Comparisons of the thermal physiology of water hyacinth biological control agents: predicting establishment and distribution pre‐ and post‐release

Abstract

AbstractInvestigations into the thermal physiology of weed biological control agents may elucidate reasons for establishment failure following release. Such studies have shown that the success of water hyacinth biological control in South Africa remains variable in the high‐lying interior Highveld region, because the control agents are restricted to establishment and development due to extreme winter conditions. To determine the importance of thermal physiology studies, both pre‐ and post‐release, this study compared the known thermal requirements of Eccritotarsus catarinensis (Carvalho) (Hemiptera: Miridae) released in 1996, with those of an agent released in 1990, Niphograpta albiguttalis (Warren) (Lepidoptera: Pyralidae) and a candidate agent, Megamelus scutellaris Berg (Hemiptera: Delphacidae), which is currently under consideration for release. The lower developmental threshold (to) and rate of development (K) were determined for N. albiguttalis and M. scutellaris, using a reduced axis regression, and incorporated into a degree‐day model which compared the number of generations that E. catarinensis, N. albiguttalis, and M. scutellaris are capable of producing annually at any given site in South Africa. The degree‐day models predicted that N. albiguttalis (K = 439.43, to = 9.866) can complete 4–11 generations per year, whereas M. scutellaris (K =502.96, to = 11.458) can only complete 0–10 generations per year, compared with E. catarinensis (K = 342, to = 10.3) which is predicted to complete 3–14 generations per year. This suggests that the candidate agent, M. scutellaris, will not fare better in establishment than the other two agents that have been released in the Highveld, and that it may not be worth releasing an agent with higher thermal requirements than the agents that already occur in these high‐lying areas. Thermal physiology studies conducted prior to release are important tools in biological control programmes, particularly those in resource‐limited countries, to prevent wasting efforts in getting an agent established.