Department of Plant Science and Research Institute for Agriculture and Life Sciences, Seoul National University, Seoul, Korea.
Phytopathology. 2012 Feb;102(2):147-57. doi: 10.1094/PHYTO-01-11-0018.
A rule-based model was developed to assess climatic risk of European canker (Neonectria galligena), which is a major disease of apple in some temperate zones. A descriptive rule was derived from published observations on climatic conditions favorable for European canker development. Fuzzy set theory was used to evaluate the descriptive rule quantitatively. The amount and frequency of rainfall and the average number of hours between 11 and 16°C/day were used as input variables whose values were matched with terms in the rule, e.g., 'high' or 'low'. The degree of a term, e.g., the state of being high or low, to a given input value was determined using a membership function that converts an input value to a number between 0 and 1. The rule was evaluated by combining the degree of the terms associated with monthly climate data. Monthly risk index values derived using the rule were combined for pairs of consecutive months over 12 months. The annual risk of European canker development was represented by the maximum risk index value for 2 months combined. The membership function parameters were adjusted iteratively to achieve a specified level of risk at Talca (Chile), Loughgall (Northern Ireland), East Malling (UK), and Sebastopol (USA), where European canker risk was known. The rule-based model was validated with data collected from Canada, Ecuador, Denmark, Germany, Norway, Poland, Sweden, the Netherlands, New Zealand, and the Pacific Northwest (USA), where European canker has been reported to occur. In these validation areas, the model's risk prediction agreed with reports of disease occurrence. The rule-based model also predicted high risk areas more reliably than the climate matching model, CLIMEX, which relies on correlations between the spatial distribution of a species and climatic conditions. The combination of a climatic rule and fuzzy sets could be used for other applications where prediction of the geographic distribution of organisms is required for climatic risk assessment.
一个基于规则的模型被开发出来，以评估欧洲溃疡（Neonectria galligena）的气候风险，这是一些温带地区苹果的主要病害。一个描述性规则是从关于有利于欧洲溃疡发展的气候条件的已发表观察中得出的。模糊集理论被用来定量评估描述性规则。降雨量的数量和频率以及 11 点至 16 摄氏度/天之间的平均小时数被用作输入变量，其值与规则中的术语相匹配，例如“高”或“低”。术语的程度，例如高低的状态，是通过使用隶属函数来确定的，该隶属函数将输入值转换为 0 到 1 之间的数字。规则是通过将与每月气候数据相关的术语的程度组合起来进行评估的。使用规则得出的每月风险指数值是通过将 12 个月中连续两个月的风险指数值组合而成的。通过将两个月的最大风险指数值组合在一起，代表了欧洲溃疡发展的年度风险。隶属函数参数通过迭代调整，以在智利塔尔卡、北爱尔兰洛格尔、英国东马勒姆和美国塞巴斯托波尔达到指定的风险水平，在这些地方，欧洲溃疡的风险是已知的。基于规则的模型在加拿大、厄瓜多尔、丹麦、德国、挪威、波兰、瑞典、荷兰、新西兰和美国太平洋西北地区收集的数据进行了验证，在这些验证区域，模型的风险预测与疾病发生的报告一致。在这些验证区域，该模型的风险预测比依赖物种空间分布与气候条件之间相关性的气候匹配模型 CLIMEX 更可靠地预测高风险区域。气候规则和模糊集的组合可用于其他需要进行气候风险评估的生物地理分布预测的应用。
