Probability of Insect Capture in a Trap Network: Low Prevalence and Detection Trapping with TrapGrid

TL;DR

This paper extends the TrapGrid model to include strict detection probability calculations, aiding interpretation of zero captures in low-prevalence insect trapping scenarios, while maintaining the original model for network sensitivity analysis.

Contribution

The paper introduces an alternative probability calculation mode in the TrapGrid model for better interpretation of zero captures in low-prevalence situations.

Findings

New strict detection probability calculation implemented

Original model remains useful for sensitivity comparisons

Enhanced interpretation of trap network results in low-prevalence contexts

Abstract

Attractant-based trap networks targeting insects are ubiquitous worldwide. These networks have diverse targets, goals, and efficiencies, but all are constrained by practical considerations like cost and available lures. An important way to balance goals and constrains is through quantitative mathematical modeling. Here we describe an extension of a computer model of trapping networks known as "TrapGrid" to include an alternative mode of calculating the probability of capture over time in a trapping network: Strict detection ("capture of one or more") compared with the average probability of capture as implemented in the original version. We suggest that this new calculation may be useful in situations of low prevalence where trap network operators wish to interpret the meaning of zero captures at a small scale. The original remains preferred for comparing the sensitivity and suitability of alternate trap networks (i.e. density of traps, their placement, lure attractiveness, etc.).