A generalized machine learning model for long-term coral reef monitoring in the Red Sea.

Coral reefs, despite covering less than 0.2 % of the ocean floor, harbor approximately 35 % of all known marine species, making their conservation critical. However, coral bleaching, exacerbated by climate change and phenomena such as El Niño, poses a significant threat to these ecosystems. This study focuses on the Red Sea, proposing a generalized machine learning approach to detect and monitor changes in coral reef cover over an 18-year period (2000-2018). Using Landsat 7 and 8 data, a Support Vector Machine (SVM) classifier was trained on depth-invariant indices (DII) derived from the Gulf of Aqaba and validated against ground truth data from Umluj. The classifier was then applied to Al Wajh, demonstrating its robustness across different sites and times. Results indicated a significant decline in coral cover: 11.4 % in the Gulf of Aqaba, 3.4 % in Umluj, and 13.6 % in Al Wajh. This study highlights the importance of continuous monitoring using generalized classifiers to mitigate the impacts of environmental changes on coral reefs.

Assessing climate-driven impacts on Karakoram glacier surge: A geospatial analysis highlighting Shishper Glacial Lake outburst flood events (2019-2022) as prime example.

The Northern Areas of Pakistan encompass the Hindukush, Karakoram, and Himalayan mountain ranges witnessing glacier surging, exacerbated by climate warming. As glaciers rapidly melt, ravines experience heightened blockage and migration, obstructing stream discharges and forming expansive ice-dammed lakes. The rupture of these natural dams triggers Glacial Lake Outburst Floods downstream in the primary glacier's ravine. The catastrophic Glacial Lake Outburst Floods in 2022 across the Karakoram ranges in Northern Pakistan prompted this study. It focuses on Shishper Glacier Lake. The aim is to provide complete flood observations and their devastating effects on downstream communities. Analysis of Landsat 08 Imagery reveals the evolution of Shishper Glacier Lake from its initiation in November 2018 to the catastrophic GLOF in May 2022. The lake reached a maximum area of 0.32 km2 in 2019 and its successive breaches on June 22, 2019, and May 29, 2020, reduced it to 0.018 km2. Draining continued until July 2021, shrinking the lake area to 0.009 km2. A noteworthy 2.73 °C temperature increase in 2022 correlated with an expansion of the lake area to 0.33 km2, culminating in the GLOF on May 7th, 2022. The study emphasizes the critical need for mapping, assessing, and monitoring surging glaciers and glacier-formed lakes in the Karakoram ranges to safeguard downstream communities from potential hazards.

Lithological discrimination based on Landsat-9 OLI sensor and field observation data: The bana an-orogenic volcano-plutonic ring complex, West Cameroon line.

The western region, encompassing the an-orogenic Bana volcano-plutonic ring complex in Cameroon, underwent comprehensive exploration involving remote sensing analysis, fieldwork investigations, petrographic, and volcanological studies. The primary objective of this work was to integrate remote sensing analysis, fieldwork, and laboratory studies to achieve accurate lithological mapping for future prospective mineral explorations in the study area. Field relationships among co-occurring rock units in the area were examined, utilizing Landsat-9 OLI data. Petrographic analysis, including the use of a polarizing microscope, was conducted on various rock units (15 samples), along with volcanological processes studies. Operational Land Imager (OLI) images of Landsat 9 were processed using algorithms including False Colour Composite (FCC), Decorrelation Stretch (DS), Band Ratio (BR) composite, Principal Component Analysis (PCA), Spectral Angle Mapper (SAM) and Constrained Energy Minimization (CEM) methods to identify distinct rock units in the Bana ring complex. As a result, the later methods permitted to identify the petrographic units of the ring complex, which primarily comprise a volcano-plutonic sequence, along with metamorphic rocks like gneisses. The volcanic units include variety of basalts, trachytes, rhyolites and volcanic tuffs, while the plutonic units including gabbros, diorites, syenites and fine-grained granites. The findings of this study accurately at 99 % have permitted to newly setup a geologic map of the study area with implications for future mineral explorations.

Integrated UAV and Satellite Multi-Spectral for Agricultural Drought Monitoring of Winter Wheat in the Seedling Stage.

Agricultural droughts are a threat to local economies, as they disrupt crops. The monitoring of agricultural droughts is of practical significance for mitigating loss. Even though satellite data have been extensively used in agricultural studies, realizing wide-range, high-resolution, and high-precision agricultural drought monitoring is still difficult. This study combined the high spatial resolution of unmanned aerial vehicle (UAV) remote sensing with the wide-range monitoring capability of Landsat-8 and employed the local average method for upscaling to match the remote sensing images of the UAVs with satellite images. Based on the measured ground data, this study employed two machine learning algorithms, namely, random forest (RF) and eXtreme Gradient Boosting (XGBoost1.5.1), to establish the inversion models for the relative soil moisture. The results showed that the XGBoost model achieved a higher accuracy for different soil depths. For a soil depth of 0-20 cm, the XGBoost model achieved the optimal result (R2 = 0.6863; root mean square error (RMSE) = 3.882%). Compared with the corresponding model for soil depth before the upscaling correction, the UAV correction can significantly improve the inversion accuracy of the relative soil moisture according to satellite remote sensing. To conclude, a map of the agricultural drought grade of winter wheat in the Huaibei Plain in China was drawn up.

Estimation of the crop evapotranspiration for Udham Singh Nagar district using modified Priestley-Taylor model and Landsat imagery.

The main challenges for utilizing daily evapotranspiration (ET) estimation in the study area revolve around the need for accurate and reliable data inputs, as well as the interpretation of ET dynamics within the context of local agricultural practices and environmental conditions. Factors such as cloud cover, atmospheric aerosols, and variations in land cover pose challenges to the precise estimation of ET from remote sensing data. This research aimed to utilize Landsat 8 and 9 datasets from the 2022-23 period in the Udham Singh Nagar district to apply the modified Priestley-Taylor (MPT) model for estimating ET. An average ET was estimated 1.33, 1.57, 1.70, 2.99, and 3.20 mm day-1 with 0.29, 0.33, 0.41, 0.69, and 1.03 standard deviation for December, January, February, March, and April month, respectively. In the validation phase, a strong correlation was found between the evaporative fraction derived from MPT and that observed by lysimeter, with R2 = 0.71, mean biased error = 0.04 mm day-1, root mean squared error = 0.62 mm day-1 and agreement index of 0.914. These results collectively support the effectiveness of the MPT model in accurately estimating ET across Udham Singh Nagar district. In essence, this research not only confirms the MPT model's capability in ET estimation but also offers detailed insights into the spatial and temporal fluctuations of energy fluxes and daily ET rates.

Integrating machine learning for enhanced wildfire severity prediction: A study in the Upper Colorado River basin.

Wildfires pose significant threats worldwide, requiring accurate prediction for mitigation. This study uses machine learning techniques to forecast wildfire severity in the Upper Colorado River basin. Datasets from 1984 to 2019 and key indicators like weather conditions and land use were employed. Random Forest outperformed Artificial Neural Network, achieving 72 % accuracy. Influential predictors include air temperature, vapor pressure deficit, NDVI, and fuel moisture. Solar radiation, SPEI, precipitation, and evapotranspiration also contribute significantly. Validation against actual severities from 2016 to 2019 showed mean prediction errors of 11.2 %, affirming the model's reliability. These results highlight the efficacy of machine learning in understanding wildfire severity, especially in vulnerable regions.

Modeling dissolved organic carbon in inland waters using an unmanned aerial vehicles-borne hyperspectral camera.

Remote sensing can provide an alternative solution to quantify Dissolved Organic Carbon (DOC) in inland waters. Sensors embedded on Unmanned Aerial Vehicles (UAV) and satellites that can capture the DOC have already shown good relationships between DOC and the Colored Dissolved Organic Matter absorption (aCDOM.) coefficients in specific spectral regions. However, since the signal recorded by the sensors is reflectance-based, DOC estimates accuracy decreases when inverting the aCDOM. coefficients to reflectance. Thus, the main objective is to study the potential of a UAV-borne hyperspectral camera to retrieve the DOC in inland waters and to develop reflectance-based models using UAV and satellite (Landsat-8 OLI and Sentinel-2 MSI) data. Ensemble based systems (EBS) were favored in this study. The EBSUAV calibration results showed that six spectral regions (543.5, 564.5, 580.5, 609.5, 660, and 684 nm) are sensitive to DOC in waters. The EBSUAV test results showed a good concordance between measured and estimated DOC with an R2 = Nash-criterion (NASH) = 0.86, and RMSE (Root Mean Squares Error) = 0.68 mg C/L. The EBSSAT test results also showed a strong concordance between measured and estimated DOC with R2 = NASH = 0.92 and RMSE = 0.74 mg C/L. The spatial distribution of DOC estimates showed no dependency to other optically active elements. Nevertheless, estimates were sensitive to haze and sun glint.

Remote sensing analysis of spatiotemporal impacts of anthropogenic influence on mountain landscape ecology in Pir Chinasi national park.

Mountain landscapes can be fragmented due to various human activities such as tourism, road construction, urbanization, and agriculture. It can also be due to natural factors such as flash floods, glacial lake outbursts, land sliding, and climate change such as rising temperatures, heavy rains, or drought.The study's objective was to analyze the mountain landscape ecology of Pir Chinasi National Park under anthropogenic influence and investigate the impact of anthropogenic activities on the vegetation. This study observed spatiotemporal changes in vegetation due to human activities and associated climate change for the past 25 years (1995-2020) around Pir Chinasi National Park, Muzaffrabad, Pakistan. A structured questionnaire was distributed to 200 residents to evaluate their perceptions of land use and its effects on local vegetation. The findings reveal that 60% of respondents perceived spatiotemporal pressure on the park. On the other hand, the Landsat-oriented Normalized Difference Vegetation Index (NDVI) was utilized for the less than 10% cloud-covered images of Landsat 5, 7, and 8 to investigate the vegetation degradation trends of the study area. During the entire study period, the mean maximum NDVI was approximately 0.28 in 1995, whereas the mean minimum NDVI was - 2.8 in 2010. QGIS 3.8.2 was used for the data presentation. The impact of temperature on vegetation was also investigated for the study period and increasing temperature trends were observed. The study found that 10.81% (1469.08 km2) of the area experienced substantial deterioration, while 23.57% (3202.39 km2) experienced minor degradation. The total area of degraded lands was 34.38% (or 4671.47 km2). A marginal improvement in plant cover was observed in 24.88% of the regions, while 9.69% of the regions experienced a major improvement. According to the NDVI-Rainfall relationships, the area was found to be significantly impacted by human pressures and activities (r ≤ 0.50) driving vegetation changes covering 24.67% of the total area (3352.03 km2). The area under the influence of climatic variability and change (r ≥ 0.50 ≥ 0.90) accounted for 55.84% (7587.26 km2), and the area under both climatic and human stressors (r ≥ 0.50 < 0.70) was 64%. Sustainable land management practices of conservation tillage, integrated pest management, and agroforestry help preserve soil health, water quality, and biodiversity while reducing erosion, pollution, and the degradation of natural resources. landscape restoration projects of reforestation, wetland restoration, soil erosion control, and the removal of invasive species are essential to achieve land degradation neutrality at the watershed scale.

Multiple remotely sensed datasets and machine learning models to predict chlorophyll-a concentration in the Nakdong River, South Korea.

The Nakdong River is a crucial water resource in South Korea, supplying water for various purposes such as potable water, irrigation, and recreation. However, the river is vulnerable to algal blooms due to the inflow of pollutants from multiple points and non-point sources. Monitoring chlorophyll-a (Chl-a) concentrations, a proxy for algal biomass is essential for assessing the trophic status of the river and managing its ecological health. This study aimed to improve the accuracy and reliability of Chl-a estimation in the Nakdong River using machine learning models (MLMs) and simultaneous use of multiple remotely sensed datasets. This study compared the performances of four MLMs: multi-layer perceptron (MLP), support vector machine (SVM), random forest (RF), and eXetreme Gradient Boosting (XGB) using three different input datasets: (1) two remotely sensed datasets (Sentinel-2 and Landsat-8), (2) standalone Sentinel-2, and (3) standalone Landsat-8. The results showed that the MLP model with multiple remotely sensed datasets outperformed other MLMs with 0.43 - 0.86 greater in R2 and 0.36 - 5.88 lower in RMSE. The MLP model demonstrated the highest performance across the range of Chl-a concentrations and predicted peaks above 20 mg/m3 relatively well compared to other models. This was likely due to the capacity of MLP to handle imbalanced datasets. The predictive map of the spatial distribution of Chl-a generated by MLP well captured the areas with high and low Chl-a concentrations. This study pointed out the impacts of imbalanced Chl-a concentration observations (dominated by low Chl-a concentrations) on the performance of MLMs. The data imbalance likely led to MLMs poorly trained for high Chl-a values, producing low prediction accuracy. In conclusion, this study demonstrated the value of multiple remotely sensed datasets in enhancing the accuracy and reliability of Chl-a estimation, mainly when using the MLP model. These findings would provide valuable insights into utilizing MLMs effectively for Chl-a monitoring.

Widespread homogenization in vegetation activities along the elevational gradients across the Himalaya over the past 40 years.

Mountainous regions are vital biodiversity hotspots with high heterogeneity, providing essential refugia for vegetation. However, climate change threatens this diversity with the potential homogenization of the distinct environmental conditions at different elevations. Here, we used a time-series (1985-2023) of Normalized Difference Vegetation Index (NDVI) from Landsat archives (30 m) to quantify vegetation changes across an elevation gradient on Himalaya Mountain. Our analysis revealed that over the past 40 years, the Himalayas have experienced widespread greening, accompanied by homogenization of vegetation across elevations. This homogenization, characterized by a reduction in the differences between high and low elevations, can be attributed to two main factors: (1) increased warming and a higher snowmelt rate at high elevations, facilitating rapid changes in high-elevation vegetation activities; and (2) higher anthropogenic disturbance at low and mid elevations, thus inhibiting low-elevation vegetation. These factors have resulted in a reduction of habitat differentiation along the mountain slopes, homogenizing vegetation and potentially threatening the unique biodiversity adapted to specific elevational zones. Our findings emphasize the urgent need for conservation strategies that prioritize the protection of heterogeneous mountain habitats to preserve their rich biodiversity in the face of climate change.

Landsat data reveal lake deoxygenation worldwide.

Dissolved oxygen (DO) is a fundamental requirement for the survival of aquatic organisms, which plays a crucial role in shaping the structure and functioning of aquatic ecosystems. However, the long-term DO change in global lakes remains unknown due to limited available data. To address this gap, we integrate Landsat data and geographic features to develop DO estimation models for global lakes using machine learning approaches. The results demonstrated that the trained random forest (RF) model has better performance (R2 = 0.72, and RMSE = 1.24 mg/L) than artificial neural network (ANN) (R2 = 0.66, and RMSE = 1.39 mg/L), support vector machine regression (SVR) (R2 = 0.62, and RMSE = 1.45 mg/L) and extreme gradient boosting (XGBoost) (R2 = 0.72, and RMSE = 1.29 mg/L). Then, we used the trained RF model to reveal the DO long-term (1984-2021) change in surface water (epilimnetic) of 351,236 global lakes with water area ≥ 0.1 km2. The results show that the average epilimnetic DO concentration of global lake was 9.72 ± 1.07 mg/L, with higher DO in the polar regions (latitude > 66.56 °) (10.87 ± 0.54 mg/L) and lower in the equatorial regions (-5 ° < latitude < 5 °) (6.29 ± 0.63 mg/L). We also find widespread deoxygenation in surface water of global lakes, with a rate of - 0.036 mg/L per decade. Meanwhile, the number of lakes and surface area that experiencing DO stress are continuously increasing, with rate of 39 and 212.85 km2, respectively. Our results offer a comprehensive dataset of DO variation spanning nearly 40 years, furnishing valuable insights for formulating effective management strategies, and enhancing the maintenance of the health of aquatic ecosystems.

Changes in the net primary production of ecosystems across Western Europe from 2015 to 2022 in response to historic drought events.

BACKGROUND: Ecosystem models are valuable tools to make climate-related assessments of change when ground-based measurements of water and carbon fluxes are not adequately detailed to realistically capture geographic variability. The Carnegie-Ames-Stanford Approach (CASA) is one such model based on satellite observations of monthly vegetation cover to estimate net primary production (NPP) of terrestrial ecosystems. RESULTS: CASA model predictions from 2015 to 2022 for Western Europe revealed several notable high and low periods in growing season NPP totals in most countries of the region. For the total land coverage of France, Greece, Italy, Portugal, and Spain, 2018 was the year with the highest terrestrial plant growth, whereas 2017 and 2019 were the years with the highest summed NPP across the UK, Germany, and Croatia. For most of Western Europe, 2022 was the year predicted with the lowest summed plant growth. Annual precipitation in most countries of Western Europe gradually declined from a high average rate in 2018 to a low average precipitation level in 2022. CONCLUSIONS: The CASA model predicted decreased growing season NPP of between - 25 and - 60% across all of Spain, southern France, and northern Italy from 2021 to 2022, and much of that plant production loss was detected in the important cropland regions of these nations.

Annual growth of terrestrial plant cover, also known as net primary production (NPP), must be maintained and managed by societies worldwide to meet their essential needs for food and fiber. A model of global NPP that uses monthly satellite images as inputs predicted that extreme drought in the years 2020 and 2021 across Western Europe, caused by greatly diminished precipitation totals and elevated temperatures, depressed plant production by between − 25 and − 60% compared to the previous five years in France, Italy, and Spain.

Urban sprawl impact assessment on the land surface temperature over the green capital of Gujarat using a geospatial approach.

In developing countries, examining land use land cover (LULC) change pattern is crucial to understanding the land surface temperature (LST) effect as urban development lacks coherent policy planning. The variability in LST is often determined by continuously changing LULC patterns. In this study, LULC change effect analysis on LST has been carried out using geometric and radiometric corrected thermal bands of multi-spectral Landsat 7 ETM + and 8 TIRS/OLI satellite imagery over Gandhinagar, Gujarat, in the years 2001 and 2022, respectively. Maximum likelihood classification (MLC) was applied to assess LULC change while an NDVI-based single-channel algorithm was used to retrieve LST using Google Earth Engine (GEE). Results showed a substantial change in built-up (+ 347.08%), barren land (- 50.74%), and vegetation (- 31.66%). With the change in LULC and impervious surfaces, the mean LST has increased by 5.47 ℃. The impact of sparse built-up was seen on vegetation and agriculture as a maximum temperature of > 47 ℃ was noticed in all LULC classes except agriculture, where the temperature reached as high as > 49 ℃ in 2022. Since Gandhinagar is developing a twin-city plan with Ahmedabad, this study could be used as a scientific basis for sustainable urban planning to overcome dynamic LULC change and LST impacts.

Integration of dynaMic water extents towards imProved lake wAter suRface Temperature (IMPART).

Lake water surface temperature (LWST) is a critical component in understanding the response of freshwater ecosystems to climate change. Traditional estimation of LWST estimation considers water surface bodies to be static. Our work proposes a novel open-source web application, IMPART, designed for estimating dynamic LWST using Landsat reflectance and MODIS temperature datasets from 2004 to 2022. Results presented globally for over 342 lakes reveal a root mean square deviation of 0.86 °C between static and dynamic LWST. Additionally, our results demonstrate that 57% of the lakes exhibit a statistically significant difference between the static and dynamic LWST values. Improved LWST will ultimately enhance our ability to comprehensively monitor and respond to the impacts of climate change on freshwater ecosystems worldwide. Furthermore, based on the Koppen-Geiger climate classification, our zonal analysis demonstrates the deviation between static and dynamic LWST. It identifies specific zones where considering waterbodies as dynamic entities is essential.

Spatiotemporal changes of urban heat island effect relative to land surface temperature: a case study of Jinan City, China.

The urban heat island (UHI) effect has become increasingly prevalent and significant with the accelerated pace of urbanization, posing challenges for urban planners and policymakers. To reveal the spatiotemporal variations of the urban heat island effect in Jinan City, this study utilized Landsat satellite images from 2009, 2014, and 2019, employing the classic Mono-Window algorithm to extract land surface temperature (LST). Additionally, Geodetector was introduced to conduct a detailed analysis of the relationship between LST in Jinan City and land cover types (vegetation, water bodies, and buildings). The results indicate a significant increase in the severity of the urban heat island effect in Jinan from 2009 to 2019, with the central urban area consistently exhibiting a high-intensity core heat island. Suburban areas of Jinan show a clear trend of merging their heat island effects with the central urban area. The combined area of strong cool island effect zones and cool island effect zones within water bodies reaches 89.7%, while the combined proportion of heat island and strong heat island effect zones in building areas is 62.2%. Vegetation cover (FVC) exerts the greatest influence among all factors on the intensity level of the urban heat island effect. These findings provide a reliable basis for decision-making related to urban planning and construction in Jinan City.

Environmental variables improve the accuracy of remote sensing estimation of soil organic carbon content.

Accurately and quickly estimating the soil organic carbon (SOC) content is crucial in the monitoring of global carbon. Environmental variables play a significant role in improving the accuracy of the SOC content estimation model. This study focuses on modeling methodologies and environmental variables, which significantly influence the SOC content estimation model. The modeling methods used in this research comprise multiple linear regression (MLR), partial least squares regression (PLSR), random forest, and support vector machines (SVM). The analyzed environmental variables include terrain, climate, soil, and vegetation cover factors. The original spectral reflectance (OSR) of Landsat 5 TM images and the spectral reflectivity after the derivative processing were combined with the above environmental variables to estimate SOC content. The results showed that: (1) The SOC content can be efficiently estimated using the OSR of Landsat 5 TM, however, the derived processing method cannot significantly improve the estimation accuracy. (2) Environmental variables can effectively improve the accuracy of SOC content estimation, with climate and soil factors producing the most significant improvements. (3) Machine learning modeling methods provide better estimation accuracy than MLR and PLSR, especially the SVM model which has the highest accuracy. According to our observations, the best estimation model in the study area was the "OSR + SVM" model (R2 = 0.9590, RMSE = 13.9887, MAE = 10.8075), which considered four environmental factors. This study highlights the significance of environmental variables in monitoring SOC content, offering insights for more precise future SOC assessments. It also provides crucial data support for soil health monitoring and sustainable agricultural development in the study area.

An advanced TSMK-FVC approach combined with Landsat 5/8 imagery for assessing the long-term effects of terrain and climate on vegetation growth.

Introduction: As an exceptional geographical entity, the vegetation of the Qinghai-Tibetan Plateau (QTP) exhibits high sensitivity to climate change. The Baima Snow Mountain National Nature Reserve (BNNR) is located in the south-eastern sector of the QTP, serving as a transition area from sub-tropical evergreen broadleaf forest to high-mountain vegetation. However, there has been limited exploration into predicting the temporal and spatial variability of vegetation cover using anti-interference methods to address outliers in long-term historical data. Additionally, the correlation between these variables and environmental factors in natural forests with complex terrain has rarely been analyzed. Methods: This study has developed an advanced approach based on TS (Theil-Sen slope estimator) MK (Mann-Kendall test)-FVC (fractional vegetation cover) to accurately evaluate and predict the time and spatial shifts in FVC within the BNNR, utilizing the GEE (Google Earth Engine). The satellite data utilized in this paper consisted of Landsat images spanning from 1986 to2020. By integrating TS and MK methodologies to monitor and assess the FVC trend, the Hurst index was employed to forecast FVC. Furthermore, the association between FVC and topographic factors was evaluated, the partial correlation between FVC and climatic influences was analyzed at the pixel level (30×30m). Results and discussion: Here are the results of this research: (1) Overall, the FVC of the BNNR exhibits a growth trend, with the mean FVC value increasing from 59.40% in 1986 to 68.67% in 2020. (2) The results based on the TS-MK algorithm showed that the percentage of the area of the study area with an increasing and decreasing trend was 59.03% (significant increase of 28.04%) and 22.13% (significant decrease of 6.42%), respectively. The coupling of the Hurst exponent with the Theil-Sen slope estimator suggests that the majority of regions within the BNNR are projected to sustain an upward trend in FVC in the future. (3) Overlaying the outcomes of TS-MK with the terrain factors revealed that the FVC changes were notably influenced by elevation. The partial correlation analysis between climate factors and vegetation changes indicated that temperature exerts a significant influence on vegetation cover, demonstrating a high spatial correlation.

Mapping agricultural tile drainage in the US Midwest using explainable random forest machine learning and satellite imagery.

There has been an increase in tile drained area across the US Midwest and other regions worldwide due to agricultural expansion, intensification, and climate variability. Despite this growth, spatially explicit tile drainage maps remain scarce, which limits the accuracy of hydrologic modeling and implementation of nutrient reduction strategies. Here, we developed a machine-learning model to provide a Spatially Explicit Estimate of Tile Drainage (SEETileDrain) across the US Midwest in 2017 at a 30-m resolution. This model used 31 satellite-derived and environmental features after removing less important and highly correlated features. It was trained with 60,938 tile and non-tile ground truth points within the Google Earth Engine cloud-computing platform. We also used multiple feature importance metrics and Accumulated Local Effects to interpret the machine learning model. The results show that our model achieved good accuracy, with 96 % of points classified correctly and an F1 score of 0.90. When tile drainage area is aggregated to the county scale, it agreed well (r2 = 0.69) with the reported area from the Ag Census. We found that Land Surface Temperature (LST) along with climate- and soil-related features were the most important factors for classification. The top-ranked feature is the median summer nighttime LST, followed by median summer soil moisture percent. This study demonstrates the potential of applying satellite remote sensing to map spatially explicit agricultural tile drainage across large regions. The results should be useful for land use change monitoring and hydrologic and nutrient models, including those designed to achieve cost-effective agricultural water and nutrient management strategies. The algorithms developed here should also be applicable for other remote sensing mapping applications.

Geological controls of mineralization occurrences in the Egyptian Eastern Desert using advanced integration of remote sensing and magnetic data.

This study presents a comprehensive analysis of mineralization exploration in the Egyptian Eastern Desert (ED), one of the most sought-after areas for those interested in mining industry, by integrating Landsat-9 images and geophysical magnetic data. Employing advanced techniques like Principal Component (PC) analysis, Minimum Noise Fraction (MNf) transform, and Band-Ratio (B-Ratio), the research focuses on mapping lithological units, hydrothermal alteration regions, and structural elements. Composite images derived from specific PC, and MNf bands, and B-Ratio exhibit superior lithological unit identification. The findings emphasize that there are significant variations in the types of rocks extend from the southern to the northern parts of the ED. Hydrothermal alteration mapping, guided by B-Ratio results, aids qualitative lithological discrimination. A novel false color composite image optimizes Landsat-9 B-Ratios, enhancing rock unit discrimination. Correlation analyses reveal associations between mineralization types and major lithological units, while exploration of the magnetic anomaly map highlights its role in correlating mineralization sites. Structural features, analyzed through Center for Exploration-Targeting Grid-Analysis (CET-GA) and Center for Exploration-Targeting Porphyry-Analysis (CET-GA) with Tilt Derivative of RTP (TDR) techniques, contribute to a robust association between regions with medium to high structural density and porphyry intrusions and mineralization. The study significantly supports the advanced exploration geoscience, providing insights into the geological structures and dynamics governing mineralization in the Egyptian ED.

Multitemporal monitoring of paramos as critical water sources in Central Colombia.

Paramos, unique and biodiverse ecosystems found solely in the high mountain regions of the tropics, are under threat. Despite their crucial role as primary water sources and significant carbon repositories in Colombia, they are deteriorating rapidly and garner less attention than other vulnerable ecosystems like the Amazon rainforest. Their fertile soil and unique climate make them prime locations for agriculture and cattle grazing, often coinciding with economically critical deposits such as coal which has led to a steady decline in paramo area. Anthropic impact was evaluated using multispectral images from Landsat and Sentinel over 37 years, on the Guerrero and Rabanal paramos in central Colombia which have experienced rapid expansion of mining and agriculture. Our analysis revealed that since 1984, the Rabanal and Guerrero paramos have lost 47.96% and 59.96% of their native vegetation respectively, replaced primarily by crops, pastures, and planted forests. We detected alterations in the spectral signatures of native vegetation near coal coking ovens, indicating a deterioration of paramo health and potential impact on ecosystem services. Consequently, human activity is reducing the extent of paramos and their efficiency as water sources and carbon sinks, potentially leading to severe regional and even global consequences.