Two days ago we enjoyed an Antarctic snowfall that covered the Palmer with a thin blanket of white. Summer snow is by no means unusual in the Ross Sea, either above or below the waves. In fact, the summertime flurry of "marine snow" is a key component of the Ross Sea carbon cycle, and one of the processes that we've come here to study.
Marine snow is the term that oceanographers use to describe aggregates of sinking particles in the water column. The aggregates include the remains of jellyfish, larvaceans, salps, and other large zooplankton; clay minerals; and tiny fragments of calcite and opaline silica from shell-bearing phytoplankton (e.g., coccolithophores and diatoms). From the window of a research sub, marine snow resembles a blizzard of white against a backdrop of dark ocean water—hence the name.
Marine snow is ultimately a byproduct of photosynthesis in sunlit surface water. During photosynthesis, phytoplankters use the energy of sunlight to incorporate carbon, nitrogen, phosphorus, iron, and other inorganic materials into their tissues. When these phytoplankton die or are eaten by zooplankton, some portion of their remains begins to sink toward the ocean floor.
Quantifying the percentage of organic detritus that sinks, and the percentage that is recycled in surface water, is one goal of the IVARS project. If marine snow carries large amounts of carbon to depth, the carbon will be sequestered from the atmospheric greenhouse for thousands of years. If the carbon is recycled near the surface, it can quickly re-enter the surface ocean and atmosphere as carbon dioxide, one of the main greenhouse gases. Thus the fate of marine snow is an important parameter in global climate and global-climate models. The fate of the other marine-snow elements is also important. Recycling of nitrogen, phosphorous, and iron provides nutrients for future generations of phytoplankton.
Many factors help determine whether a particular parcel of marine snow will reach the ocean floor or be recycled near the surface. One is the relationship between the parcel's surface area and volume. Surface area is a squared function of length X width. Volume, on the other hand, is a cubed function of length X width X height. So as particle size increases, volume increases more rapidly than surface area. For marine snow particles in seawater, volume equates to mass, and surface area equates to hydrodynamic drag. Thus larger particles have a higher ratio of mass to drag, sink more rapidly, and are more likely to reach the seafloor.
That's why the joining of individual marine snow particles into aggregates is such an important process. All else being equal, an aggregate of 10 marine snow particles will sink more rapidly than the same 10 particles individually. The formation of aggregates is promoted by the sticky nature of many marine snow particles, including jellyfish tentacles, larvacean houses, and phytoplankton themselves.
Fecal pellets provide another way to package materials into faster-sinking particles. Zooplankters generate fecal pellets by grazing on phytoplankton, digesting the usable materials, and egesting waste materials in relatively large and fast-sinking pellets. The tastiness of phytoplankters is thus another important aspect to consider when studying marine snow. For example, some evidence suggests that diatoms are more palatable to zooplankton than the haptophyte Phaeocystis antarctica, and thus more likely to be eaten and packaged into fecal pellets.
IVARS scientists, most notably Dr. Vern Asper, use two different instruments to study marine snow. One is a digital camera that we winch overboard at each of the two IVARS mooring stations. The camera takes about 2 hours to complete its roundtrip to the seafloor (650 m at Xiphias, 600 m at Calinectes), and snaps about 300 pictures of marine snow en route. Each image focuses on a narrow column about 75 centimeters from the lens that is illuminated by a strobe on either side. The pictures resemble images of the night-time sky, with star-bright particles of marine snow highlighted against a background of space-black seawater. Vern uses image-analysis software to count the number of particles in each image. The software can in some cases even identify individual particles based on size and shape. Counting and comparing the abundance of particles in each frame and between multiple profiles provides a good estimate of where the aggregates are, how big they are, and how these features change in time and space.
Sediment traps provide another way to study marine snow. Vern attaches one of these large funnels 200 meters down each IVARS mooring, where they remain throughout deployment to capture the background rain of sinking particles. Each funnel discharges downward into a series of 21 small bottles mounted on a rotating carousel. Vern programs the carousel to bring a new bottle beneath the funnel opening every two days. Thus when we retrieve the sediment traps, they hold 21 separate samples, each containing a 2-day collection of settled particles. Comparing the weight of these samples helps quantify the amount of material exported from the surface layer through time.
Data from the sediment traps and camera help quantify the amount of organic matter that sinks to depth. By subtracting this quantity from phytoplankton production values in sunlit surface waters, we can estimate 'net growth', the amount of organic material that was fixed by phytoplankton during photosynthesis. All these values can be plugged into global carbon models to help refine them. Because the Southern Ocean is the world's most critical (and poorly known) component of the marine carbon cycle, a more refined understanding of the processes involved in elemental cycling here will help improve global models of man's influence on climate.