Greetings from Christmas Island,
I figured it was probably as good of time as any to put something up on this thing as I have now been out on Christmas Island for one month. The first 2 weeks were spent simply getting to know the island and doing a lot of the old ‘poke and think’. Following my supervisors departure I have been busy over the last 2 weeks establishing an experimental design and setting up study sites.
The primary aim of this trip is to get a good start into answering the question: does the invader-invader mutualism between the yellow crazy ant and scale insects facilitate invasion of the rainforest by land snails? The hypothesis being that through the creation of enemy-free space and increase of habitat and resources (both direct and indirect effects of removing the native omnivorous red land crab) the primary invaders will have a positive impact on the snail fauna, possibly leading to secondary invasions.
Through its complex invasion history, the rainforest of Christmas Island contains many useful forest states for answering this question. Firstly, there is still currently completely intact rainforest, which has a high density of crabs and represents the state in which the majority of the island was in prior to invasion. Secondly, despite best efforts, areas of yellow crazy ant supercolonies are still present, although not as widespread or common as in the height of their invasion. Thirdly, there exists sites which are referred to as ‘ghosted’ whereby crabs have been removed from the site via migration through a supercolony but the site has never contained yellow crazy ants. And finally, there are sites which have had a supercolony which was successfully removed via baiting and has since been recolonised by red land crabs. Although there also exists almost any combination supercolony-ghosting-baiting-recovered/not recovered imaginable, it is these four rainforest states that provide the best natural experimental treatments for answering the question.
As mentioned above, the presence yellow crazy ant supercolonies result in a predictable change to the rainforest via their removal of the red land crab. The table below shows how these four experimental treatments (INTACT, SUPERCOLONY, GHOSTED, BAITED & RECOVERED) differ in their presence or absence of the ‘main players’ that could be responsible for changes seen in the land snail community.
In the first 2 weeks of poking around the rainforest we were able to locate 4 yellow crazy ants supercolonies that could be useful for this study. Based on this limiting treatment, I have been working towards replication of n = 4 for all treatments. Although not the greatest level of replication, there is always opportunity to grow that out to n = 5 if another supercolony is located in which case I will probably be at my upper limit in terms of the number of sites I can handle for the next couple of months. With the current 16 sites, you can be highly confident that they are surveyed in a detailed and rigorous manner. Sites are being set up in clusters around the located supercolonies in which a the closest usable site for the other treatments are selected, thereby ensuring a relatively even spread of site types.
So…in order to get these sites set up, I have been busy marking out site boundaries of a 1/4 ha (50×50 m) as well as getting some initial data to not only confirm that the site is of the treatment its meant to be but also just some other general characteristics (seeing as I was running the tape measure out anyway). All sites were surveyed for i) RED CRAB DENSITY – the number of occupied crab burrows in 100m2 (50 x 2 m quadrat), ii) YELLOW CRAZY ANT ACTIVITY – card counts in which white cards are placed on the ground and the number of ants crossing that card in 30 sec is recorded and repeated 11 times across a 50 m transect, iii) SEEDLING DENSITY – the number of seedlings (defined as any individual > 15 cm tall and = to or < 1 cm DDH) within 100m2 (50 x 2 m quadrat), iv) TREE DENSITY – the number of ‘trees’ (defined as any individual > 1 cm DBH and with at least partly rooted within the quadrat) within 200m2 (50 x 4 m quadrat) and v) PANDANUS DENSITY – a species not lumped in with the seedlings or trees as it has a very patchy distribution, can form very dense, almost monoculture thickets and initial observations suggested it may be a key habitat species for arboreal land snails – number of individual in 200m2 (50 x 4 m quadrat). Below are the current results in which Ghosted and Baited & Recovered are only n = 3 as I have had difficulties locating sites close to the supercolony and to make matters worse, for the last four days (and for at least a few more into the future) I haven’t had a car so I’ve only been out to the bush a couple of times after borrowing other peoples. (Apologies for the quality of the image and for it being cut off – pretty sure you just click it and you can view it in more detail)
Well this has actually taken a significant amount of time to write, so although there is probably a whole heap more that I could go into I think I’ll leave it there. Enjoy these pics of what each of these forest states actually looks like……
BAITED & RECOVERED