Physical drivers of phytoplankton production in the southern Benguela upwelling system

Investigations of primary production (PP) were undertaken in the southern Benguela ecosystem during two research surveys in October 2006 and May 2007. Significant differences in environmental conditions, as well as biomass and PP, were observed between October and May. During October, integrated biomass and PP were significantly higher, ranging from 20.43 to 355.01 mg m⁻², and 0.71 to 6.98 g C m⁻² d⁻¹, respectively, than in May, where the range was 47.92–141.79 mg m⁻², and 0.70–3.35 g C m⁻² d⁻¹, respectively. Distribution patterns indicated low biomass and PP in newly upwelled water along the coast, higher biomass and PP in the mid-shelf region, while lower values were observed at and beyond the shelf edge. Latitudinal variations showed consistently higher biomass and PP in the St. Helena Bay region compared to biomass and PP south of Cape Town. During both surveys, phytoplankton communities were comprised primarily of diatoms and small flagellates, with no significant differences. Phytoplankton adaptation to environmental variability was characterised by increased P_m^B and E_k under elevated temperatures and irradiance, while no clear relationships were evident for α^B. Generalised Additive Models (GAMs) showed that photosynthetic parameters were all significant predictors of photosynthesis rates (P_z), with P_m^B being the most important, accounting for 36.97% of the deviance in P_z. However, biomass levels and environmental conditions exerted a much greater influence on P_z, with irradiance explaining the largest proportion (68.24%) of the deviance. Multiple predictor GAMs revealed that 96.26% of the deviance in P_z could be explained by a model which included nitrate, chlorophyll a, and irradiance.     

微机GIS系统软件设计的原理与方法

本文介绍了北京大学遥感技术应用研究所(国家遥感中心技术培训部)于1988年4月推出的地理信息系统通用软件PURSIS的设计原理与方法。由于它是在IBMPC/XT,AT及系列兼容机上研制的,因此便于推广使用。为了克服微机在速度和容量上的种种限制,PURSIS系统采用了一些较复杂的技术手段,使得与其它系统相比,具有如下明显特点:(1)利用软件技术,为图形的输入和编辑创立了一个良好的交互式工作环境,使用户能方便地输入和修改各种地图。(2)采用压缩编码技术存贮图形和图像数据,从而大大减轻了微机系统的空间负担。(3)设计了高效率的数据格式转换算法,使矢量数据能快速转换为网格数据。(4)设计了COLOR 136软件,将廉价的COLOR-400卡支持的16种颜色扩充为136种颜色,成功地解决了专题图件的显示问题。(5)解决了图形数据和属性数据的联合操作问题,使用户能在图的逻辑概念上进行多种操作。此外,数字地形模型和其它数据在操作时的联系甚为紧密。(6)提供了一些常用的应用工具,如专家权重、叠加分类、专家命题等模型。使用户在这些模型支持下解决一些各自的专业问题。(7)解决了FORTRAN语言、编译DBASE Ⅲ和8086宏汇编语言之间的相互调用问题,使系统设计灵活多变,并能充分利用微机的系统功能。     

电子政务综合分析与评价平台建设初探

本文立足于政府部门决策分析的科学化和智能化,突破电子政务局限于办公自动化和政府上网等操作层面的研究,提出了在数据库管理系统及数据钻取与挖掘工具、模型库管理系统、知识库管理系统技术支撑下建立起来的电子政务综合分析与评价平台的框架结构和技术体系,该平台综合了GIS的空间分析与数据挖掘功能、空间统计功能、经典统计分析功能以及数学分析功能,能够对政务信息数据库进行多方位、深层次的数据分析与综合评价,最终对政府提供科学决策依据。     

层状地壳模型地面地下记录地震波的理论地震图和频谱计算

本文首先发展了计算层状介质地面及地下记录地震波理论地震图的部分分离变量—有限差分计算方法,其最主要的环节是引进了吸收边界条件,使计算工作量大大减少,得以在实际工作中推广应用。其次,应用矩阵方法计算地面及地下记录的地震波的理论频谱。最后给出了用上述方法得到的华北地区两种不同的理论地壳模型情况下的理论地震图和频谱计算结果。应用这些理论结果,可根据实际地震观测资料对地壳介质模型进行分析、评判和优选。     

On modelling the effects of afforestation on acidification in heterogeneous catchments at different spatial and temporal scales

A modelling approach is presented for simulating and predicting future changes in streamwater Gran alkalinity throughout a large, heterogeneous river system. The methodology is based on integrating End Member Mixing Analysis (EMMA), the Model of Acidification of Groundwater in Catchments (MAGIC) and spatial data describing the catchment characteristics stored on a Geographical Information System (GIS). These are integrated within a Functional Unit Network (FUN) to predict the changes in Gran alkalinity resulting from possible future changes in atmospheric deposition and land use (low intensity afforestation) in the River Dee catchment, NE Scotland. Model results indicate that declining sulphate and constant nitrogen deposition, combined with low intensity Scots pine (Pinus sylvestris) afforestation are unlikely to contribute significantly to streamwater acidification.     

焉耆盆地与柴达木盆地侏罗系成烃及成藏时期的比较

焉耆盆地和柴达木盆地都发育侏罗系烃源岩,形成侏罗系油气系统。由于中 新生代两者构造运动强度的不同,导致两者油气形成时期、油气运移成藏时期以及油气分布层系不同。根据烃源岩热模拟结果和储层包裹体特征的研究,证明焉耆盆地侏罗系油气系统生烃时期主要发生在中 晚侏罗世,油气成藏时期主要发生在中 晚燕山期,为早期一期成藏;柴达木北缘则分别在早侏罗世晚期 晚侏罗世早期和早第三纪末 晚第三纪早期发生了 2次生油过程,因此存在 2期油气成藏,但以晚期为主。两盆地分别是中国西北侏罗系油气系统早期成藏和晚期成藏的典型代表。     

奇异值检验中的混合模型及EM算法

本文提出了奇异值检验中的一种新模型——混合模型,推导了简单样本和一般线性模型下的参数估计公式,并引入了一种有效的迭代计算方法——EM算法。     

Repeating earthquake activities associated with the Philippine Sea plate subduction in the Kanto district, central Japan: A new plate configuration revealed by interplate aseismic slips

We detect repeating earthquakes associated with the Philippine Sea plate subduction to reveal the plate configuration. In the Kanto district, we find 140 repeating earthquake groups with 428 events by waveform similarity analysis. Most repeating earthquakes in the eastern part of the Kanto district occur with a regular time interval. They have thrust-type focal mechanisms and are distributed near the upper surface of the Philippine Sea plate. These observations indicate that the repeating earthquakes there occur as a repetition of ruptures on the isolated patches distributed on the plate boundary owing to the concentration of stress caused by aseismic slips in the surrounding areas. This shows that the distributions of repeating earthquakes suggest the aseismic slips in the surrounding areas of small patches. We determine spatial distributions of repeating earthquakes in the eastern part of the Kanto district and find that they correspond to the upper boundary of the Philippine Sea plate, that is, the upper boundary of the oceanic crust layer of the Philippine Sea plate. The plate geometry around Choshi is newly constrained by repeating earthquake data and a rather flat geometry in the eastern part of the Kanto district is revealed. The obtained geometry suggests uplift of the Philippine Sea plate due to the collision with the Pacific plate beneath Choshi.Repeating earthquakes in the western part of the Kanto district have extremely shorter recurrence times, and their focal mechanisms are not of the thrust types. These repeating earthquakes are classified as “burst type” activity and likely to occur on the preexistent fault planes which are distributed around the “collision zone” between the Philippine Sea plate and the inland plate. The variation among the repeating earthquake activities in the Kanto district indicates that regular repetition of repeating earthquakes is possible only on the plate boundary with a smooth and simple geometry.     

Limiters in Third-Generation Wind Wave Models

Third-generation ocean wave models include a so-called limiter in the integration of the source terms to guarantee numerical stability at economical numerical time steps. The original limiter has previously been associated with the sensitivity of model results to the numerical time step. More recent limiters appear to remove this sensitivity by eliminating the numerical convergence from the resulting integration scheme. This is contrary to rudimentary numerical principles as well as the underlying philosophy of third-generation wave models. The present study investigates the effects of limiters and large model time steps using time-limited wave growth test. It is shown that the conventional limiter results in stable model results even if the numerical time step violates the time scales of wave growth. Contrary to common belief, its impact is not necessarily limited to the equilibrium range of the spectrum, and the limiter systematically enhances growth rates in the intermediate stages of wave growth. Particularly initial growth errors increase significantly with increasing maximum discrete spectral frequency f max . Relaxation of the limiter is shown to reduce initial growth errors, but does so at the expense of notable errors in the spectral shape. In the present paper the limiter was relaxed by introducing a new asymmetric limiter that retains full convergence. Initial results obtained with this limiter are similar to those of the advocated nonconvergent limiters. Although this limiter still needs rigorous testing and further development, its initial results suggest that there is no justification for using nonconvergent limiters.     

DATA ANALYSIS WITH MINIMAL ASSUMPTIONS

The computer has made it possible to scrutinize data rapidly by means of graphics. This should be doneprior to the application of any model to the data, since the model must be validated before using it asa means of analyzing the data. The procedure is illustrated in terms of two examples of real experimentaldata.