ARIMA模型同MAXENT模型在自然保护区内口蹄疫疫情风险预警中的应用研究

发布时间：2018-07-02 07:19

本文选题：口蹄疫 + 自然保护区　；参考：《东北林业大学》2015年硕士论文

【摘要】：随着全球气候的变化、环境的恶化、人类活动对气候以及生境的干扰等多方面原因的影响,人们开始对所有可能影响物种丰富度的因素重视起来。野生动物源性疫病正是危害物种多样性的一个重要因素,在近20年的时间里,野生动物源性疫病在全球范围内广泛传播。口蹄疫就是一种古老的野生动物源性疫病,在近些年来发病数也成增高趋势,对于感染的动物幼崽的致死率极高,很有可能会给动物的多样性带来影响。所以对该病进行早期预防显得尤为重要。本文通过利用MAXENT模型和ARIMA模型分别对口蹄疫的分布及发病数进行预测,希望可以通过将两个模型结合起来应用,为口蹄疫的发生进行预测预警,为制定防控策略提供参考。本文首先利用MAXENT模型对口蹄疫的全球分布进行预测：通过对文献的阅读,对OIE报道的整理,统计出世界范围内口蹄疫发生地的地理坐标点,从世界气象数据库中下载环境因子,包括生物因子、海拔、平均最高温度、平均最低温度和平均温度,将环境数据用DIVA-GIS软件进行格式转换,转换成MAXENT所需要的格式,然后与坐标数据一同输入到软件中进行处预测,最后得出口蹄疫在全球范围内的可能分布区域。并且通过对ROC曲线、刀切法图的分析,得出：口蹄疫一般在旱季、雨季、年降水量较多且气候为热带或者亚热带的地区发生的比较明显。根据MAXENT梯度预测图可以看出我国的口蹄疫高危分布区在我国的西南部,这一预测结果也同我国西南部气候条件相吻合。通过对我国西南部地区野生动物资源的了解,选择偶蹄类动物较多的西藏自治区进行ARIMA时间序列预测,验证ARIMA时间序列预测模型在口蹄疫疫病预测的可靠性。实验结果显示预测值可以同实际值很好的吻合,该模型适用于口蹄疫的短期预测。通过本文的研究得出：MAXENT模型可以科学的对口蹄疫可能发生区域进行风险预测,通过直观的梯度预测图,可以简单明了的表明出口蹄疫的高危分布区,ARIMA模型可以针对某一高危分布区,通过对发病情况资料收集整理,从而进行具体发病数的短期预测,为我国以后口蹄疫疫病的防控提供较为科学的参考。
[Abstract]:With the change of global climate, the deterioration of environment, the influence of human activities on climate and habitat interference, people begin to pay attention to all possible factors affecting species richness. Wild animal borne blight is an important factor that endangers species diversity. In recent 20 years, wild animal borne disease has been widely spread in the world. Foot-and-mouth disease (FMD) is an ancient wild animal blight. In recent years, the incidence of foot-and-mouth disease (FMD) is also increasing. The mortality of infected cubs is very high, which may affect the diversity of animals. Therefore, the early prevention of the disease is particularly important. In this paper, the distribution and incidence of foot-and-mouth disease are predicted by using Maxent model and Arima model respectively. It is hoped that the two models can be combined to predict and warn the occurrence of foot-and-mouth disease and to provide a reference for the formulation of prevention and control strategy. In this paper, the global distribution of foot-and-mouth disease (FMD) is predicted by using Maxent model. By reading the literature and sorting out the reports of OIE, the geographical coordinates of foot-and-mouth disease (FMD) occurring in the world are calculated, and the environmental factors are downloaded from the world meteorological database. Including biological factors, altitude, mean maximum temperature, average minimum temperature and average temperature, converting the environmental data into the format needed by using the DIVA-GIS software, converting it into the format needed by Maxent, and then inputting the coordinate data into the software for prediction. Finally, the possible distribution of foot-and-mouth disease in the world is obtained. Through the analysis of ROC curve and knife cutting chart, it is concluded that the occurrence of foot-and-mouth disease is more obvious in dry season, rainy season, annual precipitation and tropical or subtropical climate. According to the MAXENT gradient prediction map, it can be seen that the high risk distribution area of foot-and-mouth disease in China is in the southwest of China, and the predicted results are consistent with the climate conditions of southwest China. Based on the understanding of wildlife resources in southwestern China, Arima time series prediction was carried out in Tibet Autonomous region, where there were more cloven-hoofed animals, and the reliability of Arima time series prediction model for foot-and-mouth disease was verified. The experimental results show that the predicted values are in good agreement with the actual values, and the model is suitable for the short-term prediction of foot-and-mouth disease. Through the research of this paper, it is concluded that the proportion of MAXENT model can be used to predict the risk of foot-and-mouth disease. It can be shown simply and clearly that the Arima model of high risk distribution area of foot-and-mouth disease can be used for a certain high risk distribution area, by collecting and sorting out the incidence data, so as to make short-term prediction of the specific incidence number. It provides a scientific reference for the prevention and control of foot-and-mouth disease in China.
【学位授予单位】：东北林业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：S855.3

【参考文献】