Dr. Grant Humphries, from the Zoology department at the University of Otago, New Zealand, has spent the last three years studying how a bird species called Sooty Shearwaters can help predict upcoming El Niño occurrences. After much time and research, he has figured out a way to do so using data mining.
In Dr. Humphries’ words, Sooty shearwaters (a medium sized seabird) are traditionally hunted around southern New Zealand by indigenous Māori groups. “Personal diaries from eight of these islands were gifted to us in order to look at long term trends in the population. Previous work found relationships between hunting rates and upcoming El Niño events. I was interested in examining how we could use those diaries to create population indices to predict upcoming climate events.”
The diary data consisted of hunters' ages, total time spent hunting, weather conditions, size of birds captured, estimated experience, time of the year, and time of the day. Typically, ecologists would examine these data using generalized linear modeling or mixed modeling techniques with a focus on mechanistic interactions. Humphries chose to use TreeNet, which was able to focus on the predictions to create the hunt index without constraints of having to dig for mechanisms or deal with the statistical distributions of individual predictors. TreeNet offered the ability to explore relationships between the target variables and predictors with relative ease in order to examine any interesting patterns.
Humphries was able to index seabird hunting records that could be used to predict El Niño events by nearly 14 months. In years when the numbers of birds hunted was poor, it was more likely that an El Niño event would begin within the next year or so. This represents one of the first examples of seabirds being used as predictors of climate events as opposed to them previously being used simply as monitors.