Dry Valley soils are cold and salty – basically the same environmental conditions that are used for long-term storage of DNA. Sure, Dry Valley soils may be diverse, but are all of these different microbes actively playing a role in soil ecosystems? Or do they just blow in here from other places, hang out in the soil, but never actually contribute anything? Because past studies inferred diversity based on the presence of DNA, it is it possible that the diversity of microbes that play active roles in Dry Valley ecosystem functioning is only a small subset of the microbes present in the soil? Dry Valley soils look diverse, but is the diversity functionally relevant?

Suppose for a moment that we were collecting DNA from Siberian ice cores and inferring ecological function based on the taxonomic affinities of the DNA sequences. When we encounter a chunk of frozen wooly mammoth tissue what would we infer about biological diversity and ecological function? That giant herbivores are running around and shaping Siberian ecosystem processes? Clearly our inferences would be a poor reflection of what actually goes on!

Long-term climate observations by the McMurdo Long-Term Ecological Research group are revealing increased frequency and magnitude of ‘pulse- events’ – periods of rapid warming and ice melt that lead to increased liquid water moving across/through the Dry Valley landscape. Which microbes are active in dry soils? Which microbes are responding to pulse events? Which microbes are completely dormant, waiting for more favorable conditions?

To answer these questions we set up a contained experiment that will capture the microbial response to a pulse wetting event. In the experiment we wet some soils with stable isotope labeled water (O18). The microbes that respond to the wetting event by taking up the O18 will incorporate the label into their DNA. We then extract the DNA from all the microbes in the soil and separate the strands of DNA that have incorporated the label from the strands of DNA that did not incorporate the label. We then sequence the two pools of DNA to reveal which taxa responded to the pulse and which did not – essentially, when it comes to identifying who responds to pulse events, we’ll be able to distinguish the players from the poseurs. Additionally, in order to link the response to actual ecosystem processes, we are measuring CO2 flux in the soils as a surrogate of microbial activity. For example, soils pulsed with water are expected to increase in ecological activity, and thus an increase in CO2 flux through the soil ecosystem.

This involved several steps. First, the experiment was set up conducted near the McMurdo Long-term ecological research (LTER) stoichiometry plots at F6 (Lake Fryxell, Taylor Valley). We outlined a patch of soil 8 m long and 1 m wide, with 8 replicates of 1 m2. Six PVC collars were placed in each of these 8 replicates, color coded to make it easy to tell which sampling period each represented – a control that wasn’t treated at all, so we’d know what the background of each little plot was, and collars to be sampled at 12, 24, 48, 72 and 144 hours after the treatments were added. You can see the collars here:

These collars were placed into the soil, one of each color-coded sample per plot—the locations of these colors within the small plot were randomized beforehand, and here you can see Uffe setting the collars into the soil as Zach reads off where each is supposed to go.

Once the collars were in place, Diana went around with a ruler to measure how much space was between the surface of the soil and the top of the collar—this is important because the machine that Byron would later use to measure how much CO2 was emitted from each plot needs to know how much space is being measured!

Once this information was collected, Uffe and Zach added water to the plots (the treatment), and Byron added the O18 to the experimental plots. Here you can see Uffe and Zach carefully treating their collars, and in the background Byron is taking respiration measurements:

Byron then had to stay out all week (and will have to go out once more) in order to take the respiration measurements at each time period—one 12 hours after the treatments were conducted, one 24 hours later, and so on until all the appropriate readings were taken. The microbial responses are collected in a time series of 12, 24, 48, 72, and 144 hours. The DNA sequencing and analyses will be done off the ice at Brigham Young University. Here you can see him set up to take these measurements.