【作者】 孙雯；

【导师】 黄耀；

【作者基本信息】 南京农业大学 ， 生态学， 2011， 博士

【摘要】 农业生产与气候变化密切相关,在目前全球变暖趋势愈发显著的背景下,气候变化对作物生产的影响已不容忽视。水稻是世界主要粮食作物,占据世界人口一半以上的亚洲、非洲及拉丁美洲国家人民多以水稻作为主要食物来源。中国是世界上最大的水稻生产国,水稻种植面积和总产量分别占据世界的第二和第一位。水稻在中国的产量超过了谷物类总产的40%。由于气候变化本身的区域性和季节性,以及中国水稻生产的地域性,气候变化给水稻生产带来的影响并不完全一致,具有很大的不确定性。中国以不到全球10%的农田,承载了全球近20%的人口,气候变化对其水稻生产的影响将直接关乎世界近1/5人口的粮食安全问题。毫无疑问,粮食安全将成为中国当前以及未来可预期的相当长一段时间内无法忽视的问题；如何保证粮食安全是亟待研究的重大课题。因此,本研究将采用定量化分析方法,分析气候变暖给中国地区水稻生长发育及水稻生产带来的影响,以求为确定水稻的种植区域、合理安排农作制度提供科学依据。本研究通过气候变化、水稻物候等观测及模拟数据,采用时间序列分析等统计方法及模型模拟方法,基于地理信息系统(GIS)技术,从时间、空间上阐明了水稻安全生长季变化及生长季内关键生育期极端温度胁迫的变化情况,同时在此基础上修正水稻光温生产潜力模型,评估中国历史及未来水稻光温生产潜力和产量的变化。主要研究结果为：(1)1961-2008年间,水稻安全生长季长度阶段性变化明显,延长趋势主要出现在1985以后。水稻安全生长季在东北单季稻区,长江中下游单季稻区,长江中下游双季稻区及华南双季稻区都有显著的延长趋势,分别为9d/10a、6d/10a、7d/10a和7d/10a。东北稻区安全生长季长度的增加是由安全播种期和安全成熟期的变化同时引起的,而长江中下游地区和华南地区多是由于安全播种期的提早引起的。安全播种期的提前比安全成熟期的推迟更为显著,对生长季延长的贡献也大于安全成熟期。空间上,水稻安全生长季最长出现在云贵高原地区,最短出现在东北地区北部。东北和长江中下游地区安全生长季长度的变化最为明显,增加均超过10d。(2)1961-2008年间,不同种植区水稻关键生育期低温冷害和高温热害的变化趋势具有明显的时空分布特征。总体而言,气候变暖背景下低温冷害逐渐减少,而高温热害并无增加。其中,低温冷害减少主要表现在育秧期和抽穗开花期,孕穗期无明显变化特征。水稻各种植区中,只有东北单季稻区,长江中下游双季稻区和华南双季稻区三个地区的低温冷害出现减少趋势,低温冷害指数在育秧期的减少幅度分别为0.6/10a、1.2/10a和1.6/10a,在抽穗开花期的减幅分别为0.2/10a、0.3/10a和0.4/10a。冷害减缓趋势开始显著的时间呈现出从北向南逐渐推迟的格局,育秧期东北单季稻区和长江中下游双季稻区起始于80年代中期,而华南双季稻区从90年代中期才开始显著；抽穗开花期起始时间在东北单季稻区、长江中下游双季稻区和华南双季稻区分别为70年代初期、80年代中期和90年代初期。水稻种植区高温热害并未出现增加趋势,仅有长江双季早稻区的高温热害在2000s出现的次数较为频繁,但近两年又有所缓和,因此无法断定该区的高温热害的变化趋势。通过在县级和省级尺度上对低温冷害和高温热害与水稻单产的相关关系进行分析发现,抽穗开花期极端温度胁迫指数(△TSI)升高1则会造成水稻减产291kg ha-1,而在育秧期极端温度胁迫指数(△TSI)同样升高1会水稻仅减产165kg ha-1。(3)1961-2050年间,水稻各种植区光温生产潜力的变化有所不同。东北单季稻区、长江中下游稻区和华南双季稻区1961-2008年光温生产潜力明显增长,其中长江中下游双季稻区的增幅较大,每年约增加150kg ha-1;与此相反的是,四川盆地光温生产潜力每年减少75kg ha-1,减少趋势显著。未来40年间,东北、四川盆地和云贵高原三个单季稻区以及长江中下游双季稻区均表现出极其显著的上升趋势,在A2和B2情景下的增长趋势分别为每年136kg ha-1、79kg ha-1、116kg ha-1、85kg ha-1和每年91kg ha-1、67kg ha-1、118kg ha-1、89kg ha-1。空间上看,1981-2050年间光温生产潜力在东北稻区由北向南呈阶梯状增加,在南方稻区从沿海向内陆呈环状逐渐递减。最高的值出现在云贵高原西南地区及华南的部分地区,最低值则出现在东北稻区的北部。水稻单产总体表现为增加趋势,在1961-2008年间增加趋势极其显著,但在2011-2050年间增幅则较小。空间上,水稻单产在不同阶段的分布特征不同。1981-2008年,在东北稻区由周边向中部呈环状逐渐增加,在南方稻区从东南沿海向西北内陆逐渐递增,最高的单产出现在四川盆地。未来40年水稻单产在东北地区呈现由东北向西南阶梯状递增,在南方地区则呈高低相间的带状分布。最高的单产出现在东北稻区西南部和四川盆地东部,最低的则出现在两个双季稻区。本研究主要结论：1961-2008年间,水稻安全生长季总体变长,延长趋势具有明显的时空特征,其中安全播种期的提前比安全成熟期的推迟更为显著,对生长季延长的贡献也大于安全成熟期；关键生育期低温冷害有所减少而高温热害并未增加,冷害减少主要出现在育秧期和抽穗开花期；光温生产潜力的变化在各个种植区各有不同,主要表现为1961-2008年间东北单季稻区、长江中下游稻区和华南双季稻区光温生产潜力增加而四川盆地降低,2011-2050年间所有稻区的光温生产潜力都表现出升高趋势。2011-2050年间水稻单产总体趋势增加,但增幅较小。更多还原

【Abstract】 Agricultural production is closely related to climate change. Under background of global warming, increasing political, economic and scientific attention has been drawn to the impacts of climate change on crop production. Rice is world’s major food crop, and serves as basic staple for more than half of the world’s population in Asia, Africa and Latin America. China is one of the world’s largest rice-producing countries, has the second largest area of rice cultivation and the highest rice production, and rice occupies more than 40% of the national cereals yields of China.Due to the spatial and temporal heterogeneity of climate, and the regionality and seasonality of rice planting, the climate change impacts on rice production is highly uncertainty. Agricultural land in China comprises less than 10% of world’s arable land area but feeds nearly 20% of the world population. The effects of climate change on rice production will be directly related to 1/5 of the world population’s food security. There is no doubt that food security will become a major problem in China at present and in the foreseeable future of next several decades. Food security is a major issue to be studied. In this study, by using quantitative method to analyze the influence of global warming on rice growth and production in China, we tried to provide scientific basis for determining the rice growing areas and arranging reasonable farming systems.Through observed and simulated climate and rice phenology data, using methods of statistical analysis like time series analysis, mathematical modeling and GIS technology, this study clarified spatial-temporal changes of rice growing season length and extreme temperature stress during different key growth stages of irrigated rice across China’s mainland. And by modifying the model of photosynthetic thermal productivity and adopting modified model, we estimated the historical and future rice yield and photosynthetic thermal productivity. Results of this study are presented as follows: (1) From 1961 to 2008, the change of safe growing season behaved periodically, with extending tendency after 1985. Rice safe growing season length in single rice region of Northeast and the mid-lower Yangtze River Valley, and double rice region of the mid-lower Yangtze River Valley and South China, had significantly increased with 9d/10a,6d/10a, 7d/10a and 7d/10a, respectively. Extending of growing season length in Northeast is because of the earlier safe sowing time and later mature time, and in the mid-lower Yangtze River Valley and South China is only due to the earlier of safe sowing time. Shifting of safe sowing time is more significant than that of mature time, and contributed greater to the change of growing season. Spatially, the longest safe growing season of rice appeared in Yunnan-Guizhou Plateau, and the shortest in the northern area of the Northeast China. Northeast and the mid-lower Yangtze River Valley had the most significant change trend in growing season length, with an increase of more than 10d.(2) From 1961 to 2008, low-and high-temperature stress during rice critical growth period in different planting regions distributed temporally and spatially. In general, global warming reduced low-temperature stress but did not increase high-temperature stress. Reduction of low-temperature stress mainly occurred in seedling and heading-flowering stage, no significant variation in booting. Among all planting regions, only single rice region of Northeast, double rice region of the mid-lower Yangtze River Valley and South China showed a decreasing trend of low-temperature stress, with LTSI reduction in seedling stage of 0.6/10a,1.2/10a and 1.6/10a respectively, in heading-flowering stage of 0.2/10a,0.3/10a and 0.4/10a respectively. The year when low-temperature stress starting reducing significantly appeared to be gradually later from the north to the south:during seedling stage low-temperature decreased since the mid-1980s in the Northeast and the mid-lower Yangtze River Valley, and did not drop until the 1990s in South China; during heading-flowering stage low-temperature decreased in the three regions started from the early of 1970s, the mid-1980s and the early of 1990s, respectively. No significant trends in HTSI were observed in all rice planting regions, while relatively high HTSI occurred in the mid-lower Yangtze River Valley during the 2000s. This relatively high HTSI did not suggest an increasing trend because it declined in the last two years. Through the analysis on the relation between extreme temperature stress and rice yield in county and provincial level, we founded a yield reduction of 291 kg ha-1 whenΔTSI increased by one unit in heading-flowering stage, and reduction of 165 kg ha-1 when happened in seedlings stage.(3) From 1961 to 2050, the change of photosynthetic thermal productivity behaved differently in time and space. The photosynthetic thermal productivity in single rice region of Northeast, rice region of the mid-lower Yangtze River Valley and double rice region of mid-lower Yangtze River Valley showed a significant increasing trend during 1961 to 2008. Double rice region of mid-lower Yangtze River Valley had the most significant tendency, with an annual increase of 150 kg ha-1; contrarily, the photosynthetic thermal productivity in the Sichuan Basin reduced 75 kg ha-1 each year significantly. From 2011 to 2050, the photosynthetic thermal productivity in all rice planting regions had shown a rising trend. This trend in single rice region of Northeast China, Sichuan Basin and Yunnan-Guizhou Plateau, and double rice region of the mid-lower Yangtze River Valley were significantly; with increasing trend under the A2 and B2 scenarios of 136 kg ha-1,79 kg ha-1,116 kg ha-1, 85 kg ha-1 and 91 kg ha-1,67 kg ha-1,118 kg ha-1,89 kg ha-1 each year, respectively. In space, from 1981 to 2050 the photosynthetic thermal productivity in the Northeast increased gradually from north to south, and in the South from the coastal area to the inland decreasing gradually. The highest productivity appeared in southwest Yunnan-Guizhou Plateau and parts of Southern China, while the lowest value in the northern part of northeast. Rice yield increased significantly from 1961 to 2008, but only had a small rise from 2011 to 2050. Spatially, rice yield had different distribution characteristics during the two periods. From 1981 to 2008, the rice yield in the Northeast increased gradually from surrounding area to middle annularly, and in the south from the coastal area to the inland decreasing gradually. The highest value appeared in Sichuan Basin. During the next 40 years, the rice yield in the Northeast increases from north to south, and in the south distributed zonally. The highest value appeared in southern part of northeast and eastern part of Sichuan Basin, while the lowest value in the two double rice regions.In conclusion, from 1961 to 2008, safe growing season extended generally and had obvious temporal and spatial characteristics. The forward shift of safe sowing time is more significant than backward shift of mature time, and contributed greater to the change of growing season length. During all critical growth periods, low-temperature stress reduced but high-temperature stress did not increase. Reduction of low-temperature stress mainly occurred in seedling and heading-flowering stage. The photosynthetic thermal productivity changes in planting regions were significantly different. It increased in single rice region of Northeast, rice region of the mid-lower Yangtze River Valley and double rice region of the mid-lower Yangtze River Valley, but decreased in Sichuan Basin. From 2011-2050 years, the photosynthetic thermal productivity in all planting regions had shown a rising trend. Rice yield increased generally, with an increasing trend from 1961 to 2008, but a smaller rise trend from 2011 to 2050.更多还原