Article: Johnson, C. M., Beckley, L. E., Kobryn, H., Johnson, G. E., Kerr, I., & Payne, R. (2016). Crowdsourcing modern and historical data identifies sperm whale (Physeter macrocephalus) habitat offshore of south-western Australia. Frontiers in Marine Science, 3, 167.
Sperm whales (Fig. 1) are the largest toothed predator on Earth, yet very little is known about their behaviors and their preferred ocean habitat. This is partially due to the fact that they are also the second deepest diving animal on Earth, diving as far as 2,250 meters (7,382 feet). They were also heavily hunted until the late 1990s, so their populations are greatly reduced, making them difficult to study.
One large whaling operation for these whales was off the southwest coast of Australia. The undersea canyons and coast produce conditions that attract squid, the whales’ primary food source. Parts of the territory are within Australia’s Exclusive Economic Zone, the region where a country has special rights to the marine resources of the area. Researchers and policy makers need to know how sperm whales utilize this area in order to properly manage and control how these threatened animals are treated.
The researchers in today’s article pulled information from a wide variety of sources to solve the puzzle of sperm whale behaviors off the southwest coast of Australia. They began with the voyage of the Odyssey, a scientific expedition that tracked sperm whales around the world from 2000 to 2005. Since sperm whales use loud pulses of sound, known as echolocation, to hunt, navigate, and communicate, the Odyssey was equipped with an array of underwater microphones, or hydrophones, that can detect noises from the animals up to 25 km (15.5 miles) away. The system was automated to detect and record signals that matched sperm whale noise patterns, and ran 24 hours a day. This was supplemented by daytime logs of researchers listening to the live microphone feed as well as scanning the horizon with both the naked eye and binoculars for sightings (during favorable weather). Environmental variables like temperature were recorded for each sighting or audible location of a whale. The vessel covered over 9000 km ( ~6000 miles) of the ocean near SW Australia, along the track seen in figure 2, but many areas thought to harbor sperm whales were not covered, as the ship needed to continue its worldwide voyage.
To fill in some of the gaps in the Odyssey data, the scientists turned to three online data portals to ‘crowdsource’ information. These sources included a record of reported sightings by the public, historical commercial whaling data (Fig. 3), and historical Yankee whaling data (a database of whaling activities performed by other countries near Australia).
The four sources of data were fed into a model that predicts the spatial distribution of sperm whales based on many environmental factors, including water temperature, depth, time of year, and the observation sources. This type of model is known as a species distribution model, and can be adjusted to work for many different species of animals. The model works by using the environmental factors at each of the observation sites (such as water temperature or depth) to look at a broad geographical region and make estimates as to where the whales are likely to be found. A simplified example of this would be noticing that your friend like to frequent coffee shops on rainy mornings. While on a trip together, if she isn’t in your hotel room in the morning and it’s raining, your first thought might be to check a local coffee shop.
The model performs a similar analysis, but for many different ‘preferences’ and can use statistics to determine the best place to look. Maybe it’s important the coffee shop have bagels, but less important that they have iced coffee, or vice versa. By tweaking the weight or significance of each factor slightly in many different runs of the model, the researchers can determine the most probable areas a sperm whale would be. In the end, the Yankee whaling data was not used in the final model, as the locations recorded were somewhat inaccurate. The model produces a map of the environments most suitable for sperm whales (Fig. 4). The area determined suitable is at least 19,442 km2, with much of it inside Australia’s Exclusive Economic Zone. There are also already some marine protected areas covering some of the suitable habitat.
By combining data from several data sources, these researches expanded the usefulness of each data source. The map the species distribution model produces can be given to policy makers so they can create laws to better protect sperm whales. Other researchers can use the map to focus their studies on areas where whales are likely to be found, reducing time and money spent. The authors also suggest that autonomous vehicles that spend days or weeks at a time surveying could be equipped with similar acoustic equipment to locate and track sperm whales, providing an even better picture of their habitat usage.