Counting dolphins

Written by Claire Lacey

Exciting times here at “spinner dolphin project HQ”, as we make final preparations to get our dolphin surveys underway. These trips are a collaboration with NOAA / PIFSC and will be running until the end of the year or so, as we collect the data we need to get an estimate of how many spinner dolphins use the coastal waters around Oʻahu.

There are a few different ways to count animals, and very often it is not as easy as it initially appears. The most straightforward way is just to do a direct count. If you have sheep (for example), you can round them all up into a pen, count them directly and you have your answer. Easy! But, of course, this doesn’t work with populations of wild animals.

Another way of counting individual animals is to carry out a mark-recapture study - which is what Liah is doing with the photo-id work with the spinner dolphins of the Wai'anae coast. This method uses naturally occurring marks on the dolphins' fins – such as nicks or scars - to recognise individual animals as you can see in the photograph below:

This image shows a very well-marked spinner dolphin dorsal fin. Photo was taken under research permit 21476.

Each time you see a group of dolphins, the aim is to try to take a photo of every animal in the group. When you look at the pictures afterwards, you can work out what proportion of the animals have a recognisable mark; for example, you know there were 20 dolphins in the group you saw, but maybe only seven of them have a fin you can recognise again.

One sample is, of course, not enough; so you go out throughout the year and you look for animals, each time trying to photograph the entire population, and each time looking at the proportion of recognisable animals out of the total. After a good number of visits, you will be able to use this information to estimate the size of your dolphin population. This is a great method if you can repeatedly see the same animals and you have a good handle on where the population can be found – but it does rely on you being able to sample the entire population each time.

Outside of the leeward coast, dolphins are not seen regularly enough to get enough re-samples to use the photo ID method, so to get an estimate of how many animals there may be around the full Oʻahu coastline, we will be using distance sampling. This method involves designing a set of transect lines to cover the entire area of interest (see below).

Example of the zig-zag transect lines we will be using to survey the coastal waters of O`ahu for spinner dolphins

We will sail each of these lines in turn, looking for animals. Every time spinner dolphins are seen, we’ll note down information about how far away the animals are from the boat when they are first sighted (the “distance” part of “distance sampling”). We also record information on group size, behaviour, and so on. The distances to dolphins are used to build a detection function (see below ). This is a simple model which allows us to estimate how many groups of dolphins we may have missed. The method assumes you see all the dolphins which are right on the transect line you are sailing (or at zero distance from the boat), and that you see fewer dolphin groups further away from the boat.

This shows what a simple detection function may look like: more dolphin groups are seen close to the trackline.

We’ll be doing more than one trip around the island (maybe as many as 17, depending on how many groups of animals we come across), to collect enough data to build our detection function. Of course, lots of other factors, such as weather conditions can influence how far away you can see dolphins – but more about this in a future blog post!

We’ll be updating the spinner dolphin Facebook page with information about the surveys as we go – so if you’re interested, give the page a follow and feel free to say hello!

For more information about MMRPs spinner dolphin work, see our main project page.