A note on testing separability in spatial-temporal marked point processes

Summary .   In environmental risk analysis, it is common to assume the stochastic independence (or separability) between the marks associated with the random events of a spatial-temporal point process. Schoenberg (2004, Biometrics 60, 471–481) proposed several test statistics for this hypothesis and used simulated data to evaluate their performance. He found that a Cramér-von Mises-type test is powerful to detect gradual departures from separability although it is not uniformly powerful over a large class of alternative models. We present a semiparametric approach to model alternative hypotheses to separability and derive a score test statistic. We show that there is a relationship between this score test and some of the test statistics proposed by Schoenberg. Specifically, all are different versions of weighted Cramér-von Mises-type statistics. This gives some insight into the reasons for the similarities and differences between the test statistics' performance. We also point out some difficulties in controlling the type I error probability in Schoenberg's residual test.     

Spatio-temporal variation in coral recruitment at different scales on Heron Reef, southern Great Barrier Reef

 Recruitment of scleractinian corals on settlement plates at Heron Island, Great Barrier Reef, was examined over four years (September 1991–September 1995) to quantify spatio-temporal patterns at different scales and to assess post settlement mortality. Recruitment was dominated by pocilloporid corals which accounted for 80.1% of the 8627 spat counted, whereas non-isoporan acroporids represented only 16.4%. Poritids, faviids and isoporan acroporids rarely recruited to the plates (3.5%), despite their obvious abundance as adults on the reef. Recruitment patterns on the plates indicate strong space-time interactions as evidenced by patchy recruitment of both pocilloporid and acroporid spat. Interactions were found between space (on the scale of 10² m, i.e. sites within zones, and 10¹ m, i.e. racks within sites) and time (on the scale of years) for pocilloporids and between space (on the scale of 10³ m, i.e. zones, and 10² m) and time (on the scale of years) for acroporids. Post-recruitment mortality of acroporid spat in the period 3–10 months after their major spawning was dependent on their initial recruitment density, but pocilloporid mortality was either independent of initial recruitment density or, more likely, obscured by additional recruitment of pocilloporids to plates between late February and September. High rates of recruitment and growth by other sessile organisms, particularly bryozoans and oysters, appear to result in increased post-recruitment mortality and limit recruitment of scleractinian corals on settlement plates. The work reinforces an emerging picture that coral recruitment patterns are determined by mechanisms that manifest over a large range of spatial scales. Accepted: 1 September 1997     

A spatiotemporal stochastic process model for species spread

We use a spatiotemporal Markov process to model the spread of an ecological population through its environment over time. Available habitat is divided into sites, and a parametric function of spatial variables is used to model the probability that one site is colonized from another. This allows us both to make predictions about the future spread of a population, and to determine which are the important factors governing colonizations. The model evolves in discrete time, allowing the population distribution to change seasonally in accordance with breeding patterns. Discrete time formulations are natural for ecological populations, but are problematic due to difficulties of fitting and predicting over irregular time intervals. The model described here can accommodate years of missing data and can therefore fit and predict at irregular intervals. Two methods of approximating the likelihood are described and applied to ornithological survey data for the woodlark, Lullula arborea, from Thetford Forest in the U.K.     

巨桉连栽对土壤微生物生物量和数量的影响

为探讨巨桉连栽对土壤微生物生物量和数量的影响,采用时空互换法,研究了马尾松(Pinus massoniana)林和不同连栽代次的巨桉(Eucalyptus grandis)人工林的微生物生物量、数量与土壤理化性质的关系。结果表明,巨桉一代林的土壤微生物生物量碳、生物量氮和土壤真菌、细菌数量与马尾松林的差异不显著,放线菌数量则显著增加。随巨桉连栽代次增加,土壤微生物生物量碳、生物量氮和土壤细菌、放线菌、真菌数量均递减。回归分析表明,土壤全磷能独立解释微生物生物量氮、细菌和真菌数量71.7%、86.1%和63.0%的变异,与总孔隙度共同解释微生物生物量碳87.9%的变异,与全氮共同解释放线菌数量89.6%的变异。可见,土壤全磷较大程度解释了微生物的变化。     

The Knox method and other tests for space-time interaction

The Knox method, as well as other tests for space-time interaction, are biased when there are geographical population shifts, i.e., when there are different percent population growths in different regions. In this paper, the size of the population shift bias is investigated for the Knox test, and it is shown that it can be a considerable problem. A Monte Carlo method for constructing unbiased space-time interaction tests is then presented and illustrated on the Knox test as well as for a combined Knox test. Practical implications are discussed in terms of the interpretation of past results and the design of future studies.     

Predicting protein structural classes with pseudo amino acid composition: an approach using geometric moments of cellular automaton image

A novel approach was developed for predicting the structural classes of proteins based on their sequences. It was assumed that proteins belonging to the same structural class must bear some sort of similar texture on the images generated by the cellular automaton evolving rule [Wolfram, S., 1984. Cellular automation as models of complexity. Nature 311, 419-424]. Based on this, two geometric invariant moment factors derived from the image functions were used as the pseudo amino acid components [Chou, K.C., 2001. Prediction of protein cellular attributes using pseudo amino acid composition. Proteins: Struct., Funct., Genet. (Erratum: ibid., 2001, vol. 44, 60) 43, 246-255] to formulate the protein samples for statistical prediction. The success rates thus obtained on a previously constructed benchmark dataset are quite promising, implying that the cellular automaton image can help to reveal some inherent and subtle features deeply hidden in a pile of long and complicated amino acid sequences.