Current carbon sequestration technologies are not meeting targets to deliver the 2050 global net zero goal. Hence, the increasing campaign in nature-based solutions (NBS) rather than depending on engineered sequestration technologies, which is by far harder to scale up. Though amplifying the NBS that are already inherent in the environment is a matter of increased activities, the current changing climatic conditions made it complicated such as site-targeted mangrove rehabilitation, regenerative forestry, and restorative agriculture. These land-based solutions comprise approximately half of the global total carbon sequestered, while the other 50% is naturally sinking in the ocean deep.

Harnessing potential ocean productivity is a huge leap in the world’s carbon sequestration thrust. Not only does this process happen in the natural world, but quantification, monitoring, and forecasting activities can aid future policymaking to amplify productivity in our ocean–which is also expected to serve as a driver to the placement of programs geared toward water quality preservation, conservation, and treatment in inland waters. This study focuses on the quantification and forecasting of the Biological Pump potential in the Philippine Sea, specifically inside the Exclusive Economic Zone (EEZ). Variabilities and disturbances such as increased sea surface temperature, and considering the geographic location of the Philippines, receiving high frequency of annual typhoons, were investigated to affect ocean productivity. Spatio-temporal maps were generated to provide visualization for the trends and phenomena before, during, and after typhoon occurrence for the years 2019 until 2021. The normal scenario for typhoons was reflected in 2019 while both 2020 and 2021 years recorded rare high-intensity supertyphoons Rolly (Goni) and Odette (Rai), while is respectively. NASA Ocean Biology Processing Group (OBPG) Ocean Color data were used to produce spatiotemporal maps for both chlorophyll (CHL) and Sea surface temperature (SST). Correlating these maps with typhoon occurrence, and SST, the Biological Pump potential annual estimate was generated.

Super-resolution (SR) is a well-established technique used to enhance the resolution of low-resolution images. In this paper, we introduce a novel approach for super-resolution of Sentinel-2 10m RGB images using higher-resolution Venus 5m RGB images. The proposed method takes advantage of a modified SRResNet network, integrates perceptual loss based on the VGG network, and incorporates a learning rate decay strategy for improved performance. By leveraging the higher-resolution Venus 5m RGB images as a reference image, this approach aims to generate high-quality, super-resolution images of the Sentinel-2 10m RGB images. The modified SRResNet network is designed to capture and learn the underlying patterns and details present in the Venus images, enabling it to effectively enhance the resolution of the Sentinel-2 images. In addition, the inclusion of perceptual loss based on the VGG network helps to preserve important visual features and maintain the overall image quality. The learning rate decay strategy ensures the network converges to an optimal solution by gradually reducing the learning rate during the training process. Our research contributes to the field of super-resolution by offering a novel approach specifically tailored for enhancing the resolution of Sentinel-2 10m RGB images using Venus 5m RGB images. The proposed methodology has the potential to benefit various applications, such as remote sensing, land cover analysis, and environmental monitoring, where high-resolution imagery is crucial for accurate and detailed analysis. In summary, our approach presents a promising solution for the super-resolution of Sentinel-2 10m RGB images, providing an effective means to obtain higher-resolution imagery by leveraging the complementary information from Venus 5m RGB images. We used a SEN2VENµS dataset for this research. The SEN2VENµS dataset comprises cloud-free surface reflectance patches obtained from Sentinel-2 imagery. Notably, these patches are accompanied by corresponding reference surface reflectance patches captured at a remarkable 5-meter resolution by the VENµS Micro-Satellite on the same acquisition day. To assess the effectiveness of the proposed approach, we evaluate it using widely used metrics such as mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). These metrics provide quantitative measurements of the quality and fidelity of the super-resolution images. Experimental results demonstrate the effectiveness of our proposed approach in achieving improved super-resolution performance compared to existing methods. As an example, the proposed method achieved a PSNR of 35.70 and an SSIM of 0.94 on the training dataset, outperforming the bicubic interpolation method, which yielded a PSNR of 29.53 and an SSIM of 0.92. On the validation dataset, our approach achieved a PSNR of 40.3809 and an SSIM of 0.98, while the bicubic interpolation method achieved a PSNR of 34.26 and an SSIM of 0.94. Finally, on the test dataset, our approach achieved a PSNR of 29.8231 and an SSIM of 0.90, whereas the bicubic interpolation method yielded a PSNR of 26.99 and an SSIM of 0.85. The evaluation, based on MSE, PSNR, and SSIM metrics showcase the enhanced visual quality, increased image resolution, and improved similarity to the reference Venus images.

Over the past few years, global warming has had increasingly noticeable effects, especially through the melting of the polar ice caps. This has caused sea levels to rise, which puts coastal cities and islands at risk of flooding. To combat this issue, monitoring and examining water changes has proven effective in predicting natural disasters caused by rising sea levels. One crucial factor in understanding the impact of global warming is sea surface height (SSH). Measuring SSH can provide valuable information about ocean-level changes. This research used data from the Jason 2 altimetry radar satellite, which provided 36 cycle periods per year, to investigate water heights around the Hawaiian Islands in 2019. To accurately evaluate water height variations, a specific area near the Pacific Ocean close to the Hawaiian Islands was selected. By processing the collected satellite data, a water height chart was created, which revealed an overall increase in height over one year. This analysis provided insight into changing ocean levels in the region, highlighting the urgency of addressing potential threats faced by coastal communities. The study also explored several influential factors contributing to water height variations, such as temperature, precipitation, air pressure, and humidity in Google Earth engine cloud-based platform. Machine learning algorithms, including MLPR and XGBOOST, were used to model water height within the specified range. The results showed that the XGBOOST algorithm was superior in accurately predicting water height, with an impressive R-square value of 0.95. In comparison, the MLPR algorithm achieved an R-square value of 0.91. These findings underscore the effectiveness of using advanced machine learning techniques to understand and model the complex dynamics of water height fluctuations in response to climate change factors. By utilizing these insights, policymakers, scientists, and local authorities can make informed decisions and develop resilient strategies to mitigate the risks associated with rising sea levels. Such proactive measures are crucial for safeguarding vulnerable coastal cities and islands from the increasing frequency and severity of natural disasters exacerbated by global warming.

Foundation models have demonstrated unparalleled performance across diverse tasks involving vision, language, and multimodal domains. Notably, visual foundation models have surpassed the capabilities of prior task-specific models in various dense prediction tasks. Yet, these models still target general benchmarks and are evaluated on curated datasets. The practical adaptation of such models to domain-specific tasks remains an area that has received relatively limited attention.

The domain of remote sensing imagery holds significant importance due to its critical real-world applications, such as instance segmentation of buildings, which enables precise identification and analysis of buildings' footprints for various applications, including urban planning and infrastructure development. While Convolutional Neural Networks (CNNs) have demonstrated remarkable capabilities in buildings' rooftop instance segmentation task and have led to the development of cutting-edge models in this domain, they do have certain limitations. One prominent constraint is the limited generalization, where deploying a highly accurate CNN-based buildings footprint segmentation model on unseen data may lead to reduced performance. One may resort to fine-tuning or retraining to enhance the model's performance.

For this aim, we present a novel approach to adapt foundation models to address existing models' generalization dropback. Among several models, our focus centers on the Segment Anything Model (SAM), a potent foundation model renowned for its prowess in class-agnostic image segmentation capabilities. We start by identifying the limitations of SAM, revealing its suboptimal performance when applied to remote sensing imagery. Moreover, SAM does not offer recognition abilities and thus fails to classify and tag localized objects. To overcome this, we propose to adapt SAM via prompt engineering. Concretely, our investigation delves into 14 distinct single and composite prompting strategies, encompassing a novel approach that enhances SAM performance by integrating a pre-trained CNN as a prompt generator. To the best of our knowledge, this is the first attempt to augment SAM with a CNN-based prompt generator that offers recognition capabilities.

Via a thorough quantitative and qualitative analysis, we evaluate each scenario using three remote sensing datasets: WHU Building Dataset, Massachusetts Buildings Dataset, and AICrowd Mapping Challenge. Our results highlight a substantial enhancement in SAM's buildings segmentation accuracy. Specifically, for out-of-distribution performance on the WHU dataset, we observed a 5.47% and 4.81% improvement in both IoU and F1-score, respectively. We also witnessed 2.72% and 1.58% enhancement in terms of True-Positive-IoU and True-Positive-F1-score, respectively, for in-distribution performance on the WHU dataset.

We hope this work will inspire the broader academic community to explore the potential of foundation models for domain-specific tasks, as we intend to release this work code repository.

Assessing the environmental impact of the mining industry plays a critical role in understanding and mitigating the ecological consequences of extractive activities. This paper presents MineSegSAT, a model that presents a novel approach to predicting environmentally impacted areas within mining land regions using the SegFormer deep learning segmentation architecture trained on Sentinel-2 data. The data was collected from non-overlapping regions over Western Canada in 2021 containing areas of land that have been environmentally impacted by mining activities that were identified from high-resolution satellite imagery in 2021 (https://doi.org/10.1038/s43247-023-00805-6). The SegFormer architecture, a state-of-the-art semantic segmentation framework, is employed to leverage its advanced spatial understanding capabilities for accurate land cover classification. We investigate the efficacy of loss functions including Dice, Tversky, and Lovasz loss respectively and evaluate model performance based on F1-score, precision, recall, and accuracy. The trained model was utilized for inference over the same areas in the ensuing year to identify potential areas of expansion or contraction over these same periods. The Sentinel-2 data is made available on Amazon Web Services through a collaboration with Earth Daily Analytics which provides corrected and tiled analytics-ready data on the AWS platform. The model and ongoing API to access the data on AWS allow the creation of an automated tool to monitor the extent of disturbed areas surrounding known mining sites to ensure compliance with their environmental impact goals.

CartoDEM Version 3 Release 1 is an openly accessible dataset for digital elevation model (DEM) and is currently the most reliable dataset for the relatively plain region in India specifically. The presented study is to evaluate the CartoDEM with respect to the two openly accessible space-borne LiDAR datasets from the LiDAR sensors namely, Advanced Topographic Laser Altimeter System (ATLAS) onboard the Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) and Global Ecosystem Dynamics Investigation (GEDI) over the International Space Station (ISS). The difference i.e. deviations were computed for the CartoDEM and the LiDAR footprint elevation values for each of the two datasets namely, ICESat-2 and GEDI. The difference values were filtered for footprints with differences between 0 to 2.5 m between the DEM and LiDAR elevation values. Besides this, an overall estimate is also done for the elevation values obtained over the surface i.e. ground as well as the objects over the terrain, such as the trees and buildings. The root mean square (RMSE) is observed as 1.16 m and 1.74 m for ICESat-2 and GEDI datasets for the points/footprints on the terrain. Whereas considering similar constraints and noise removal for the two LiDAR datasets the RMSE is obtained as 1.78m and 5.48m for the ICESat-2 and GEDI footprints on the surface (terrain/object) respectively. The study reveals that the CartoDEM is highly accurate in the plains when validated with respect to the ICESat-2 datasets, which work on the photon counting technique in contrast to the waveform technique used in GEDI. The use of ground control points (GCPs) acquired by National Remote Sensing Centre (NRSC) using Differential Global Positioning System (DGPS), as part of the GCP Library (GCPL) in the preparation of CartoDEM, from triangulated Cartosat-1 stereo datasets followed by DEM editing in the preparation of version 3, are also the important factors which contributed to its higher accuracy. Further, it is observed that the ICESat-2 performance is better than the GEDI mission for the terrain height. Thus it is depicted that the spaceborne lidar datasets from ICESat-2 can be utilized for the validation of DEMs. ICESat-2 can be used as a measure of height for applications where a primary input of DEM or elevation is required to be utilized or verified for engineering or modeling-based applications, in traditional or Remote Sensing and GIS (RS&GIS) studies as well as analysis for decision making.

Hydrothermal alteration mapping is considered as a widely adopted step in mineral exploration of numerous ore deposits. In this work the wavelength mapping and relative absorption band depth methods were applied to map hydrothermal alterations in a site from the abandoned mine of Idikel, western Anti-Atlas, Morocco. Fe2+/Fe3+, Al-OH, Mg-Fe-OH/ CO3 hydrothermal alteration minerals were targeted in the basis of HyMap airborne imaging spectroscopy data. Using wavelength mapping approach, the 900 to 1205 nm, 2094 to 2217 nm, and 2264 to 2318 nm ranges were selected to map Fe2+/Fe3+, Al-OH, Mg-Fe-OH/ CO3 manifestations, respectively. By carefully selecting these spectral ranges, this study aimed to achieve accurate and reliable mapping of hydrothermal alteration features. The highest interpolated depth of Al-OH features was matched with a major cluster of pixels at 2200 nm. Mg-Fe-OH/ CO3 highest interpolated depth was depicted with 2300 nm. Fe2+/Fe3+ highest interpolated depth was depicted between 900 and 1000. The relative absorption band depth method was also applied to enhance the detectability of hydrothermal alteration minerals. This method involves assessing the depth of absorption bands associated with the target minerals, allowing for a detailed characterization of the alteration features. The combination of both wavelength mapping and enhancement methods contributed to a comprehensive and robust hydrothermal alteration mapping process. The identification of Fe2+/Fe3+, Al-OH, and Mg-Fe-OH/CO3 manifestations provided valuable insights into potential mineralization zones within the study area. Overall, this research contributes to the advancement of hydrothermal alteration mapping using hyperspectral data by selecting the required HyMap bands for mapping targeted alterations. The combination of wavelength mapping and enhancement methods proves to be a powerful approach for accurately identifying and characterizing hydrothermal alteration features using the specific hyperspectral channels. The findings from this study can aid future mineral exploration endeavors in similar geological settings, providing valuable guidance for locating potential mineral resources in mountainous and challenging terrains.

Access to high spatial resolution satellite images enables more accurate and detailed analysis of these images. Furthermore, it facilitates easier decision-making on a wide range of issues. Nevertheless, there are commercial satellites such as Worldview that have provided a spatial resolution of fewer than 2.0 meters, but using them for large areas or multi-temporal analysis of an area brings huge costs. So, to tackle these limitations and access free satellite images with higher spatial resolution, there are challenges that are known as single-image super-resolution (SISR). The Sentinel-2 satellites were launched by the European Space Agency (ESA) to monitor the Earth which has enabled access to free multi-spectral images, five-day time coverage, and global spatial coverage has been among the achievements of this launch. Also, it led to the creation of a new flow in the field of space businesses. These satellites have provided bands with various spatial resolutions which the Red, Green, Blue, and NIR bands have the highest spatial resolution by 10 m. In this study, therefore, to recover high-frequency details, increase the spatial resolution, and cut down costs Sentinel-2 images have been considered. Additionally, a model based on Enhanced Super-Resolution Generative Adversarial Network (ESRGAN) has been introduced to increase the resolution of 10 m bands to 2.5 m. In the proposed model, to preserve the spectral information of the images, changes were made in the loss function. Also, since there is no way to obtain higher-resolution (HR) images in the conditions of the Sentinel-2 acquisition image, we preferred instead to simulate data, to use a sensor with a higher spatial resolution that is similar in spectral bands to Sentinel-2 as a reference and HR image. Hence, Sentinel-Worldview image pairs were prepared and the network was trained. Finally, the evaluation of the results obtained, showed that while maintaining the visual appearance, it was able to maintain some spectral features of the image as well. The average Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Spectral Angle Mapper (SAM) metrics of the proposed model from the test dataset were 42.20 dB, 0.91, 0.08 radians, respectively.

Wetlands are valuable natural resources which provide many services to both the environment and humans. Over the past several decades, climatic change and human activities have a considerable impact on the water level of wetlands. Lake Zarivar, located in the northwestern region of Iran, represents a significant ecological unit and aquatic ecosystem. Given its unique characteristics, it is imperative to conduct thorough investigations using remote sensing techniques into the alterations occurring within its water body. In this study, the relationship between seasonal variations in Zarivar Lake's water surface area and meteorological variables, including precipitation, evapotranspiration, and lake surface water temperature (LSWT) from 2015 to 2022 has been analyzed. To this end, the Google Earth Engine (GEE) cloud platform, a powerful and fast tool for processing the time series of images, were used. The water body was extracted by utilizing the average images of the dual-polarized SAR Sentinel-1 imagery for each season. The edge of the lake was identified by applying the Otsu threshold. Furthermore, meteorological parameters encompass the utilization of the Landsat-8 satellite's Thermal band to determine LSWT, the Chirps rainfall model data for assessing precipitation levels, and the employment of MODIS evapotranspiration products in the form of 8-day data. The results demonstrated the logic of the relationship between changes in the water zone's area and meteorological variables. The correlation coefficient between water area of the lake and precipitation data were be 0.67 and 0.73 in the winter and spring seasons, respectively, and 0.29 and 0.30 in the summer and fall seasons, respectively. On the other hand, there was a strong relationship between the water area and the LSWT. As a result, the correlation coefficient had a value of -0.21 in the winter, when there is a decline in LSWT and an increase in snow and rain, and a value of 0.45 in the spring, when there is an increase in LSWT and considerable rainfall in spring. It reached 0.13 in the summer because of the higher LSWT and lower precipitation, and it dropped to -0.15 in the fall because of the lower LSWT and higher precipitation. All seasons have positive values for the correlation coefficient between the water body area and the evapotranspiration parameters. Due to the persistent rains in the winter and spring, evapotranspiration also increased, reaching values of 0.50 and 0.59 in the respective seasons. The relationship between water area and evapotranspiration is also in the same direction throughout the summer, with a value of 0.60 due to the rise in LSWT and significant rainfall in spring. Due to the lack of precipitation and the consumption of lake water in the summer, the relationship between evapotranspiration and the area of the water zone increased to 0.45 in the fall.

Turkey is an earthquake country where disasters occur frequently due to its geographical location. Earthquakes cause spatial changes in urban and rural areas,which can be monitored and analyzed fast and accurately with Remote Sensing and Geographical Information Technologies.The study aims to determine the spatial changesin the city center of Hatay, Antakya,which was heavily hit by two large earthquakes of magnitude 7.7 and 7.6 on 06.02.2023. It was accepted that the structures added to the building stock from 2000 to 2002 were built according to the earthquake regulation published in 1997, and the damaged structures after the earthquake have been evaluated in this context. During the period leading up to the publication of the new building inspection law, the structures built according to the earthquake regulation were identified,and it has been observed that changes are taking place in these areas. To fulfill this aim, 2000 and 2002 images of LANDSAT 7 were usedto perform NBDI (Normalized Building Difference Index) analysis,and the image diffe-rencing was appliedby subtracting the image of the year 2000 from the year 2002. The struc-tural area changes werethen determined by comparing them with the 2023 SENTINEL-2 satellite image created after the earthquake. LANDSAT 7 data within the scope of the study USGS.gov (United States Geological Survey (.gov)) from the website and SENTINEL-2 data were obtained from the SENTINELS COPERNICUS (Sentinel Online) website by selecting certain dates within the scope of the topic. The main difficulties that may affect the results within the scope of the study can be expressed as satellite images obtained from the website and the possibility that geographical data sources are incorrect. The results showed that the structures built according to the earth-quake decrees between 2000 and 2002 have undergone changes after the earthquake that occurred in 2023 and that the regulation is not sufficient in taking precautions against earthquakes. In this context, earthquake regulations should be reviewed, their application should be done correctly during the construction of the structure and it should be ensured that the structures are not damaged. This study, which is based on the spatial change analysis after the earthquake, is an important and useful study in determining the points where the earthquake regulations in question are insufficient in the areas that need to be rebuilt in the city center of Antakya. In addition, it will be an important analysis example for the city planning studies expected to be carried out by local governments for the earthquake zone. Policy makers and practitioners can develop strategic planning approaches for the needs by dividing land determination and urban planning studies into regions within the framework of these areas by taking advantage of the image of the results presented in the study. By determining the damaged urban use areas located in the areas where change detection is observed within the study area, city planning intervention decisions related to this area can be developed.