【作者】 王倩；

【导师】 毕军； 刘苗苗；

【作者基本信息】 南京大学 ， 环境规划与管理， 2021， 硕士

【摘要】 近地面臭氧由于其较强的氧化性,会对农业生产造成损害,导致农作物减产。当前有关臭氧胁迫对农作物生产影响的研究在发展中国家普遍受到方法学和数据方面的限制,在进行历史臭氧减产效应评估时空间尺度不够精细,时间趋势分析不足,在探究未来粮食生产安全影响时对气候变化与大气污染的复杂交互机制考虑不够充分。因此,本研究以提高臭氧对区域农作物产量损失评估的精准性,探究时空变化趋势目标,以期为区域臭氧污染管控与农业生产管理提供科学建议。其次,亟需借鉴国际成熟经验,发展耦合臭氧与环境多要素的评估模型,合理开展臭氧对农业生产的影响评估,以期为气候变化和空气污染背景下未来农作物产量预测分析提供支撑。本研究选择江苏省作为研究对象,利用农作物生长发育旬值数据、臭氧模拟数据、气象地面观测数据、农作物年产量等数据,结合空间分析、统计建模等方法,表征了2007-2019年江苏省97个县级行政区冬小麦和水稻生长期的臭氧暴露剂量AOT40、温度和降水量。在此基础上,分别构建了基于暴露试验剂量-响应函数和面板回归模型两种评估臭氧对农作物产量损失的方法,精细化评估了历史臭氧污染导致的江苏省水稻和冬小麦产量损失范围及其时空变化特征,探究了统计方法评估结果的准确度。主要研究结论如下:(1)研究时期内臭氧和气候要素的时空变化特征显示,冬小麦生长期温度峰值较多分布在苏南地区,其次是徐州;水稻生长期温度峰值主要分布在苏南地区。两种农作物生长期内降水量均呈现从北向南逐渐增加的趋势。总体上臭氧暴露剂量随时间呈增长趋势,特别是在2013年后。空间分布上,冬小麦生长期苏北地区,特别是徐州臭氧污染最为严重;水稻生长期臭氧污染在苏南地区较为严重,东部沿海地区浓度相对较低。AOT40随时间变化趋势与温度变化呈现显著正相关,与降水量变化变化呈现负相关。(2)基于暴露试验剂量-响应关系的评估结果表明,2007-2019年,臭氧污染导致江苏省冬小麦年相对产量损失16.56%-44.48%,年产量损失达3.83×10~6-7.58×10~6 t,相当于3-6千万人一年的粮食消费量。臭氧污染导致江苏省水稻年相对产量损失6.21%-18.39%,年产量损失达1.82×10~6-3.13×10~6 t,相当于1.5-2.5千万人一年的粮食消费量。本身越不发达,经济更依靠农业的区县受臭氧污染导致的产量损失更为严重,大气污染加剧了区域发展的不平衡。(3)基于面板回归模型的拟合结果表示,臭氧及臭氧和气候要素的交互项均对冬小麦和水稻单产影响显著,表明冬小麦和水稻对臭氧的响应均取决于农作物生长期的温度和水分状况。2007-2019年,臭氧污染导致冬小麦年平均相对产量损失18.30%(7.21%-31.55%),冬小麦面板回归模型具有相对较好的拟合效果。水稻面板回归模型的产量损失结果和响应系数与暴露试验存在差异。未来可能需要进一步扩展纳入模型的地区和时间范围,以及使用不同的臭氧和气象因素表征指标,进一步搭建模型稳健、结果可靠的计量关系,从而支撑未来气候变化和大气污染不同情景下农作物产量的预测分析。本研究精细化评估了江苏省历史臭氧污染导致的农作物产量损失,构建及验证了耦合臭氧与气象要素的冬小麦面板回归模型的科学性,拓展了统计模型在我国臭氧减产效应评估方面的应用。研究成果可为全国以及其他地区量化、精确化农业风险水品提供科学参考,为制定臭氧污染防治政策提供有效支撑。更多还原

【Abstract】 Ground level ozone,due to its oxidation,can cause damage to agricultural production and reduce yield.At present,the research on the impact of O₃-induced crop yield losses in developing countries is generally limited by methodology and data.In analysis of ozone damage to historical crop yields,the spatial scale is not precise enough,the time trend analysis is insufficient.The complex interaction mechanism between climate change and air pollution is not fully considered when exploring the future food production safety.Therefore,this study is to improve the accuracy of crop loss from O3 and make suggestions for regional ozone pollution control and agricultural production management.Secondly,it is urgent to learn from the international experience to assess O3-induced yield for crop,using the statistical model with ozone and environmental factors.It is helpful to predict and analyze the future crop yield under the climate change and air pollution.In this study,we selected Jiangsu province as research area.Based on ten day data of crop growth and development,ozone monitoring data,meteorological ground observation data,annual crop yield and other data,combined with spatial analysis,statistical modeling and other methods,the ozone exposure dose（AOT40）,temperature and precipitation of 97 county of Jiangsu province during the growing season of winter wheat and rice in 2007-2019 were characterized.On this basis,two methods based on the dose-response functions and panel regression model were constructed to evaluate the yield loss of rice and winter wheat induced by ozone pollution in Jiangsu province.The range and spatiotemporal variation characteristics of yield loss of rice and winter wheat caused by ozone pollution in history were refined,and the accuracy of statistical methods was explored.The main conclusions are as follows.（1）The temporal and spatial variation characteristics of ozone and climate factors in the study period showed that the peak temperature during winter wheat growing seasons was mainly distributed in southern Jiangsu,followed by Xuzhou.And the peak temperature during rice growing season was mainly distributed in southern Jiangsu.The precipitation of the two crops increased gradually from north to south.In general,the ozone exposure dose increased with time,especially after 2013.In terms of spatial distribution,ozone pollution is the most serious in Northern Jiangsu during winter wheat growing seasons,especially in Xuzhou.Ozone pollution is more serious in southern Jiangsu during rice growing seasons,and the concentration is relatively lower in eastern coastal areas.The variation trend of AOT40 with time is significantly positively correlated with temperature,and negatively correlated with precipitation.（2）The results showed that the annual relative yield loss of winter wheat ranged from 16.56%to 44.48%in 2007-2019.The annual yield loss was 3.83×10⁶-7.58×10⁶ t,supporting grain consumption of 30-60 million people per year.The annual relative yield loss of rice in Jiangsu province ranged from 6.21%to 18.39%.The annual yield loss was 1.82×10⁶-3.13×10⁶ t,supporting grain consumption of 15-25 million people per year.The less developed the counties are,the more dependent the economy is on agriculture,the more serious the output loss caused by ozone pollution,and the air pollution aggravates the imbalance of regional development.（3）The fitting results based on panel regression model show that ozone and the interaction between ozone and climate factors have significant effects on winter wheat and rice yield,indicating that the response of winter wheat and rice to ozone depends on the temperature and water status of crop growth period.From 2007 to 2019,the annual relative yield loss of winter wheat caused by ozone pollution is 18.30%（7.21%-31.55%）,and the panel regression model of winter wheat has relatively good fitting effect.The results of yield loss and response coefficient based on rice panel regression model were quite different from those of exposure chamber.In the future,it may be necessary to further expand the area and time range included in the model,and use different ozone and meteorological indicators to further build a robust and reliable measurement relationship of the model.So as to support the prediction and analysis of crop yield under different scenarios of climate change and air pollution in the future.Through the construction of high spatial-temporal resolution and long-time scale ozone simulation data,this study finely evaluated the loss of crop yield caused by historical ozone pollution in Jiangsu province.This study constructs and verifies the scientific nature of the winter wheat panel regression model coupled with ozone and meteorological factors,and expands the application of ozone yield reduction effect evaluation method in China.The research results can provide scientific reference for the quantification and precision of agricultural risk in China and other regions,and provide effective support for the formulation of ozone pollution control policies.更多还原