Potential Effects of Climate Change on Tritrophic Interactions in Crucifer Farming Systems of Mount Kilimanjaro and Taita Hills

Abstract

Diamondback moth (DBM) (Plutella xylostella L.), a pest of cruciferous vegetables worldwide, is resistant to major groups of insecticides, and attention has shifted to biological control using parasitoids. However, DBM and its parasitoid species have individual climatic requirements which, when not met adequately due to climate change, will potentially disrupt biological control of the pest. The response of this pest and its parasitoids to climate change at a local scale is less documented than in international scale and is largely ignored in a region where climate patterns have been changing in a way that is likely to interfere with the insect life history and the pest incidence. This study was therefore conducted to determine the likely changes in DBM abundance, the parasitic interaction between the pest and its key parasitoid species Diadegma semiclausum and Cotesia vestalis, and the feeding relationship between the pest and host plants, in response to climate change along altitudinal gradients. Potential effects of climate change on future distributions of the three insects were investigated to establish possible future trend of the pest levels necessary for developing adaptive pest control strategies. Field surveys were conducted for two years on two transects laid along the altitudinal gradients of Mount Kilimanjaro and Taita hills. Based on altitude, each transect was sub-divided into low, medium and high zones within which farmer-owned crucifer vegetable farm plots were identified for the study and marked by Global Positioning System (GPS). Thermohygrometers and rain gauges were installed on farms for recording daily values of temperature, relative humidity and rainfall. The crucifer plants were sampled for larvae and pupae of DBM which were all collected and taken to laboratory for parasitism assessment and life table studies. The pest abundance and parasitism were subjected to analysis of variance (ANOVA) to determine differences among seasons. Biodiversity.R, a statistical package for community ecology, was used to calculate the diversity of xvi parasitoid species. Generalized linear models and regression analyses were adopted for assessing the weather effects on crop damage and parasitism. Colonies of DBM, D. semiclausum and C. vestalis were initiated in laboratory for generating life history datasets at 10, 12.5, 15, 20, 25, 30 and 35°C. Temperature-driven phenology models for the pest and its respective parasitoid species were developed based on laboratory experimental data obtained at constant temperatures and validated through life table data collected outdoor under field temperature conditions. All model development was implemented through the Insect Life Cycle Modeling (ILCYM) software version 3. The generated phenology models, the recorded baseline (2013) field temperatures and future temperatures (2055) downscaled from AFRICLIM database; and the digital georeferenced topographical maps of the transects, were loaded in index interpolator, a sub-module of ILCYM for calculating, interpolating and mapping potential population growth of the insects, through establishment, generation and activity indices. The results showed that DBM abundance was least during long rains and highest during hot dry and short rainy seasons in the two transects. Besides D. semiclausum and C. vestalis, other parasitoid species emerged from parasitized DBM. Cotesia vestalis and Oomyzus sokolowskii, a parasitoid, contributed significantly to DBM parasitism (O. sokolowskii: F = 32.69, df = 1, 101, P < 0.0001***; C. vestalis: F = 27.74, df = 1, 101, P < 0.0001***) compared to other parasitoids in the low zone of Mt. Kilimanjaro, at 43% and 40.7%, respectively. Diadegma semiclausum was the most dominant parasitoid in medium and high zones of both transects. Diversity of parasitoid species declined considerably from low to medium zones of Taita hills probably because of great competitiveness and dominance of D. semiclausum released from 2002 to 2004 in Taita hills. Crop damage increased with DBM abundance in the two transects and declined with rainfall. The DBM parasitism by C. vestalis varied significantly with temperature (F =.....................