# Ensemble Distribution Modeling of the Globally Invasive Asian Cycad Scale, Aulacaspis yasumatsui Takagi, 1977 (Hemiptera: Diaspididae)

## TL;DR

This paper creates a new model to predict the global spread of an invasive scale insect that threatens cycad plants.

## Contribution

The study introduces the first ensemble species distribution model for Aulacaspis yasumatsui using multiple algorithms and new environmental data.

## Key findings

- The ensemble model predicts a likely invasion of A. yasumatsui in tropical and subtropical regions of Africa and Central America.
- Dispersal is also highly possible in northern areas of Europe and China according to the model results.

## Abstract

Aulacaspis yasumatsui is an invasive scale insect of great economic importance to the horticulture industry, affecting plant species in the Cycas genus. Prior attempts to model the ongoing spatial expansion of A. yasumatsui were based on the MaxEnt algorithm and WorldClim bioclimatic covariates to predict shifts in its distribution owing to future climate scenarios. Here, we develop the first ensemble species distribution model including recent occurrence records and environmental predictors not analyzed in previous efforts. We combine the outcomes of six different algorithms in order to improve model accuracy and minimize model error. The findings support the likely invasion of A. yasumatsui in tropical and subtropical areas of Africa and Central America. Dispersal is also highly possible through northern areas of Europe and China.

Species distribution models (SDMs) have become an important tool to inform conservation and pest surveillance programs about the potential biological invasion of insect pests. Nonetheless, to be operational, SDMs need to incorporate multiple environmental covariates and a representative number of occurrence points depicting the species’ ecological niche. The algorithm of choice, model of choice, and comparison can also have a great effect on the final prediction output. We created a dataset based on previously published records, plus 36 new occurrences and 37 environmental predictors, to generate the first global ensemble distribution model for Aulacaspis yasumatsui. We employed a strategy that aggregates SDMs with the best performance (i.e., greater accuracy) from six different algorithms, resulting in an averaged and weighted model, i.e., the ensemble model. We then selected models from algorithms whose true skill statistic (TSS) was above 0.5 in order to map the potential global distribution of A. yasumatsui. Our results suggest that covariate selection and the individual model algorithms used in the ensemble may be more important for achieving an accurate SDM than the number of occurrence points.