The major challenge to sustainable use of Bt cotton is the risk that the target pests, Helicoverpa spp, may evolve resistance to the engineered toxins. Resistance to conventional Bt sprays has evolved in field populations of other moths (e.g. Plutella xylostella), H. armigera has consistently developed resistance to synthetic pesticides in the field, and cultures of Bt resistant strains of H. armigera have been generated in the lab. Bt resistance concerns are thus well-founded. Much effort has therefore been devoted to developing and implementing pre-emptive resistance management strategies, most notably based on the use of refuges to maintain sources of susceptible moths in the population which will mate with potentially resistant individuals produced in Bt crops - thus dampening the development of resistance. This project aimed to help identify the most productive refuge options. Such information is essential to allow robust estimates of refuge sizes. Crops considered in CSE90C included sprayed (non Bt) conventional cotton, unsprayed conventional cotton, pigeon pea, sorghum, maize and soy bean.

One of the major criteria defining effective refuges is that they will generate enough susceptible moths to ensure that matings between resistant survivors from Bt crops are extremely unlikely. But our knowledge underpinning the optimal placement of refuges within a landscape and how well the moths generated there disperse to Bt crops is very limited. Some studies have simulated movements of H. armigera from refuges to transgenic crops using the model HEAPS and argued that dispersal from refuges can be patchy according to wind speed and direction and spatial distribution of crops. The qualities of plant hosts at source and sink, aggregative / synchronous movement behaviours of the moths and limits of simple diffusion are also considered likely to be important. But empirical data from the field on all this are scarce. In CSE90C, we sought to use strontium to mark moths in refuges and set traps to recapture them in nearby Ingard cotton crops.

Many questions about the seasonal abundance and resistance dynamics of Helicoverpa require knowledge of which crops contribute to local populations. While we can infer something about the relative importance of different crops from pupal numbers, the definitive answers require that we can assign moths to probable crop origin. We intended to do this, and thus infer local movements of moths, using the ratios of carbon isotopes that vary between C3 (cotton, legumes) and C4 (maize, sorghum) plants and thus are likely to be transferred to moths reared on them.

Despite some previous research on cultivation and its usefulness in reducing the abundance of over-wintering pupae of Helicoverpa (“pupae busting”), questions often arise as to the best tillage methods to use to achieve this. We intended in CSE90C to establish field trials, in collaboration with soil scientists, to determine optimum methods amongst commonly available machinery to destroy pupae, whilst taking into consideration the impact of such methods on soil structure and fertility.