Factors affecting the control of Dactylopius, an invasive pest of cactus crop: artificial intelligence-based meta-analysis

Introduction The cochineal scale insect scientifically known as Dactylopius opuntia was first introduced as biological control methods in regions where cactus was considered an invasive weed. Today, D. opuntiae insects is seen as one of the most important limitations to edible cactus production. In the present study, a meta-analysis was performed to assess the effectiveness of various control methods against D. opuntiae and the associated key factors. Methods The Artificial Intelligence (AI)-based tools were employed to screen scientific production reporting the effectiveness of the fight against D. opuntiae , while robust variance estimation (RVE) and proportional meta-regressions were applied to deal with the abundance of effect sizes within a single study and perform subgroup meta-analysis respectively. Results The effect sizes were pooled to 4.19 ± 0.08, I 2 = 99.99%, for the nymphs and to 3.99 ± 3 0.09 I 2 = 99.99%, for the adult insects. The proportional meta-analysis pooled the proportion of nymph’s mortality by 63%, I 2 = 99.1%, and adult mortality by 74%, I 2 = 98.6%. In subgroup analysis, a significant difference (p < 0.001) was observed among study locations (countries) and trial durations (times), suggesting that these factors can influence the effectiveness of the fight against Dactylopius pest. Importantly, non-chemical control means yielded higher proportion of adult cochineal mortality compared to chemical method. Discussion Finally, the scientific information obtained in the present study provides valuable insight for decision-makers towards effective, and sustainable control of white cochineal pest. Therefore, consideration of environmental influence (location) and the duration of treatments in this control is recommended.