In this study, we employ in situ, meteorological, and remote sensing data to estimate chlorophyll-a concentration at different depths in a South American freshwater ecosystem, focusing specifically on a lake in southern Chile known as Lake Maihue. For our analysis, we explored four different scenarios using three deep learning and traditional statistical models. These scenarios involved using field data (Scenario 1), meteorological variables (Scenario 2), and satellite data (Scenarios 3.1 and 3.2) to predict chlorophyll-a levels in Lake Maihue at three different depths (0, 15, and 30 m). Our choice of models included SARIMAX, DGLM, and LSTM, all of which showed promising statistical performance in predicting chlorophyll-a concentrations in this lake. Validation metrics for these models indicated their effectiveness in predicting chlorophyll levels, which serve as valuable indicators of the presence of algae in the water body. The coefficient of determination values ranged from 0.30 to 0.98, with the DGLM model showing the most favorable statistics in all scenarios tested. It is worth noting that the LSTM model yielded comparatively lower metrics, mainly due to the limitations of the available training data. The models employed, which use traditional statistical and machine learning models and meteorological and remote sensing data, have great potential for application in lakes in Chile and the rest of the world with similar characteristics. In addition, these results constitute a fundamental resource for decision-makers involved in the protection and conservation of water resource quality.

Patagonia is commonly seen as an exceptionally pristine area because of its wildlife and practically unpolluted waters. However, during the twentieth century the burning of natural forests was one of the most important human activities in Northern Chilean Patagonia. Some estimations indicate that three million hectares were burned during the first three decades of the century. Hence the objective of this study was to evaluate the impacts of the historical fires in Lake Burgos (45º42′S) and Lake Thompson (45º38′S) in Chilean Patagonia. The impact was measured by evaluating chironomid assemblage since they are sensitive enough to be used as an indicator of aquatic ecosystem health. Fires have a direct and drastic effect on a lake watershed but also indirectly affect a lake ecosystem, changing sedimentation patterns or increasing nutrient inputs. In the studied lakes the periods with higher prevalence of fires were identified by charcoal analysis, while organic matter and magnetic susceptibility allowed the confirmation of pre-fire and post-fire periods. The chironomid composition was evaluated through a PCA and an analysis of similarity (ANOSIM) to test the significance among periods while a Detrended Correspondence Analysis was applied to the chironomid assemblage downcore to assess compositional structure and taxa turnover. In Lake Burgos the ANOSIM test indicated significant differences between the pre-fire and fire periods (p < 0.05), while in Lake Thompson differences were not significant. However, in Lake Thompson the PCA clearly separated the pre-fire from the fire period but not the fire from the post-fire periods. In both lakes chironomid composition changed in relation to the period of higher prevalence of fires, which in turn implies catchment changes, pollution, and other anthropogenic impacts. Particularly a marked change in mesotrophic/eutrophic taxa was detected, reflecting an increase in nutrient input due to deforestation. Our findings point out that the lacustrine ecosystems are still affected by the impact of fires and the subsequent increase in nutrient supply that occurred almost 50 years ago. No sign of reverting to pre-disturbance conditions was observed, which makes these lakes highly sensitive to current human-induced impacts.

Stable isotopes in mollusk shells have been widely used for hydrological balance reconstructions. However, their use is restricted to lakes that preserve the calcareous material of the shells. Lago Cisnes is located in Patagonia (47°S) and shows a continuous record of three species of mollusk during the past 5000 years. The isotopic records of δ¹⁸O in Pisidium sp., Lymnaea sp., and Biomphalaria sp. show discrepancies among them, which can be explained by the differential effect that evaporation has on the habitat where each species lives. Between 1800 and 500 cal. yr BP, the three species show similar isotopic variations, suggesting that climatic condition affecting in the same way the different microhabitats in the lake. Around 1700 cal. yr BP, an enrichment of ¹⁸O on mollusks shells indicates drier conditions that prevails until 1100 cal. yr BP. Later on, isotopic signal tends to decrease, suggesting a humid period between 750 and 500 cal. yr BP. Such humid conditions lasted until 170 cal. yr BP, which were evidenced only by Biomphalaria sp. and Pisidium sp. Climate variability during the late Holocene in Lago Cisnes is in agreement with marine records from northern Patagonia, which would suggest westerlies weakening during a northward migration after 1100 cal. yr BP and/or an important temperature control on the evaporation, where low temperatures could decrease the evaporation driven by the westerlies. Additional records in this area would be requested to clarify the westerlies effects on the east flank of the Andes.

Paleolimnological reconstructions from the mid and high latitudes in the Southern Hemisphere are still relatively scarce. Anthropogenic impacts have evidenced trophic state changes and an increase in cyanobacterial blooms in the lacustrine system of San Pedro de la Paz in the last decades. Here, we reconstructed primary production and sedimentological changes spanning the past 2500 years in two coastal lakes in Mediterranean Chile. A multiproxy approach including sedimentological, biogenic silica, carbon and nitrogen isotopes and fossil pigments analysis in sediment cores was performed in Laguna Grande (LGSP) and Laguna Chica de San Pedro (LCSP). A marked change in the sedimentology of the lakes, likely related to the terrigenous sediment inputs derived by a transition from an arid condition in the mid-Holocene to a more humid condition in the late Holocene that favoured arboreal forest establishment at 100 BC–AD 150. A period of low primary production was identified between 850 to 1050 AC for LCSP, suggesting moist and cold conditions that were possibly related to La Niña events. In recent decades, there have been increases in primary production, probably resulting from anthropogenic disturbances. These likely include the clearance of native vegetation, the introduction of exotic tree species, and urbanisation, which in turn, resulted in nutrient inputs and hence eutrophication. We conclude that an integrated management program for both lakes is urgently needed.