Measuring inhospitable lakes
“Are you sure this is the ramp?” my colleague, Dr. Jake Beaulieu asked the head field researcher, Adam Balz, as we drove up to the site.
The three of us took in the view of the crumbling asphalt inclined plane that disappeared into the lake. According to the map, this was the boat launch. But the usage of the lake had changed from allowing motorized craft to “paddle craft only” several years ago, and now the disused ramp was covered in layers of sandy sediment.
“I’m afraid this is our best bet” Adam responded.
We backed the SUV with the boat on its trailer as far down the former ramp as we could, weighing our worry about getting the boat into the water against our worry about failing to get the SUV back out of the water. But we did it – after much shifting and shoving the boat was free, and we were in the clear to start our main task: measuring emissions of greenhouse gases from this lake, one of 32 we planned to survey that summer.
Why did we choose such an inconvenient lake? The 32 lakes included in our study were carefully picked so that we could test a hypothesis: do the greenhouse gas emissions of a given lake depend on that lake’s productivity? Productivity in this context means the amount of plant and algal growth in the lake due to elevated nutrient levels, also referred to as eutrophication.
Connecting the dots between cause and effect
High productivity in lakes often leads to favorable conditions for greenhouse gas production. The most important impact of increased plant and algal growth is on the production and emission of methane, a greenhouse gas that is more potent than carbon dioxide: a given amount of methane causes 28 times more warming than the same amount of carbon dioxide over 100 years, on average. When algae and other aquatic flora die, they sink to the bottom of the lake. Bacteria at the bottom of the lake decompose this biomass, which (1) depletes oxygen levels, and (2) provides a ready source of carbon, two effects that enhance conditions for methane production by specialized microbes.
The 32 Midwest lakes we chose to measure had different amounts of agriculture in their watersheds. We expected that lakes draining a lot of farmland would be more productive and have higher greenhouse gas emissions than lakes draining forests and grasslands. This is because a portion of the fertilizers applied to provide nutrients for crops are washed away when it rains, and eventually those nutrients end up in streams, rivers, lakes, and even the ocean.
Lakes going green: not a good thing for their carbon footprint
Globally, lakes emit the equivalent of ~ 20% of total fossil fuel greenhouse gas emissions, and lake emissions are predicted to increase as eutrophication levels continue to rise. This finding was reported in a recent paper that combined the 32 lake measurements we collected (including the measurements from the inhospitable “boat ramp” lake) with hundreds of other data sets of lake greenhouse gas measurements (DelSontro et al., 2018).
The authors used computer simulations to look at the relationship between productivity and greenhouse gas emissions. For methane, they found that chlorophyll, the green pigment in plant cells that absorbs light for photosynthesis, was a strong predictor of emissions: a 10-fold increase in chlorophyll levels lead to a six-fold increase of methane emissions. That is, an increase in algae tends to cause an increase in greenhouse gas emissions.
We can’t measure everywhere all the time: representativeness reduces bias
The relationships Dr. Tonya DelSontro’s paper establishes between lake characteristics and greenhouse gas emissions has two key implications when it comes to our understanding of the role lakes play in climate change.
The first is improving our current global estimate of total greenhouse gas emissions from lakes. Because we can’t measure greenhouse gas emissions from all the lakes in the world (more than 100 million!), the estimate of total emission relies on an “upscaling” approach. The traditional approach is to take the dataset of all the measured (and reported) greenhouse gas emission rates per unit area, average them, then multiply that average by an estimate of the global total lake surface area.
A potential problem with this traditional approach is if your dataset is biased in some way: What if the dataset includes mostly highly productive lakes, because those tend to be closer to cities (and universities), thus more convenient to measure?
DelSontro and her coauthors took this into consideration, employing an upscaling method that weighed lakes according to their productivity and their size. Their results indicated lower overall greenhouse gas emission rates from lakes than the previous best estimate based on the traditional approach.
Improving our ability to predict future emissions is the second major advantage of establishing a relationship between lake greenhouse gas emissions and productivity. The modeled relationship between chlorophyll levels and greenhouse gas emissions enables us to predict what greenhouse gas emissions will likely be in response to different future scenarios of increasing or decreasing lake productivity due to changes in farming practices and climate.
The challenge continues: still hard, inhospitable work ahead
All 32 of the lakes in our survey were measured in the summer. Logistics, safety concerns, lack of training, and researcher preference (and comfort!) are some of the barriers to making year-round measurements of lake emissions. Similar scenarios exist in many other fields as well. There are still so many exciting frontiers in science that need creative, adventurous minds to find solutions, estimate the bias, and sometimes just do the hard work needed to understand our environment and the change it is undergoing.
Disclaimer: This work is not a product of the United States Government or the U.S. Environmental Protection Agency. The author is not doing this work in any governmental capacity. The views expressed are his/her own and do not necessarily represent those of the United States or U.S. EPA.