/p\u003e \u003cp\u003eDeformation monitoring results and analysis\u003c/p\u003e \u003cp\u003eUsing data from 135 dual-polarization (VV/VH) Sentinel-1A images, covering 72 time periods from January 2018 to December 2023, deformation in the line-of-sight (LOS) was obtained using SBAS and PS methods. The cumulative surface subsidence data provides insights into the temporal\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eevolution, magnitude, subsidence trends, and the distribution of major subsidence areas within the study region. The time series of cumulative surface subsidence for the study area is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. From January 2018 to December 2023, the study area exhibited a maximum cumulative subsidence of -123.9 mm and a maximum cumulative uplift of 48.41 mm, with an increasing trend observed annually.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAs shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e, the time series deformation of pixel points located in the subsidence area (P1 point in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e) using the two methods (SBAS and PS) indicates good consistency. The results of SBAS and PS were imported into ArcGIS, and 20,000 random points were generated for overlay analysis. After removing invalid points, 9,050 overlay points remained for correlation analysis, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e(a). The correlation coefficient (R\u0026sup2;) between the deformation rates obtained from the two monitoring results is 0.89, demonstrating a strong correlation.\u003c/p\u003e \u003cp\u003eThe histogram in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e(b) shows the deformation rate differences between SBAS and PS for the same pixel points. Most differences fall within \u0026plusmn;\u0026thinsp;4 mm/yr, with a standard deviation of 0.75 mm/yr and a mean value of -1.34 mm/yr. This suggests that, although there are slight discrepancies\u003c/p\u003e \u003cp\u003ebetween the two methods, both provide reliable ground deformation monitoring results. Overall, the SBAS monitoring results for surface deformation are robust and can be effectively combined with temperature and vegetation coverage data to identify smoke fugitive channels in coal fire areas. This combined method improves the precision and dependability of coal fire detection and monitoring, offering essential insights for future research and mitigation efforts.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eTo analyze the performance of this method in detecting smoke fugitive channels in different regions, this section overlays the deformation time series, LST, and FVC. Additionally, the impacts of coal fires, mining activities, and other factors were analyzed using LST, FVC, and deformation data from characteristic points in various regions. This analysis reveals the evolutionary characteristics of smoke fugitive channels in different areas.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e shows the identified areas where sparse vegetation coverage and high-temperature anomalies overlap, along with subsidence monitoring data. It can be seen that the subsidence points and the overlapping areas are largely consistent. Nine characteristic points were selected from A1 to D2. The time series data for subsidence, LST, and FVC were extracted and overlaid, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e, with the time series of LST (red line), FVC (green line), and deformation (blue line). The distribution of the characteristic points and field survey data classify them into four categories. The first category includes A1, A2, and A3, located at smoke fugitive channel points in coal fire areas. The second category includes B1 and B2, which are in high-temperature areas with normal vegetation coverage. The third category includes C1 and C2, which are in low vegetation areas with normal temperatures. The fourth category includes D1 and D2, which are control points located far from the fire area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eCharacteristic points A1, A2, and A3 are located in coal fire areas. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e, the ground subsidence at category A points is significant and increases over time, indicating that these areas are severely affected by coal fires. Additionally, the surface temperature (LST) is generally high, particularly outside of the summer months, but weak during summer. Most coal fire anomalies fall into this category, explaining why single thermal infrared imagery is ineffective for detecting coal fires in summer (Song \u0026amp; Kuenzer, 2014). The vegetation cover (FVC) is generally low and shows little variation over time, indicating that the vegetation is severely affected by coal fires and difficult to recover. A significant correlation exists between temperature, FVC, and ground subsidence. The data distribution for category A points shows a large median subsidence with a wide range, indicating significant and unstable ground subsidence in the area(Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e); the LST median is high with large fluctuations, indicating drastic temperature changes; the FVC median is low with a small range, indicating minimal vegetation cover. According to field surveys, this area was affected by spontaneous combustion of residual coal in inadequately mined underground spaces. These areas can be marked as smoke fugitive channels.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBased on field investigations, characteristic points B1 and B2 are covered with coal piles. As coal accumulates, it slowly oxidizes upon contact with air, releasing heat and causing thermal anomalies. However, these coal piles do not combust, resulting in relatively low heat release. The surrounding areas have high vegetation cover with significant variation, likely influenced by seasonal changes. Additionally, ground subsidence was observed at B1 and B2, but the magnitude is small, and the region remains relatively stable. Therefore, it can be concluded that these points are not smoke fugitive channels from coal fires.\u003c/p\u003e \u003cp\u003eCharacteristic point C1 is located in the Xinxing coal mine, and C2 is located in the Shunda coal mine. These areas experience significant human activities, such as mining operations and exploration projects. The low vegetation cover at these points may be due to these activities or other complex environmental factors. The ground subsidence in these areas is minimal, indicating stable surface conditions. The temperature variations are normal and not significantly affected by coal fires. Therefore, in the proposed method, these areas can be excluded from influencing the detection results of smoke fugitive channels caused by coal fires.\u003c/p\u003e \u003cp\u003eCharacteristic points D1 and D2 in category D exhibit the smallest median subsidence and the smallest range of variation, indicating extremely stable surface conditions unaffected by coal fires. The median LST is relatively low with a small range of temperature variation, showing that these areas have stable temperatures without significant high-temperature anomalies. The FVC median is the highest with a large range of variation, indicating good vegetation cover that is healthy and responsive to seasonal changes. According to field surveys, these areas have a stable ecological environment, unaffected by coal fires or other human activities, making them suitable as control points. The stability of category D points provides a baseline reference for understanding environmental changes in areas affected by coal fires.\u003c/p\u003e \u003cp\u003eThe analysis above indicates that the deformation and heat release processes in the study area are complex. The combustion of coal and mining activities alter the underground stress state, contributing to observed subsidence. High-temperature anomalies and sparse vegetation coverage areas include those caused by coal combustion and heat accumulation from other human activities or surface objects. Therefore, identifying smoke fugitive channels caused by coal fires cannot depend on a single data source, as done in previous studies (Deng et al., 2021; He et al., 2020; Li et al., 2018). Although the method proposed in this paper is suitable for both mining and non-mining areas, it still requires thorough analysis and field verification to ensure its accuracy and applicability.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study proposes a novel comprehensive analysis method utilizing multi-source remote sensing technology to detect smoke fugitive channels caused by coal fires. The analysis employed twenty-nine Landsat-8 satellite images from the Sulabulak fire area in China to retrieve vegetation coverage (FVC) and land surface temperature (LST). Sparse vegetation coverage and high-temperature anomaly areas were extracted and overlaid to identify smoke fugitive channels. Additionally, 135 dual-polarized Sentinel-1A images were used to obtain ground deformation data via SBAS-InSAR and PS-InSAR techniques. Field survey data validated the identification of smoke fugitive channels, showing a high degree of agreement at nine characteristic points within the overlapping and subsidence areas. This confirms the method's accuracy in determining the spatial distribution of smoke fugitive channels. From the results of this study, we can draw the following conclusions:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eBy analyzing Landsat-8 and Sentinel-1A data for the last decade, we found that the area of sparse vegetation cover in the Sulabulak fire area has increased in size to 2.77 km\u003csup\u003e2\u003c/sup\u003e, and the area of high-temperature anomalies has expanded to 0.75 km\u003csup\u003e2\u003c/sup\u003e, with a cumulative surface deposition of -123.9 mm. These data indicate an increase in the intensity of burning in the coal fires and an increase in the extent of smoke escape routes.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eCombining surface deformation data from SBAS-InSAR technology verified the effectiveness of remote sensing in detecting smoke fugitive channels, particularly in identifying ground subsidence related to coal fire activities.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eThe analysis showed that ground subsidence is primarily concentrated in coal fire areas and smoke fugitive channels, with a high degree of overlap with high-temperature anomaly areas and sparse vegetation coverage. This suggests that coal fire combustion and the presence of smoke fugitive channels are major factors causing ground subsidence, highlighting the need to comprehensively consider various environmental and human activity factors when identifying and evaluating coal fire regions.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eWhile this method has been successful in the Sulabulak fire area, the complexity of coal field regions necessitates further field validation and detailed analysis to ensure the method's generalizability and accuracy. Future research will extend this method to different types and scales of coal fire areas, considering factors such as terrain, climatic conditions, and human activities. Additionally, based on the identified smoke fugitive channel areas, we will estimate carbon dioxide emissions using coal sample combustion characteristics, soil temperature, and moisture parameters, which will be a focus of our future research.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e \u003cstrong\u003eEthics approval\u003c/strong\u003e \u003cp\u003eAll authors have read, understood, and have complied as applicable with the statement on \u0026ldquo;Ethical responsibilities of Authors\u0026rdquo; as found in the Instructions for Authors.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent to participate\u003c/strong\u003e \u003cp\u003eInformed consent was obtained from all individual participants included in the study.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCompeting interests\u003c/strong\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research was financially supported by Innovation Leading Talent project (grant no.2023TSYCLJ0003) funded by the Xinjiang Department of Science and Technology of China.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eZhicheng Yang：Writing \u0026ndash; original draft, Conceptualization, Investigation, Methodology. Qiang Zeng: Supervision, Investigation, Funding acquisition, Writing \u0026ndash; review and editing.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets are available upon reasonable request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAditiya, A., Ito, T., 2023. Present-day land subsidence over Semarang revealed by time series InSAR new small baseline subset technique. International Journal of Applied Earth Observation and Geoinformation 125, 103579. https://doi.org/10.1016/j.jag.2023.103579\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBiswal, S.S., Raval, S., Gorai, A.K., 2019. Delineation and mapping of coal mine fire using remote sensing data - a review. Int. J. Remote Sens. 40, 6499\u0026ndash;6529. https://doi.org/10.1080/01431161.2018.1547455\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarroll, M., Gao, C.X., Campbell, T.C.H., Smith, C.L., Dimitriadis, C., Berger, E., Maybery, D., Ikin, J., Abramson, M.J., Sim, M.R., McFarlane, A., Smith, K., Guo, Y., 2022. Impacts of coal mine fire-related PM2.5 on the utilisation of ambulance and hospital services for mental health conditions. Atmospheric Pollution Research 13, 101415. https://doi.org/10.1016/j.apr.2022.101415\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, M., Xue, Yayong, Xue, Yibo, Peng, J., Guo, J., Liang, H., 2024. Assessing the effects of climate and human activity on vegetation change in Northern China. Environmental Research 247, 118233. https://doi.org/10.1016/j.envres.2024.118233\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, X., Peng, J., Song, Z., Zheng, Y., Zhang, B., 2022. Monitoring Persistent Coal Fire Using Landsat Time Series Data From 1986 to 2020. IEEE Trans. Geosci. Remote Sensing 60, 5001616. https://doi.org/10.1109/TGRS.2022.3142350\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaverie, M., Ju, J., Masek, J.G., Dungan, J.L., Vermote, E.F., Roger, J.-C., Skakun, S.V., Justice, C., 2018. The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sens. Environ. 219, 145\u0026ndash;161. https://doi.org/10.1016/j.rse.2018.09.002\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrawford, C.J., Roy, D.P., Arab, S., Barnes, C., Vermote, E., Hulley, G., Gerace, A., Choate, M., Engebretson, C., Micijevic, E., Schmidt, G., Anderson, C., Anderson, M., Bouchard, M., Cook, B., Dittmeier, R., Howard, D., Jenkerson, C., Kim, M., Kleyians, T., Maiersperger, T., Mueller, C., Neigh, C., Owen, L., Page, B., Pahlevan, N., Rengarajan, R., Roger, J.-C., Sayler, K., Scaramuzza, P., Skakun, S., Yan, L., Zhang, H.K., Zhu, Z., Zahn, S., 2023. The 50-year Landsat collection 2 archive. Science of Remote Sensing 8, 100103. https://doi.org/10.1016/j.srs.2023.100103\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeng, J., Ge, S., Qi, H., Zhou, F., Shi, B., 2021. Underground coal fire emission of spontaneous combustion, Sandaoba coalfield in Xinjiang, China: Investigation and analysis. Science of The Total Environment 777, 146080. https://doi.org/10.1016/j.scitotenv.2021.146080\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDu, X., Sun, D., Li, F., Tong, J., 2022. A Study on the Propagation Trend of Underground Coal Fires Based on Night-Time Thermal Infrared Remote Sensing Technology. Sustainability 14, 14741. https://doi.org/10.3390/su142214741\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDwyer, J.L., Roy, D.P., Sauer, B., Jenkerson, C.B., Zhang, H.K., Lymburner, L., 2018. Analysis Ready Data: Enabling Analysis of the Landsat Archive. Remote Sens. 10, 1363. https://doi.org/10.3390/rs10091363\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalve, J.M., S\u0026aacute;nchez, J.M., Garc\u0026iacute;a-Santos, V., Gonz\u0026aacute;lez-Piqueras, J., Calera, A., Villodre, J., 2022. Assessment of Land Surface Temperature Estimates from Landsat 8-TIRS in A High-Contrast Semiarid Agroecosystem. Algorithms Intercomparison. Remote Sensing 14, 1843. https://doi.org/10.3390/rs14081843\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, X., Yang, X., Luo, Z., Guan, T., 2020. Application of unmanned aerial vehicle (UAV) thermal infrared remote sensing to identify coal fires in the Huojitu coal mine in Shenmu city, China. Sci Rep 10, 13895. https://doi.org/10.1038/s41598-020-70964-5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHooper, A., Bekaert, D., Spaans, K., Arıkan, M., 2012. Recent advances in SAR interferometry time series analysis for measuring crustal deformation. Tectonophysics 514\u0026ndash;517, 1\u0026ndash;13. https://doi.org/10.1016/j.tecto.2011.10.013\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang, L., Lin, H., Ma, J., Kong, B., Wang, Y., 2011. Potential of small-baseline SAR interferometry for monitoring land subsidence related to underground coal fires: Wuda (Northern China) case study. Remote Sensing of Environment 115, 257\u0026ndash;268. https://doi.org/10.1016/j.rse.2010.08.008\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang, W., Wu, J., Gu, L., Yang, B., Chen, Q., Liu, X., 2011. Monitoring Method of Underground Coal Fire Based on Night Thermal Infrared Remote Sensing Technology. Spectrosc. Spectr. Anal. 31, 357\u0026ndash;361. https://doi.org/10.3964/j.issn.1000-0593(2011)02-0357-05\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKelly, J., Kljun, N., Olsson, P.-O., Mihai, L., Liljeblad, B., Weslien, P., Klemedtsson, L., Eklundh, L., 2019. Challenges and Best Practices for Deriving Temperature Data from an Uncalibrated UAV Thermal Infrared Camera. Remote Sens. 11, 567. https://doi.org/10.3390/rs11050567\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuenzer, C., Zhang, J., Tetzlaff, A., van Dijk, P., Voigt, S., Mehl, H., Wagner, W., 2007. Uncontrolled coal fires and their environmental impacts: Investigating two arid mining regions in north-central China. Appl. Geogr. 27, 42\u0026ndash;62. https://doi.org/10.1016/j.apgeog.2006.09.007\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, F., Yang, W., Liu, X., Sun, G., Liu, J., 2018. Using high-resolution UAV-borne thermal infrared imagery to detect coal fires in Majiliang mine, Datong coalfield, Northern China. Remote Sens. Lett. 9, 71\u0026ndash;80. https://doi.org/10.1080/2150704X.2017.1392632\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang, Y., Yang, Y., Guo, S., Tian, F., Wang, S., 2023. Combustion mechanism and control approaches of underground coal fires: a review. Int J Coal Sci Technol 10, 24. https://doi.org/10.1007/s40789-023-00581-w\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, J., Wang, Y., Li, Y., Dang, L., Liu, X., Zhao, H., Yan, S., 2019. Underground Coal Fires Identification and Monitoring Using Time-Series InSAR With Persistent and Distributed Scatterers: A Case Study of Miquan Coal Fire Zone in Xinjiang, China. IEEE Access 7, 164492\u0026ndash;164506. https://doi.org/10.1109/ACCESS.2019.2952363\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartha, T.R., Guha, A., Kumar, K.V., Kamaraju, M.V.V., Raju, E.V.R., 2010. Recent coal-fire and land-use status of Jharia Coalfield, India from satellite data. Int. J. Remote Sens. 31, 3243\u0026ndash;3262. https://doi.org/10.1080/01431160903159340\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMasek, J.G., Wulder, M.A., Markham, B., McCorkel, J., Crawford, C.J., Storey, J., Jenstrom, D.T., 2020. Landsat 9: Empowering open science and applications through continuity. Remote Sens. Environ. 248, 111968. https://doi.org/10.1016/j.rse.2020.111968\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcKenna, P., Erskine, P.D., Lechner, A.M., Phinn, S., 2018. Measuring fire severity using UAV imagery in semi-arid Central Queensland, Australia (vol 38, pg 4244, 2017). Int. J. Remote Sens. 39, 4285\u0026ndash;4285. https://doi.org/10.1080/01431161.2018.1443780\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeghraj, K.C., Leigh, L., Pinto, C.T., Kaewmanee, M., 2023. Method of Validating Satellite Surface Reflectance Product Using Empirical Line Method. Remote Sens. 15, 2240. https://doi.org/10.3390/rs15092240\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinx, J.C., Lamb, W.F., Andrew, R.M., Canadell, J.G., Crippa, M., Doebbeling, N., Forster, P.M., Guizzardi, D., Olivier, J., Peters, G.P., Pongratz, J., Reisinger, A., Rigby, M., Saunois, M., Smith, S.J., Solazzo, E., Tian, H., 2021. A comprehensive and synthetic dataset for global, regional, and national greenhouse gas emissions by sector 1970\u0026ndash;2018 with an extension to 2019. Earth Syst. Sci. Data 13, 5213\u0026ndash;5252. https://doi.org/10.5194/essd-13-5213-2021\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishra, R.K., Bahuguna, P.P., Singh, V.K., 2011. Detection of coal mine fire in Jharia Coal Field using Landsat-7 ETM\u0026thinsp;+\u0026thinsp;data. Int. J. Coal Geol. 86, 73\u0026ndash;78. https://doi.org/10.1016/j.coal.2010.12.010\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamzan, U., Fan, H., Aeman, H., Ali, M., A. A. Al-qaness, M., 2022. Combined analysis of PS-InSAR and hypsometry integral (HI) for comparing seismic vulnerability and assessment of various regions of Pakistan. Sci Rep 12, 22423. https://doi.org/10.1038/s41598-022-26159-1\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiyas, M.J., Syed, T.H., Kumar, H., Kuenzer, C., 2021. Detecting and Analyzing the Evolution of Subsidence Due to Coal Fires in Jharia Coalfield, India Using Sentinel-1 SAR Data. Remote Sens. 13, 1521. https://doi.org/10.3390/rs13081521\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoy, D., Singh, G., Seo, Y.-C., 2019. Coal mine fire effects on carcinogenicity and non-carcinogenicity human health risks. Environmental Pollution 254, 113091. https://doi.org/10.1016/j.envpol.2019.113091\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoy, P., Guha, A., Kumar, K.V., 2015. An approach of surface coal fire detection from ASTER and Landsat-8 thermal data: Jharia coal field, India. Int. J. Appl. Earth Obs. Geoinf. 39, 120\u0026ndash;127. https://doi.org/10.1016/j.jag.2015.03.009\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchroeder, W., Oliva, P., Giglio, L., Quayle, B., Lorenz, E., Morelli, F., 2016. Active fire detection using Landsat-8/OLI data. Remote Sens. Environ. 185, 210\u0026ndash;220. https://doi.org/10.1016/j.rse.2015.08.032\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSelvakumaran, S., Plank, S., Gei\u0026szlig;, C., Rossi, C., Middleton, C., 2018. Remote monitoring to predict bridge scour failure using Interferometric Synthetic Aperture Radar (InSAR) stacking techniques. International Journal of Applied Earth Observation and Geoinformation 73, 463\u0026ndash;470. https://doi.org/10.1016/j.jag.2018.07.004\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong, Z., Kuenzer, C., 2014. Coal fires in China over the last decade: A comprehensive review. International Journal of Coal Geology 133, 72\u0026ndash;99. https://doi.org/10.1016/j.coal.2014.09.004\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTian, H., Tao, Y., Kou, P., Alonso, A., Luo, X., Gong, C., Fan, Y., Lei, C., Gou, Y., 2023. Monitoring and evaluation of gully erosion in China\u0026rsquo;s largest loess tableland based on SBAS-InSAR. Nat Hazards 117, 2435\u0026ndash;2454. https://doi.org/10.1007/s11069-023-05950-x\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTorres, R., Snoeij, P., Geudtner, D., Bibby, D., Davidson, M., Attema, E., Potin, P., Rommen, B., Floury, N., Brown, M., Traver, I.N., Deghaye, P., Duesmann, B., Rosich, B., Miranda, N., Bruno, C., L\u0026rsquo;Abbate, M., Croci, R., Pietropaolo, A., Huchler, M., Rostan, F., 2012. GMES Sentinel-1 mission. Remote Sens. Environ. 120, 9\u0026ndash;24. https://doi.org/10.1016/j.rse.2011.05.028\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, T., Shi, J., Ma, Y., Husi, L., Comyn-Platt, E., Ji, D., Zhao, T., Xiong, C., 2019. Recovering Land Surface Temperature Under Cloudy Skies Considering the Solar-Cloud-Satellite Geometry: Application to MODIS and Landsat-8 Data. J. Geophys. Res.-Atmos. 124, 3401\u0026ndash;3416. https://doi.org/10.1029/2018JD028976\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Z., Zhou, J., Liu, S., Li, M., Zhang, X., Huang, Z., Dong, W., Ma, J., Ai, L., 2022. A Land Surface Temperature Retrieval Method for UAV Broadband Thermal Imager Data. IEEE Geosci. Remote Sens. Lett. 19. https://doi.org/10.1109/LGRS.2021.3100586\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan, S., Shi, K., Li, Y., Liu, J., Zhao, H., 2020. Integration of satellite remote sensing data in underground coal fire detection: A case study of the Fukang region, Xinjiang, China. Front. Earth Sci. 14, 1\u0026ndash;12. https://doi.org/10.1007/s11707-019-0757-9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, B., She, J., Liu, G., Ma, D., Zhang, R., Zhou, Z., Zhang, B., 2022. Coal fire identification and state assessment by integrating multitemporal thermal infrared and InSAR remote sensing data: A case study of Midong District, Urumqi, China. ISPRS Journal of Photogrammetry and Remote Sensing 190, 144\u0026ndash;164. https://doi.org/10.1016/j.isprsjprs.2022.06.007\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, H., Thanh, H.V., Han, F., Wang, Y., Zhang, Xun, Zhao, R., Sasaki, K., Zhang, Xiaoming, 2023. Development of an in-situ gel from CO2-captured complex solution and inhibiting coal spontaneous combustion: A case study in thermal engineering. Case Studies in Thermal Engineering 50, 103423. https://doi.org/10.1016/j.csite.2023.103423\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, P., Qian, X., Guo, S., Wang, B., Xia, J., Zheng, X., 2023. A New Method for Continuous Track Monitoring in Regions of Differential Land Subsidence Rate Using the Integration of PS-InSAR and SBAS-InSAR. Remote Sensing 15, 3298. https://doi.org/10.3390/rs15133298\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, Z., Hu, C., Wu, Z., Zhang, Zhen, Yang, S., Yang, W., 2023. Monitoring and analysis of ground subsidence in Shanghai based on PS-InSAR and SBAS-InSAR technologies. Sci Rep 13, 8031. https://doi.org/10.1038/s41598-023-35152-1\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, L., Zhang, D., Wang, J., Huang, Z., Pan, D., 2013. Mapping Land Subsidence Related to Underground Coal Fires in the Wuda Coalfield (Northern China) Using a Small Stack of ALOS PALSAR Differential Interferograms. Remote Sensing 5, 1152\u0026ndash;1176. https://doi.org/10.3390/rs5031152\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng, Q.: Causes, Monitoring, Extinction, and Eco-environmental Impacts of Underground Coal Fires: A Comprehensive Perspective, EGU General Assembly 2023, Vienna, Austria, 24\u0026ndash;28 Apr 2023, EGU23-914, https://doi.org/10.5194/egusphere-egu23-914, 2023.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Smoke fugitive channels identification, Underground coal fires, Vegetation coverage, Land surface temperature, InSAR Technique","lastPublishedDoi":"10.21203/rs.3.rs-4856299/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4856299/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUnderground coal fires are a pervasive global environmental issue, especially in coal-rich regions, causing significant environmental damage, safety hazards, and economic losses. These fires release smoke containing carbon dioxide and other harmful gases, exacerbating climate change. This study presents a novel comprehensive analysis method using multi-source remote sensing technology to detect smoke fugitive channels caused by coal fires. We utilized 29 Landsat-8 satellite images of the Sulabulak fire area in China to retrieve vegetation coverage (FVC) and land surface temperature (LST), identifying sparse vegetation and high-temperature anomaly areas. Additionally, 135 dual-polarized Sentinel-1A images were used to obtain surface deformation through SBAS-InSAR and PS-InSAR techniques. The integration of these datasets, validated by field survey data, revealed a high degree of overlap between the identified smoke fugitive channels and subsidence areas. Our results demonstrate an annual increase in sparse vegetation areas, high-temperature anomalies, and ground subsidence, indicating intensified coal fire combustion and expanding smoke fugitive channels. This method's effectiveness in identifying coal fire areas underscores its potential for enhancing coal fire monitoring and management, contributing to more accurate carbon emission estimates and improved mitigation strategies.\u003c/p\u003e","manuscriptTitle":"Application of Multi-Source Remote Sensing Fusion for Identifying Smoke Fugitive Channels in the Sulabulak Fire Area, Urumqi, China","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-18 01:50:39","doi":"10.21203/rs.3.rs-4856299/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"09088328-56dd-44aa-8bec-35dd7c744770","owner":[],"postedDate":"September 18th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-09-24T15:38:55+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-18 01:50:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4856299","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4856299","identity":"rs-4856299","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}