Which Pipit to Report?

I was doing some research using the ATSB’s wildlife strike data file and came across a number of interesting little data artifacts. My goal was to maximise the number of reports with standard species names (common names, not the Latin) and this required a combination of Python programming and manual mapping.

Through this process, I noticed that one species in particular kept getting thrown up as “non-compliant” - meaning that it was not found in the standard list of bird names. It involved almost 150 strike reports and on the face of it, looked like a legit bird name.

That bird was the Richard’s Pipit.

I searched the Internet and found it was indeed a real bird but it took a little bit more digging to found out why it was not being processed as a standard bird name…

So, what?

I guess the most immediate impact of mistakenly labelling one species as two is that it goes under-reported. The Richard’s Pipit, again, not a bird that actually exists in Australia, is currently the 16th most reported struck bird.

With some relatively minor data management, the Australasian Pipit becomes the 6th most reported struck species.

What’s the Lesson?

There are two ways to approach this issue. The first is consistency. If your wildlife strike database includes “old” names for species, then a certain amount of consistency needs to be applied. That could mean amended the database to reflect the new name (not too hard when modern electronic records are maintained) or it could mean retaining the old name.

The second approach is acknowledgement. People using wildlife strike data need to acknowledge these changes are adjust their analysis appropriately.

Obviously, a combination of these approaches would be best.

Image credits: Richard’s Pipit (JJ Harrison via Wikipedia) & Australiasian Pipit (Sommerdrought via Wikipedia)