Development of a biological control-based integrated management of Plutella xylostella (Linnaeus) (Lepidoptera: Plutellidae) in South Africa

Abstract

This study investigated the potential of developing a biological control-based integrated management of diamondback moth, Plutella xylostella (Linnaeus) (Lepidoptera: Plutellidae), a key insect pest of Brassica crops in South Africa. This study used data sets collected at weekly intervals on unsprayed cabbage fields during February 2002–January 2008 to critically examine: 1) the roles of temperature and parasitoid diversity on population regulation of P. xylostella; 2) the effect of hyperparasitoids on biological control of P. xylostella; 3) the influence of parasitism levels on the ability of synthetic sex pheromone traps to forecast infestations; and 4) the potential of using ratios of parasitoid cocoons to pest infestations to estimate field parasitism levels. Adult and immature P. xylostella were recorded throughout the year. But, moth catches and larval and pupal infestations were high only during September–October peaking at (mean ± SD) 36.2 ± 18.65 and 10.6 ± 9.59, respectively. Five species of indigenous primary parasitic Hymenoptera were recorded, and parasitism levels were higher (≥50 %) during late October–May than during June–early October (<50 %). Parasitism levels were positively influenced by average field temperatures where 50 % parasitism rates corresponded with field temperatures of 20 °C. A positive influence of parasitoid diversity on parasitism was only observed during September–early October; thus, the aggregation of primary parasitoids prevented further increases of P. xylostella population density during this period. However, only Cotesia vestalis (Haliday) (Braconidae), Oomyzus sokolowskii (Kurdjumov) (Eulophidae) and Diadromus collaris (Gravenhorst) (Ichneumonidae) showed significant numerical responses. In contrast, regulation of infestations at low levels, mainly <1 P. xylostella larvae and pupae per plant per week, during November–May was due to C. vestalis and O. sokolowskii. Larvae and pupae of C. vestalis, which accounted for 78.26 % of total primary parasitism, were attacked by obligate hyperparasitoids Mesochorus sp. (Ichneumonidae), Eurytoma sp. (Eurytomidae) and Pteromalus sp. (Pteromalidae). Their impact on C. vestalis population density was significant at times, but that did not reduce total primary parasitism levels nor resulted in higher P. xylostella infestations. This is because O. sokolowskii and D. collaris increased their contributions to primary parasitism as C. vestalis population declined. The ability of male moth catches to forecast infestations was better during periods of low than during high parasitism. Thus, the ability of pheromone trap catches to forecast infestations depends on survival of P. xylostella immature stages. Since moth catches were significantly low during high parasitism than during low parasitism period, a pheromone-based action threshold of 8.6 moths per trap for two consecutive weeks is suggested. It was demonstrated that ratios of parasitoid cocoons to infestations can be used as a simple and practical method to estimate background parasitism levels in the field. A 20 % ratio of parasitoid cocoons to infestations corresponded with 50 % parasitism level, above which they are considered high enough to regulate pest population. This thesis provides evidence that biological control-based integrated management of P. xylostella is feasible using indigenous parasitoids in South Africa, and that the methods developed herein to estimate the impact of parasitoids on the pest population can be used in decision making by growers on insecticide applications.