The impact of soil water and nitrogen variability on the fitness and performance of Neolema abbreviata Larcordaire (Chrysomelidae) a biological control agent for Tradescantia fluminensis.

Abstract

Tradescantia fluminensis Vell. (Commelinaceae) is a plant of Neotropical origin native to the southern parts of Brazil bordering Argentina. In South Africa, it is classified as a category 1B invader species in the National Environmental Management Biodiversity Act (NEMBA) owing to its incipient phase of invasion. The occurrence of naturalised populations of T. fluminensis has so far been confirmed in all provinces except the Free State, Northern Cape and North West. In cognisance of the devastating effects of invasive alien plants on native biodiversity, ecosystem health and ultimately provision of ecosystem services, several control methods have been employed with varying degrees of success. Classical biological control, which involves the release of exotic natural enemies (pathogens and herbivorous insects), is one such method widely-used because of its relatively low costs and minimal non-target effects. For T. fluminensis, Neolema abbreviata (Larcodaire) Coleoptera: Chrysomelidae) is one agent that is earmarked for release in South Africa following a successful introduction in New Zealand. However, little is known how novel environments presented by soil water and nutrient gradients may indirectly influence its herbivore performance and life-history through alterations in host-plant quality. In this era of global climate change where anthropogenic activities have led to changes in rainfall patterns and biogeochemical cycles of major elements such as nitrogen, investigation of species responses to such is important. Results from my study show that both water and nitrogen (N) variability influenced plant biomass accumulation, foliar N content and subsequent herbivore performance, and life-history traits of both adult and larval N. abbreviata. The longest vines were on plants that had optimal irrigation under excess fertiliser whilst severely water stressed plants that had excess fertiliser had the shortest vines. Foliar N content was highest in plants that had excess fertiliser under both pulsed and optimal irrigation whilst lowest foliar N content was in plants under optimal irrigation without any fertiliser. Optimally irrigated plants that received moderate fertiliser had their highest rate of egg deposition in both no-choice and multi-choice conditions suggesting quality-based host ranking behaviour in N. abbreviata. The consequent larval performance traits which included weight gain and time to pupation were superior in this treatment thereby providing support for the preference-performance hypothesis (PPH). Feeding patterns between larvae and adults among plant treatments were largely similar suggesting uniform nutritional requirements across the life-stages. There were limited parental effects of plant quality on the life-history traits in both larvae and adults across F1 and F2 generations. In reciprocal diet transplant experiments, there were no significant responses to parental diet effects on larval weight, mortality, feeding damage, pupal weight and days to pupation. However, there were significant parental diet x test diet interactions with offspring from parents fed on high N plants generally performing better on low N test plants in traits such as larval weight gain and final pupal weights. Oviposition selection, feeding weight and longevity did not respond to the effects of parental diet nor its interaction with test diet, unlike the case with larval traits. There were significant correlations between pupal weight and number of days to pupation, pupal weight and eclosion success. I conducted a 3 x 3 full factorial experiment to determine the impact of water and fertiliser variability on the performance of Neolema abbreviata (and its host plant Tradescantia fluminensis. My results show differential responses to parental diet between larvae and adults of the same generation among an insect species with both actively feeding larval and adult life-stages. However, there was no correlation between adult weight and longevity. Overall, my thesis contributes to the growing body of literature on the impacts of anthropogenic global change on plant-insect interactions. It will also assist land managers when applying biological control of T. fluminensis. Furthermore, my results show the implications on the successful biological control (mass-rearing and field release) of T. fluminensis resulting from variable nitrogen and water conditions.