随机区组设计的多元分析

本研究首次把经典随机区组设计的统计分析方法由单变元推广到多变元,使得试验的多性状记录能综合在一起进行分析,以得出考虑了性状间相关性在内的统计结论,并结合橡胶幼苗肥料试验实例给出了MINITAB统计程序。     

基于业务流的混合模式数据监视平台的设计与实现

研究了具有初始财富的投资者如何最大化终端资产和消费的期望效用,首先通过交易费用函数建立带交易费的连续时间投资与消费模型,然后运用鞅分析和对偶理论证明了：在有效市场中,如果投资者积极交易,则只会降低终端财富的期望值,并得到了最优投资消费组合过程和终端资产.     

Cephalopod Beaks from the Stomach of a Jackass Penguin

In order to determine the potential of the earthworm Eudrilus eugeniae as waste processor and protein producer, the life history of this species was studied. The development, growth and reproduction of E. eugeniae were studied. Cattle dung was used as substrate with a moisture content of 70–80% and a temperature of 25°C. Data were gathered over a period of 300 days. It was found that cocoon production started within 24 h after copulation and can be sustained for at least 300 days. Cocoons are produced at an average rate of 1,65 cocoons per worm per day. The mean incubation period of cocoons is 16,6 days with a hatching success of 84% and 2,7 hatchlings per cocoon that hatched. Sexual maturity is attained by the offspring within 40 to 50 days after hatching. E. eugeniae is compared to other vermicomposting species.     

Abundance and home ranges of feral cats in an urban conservancy where there is supplemental feeding: a case study from South Africa

There is much debate surrounding the impact of feral cats (Felis catus) on wildlife. Conservancies areusually areas where indigenous flora and fauna are protected and aliens excluded or managed. The University of KwaZulu-Natal’s Howard College campus (HCC) is an urban conservancy containing feral cats that are presently not managed, and little is known about their ecology and behaviour. Consequently a feral cat population census was conducted, and their home range investigated. Estimates of the overall campus feral cat population numbers ranged between 23.4–40.0 cats/km² with a minimum of 55 identified as resident. They were not randomly distributed in the study area, with spacing patterns being related to resource availability. Home range area and core distribution of eight radio-collared cats were determined over 13 months. Total home range areas were relatively small, with considerable overlap between them. Home ranges were clustered in areas with permanent feeding stations and these were also within the cats’ core ranges. Supplemental food resources appear to have a major influence on numbers, home and core range area, and behavior of cats. It is clear that cat densities grow to high levels with reliable and abundant food supply and only ad hoc sterilization. This has implications for their management in the HCC urban conservancy.     

Validation of an efficient visual method for estimating leaf area index in clonal Eucalyptus plantations

Leaf area index (LAI) is a key ecophysiological parameter in forest stands because it characterises the interface between atmospheric processes and plant physiology. Several indirect methods for estimating LAI have been developed. However, these methods have limitations that can affect the estimates. This study aimed to evaluate the accuracy and applicability of a visual method for estimating LAI in clonal Eucalyptus grandis × E. urophylla plantations and to compare it with hemispherical photography, ceptometer and LAI-2000^® estimates. Destructive sampling for direct determination of the actual LAI was performed in 22 plots at two geographical locations in Brazil. Actual LAI values were then used to develop a field guide with photographic images representing an LAI range of 1.0–5.0 m² m^?2 (leaf area/ground area). The visual LAI estimation guide was evaluated with 17 observers in the field. The average difference between actual LAI and visual LAI estimation was 12% and the absolute difference between the two methods was less than or equal to 0.5 m² m^?2 in 77% of plots. Pearson’s correlation coefficients were high between actual LAI and hemispherical photographs (0.8), visual estimation (0.93) and LAI-2000^® (0.99) and low for the ceptometer (0.18). However, absolute values differed among methods, with the average difference between the actual and estimated LAI of [12]% for visual estimation, 28% for the LAI-2000^®, 37% for the ceptometer and ?43% for hemispherical photographs. The LAI-2000^® and ceptometer overestimated LAI in all plots, whereas hemispherical photographs underestimated the values in all measurements, showing that these methods need calibration to be used. No differences were observed between actual LAI and visual estimates across stand ages of 2–8 years and LAI of 1.5–5.3 m² m^?2 (P > 0.05). The results show that visual estimation of LAI in Eucalyptus stands is a practical method that is unaffected by atmospheric characteristics and can be used on an operational scale.     

基于NLPCA-RBF神经网络的番茄蒸散量预测

蒸散量(ET)是水文循环中的重要组成部分。精确的ET预测在水资源管理和灌溉系统设计等方面的研究是十分必要的。利用非线性主成分分析法(NLPCA)和径向基(RBF)神经网络组成的模型(NLPCA-RBF)对番茄蒸散量进行估算。在既保证ET影响因素信息完整,又可消除影响因素之间相关性的前提下,利用NLPCA将影响ET的7个气象因素简化为3个综合成分,并以此为网络训练的输入数据,根据实测的蒸散量作为网络输出建立了RBF神经网络,并且经非训练样本点数据检验。结果表明,与传统RBF网络模型较,NLPCA-RBF网络预测模型能够更好的反应影响因子与蒸散量之间的关系,取得更为精确的结果。     

利用非参数方法估计山东半岛南部海域秋季游泳生物群落物种丰富度

为减少渔业资源物种丰富度观测的偏差,以山东半岛南部海域为例,根据2016年10月对该海域进行的底拖网调查数据,运用Chao、一阶刀切(Jackknife1)、二阶刀切(Jackknife2)和自助法(Bootstrap)4种非参数方法估算了该海域鱼类群落物种丰富度,并探讨了影响非参数方法估算的因素及非参数方法的应用。结果表明,4种非参数方法估算鱼类群落物种数目呈现出Jackknife2ChaoJackknife1Bootstrap,依次为265种、248种、235种、206种;不同估算方法的标准差范围有较大差别,Chao最大,Jackknife居中,Bootstrap最小。通过MDS多维标度排序将该海域划分两个群落,两群落物种数有显著差异,群落Ⅰ物种数目估计值为250种、238种、219种、191种,群落Ⅱ物种数目为86种、100种,4种非参数估算方法在不同群落表现基本一致。综合比较不同群落的物种估算结果,认为采样强度是影响4种非参数估计方法表现优劣的主要因素,采样充足是使用非参数估计方法的前提。本研究为不同采样努力量下物种丰富度的比较提供了方法参考和比较标准。     

湖北省畜牧业温室气体排放潜力

根据湖北省2007-2014年畜禽饲养量,按照《省级温室气体清单编制指南(试行)》要求,评估了湖北省2007-2014年畜禽养殖过程中的温室气体(GHG)排放潜力,并比较了2010年湖北省各地区的温室气体排放潜力以及各畜禽肠道甲烷(CH_4)、粪便CH_4、粪便氧化亚氮(N_2O)的排放状况。结果表明:(1)2007-2014年湖北省温室气体排放潜力总体呈现上升趋势,2014年达1 535.01万t CO2-eq,增幅11.50%;(2)2010年湖北省襄阳、孝感、黄冈和恩施的畜牧业温室气体排放潜力最大,占全省的58.81%;(3)非奶牛、水牛对肠道CH_4排放的贡献率最大,分别为43%、31%;猪是粪便CH_4和N_2O排放的主要来源,分别占粪便CH_4和N_2O排放潜力的83%和39%。因此,湖北省各地区应在保证畜牧业持续发展的同时,积极采取温室气体减排措施;针对不同畜禽种类、不同地理区域,应当有的放矢,因地制宜。     

Stand-scale transpiration estimates in a Moso bamboo forest: (I) Applicability of sap flux measurements

The applicability of sap flux (F_d) measurements to bamboo forests has not been studied. This study was undertaken to establish an optimal and effective design for stand-scale transpiration (E) estimates in a Moso bamboo forest. To this aim, we validated F_d measurements in Moso bamboos in a cut bamboo experiment. In addition, we analyzed how sample sizes affect the reliability of E estimates calculated from F_d and conducting culm area (A_{S_b}). In the cut bamboo experiments, we found that F_d measurement using a 10 mm probe was a valid means of determining the water-use behavior of a Moso bamboo, although a specific correction was needed. Furthermore, we calculated E from stand A_{S_b} (A_{S_stand}) and mean stand F_d (J_S). Employing Monte Carlo analysis, we examined potential errors associated with sample size in E, A_{S_stand}, and J_S using an original dataset with A_{S_b} and F_d measured for 40 and 16 individuals, respectively. Consequently, we determined the optimal sample size for both A_{S_stand} and J_S estimates as 11. The optimal sample sizes for J_S were almost the same under different vapor pressure deficit and soil moisture conditions. The optimal sample size for J_S at the study site was less than that of a coniferous plantation in the same region probably owing to small individual-to-individual variations in sap flux in the Moso bamboo forest. Our study concludes that sap flux measurements are an applicable technique for assessing water use in Moso bamboo forests.     

我国农作物秸秆转化生物能源潜力评估

农作物秸秆是一种价格低廉的可再生资源,燃料化是其资源化利用的重要途径。以农作物经济产量为依据,对我国各类农作物秸秆的产量和可收集利用量进行估算,进而评估其生物转化甲烷和燃料乙醇的潜力。2014年我国农作物秸秆产量达72 836.9万t,其中可收集利用量达60 986.1万t。玉米秸秆、水稻秸秆和小麦秸秆为前三大秸秆资源。河南省、黑龙江省和山东省的秸秆资源是最为丰富。以热值估算,仅玉米秸秆、水稻秸秆和小麦秸秆三大秸秆可折合标准煤量20 177.5万t。基于微生物发酵工艺估算,我国主要农作物秸秆可生产甲烷1 286.8亿m~3,或可生产燃料乙醇1 823.5亿升。我国农作物秸秆的可收集利用量巨大,以其为原料生产生物燃料具有广泛的开发前景。