Home » Uncategorized » Photographic Mark-Recapture of Gold-Dust Day Geckos in Hawai’i

Photographic Mark-Recapture of Gold-Dust Day Geckos in Hawai’i

As you read the title of this post, you may be wondering to yourself how someone with a background in dendroecology and biogeochemical cycling got herself involved in a project on mark-recapture of lizards. This is a reasonable question, the short answer to which is, she married a herpetologist. The long answer is outlined below 🙂

The Beginning Since I only teach in the mornings this semester, I often work from home in the afternoons. I quickly noticed that our front lanai is frequented by a number of a rather adorable lizards, easily identified as Madagascar gold-dust day geckos, Phelsuma laticauda. Their common name comes from the ‘dusting’ of yellow along their back behind the head. Mostly out of casual boredom, I started photographing them, trying to figure out if I was seeing the same one or two over and over, or if they were all different. These little guys are super adorable and not so shy that you can’t get a few good photos every now and then, which quickly filled up my Facebook page. I lured them with oranges, I made a gecko blind out of the bathroom window, I even caught a couple of them having a romantic moment on the lanai roof…..

IMG_2432 copy

Geckos love oranges

To help assess my lanai’s population size, one of my Anthony’s herp colleagues at UH-Manoa, Dr. Amber Wright sent me a paper on a new technique called photographic mark recapture. In 2012, some scientist out of Dartmouth published a paper on using pattern-recognition software to non-invasively assess population sizes of giraffes. Essentially, you ‘mark’ the animals by identifying their particular pattern in a photo, then ‘recapture’ by rephotographing the population at a later date and identifying which animals you’ve seen before and which are new. The program the authors developed is called Wild-ID and is available online for free. Amber sent me the paper, and I got started identifying the Gerhart-Barley Lanai (GBL) gecko population. Though not as large as the giraffes, the day geckos have unique markers on their backsides in the form of red splotches. Some are different enough as to be identifiable by eye. Others are quite similar, though, so the program is helpful. So far, the GBL population consists of nine individuals (see photos below). UH Manoa biology grad student Áki Lárusen had the fantastic idea of naming geckos ‘Paddy’ in reference to their awesome toe-pads, which allow them to walk up walls and ceilings. All GBL geckos are now called Paddy (or Paddington Gecko, if we’re being formal) until I get to know them well enough for a unique name. You may wonder if I am joking on that, so I will tell you now that I am 100% not joking.

This slideshow requires JavaScript.

When Obsession Becomes Science My lanai gecko-hunting became a frequent topic of conversation, which is how I discovered Amber was putting one of her undergrads, Bailee, on a proof-of-concept project to see if this sort of ID would be feasible for real work on the gold-dust day geckos. As the only person who, to our knowledge, had used it on this species, I started showing Bailee the ropes of how to use the Wild-ID program. This initial how-to quickly became me joining the Wright Lab for their preliminary work on testing photographic mark-recapture against traditional methods. Currently, we are working with a population of geckos living near the Biology building on the UH-Manoa campus. We capture the geckos using a noose (which is harder than it sounds – they are curious, but timid little things), then we record their sex, weight, and length. To test the mark-recapture methods, we take high-resolution, up-close photos for the Wild-ID program, mark each gecko’s neck with a unique color pattern of non-toxic paint (which will last only until the gecko next molts), and mark each gecko with a unique pattern of visible implant elastomer (VIE) tags (which are longer-term than paint, but not always permanent).

This slideshow requires JavaScript.

Why use photographic mark-recapture when there are clearly other methods proven effective?

There are FOUR geckos in this picture.

There are FOUR geckos in this picture.

1) Taking photos is less intrusive and physically invasive. VIE tags require injection of the elastomer under the skin, which doesn’t technically hurt them (they don’t even flinch when the procedure is performed), but would be nice to avoid if there’s a less invasive alternative. Having to catch geckos involves snagging them with a noose (which I imagine is upsetting) and handling them a lot, which can damage their tails (geckos can lose tail parts as a defense mechanism against predation) or injure them in other unintended ways. With the paint tags, we could do the ‘recapturing’ by photo since they are visible on the back (assuming the gecko hasn’t molted), but the VIE tags require physical recapturing since the tags are on the gecko’s underbelly. Photographic mark-recapture avoids all these problems. With a good SLR and a macro lens, you can get fantastic photos (see Bailee’s below) without even startling the geckos.

2) Taking photos doesn’t impact the geckos interactions with each other or other species. The paint marks potentially make the geckos more conspicuous to predators (although they are already rather brightly colored), or more/less desirable to each other as mates.

3) By avoiding all the time, energy, and costs associated with traditional marking, you can potentially get a much larger data set. On our couple of trips to the UH-Manoa population, we see far more geckos than we are able to capture and tag, meaning photographic mark-recapture might give us a much larger pool to work with. Currently, we have VIE tagged and paint marked 12 geckos in two trips. On the second trip, we didn’t recapture any of the geckos from our first trip, and in subsequent photo forays, we have only once rediscovered a tagged gecko. This suggests we haven’t yet really made a dent in the UH campus population with only 12 tags. Bailee is focusing on testing the Wild-ID program to make sure it correctly identifies photos of different geckos as unique individuals, and identifies photos of the same gecko from different observations as the same individual (thereby avoiding false positives and false negatives, respectively).

See how many unique back patterns you can ID in our photos of the campus population!

This slideshow requires JavaScript.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: