How climate change alters lake oxygen levels

[Music] Welcome to Science Sessions, the podcast of the proceedings of the National Academy of Sciences, where we connect you with Academy members, researchers, and policymakers. Join us as we explore the stories behind the science. I'm Paul Gabrielson. In the course of the seasons, lakes can undergo cycles of oxygenation as seasonal temperatures, biochemical processes, and even ice cover affect the amount of oxygen circulating in the lake. So, how does climate change alter those cycles? In a recent PNAS study, Joachim Jonssen of the University of Helsinki and colleagues studied long-term measurements of northern hemisphere lakes and found that oxygenation levels in lakes have indeed changed in recent decades with some types of lakes more affected than others. Joachim introduces to how lakes change through the year with respect to oxygenation and ice cover. Just like for us humans, oxygen is vital for the survival of many aquatic organisms like fish. Most of the world's lakes are seasonally ice covered. And so, there are seasonal cycles of temperature and light and wind and rain. In summer, there's warm, light water on top of cool, heavy water. And these layers they do not really mix in the summer. In the bottom layer, there's a limited oxygen production from photosynthesis because you're deep down it's dark. And there's also no supply from a surface layer. So, respiration, which is oxygen consumption, slowly decreases the oxygen concentration. And in the fall, when the surface cools down, it's more easy to mix the lake and for oxygen to enter from the atmosphere. And so, you see an increase in oxygen in the fall. And then when the ice comes on, it's the same essentially as in summer. The ice prevents oxygen from going in. And if there's snow on top of the ice that also prevents photosynthesis, oxygen production. So, in winter, in ice covered lakes, the oxygen goes down again. And this continues until the spring when the snow melts and the light penetrates through the ice. And photosynthesis starts again. And then when the ice melts completely, the wind mixing also introduces oxygen in the water. So, there's a seasonal cycle of oxygen decline in the summer and an increase during the fall and spring, mixing periods and then oxygen depletion again in the winter. There's also very shallow lakes, they stratify in the summer, but then they also mix again very easily. So, there the oxygen depletion is more intermittent. Before this study, what was the prevailing hypothesis about how climate change would affect lake oxygenation? So, climate change has effects in both the summer and the winter. And what is happening in summer is that the lake surface is getting warmer, which makes it more difficult to mix the lake. Summer stratification lasts longer in lakes, and therefore there's more time for oxygen depletion in the bottom waters. By the end of the summer, the oxygen concentration is lower when you look at it over many years, over decades. And our hypothesis for the winter was that the opposite is happening because the same climate warming is causing the ice cover periods to decrease rapidly. So, we have shorter ice cover in lakes. There is less time for the oxygen depletion in winter. So, by the end of winter, you'll end up with more oxygen. What did you do in your study? Can you describe the scale of the study and the sample size? We compiled a big data set of water chemistry measurements. So, oxygen also water temperature nutrients and a number of other variables of lakes across the northern hemisphere. So, seasonally ice covered lakes. In total, we collected about 6.6 million observations from over 19,000 lakes between 1960 and 2022. We did not do any measurements ourselves. This data set really represents the collective work and effort of many, many dedicated people going out on the lakes and analyzing water samples in the laboratory and then processing the data also and doing so consistently over many years. And this is costly. It's expensive to do this. But it's really important to get a baseline and to track these long-term changes that we are causing. If it wasn't for this long-term monitoring since the 1960s, we would really be at a loss to why suddenly fish winter kills are increasing or why the drinking water quality is going down. Then we did statistical analysis to see how the average concentration of dissolved oxygen under ice has changed over the years. We were also interested in how the seasonal cycle of oxygen has shifted specifically during these really critical periods of lake mixing in the fall in the spring. And also, whether we see different trends in lakes of different size or in nutrient porous versus nutrient-rich lakes. And finally, we were interested in whether the rate of oxygen consumption has changed over the years because a lake has become more nutrient-rich over time. What did you find? What was the difference between large and small lakes? Our analysis confirmed what we all read new about oxygen in lakes, which gives us some confidence in the data set. For example, under ice, oxygen concentrations are generally lower in smaller lakes and also they are lower in nutrient-rich lakes and organic-rich lakes. And as a result, the consumption rates tend to be higher in these types of lakes. And we also see very clearly the seasonal cycle of oxygen with maximum in spring and fall during a mixing and the minimum in the winter and the deice and also in the summer. So when we look at long-term trends, we saw that oxygen had increased in the larger lakes. So that's lakes over a thousand hectares in size, but it had decreased in smaller lakes less than 10 hectares in size. And this is despite a significant decrease in the duration of ice cover across lakes of all sizes. So for the larger lakes, we see a shorter ice cover and this does indeed improve oxygen availability in winter. But for the smaller lakes, we had to find an explanation for the negative trend. In the smaller lakes, it was a very strong negative oxygen trend in summer, so oxygen loss, which persists through the fall mixing period and then partly into the winter season. These lakes retain a long-term memory of oxygen conditions in the summer that does not really reset in the fall. And we also see this in the oxygen concentrations during mixing, which oxygen doesn't really reach 100% saturation in smaller lakes. And we also found that the contributing factor for this divergence of the oxygen trends is that ice cover shortening is about twice as strong in the largest lakes compared to the smallest lakes. And summer oxygen loss is generally greatest in small lakes also. We see this oxygen increase in the larger lakes because we have this significant ice cover shortening, but also the deoxygenation small lakes because of this ecological memory from the summer. What are the consequences of deoxygenation of small lakes? The consequences can be pretty severe. Deoxygenation limits the winter habitat of aquatic organisms that need oxygen to breathe. So Lake Trial prefers oxygen concentrations higher than 5 milligrams per liter. So when a greater portion of the water volume in the lake becomes oxygen depleted, they can be squeezed out of their preferred winter habitat. And in the really extreme cases where the lake becomes very oxygen depleted, you can have a mass mortality event or winter kill when many, many fish die. These events become more likely in small lakes as summer deoxygenation becomes more severe. We see an increase in the percentage of lakes that experience anoxia, so that is a complete depletion of oxygen in the bottom water. So we've gone from 13% of lakes experiencing anoxia prior to 1970 to over 50% today. Anoxia has all kinds of detrimental consequences. One is greenhouse gases, so when there is anoxia tends to be more methane production. And this methane can be released when the ice comes off in spring. Anoxia promotes the release of nutrients from the sediment. And this allows for the growth, for example, of cyanobacterial blooms or blooms of blue-green algae, which can be toxic. And all these ecological consequences also have major socioeconomic impacts. So we use lakes for recreational purposes, fishing and water sports, but also for irrigation and drinking water. So water quality is really important for human use. So people have cottageage by the lake. They go on annual fishing trips. This is all very culturally important besides from the economic importance. And our results do indicate it is possible to mitigate the long-term oxygen decline caused by this climate change by, for example, reducing the nutrient concentrations in the lake. We do not see winter oxygen loss in the really nutrient-poor small lakes, but these restoration measurements they are often costly. And restoring hundreds of thousands of small lakes, it's probably not feasible. What are the caveats and limitations of the study? This is a broad statistical analysis. So we can say with certainty how oxygen concentrations have changed or will change in a specific lake if there are no measurements. So it's a matter of probability. So future projections are limited based on our results because we don't yet know in a general sense what happens when lakes become completely ice-free. We don't have many examples of that. And so understanding what happens when a lake becomes completely ice-free is also a topic of future study or maybe a model. Thanks for tuning in to Science Sessions. You can subscribe to Science Sessions on iTunes, Spotify, or wherever you get your podcasts. If you liked this episode, please consider leaving a review and helping us spread the word.