Chlorophyll-a (Chl-a) is widely used as a proxy for phytoplankton biomass and lake trophic status. This exploratory study compared Random Forest, XGBoost, and a one-step Long Short-Term Memory (LSTM) network using 1012 station–date observations and 43 contemporaneous water quality predictors from Leon County, Florida, during 2014–2024. Initial shuffled five-fold validation favored XGBoost (mean RMSE 10.45 µg L−1; mean R2 0.484), but shared stations across folds and prior manual configuration selection limit that comparison. Additional validation held out entire stations in five outer folds and selected configurations using three station-blocked inner folds. Pooled outer RMSE was 11.92, 10.49, and 10.91 µg L−1 for Random Forest, XGBoost, and LSTM, respectively; the corresponding R2 values were 0.371, 0.513, and 0.473. Six expanding training periods ending in 2017–2022, each assessed over the next two years, evaluated sensitivity to chronological cutoff. These tests complement the original 2014–2019 versus 2020–2024 split, for which R2 was 0.054–0.071. Random Forest had the lowest RMSE in all six windows, while LSTM had negative R2 in four. Spatial transfer remained limited for Lakes Munson and Iamonia. Nested evaluation separates the new configuration search from each outer assessment, but the dataset and predictor panel had already been examined; the reanalysis is not an untouched external validation. The findings describe conditional concentration estimation using contemporaneous measurements, not advance forecasting or assured transfer across contrasting lakes.
Ancient irrigation tanks are a defining component of Sri Lanka’s freshwater landscape, but their CO2 flux dynamics remain poorly characterized and are underrepresented in carbon databases. This study quantified hourly daytime CO2 fluxes and examined associations between CO2 flux and environmental variables in Thalangama Tank (TG) and Huruluwewa Tank (HW), which are situated in contrasting hydroclimatic zones. An automated floating chamber was deployed monthly in each tank between January 2025 and January 2026 to capture the monsoon seasons, while water quality parameters and hydro-meteorological data were monitored simultaneously. TG, a small and shallow tank located in the Wet Zone, showed predominantly positive daytime CO2 fluxes across the sampled campaigns, and no significant seasonal dynamics were observed. In contrast, HW, a large but shallow Dry Zone tank, showed both daytime CO2 uptake and emission, and daytime CO2 flux varied significantly among seasons. Daytime net emission was observed during the first inter-monsoon and northeast monsoon periods, while CO2 uptake was observed during the southwest monsoon and second inter-monsoon periods. At HW, daytime CO2 flux was strongly associated with the ionic gradient, moderated by strong hydrometeorological forcing. Conversely, CO2 flux in TG was weakly related to the ionic gradient but associated with the metabolic balance and rainfall variation. These contrasting patterns suggest that hydrometeorological and physicochemical conditions are important factors associated with daytime CO2 exchange in tropical irrigation tanks. Also emphasize the potential for significant temporal variability in CO2 flux among tropical artificial freshwater systems.
Regional background concentrations of ten priority potential pollutants (Hg, Cd, Zn, Pb, Cu, Cr, Co, Ni, Fe, and Mn) were established for the first time in the water and bottom sediments of 18 lakes in the central Baraba Lowland. Comparison of natural and technogenic limnosystems using three normalization standards (regional background concentrations (RBC), upper continental crust (UCC), and Post-Archean Australian shale (PAAS)) showed that the choice of reference fundamentally influences enrichment/depletion interpretations. Concentration coefficients for Mg, Si, Sr, Pb, and Mn are standard-independent («stable»), whereas those for Cd, Sb, Cr, Cu, and Hg are standard-dependent («unstable») and require explicit specification of the reference. Despite similar bulk sediment geochemistry, spheroidal and irregular Si–Fe slag particles found in technogenic limnosystems provide direct mineralogical evidence of aerotechnogenic input. Alkaline, bicarbonate-sodium waters act as an effective geochemical barrier, sequestering elements in the solid phase and supporting self-purification. Sapropels meet Class I of State Standard GOST R 54000-2010 for all elements except Zn (Class II), confirming their resource potential under proper quality control. Future work should expand the list of analytics and include all components (water, sediments, biota, and catchment soils) to build a comprehensive model of limnosystem functioning under anthropogenic load.
This article discusses the use of thermal and mineral waters in Poland in balneology, rehabilitation, and recreation. It highlights their importance for health, economy, and tourism, as well as the country’s significant hydrogeothermal potential, particularly in the Carpathians, the Sudetes, and the Polish Lowlands. The authors present a classification of waters, their chemical and thermal properties, and their main applications: therapeutic baths, drinking cures, inhalations, peloidotherapy, and hydrotherapy. A significant portion of the study reviews the latest global research confirming the effectiveness of balneotherapy in the treatment of musculoskeletal and skin disorders, indicating the need for further clinical research. The paper also discusses Polish spas and recreational centers using thermal waters, pointing to their growing role in health care and wellness tourism. Limitations to the development of balneotherapy, such as high drilling costs, geological risks, and the need to maintain therapeutic water parameters, are highlighted.
Chlorophyll-a is a widely used indicator for assessing the trophic status and water quality of aquatic ecosystems because of its close relationship with phytoplankton biomass. This study aimed to identify the physicochemical, biological, and temporal variables associated with chlorophyll-a concentrations in Lake Gatun during the 2017–2023 period using the Light Gradient Boosting Machine (LightGBM) algorithm and a Generalized Linear Model (GLM). A total of 25 predictor variables describing water quality were analyzed using data collected from 14 monitoring stations distributed throughout Lake Gatun within the Panama Canal watershed. The LightGBM model achieved satisfactory predictive performance, with an RMSE of 4.58, an MAE of 3.26, and a coefficient of determination (R2) of 0.42 on the testing dataset. Variable importance analysis identified turbidity, dissolved oxygen, and water transparency as the most influential predictors of chlorophyll-a concentrations. These findings improve our understanding of the factors associated with chlorophyll-a variability and demonstrate the potential of machine learning models as decision-support tools for monitoring and managing aquatic ecosystems.
