Comparison of PlanetScope and Sentinel-2 satellite observations in mapping small-scale forest fires

Abstract. Forest fires are natural events that occur in numerous ecosystems worldwide and cause significant damage to human, ecological and socio-economic factors. It is also crucial to obtain useful information on the distribution and density of burned areas on large scale. An efficient way to map large regions is through remote sensing (RS). Nevertheless, the complex scenario and similar spectral signature of features in multispectral bands can lead to many false positives, making it difficult to extract the burned areas accurately. Multispectral data from Sentinel-2 satellite images allow the development of novel burned area indices, as more spectral data is recorded in the Red-Edge region. This research aims to develop a new burned area detection index (BADI) at 20 m spatial resolution in the google earth engine platform to detect the wildfire-affected areas in southwest of Iran using Sentinel-2 satellite imagery. The BADI spectral index has been specially designed to take benefit of the Sentinel-2 spectral bands and use a spectral combination of bands that are reasonable for post-fire burned regions detection. The final results indicated that the proposed index by applying a post-processing stage works well in the case of the study area to identify the burned areas. At the same time, it can satisfactorily suppress the complicated and irrelevant changes in the scene. Furthermore, the BADI index is rapid and can provide the burned areas map in near real-time. According to the Copernicus Emergency Management Service (EMS) reference data, maps of the burned areas were produced with a kappa coefficient of 0.92 and an overall accuracy of 92.15%, which demonstrated a good result in comparison to similar spectral indices.

Satellite-based land-use classification plays a crucial role in various Earth observation applications, ranging from environmental monitoring to disaster management. This study presents a comparative analysis of machine learning techniques applied to land cover classification using Landsat-9 and Sentinel-2 satellite imagery in the Reyhanlı district in southern Türkiye. Three different classification algorithms, Random Forest (RF), Support Vector Machine (SVM), and Maximum Likelihood Classification (MLC), were evaluated for their ability to distinguish different land cover classes. High resolution multispectral satellite imagery processed under the same conditions using Geographic Information System (GIS) software was utilized in this study. Visual inspection and statistical evaluation, including overall accuracy and kappa coefficient, were employed to assess classification performance. The classification of Sentinel-2 and Landsat-9 satellite imagery using different machine learning algorithms resulted in the highest overall accuracy (OA = 0.911, Kappa = 0.879) for Sentinel 2 imagery with the RF algorithm. These findings highlight the importance of satellite image selection and algorithm optimization for accurate land cover mapping. This study provides valuable insights for local planners and authorities and underscores the potential of Sentinel-2 imagery combined with machine learning techniques for effective land-use classification and monitoring.

Coconut farming in the Brazilian semi-arid region, especially in the state of Ceará, plays a significant economic role but faces challenges such as the limited use of modern agricultural technologies. The use of remote sensing techniques, through both multispectral satellite imagery and radar data, offers an opportunity to efficiently, quickly, and accurately monitor crop conditions, promoting more effective and sustainable management. This study aimed to map coconut plantations in the municipality of Paraipaba, in the state of Ceará, using machine learning techniques. Multispectral images from Sentinel-2 satellites were used, covering the period from January 1, 2024, to March 8, 2025, and radar images from Sentinel-1 satellites, from January 1, 2023, to May 30, 2024. The study employed Support Vector Machine (SVM) and Random Forest (RF) classifiers, with RF demonstrating superior accuracy in both multispectral and radar images. The RF classifier achieved an accuracy of 97.8% and a kappa coefficient of 94.2% with multispectral images, outperforming SVM, which yielded an accuracy of 86.3% and a kappa of 71.9%. For radar imagery, RF achieved a global accuracy of 98.6% and a kappa of 93.3%. This study highlights the effectiveness of machine learning techniques, particularly RF, for the accurate and efficient mapping of coconut plantations using Sentinel-1 and Sentinel-2 satellite imagery

Development of an emission estimation method with satellite observations for significant forest fires and comparison with global fire emission inventories: Application to catastrophic fires of summer 2021 over the Eastern Mediterranean

Due to the damage they cause to the environment, forest fires have an important place among the disasters that occur around world. In recent years, forest fires have increased in frequency, size and intensity, especially in Mediterranean countries. Preventive measures should be taken and risk reduction should be implemented so that natural or man-made risks do not turn into a catastrophe disaster. After a disaster commences, the implementation of evacuation plans for the settlement, when necessary, is of great importance in this context. One of these forest fires started on July 23, 2018 in the popular holiday resort of Mati in Greece. Mati located within the borders of the Attica region and 29km east of the Athens, was examined within the scope of this study. The forest fire that took place in the said regions affected a very large area and the fires caused the death of 103 people and the destruction of approximately 4,000 houses, including thousands of vehicles. In the study, data processing and evaluation using Sentinel-2 satellite images from the Copernicus program of the European Space Agency (ESA), SNAP software, an open source software developed by ESA and the ArcMap program were used for subsequent statistical calculations. As a result, it was determined how much the area was burned with the help of Sentinel-2 satellites and a study was carried out on the mapping of the affected areas. In addition, the relationship between disaster risk reduction activities has been examined.

Using sentinel-2 satellite imagery to develop microphytobenthos-based water quality indices in estuaries

The study utilizes the application of Sentinel-1 and Sentinel-2 satellite imagery between 2018 and 2022 in assessing flood extent and damages in Yenagoa, Bayelsa State, Nigeria, using machine learning techniques applied to Sentinel-1 and Sentinel-2 satellite imagery from 2018 to 2022.Synthetic Aperture Radar (SAR) data from Sentinel-1, with its all-weather capabilities, enabled the detection and mapping of the flood extent using machine learning algorithms, while Sentinel-2 multispectral images facilitated land-use classification before and after the flood events using support vector machines (SVM).The Shuttle Radar Topographic Mission (SRTM) and geological map were also used.This study, carried out using Google Earth Engine (GEE), Python, JavaScript, and ArcGIS 10.5, reveals a tremendous increase in the flood-affected areas, which expanded from 54.92 km in 2018 to 90.15 km in 2022.It found that the main drivers of such events were increased rainfall and rapid urbanization.The DEM data extracted from SRTM showed that the low-lying areas, specifically those with an elevation range of -6 m to 7 m (gentle slope, range from 1 to 9), are the areas most prone to flooding.The geological composition, described as a swampy deltaic plain, contributed to prolonging the duration and severity of the flood.Machine learning analysis using Sentinel-2 imagery showed that vegetated and built-up classes are highly flooded, thus bringing socio-economic losses due to displacement of households and economic loss.This study has brought out the vital role of machine learning and remote sensing in flood detection and monitoring, besides the urgent need for data-driven flood risk management strategies integrating regional topography, land use dynamics, and geological factors.

Monitoring burned areas in Thailand and other tropical countries during the post-harvest season is becoming increasingly important. High-resolution remote sensing data from Sentinel-2 satellites, which have a short revisit time, is ideal for accurately and efficiently mapping burned regions. However, automating the mapping of agriculture residual on a national scale is challenging due to the volume of information and level of detail involved. In this study, a Sentinel-2A Level-1C Multispectral Instrument image (MSI) from February 27, 2018 was combined with object-based image analysis (OBIA) algorithms to identify burned areas in Mae Chaem, Chom Thong, Hod, Mae Sariang, and Mae La Noi Districts in Chiang Mai, Thailand. OBIA techniques were used to classify forest, agricultural, water bodies, newly burned, and old burned regions. The segmentation scale parameter value of 50 was obtained using only the original Sentinel-2A band in red, green, blue, near infrared (NIR), and Normalized Difference Vegetation Index (NDVI). The accuracy of the produced maps was assessed using an existing burned area dataset, and the burned area identified through OBIA was found to be 85.2% accurate compared to 500 random burned points from the dataset. These results suggest that the combination of OBIA and Sentinel-2A with a 10 m spatial resolution is very effective and promising for the process of burned area mapping.

This study compares the capability of Sentinel-1, Sentinel-2, and PlanetScope (PS) satellites in monitoring the variations of surface water of Dai Lai Lake, located in North Vietnam, for the 2018–2023 period. The analysis involves the utilization of Google Earth Engine to partially process Sentinel-1 and Sentinel-2 observations, while PS observations are processed using local computers, to generate VH-polarized backscatter coefficient, Normalized Difference Water Index (NDWI), and Modified of Normalized Difference Water Index (MNDWI) maps. The method for making binary water/non-water maps primarily employs the Otsu algorithm on each single map derived from the previous step. Findings reveal that the lake's water extent remains relatively stable over the 6-year period, and is not strongly affected by the seasonal cycle. Although the spatial distribution patterns of the lake exhibit significant similarity, average water extent of the lake derived from 3-m resolution PS imagery is about 2.17 and 5.60% more than that obtained from 10-m resolution Sentinel-2 and Sentinel-1 imagery, respectively. PS observations are effective for monitoring small lakes, but it is advised to check the quality of its NIR band. Sentinel-2 observations prove great effectiveness for lake monitoring, using both NDWI and MNDWI. For Sentinel-1 observations, potential misclassifications could arise due to similarities in VH-polarized backscatter coefficients between water surfaces and other flat surfaces.

The Mediterranean region of the Türkiye is among the areas most susceptible to forest fires. Although the reasons for fire outbreaks vary over time, these events consistently result in the loss of forest resources and significant ecological damage. Forest fires reduce or eliminate many forms of vegetation from the land surface. The Kavaklıdere-Muğla-Yatağan-Yılanlı fire, which occurred between 2 and 8 August 2021, affected a large area. Therefore, the study aims to investigate the land use and land cover (LULC) changes in 2020, 2021, and 2024 within Kavaklıdere district of Muğla. In this study, six LULC classes, Agriculture (A), Bare and Other (BO), Forest (F), Urban (U), Water (W) and Burnt Area (BA) were identified using Sentinel-2 satellite imagery and random forest classification technique on the Google Earth Engine (GEE) platform. In addition, land surface temperature (LST) data were obtained using the split-window algorithm applied to Landsat-8 data. The findings indicated that LULC changes are clearly in fire-affected areas, with LST values are notably higher in burned areas than in other classes. The results demonstrate the utility of remote sensing techniques for monitoring post-fire changes in land cover and surface temperature.

Sentinel-2 satellite and HYSPLIT model suggest that local cereal harvesting substantially contribute to peak Alternaria spore concentrations

Wildfire is a natural disaster that harms human and animal habitats and the socio-economy. Remote sensing techniques are commonly used in the research of natural disasters, natural resources and monitoring. Timely and accurate estimation of the location of forest fires is particularly important for post-fire management and decision-making. Sentinel-2 satellite images of the European Space Agency ‘ESA’ were used to estimate the area affected by forest fires at Bayan-Uul and Bayandun soums in Dornod province, and classified the burn severity levels and comparison with other influencing factors in this study. The normalized burn ratio ‘NBR' and indices on pre-fire and post-fire were calculated. The total burned area was calculated as 58,131.6 ha, and low, moderate-low, moderate-high, and high burn severity levels cover 15,423.7 ha (26.3%), 29,529.4 ha (50.4%), 13,160.2 ha (22.5%), and 18.3 ha (0.03%), respectively. The 87.6% of the burned area is situated in Mongolian territory, while the remaining area (12.4%) belongs to the Russian Federation. Comparing 10 natural and geographical factors that can influence the burn severity and calculating the correlation coefficients by Pearson. Four of them related a positive lower, and six of them related negative lower. The weak relationships of Normalized Difference Vegetation Index ‘NDVI’, elevation were 0.4 and 0.23. However, the precipitation correlation was -0.22 (negative weak). The distribution of wildfire is strongly influenced by the wind, and the correlation coefficient demonstrates a negative correlation with no effect on combustion. The conditions of socio-economic and ecological disastrous consequences such as loss of plant species and resources, changes in plant structure, depletion of pasture resources, extinction of rare animals and plants, reduction of forest resources, and large-scale air pollution resulting in the loss of human and animal in post-fire. Therefore, this research is important to due for studying the burning, distribution and, coverage area of the fire, and create conditions for the prevention of future risks.

The Sentinel-2 satellites are providing an unparalleled wealth of high-resolution remotely sensed information with a short revisit cycle, which is ideal for mapping burned areas both accurately and timely. However, the high detail and volume of information provided actually encumbers the automation of the mapping process, at least for the level of automation required to map systematically wildfires on a national level. This paper proposes a fully automated methodology for mapping burn scars using Sentinel-2 data. Information extracted from a pair of Sentinel-2 images, one pre-fire and one post-fire, is jointly used to automatically label a set of training patterns via two empirical rules. An initial pixel-based classification is derived using this training set by means of a Support Vector Machine (SVM) classifier. The latter is subsequently smoothed following a multiple spectral-spatial classification (MSSC) approach, which increases the mapping accuracy and thematic consistency of the final burned area delineation. The proposed methodology was tested on six recent wildfire events in Greece, selected to cover representative cases of the Greek ecosystems and to present challenges in burned area mapping. The lowest classification accuracy achieved was 92%, whereas Matthews correlation coefficient (MCC) was greater or equal to 0.85.

In abrupt fire disturbances, high quality images suitable for wildfire damage assessment can be difficult to acquire. Quantifying wildfire burn area and severity are essential measures for quick short-term disaster response and efficient long-term disaster restoration. Planetscope (PS) imagery offers 3 m spatial and daily temporal resolution, which can overcome the spatio-temporal resolution tradeoff of conventional satellites, albeit at the cost of spectral resolution. This study investigated the potential of augmenting PS imagery by integrating the spectral information from Sentinel-2 (S2) differenced Normalized Burn Ratio (dNBR) to PS differenced Normalized Difference Vegetation Index (dNDVI) using histogram matching, specifically for wildfire burn area and severity assessment of the Okgye wildfire which occurred on April 4th, 2019. Due to the difficulty in acquiring reference data, the results of the study were compared to the wildfire burn area reported by Ministry of the Interior and Safety. The burn area estimates from this study demonstrated that the histogram-matched (HM) PS dNDVI image produced more accurate burn area estimates and more descriptive burn severity intervals in contrast to conventional methods using S2. The HM PS dNDVI image returned an error of only 0.691% whereas the S2 dNDVI and dNBR images overestimated the wildfire burn area by 5.32% and 106%, respectively. These improvements using PS were largely due to the higher spatial resolution, allowing for the detection of sparsely distributed patches of land and narrow roads, which were indistinguishable using S2 dNBR. In addition, the integration of spectral information from S2 in the PS image resolved saturation effects in areas of low and high burn severity.

Abstract. Following events such as extreme precipitation, floods, storms and hurricanes, litter accumulation in rivers increases. After these events, it is important to identify the areas of litter accumulation in rivers to protect the aquatic ecosystem, prevent secondary disasters from occurring, and prevent this litter from being transported to the seas and oceans. This study aims to determine the areas of litter accumulation in rivers of different lengths after extreme precipitation events in different parts of the world using remote sensing data. Litter accumulations were investigated in the Potpecko River in Serbia in January 2021, the Yangtze River in China in August 2018, the Drina River in Bosnia-Herzegovina in March 2021, and the Ezine River in Türkiye in August 2021. Medium spatial resolution Sentinel-2 and high spatial resolution Pleiades satellite imagery were used. Normalized Difference Water Index (NDWI) and Random Forest (RF) classification algorithms were used to extract information from the images. As a result of the classifications, the overall accuracy was 91.7%, 78.3%, and 95.2% for the Drina, Yangtze, and Potpecko rivers, respectively. In the Ezine River, tree stumps transported by the river were detected with an overall accuracy of 76.9%. Higher resolution Pleiades imagery was more effective in detecting smaller details, while Sentinel-2 imagery was able to detect larger litter accumulations with high accuracy.