Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing Data

Summary .  We study quantile regression (QR) for longitudinal measurements with nonignorable intermittent missing data and dropout. Compared to conventional mean regression, quantile regression can characterize the entire conditional distribution of the outcome variable, and is more robust to outliers and misspecification of the error distribution. We account for the within-subject correlation by introducing a   ℓ₂   penalty in the usual QR check function to shrink the subject-specific intercepts and slopes toward the common population values. The informative missing data are assumed to be related to the longitudinal outcome process through the shared latent random effects. We assess the performance of the proposed method using simulation studies, and illustrate it with data from a pediatric AIDS clinical trial.     

Comparing Stochastically Restricted Linear Estimators in a Regression Model

Conditions for superiority of the minimum dispersion estimator over another with respect to the covariance matrix are derived when the vector parameter of a regression model is subject to competing stochastic restrictions. The restrictions may also consist both of a deterministic part and a stochastic part.     

Species co‐occurrence patterns and dietary resource competition in primates

Diamond (Assembly of species communities. In: Cody ML, Diamond JM, editors. Ecology and evolution of communities. Cambridge: Belknap. p 342–444 ( 1975 )) argued that interspecific competition between species occupying similar niches results in a nonrandom pattern of species distributions. In particular, some species pairs may never be found in the same community due to competitive exclusion. Rigorous analytical methods have been developed to investigate the possible role that interspecific competition has on the evolution of communities. Many studies that have implemented these methods have shown support for Diamond's assembly rules, yet there are numerous exceptions. We build on this previous research by examining the co‐occurrence patterns of primate species in 109 communities from across the world. We used EcoSim to calculate a checkerboard (C) score for each region. The C score provides a measure of the proportion of species pairs that do not co‐occur in a set of communities. High C scores indicate that species are nonrandomly distributed throughout a region, and interspecific competition may be driving patterns of competitive exclusion. We conducted two sets of analyses. One included all primate species per region, and the second analysis assigned each species to one of four dietary guilds: frugivores, folivores, insectivores, and frugivore‐insectivores. Using all species per region, we found significantly high C scores in 9 of 10 regions examined. For frugivores, we found significantly high‐C scores in more than 50% of regions. In contrast, only 23% of regions exhibited significantly high‐C scores for folivores. Our results suggest that communities are nonrandomly structured and may be the result of greater levels of interspecific competition between frugivores compared to folivores. Am J Phys Anthropol, 2010. © 2010 Wiley‐Liss, Inc.     

A physiological analogy of the niche for projecting the potential distribution of plants

Aim  To develop a physiologically based model of the plant niche for use in species distribution modelling. Location  Europe. Methods  We link the Thornley transport resistance (TTR) model with functions which describe how the TTR’s model parameters are influenced by abiotic environmental factors. The TTR model considers how carbon and nutrient uptake, and the allocation of these assimilates, influence growth. We use indirect statistical methods to estimate the model parameters from a high resolution data set on tree distribution for 22 European tree species. Results  We infer, from distribution data and abiotic forcing data, the physiological niche dimensions of 22 European tree species. We found that the model fits were reasonable (AUC: 0.79–0.964). The projected distributions were characterized by a false positive rate of 0.19 and a false negative rate 0.12. The fitted models are used to generate projections of the environmental factors that limit the range boundaries of the study species. Main conclusions  We show that physiological models can be used to derive physiological niche dimensions from species distribution data. Future work should focus on including prior information on physiological rates into the parameter estimation process. Application of the TTR model to species distribution modelling suggests new avenues for establishing explicit links between distribution and physiology, and for generating hypotheses about how ecophysiological processes influence the distribution of plants.     

Rates of small-scale species mobility in alvar limestone grassland

Abstract. Small-scale species frequency and cumulative species frequency were studied in four plots in limestone grassland of the Veronica spicata-Avenula pratensis association on Stora Alvaret on the Baltic island of Öland, Sweden. Species mobility was expressed as increase in cumulative species frequency in 20 subplots of 100 cm². Observed cumulative frequencies from 1985–1989 in all four plots, and from 1985–1995 in one plot were compared with values following from two null models, a ‘minimal mobility’ model and a random mobility model. In ca. 50 % of the cases the observed cumulative frequency was not significantly different from the random expectation. However, in many such cases the mean annual frequency was either very high or very low. Three ways of calculating the mobility rate are presented though only one is used: (observed cumulative frequency -lowest annual frequency) / expected cumulative frequency. Values × 100 range from 0 to 100. There were slight differences between the four plots which were interpreted in terms of differences in grazing intensity and soil depth. It is stressed that the idea of the Carousel model has never been meant to suggest that all species would show random mobility, which we now quantify, but that species differ in their mobility rate and that the mean rate is much higher than generally realized.     

Evaluation of past and potential phosphorus uptake at the Orlando Easterly Wetland

The Orlando Easterly Wetland (OEW), located near Christmas, Florida, USA, is among the longer-lived treatment wetlands in the United States. It was established in the late 1980s to reduce nitrogen and phosphorus concentrations from tertiary treated wastewater bound for the St. Johns River. A goal of 0.07 mg/l total phosphorus concentration has been set by the regulating agency (St. Johns River Water Management District). In order to understand and define the operating conditions for which this target could be met, a systematic study of historic phosphorus uptake was performed using a traditional first-order model. Phosphorus uptake performance is shown to correlate well with hydraulic performance for two parallel upstream cells. The first-order model is enhanced with predictive capabilities that acknowledge the correlation between the phosphorus uptake rate constant and the hydraulic loading rate observed in the system. Inherent limitations with the first-order modeling approach are addressed and uncertainty in model performance is used to bound predictions.     

Characterization of Viticultural Terroirs using a Simple Field Model Based on Soil Depth – II. Validation of the Grape Yield and Berry Quality in the Anjou Vineyard (France)

Soil and climate are major constituents of the French notion of Terroir. This concept implies that there is a strong relationship between the composition of the grape, the characteristics of the wine and the territory of production. To study this link, a new method of characterization of the Terroir, including geological and pedological factors, was investigated. It uses a field model based on depth and clay content of soil, together with the degree of weathering of the parent rock. Consequently, for every type of parent rock belonging to a given geologic stage, there are a series of soils that show different stages of pedological evolution. According to the model, three kinds of soils are distinguished with regards to the weathering intensity of the parent rock, that are named weakly weathered rock (WWR), moderately weathered rock (MWR) and strongly weathered rock (SWR). By hypothesis, each soil type is considered as a homogeneous unit for vine production from the viewpoint of ecophysiological factors. Each terroir unit defined by this method is called a Basic Terroir Unit (BTU). To validate this hypothesis, experimental plots planted with Chenin and Cabernet Franc vines were studied over three consecutive seasons (2000–2002), in the Anjou vineyard (Loire Valley – France). The major BTUs developed on the two most important geological systems of Anjou (Brioverian and Ordovician–Devonian), were studied. Results showed that the berries of vines cultivated in WWR were significantly smaller, richer in sugars and anthocyanins and had a Total Phenolic Index higher than those of the vines cultivated in SWR. They also had a lower titratable acidity. Cabernet Franc vines cultivated in MWR had berries with sugar and anthocyanin contents but also total phenolics very close to those of WWR. With Chenin vines there was a good relationship between the global pool of free aromas of berries and the BTU type. The study showed significant relationships between the quality of grapes and the measured values of several ecophysiological variables such as the water supply regime or the timing of budburst.     

Application of a photosystem II model for analysis of fluorescence induction curves in the 100 ns to 10 s time domain after excitation with a saturating light pulse

A mathematical model of photosystem II (PSII) events was used to analyze chlorophyll fluorescence transients in the time domain from 100 ns to 10 s after excitation with a saturating 10-ns flash, applied as a part of specialized illumination protocol, using preparations of a thermophilic strain of the unicellular green alga, Chlorella pyrenoidosa Chick (using both intact and diuron-treated cells). Analysis of simulation results has proven that particular attention should be given to flash-induced recombination processes, including nonradiative recombination in PSII, while subsequent charge transfer along the electron transport chain of thylakoid membrane can be adequately described by a single reaction of quinone reoxidation. The PSII model was extended by taking inhibition by diuron of the electron transport in the acceptor side of PSII into account, which allowed simulation of fluorescence induction curves observed in the presence of this inhibitor. The model parameters were determined (stromal pH, rate constants of nonradiative recombination, and the initial reduction state of the quinone pool) which provided adequate simulation of experimentally observed ratios of the maximal and initial fluorescence levels (F _m/F ₀).