Catch per unit effort standardization of the eastern Bering Sea walleye pollock (Theragra chalcogramma) fleet

A general linear model (GLM) was used to standardize catch per unit effort (CPUE) data for Alaska walleye pollock (Theragra chalcogramma) from the Bering Sea fleet for the years 1995–1999. Data were stratified temporally by year and season and spatially by area using either Alaska Department of Fish and Game (ADF&G) or National Marine Fisheries Service (NMFS) reporting areas. Four factors were used: vessel identification (ID) number, vessel speed, percentage of pollock by weight in the haul (a measure of targeting), and whether most of the haul took place before or after sunset. At least 29 combinations of main effects, quadratic covariates, and interactions were tested for each year/area/season stratum. GLM models explained from 31 to 48% of the total sums of squares. Vessel identification number was included in all models and explained the most variability. Of the remaining factors, the square of the percentage of pollock in the haul was included in most models, following an F-test to determine parsimony. Analysis of the vessel identification number coefficients indicated that larger vessels tended to have higher CPUEs; and that this relationship differed between dedicated catcher vessels and offshore catcher processors. Coefficient estimates and response surfaces generally indicated increased CPUEs with the percentage of pollock in the haul and showed mixed results with vessel speed. The vessel identification number incorporated most vessel characteristics, leaving vessel speed primarily as a fitting variable with less biological meaning. The year/area/season stratification procedure was found to be necessary due to the unbalanced design, which otherwise would have factor levels with no data in a large combined model. In addition, the stratification procedure reduced the variability in CPUE substantially.     

A generalized linear mixed model analysis of a multi-vessel fishery resource survey

A generalized linear mixed model (GLMM) that treats year and spatial cell as fixed effects while treating vessel as a random effect is used to examine fishing power among chartered industry-based vessels and a research trawler, the FRV Miller Freeman, for bottom trawl surveys on the upper continental slope of U.S. West coast. A Bernoulli distribution is used to model the probability of a non-zero haul and the gamma distribution to model the non-zero catch rates of four groundfish species. The use of vessel as a random effect allows the data for the various vessels to be combined and a single continuous time-series of biomass indices to be developed for stock assessment purposes. The GLMMs fit the data reasonably well. Among the different models examined, the GLMM incorporating a random vessel × year effect had the smallest ΔAIC and was thus chosen as the best model. Also, estimated random effects coefficients associated with the industry-based vessels and the FRV Miller Freeman for each year suggests that these vessels can be assumed to be from a common random effects distribution. These results suggest that combining data from the chartered industry-based vessels and from the research trawler may be appropriate to develop indices of abundance for stock assessment purposes. Finally, an evaluation of variances associated with abundance indices from the different models indicate that analyzing these data as a fixed effect GLM may underestimate the level of variability due to ignoring the grouped nature of tows within vessels. As such, use of a mixed model approach with vessel as a random effect is a reasonable approach to developing abundance indices and their variances.     

Remedies for pseudoreplication

Pseudoreplication is the failure of a statistical analysis to properly incorporate the true structure of randomness present in the data. It has been well documented and studied in the ecological literature but has received little attention in the fisheries literature. Avoiding pseudoreplication in analyses of fisheries data can be difficult due to the complexity of the statistical procedures required. However, recent developments in statistical methodology are decreasing the extent to which pseudoreplication has to be tolerated. Seven examples are given here, beginning with simple design-based remedies and progressing to more challenging examples including the model-based remedies of mixed-effects modelling, generalized linear mixed models, state-space models, and geostatistics.     

中文农业搜索引擎字符编码识别

针对农业网页中汉字编码标识混乱的情况,提出了一种综合运用编码规则和网页文本特征的字符编码识别模型。利用卡方检验算法,结合最小二乘多元线性回归方法,得到了基于网页文本特征的字符识别模型。实验结果显示,在适当的选取阈值（r ＝1,阈值＝属于某一编码的字符数/网页总字符数）和文本特征数（≥65）的基础上,模型准确率达到100％,且结果稳定。     

高斯烟雨扩散模型在空气中PM2.5实际问题的应用

通过建立高斯烟雨扩散模型并采用多元线性回归等数学方法,进一步探索产生雾霾的主要因素PM2.5的形成、扩散过程以及与风速之间的关系。以武汉市和西安市2013年AQI监测数据为基础,得到空气中PM2.5与其他污染物之间的相关性。预测了西安市某一区域PM2.5浓度骤增并持续数小时的情况下,污染严重和相对安全的区域。为研究空气中PM2.5成因、监测和治理提供了一定的理论依据。     

高割龄橡胶树 PR107高产群体性状指标量化分析

为探索高割龄橡胶树高产栽培技术,采用相关、通径及线性回归分析,对海南省大丰农场23块21、23 a高割龄PR107胶园及其9个高产群体主要经济性状与产量进行统计分析。研究表明,在各主要经济性状中,对产量影响最大的是每公顷有效割株数,其次是年均割次株产干胶量,再之是年割胶刀数。生产中要以每公顷有效割株数为中心来合理安排割胶强度,对年割胶刀数进行合理的促控,保持适宜的高产群体,开割中后期注意刺激剂的施用,以既增加年均每割次株产干胶量,又不影响在割率才有利于高产。运用微机对三性状产量效应数学模型进行仿真模拟,得出高割龄PR107年产量≥1500 kg/hm2的栽培措施为：年均每割次株产干胶量79．82～85．26 g,年割胶刀数60．84～61．19,每公顷有效割株数393．52～422．37株。三因子取平均值时年单产预估为1988．10 kg/hm2。     

线性旋转变压器研究

文章阐述了线性旋转变压器的工作原理,并对其基本结构、种类做了介绍,说明了线性旋转变压器的实用性。     

我国耕地保护绩效定量研究

根据修正后的耕地面积数据(1978-2012年)和耕地保护相关的经济社会因素,构建耕地数量变化的多元线性回归模型,在对耕地保护政策发展历程、现有体系分析的基础上,依据该模型对我国35年来耕地保护政策的绩效进行了分析.结果显示,我国耕地保护政策绩效显著,35年共遏制耕地减少面积250.96万hm2,遏制耕地减少贡献率达17.35％;耕地保护政策绩效有着明显的时间差异性,不同阶段政策遏制耕地减少的贡献率差异明显,但在耕地保护政策发展的4个阶段,政策遏制耕地减少的贡献率是逐渐上升的,说明耕地保护政策在不断的发展,耕地保护政策的绩效在逐渐发挥.在2003-2012年阶段,耕地减少面积逐渐稳定并降至多年来的最低点,同时政策遏制耕地减少的贡献率达到峰值,表明耕地数量保护的目标绩效显著,达到了顶峰.建议未来的耕地保护工作需要突出耕地的质量建设和生态建设两个着重点.