Seasonal overturn and stratification changes drive deep-water warming in one of Earth's largest lakes

Eighty-four percent of Earth's non-frozen surface freshwater resides in the ten largest lakes. These lakes are highly sensitive to environmental changes and serve as valuable climate change sentinels. While surface water temperature warming is well-documented, our understanding of deep-water temperature changes in large lakes is limited due to a scarcity of long-term subsurface observations. This study addresses this gap by analyzing three decades of high-frequency (3-hourly and hourly) subsurface water temperature data from Lake Michigan. The primary goal is to understand how deep waters in this large lake respond to climate trends, specifically examining how changes in surface temperatures and extended summer periods affect winter subsurface characteristics. High-resolution data allows for a detailed examination of relationships between fall overturn, winter cooling period duration, and subsurface temperatures. This research is crucial because shifts in the thermal regimes of large lakes will have significant ecological consequences on the world's surface freshwater resources. The study specifically investigates how surface warming trends translate into deeper thermal changes, examining the interplay between fall overturn, winter cooling duration, and subsurface temperatures, to improve our understanding of climate impacts on this vital freshwater resource.

Previous research has shown that global lake surface water temperatures (LSWT) have warmed by an average of 0.21 °C/decade. Some lakes are warming faster than ocean and regional air temperatures. While surface warming trends are well-established through satellite and in situ measurements, subsurface data is sparse. Limited studies on deep lake subsurface temperatures exist, hindering a comprehensive understanding of climate change impacts at depth. Projected connections between climate change-induced thermal structure changes and mixing regime shifts have been explored through numerical modeling, but observations are needed to confirm these projections. Existing datasets often lack the vertical resolution, temporal frequency, or duration needed for long-term analysis. Studies in other large lakes, such as Lake Baikal, Lake Tanganyika, and Lake Superior, have documented changes in thermal structure and their impacts on ecosystems, providing context for the Lake Michigan study.

Thirty years of nearly continuous hourly and 3-hourly subsurface water temperature measurements from a high vertical-resolution thermistor string in 150 m of water in southern Lake Michigan were analyzed. Satellite-derived surface temperatures from the Great Lakes Surface Environmental Analysis (GLSEA) were used to contextualize the subsurface data. Three statistical approaches (linear regression, Theil-Sen estimator, and Seasonal Trend decomposition using Loess (STL)) were used to analyze both surface and subsurface temperature trends. Monthly trends were also analyzed to identify seasonal variations in warming and cooling rates. For the deep-water dynamics analysis, data from the 110m transect were used to determine the fall overturn date, the duration of the annual cooling period, the minimum temperature, and the duration and temperature of the summer stratification period. Overlake meteorological data (air temperature, wind speed, cloud cover, and shortwave radiation) were collected from various sources to understand the atmospheric drivers of water temperature changes. The data were analyzed using linear regressions and the STL decomposition method. A sensitivity analysis was performed to verify the robustness of the results across different depths.

The analysis revealed year-round surface warming rates at the mooring location ranging from 0.40 to 0.49 °C/decade, consistent with the lake-average trend. Subsurface warming trends were observed below the thermocline at depths of 60–100 m, with warming rates generally less than at the surface. At 110 m, the warming trend was less pronounced and showed high interannual variability. No significant trends were found at 140 m. Subsurface temperatures showed strong seasonal trends, with peak warming coinciding with fall overturn. The extended summer stratification period, indicated by warming trends in the fall at the surface, delayed the fall turnover date and resulted in delayed arrival of warmer waters at depth. This led to relative cooling trends below the thermocline in the fall and winter warming trends below 60 m from January through April. Long-term warming trends in water temperature were consistent with observed increases in overlake air temperature, wind speed, and shortwave radiation, coupled with a decrease in cloud cover. The 1997-1998 El Niño event marked a notable shift in the deep-water dynamics, with a delayed fall overturn, a shorter cooling period, and an extended summer stratification. This shift in thermal conditions was linked to changes in atmospheric conditions. The delay in fall overturn triggered a cascade of subsurface changes, including shorter cooling periods, higher minimum bottom temperatures, and increased summer stratification temperatures. The study shows how shifts in surface and subsurface water temperatures are influencing the lake's thermal structure, with implications for its ecosystem.

The findings demonstrate a clear link between surface warming trends and changes in deep-water temperatures in Lake Michigan. The observed cascade of events—delayed overturn, shorter cooling periods, higher minimum winter temperatures, and increased summer stratification temperatures—highlights the impact of climate change on the lake's thermal dynamics. These changes have significant ecological implications, potentially affecting dissolved oxygen levels, food web structures, and the prevalence of invasive species. The results are consistent with modeling studies predicting mixing regime shifts in large lakes under climate change, suggesting that Lake Michigan may be transitioning from a dimictic to a warm monomictic state. This shift could have profound consequences for the lake's ecosystem. The study highlights the importance of long-term, high-frequency monitoring of subsurface water temperatures to understand the impacts of climate change on large lakes and their ecosystems.

This study provides compelling evidence of deep-water warming in Lake Michigan driven by changes in seasonal overturn and stratification. The observed cascade effect from surface to deep water illustrates the interconnectedness of climate change impacts within large lake systems. Future research should focus on exploring the long-term ecological consequences of these thermal shifts and refining the predictive capabilities of lake mixing regime models.

The study's findings are based on data from a single location in Lake Michigan. While this location is representative of the southern basin, spatial variability in thermal responses across the entire lake may exist. The analysis relied on satellite-derived surface temperatures for part of the study period, which may have limitations in accuracy compared to in situ measurements. The study's analysis is largely descriptive, offering limited prediction about future thermal conditions.