Lake surface temperature (LST) is an important attribute that highlights regional weather and climate variability and trends. The spatial resolution and thermal sensors on Landsat platforms provide the capability of monitoring the temporal and spatial distribution of lake surface temperature on small- to medium-sized lakes. In this study, a retrieval algorithm was applied to the thermal bands of Landsat archives to generate a LST dataset (North Slave LST dataset) for 535 lakes in the North Slave Region (NSR) of the Northwest Territories (NWT), Canada, for the period of 1984 to 2021. North Slave LST was retrieved from Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI) and the Thermal Infrared Sensor (TIRS); however, most of the dataset was created from the thermal bands of Landsat 5 (43 %) due to its longevity (1984–2013). Cloud masks were applied to Landsat images to eliminate cloud cover. In addition, a 100 m inward buffer was applied to lakes to prevent pixel mixing with shorelines. To evaluate the algorithm applied, retrieved LST was compared with in situ data and Moderate Resolution Imaging Spectroradiometer (MODIS) LST observations. A good agreement was observed between in situ observations and North Slave LST, with a mean bias of 0.12 ^∘C and a root mean squared deviation (RMSD) of 1.7 ^∘C. The North Slave LST dataset contains more available data for warmer months (May to September; 57.3 %) compared to colder months (October to April). The average number of images per year for each lake across the NSR ranged from 20 to 45. The North Slave LST dataset, available at https://doi.org/10.5683/SP3/J4GMC2 (Attiah et al., 2022), will provide communities, scientists, and stakeholders with spatial and temporal changing temperature trends on lakes for the past 38 years.

The Canadian Lake Ice Model has been adapted and distributed at a high spatial resolution (∼50 m) to simulate Lake Ice Thickness (LIT) and phenology across small to medium‐sized lakes in this study. The model was applied to simulate the spatial variability of LIT, Ice Cover Duration (ICD), and Lake Ice Phenology across 500 predominantly small to medium‐sized lakes in the North Slave Region (NSR) of the Northwest Territories (NWT), Canada, from 1984 to 2022. The model utilizes a 30 m grid lake surface temperature (LST) data set derived from the North Slave LST data set, along with climate inputs from the European Center for Medium‐Range Weather Forecasts Reanalysis v5 (ERA5), including wind speed, relative humidity, snowfall, and cloud cover. These inputs provide surface fluxes to the model, driving its unsteady heat equation to produce daily LIT, annual ICD, and annual freeze‐up and break‐up dates on a ∼50 m grid resolution. Validation against in situ measurements shows a root mean square deviation of 2.7–7 cm for LIT. Trend analysis revealed a significant decline in LIT (−0.26 cm/year to −0.10 cm/year) and ICD (−0.40 days/year to −0.15 days/year). The highest rates of LIT decline are observed in the early and later months of ice formation. Freeze‐up timings are primarily influenced by depth, while geographic factors such as latitude and elevation affect break‐up timing in the NSR. This distributed modeling approach provides a robust assessment of lake ice variability and trends under changing climate and weather conditions.

1. Urban heat anomalies, driven by land cover transformation and human activity, are an increasingly urgent challenge in rapidly urbanizing regions worldwide (Marchiori et al., 2012; Nasar-U-Minall...

Turbidity is a crucial indicator of water quality and ecosystem health in large northern lakes. However, logistical reasons, high costs and the spatial complexity of lakes result in a scarcity of observations, hindering long-term monitoring efforts. In this study, we develop a dual-threshold Random Forest (RF) model optimized for low- and high-turbidity regimes to retrieve turbidity from Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) surface reflectance and band ratios over Great Slave Lake (GSL) from 2002 to 2024. Our objectives were to create a turbidity dataset from MODIS and VIIRS, assess sensor performance against in-situ measurements and among sensors, and investigate the spatial, seasonal, and interannual dynamics of turbidity. The trained RF models performed robustly across both sensors and turbid regimes: R² = 0.80 - 0.83, r = 0.89-0.91, IA = 0.94-0.95, MBE within ±0.30 NTU, with RMSE 2.3-2.8 NTU (low-turbidity areas) and 11.4 -14.1 NTU (high-turbidity areas). VIIRS consistently slightly outperformed MODIS, with lower RMSE and higher R², and has consistently higher correlations with input ratios. We observe a significant increasing trend of turbidity across the lake up to ~2 NTU/yr. Both sensors identified persistent high-turbidity zones in GSL’s southeastern region and nearshore areas, associated with river inflow from the Slave River, shallow bathymetry, and wind-driven resuspension, with plume extent expanded during high-inflow years such as 2020 and 2022. Collectively, these results provide the first harmonized continuous MODIS-VIIRS turbidity record for GSL (https://doi.org/10.20383/103.01563), quantify sensor-specific performance, and link satellite retrievals to primary limnological drivers.

Remote sensing data, particularly from Sentinel-2 satellites, offers valuable insights into surface mining activities. This study evaluates the effectiveness of Sentinel-2 imagery in detecting and monitoring sand and gravel extraction sites across five mining areas in Schleswig-Flensburg, Germany, from 2015 to 2019. Three machine learning algorithms—Random Forest (RF), Support Vector Machines (SVM), and Artificial Neural Networks (ANN)—were compared to determine the most accurate classification approach. Models were trained using three different training data scenarios to assess their performance. While RF and SVM demonstrated greater robustness than ANN, SVM outperformed the others when validated against ground truth data. Consequently, an optimized SVM-based classification model was developed and implemented in R to analyze temporal changes in the study areas over five years. The findings highlight the potential of integrating Sentinel-2 imagery with machine learning techniques for accurate and efficient monitoring of surface mining activities, offering a scalable approach for environmental management and land-use planning.