Palo Verde, pt. 3: Faculty-led projects

Tuesday 2/22 through Thursday 2/24, we worked on faculty-led projects with Jordan, a professor from Tulane, and Ignacio, a graduate student at the University of Costa Rica. On Tuesday and Wednesday we did data collection, splitting up into two groups and working on one of the two projects each day. I worked on Ignacio's project first, and collected data on Green Lynx Spiders all day. We were testing to see whether spiders gain experience with age by looking for relationships between spider age (juvenile/adult) and responses when encountered with a predator.

A green lynx spider (Peucetia viridans)

Ruellia inundata, the plant where green lynx spiders live. Rather than building webs, they perch on the plants, wait for pollinators to pass by, and then pounce on them.

We did this by simulating predation with a false bird beak (aka a stick sharpened to a point and wrapped in black electrical tape) and measuring how close we could get to the spider before eliciting a response, the response distance, response type (run, jump, shift, etc.), and response location (leaf, stem, flower, or ground). 

A "simulated predation event." Chelsea is moving the stick, or model bird beak, towards the spider, and Carolyn is waiting to measure and record its response distance and type.Photo by my professor, Mau.

While 14 of us were gathering data all day, a smaller group of six was responsible for compiling all the data, writing it up, and presenting it to the class. I was in the small group for Ignacio's project, so after a long morning of finding, poking, and observing 50 or so spiders, I spent the afternoon making a giant dataset of all of our observations from day one.

On Wednesday, the other group was collecting spider data with Ignacio, and my half of the class went with Jordan to work on his project. He was observing bird behavior based on population density to see if behaviors that decrease vigilance toward predators (e.g. foraging) is more prevalent in areas of higher bird density, overall looking at the osts and benefits of social behavior in birds. We spent the day observing Northern Jacanas at three different sites in Palo Verde, focusing on one individual at a time for five minutes of observation. We worked in pairs so that one person could record the behaviors that the person with binoculars observed; however, I was working with Rukhshana, and she got sick that morning, so I ended up working in a group of three with Ashur and Inca.

A really terrible picture of Jacana spinosa, the Northern Jacana. If you look closely you can see one on the right hand side about half way down. Everything else in the picture is water hyacinth in the marsh. Northern Jacanas are actually really cool looking, and Google images can probably show you much better stuff than this!

Birdwatching in the marsh at one of our lookout points

Justin simulating the Jacana's aggressive wing-display behavior on Matt.

The big croc!

Because both small groups had to present on the research the next day, we only gathered data in the morning so we could work on analyzing the data later in the day. Although we were only out for about five hours, it was an exciting morning. First, Inca got chased by a herd of cattle that live out in the marsh to keep the invasive grasses under control, which was scary! Then, at our last observation site, we saw some crocodiles out in the marsh from the bird observation tower, which was really cool. We found a small one at first but then went back down onto the ground to tell other people, and Ashley came over and spotted one that was at least 6-8 feet long off in the distance. We watched it for a while as it swam around in the marsh and over to a flock of whistling ducks, which we thought it might try to attack. Unfortunately we didn't get to see the crocodile eat anything, but it was awesome to watch nonetheless.

After lunch,  I spent the afternoon compiling the second day's spider data and running statistical tests on the complete dataset. I gained a lot of respect for professors like Bob Kaplan at Reed, who have to piece together huge datasets for class-wide ecology experiments... it's hard work! You have to be very careful to try to notice errors in data input by all of the contributors, figure out what they did wrong, and fix it, as well as making sure everyone's data is consistent in units and format. It took me a few hours just to get everyone's data into a usable form to run statistical tests, since for some of the observational components (e.g. response type) people had recorded a lot of things that were not in our list of options or were entered differently. When the dataset was complete, I spent the rest of the afternoon analyzing it with eight different tests. Then after dinner my group of six met to discuss and interpret our results, exchange notes on some literature we had read, and put together our presentation to give the next day. It took a long time to get everything together, but we finished up our presentation by midnight (which is very late for us!). In the end we found that adult spiders had significantly lower predator distances at the time of response than juvenile spiders, and that while some adult spiders had no response whatsoever (they stayed still even when the "predator" was at a distance of 0), juvenile spiders always responded in some way. These suggest that the adult spiders may be better able to identify our "predators" as a non-threat, since they were not entirely realistic replicas of bird beaks and since they were approaching the spider very slowly. This would support the experience hypothesis, that older, more experienced spiders can more accurately evaluate the costs and benefits of fleeing from a predator and losing their perch on the plant, which they require to catch their food (pollinators like honey bees). Our presentation the next morning went relatlively smoothly, and I was impressed that we could actually gather enough data in a day and a half to get interesting results. Prior to the FLPs I had been pretty skeptical of the OTS independent projects, in which we only have four days to gather data; the shortest IP I've ever done at Reed has been about six weeks, and all of them have felt like I could have used more time if it were available. However, after seeing how much we could pull together in just two days, I was feeling more optimistic about our IPs, which were our next big task at Palo Verde. 

An unrelated picture of Palo Verde as seen from the dining hall, just because I want to post it. Haha.