DEVELOPING METHODS TO PREDICT RISK OF CYLINDROCLADIUM QUINQUESEPTATUM LEAF BLIGHT ON EUCALYPTS IN MAINLAND SOUTHEAST ASIA AND AROUND THE WORLD KM OLD, TH BOOTH, T JOVANOVIC and MJ DUDZINSKI CSIRO Forestry and Forest Products, PO Box E4008 Kingston, Canberra, Australia ACT 2604 Background and objectives Eucalypts and other Australian tree species are of major economic, social and environmental importance to many countries. The most commonly grown eucalypt species in South East Asia is E. camaidulensis. The future of eucalypts, and specifically E. camaidulensis, in Vietnam has been brought into question by the appearance in the late 1980s of severe leaf blight epidemics in south-east and central Vietnam. Badly affected trees suffer severe reductions in growth rate and crowns and even main stems become deformed with losses in merchantable volume. A similar syndrome was observed on a local scale in northern Queensland (Australia) in 1989. Several pathogens are associated with disease in both countries with Cyiindociadium quinqueseptatum (CLE3) being an important causal agent. With the support of the Australian Centre for International Agricultural Research (ACIAR) a major Project is being undertaken to minimise disease impacts on eucalypts in South East Asia by gaining a thorough understanding of the etiology and epidemiology of leaf blight diseases, developing screening methods and locating sources of resistance which will be tested in complementary trials. One sub-project, which is briefly described here, is developing methods to predict disease hazard risk at sites in mainland South East Asia and around the world. Materiais and methods Climatic interpolation and the use Of simple simulation models have greatly improved techniques for matching tree species to sites [I]. Recent advances in mathematical interpolation allow mean Climatic conditions to be reliably estimated for most locations in the world, whilst computer models allow cornp[ex growth-environment interactions to be simulated in more realistic ways. The disease hazard sub-project is applying similar methods to identify areas of high risk for leaf blight diseases. A literature review was first carried out to determine which environmental factors were associated with CLB infections. Very simple rules related to temperature and rainfall were developed to identify high risk areas. These rules were used in several climatic mapping programs previously developed identify areas suitable for planting different species. These PC-based programs include intteorpolated data estimated for tens of thousands of locations in regular grids. Results and conclusions Although previous studies suggest that CLB infection is Probably related to very complex short term conditions such as hourly leaf wetness, preliminary analyses suggest that a simple description of CLB requirements can provide useful broadscale estimates of susceptible areas in South East Asia, India, Africa, Australia and Latin America. Survey data are now being collated to develop more detailed descriptions of CLB requirements. A simulation mapping program has been developed for South East Asia which includes data on maximum temperature, minimum temperature, precipitation, evaporation and solar radiation for 56 000 locations. This will be used to develop more complex models of CLB infection. Reference 1. Booth TH (ed.) 1996. Proceedings of an international workshop held in Bangkok. ACIAR Proceedings No. 63, 126 p.