More Results and Final Thoughts

The last part of my project was to perform statistical tests with the Tmax and Salpa thompsoni abundance values. The goal was to look for significant spatial variation for either variable as well as relationships between the two throughout the 21-year time series. I cut my data set down for these tests in hopes that I could isolate the effect of Upper Circumpolar Deep Water from the many other environmental parameters that influence salps. I chose to focus on the middle part of the sampling grid: the 400, 300, and 200 lines. In the north salps are highly abundant because of the ice-free conditions, and in the far south salps are particularly rare due to greater sea ice extent.

I compared each sampling line’s shelf stations using a Mann-Whitney rank sum test. This is a nonparametric statistical test that compares medians rather than means. I could not use standard T-tests because of the skewed nature of my data set. The 300 line had median Tmax values significantly greater than both the 200 and 400 line shelf stations (P < 0.01). Additionally, Mann-Whitney tests show 300 line Salp thompsoni densities to be significantly higher than those on the 200 line (P = 0.0001) but not the 400 line (P = 0.121). These results are not very clear, but I also examined frequency of UCDW and salp presence. 49% of CTD casts on the 300 line shelf had a Tmax greater than 1.6 degrees Celsius. The next highest frequency of this warm water was the 400 line with only 29% presence. Additionally, salps were present in 58% of 300 line shelf stations samples, a higher percentage than every other except for the northernmost line. This led me to look at salp presence trends on the 300 line in relation to UCDW presence. A Pearson correlation showed that the number times salps were present on the 300 line in a given year was positively related to the number of stations with a Tmax greater than 1.6 degrees Celsius (r = 0.492; P = 0.0233). There was not a similar significant correlation on other sampling lines, suggesting that UCDW may be having an effect on the 300 line salp population in particular.

What does this mean in trying to find a mechanistic relationship between UCDW and salps of the Western Antarctic Peninsula shelf? I think that the most likely connection is UCDW flooding events transporting animals from the Antarctic Circumpolar Current into shallower waters where they are less common. The 300 line is the key location for UCDW intrusions. It has does not have particularly high salp densities, but the animals are present very often. The density numbers on the 300 line would probably be higher if UCDW intrusions were prompting previously present salps to begin reproducing, However the elevated frequency of salp presence, particularly in years of UCDW intrusions, makes it likely that the water mass is affecting the gelatinous critters.

Salps are common on the 300 line shelf, but not in high density like they were here


I learned so much working on this project and had a ton of fun. Thank you to everyone who helped me along the way, particularly to Dr. Debbie Steinberg and to the Monroe Program for giving me this opportunity.

The 2015 PAL LTER zooplankton team. Thanks to all of you!



  1. Jack, this sounds like it was a very interesting project for you to undertake. What do you think was the greatest challenge with working on a project such as this? Is there anything you wish you had done differently, looking back on your work?