Having spent a few days really looking at the data we have gathered, we have discovered some interesting results, that have given us lots to think about .
Firstly, before presenting some results, I want to stress that when estimating the location of an animal using radio tracking (and triangulation) there is a certain amount of error involved that needs to be taken into consideration, however, this should not detract from the biological significance of the data. For reference, the mean difference between the GPS location and an estimated value is 78m (based on the a pilot study and the fact that each time we located a den site and obtain an accurate GPS location, we also generate an estimate through triangulation).
For fear of missing things of biological significance we re-looked at P33’s home range size. Previously, this had only been calculated using GPS locations of den sites. However, on some occasions we were unable to locate the den site and took only a triangulation estimate, these were originally excluded from the home range size. The graphs below give a more show the home range sizes of P33 (female, top graph) and P27 (male, bottom graph) and how strictly excluding data could bias results.
The gaps in the graph for P33 are days where no locations were gathered. The days where we couldn’t accurately locate/access the den site correlate with the days there was a substantial increase in home range size.
‘A’ stands for estimates taken when P27 was active. There are four activity estimates that results in a sharp increase in home range size and three of these are on nights he went to a new den site. However, the den site was no more than 150m from the last one.
As was mentioned in a previous post, more activity has been recorded for P27 compared to P33, despite a similar sampling effort. Activity estimates seem to have a comparatively large impact on home range size for P27. This could be a combination of him being active even when he returns to the same den site, and him moving around more when changing den sites. This was not seen for P33, despite the mean distance between an old and new den site being 89m for P27 and 76m for P33.
Interestingly, though, after our reevaluation of the data, the home range estimate for P33 increased considerably to 20ha, to 17 ha for P27. Originally, based on GPS locations of den sites alone, they stood at 2.5ha and 4.7ha respectively. This warns that den site location alone may not be enough to accurately estimate home range size in this species, but could be enough to indicate when the home range is stable.
Another observation that we have made with P27 is that the selection of den sites are a lot different to that of P33 and P34 (for the short time she was tracked). Whereas both females used very small tree hollows, often climbed higher up inside the trunk, or even climbing the outside of the trunk, P27 prefers large, open, often rotting tree trunks, where there is potential for him to dig a burrow. There are some examples in the photos below.
P33 tended to select small tree hollows at the base of the tree
In the above photo P27 may have dug a burrow into the soil inside the rotting trunk as the signal was at ground level and with torches peering inside, we could not see him.
On a few occasions P27 has been very easy to spot
At this den site P27 climbed up the mud and dug himself into a burrow at the top.
These differences in behavior maybe due to the provenance of the animal and differences in the habitat it lived in. For example released Loris have been know to travel large distances and found in a habitat the matches that the were originally taken from. However, as well as raising interesting ecological questions about the species, it is also reassuring to know that so far all the release candidates seem to be surviving and doing well.