Trends and abrupt changes in 104 years of ice cover and water temperature in a dimictic lake in response to air temperature, wind speed, and water clarity drivers

The one-dimensional hydrodynamic ice model, DYRESM-WQ-I, was modified to simulate ice cover and thermal structure of dimictic Lake Mendota, Wisconsin, USA, over a continuous 104-year period (1911–2014). The model results were then used to examine the drivers of changes in ice cover and water temperature, focusing on the responses to shifts in air temperature, wind speed, and water clarity at multiyear timescales. Observations of the drivers include a change in the trend of warming air temperatures from 0.081 °C per decade before 1981 to 0.334 °C per decade thereafter, as well as a shift in mean wind speed from 4.44 m s−1 before 1994 to 3.74 m s−1 thereafter. Observations show that Lake Mendota has experienced significant changes in ice cover: later ice-on date(9.0 days later per century), earlier ice-off date (12.3 days per century), decreasing ice cover duration (21.3 days per century), while model simulations indicate a change in maximum ice thickness (12.7 cm decrease per century). Model simulations also show changes in the lake thermal regime of earlier stratification onset (12.3 days per century), later fall turnover (14.6 days per century), longer stratification duration (26.8 days per century), and decreasing summer hypolimnetic temperatures (−1.4 °C per century). Correlation analysis of lake variables and driving variables revealed ice cover variables, stratification onset, epilimnetic temperature, and hypolimnetic temperature were most closely correlated with air temperature, whereas freeze-over water temperature, hypolimnetic heating, and fall turnover date were more closely correlated with wind speed. Each lake variable (i.e., ice-on and ice-off dates, ice cover duration, maximum ice thickness, freeze-over water temperature, stratification onset, fall turnover date, stratification duration, epilimnion temperature, hypolimnion temperature, and hypolimnetic heating) was averaged for the three periods (1911–1980, 1981–1993, and 1994–2014) delineated by abrupt changes in air temperature and wind speed. Average summer hypolimnetic temperature and fall turnover date exhibit significant differences between the third period and the first two periods. Changes in ice cover (ice-on and ice-off dates, ice cover duration, and maximum ice thickness) exhibit an abrupt change after 1994, which was related in part to the warm El Niño winter of 1997–1998. Under-ice water temperature, freeze-over water temperature, hypolimnetic temperature, fall turnover date, and stratification duration demonstrate a significant difference in the third period (1994–2014), when air temperature was warmest and wind speeds decreased rather abruptly. The trends in ice cover and water temperature demonstrate responses to both long-term and abrupt changes in meteorological conditions that can be complemented with numerical modeling to better understand how these variables will respond in a future climate.

Trends and abrupt changes in 104-years of ice cover and water temperature in a dimictic lake in response to air temperature, wind speed, and water clarity drivers

The one-dimensional hydrodynamic-ice model, DYRESM-WQ-I, was modified to simulate ice cover and thermal structure of dimictic Lake Mendota, WI, USA, over a continuous 104-year period (1911–2014). The model results were then used to examine the drivers of changes in ice cover and water temperature, focusing on the responses to shifts in air temperature, wind speed, and water clarity at multi-year time scales. Observations of the drivers include a change in the trend of warming air temperatures from 0.081 °C per decade before 1981 to 0.334 °C per decade thereafter, as well as a shift in mean wind speed from 4.44 m s−1 to 3.74 m s−1 in 1994. Observations show that Lake Mendota has experienced significant changes in ice cover: later ice on (9.0 days later per century), earlier ice-off (12.3 days per century), decreasing ice cover duration (21.3 days per century), while model simulations indicate a change in maximum ice thickness (12.7 cm decrease per century). Model simulations also show changes in the lake thermal regime of: earlier stratification onset (12.3 days per century), later fall turnover (14.6 days per century), longer stratification duration (26.8 days per century), and decreasing summer hypolimnetic temperatures (−1.4 C per century). Correlation analysis of lake variables and driving variables revealed ice cover variables, stratification onset, epilimnetic temperature, and hypolimnetic temperature were most closely correlated with air temperature, whereas freeze-over water temperature, hypolimnetic heating, and fall turnover date were more closely correlated with wind speed. Each lake variable (i.e., ice-on and ice-off dates, ice cover duration, maximum ice thickness, freeze-over water temperature, stratification onset, fall turnover date, stratification duration, epilimnion temperature, hypolimnion temperature, and hypolimnetic heating) was averaged for the three periods (1911–1980, 1981–1993 and 1994–2014) delineated by abrupt changes in air temperature and wind speed. Average summer hypolimnetic temperature and fall turnover date exhibit significant differences between the third period and the first two periods. Changes in ice cover (ice-on and ice-off dates, ice cover duration, and maximum ice thickness) exhibit an abrubt change after 1994 which was related in part to the warm El Niño winter of 1997–1998. Under-ice water temperature, freeze-over water temperature, hypolimnetic temperature, fall turnover date, and stratification duration demonstrate a significant difference in the third period (1994–2014), when air temperature was warmest and wind speeds decreased rather abruptly. The trends in ice cover and water temperature demonstrate responses to both long-term and abrupt changes in meteorological conditions that can be complemented with numerical modelling to better understand how these variables will respond in a future climate.

Partitioning Light Attenuation in an Acidic Lake

Although a number of researchers have reported that acidification of lakes is accompanied by an increase in transparency, there has been no systematic evaluation of the processes responsible for this transformation. In this study we partitioned the attenuation of light in acidic Dart's Lake, located in the Adirondack region of New York from May to September 1982. We observed that changes in light attenuation (Kd) and light absorption (a) were regulated largely by "gelbstoff." Substantial decreases in Kd and a occurred through the study period and were correlated with a depletion in the concentration of dissolved organic carbon (DOC). In-lake concentrations of DOC were controlled by terrigeneous loading and in-lake processes. The decrease in DOC and the attendant decreases in a and Kd were coupled to a loss of Al from the water column of the lake. We suggest that coagulation/adsorption of DOC by Al may have contributed to increases in lake clarity. Increased transparency is significant because it enhances hypolimnetic heating and decreases the thermal stability of lakes.

Effects of Changes in pH on Transparency and Thermal Regimes of Lohi Lake, near Sudbury, Ontario

Changes in Secchi transparency (depth) and thermal regimes accompanying experimental neutralization and subsequent re-acidification of Lohi Lake, near Sudbury, Ontario are described. Secchi depth was not correlated with chlorophyll concentrations, phytoplankton biomass or total phosphorus, but it was negatively correlated with pH. Secchi depth was positively correlated with thermocline depth, epilimnetic thickness, and hypolimnetic heating rates suggesting that acidification should result in deeper mixing zones and warmer hypolimnia in lakes. Several possible implications of changes in transparency and thermal regimes produced by acidification are presented.