4.6.4
DEVELOPMENT OF A QUANTITATIVE RISK ASSESSMENT MODEL FOR THE IMPORTATION OF US MILLING WHEAT CONTAINING TELIOSPORES OF TILLETIA CONTROVERSA

GL PETERSON1 and RJ STEFANSKI2

1USDA-Agricultural Research Service, Fort Detrick, MD 21702, USA; 2USDA-World Agriculture Outlook Board, Washington, DC, USA

Backgound and objectives
Interest in preventing the establishment of economically important plant diseases has increased markedly with the globalization of trade and founding of the World Trade Organization. Many countries are examining their current phytosanitary regulations from a perspective of affording adequate protection and eliminating unwarranted trade barriers. China and Brazil are currently evaluating restrictions on importing wheat containing teliospores of Tilletia controversa Kuhn (TCK), the causal agent of dwarf bunt of fall sown wheat. Risk analysis has become an accepted process for addressing such issues and guidelines have been established by FAO, NAPPO and EPPO, and accepted by FAO, IPPC, NAFTA and GATT. Our objective was to develop a quantitative TCK pest risk-assessment model (TCK-PRA) to evaluate the potential risk to China and Brazil, of introduction, establishment and yield loss from the importation of US milling wheat containing teliospores of TCK.

Materials and methods
A TCK-PRA model was developed by a multi-national task force consisting of US, Swedish, Canadian, Mexican and German experts from government agencies, universities and industry. Using historic and estimated data expressed as frequency distributions, the TCK-PRA model examines pathways of teliospore diversion from transport and milling processes, through grain handling, mill losses, millfeed handling and dispersion of manure to fields. Based on the epidemiological requirements of TCK, a geophytopathological (GP) model [1] was developed utilizing 15 years of climatic data obtained from the UN World Meteorological Organization's telecommunications network, the Crop Environment Resource Synthesis-WHEAT crop model and the MODWHT3 model (wheat growth), to quantify potential conducive hectares in the US, Brazil and China. The GP model differentiates between areas with introduction potential at any level for TCK, and those conducive to economic yield loss. Employing results of a 3-year study to determine the number of teliospores needed to incite disease [2], the TCK-PRA model assessed the number of teliospores escaping the milling process and the likelihood of arrival at a conducive environment in sufficient quantity to establish dwarf bunt. Results were determined by 100,000 Monte Carlo simulations per scenario and reported as the probability of introduction.

Result and conclusions
The GP model analyses of the US correlated well with known areas of dwarf bunt occurrence. When applied to Brazil, the probability of TCK establishment after 100,000 iterations was zero. The absence of conducive environment was the limiting model element. When the Chinese data were evaluated, limited areas in Sichuan, Guihou, Hunan and Gansu were identified by the GP model as regions with TCK introduction potential. Only a limited area in Gansu was identified in the GP model yield-loss scenario. Though regions for potential establishment were identified, the TCK-PRA model detected zero likelihood of TCK establishment in China. The limiting element was the levels of TCK contamination in US grain shipments, which failed to reach the levels necessary to cause disease.

References
1. Coakley SM, McDaniel LR, Line RF, 1988. Climatic Change 12, 57-75.
2. Goates BJ, Peterson GL, 1996. Phytopathology 87, S34.