Using Digital Image Analysis to Quantify Small Arthropod Vectors

Abstract

Quantifying arthropod vectors can be a time-consuming process. Here, we describe a technique to count large samples of small arthropods using ImageJ. ImageJ is an open source image processing software, produced by the National Institutes of Health, with a straightforward interface that has proven useful in quantifying small organisms (i.e., cells, pollen, eggs). In 2017, we deployed CDC light traps baited with carbon dioxide among seven sites to capture black flies (Diptera: Simuliidae). Samples of the captured specimens were photographed, and then quantified manually and automatically, using ImageJ. We compared the accuracy of three types of automated counts to manual counts of black flies using an information-theoretic approach. We found that changing the particle size produced counts closest to those obtained by manual counts. Even over a large range of values, from tens to thousands of flies, our automated counts were often identical to and almost always within 5% of the manual counts. When different, automated counts were usually slightly less than manual counts, and thus conservative estimates. This automated technique is simple, repeatable, requires minimal training, and can reduce the time needed to quantify small arthropods such as black flies.

Keywords: ImageJ; Simuliidae; arthropod counts; automatic counts; black flies.