【作者】 钱龙；

【导师】 王修贵；

【作者基本信息】 武汉大学 ， 水利水电工程， 2017， 博士

【摘要】 适宜的农田水分条件是农作物正常生长的基本保障。强降雨天气和农田排水不畅会使农田地下水位过高(渍害)或田面淹水(涝害)。当渍害和涝害影响了农作物的正常生长和产量形成时,就形成了涝渍胁迫。涝渍胁迫严重制约我国易涝易渍地区的农业生产活动。近年来,受全球气候变化的影响,我国洪涝渍灾害发生频率越来越高,影响范围越来越广,也给我国的农业生产带来了日益严重的威胁。长江中下游平原是我国三大棉产区之一,棉花种植面积约为我国棉花种植总面积的三分之一。受亚热带季风气候的影响,长江中下游平原在棉花的关键生育期内(6～8月)降水集中、雨量大、历时长,如遇农田排水系统不完善或管理运行不当,容易产生涝渍胁迫。由于棉花是一种对涝渍胁迫敏感的作物,涝渍胁迫会对当地的棉花生产造成严重影响。因此,根据当地农田的涝渍成灾特点,研究涝渍胁迫下的棉花响应规律、探索合理的农田排水指标以及提出涝渍胁迫条件下的棉花产量的模拟方法具有重要的现实意义。本文依托于国家自然科学基金项目“涝渍胁迫条件下旱作物水分生产函数研究”以及国家“十二五”科技支撑计划“农田除涝减灾工程综合控制技术及工程模式研究”,选取地处长江中下游平原的湖北省作为研究区域,将棉花作为研究对象。以2003至2011年连续多年开展了测坑涝渍试验的实测资料为基础,综合应用方差分析、回归分析、相关分析、结构方程模型以及动态产量模型等多种统计学方法和模型模拟方法,分析了不同涝渍发生生育期(苗期、蕾期、花铃期和吐絮期)和不同涝渍胁迫形式(单涝、单渍和先涝后渍)对棉花形态生长及产量的影响,提出了多生育期先涝后渍下的综合排水指标,并构建了涝渍胁迫下的经验型动态产量模型和机理型动态产量模型。主要得出了以下研究成果:(1)在棉花形态生长的响应方面,单渍胁迫即使历时更长,对棉花形态生长的影响仍小于单涝胁迫,而先涝后渍胁迫的抑制作用最大。先涝后渍胁迫发生在蕾期和花龄期内时均会显著(p<0.05)地抑制棉花的形态生长,而发生在吐絮期内时抑制作用很小。叶面积的生长对先涝后渍胁迫最为敏感,其余依次是茎粗和株高。在棉花产量的响应方面,单渍胁迫的减产作用小于单涝胁迫和先涝后渍胁迫。花铃期内遭受先涝后渍胁迫会导致棉花显著减产,蕾期次之,而吐絮期内减产作用较小。籽棉产量受单涝、单渍和先涝后渍胁迫的减产作用比干物质产量大。(2)运用结构方程模型构建了棉花在不同生育期内的“涝渍胁迫—高温天气—棉花生长/产量”交互关系模型。模型运行结果表明,加入对涝渍胁迫期间高温天气的考虑会影响棉花对涝渍最敏感生育期的判定。在考虑高温影响后,棉花花铃期是对涝渍最敏感的生育期,其涝渍敏感程度为苗期、蕾期和吐絮期的3倍以上。高温和涝渍胁迫均会对棉花产量造成显著的影响,但两者间的交互作用仅在部分试验中出现,这可能与涝渍影响的程度有关。另外,涝渍胁迫会显著影响棉花的生殖生长但不会显著影响营养生长,这可能与涝渍胁迫后营养生长的恢复生长有关。(3)在先涝后渍胁迫下,基于涝水和渍水时间划分的排水指标比基于地表水和地下水空间动态过程划分的排水指标更为合理。单渍的减产作用是单涝的0.759倍。先期的涝会增强棉花对后续渍的适应性,先涝后渍胁迫下渍的减产作用是涝的0.293倍。基于以上分析建立的综合排水指标CSFEW30与棉花减产率之间呈极显著的线性关系(R2=0.629,n=15,p<0.001)。(4)以水分亏缺条件下常用的两个动态产量模型为基础,建立了涝渍胁迫下的动态产量模型——改进Morgan模型和改进CROPR模型。两个改进模型均采用了与农田水位动态直接关联的作物响应函数,因此与农田排水中的水位管理衔接良好。两个模型各有优势,应根据实际情况进行选择。改进Morgan模型是经验型模型,所需资料少,模型模拟精度较高。改进CROPR模型是机理型模型,较好地反映了作物生长和响应的机理,模型模拟精度高。建立的两个改进模型均能较好地模拟涝渍胁迫下棉花的干物质生长过程及籽棉产量。综上所述,本文研究成果揭示了不同涝渍胁迫形式和不同涝渍发生生育期等多种客观存在的涝渍条件下棉花生长及产量的响应规律,提出了多生育期先涝后渍胁迫下的农田排水指标,并构建了与农田排水的水位管理密切关联的涝渍胁迫下的动态产量模型。研究成果可为洪涝渍灾害下棉田的排水规划、设计和运行管理提供参考依据。更多还原

【Abstract】 Appropriate water condition in the fields is the basic guarantee for normal growth of field crops.Heavy rains and poor field drainage usually bring perched groundwater(waterlogging)and ponded water(submergence)problems to the fields.When waterlogging and submergence hinder the growth and yield of field crops,then soil aeration stress occurs.Aeration stress seriously restrict the agricultural production in the districts liable to flooding.China is one of the countries in which flooding disasters frequently occur.As a result of global climate change,flooding disasters occur more frequently in China and bring greater damage to local agricultural production.The middle-lower reach of the Yangtze river plain is one of the three main regions for producting cotton in China,and its cultivating areas for cotton is about one third of the total cultivating areas for cotton in China.Due to the effect of subtropical monsoon climate,the middle-lower reach of the Yangtze river plain receives concentrated and heavy rains from June to August when local cotton plants are through critical growth stages.When drainage systems are defective and field water table managements are improper,aeration stress is likely to occur in the cotton fields.Considering that cotton is sensitive to aeration stress,the aeration stress frequently occurring from June to August in the middle-lower reach of the Yangtze river plain poses a great threat to local cotton production.Therefore,it is crucial to(ⅰ)study the responses of cotton to aeration stress,(ⅱ)explore the rational field drainage index,and(ⅲ)propose new yield-simulation methods under aeration stress.This study is financially supported by two State Natural Science Funds named"Study on dry crop water production function under water logging stress" and "Study on a unified model for crop water production function under water logging and drought stress".Hubei province,which is located in the middle-lower reach of the Yangtze river plain,was the study area;and cotton was the experimental crop.Lysimeter experiments under aeration stress were continuously performed from 2003 to 2011.Based on the data from these experiments,many statistical and simulation methods including variance analysis,regression analysis,correlation analysis,structural equation modeling and dynamic yield modeling were performed.Consequently,the response of cotton growth and yield under aeration stress with respect to different stress occurring time(at seedling stage,budding stage,flowering stage and boll-opening stage)and different stress forms(waterlogging,submergence,and submergence followed by waterlogging)was analyzed.Anew drainage index applicable to submergence followed by waterlogging during multi-growth stages was proposed.Finally,an empirical and dynamic yield model(improved Morgan model)as well as a mechanism and dynamic yield model(improved CROPR model)under aeration stress were established.The main results are as follows:(1)Waterlogging inhibited the growth of cotton morphology less than submergence even with a longer duration,while submergence followed by waterlogging stress inhibited the most.The aeration stress occurring in the budding stage and flowering stage inhibited the growth of morphology characteristics significantly(p<0.05),while the inhibition in the boll-opening stage was not significant.Leaf area index was the most sensitive morphology index describing the response of cotton to waterlogging stress,followed by stem diameter and plant height.The average yield reduction rate of waterlogging was smaller than that of submergence and submergence followed by waterlogging stress.Aeration stresses occurring in the flowering stage reduced cotton yield significantly,but little yield-reducing effect was found in the boll-opening stage.And the yield-reducing effect of aeration stress at the budding stage was between other two growth stages.The adverse effect of aeration stress on seed cotton yield was greater than that on dry matter yield.(2)Structural equation modeling(SEM)was performed to reveal the relationships among aeration stress,hot weather,cotton growth and yield at various cotton growth stages.The model results showed that including high temperatures would affect the identification of the growth stage most sensitive to aeration stress.When high temperatures were included,cotton sensitivity to aeration stress at the flowering stage was three-fold higher than those at other stages.Both aeration stress and high temperaures had significant effects on cotton yield while the interaction between them was only significant in one experiment,which may be related to the magnitude of aeration stress effect.Cotton reproductive growth was significantly impeded by aeration stress while vegetative growth was not.This result was probably caused by the recovery of vegetative growth,which was also demonstrated in this study.(3)Under submergence followed by waterlogging,the drainage indices based on division of submergence duration and waterlogging duration,compared with the indices based on division of surface and subsurface,can describe the relationship between the aeration-stress degrees and yield reduction more stably over different years.In the process of submergence followed by waterlogging,submergence would reduce the yield-reducing effect of waterlogging on cotton yield.And when this effect of submergence on waterlogging was included,the yield reduction of waterlogging was 0.293 times as much as that of submergence.Based on the above analysis,a new index named CSFEW30 used under submergence followed by waterlogging during multiple growth stages has been proposed.And this drainage index was proved to be significantly(R2=0.629,n=15,p<0.001)related to cotton yield reduction rate.(4)Two dynamic models(Morgan model and CROPR model),which are commonly used for simulating crop yield response under water deficit were improved to apply under aeration stress.The improved Morgan model is an empirical model while the improved CROPR model is a mechanism model.Both of these two improved models have adopted crop response functions that are well connected with field water tables;thus,they are practical in field drainage.Each of the two models has its own advantages,so one should adopt the proper model according to actual conditions.To be specific,the improved Morgan is an empirical model which only requires a small amount of data and its simulation accuracy is acceptable.The improved CROPR is a mechanism model which emphasize the crop response mechanism under aeration stress;thus,it performs better than the improved Morgan.Both of the two improved models performed well in simulating cotton dry matter growth and seed cotton yield under aeration stress.In summary,the results of this study have revealed the response of cotton growth and yield under various existing aeration-stress situations.And also,a drainage index applied under submergence followed by waterlogging during various cotton growth stages has been proposed.Moreover,the dynamic yield models that are closely related to field water table management have been established.The results of this study could offer valuable references for the planning,design and management of field drainage under flooding disasters.更多还原