Monday, October 19, 2015

"It always rains when..."

Out in the bush, there is a distinct lack of meteorology happening. We can tell you the high and low temperature as well as the rainfall in camp for nearly every single day of the past 27 years we’ve been in the Mara…but when you ask about looking into the future, there are few resources for getting a firm handle on that.

So what’s someone to do when you’re trying to anticipate which clothes to wear, or whether to bring your rain boots and coat on observations, just in case you get stuck this time and need to stand for hours digging the car out of mud?

Well, as in many situations we face in the field, the answer has become, “Figure out how to do it yourself!”

Serena camp has recently begun using the following metric: “It always rains when Emily leaves her towel out to dry overnight.”

Philomen (who works at Serena camp) has his own: “It always rains on the full moon.”

In the year I was in the field, I developed my own method. There are three ingredients – as they begin to add up, you know it’s coming; when you have all three, you start bringing your rain boots into the car every time you drive out of camp.

1.     Hotter than normal days. We’re talking 30+ degrees Celsius (86 degrees Fahrenheit for all those not inclined towards metric thinking)

2.     Large, fluffy, white clouds gathering in the afternoon on the north-east horizon. 
3.     Siafu. All over. These army ants carve ant-sized trenches in the ground as they march their long lines through camp. (They even build tunnels with their bodies!) We could swear they congregate around water sources, and seem to appear right before rain. 

So who needs meteorologists when you can plainly see when Emily leaves her towel out to dry, or can even better – when you know exactly when the next full moon is? Or what about those clouds and temperature and ants…that has to be accurate, right? What happens then when a full moon doesn’t bring rain, or a hot day with fluffy clouds and a line of ants doesn’t equate to a downpour in the next 24-hours? Well…. “that was an exception”, right?

Because do I actually have the data to show you that it rained significantly more often when my three ingredients came together than when they didn’t? Can I tell you the parameters of what constituted as ‘hot’ or the time at which the clouds began to gather, or the number of siafu considered ‘all over’? (Trust me, it doesn’t take many to feel that way.) Have we presented any alternative hypotheses? (Let’s all agree that Emily’s towel doesn’t warrant an alternative hypothesis).

I must say, I would have to answer ‘no’ to all of these; as sure as I am that it ‘always’ rains when those three factors occur simultaneously.

What’s going on then? Why are we so sure of our predictive powers?

It’s called ‘confirmation bias’, “a tendency to search for or interpret information in a way that confirms one's preconceptions”, according to ScienceDaily. It’s all too easy to fall prey to this attractive mental teaser. It’s amazing how many pieces seem to fall into place or how often you experience a certain thing because you were made aware of it. But this is one of the greatest assets of science – we uphold strict standards of hypothesizing, designing solid experiments, running them over and over again to achieve a high enough sample size that is representative of the whole.

In the field, it’s particularly important that we all remain conscious of this – when you spend 365 nearly consecutive days in the Mara, when the changing of the landscape and wildlife is so readily apparent each day, you start to think you understand all the patterns.

For our group, it’s a valuable reason we tell stories of past years, to remind ourselves how different each of the years can be, that our 1/27th of the field life of the project is still relatively small. To avoid confirmation bias, we have to keep our personal interpretation and expectation out of the equation, and diligently gather data every day following the procedures that have been in place across time.

Hyena research involves great sunrises and sunsets, pummeling rainstorms, and adorable cubs; however, those adorable cubs are able to enter our long-term research picture because of these protocols that help us avoid bias in our data.

I confess: even knowing this, I still pay attention to temperature, clouds, and ants. It has led to having my boots in the car on far more days than are needed…but I’ve rarely been caught without them at least! 

Posted by Hadley Couraud


JaimeT said...

When I was out there someone told me that it always rains when you hear the ground hornbills in the morning. It was true (as long as I ignored the mornings when I heard the ground hornbills and it didn't rain)!

Laurena Jenkins Hoffmeyer said...

I arrived today and am staying at the Mara Serena for 3 nights. We passed one of your research vehicles today after the rain and were hoping to get your attention to ask you how you liked the outcome of the MSU/U of M football game. Even in Kenya, as a graduate of MSU, people were letting me know the outcome. Hope to see you while out and about. Laurena

Michigan State University | College of Natural Science