运用多时相直接分类法对土地利用进行遥感动态监测的研究

采用多时相直接分类法对贵州省 (原 )安顺市从1987～1995年间的土地利用变化状况进行了遥感动态监测。与以往方法相比 ,该方法不仅可以直接获得变化类型及其数量 ,而且可以避免分类后比较法所容易出现的逻辑性变化探测错误。对不同分类波段组合的试验表明 ,经过差值、比值处理的前3种波段组合 ,具有较差的分类效果 (总体精度只有30 %～40 % ) ,而那些能较好地保留原始信息的波段组合 ,则具有相对较好的分类效果 (总体精度超过70 % )。另外 ,分类效果还明显地受到所采用的辐射校正方法的影响 ,经过暗组亮组辐射校正的影像要明显比只经过简单辐射校正(直方图法 )的影像具有更好的分类效果 ,总体精度提高了16.6 %。该方法的应用结果表明 ,研究区在8年间经历了一些明显的土地利用变化 ,例如城镇扩展、耕地减少、菜地变迁等。     

水中几种有机污染物的SPE-HPLC联用测定技术

酚类及硝基苯是水中一类重要的优先控制污染物。通过对固相萃取 (SPE)技术与高效液相色谱 (HPLC)联用技术的研究 ,得出对于含酚类及硝基苯的水样采用 2 0 0mg的键合硅胶C1 8柱在水样pH值为 3 .0和过柱水样为 40mL时可得到较好的萃取及测定结果 ,从而提出一种对水中酚类物质及硝基苯测定具有灵敏度高、准确性好的分析方法。     

Ground gleaning in horseshoe bats: comparative evidence from<Emphasis Type="Italic"> Rhinolophus blasii</Emphasis>,<Emphasis Type="Italic"> R. euryale</Emphasis> and<Emphasis Type="Italic"> R. mehelyi</Emphasis>

The 71 species of horseshoe bat (genus Rhinolophus) use echolocation calls with long constant-frequency (CF) components to detect and localize fluttering insects which they seize in aerial captures or glean from foliage. Here we describe ground-gleaning as an additional prey-capture strategy for horseshoe bats. This study presents the first record and experimental evidence for ground-gleaning in the little-studied Blasius horseshoe bat (Rhinolophus blasii). The gleaning bouts in a flight tent included landing, quadrupedal walking and take-off from the ground. The bats emitted echolocation calls continuously during all phases of prey capture. Both spontaneously and in a choice experiment, all six individuals attacked only fluttering insects and never motionless prey. These data suggest that R. blasii performs ground-gleaning largely by relying on the same prey-detection strategy and echolocation behaviour that it and other horseshoe bats use for aerial hawking.We also studied the Mediterranean horseshoe bat (R. euryale) in the flight tent. All four individuals never gleaned prey from the ground, though they appeared to be well able to detect fluttering moths on the ground. It is not known yet whether ground-gleaning plays a role in Mehelys horseshoe bat (R. mehelyi). In a performance test, we measured the ability of these three European species of middle-sized horseshoe bats (R. euryale, R. mehelyi and R. blasii) to take-off from the ground. All were able to take flight even in a confined space; i.e. the willingness to ground-glean in R. blasii is not related to a superior take-off performance. In contrast to ground-gleaning bats of other phylogenetic lineages, R. blasii appears not to be a specialist, but rather shows a remarkable behavioural flexibility in prey-capture strategies and abilities. We suggest that the key innovation of CF echolocation paired with behavioural flexibility in foraging strategies might explain the evolutionary success of Rhinolophus as the second largest genus of bat.Communicated by T. Czeschlik     

Maximum likelihood method for parameter estimation in non-linear models with below detection data

The maximum likelihood (ML) method for regression analyzes of censored data (below detection limit) for nonlinear models is presented. The proposed ML method has been translated into an equivalent least squares method (ML-LS). A two stage iterative algorithm is proposed to estimate statistical parameters from the derived least squares translation. The developed algorithm is applied to a nonlinear model for prediction of ambient air CO concentration in terms of concentrations of respirable particulate matter (RSPM) and NO₂. It has been shown that if censored data are ignored or estimated through simplifications such as (i) censored data are equal to detection limit, (ii) censored data are half of the difference between detection limit and lower limit (e.g., zero or background level) or (iii) censored data are equal to lower limit, this can cause significant bias in estimated parameters. The developed ML-LS method provided better estimates of parameters than any of the simplifications in censored data.     

人的失误本质及检测的研究

对人的失误本质、定义及特点作了系统的分析研究。针对人的失误产生条件、原因及类型，提出了检测失误与控制失误的技术     

Status report on flood warning systems in the United States

One of the major changes in flash-flood mitigation in the past decade is the number of communities that have implemented warning systems. The authors conducted a survey of 18 early-warning systems in the United States developed by communities or regions to provide protection against flash floods or dam failures. Problems revealed by the study included the following: equipment malfunctions, inadequate maintenance funding, inconsistent levels of protection and expenditure, inconsistent levels of expectations and formalization, varying levels of local commitment to the systems, underemphasis on response capability, and a tendency to over-rely on warning systems. The study also revealed some unanticipated benefits experienced by the survey communities: the warning systems serve as valuable data collection tools, a great deal of interagency cooperation has been demonstrated, and warning systems offer increased alternatives to structural modification projects. The interjurisdictional nature of drainage basins, the evolving roles of the various federal agencies involved in flood mitigation, and the lack of governmental standards of operations for flood warning systems are issues that must be considered as communities make decisions regarding the adoption of warning systems. The record on these systems is too short for a precise assessment of how successful they are; however, results of the study indicate that if the goal of reducing loss of life and property from flooding is to be achieved, warning systems must be only one part of a comprehensive flood loss reduction program.     

加入WTO对我国检测业的影响及对策

本文就加入WTO后我国检测业所面临的形势进行了分析，并就我国检测工作现状、存在的问题和如何应对入世后的挑战提出了对策。     

基于多传感器融合的林火监测

为了提高近距离火灾监测的准确率,建立了基于Arduino平台的多传感器实时监测系统.此系统安装在移动机器人身上以探测火灾.在林火发生期间,会产生CO、C02明火火焰及其他产物,并引起周围环境温度的升高.因此,选择合适的传感器,检测出以上参数,就有可能据此判断实际环境是否有火.通过在Arduino上搭建火焰传感器、温度传感器、气体传感器和烟雾传感器,可以实时监测环境参数.在无火和有火环境中进行了多次试验,进行数据采集,得到了大量原始数据.无火环境的数据是在不同的天气条件下测得的;有火环境由试验火堆模拟得到.在模拟的过程中,进行人为操作以模拟不同的火情.如通过浇湿底部的可燃物模拟预热阶段,试验数据因此更有代表性.数据分析表明,单个传感器的输出值波动大,且在有火环境和无火环境中的输出值有重叠.因此,用单一传感器来检测火灾的准确率很低.而同时分析3个传感器的输出值时,其输出值随所检测火堆的不同呈现出一致的变化规律.最后,利用神经网络进行多传感器数据融合.涉及5个输入变量,由神经网络实现对多变量的非线性问题进行模式识别.将前述试验所得数据划分为训练数据和测试数据,两类数据均包含一定比例的有火样本和无火样本.用训练数据对BP神经网络进行训练,可得到林火识别模型.用测试数据检验模型,结果表明,该BP神经网络对试验火的识别准确率为98.625％.     

市售贝类麻痹性贝毒快速生化检测方法探究

为了研究快速检测海产品中低浓度麻痹性贝毒(paralytic shellfish poisoning,PSP)的生化方法,在传统小白鼠法的基础上,使用浓度(以STXeq计)为1和0.2μg/kg的市售双壳贝类PSP提取液对小鼠进行腹腔注射,分别研究暴露15、60、120min后小鼠血液中ACh、AChE、NO和NOS的变化情况.结果发现,低浓度(0.2μg/kg)、短时间(15min)条件下仅ACh含量发生显著变化[与对照组相比p0.05,含量(以蛋白计)为(141.2±14.8)μg/mg];而NO与NOS却在长时间(120min)暴露后才发生改变[与对照组相比p0.05,分别为(68.7±3.8)μmol/g及(40.1±4.9)U/mg].在高浓度下,3个时间点的ACh含量也均发生显著变化.结果表明,在低浓度及短时间条件下反映PSP效应的指标仅是ACh,从而可以尝试选择其作为快速检测PSP的生化指标.