Bayesian Hierarchical Distributed Lag Models for Summer Ozone Exposure and Cardio-Respiratory Mortality

In this paper, we develop Bayesian hierarchical distributed lag models for estimating associations between daily variations in summer ozone levels and daily variations in cardiovascular and respiratory (CVDRESP) mortality counts for 19 U.S. large cities included in the National Morbidity Mortality Air Pollution Study (NMMAPS) for the period 1987 - 1994. At the first stage, we define a semi-parametric distributed lag Poisson regression model to estimate city-specific relative rates of CVDRESP associated with short-term exposure to summer ozone. At the second stage, we specify a class of distributions for the true city-specific relative rates to estimate an overall effect by taking into account the variability within and across cities. We perform the calculations with respect to several random effects distributions (normal, t-student, and mixture of normal), thus relaxing the common assumption of a two-stage normal-normal hierarchical model. We assess the sensitivity of the results to: 1) lag structure for ozone exposure; 2) degree of adjustment for long-term trends; 3) inclusion of other pollutants in the model;4) heat waves; 5) random effects distributions; and 6) prior hyperparameters. On average across cities, we found that a 10ppb increase in summer ozone level for every day in the previous week is associated with 1.25 percent increase in CVDRESP mortality (95% posterior regions: 0.47, 2.03). The relative rate estimates are also positive and statistically significant at lags 0, 1, and 2. We found that associations between summer ozone and CVDRESP mortality are sensitive to the confounding adjustment for PM_10, but are robust to: 1) the adjustment for long-term trends, other gaseous pollutants (NO_2, SO_2, and CO); 2) the distributional assumptions at the second stage of the hierarchical model; and 3) the prior distributions on all unknown parameters. Bayesian hierarchical distributed lag models and their application to the NMMAPS data allow us estimation of an acute health effect associated with exposure to ambient air pollution in the last few days on average across several locations. The application of these methods and the systematic assessment of the sensitivity of findings to model assumptions provide important epidemiological evidence for future air quality regulations.

Risk assessment for invasive exotic plants using multi-criteria risk models

Invasive exotic plants have altered natural ecosystems across much of North America. In the Midwest, the presence of invasive plants is increasing rapidly, causing changes in ecosystem patterns and processes. Early detection has become a key component in invasive plant management and in the detection of ecosystem change. Risk assessment through predictive modeling has been a useful resource for monitoring and assisting with treatment decisions for invasive plants. Predictive models were developed to assist with early detection of ten target invasive plants in the Great Lakes Network of the National Park Service and for garlic mustard throughout the Upper Peninsula of Michigan. These multi-criteria risk models utilize geographic information system (GIS) data to predict the areas at highest risk for three phases of invasion: introduction, establishment, and spread. An accuracy assessment of the models for the ten target plants in the Great Lakes Network showed an average overall accuracy of 86.3%. The model developed for garlic mustard in the Upper Peninsula resulted in an accuracy of 99.0%. Used as one of many resources, the risk maps created from the model outputs will assist with the detection of ecosystem change, the monitoring of plant invasions, and the management of invasive plants through prioritized control efforts.

Erupted magma volume estimates at Santiaguito and Pacaya Volcanoes, Guatemala using digital elevation models

High resolution digital elevation models (DEMs) of Santiaguito and Pacaya volcanoes, Guatemala, were used to estimate volume changes and eruption rates between 1954 and 2001. The DEMs were generated from contour maps and aerial photography, which were analyzed in ArcGIS 9.0®. Because both volcanoes were growing substantially over the five decade period, they provide a good data set for exploring effective methodology for estimating volume changes. The analysis shows that the Santiaguito dome complex grew by 0.78 ± 0.07 km3 (0.52 ± 0.05 m3 s-1) over the 1954-2001 period with nearly all the growth occurring on the El Brujo (1958-75) and Caliente domes (1971-2001). Adding information from field data prior to 1954, the total volume extruded from Santiaguito since 1922 is estimated at 1.48 ± 0.19 km3. Santiaguito’s growth rate is lower than most other volcanic domes, but it has been sustained over a much longer period and has undergone a change toward more exogenous and progressively slower extrusion with time. At Santiaguito some of the material being added at the dome is subsequently transported downstream by block and ash flows, mudflows and floods, creating channel shifting and areas of aggradation and erosion. At Pacaya volcano a total volume of 0.21 ± 0.05 km3 was erupted between 1961 and 2001 for an average extrusion rate of 0.17 ± 0.04 m3 s-1. Both the Santiaguito and Pacaya eruption rate estimates reported here are minima, because they do not include estimates of materials which are transported downslope after eruption and data on ashfall which may result in significant volumes of material spread over broad areas. Regular analysis of high resolution DEMs using the methods outlined here, would help quantify the effects of fluvial changes to downstream populated areas, as well as assist in tracking hazards related to dome collapse and eruption.

Using water balance models to approximate the effects of climate change on spring catchment discharge : Mt. Hanang, Tanzania

This project addresses the potential impacts of changing climate on dry-season water storage and discharge from a small, mountain catchment in Tanzania. Villagers and water managers around the catchment have experienced worsening water scarcity and attribute it to increasing population and demand, but very little has been done to understand the physical characteristics and hydrological behavior of the spring catchment. The physical nature of the aquifer was characterized and water balance models were calibrated to discharge observations so as to be able to explore relative changes in aquifer storage resulting from climate changes. To characterize the shallow aquifer supplying water to the Jandu spring, water quality and geochemistry data were analyzed, discharge recession analysis was performed, and two water balance models were developed and tested. Jandu geochemistry suggests a shallow, meteorically-recharged aquifer system with short circulation times. Baseflow recession analysis showed that the catchment behavior could be represented by a linear storage model with an average recession constant of 0.151/month from 2004-2010. Two modified Thornthwaite-Mather Water Balance (TMWB) models were calibrated using historic rainfall and discharge data and shown to reproduce dry-season flows with Nash-Sutcliffe efficiencies between 0.86 and 0.91. The modified TMWB models were then used to examine the impacts of nineteen, perturbed climate scenarios to test the potential impacts of regional climate change on catchment storage during the dry season. Forcing the models with realistic scenarios for average monthly temperature, annual precipitation, and seasonal rainfall distribution demonstrated that even small climate changes might adversely impact aquifer storage conditions at the onset of the dry season. The scale of the change was dependent on the direction (increasing vs. decreasing) and magnitude of climate change (temperature and precipitation). This study demonstrates that small, mountain aquifer characterization is possible using simple water quality parameters, recession analysis can be integrated into modeling aquifer storage parameters, and water balance models can accurately reproduce dry-season discharges and might be useful tools to assess climate change impacts. However, uncertainty in current climate projections and lack of data for testing the predictive capabilities of the model beyond the present data set, make the forecasts of changes in discharge also uncertain. The hydrologic tools used herein offer promise for future research in understanding small, shallow, mountainous aquifers and could potentially be developed and used by water resource professionals to assess climatic influences on local hydrologic systems.

Hotshots, hot spots, and female preference: exploring lek formation models with a bower-building cichlid fish

In many animals, males congregate in leks that females visit for the sole purpose of mating. We observed male and female behavior on 3 different-sized leks of the bower-building cichlid fish Nyassachromis cf. microcephalus to test predictions of 3 prominent lek models: the "hotshot," "hot spot," and "female preference" models. In this system, we were able to refine these predictions by distinguishing between indirect mate choice, by which females restrict their set of potential mates in the absence of individual male assessment, and direct mate choice, by which females assess males and their territories through dyadic behavioral interactions. On no lek were males holding central territories favored by indirect or direct mate choice, contrary to the prediction of the hotshot model that leks form because inferior males establish territories surrounding hotshot males preferred by females. Average female encounter rate of males increased with lek size, a pattern typically interpreted as evidence that leks form through female preference for lekking males, rather than because males congregate in hot spots of high female density. Female propensity to engage in premating behavior once courted did not increase with lek size, suggesting female preference for males on larger leks operated through indirect choice rather than direct choice based on individual assessment. The frequency of male-male competitive interactions increased with lek size, whereas their foraging rate decreased, implying a cost to males maintaining territories on larger leks. Together these data most strongly support the female preference model, where females may benefit through indirect mate choice for males able to meet the competitive cost of occupying larger leks.

Portfolio-optimization models for small investors

Since 2010, the client base of online-trading service providers has grown significantly. Such companies enable small investors to access the stock market at advantageous rates. Because small investors buy and sell stocks in moderate amounts, they should consider fixed transaction costs, integral transaction units, and dividends when selecting their portfolio. In this paper, we consider the small investor’s problem of investing capital in stocks in a way that maximizes the expected portfolio return and guarantees that the portfolio risk does not exceed a prescribed risk level. Portfolio-optimization models known from the literature are in general designed for institutional investors and do not consider the specific constraints of small investors. We therefore extend four well-known portfolio-optimization models to make them applicable for small investors. We consider one nonlinear model that uses variance as a risk measure and three linear models that use the mean absolute deviation from the portfolio return, the maximum loss, and the conditional value-at-risk as risk measures. We extend all models to consider piecewise-constant transaction costs, integral transaction units, and dividends. In an out-of-sample experiment based on Swiss stock-market data and the cost structure of the online-trading service provider Swissquote, we apply both the basic models and the extended models; the former represent the perspective of an institutional investor, and the latter the perspective of a small investor. The basic models compute portfolios that yield on average a slightly higher return than the portfolios computed with the extended models. However, all generated portfolios yield on average a higher return than the Swiss performance index. There are considerable differences between the four risk measures with respect to the mean realized portfolio return and the standard deviation of the realized portfolio return.

LakeMIP Kivu: evaluating the representation of a large, deep tropical lake by a set of one-dimensional lake models

The African great lakes are of utmost importance for the local economy (fishing), as well as being essential to the survival of the local people. During the past decades, these lakes experienced fast changes in ecosystem structure and functioning, and their future evolution is a major concern. In this study, for the first time a set of one-dimensional lake models are evaluated for Lake Kivu (2.28°S; 28.98°E), East Africa. The unique limnology of this meromictic lake, with the importance of salinity and subsurface springs in a tropical high-altitude climate, presents a worthy challenge to the seven models involved in the Lake Model Intercomparison Project (LakeMIP). Meteorological observations from two automatic weather stations are used to drive the models, whereas a unique dataset, containing over 150 temperature profiles recorded since 2002, is used to assess the model’s performance. Simulations are performed over the freshwater layer only (60 m) and over the average lake depth (240 m), since salinity increases with depth below 60 m in Lake Kivu and some lake models do not account for the influence of salinity upon lake stratification. All models are able to reproduce the mixing seasonality in Lake Kivu, as well as the magnitude and seasonal cycle of the lake enthalpy change. Differences between the models can be ascribed to variations in the treatment of the radiative forcing and the computation of the turbulent heat fluxes. Fluctuations in wind velocity and solar radiation explain inter-annual variability of observed water column temperatures. The good agreement between the deep simulations and the observed meromictic stratification also shows that a subset of models is able to account for the salinity- and geothermal-induced effects upon deep-water stratification. Finally, based on the strengths and weaknesses discerned in this study, an informed choice of a one-dimensional lake model for a given research purpose becomes possible.

Portfolio-optimization models for small investors

Since 2010, the client base of online-trading service providers has grown significantly. Such companies enable small investors to access the stock market at advantageous rates. Because small investors buy and sell stocks in moderate amounts, they should consider fixed transaction costs, integral transaction units, and dividends when selecting their portfolio. In this paper, we consider the small investor’s problem of investing capital in stocks in a way that maximizes the expected portfolio return and guarantees that the portfolio risk does not exceed a prescribed risk level. Portfolio-optimization models known from the literature are in general designed for institutional investors and do not consider the specific constraints of small investors. We therefore extend four well-known portfolio-optimization models to make them applicable for small investors. We consider one nonlinear model that uses variance as a risk measure and three linear models that use the mean absolute deviation from the portfolio return, the maximum loss, and the conditional value-at-risk as risk measures. We extend all models to consider piecewise-constant transaction costs, integral transaction units, and dividends. In an out-of-sample experiment based on Swiss stock-market data and the cost structure of the online-trading service provider Swissquote, we apply both the basic models and the extended models; the former represent the perspective of an institutional investor, and the latter the perspective of a small investor. The basic models compute portfolios that yield on average a slightly higher return than the portfolios computed with the extended models. However, all generated portfolios yield on average a higher return than the Swiss performance index. There are considerable differences between the four risk measures with respect to the mean realized portfolio return and the standard deviation of the realized portfolio return.

Reproducibility and accuracy of linear measurements on dental models derived from cone-beam computed tomography compared with digital dental casts.

INTRODUCTION The aim of this study was to determine the reproducibility and accuracy of linear measurements on 2 types of dental models derived from cone-beam computed tomography (CBCT) scans: CBCT images, and Anatomodels (InVivoDental, San Jose, Calif); these were compared with digital models generated from dental impressions (Digimodels; Orthoproof, Nieuwegein, The Netherlands). The Digimodels were used as the reference standard. METHODS The 3 types of digital models were made from 10 subjects. Four examiners repeated 37 linear tooth and arch measurements 10 times. Paired t tests and the intraclass correlation coefficient were performed to determine the reproducibility and accuracy of the measurements. RESULTS The CBCT images showed significantly smaller intraclass correlation coefficient values and larger duplicate measurement errors compared with the corresponding values for Digimodels and Anatomodels. The average difference between measurements on CBCT images and Digimodels ranged from -0.4 to 1.65 mm, with limits of agreement values up to 1.3 mm for crown-width measurements. The average difference between Anatomodels and Digimodels ranged from -0.42 to 0.84 mm with limits of agreement values up to 1.65 mm. CONCLUSIONS Statistically significant differences between measurements on Digimodels and Anatomodels, and between Digimodels and CBCT images, were found. Although the mean differences might be clinically acceptable, the random errors were relatively large compared with corresponding measurements reported in the literature for both Anatomodels and CBCT images, and might be clinically important. Therefore, with the CBCT settings used in this study, measurements made directly on CBCT images and Anatomodels are not as accurate as measurements on Digimodels.

Selection bias and the cross-validation of regression models for prediction

Strategies are compared for the development of a linear regression model with stochastic (multivariate normal) regressor variables and the subsequent assessment of its predictive ability. Bias and mean squared error of four estimators of predictive performance are evaluated in simulated samples of 32 population correlation matrices. Models including all of the available predictors are compared with those obtained using selected subsets. The subset selection procedures investigated include two stopping rules, C$\sb{\rm p}$ and S$\sb{\rm p}$, each combined with an 'all possible subsets' or 'forward selection' of variables. The estimators of performance utilized include parametric (MSEP$\sb{\rm m}$) and non-parametric (PRESS) assessments in the entire sample, and two data splitting estimates restricted to a random or balanced (Snee's DUPLEX) 'validation' half sample. The simulations were performed as a designed experiment, with population correlation matrices representing a broad range of data structures.^ The techniques examined for subset selection do not generally result in improved predictions relative to the full model. Approaches using 'forward selection' result in slightly smaller prediction errors and less biased estimators of predictive accuracy than 'all possible subsets' approaches but no differences are detected between the performances of C$\sb{\rm p}$ and S$\sb{\rm p}$. In every case, prediction errors of models obtained by subset selection in either of the half splits exceed those obtained using all predictors and the entire sample.^ Only the random split estimator is conditionally (on $\\beta$) unbiased, however MSEP$\sb{\rm m}$ is unbiased on average and PRESS is nearly so in unselected (fixed form) models. When subset selection techniques are used, MSEP$\sb{\rm m}$ and PRESS always underestimate prediction errors, by as much as 27 percent (on average) in small samples. Despite their bias, the mean squared errors (MSE) of these estimators are at least 30 percent less than that of the unbiased random split estimator. The DUPLEX split estimator suffers from large MSE as well as bias, and seems of little value within the context of stochastic regressor variables.^ To maximize predictive accuracy while retaining a reliable estimate of that accuracy, it is recommended that the entire sample be used for model development, and a leave-one-out statistic (e.g. PRESS) be used for assessment. ^

Revised age models and foraminiferal records obtained for the Last Interglacial period in six marine sediment cores

The dataset contains the revised age models and foraminiferal records obtained for the Last Interglacial period in six marine sediment cores: - the Southern Ocean core MD02-2488 (age model, sea surface temperatures, benthic d18O and d13C for the period 136-108 ka), - the North Atlantic core MD95-2042 (age model, planktic d18O, benthic d18O and d13C for the period 135-110 ka), - the North Atlantic core ODP 980 (age model, planktic d18O, sea surface temperatures, seawater d18O, benthic d18O and d13C, ice-rafted detritus for the period 135-110 ka), - the North Atlantic core CH69-K09 (age model, planktic d18O, sea surface temperatures, seawater d18O, benthic d18O and d13C, ice-rafted detritus for the period 135-110 ka), - the Norwegian Sea core MD95-2010 (age model, percentage of Neogloboquadrina pachyderma sinistral, sea surface temperatures, benthic d18O, ice-rafted detritus for the period 134-110 ka), - the Labrador Sea core EW9302-JPC2 (age model, percentage of Neogloboquadrina pachyderma sinistral, sea surface temperatures, benthic d18O for the period 134-110 ka).

Probabilistic models for the prediction of a ship performance in dynamic ice

We introduce two probabilistic, data-driven models that predict a ship's speed and the situations where a ship is probable to get stuck in ice based on the joint effect of ice features such as the thickness and concentration of level ice, ice ridges, rafted ice, moreover ice compression is considered. To develop the models to datasets were utilized. First, the data from the Automatic Identification System about the performance of a selected ship was used. Second, a numerical ice model HELMI, developed in the Finnish Meteorological Institute, provided information about the ice field. The relations between the ice conditions and ship movements were established using Bayesian learning algorithms. The case study presented in this paper considers a single and unassisted trip of an ice-strengthened bulk carrier between two Finnish ports in the presence of challenging ice conditions, which varied in time and space. The obtained results show good prediction power of the models. This means, on average 80% for predicting the ship's speed within specified bins, and above 90% for predicting cases where a ship may get stuck in ice. We expect this new approach to facilitate the safe and effective route selection problem for ice-covered waters where the ship performance is reflected in the objective function.

Age models, sedimentology and geochemistry from sediment cores in Drake Passage throughflow

The Drake Passage (DP) is the major geographic constriction for the Antarctic Circumpolar Current (ACC) and exerts a strong control on the exchange of physical, chemical, and biological properties between the Atlantic, Pacific, and Indian Ocean basins. Resolving changes in the flow of circumpolar water masses through this gateway is, therefore, crucial for advancing our understanding of the Southern Ocean's role in global ocean and climate variability. Here, we reconstruct changes in DP throughflow dynamics over the past 65,000 y based on grain size and geochemical properties of sediment records from the southernmost continental margin of South America. Combined with published sediment records from the Scotia Sea, we argue for a considerable total reduction of DP transport and reveal an up to ~40% decrease in flow speed along the northernmost ACC pathway entering the DP during glacial times. Superimposed on this long-term decrease are high-amplitude, millennial-scale variations, which parallel Southern Ocean and Antarctic temperature patterns. The glacial intervals of strong weakening of the ACC entering the DP imply an enhanced export of northern ACC surface and intermediate waters into the South Pacific Gyre and reduced Pacific-Atlantic exchange through the DP ("cold water route"). We conclude that changes in DP throughflow play a critical role for the global meridional overturning circulation and interbasin exchange in the Southern Ocean, most likely regulated by variations in the westerly wind field and changes in Antarctic sea ice extent.

Age models and stable isotope ratios of SO12 and SO35 sediment cores

Based on organic carbon accumulation rates, nine time slices of oceanic export paleoproductivity (Pnew) are presented which depict the variability of Pnew on a global scale through the last 30,000 years and document that the basic distribution patterns did not change through glacial and interglacial times. However, the glacial ocean shows an increased contrast of high- versus low-productivity zones. d13C values of near-surface-dwelling planktonic foraminifera Globigerinoides ruber suggest that the same contrast applies to the glacial nutrient inventories of the ambient surface waters, with a significant glacial transfer of PO4 from low- to high-productivity zones. In this way, glacial Pnew increased by a global average of about 2-4 Gt C/yr and led, via an enhanced CaCO3 dissolution and alkalinity in the deep ocean, to a significant extraction of CO2 from the surface water and the atrnosphere.