header image

New survey site, habitat data, and muddy mishap

Posted by: | August 6, 2012 | No Comment |

 Last week we journeyed once again into the South Slough of Coos Bay for another Cockle sampling. This time we went to Valino Island, which represents a more heterogeneous habitat than our previous survey at Indian Point. That is, the substrate at Valino Island ranges from a softer, muddier consistency to a firmer, sandier consistency whereas Indian Point is more consistently sandy. Much of Valino Island is muddy, so the raking and walking were a bit more arduous. There was also considerably more eel grass at Valino, so we wanted to take some extra care to minimize our impact on it. Nonetheless, we managed to rake up around 120 cockles, which I measured, weighed, and put into the database. The cockles seem to be less abundant in the muddy areas with eel grass and more abundant in sandier areas closer to water sources like channels and pools. We’ve thus decided to expand the study to include habitat as a factor for cockle distribution. Since we didn’t take habitat info down in our initial surveys, we took some extra time and revisited each quadrat at each site and recorded various habitat aspects: substrate type (sand, mud, smud [sand+mud]), presence/absence of eel grass, presence/absence of gaper clams. (note: for a more detailed description of our field methods, see blog entry for 7-23-2012). In my last blog entry I mentioned starting in on some stat analyses. Due to last week’s fieldwork, we weren’t able to start much in-depth interpretation, but we did decide to collect some extra cockles at each field site to boost our sample size. We randomly selected approx 20 quadrats at each site and raked up about 70 extra cockles per site. This should help beef up our data for further analyses. So far it looks like nonparametric hypothesis tests are the way to go, because our sample sizes aren’t large enough to assume a normal distribution. Something I’ve learned through these initial stat analyses is that our sample sizes need to be very large in order for standard parametric tests (T-test, ANOVA, etc) to be valid. This morning we journeyed back to Valino Island to collect habitat data and more cockles. My confidence for walking on soft mud has increased—to a fault: at one point I let my guard down, my foot got stuck, and I fell right on my backside in the mud. Luckily my chest waders saved me from total disaster. Nothing hurt but my pride, and I may have an interesting photo to share. The rest of the week I’ll be generating pivot tables, which are very helpful/powerful when looking at data (more about those later when I learn more) and doing some more in-depth stat analyses with the new habitat info.

Print Friendly, PDF & Email
under: Eric Post

Leave a response

Your response: