格里芬双列杂交育种法3配合力分析的SAS实施

论述了如何应用SAS过程解决格里芬双列杂交育种法3的配合力分析问题，主要用到iml过程及各种平方和的变形公式，并以实例示明具体计算法。     

Application of thermal sums concept to estimate the time to harvest new banana hybrids for export

Application of the thermal sum concept was developed to determine the optimal harvesting stage of new banana hybrids to be grown for export. It was tested on two triploid hybrid bananas, FlhorBan 916 (F916) and FlhorBan 918 (F918), created by CIRAD's banana breeding programme, using two different approaches. The first approach was used with F916 and involved calculating the base temperature of bunches sampled at two sites at the ripening stage, and then determining the thermal sum at which the stage of maturity would be identical to that of the control Cavendish export banana. The second approach was used to assess the harvest stage of F918 and involved calculating the two thermal parameters directly, but using more plants and a longer period. Using the linear regression model, the estimated thermal parameters were a thermal sum of 680 degree-days (dd) at a base temperature of 17.0 °C for cv. F916, and 970 dd at 13.9 °C for cv. F918. This easy-to-use method provides quick and reliable calculations of the two thermal parameters required at a specific harvesting stage for a given banana variety in tropical climate conditions. Determining these two values is an essential step for gaining insight into the agronomic features of a new variety and its potential for export.     

Modelling species competition in mixtures of perennial sow-thistle and spring barley based on shoot radiation use efficiency

Abstract

Perennial sow-thistle (Sonchus arvensis L.) may be a serious weed in organic and conventional farming. To assess the effects of radiation acquisition and resource allocation on competitive ability, S. arvensis was grown together with spring barley (Hordeum distichon L.) in six mixtures in a replacement series with initial above-ground biomass proportions of S. arvensis ranging from 2% to 96%. A one-season experiment was made outdoors in boxes in Uppsala, Sweden, at a low level of nitrogen supply (5 g N m^?2). The study tested the predictability of shoot biomass of each species based on two principal assumptions: (i) growth model parameters derived from species in monocultures could be applied in mixtures, and (ii) radiation in the mixed stand was partitioned between species proportional to their leaf area. Calibration of two parameters, for scaling of shoot radiation use efficiency and radiation partitioning respectively, were the base for the evaluation. When the coefficients were close to unity, which was the case for all mixtures dominated by barley, and for one of the mixtures with high proportion of S. arvensis in the early season, observed and predicted shoot biomass coincided well. For the evenly composed mixtures, total shoot biomass was underestimated (the scaling coefficient of shoot radiation use efficiency was>1), whereas the relative composition among species was predicted well. In the late season the principal model assumptions were not applicable to S. arvensis, likely due to increasing root allocation not accounted for in the model. Sonchus arvensis in mixtures with high proportions was planted early in relation to sowing of barley, which resulted in a comparably late development stage of S. arvensis. Consequently the relation between species development stages varied with species composition suggesting a need to introduce effects of differences in development stage into the model.     

微粒群优化求解van Genuchten方程参数

van Genuchten方程参数的求解是复杂的非线性拟合问题,用传统方法难以有效解决。应用微粒群优化目的在于找到一条求解van Genuchten方程参数新的有效途径。以粉壤土吸湿和脱湿实测数据为实验,利用VisualBasic.NET语言进行编程,首先建立目标函数,然后计算每个微粒的适应值,通过微粒的自动寻优过程,找到最小误差平方和,最后记录下与之对应的参数,即为方程所求得解。通过最小误差平方和与其他方法进行比较,表明微粒群优化程序设计简洁、运行时间短、拟合精度高,是一种求解van Genuchten方程参数的有效方法。     

非线性模型最佳拟合方法的研究

本文对非线性模型具有代表性的拟合方法进行了比较研究，分析各种拟合方法存在的利与弊，研究结果认为三次设计和改进单纯形法都是非线性模型的最佳拟合方法。     

影响阿克苏棉区棉花气象产量的温度因子研究

通过阿克苏棉区棉花产量指数与以 0 .1℃为间隔的连续变动温度界限值所形成的 9种积温(或温度和 )的因子序列的相关性分析 ,得到以下几点结论 :(1 )相关性以日最高温度的温度和为最好 ,日平均温度积温次之 ,日最低温度的温度和最差。 (2 )在三种积温 (或温度和 )类型中 ,以有效积温 (或温度和 )与棉花产量指数的相关性最佳 ,气候积温 (或温度和 )和活动积温 (或温度和 )极为接近。(3 )≥ 2 8.0℃的日最高温度的三种温度和是评价阿克苏棉区棉花生长热量状况对产量影响程度大小的很好的热量指标。 (4)所选的 6个积温 (或温度和 )因子与棉花产量指数的相关具有较好的稳定可靠性     

论退耕还林地替代天然林地种植草果的生态学意义

草果对自然环境务件有特殊要求,在天然林下种植草果有不少弊端。退耕还林工程使替代天然林种植草果成为可能,有利于保护原有天然林和巩固退耕还林成果。     

灌区节水改造综合评估指标权重确定方法研究

为了科学合理地确定指标权重,从而对灌区节水改造项目综合后评估作出理想的评价结果,在阐述基于离差平方和的综合赋权法和基于博弈论的综合赋权法原理的基础上,将这两种综合集成赋权法分别应用于关中洛惠渠等4个灌区的节水改造项目综合评估,发现两种综合赋权法得出的权重向量大小排序结果并不相同,但根据两种权重值计算的多个灌区综合评价值大小排序结果相同.对这一现象进行的进一步分析表明,两种综合集成赋权法的指导思想虽然不同,但其原理都是科学合理的.在确定灌区评价指标权重时,若为了对各灌区评价对象充分拉开档次,可选用基于离差平方和的综合赋权法,当希望尽可能保留各主客观权重值的信息时,可选用基于博弈论的综合赋权法.     

非平稳高斯序列最大值与部分和联合的几乎处处收敛

假定(Xn,n≥1)为标准化非平稳高斯序列,Mn=max{Xi,1≤i≤n},Sn=∑ni=1Xi分别为对应的最大值与部分和,在协方差函数满足适当条件下,得到了最大值与部分和联合的几乎处处收敛定理.     

确定监测区域建立森林郁闭度估测方程最优样地的研究

在对监测区域样地进行合理分类的基础上,需抽取一定数量能代表监测区域森林资源分布状况的最优样地,根据所抽样地对应影响郁闭度估测的主要遥感和GIS因子,可建立以样地为单位的郁闭度估测方程. 最优样地是指能代表监测区域森林种类及分布状况的样地,它包含样地的数量及样地的代表性两个方面的含义. 如何抽取最优样地属多目标优化问题. 该文在理论描述特定监测区域最优样地数学模型的基础上,通过实例进行了系统研究与分析,有效解决了最优样地的抽样问题,所得结果可用于指导生产实践.