水环境模拟中变异函数的稳健性分析

文章将稳健统计学（RS）引入地质统计学（GS）,对廊坊市地下水现场采样数据中Cl^-运用影响系数法和估计邻域法识别并处理特异值,取得了较好的稳健效果。在此基础上,用Cressie-Hawkins法、中位调节法两种稳健统计学方法系统的对进行特异值处理前后的数据作了处理和分析。结果表明,变量Cl^-离子经ICM法处理特异值后再进行稳健计算方法中以MA法有较大的优势,而且获得稳健变异函数γR（h）及其估计量对于改善变异函数的结构性和理论模型的主要参数有明显效果。因此,稳健地质统计学的方法可以移植到水环境科学中应用,这对加强水环境空间变异性以及稳健变异函数的基础研究有着重要的应用价值和学术价值。     

常用玻璃量器衡量法K(t)值数学模型的建立与探讨

介绍常用玻璃量器的检定原理和方法,并详细叙述了修正系数K(t)的数学模型建立的全过程。全面分析了模型中所有参数的相关性,在此基础上对模型进行了优化,去除了模型中次要参数,大大简化了数学模型,同时,修正了JJG196-2006《常用玻璃量器检定规程》中忽略大气压力参数的影响,导致可能出现检定结论出错的问题,并给出了优化后的数学模型表达式,分别给出了大气压力在110～50KPa之间的钠钙玻璃和硼硅玻璃量器的K(t)值表。     

A derivation of the statistical characteristics of forest fires

The analysis of large data sets concerning fires in various forested areas of the world has pointed out that burned areas can often be described by different power-law distributions for small, medium and large fires and that a scaling law for the time intervals separating successive fires is fulfilled. The attempts of deriving such statistical laws from purely theoretical arguments have not been fully successful so far, most likely because important physical and/or biological factors controlling forest fires were not taken into account. By contrast, the two-layer spatially extended forest model we propose in this paper encapsulates the main characteristics of vegetational growth and fire ignition and propagation, and supports the empirically discovered statistical laws. Since the model is fully deterministic and spatially homogeneous, the emergence of the power and scaling laws does not seem to necessarily require meteorological randomness and geophysical heterogeneity, although these factors certainly amplify the chaoticity of the fires. Moreover, the analysis suggests that the existence of different power-laws for fires of various scale might be due to the two-layer structure of the forest which allows the formation of different kinds of fires, i.e. surface, crown, and mixed fires.     

Refuge‐Effect Hypothesis and the Demise of the Dodo

The Dodo was last sighted on the inshore island of Ile d'Ambre in 1662, nearly 25 years after the previous sighting on the mainland of Mauritius. It has been suggested that its survival on the inshore island is representative of the refuge effect. Understanding what constitutes significant persistence is fundamental to conservation. I tested the refuge‐effect hypothesis for the persistence of the Dodo (Raphus cucullatus) on an inshore island beyond that of the mainland population. For a location to be considered a refuge, most current definitions suggest that both spatial and temporal isolation from the cause of disturbance are required. These results suggest the island was not a refuge for the Dodo because the sighting in 1662 was not temporally isolated from that of the mainland sightings. Furthermore, with only approximately 350 m separating Ile d'Ambre from the mainland of Mauritius, it is unlikely this population of Dodos was spatially isolated. Hipótesis del Efecto Refugio y la Desaparición del Dodo     

Asymptotics in adaptive cluster sampling

In this article we consider asymptotic properties of the Horvitz-Thompson and Hansen-Hurwitz types of estimators under the adaptive cluster sampling variants obtained by selecting the initial sample by simple random sampling without replacement and by unequal probability sampling with replacement. We develop an asymptotic framework, which basically assumes that the number of units in the initial sample, as well as the number of units and networks in the population tend to infinity, but that the network sizes are bounded. Using this framework we prove that under each of the two variants of adaptive sampling above mentioned, both the Horvitz-Thompson and Hansen-Hurwitz types of estimators are design-consistent and asymptotically normally distributed. In addition we show that the ordinary estimators of their variances are also design-consistent estimators.     

Using hidden Markov chains and empirical Bayes change-point estimation for transect data

Consider a lattice of locations in one dimension at which data are observed. We model the data as a random hierarchical process. The hidden process is assumed to have a (prior) distribution that is derived from a two-state Markov chain. The states correspond to the mean values (high and low) of the observed data. Conditional on the states, the observations are modelled, for example, as independent Gaussian random variables with identical variances. In this model, there are four free parameters: the Gaussian variance, the high and low mean values, and the transition probability in the Markov chain. A parametric empirical Bayes approach requires estimation of these four parameters from the marginal (unconditional) distribution of the data and we use the EM-algorithm to do this. From the posterior of the hidden process, we use simulated annealing to find the maximum a posteriori (MAP) estimate. Using a Gibbs sampler, we also obtain the maximum marginal posterior probability (MMPP) estimate of the hidden process. We use these methods to determine where change-points occur in spatial transects through grassland vegetation, a problem of considerable interest to plant ecologists.