Invasive mussels: use ANOVA or another type of analysis

#1
Hope this is right place for this:
I am a PhD student studying performance and population dynamics of invasive freshwater mussels (zebra and quagga mussels or Z and Q). I'm planning field work for my second field season and I need to determine how I will analyze the data I collect.

For my first experiment, I will have two different species of mussels arranged in groups in PVC tubes (eg ZZ, ZQ, QQ, QZ)(water is flowing in one direction). I want to see if the species at the front of the tube depletes the available food for the species at the back of the tube. I was thinking of doing an ANOVA with two fixed factors (position in the tube, and Other Species in the tube) with two levels each (front or back of tube; and Z or Q mussels). However, I fear that the fixed factor "other species" will be skewed, because really, the mussel in the back of the tube should have no effect on the mussels at the front.

Because my experience in biostats is kind of limited, I'm not sure what I should use instead. Maybe PCA?

I have other questions but I figure this one is confusing enough. Let me know if any other details are needed.

Thanks.
-N
 
#2
Maybe a one way ANOVA with four groups: z at front, q at front, x species behind z, and x species behind q?

or with three groups: front, x species behind z, and x species behind q?
 
#4
Ok this chapter is divided in two parts

Part 1 - Looking at the depletion of food particles in the water. To do this I'll be using peristaltic pumps to pull water samples from the tube at the front, middle (between the two groups of mussels), and end of the tube. This is going to be a real-time experiment, meaning I'll have to stay with the tubes for a few hours while I collect water samples. I'll use chlorophyll a content as a proxy for food level. I plan to have tubes with all combinations of species (ZQ, ZZ, QQ, QZ) as well as combinations with dead mussel shells to determine if there is any effect of nearby neighbors that are not actively filtering the water (DZ, QZ).

Part 2 - Looking at the effects of this depletion on mussels. To do this, I will deploy tubes at several locations and collect data on shell growth and biomass at the end of the season (which runs from June to October). I will use the same combinations as the short term tubes. Altogether its 18 tubes (6 treatment types, 3 replicates each).

I'm also considering deploying "double barreled" devices with two side-by-side PVC pipes with one species in each and determining their filtering abilities when they both have access to the same food availability. That is to say the mussels in single-tubes have sequential access to food, while the double-tubes would have simultaneous access.

Does this make sense?