Go beyond species lists in your surveys and BioBlitzes. Tweak Four: how often did you misidentify?

This article is one in a series about four simple tweaks you can make to your surveys and BioBlitz events that will make them much more useful for long-term monitoring.

Introduction

Tweak One: document your effort

Tweak Two: subdivide your effort

Tweak Three: how much did you overlook?

Tweak Four: how often did you misidentify?

Tweak Four: how often did you misidentify?

Identification errors are another problem with using simple species lists for long-term monitoring. Everyone makes mistakes, including experts. Throw those experts into a hectic, crowded BioBlitz environment and the chances of mistakes become higher. This is OK, just like it was OK if your survey didn’t find all the species at a site. In both cases, what’s most important is that you survey in a way that lets people estimate *how many** misidentifications are made.

There are two ways to assess the probability of identification errors. We already did one method in Tweak Three where we had people independently and simultaneously count birds at the same time. However, this can mix up detection and identifications errors.

The easier thing to do is collect evidence, like photos, recordings, or specimens (if you have a collecting permit). That can be as simple as “This is plant species X and here’s a photo of it” or “This is bird species Y and here’s a recording of it singing”, where X and Y are the species names of your identifications. You certainly don’t need evidence every time you find something, so long as you get enough evidence that you, and others, can assess how often you made mistakes.

The key step is honestly recording what you thought a species was when you photographed it, not later after you thought more about it, or asked others, or looked it up in a book. You shouldn’t be afraid or embarrassed about being wrong. Also, any time you’re not 100% certain of your identification, you should always put a question mark after your identification.

You can wreck your data by guessing identifications without using a question mark to indicate that it’s a guess. That ruins any chance of estimating how often you’ve misidentified species. Similarly, if you use your photos to look up a species ID later and then “fix” your original notes, that also damages your data set. Always retain your original field identifications alongside your corrected identifications.

A simple way to check if your identifications are correct is to load your survey’s photos and audio recordings onto iNaturalist and see whether others agree with all your identifications. You should always upload your observations with your original field identifications. On iNaturalist you can then correct that identification yourself, if necessary, and it will display the old identification crossed out and replaced by your preferred identification. Other experts can then offer their identifications.