【作者】 高阳；

【导师】 段爱旺；

【作者基本信息】 中国农业科学院 ， 农业水土工程， 2009， 博士

【摘要】 间作种植具有充分利用多种资源、提高单位面积作物产量的优点。近些年来,随着人口的快速增加、耕地的不断减少和水资源的日益紧缺,我国的粮食供给安全受到了严重的威胁。在这样的背景下,深入研究间作冠层对辐射的截获与利用、间作群体的蒸腾耗散规律、间作模式下根系的分布及吸水规律,对于优化间作群体的组成结构、提高间作模式管理水平、保证间作种植具有较高的资源利用效率和产量水平都具有重要的理论和现实意义。本项研究于2006-2008年在河南商丘农田生态系统国家野外科学观测研究站进行,选取玉米/大豆间作这种在我国北方地区广泛采用的间作种植模式作为研究对象,设置4个试验处理,分别为玉米单作（SM）、大豆单作（SSB）、玉米/大豆1:3间作(I₁₃)和玉米/大豆2:3间作(I₂₃),重复4次。三年间,通过较为系统的田间试验,对玉米/大豆间作群体的光环境特性、土壤蒸发、植株蒸腾、根系分布特征、土壤水分运动以及资源利用等问题进行了试验研究和定量模拟,取得的主要研究结果如下:（1）在生育早期,I₁₃和I₂₃处理的大豆条带边行（与玉米相邻行）底部的PAR透射率高于大豆条带内行底部的PAR透射率,而I₂₃处理玉米行间底部的PAR透射率高于边行;进入生育后期,冠层底部不同位置处的PAR透射率差异不明显,平均透射率均小于7%; 4个试验处理的消光系数（K）的平均值分别为:SM处理0.46、SSB处理0.59、I₁₃处理0.51、I₂₃处理0.50。（2）在群体生育早期（LAI<1）,I₁₃和I₂₃处理小区内不同行位置处的土壤蒸发量相差不大。但随着群体的生长发育（LAI≥1）,不同位置处的土壤蒸发量呈现出一定的差异。土壤蒸发过程受表层土壤含水率和叶面积指数影响较大,相对土面蒸发强度E/ET0与表土土壤含水率呈指数性正相关关系,与叶面积指数则呈指数性负相关关系。I₂₃处理间作群体内,玉米和大豆植株的茎流速率在晴天呈单峰曲线,而在阴天则呈多峰曲线。植株的茎流受诸多环境因子的影响,其中太阳辐射是影响植株茎流的最主要的气象因子。玉米和大豆的单株日茎流量与多个气象因子间存在较好的相关关系,达到极显著水平。茎流观测期内（2008年6月1日-6月30日）,I₂₃处理玉米植株的日均蒸腾量（茎流量）（1.44 mm d-1）为大豆日均蒸腾量(0.79 mm d^-1)的1.8倍,玉米和大豆的蒸腾量分别为间作群体总蒸腾量的64%和36%。利用ERIN模型模拟间作群体的土壤蒸发和植株蒸腾时,阻力项的确定是至关重要的。I₂₃处理间作群体内,玉米/大豆间作群体的土壤表面阻力为300 s m^-1,间作冠层的空气动力学阻力r^a_a和r^s_a分别为4.50和42.41 s m^-1,冠层边界层阻力分别为12.19和38.89 s m^-1,玉米和大豆的气孔阻力分别为81.32和64.92 s m^-1。模拟结果表明,间作群体内玉米或大豆日蒸腾量的模拟值均高于实测值,土壤蒸发的模拟值为实测值的94.67%。相关分析表明,观测值和模拟值间有较好的相关关系,相关系数在0.8以上。ERIN模型能够较为准确地模拟间作群体的土壤蒸发和作物蒸腾,亦可以较为准确地确定蒸腾量在间作作物间的分配。（3）采用冲洗土壤剖面法来获取根系的分布模式。结果表明,水分充足条件下,I₂₃处理间作群体内,玉米的根系深度要大于大豆根系;玉米根系不仅分布于玉米条带下方的区域,而且可以扩展到邻近的大豆条带内行下方的区域,而大豆根系则水平分布于大豆行对应的相对有限的区域内。由于犁底层和粘土夹层（深度约为30 cm）的存在,16²2 cm土层内玉米和大豆根系的侧向生长距离最远,而根长密度则主要分布在近地表处（0³0 cm）以及作物行处。拟合结果表明,指数模型可以很好地描述I₂₃处理间作群体内作物根长密度的二维分布状况。在生长过程中,I₂₃处理间作条带内不同区域的水分状况变化幅度依次为:玉米区域>大豆区域>条带行间。这表明在水分充足条件下,即使根系在大部分土体中都有分布,但同一时间内并非所有根区都具有等效的吸水作用,作物会优先在间作群体中相应的土壤区域内吸水,而后才会从根系混合区域内吸收水分。间作群体内,不同作物生长到一定程度后根系会出现混合交叉分布,要准确地确定间作群体中不同作物的根系密度分布状况是比较困难的。本文利用较为容易获得的参数建立了I₂₃处理间作作物根系的二维分布模式,并依据所建立的二维根系吸水模型,利用用HYDRUS-2D软件模拟了I₂₃处理间作条件下的土壤水分运动。模拟结果表明,所建立的二维根系吸水模型能够比较好的模拟间作群体的土壤水分运动规律。（4）利用Logistic方程拟合了单作和间作条件下玉米和大豆单株干物质积累的动态过程,相关性均达到极显著水平。由于边际效应的作用,I₁₃和I₂₃处理间作群体内单株玉米的地上部干物质量和籽粒产量都要明显高于单作玉米,I₁₃和I₂₃处理间的差异并不显著。然而不同种植模式下单株大豆的地上部干物质量和籽粒产量的差异并不显著。I₁₃和I₂₃处理内玉米的干物质转换率高于SM处理,而SSB处理大豆的干物质转换率则高于间作处理,I₁₃和I₂₃处理间的差异并不显著。I₁₃和I₂₃处理玉米的籽粒产量分别为SM处理的83%和95%,表明在间作种植模式下,玉米的增产效果并不足以弥补大豆占用面积所引起的籽粒产量下降。I₁₃和I₂₃处理大豆的籽粒产量分别为SSB处理的82%和76%,表明当大豆与玉米间作时,玉米的遮荫导致大豆籽粒产量下降。但是,玉米/大豆间作的总籽粒产量仍要明显高于单作玉米和单作大豆,表明间作种植能够增加群体的干物质积累量,提高群体总产量。（5）利用辐射截获率（F）、辐射利用效率（RUE）和收获指数（HI）3个指标,评价了玉米/大豆间作群体对辐射的截获与利用状况。3个生育期内,SM、I₁₃和I₂₃处理玉米收获指数（HI）的平均值分别为0.42、0.45和0.44;SSB、I₁₃和I₂₃处理大豆的收获指数（HI）的平均值分别为0.40、0.35和0.36;I₁₃和I₂₃处理玉米的RUE分别为3.14和3.13 g MJ^-1,略低于SM处理的RUE(3.18 g MJ^-1)。I₁₃和I₂₃处理大豆的RUE分别为1.65和1.63 g MJ^-1,略高于SSB处理的RUE(1.55 g MJ^-1)。I₁₃处理的RUE(2.82 g MJ^-1)略高于和I₂₃处理(2.78 g MJ^-1)。I₁₃和I₂₃处理群体的RUE比SM处理低11%和13%,但要比SSB处理高82%和79%。数据显示,间作种植模式能够通过更有效地利用辐射来增加群体的产量。3个生长季内,I₁₃处理ETc的平均值（495.84 mm）略低于I₂₃处理（506.16 mm）。I₁₃处理的ETc分别比SM和SSB处理高15%和6%;I₂₃处理的ETc分别比SM和SSB处理高17%和9%。I₁₃和I₂₃处理间作群体的WUE(I₁₃:20.24 kg ha^-1 mm^-1;I₂₃:21.97 kg ha^-1 mm^-1 )略低于SM处理( 22.67 kg ha^-1 mm^-1),但明显高于SSB处理(5.07 kg ha^-1 mm^-1)。I₁₃和I₂₃处理的土地当量比（LER）分别为1.65和1.71,表明单作种植要多利用65%和71%的土地才能得到与I₁₃和I₂₃处理间作种植相同的产量。试验数据同时指出,玉米/大豆条带间作种植具有明显的产量优势,可以显著提高土地利用效率。更多还原

【Abstract】 Intercropping is an effective way of using several kinds of natural resources more effectively and improving yield per unit of cultivated land. In recent years, natural resources, such as arable land and water, are becoming more and more limited with quick social development and rapid growth of industry and population, which constitutes a serious threat on food supply security in China. Therefore, it is necessary meaningful to study the radiation interception and utilization by intercropping canopies, the soil evaporation, plant transpiration, and root water uptake dynamics in intercropping systems for optimizing components, improving management and rising resources use efficiency and output of intercropping systems.Field experiments were conducted at Shangqiu Agro-ecosystem Experimental Station in 2006-2008 years to study the maize/soybean intercropping system, a widely practiced intercropping model in Northern China. The field experiment comprised of four treatments with four replications. Four treatments were set as: sole maize （SM）, sole soybean （SSB）, maize/soybean 1:3 intercropping (three rows of soybean by every one row of maize, I₁₃) and maize/soybean 2:3 intercropping (three rows of soybean by every two rows of maize, I₂₃). Light environment characteristics, soil evaporation, plant transpiration, root distribution, soil water movement and resources utilization efficiency in the four different cropping systems were investigated and simulated with field experimental data got in three years. The main results were as following:（1） About the PAR radiation interception and distribution in intercropping systemsIn earlier growing stage, photosynthetically active radiation （PAR） transmittance at the bottom of edge row of soybean strip adjacent to maize row in I₁₃ and I₂₃ treatments was higher than that of inner row of soybean strip, while it was in adverse for maize strip in I₂₃ treatment. In later growing stage, PAR transmittance at the bottom of the intercropping canopies did not vary significantly, and the averaged PAR transmittance was less than 7%. The averaged extinction coefficient （K） of treatment SM, SSB, I₁₃ and I₂₃ over three growing seasons was 0.46, 0.59, 0.51 and 0.50, respectively.（2） About soil evaporation and plant transpirationIn earlier growing stage （LAI<1）, there were little differences for soil evaporation at different locations in I₁₃ and I₂₃ treatments. With development of crop canopy （after LAI≥1）, the differences among soil evaporation rates at different locations became more and more significant. Soil evaporation rate was mainly influenced by water content in surface layer of soil and leaf area index （LAI）. The relative soil evaporation intensity （E/ET0） showed a good positive exponential growth relation with water content in 0-5 cm layer of soil, and a good negative exponential growth relation with LAI.The diurnal variation of maize and soybean plant sap flow in I₂₃ treatment fitted a single-peak curve in sunny day and multi-peak curve in cloudy day, respectively. Sap flow was influenced by many environmental factors, but showed a closest relationship with solar radiation. The correlations between daily sap flow of maize or soybean and environmental factors was extremely significant. During observation period of sap flow （June 1st– June 30th , 2008）, mean diurnal transpiration value for maize plant （1.44 mm d-1） in I₂₃ treatment was 1.8 time of that for soybean plant （0.79 mm d-1）, and maize transpiration and soybean transpiration contributed 64% and 36% to the total transpiration in I₂₃ treatment, respectively.It is very important to determine correctively relevant resistances for modeling soil evaporation and plant transpiration in intercropping systems with ERIN model. In maize/soybean 2:3 intercropping system (I₂₃ treatment), surface resistance of substrate soil was 300 sm^-1, aerodynamic resistance r^a_a and r^s_a was set as 4.50 and 42.41 sm^-1, boundary layer resistances as 12.19 and 38.89 sm^-1 for maize and soybean canopies, and stomatal resistance as 81.32 and 64.92 sm^-1 for maize and soybean leaves, respectively. Results indicated that simulated maize or soybean transpiration value was higher than measured value, and calculated soil evaporation value 94.67% of measured value in I₂₃ treatment. Statistics analysis showed that there was a significant correlation relationship between measured values and simulated values with a correlation coefficient of greater than 0.8. Simulated results showed ERIN model in relative excellent for estimating soil evaporation and plant transpiration of intercropping systems, and for determining transpiration contribution of each component in an intercropping system.（3） About root distribution and water uptakeRoot growth and distribution of intercropping system was investigated with washing out, numbering, sampling and measuring growing roots from a thin practical soil profile on site. Results indicated that developed depth for intercropped maize was greater than that for intercropped soybean in I₂₃ treatment under full irrigation. In maize/soybean 2:3 intercropping system (I₂₃ treatment), maize roots not only developped horizontally in the soil zone just under the maize strip, but also extended in adjacent soil zone and can reach the position just vertical under the middle row of the soybean strip. For soybean, however, most roots developped horizontally in the soil zone just under the soybean strip. The greatest horizontal extension distance for maize and soybean roots appeared in 16²2 cm soil layer from surface, relative to the existing of a plough pan and clay intercalation.Root length density （RLD） of maize and soybean in I₂₃ treatment were primarily distributed in surface soil layer of 0³0 cm and in soil zone under crop rows. An exponential model fitted very well to the 2D distribution of root length density for I₂₃ treatment. In maize-soybean 2:3 intercropping system (I₂₃ treatment), the ranges of soil water content change were in the order: maize zone>soybean zone>middle zone. This order indicated that each crop preferentially absorbed soil water from soil zone just under its strip, and then from adjacent mixed zone in intercropping system.It was very difficult to determine accurately root density distribution in intercropping system, because of mixed root distribution of different crops. A two-dimensional root distribution and water uptake model for intercropped crops was developed with some easily acquired parameters. Soil water transport in intercropping system (I₂₃ treatment) was simulated with HYDRUS software package and the two-dimensional root water uptake model. Results showed that the model was suitable for simulating soil water transport in intercropping system.（4） About biomass accumulation and yieldBiomass accumulation of maize and soybean was fitted with Logistic equation for both the intercropping and monocultures, and correlation reached very significant level （P<0.01）. Mean above-ground biomass accumulations and grain yield of each intercropped maize plant were higher than that of each sole maize plant because of edge effect. There was no significant difference for biomass and yield between I₁₃ and I₂₃ treatments. However, there was little difference between intercropped soybean plant and sole soybean plant. Comparing with sole maize, the intercropped maize had higher translocation rate of biomass. Translocation rate of biomass of the sole soybean was slightly higher than that of the intercropped soybean. There was no significant difference for translocation rate of biomass of soybean in I₁₃ and I₂₃ treatments.Grain maize yield for I₁₃ and I₂₃ treatments was 83% and 95% of that for SM treatment, respectively, which indicated that the grain yield increase caused by edge-effect was not enough to compensate the grain yield lose caused by maize occupied area decrease. Grain soybean yield for treatment I₁₃ and I₂₃ was 82% and 76% of that for SSB treatment, respectively, wihch showed that the shading effects of maize to soybean in maize/soybean intercropping system. The total yield of maize/soybean intercropping system, however, was significantly higher than the yield of each sole cropping, which showed that the advances of intercropping in increasing biomass and yield for per unit of cultivated land.（5） About resource use efficiencyPAR interception and utilization by maize/soybean intercropping canopy were evaluated with the fraction of radiation intercepted （F）, radiation use efficiency （RUE） and harvest index （HI）. The averaged maize HI for treatment SM, I₁₃ and I₂₃ over three growing seasons was 0.42, 0.45 and 0.44, respectively; the averaged soybean HI for treatment SSB, I₁₃ and I₂₃ was 0.40, 0.35 and 0.36, respectively. Maize RUE of treatment I₁₃ and I₂₃ was 3.14 and 3.13 g MJ^-1, slightly less than that of treatment SM (3.18 g MJ^-1). Soybean RUE of treatment I₁₃ and I₂₃ was 1.65 and 1.63 g MJ^-1, slightly higher than that of treatment SSB (1.55 g MJ^-1). RUE of treatment I₁₃ (2.82 g MJ^-1) was slightly higher than that of treatment I₂₃ (2.78 g MJ^-1). RUE of treatment I₁₃ and I₂₃ was less than that of treatment SM by 11% and 13%, but higher than that of treatment SSB by 82% and 79%, respectively. The results indicated that intercropping may increase grain yield per unit of land by more efficient radiation utilization.Mean evapotranspiration （ETc） of I₁₃ treatment （495.84 mm） was slightly lower than that of treatment I₂₃ （506.16 mm） over three growing seasons. ETc of I₁₃ treatment was 15% and 6% higher than that of SM and SSB treatments, respectively. ETc of I₂₃ treatment was 17% and 9% higher than that of SM and SSB treatments, respectively. Water use efficiency （WUE） of I₁₃ and I₂₃ treatments (I₁₃:20.24 kg ha^-1 mm^-1;I₂₃:21.97 kg ha^-1 mm^-1) was slightly lower than that of SM treatment (22.67 kg ha^-1 mm^-1), but significantly higher than that of SSB treatment (5.07 kg ha^-1 mm^-1). Land equivalent ratio （LER） of I₁₃ and I₂₃ treatments was 1.65 and 1.71, which derived that 65% and 71% more land in the monocultures was required than that in the intercropping for producing same grain, and that maize/soybean intercropping has significant advantage of improving yield and land use efficiency.更多还原