Adjusting for Ordinal Covariates by Inducing a Partial Ordering

In the context of randomized clinical trials, multiplicity arises in many forms. One prominent example is when a key endpoint is measured and analyzed both at baseline and after treatment. It is common to analyze each separately, but more efficient to adjust the post‐treatment comparisons for the baseline values. Adjustment techniques generally treat the covariate (baseline value, in this case) as either nominal or continuous. Either is problematic when applied to an ordinal covariate, the former because it fails to exploit the natural ordering and the latter because it relies on an artifical notion of linear prediction and differences between values. We propose new methods for adjusting for ordinal covariates without having to treat them as nominal or continuous. Specifically, the information‐preserving composite endpoint consists of the pair of values for each patient, one at baseline and one after treatment. Some of these patterns will indicate more improvement than others, yet some pairs of patterns are not comparable. Hence, the ordering is only partial. We develop an approach to testing and deriving estimators of magnitudes of the treatment effect based on comparing each observation in one group to each observation in the other group to which it is comparable. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

基于Maxent的大洋臀纹粉蚧和南洋臀纹粉蚧在中国的适生区分析

大洋臀纹粉蚧Planococcus minor Maskell和南洋臀纹粉蚧P.lilacinus Cockerell是我国有重要检疫意义的有害生物。这两种粉蚧从东盟进口水果口岸检疫中频繁截获,且已在广东、海南、云南等地发现大洋臀纹粉蚧入侵,这对我国热带、亚热带的水果和观赏植物等已构成严重威胁。本研究以大洋臀纹粉蚧和南洋臀纹粉蚧在国内外的分布数据为基础,利用Maxent生态位模型和Arc GIS对两种粉蚧在中国的潜在适生区进行预测。预测结果表明,大洋臀纹粉蚧和南洋臀纹粉蚧的适生区主要分布在长江流域以南地区,其适生区面积分别占全国面积的22.14%、18.17%,其中高度适生区域集中分布在广西、广东、云南、福建、四川、台湾、海南等地区,与我国热带水果的主要种植区域具有高度的一致性。大洋臀纹粉蚧已成功入侵,一旦南洋臀纹粉蚧传入便可迅速扩散蔓延,并对我国热带水果生产造成严重危害。因此,各口岸应加强对东盟进口水果的检验检疫,预防其新的进入和扩散。     

Predicted avian responses to bioenergy development scenarios in an intensive agricultural landscape

Conversion of native prairie to agriculture has increased food and bioenergy production but decreased wildlife habitat. However, enrollment of highly erodible cropland in conservation programs has compensated for some grassland loss. In the future, climate change and production of second-generation perennial biofuel crops could further transform agricultural landscapes and increase or decrease grassland area. Switchgrass (Panicum virgatum) is an alternative biofuel feedstock that may be economically and environmentally superior to maize (Zea mays) grain for ethanol production on marginally productive lands. Switchgrass could benefit farmers economically and increase grassland area, but there is uncertainty as to how conversions between rowcrops, switchgrass monocultures and conservation grasslands might occur and affect wildlife. To explore potential impacts on grassland birds, we developed four agricultural land-use change scenarios for an intensively cultivated landscape, each driven by potential future climatic changes and ensuing irrigation limitations, ethanol demand, commodity prices, and continuation of a conservation program. For each scenario, we calculated changes in area for landcover classes and predicted changes in grassland bird abundances. Overall, birds responded positively to the replacement of rowcrops with switchgrass and negatively to the conversion of conservation grasslands to switchgrass or rowcrops. Landscape context and interactions between climate, crop water use, and irrigation availability could influence future land-use, and subsequently, avian habitat quality and quantity. Switchgrass is likely to provide higher quality avian habitat than rowcrops but lower quality habitat than conservation grasslands, and therefore, may most benefit birds in heavily cultivated, irrigation dependent landscapes under warmer and drier conditions, where economic profitability may also encourage conversions to drought tolerant bioenergy feedstocks.     

Evaluating distributional shifts in home range estimates

A variety of methods are commonly used to quantify animal home ranges using location data acquired with telemetry. High‐volume location data from global positioning system (GPS) technology provide researchers the opportunity to identify various intensities of use within home ranges, typically quantified through utilization distributions (UDs). However, the wide range of variability evident within UDs constructed with modern home range estimators is often overlooked or ignored during home range comparisons, and challenges may arise when summarizing distributional shifts among multiple UDs. We describe an approach to gain additional insight into home range changes by comparing UDs across isopleths and summarizing comparisons into meaningful results. To demonstrate the efficacy of this approach, we used GPS location data from 16 bighorn sheep (Ovis canadensis) to identify distributional changes before and after habitat alterations, and we discuss advantages in its application when comparing home range size, overlap, and joint‐space use. We found a consistent increase in bighorn sheep home range size when measured across home range levels, but that home range overlap and similarity values decreased when examined at increasing core levels. Our results highlight the benefit of conducting multiscale assessments when comparing distributions, and we encourage researchers to expand comparative home range analyses to gain a more comprehensive evaluation of distributional changes and to evaluate comparisons across home range levels.     

Cancer Cluster Investigation in Residents Near a Municipal Landfill

While there are several studies of the health status of residents near hazardous waste landfills, relatively few have focused on the health of individuals living near municipal landfills. We assessed whether there were increased incidence rates of 18 cancer types and all cancers combined among residents near the Northampton Regional Landfill in western Massachusetts by conducting analyses for the cities of Northampton and Easthampton, as well as each census tract in these cities, between years 1982 and 2004. We also examined incidence rates within a 1-mile radius of the landfill and examined spatial patterns of all cancer cases in Northampton and Easthampton. The majority of cancer types occurred at, near, or below expected rates. No statistically significant association persisted over time. Within 1 mile of the landfill, all cancers combined occurred at rates comparable to the statewide average (SIR = 94, 95% CI: 69, 120) from 1997 to 2004, as did all individual cancer types analyzed. There were no unusual spatial patterns of cases in the area surrounding the landfill for the entire study period or four individual periods examined. The results do not support the hypothesis that residing near the investigated municipal landfill was associated with increased cancer risk.     

Sharing natural resources: mountain gorillas and people in the Parc National des Volcans,Rwanda

The compatibility of natural resource use by people and mountain gorillas (Gorilla beringei beringei) within the Parc National des Volcans was studied. The distribution of gorillas was modelled using a Maximum Entropy algorithm. Biophysical predictor variables were trained with daily GPS locations of gorillas during 2006. Elevation, as a climate surrogate, was the best predictor (58%) of the occurrence of gorillas. The mid‐altitudes (2500–3500 m a.s.l.) contained the bulk of the gorilla groups. Incoming solar radiation, as proxy for comfortable nesting sites, was the second best predictor (17%). Vegetation types, as foliage provider, (13%) and slope steepness for providing security (12%) were contributing predictors. The modelled and actual gorilla distributions were together overlaid with people’s resource use in the park. Both people and gorillas were congregated in the areas identified as most suitable for gorillas. However, within these areas spatial segregation was found between human natural resource‐users and gorillas. Therefore, the number of gorillas is likely to be limited by the human natural resource use within the park. A perimeter fence, the introduction of community‐based natural resource management, and a buffer zone are discussed as short‐, medium‐ and long‐term mitigation measures.     

Multiple McNemar Tests

Summary Methods for performing multiple tests of paired proportions are described. A broadly applicable method using McNemar's exact test and the exact distributions of all test statistics is developed; the method controls the familywise error rate in the strong sense under minimal assumptions. A closed form (not simulation‐based) algorithm for carrying out the method is provided. A bootstrap alternative is developed to account for correlation structures. Operating characteristics of these and other methods are evaluated via a simulation study. Applications to multiple comparisons of predictive models for disease classification and to postmarket surveillance of adverse events are given.     

Comparative analysis of marine ecosystems: international production modelling workshop

Understanding the drivers that dictate the productivity of marine ecosystems continues to be a globally important issue. A vast literature identifies three main processes that regulate the production dynamics of such ecosystems: biophysical, exploitative and trophodynamic. Exploring the prominence among this ‘triad’ of drivers, through a synthetic analysis, is critical for understanding how marine ecosystems function and subsequently produce fisheries resources of interest to humans. To explore this topic further, an international workshop was held on 10–14 May 2010, at the National Academy of Science''s Jonsson Center in Woods Hole, MA, USA. The workshop compiled the data required to develop production models at different hierarchical levels (e.g. species, guild, ecosystem) for many of the major Northern Hemisphere marine ecosystems that have supported notable fisheries. Analyses focused on comparable total system biomass production, functionally equivalent species production, or simulation studies for 11 different marine fishery ecosystems. Workshop activities also led to new analytical tools. Preliminary results suggested common patterns driving overall fisheries production in these ecosystems, but also highlighted variation in the relative importance of each among ecosystems.     

Fitting the bill: do different winter food resources influence juniper titmouse (Baeolophus ridgwayi) bill morphology?

Divergent adaptive selection is a prominent mechanism influencing patterns of morphological diversity. We used the juniper titmouse [Baeolophus ridgwayi (Richmond, 1902)], a nonmigratory passerine that inhabits woodlands throughout western North America, to investigate variation in bill morphology in relation to diet and geography. We gathered data from museum specimens and used morphometric techniques to determine the relative strength of support for competing hypotheses using information theoretics: (1) differences in bill morphology are predicted by a key winter food resource that each regional population consumes (seeds of different juniper tree species); or (2) bill morphology scales with body size, and both increase along a latitudinal gradient. Juniper species emerged as the variable with the most support explaining variation in bill size, supporting the hypothesis that seed sizes influence bill size, independently of body size. The shape analysis revealed no distinct patterns in bill shape variation, but employed a powerful method for evaluating the strength of support for numerous candidate models. The differences in bill size of juniper titmice across their range are likely to reflect adaptive variation, because bill morphology is highly heritable in birds, juniper titmouse gene flow appears to be relatively low, and there is a clear mechanistic explanation for why bill sizes may differ among the ranges of the three juniper species. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 101 , 667–679.