Data selection using support vector regression

Geophysical data sets are growing at an ever-increasing rate, requiring computationally efficient data selection(thinning)methods to preserve essential information. Satellites, such as Wind Sat, provide large data sets for assessing the accuracy and computational efficiency of data selection techniques. A new data thinning technique, based on support vector regression(SVR), is developed and tested. To manage large on-line satellite data streams, observations from Wind Sat are formed into subsets by Voronoi tessellation and then each is thinned by SVR(TSVR). Three experiments are performed. The first confirms the viability of TSVR for a relatively small sample, comparing it to several commonly used data thinning methods(random selection, averaging and Barnes filtering), producing a 10% thinning rate(90% data reduction), low mean absolute errors(MAE) and large correlations with the original data. A second experiment, using a larger dataset, shows TSVR retrievals with MAE < 1 m s-1and correlations 0.98. TSVR was an order of magnitude faster than the commonly used thinning methods. A third experiment applies a two-stage pipeline to TSVR, to accommodate online data. The pipeline subsets reconstruct the wind field with the same accuracy as the second experiment, is an order of magnitude faster than the nonpipeline TSVR. Therefore, pipeline TSVR is two orders of magnitude faster than commonly used thinning methods that ingest the entire data set. This study demonstrates that TSVR pipeline thinning is an accurate and computationally efficient alternative to commonly used data selection techniques.     

Growth performance of the clam, Meretrix meretrix, breedingselection populations cultured in different conditions

A breeding-selection program for the clam,Meretrix meretrix,was conducted since 2004.Two of the selection populations were generated with the shell color pattern as an additional selection criterion and named as SP(purple stripes)population and SB(black dots)population.The third-generation SP and SB populations(08G3SP and 08G3SB,respectively)were cultured at two commercial clam farms and a nursery pond and their shell lengths were compared.08G3SB clams had significantly larger sizes than 08G3SP clams at commercial clam farms(p<0.05),yet 08G3SB individuals were significantly smaller than 08G3SP individuals at the nursery pond(p<0.05).Then,we examined the growth of the fourth-generation SP and SB populations(10G4SP and 10G4SB,respectively)at a commercial farm,and found that the shell lengths of the 10G4SB clams increased at a significantly higher growth rate than the 10G4SP clams(p<0.05)from May to September,when the water temperature was between 24.2–27.5 C,while 10G4SB lost the growth advantage in the other months.These results indicated that SP and SB populations responded differently to environmental factors,so it is beneficial for the clam farmers to select a suitable population according to the culture environment.Furthermore,a diallel mating of the SB and SP populations was designed to investigate whether their hybrid population would show heterosis.However,the heterosis was not shown in this study,which might result from the slight genetic divergence between SB and SP populations.     

玉树地震地表破裂调查与灾后重建避让选址研究

2010年4月14日07时49分40.7秒,青海省玉树县发生Ms7.1级地震。通过现场调查和分析,玉树地震的发震断裂是甘孜—玉树活动断裂的玉树段,地震产生的地表破裂长度为23km,走向北西西—北西,总体表现为左旋走滑特征,可进一步分为3段:西段呈连续延伸的左旋走滑破裂,在隆宝镇的"郭央烟宋多"附近(坐标:北纬33°03′11″、东经96°51′26″)最大水平位移达1.75m,可定为宏观震中;中段主要位于县城南侧,由多条右阶斜列的破裂组成;东段表现为逆冲兼左旋走滑。玉树地震除了造成大量房屋破坏外,次生地质灾害主要包括:地震砂土液化及其引起的公路变形、地震滑坡、地震诱发水渠破坏及其链生土质滑坡和泥石流等。通过对内外动力地质灾害的综合分析,初步提出了玉树地震灾区灾后重建避让选址的建议。     

土壤铅含量高光谱遥感反演中波段选择方法研究

利用高光谱遥感数据进行了南京郊外土壤重金属元素铅的含量反演,由于高光谱数据波段众多,波段选择或变换至关重要。比较了基于次贪婪的前向选择模型的最小角度拟合和基于遗传算法进行波段选择的最小二乘和偏最小二乘拟合,结果发现基于遗传算法的偏最小二乘反演结果优于全波段的偏最小二乘,表明波段选择在高光谱反演重金属中是有益的。尽管采取了波段选择后的各方法在反演时均能达到70%以上的训练精度,但因遗传算法搜索的解空间范围更宽广,使得基于遗传算法的偏最小二乘优于前向选择模型的最小角度拟合。最后还比较了基于遗传算法的普通最小二乘和偏最小二乘拟合,结果表明偏最小二乘更优,因此在高光谱反演重金属含量当中,偏最小二乘精度较高,而在波段选择方法中,遗传算法更优。     

勘察设计企业并购模式的选择

随着我国社会主义市场经济体制改革的不断深化,加入WTO后与国际市场融入度的不断加强,企业生存与发展的环境日益呈现出市场一体化、需求多元化、竞争激烈化的特征,为了拓展企业的业务范围、扩大企业规模以及提高市场的竞争实力以应对日益激烈的市场竞争,勘察设计企业面对严峻的市场和竞争环境,结合企业自身的资源、环境和并购目的选择合适的并购模式,通过借助并购等资本运营手段,将企业做大做强,增强其综合实力,有效参与全球竞争。     

安徽省大别山区扶贫开发的产业选择和发展对策

大别山区扶贫开发和功能区生态环境保护是目前人们所广泛关注的重要课题。合理发展产业经济是实现国家重要生态功能区环境保护和脱贫的重要途径。结合大别山区贫困县的生态功能定位和产业基础,构建生态产业集中发展区和生态经济综合配套区,以六安市为例选择优势主导产业、各贫困县的生态产业,且提出生态补偿式扶贫开发中产业具体发展对策。     

Evaluation of Three Harvest Control Rules for Bigeye Tuna（Thunnus obesus） Fisheries in the Indian Ocean

The stock of Bigeye tuna（Thunnus obesus） in the Indian Ocean supports an important international fishery and is considered to be fully exploited. The responsible management agency, the Indian Ocean Tuna Commission（IOTC）, does not have an explicit management decision-making framework in place to prevent over-fishing. In this study, we evaluated three harvest control rules, i） constant fishing mortality（CF）, from 0.2 to 0.6, ii） constant catch（CC）, from 60000 to 140000 t, and iii） constant escapement（CE）, from 0.3 to 0.7. The population dynamics simulated by the operating model was based on the most recent stock assessment using Stock Synthesis version Ⅲ（SS3）. Three simulation scenarios（low, medium and high productivity） were designed to cover possible uncertainty in the stock assessment and biological parameters. Performances of three harvest control rules were compared on the basis of three management objectives（over 3, 10 and 25 years）： i） the probability of maintaining spawning stock biomass above a level that can sustain maximum sustainable yield（MSY） on average, ii） the probability of achieving average catches between 0.8 MSY and 1.0 MSY, and iii） inter-annual variability in catches. The constant escapement strategy（CE=0.5）, constant fishing mortality strategy（F=0.4） and constant catch（CC=80000） were the most rational among the respective management scenarios. It is concluded that the short-term annual catch is suggested at 80000 t, and the potential total allowable catch for a stable yield could be set at 120000 t once the stock had recovered successfully. All the strategies considered in this study to achieve a ‘tolerable＇ balance between resource conservation and utilization have been based around the management objectives of the IOTC.     

Evaluation of three harvest control rules for Bigeye Tuna (Thunnus obesus) fisheries in the Indian Ocean

The stock of Bigeye tuna(Thunnus obesus) in the Indian Ocean supports an important international fishery and is considered to be fully exploited. The responsible management agency, the Indian Ocean Tuna Commission(IOTC), does not have an explicit management decision-making framework in place to prevent over-fishing. In this study, we evaluated three harvest control rules, i) constant fishing mortality(CF), from 0.2 to 0.6, ii) constant catch(CC), from 60000 to 140000 t, and iii) constant escapement(CE), from 0.3 to 0.7. The population dynamics simulated by the operating model was based on the most recent stock assessment using Stock Synthesis version Ⅲ(SS3). Three simulation scenarios(low, medium and high productivity) were designed to cover possible uncertainty in the stock assessment and biological parameters. Performances of three harvest control rules were compared on the basis of three management objectives(over 3, 10 and 25 years): i) the probability of maintaining spawning stock biomass above a level that can sustain maximum sustainable yield(MSY) on average, ii) the probability of achieving average catches between 0.8 MSY and 1.0 MSY, and iii) inter-annual variability in catches. The constant escapement strategy(CE=0.5), constant fishing mortality strategy(F=0.4) and constant catch(CC=80000) were the most rational among the respective management scenarios. It is concluded that the short-term annual catch is suggested at 80000 t, and the potential total allowable catch for a stable yield could be set at 120000 t once the stock had recovered successfully. All the strategies considered in this study to achieve a ‘tolerable' balance between resource conservation and utilization have been based around the management objectives of the IOTC.     

基于特征选取及广义傅立叶分形的SAR图像海洋溢油检测算法

针对海上溢油SAR图像中油膜与类油膜的识别问题,提出了一种结合傅立叶分形与特征提取的检测算法。由于分形特征可以具有无穷多的细节,并在不同的研究尺度存在自仿射特性。这与油膜及类油膜表面的几何形貌特征非常吻合。该算法通过计算样本的傅立叶分形特征,组成油膜与类油膜的特征空间。然后,应用基于差分进化的特征选取方法将利于分类的重要特征值筛选出来。再利用重要特征值对原有样本进行分类。实验表明,经特征选取的分形特征向量能够以100%的准确率将两类样本区分开。该算法在选取重要特征的同时实现了对高维特征空间降维的目的,该思想可以应用于其他的基于高维特征的识别系统中,具有普遍的适用性。