【作者】 Tomoro Eric-Désiré；

【导师】 诸叶平；

【作者基本信息】 中国农业科学院 ， 作物信息科学， 2014， 博士

【摘要】 在中国,玉米是第二大粮食作物,它在农业生产和国民经济中具有重要的作用。然而,这种生产可能会受到区域气候变化的影响。相关研究表明,与2000年相比全国年平均气温将在2020年上升1.3-2.1℃,2050年上升2.3-3.3℃。这种变化可能会极大地影响河北地区的玉米生产,继而影响世界玉米产量并给世界粮食供应造成巨大损失。本文利用已知区域气候影响研究(PRECIS)的A2和B2情景下,通过作物模型模拟,讨论了气候变化对中国河北省玉米生长和产量的影响,并提出了一些相应的管理措施和战略。本研究选择了1980至2010年9个河北气象站的气象数据作为数据基础,采用中国农业科学院农业信息研究所农业智能实验室研制的玉米生产仿真系统(MPES),来模拟玉米在未来气候条件下的生长情况,同时还进行了该模型对天气的敏感性分析。此研究的天气预报数据由PRECIS模拟驱动(1961年至1990年)。MPES模型仿真生长的结果是通过连通性验证技术和模拟分析在SRES A2和B2情景下预测21世纪2020s、2050s和2070s天气得到的。因此,为了得到更好地预测产量,考虑到了播种和种植的调整类型和新品种的适应措施。仿真数据采自于河北省不同区域气象观测站以及中国气象局。MPES模型归中国农业科学院农业信息研究所农业智能实验室所有。研究的主要结果如下：1、通过改变初始天气数据获得MPES模型对天气的敏感性。建立基础未来气候评价模型评估气候适应性。该MPES模型是基于玉米在河北省气候条件的适应性和验证以及玉米参数而设计的。2、模型对玉米产量模拟很好,除了几年的仿真模拟值和观测值的生育期偏差外,大部分站点夏玉米模拟的归一化的均方根NRMSE是9.81%。9个站点测量到的数据与模拟比较相关系数R2的值是0.70,置信度σ=0.01。所有站点的模拟结果也很显著。模拟年际间变化很大,不同站点年份产量可能或高或低,但整体误差较小。该模型在多数情况下展示了良好的性能。3、MPES仿真模型可以有效的再现品种差别,并对产生更高产量的最佳生长期进行模拟。然而,有些结果出现了允许误差。模型模拟情景与观察到的玉米产量相比说明可以达到产量最大化。4、MPES模型对温度和降水敏感。可以很好地模拟未来气候变化对玉米生产的影响。遗传参数的敏感性分析表明模型对玉米生长参数更加敏感,从而为确定玉米育种的未来发展方向提供理论依据。5、分析并修订了青县,曲周等7个站点观测到的气象数据来模拟每个站点玉米生长期和产量,并与实测值进行比较。结果表明,区域气候PRECIS模型与MPES模型参数兼容,可以用于评估未来气候变化度作物的影响。6、在A2、B2情景下对滦县,深州,宁晋和其他6个站点2070年气候因素的各种变化进行了预测并加以评价。A2情景比B2情景温度升高更多,A2情景下除了大河和怀来2个站点的降水量呈小幅下降趋势外,其他站点降水量增加。总体来说,大多站点降水量有增加的趋势。7、气候变化将导致玉米生长期变短,A2情景的生育期比B2的长。预测21世纪70年代A2情景下产量平均下降20%,B2平均下降11%。8、适当调整播种和耕种方法及引进新品种可以有效地提高玉米的产量。需要改进农业管理系统,以及提高肥科利用率来促进未来农业可持续发展。更多还原

【Abstract】 Maize (Zea mays L.) is the second-largest food crop in China after rice (China Statistics Yearbook2012), and it plays an important role in agricultural production and in the national economy in the country. However, this production may be affected by climate changes in regional areas. The future prediction indicates that the nationwide annual mean air temperature would increase by1.3-2.1℃in2020and2.3-3.3℃in2050as compared with that in2000. This change may dramatically affect maize production in this part of the world and thus will decrease the world maize production, which will have huge damage on world’s food supply. This thesis discussed the effect of climate change on maize growth and the impact on the predictive production in Hebei Province of China using the Providing Regional Climates for Impacts Studies (PRECIS) climate prediction under SRES A2and B2scenarios, and then proposed some appropriate adaptations and strategies. About9sites were selected and data from the meteorological station in Hebei province were analyzed in this study, baseline1980to2010. The Maize Production Emulation System (MPES) simulation model was used to simulate the maize growth under extreme climate scenarios, which is the propriety of the Agriculture Intelligent Technology Laboratory of the Agricultural Information Institute of Chinese Academy of Agricultural Science. The applicability of MPES model was evaluated in China, and its sensitivity analysis to the weather was checked. The weather forecast data used for this research were derived by PRECIS simulation baseline (1961-1990). The results of the simulation growth using MPES model is a combination of both connectivity verification techniques and simulation analysis of each site in the SRES A2, B2scenarios under2020s,2050s and2070s future weather. Hence, the adjustment type of sowing and cultivation, the introduction of measures to adapt to the new breed to predict better yield were considered. The simulation data was collected from different meteorological stations across Hebei Province area and some from the Nation Statistics Bureau of China. These data include historical observed field information, daily meteorological data and soil parameters data. The data obtained was used to build a database about growth period and simulated yield. The settings and calibration of MPES model were based on further validation and analysis of adaptation and weather sensitivity were established by using the combination of temperature settings, rain and crops parameters. The main results were as follows:1. The sensibility to weather of MPES was obtained by changing the initial weather inputs. Its capacity of climate adaptation was also evaluated by setting up a basis as a model for evaluation of future climate conditions.2. The maize yield has excellent simulation with the Normalized Root Mean Square Error (NRMSE) of9.81%, for the simulation to the growing period of summer maize in most of the sites, apart from a few years’ simulation of growth period day’s deviation. The trends of the simulated and observed values were quite close. The measured data in9locations compared with the simulated gave a correlation coefficient R2of0.70with a confidence level σ=0.01. Simulation performance results were also significant in all sites. Simulation for different years showed wide variation, very high or low depending on the sites performance, but with a small error in the set. The model has shown its effectiveness in most of the study areas.3. The MPES model can reproduce the effect of variations in seed and simulate the best period of growth for better performance. However, some results showed some errors with reasonable limits. The performance of the model expressed by a better applicability of the simulation compared to the observed maize yield can be maximized.4. The MPES model is sensitive to temperature and precipitation. It can provide better approach for future climate changes impacts on maize production. The Sensitivity analyses of genetic parameters have demonstrated that the model is more sensitive to maize growth parameters, which provide a theoretical basis for determining the future direction of maize breeding.5. Observed weather data in Qingxian, Quzhou and7other sites were analyzed and revised to simulate the growth and yield of maize in each site when compared with the measured values. The results showed that the climate model SRES were compatible with the parameters of the model MPES and capacity can be adapted to assess the impact of future climate change.6. Prediction in Luanxian, Shenzhou, Ningji and other6sites under A2and B2scenarios in2070s of different changes in climatic factors were evaluated. In addition, the A2scenario, which provides a higher temperature increase in scenario B2, predicted an increase in precipitation, except for Dahe and Huailai under scenario A2which showed a slight downward trend sites. Overall, most sites showed a tendency to increased precipitation.7. Climate change will lead to a shorter period of maize growth. Maize growth period under A2scenario is longer than that of B2scenario. The predictions in yield provided a tendency to an average of20%decrease in A2scenario in2070s, as mean reduction of B2scenario is11%.8. To adjust appropriately seeding and cultivation, the introduction of new varieties can effectively improve the yield of maize. The management system of agriculture must be improved and the efficiency of fertilizers is necessary to be promoted for future agricultural sustainability.更多还原