Greenhouse Gas Research: Test Runs and Analyzer Adventures

The first half of summer research has flown by with all sorts of adventures, including nighttime sampling, pool noodle shopping, and troubleshooting a faulty detector on the greenhouse gas analyzer.   A few weeks ago I added the finishing touches to my chambers by installing tiny fans inside the head-space and covering the outside with a sheet of aluminum foil.   The fans dangle inside the chamber and circulate the gases that accumulate, while the aluminum foil prevents the chamber from getting too hot by reflecting sunlight.  The finished products look like a cross between an elementary school science fair project and props from a low-budget sci-fi film. Nevertheless, the chambers appear to be functional and have proven easy to use.

Three of the four chambers, tethered to the Keck Dock

Three of the four chambers, tethered to the Keck Lab dock


With my flux chambers fully outfitted, I began sampling on Lake Matoaka to get an idea of what level of emissions we might see and to test how a few factors might influence greenhouse gas fluxes.  Sampling with my flux chambers takes just over an hour each time.  After plugging in the fan, I lower the chamber until it is floating on the water surface, with about 8 cm of bucket submerged.  After making sure everything is sealed, I use a syringe to take a sample from a valve fitted with a stopcock.  This is my “time 0” sample, which gives the concentrations of gases that were already present in the chamber when the sampling started. By taking samples at 20 minutes, 40 minutes, and 60 minutes, I can calculate the rate that gases are being released from the water.

I’ve spent the first several weeks experimenting with my chambers to test a few different questions.  First I started with the chambers tied to the Keck lab dock to check for consistency and to see what concentrations of greenhouse gases might be being produced on the edge of the lake.  Next I paddled out to the floating “algae dock” in the center of the lake and launched my chambers from there two mornings, investigating how emissions might be different in the center of the lake.  Lastly I ran my chambers three times from the boathouse dock near the Matoaka amphitheater, once at 9:00 AM, once at 3:00 PM and once at 11:00 PM.  This let me to look into how time of day might affect emissions from the lake, and what trends might be present during night vs. day.  (Special thanks to my roommate Julia for being my nighttime sampling assistant!)

Setting up chambers for sampling at night from the boathouse dock

Setting up chambers for sampling at night from the boathouse dock


While these short trials were mostly just to get a feel for running my chambers and much more research is needed in order to conclude anything about greenhouse gas emissions from Lake Matoaka, there were a few interesting observations from my first trials about each of the three greenhouse gases:

Methane (CH4):  Methane can be released by diffusion (steady release into the air from the water column) or ebullition (bubbles coming directly from the sediments and causing bursts of methane release).  In some areas near the edge of the lake, these bubbles were very visible while sampling, and corresponded to high fluxes of methane that varied a lot from place to place.  In the center of the lake, we saw lower methane fluxes that seemed to be dominated by diffusion, with near-perfectly linear rates of methane accumulation.   Warmer temperatures during sampling also seemed to give higher methane fluxes.

Carbon Dioxide (CO2):  While CO2 seemed to accumulate linearly during cool, shady mornings, there was more variation in CO2 fluxes at other times of day.  In both the afternoon and night sample from the boathouse dock, CO2 fluxes were negative, indicating that the lake was actually absorbing CO2 during this time, likely due to algae taking in CO2 during photosynthesis throughout the day.   A lot of different factors appear to be working together to control CO2 fluxes, likely including sunlight, temperature, nutrients and more.

Nitrous Oxide (N2O):  Only very low concentrations of N2O were detected, and these concentrations showed no clear increase over time.  It’s likely that Lake Matoaka isn’t producing any significant quantities of nitrous oxide, but smaller ponds with high levels of nitrogen (such as the Crim Dell) may have more favorable conditions for N2O production.

Here are two graphs from my time-of-day experiment at the boathouse dock.  Methane peaked in the afternoon, while carbon dioxide flux was only positive in the morning.

GHG graphs4

Inevitably, the project was stalled last week due to a small roadblock: the nitrous oxide detector on the greenhouse gas analyzer starting showing signs of an issue.  After emails, calls, and visits from tech support, the problem hasn’t entirely been resolved, but the analyzer has been deemed “stable” to keep running samples, so long as we keep an eye on how much the nitrous oxide standards are drifting throughout the day.   The faulty detector has been an opportunity for me to learn a lot more about how the analyzer functions (and what might make it cease to function).  Fingers crossed that it begins behaving better in the next week, and that we can figure out what might be the root cause of the problem!

In the meantime, sampling will resume.  With test runs in Lake Matoaka behind us, its time to do some experiments on William and Mary’s other favorite body of water: the Crim Dell.  Because it is shallower, smaller, and has higher inputs of nitrogen, it will be interesting to see how the Crim Dell differs from Lake Matoaka in its greenhouse gas fluxes.  Results coming soon!



  1. acgerhard says:

    I’m curious- did you have to take any special precautions in deploying the floats to make sure that the sediment wasn’t disturbed unduly, or are the floats far enough off of the lake’s bottom to avoid that issue altogether?

  2. carolina says:

    Good question! So far I’ve been deploying my chambers in areas where the water is deep enough so that the the sediment isn’t disturbed. The chambers are only submerged 8 cm, so dragging on the bottom hasn’t been an issue. But the sediment is very sensitive to disturbance. If I touch my paddle to the bottom, or if a turtle or fish swims down into the sediment, a very visible wave of bubbles is released.