Taking Pictures of Particles in River Water and a Visit from NSF

This past week was very busy.  I learned how to use a new piece of equipment called a FlowCAM.  It analyzes particles in water, such as algae, bacteria, and organic matter, and provides various metrics on each particle.  It can process tens of thousands of particles within a half an hour.  The part that excites me most about this equipment is that it takes a picture of each particle that it takes measurements on.  While the system is running you can see the particles in real time.  It was very interesting to watch the particles and try to identify some of them using a referencecontaining illustrations of common cells and algae found in North America.  I learned a great deal about the operation of the FlowCAM, including the issue of it getting clogged! 

While the data produced by the FlowCAM is interesting, it is fairly meaningless unless some conclusions can be drawn from it.  However, with the FlowCAM reporting over 20 different metrics on each particle, it is hard to draw conlusions by simply plotting some data and looking for correlations.  Therefore some sort of multivariate statistical analysis is necessary.  One such method that was previously identified as a good method is principal component analysis, or PCA.  The function of PCA is to identify those variables which are causing the greatest variability in the data.  Those variables identified, in order of greatest to least contribution to variablility, are called principal components.

In order to perform PCA it is necessary to use some sort of mathematical analysis software.  In our case we have been using MATLAB.  Since I have not previously used MATLAB, I took some more time this week to work through some tutorials and familiarize myself with some of the basics so that I could understand how to perform PCA in MATLAB.  Luckily, it was a sucess and I was able to recreate some results that were obtained last year.

On Thursday CMOP experienced their annual NSF site visit.  I was fortunate enough to be able to have lunch with one of the committee members, Dr. Dana. K. Savidge.  We had a nice talk about her work in geophysics.  It was also very informative to watch the talks given by the numerous researchers at CMOP.  I was able to get a much better picture of how all sorts of projects at CMOP tie together and how great of an impact the research has on the overall understanding of coastal margin ecosystems.