Investigating the consistency between proxy-based reconstructions and climate models using data assimilation: a mid-Holocene case study

The mid-Holocene (6 kyr BP; thousand years before present) is a key period to study the consistency between model results and proxy-based reconstruction data as it corresponds to a standard test for models and a reasonable number of proxy-based records is available. Taking advantage of this relatively large amount of information, we have compared a compilation of 50 air and sea surface temperature reconstructions with the results of three simulations performed with general circulation models and one carried out with LOVECLIM, a model of intermediate complexity. The conclusions derived from this analysis confirm that models and data agree on the large-scale spatial pattern but the models underestimate the magnitude of some observed changes and that large discrepancies are observed at the local scale. To further investigate the origin of those inconsistencies, we have constrained LOVECLIM to follow the signal recorded by the proxies selected in the compilation using a data-assimilation method based on a particle filter. In one simulation, all the 50 proxy-based records are used while in the other two only the continental or oceanic proxy-based records constrain the model results. As expected, data assimilation leads to improving the consistency between model results and the reconstructions. In particular, this is achieved in a robust way in all the experiments through a strengthening of the westerlies at midlatitude that warms up northern Europe. Furthermore, the comparison of the LOVECLIM simulations with and without data assimilation has also objectively identified 16 proxy-based paleoclimate records whose reconstructed signal is either incompatible with the signal recorded by some other proxy-based records or with model physics.

Development and validation of energy expenditure prediction models based on GT3X accelerometer data in 5- to 9-year-old children

Background: Accelerometry has been established as an objective method that can be used to assess physical activity behavior in large groups. The purpose of the current study was to provide a validated equation to translate accelerometer counts of the triaxial GT3X into energy expenditure in young children. Methods: Thirty-two children aged 5–9 years performed locomotor and play activities that are typical for their age group. Children wore a GT3X accelerometer and their energy expenditure was measured with indirect calorimetry. Twenty-one children were randomly selected to serve as development group. A cubic 2-regression model involving separate equations for locomotor and play activities was developed on the basis of model fit. It was then validated using data of the remaining children and compared with a linear 2-regression model and a linear 1-regression model. Results: All 3 regression models produced strong correlations between predicted and measured MET values. Agreement was acceptable for the cubic model and good for both linear regression approaches. Conclusions: The current linear 1-regression model provides valid estimates of energy expenditure for ActiGraph GT3X data for 5- to 9-year-old children and shows equal or better predictive validity than a cubic or a linear 2-regression model.

Design and analysis for three level hierarchical structures with count response

In numerous intervention studies and education field trials, random assignment to treatment occurs in clusters rather than at the level of observation. This departure of random assignment of units may be due to logistics, political feasibility, or ecological validity. Data within the same cluster or grouping are often correlated. Application of traditional regression techniques, which assume independence between observations, to clustered data produce consistent parameter estimates. However such estimators are often inefficient as compared to methods which incorporate the clustered nature of the data into the estimation procedure (Neuhaus 1993).1 Multilevel models, also known as random effects or random components models, can be used to account for the clustering of data by estimating higher level, or group, as well as lower level, or individual variation. Designing a study, in which the unit of observation is nested within higher level groupings, requires the determination of sample sizes at each level. This study investigates the design and analysis of various sampling strategies for a 3-level repeated measures design on the parameter estimates when the outcome variable of interest follows a Poisson distribution. ^ Results study suggest that second order PQL estimation produces the least biased estimates in the 3-level multilevel Poisson model followed by first order PQL and then second and first order MQL. The MQL estimates of both fixed and random parameters are generally satisfactory when the level 2 and level 3 variation is less than 0.10. However, as the higher level error variance increases, the MQL estimates become increasingly biased. If convergence of the estimation algorithm is not obtained by PQL procedure and higher level error variance is large, the estimates may be significantly biased. In this case bias correction techniques such as bootstrapping should be considered as an alternative procedure. For larger sample sizes, those structures with 20 or more units sampled at levels with normally distributed random errors produced more stable estimates with less sampling variance than structures with an increased number of level 1 units. For small sample sizes, sampling fewer units at the level with Poisson variation produces less sampling variation, however this criterion is no longer important when sample sizes are large. ^ 1Neuhaus J (1993). “Estimation efficiency and Tests of Covariate Effects with Clustered Binary Data”. Biometrics , 49, 989–996^

Sequential updating via dynamic hierarchical models: A Bayesian approach to longitudinal data analysis

Most statistical analysis, theory and practice, is concerned with static models; models with a proposed set of parameters whose values are fixed across observational units. Static models implicitly assume that the quantified relationships remain the same across the design space of the data. While this is reasonable under many circumstances this can be a dangerous assumption when dealing with sequentially ordered data. The mere passage of time always brings fresh considerations and the interrelationships among parameters, or subsets of parameters, may need to be continually revised. ^ When data are gathered sequentially dynamic interim monitoring may be useful as new subject-specific parameters are introduced with each new observational unit. Sequential imputation via dynamic hierarchical models is an efficient strategy for handling missing data and analyzing longitudinal studies. Dynamic conditional independence models offers a flexible framework that exploits the Bayesian updating scheme for capturing the evolution of both the population and individual effects over time. While static models often describe aggregate information well they often do not reflect conflicts in the information at the individual level. Dynamic models prove advantageous over static models in capturing both individual and aggregate trends. Computations for such models can be carried out via the Gibbs sampler. An application using a small sample repeated measures normally distributed growth curve data is presented. ^

Matrix models for quantifying competitive intransitivity from species abundance data

In a network of competing species, a competitive intransitivity occurs when the ranking of competitive abilities does not follow a linear hierarchy (A > B > C but C > A). A variety of mathematical models suggests that intransitive networks can prevent or slow down competitive exclusion and maintain biodiversity by enhancing species coexistence. However, it has been difficult to assess empirically the relative importance of intransitive competition because a large number of pairwise species competition experiments are needed to construct a competition matrix that is used to parameterize existing models. Here we introduce a statistical framework for evaluating the contribution of intransitivity to community structure using species abundance matrices that are commonly generated from replicated sampling of species assemblages. We provide metrics and analytical methods for using abundance matrices to estimate species competition and patch transition matrices by using reverse-engineering and a colonization-competition model. These matrices provide complementary metrics to estimate the degree of intransitivity in the competition network of the sampled communities. Benchmark tests reveal that the proposed methods could successfully detect intransitive competition networks, even in the absence of direct measures of pairwise competitive strength. To illustrate the approach, we analyzed patterns of abundance and biomass of five species of necrophagous Diptera and eight species of their hymenopteran parasitoids that co-occur in beech forests in Germany. We found evidence for a strong competitive hierarchy within communities of flies and parasitoids. However, for parasitoids, there was a tendency towards increasing intransitivity in higher weight classes, which represented larger resource patches. These tests provide novel methods for empirically estimating the degree of intransitivity in competitive networks from observational datasets. They can be applied to experimental measures of pairwise species interactions, as well as to spatio-temporal samples of assemblages in homogenous environments or environmental gradients.

Pre-therapy inflammation and coagulation activation and long-term CD4 count responses to the initiation of antiretroviral therapy

OBJECTIVES Pre-antiretroviral therapy (ART) inflammation and coagulation activation predict clinical outcomes in HIV-positive individuals. We assessed whether pre-ART inflammatory marker levels predicted the CD4 count response to ART. METHODS Analyses were based on data from the Strategic Management of Antiretroviral Therapy (SMART) trial, an international trial evaluating continuous vs. interrupted ART, and the Flexible Initial Retrovirus Suppressive Therapies (FIRST) trial, evaluating three first-line ART regimens with at least two drug classes. For this analysis, participants had to be ART-naïve or off ART at randomization and (re)starting ART and have C-reactive protein (CRP), interleukin-6 (IL-6) and D-dimer measured pre-ART. Using random effects linear models, we assessed the association between each of the biomarker levels, categorized as quartiles, and change in CD4 count from ART initiation to 24 months post-ART. Analyses adjusted for CD4 count at ART initiation (baseline), study arm, follow-up time and other known confounders. RESULTS Overall, 1084 individuals [659 from SMART (26% ART naïve) and 425 from FIRST] met the eligibility criteria, providing 8264 CD4 count measurements. Seventy-five per cent of individuals were male with the mean age of 42 years. The median (interquartile range) baseline CD4 counts were 416 (350-530) and 100 (22-300) cells/μL in SMART and FIRST, respectively. All of the biomarkers were inversely associated with baseline CD4 count in FIRST but not in SMART. In adjusted models, there was no clear relationship between changing biomarker levels and mean change in CD4 count post-ART (P for trend: CRP, P = 0.97; IL-6, P = 0.25; and D-dimer, P = 0.29). CONCLUSIONS Pre-ART inflammation and coagulation activation do not predict CD4 count response to ART and appear to influence the risk of clinical outcomes through other mechanisms than blunting long-term CD4 count gain.

Achieving a more realistic assessment of rockfall hazards by coupling three-dimensional process models and field-based tree-ring data

Sound knowledge of the spatial and temporal patterns of rockfalls is fundamental for the management of this very common hazard in mountain environments. Process-based, three-dimensional simulation models are nowadays capable of reproducing the spatial distribution of rockfall occurrences with reasonable accuracy through the simulation of numerous individual trajectories on highly-resolved digital terrain models. At the same time, however, simulation models typically fail to quantify the ‘real’ frequency of rockfalls (in terms of return intervals). The analysis of impact scars on trees, in contrast, yields real rockfall frequencies, but trees may not be present at the location of interest and rare trajectories may not necessarily be captured due to the limited age of forest stands. In this article, we demonstrate that the coupling of modeling with tree-ring techniques may overcome the limitations inherent to both approaches. Based on the analysis of 64 cells (40 m × 40 m) of a rockfall slope located above a 1631-m long road section in the Swiss Alps, we illustrate results from 488 rockfalls detected in 1260 trees. We illustrate that tree impact data cannot only be used (i) to reconstruct the real frequency of rockfalls for individual cells, but that they also serve (ii) the calibration of the rockfall model Rockyfor3D, as well as (iii) the transformation of simulated trajectories into real frequencies. Calibrated simulation results are in good agreement with real rockfall frequencies and exhibit significant differences in rockfall activity between the cells (zones) along the road section. Real frequencies, expressed as rock passages per meter road section, also enable quantification and direct comparison of the hazard potential between the zones. The contribution provides an approach for hazard zoning procedures that complements traditional methods with a quantification of rockfall frequencies in terms of return intervals through a systematic inclusion of impact records in trees.

Effect of low count sums on quantitative environmental reconstructions: an example using subfossil chironomids

The concentrations of chironomid remains in lake sediments are very variable and, therefore, chironomid stratigraphies often include samples with a low number of counts. Thus, the effect of low count sums on reconstructed temperatures is an important issue when applying chironomid‐temperature inference models. Using an existing data set, we simulated low count sums by randomly picking subsets of head capsules from surface‐sediment samples with a high number of specimens. Subsequently, a chironomid‐temperature inference model was used to assess how the inferred temperatures are affected by low counts. The simulations indicate that the variability of inferred temperatures increases progressively with decreasing count sums. At counts below 50 specimens, a further reduction in count sum can cause a disproportionate increase in the variation of inferred temperatures, whereas at higher count sums the inferences are more stable. Furthermore, low count samples may consistently infer too low or too high temperatures and, therefore, produce a systematic error in a reconstruction. Smoothing reconstructed temperatures downcore is proposed as a possible way to compensate for the high variability due to low count sums. By combining adjacent samples in a stratigraphy, to produce samples of a more reliable size, it is possible to assess if low counts cause a systematic error in inferred temperatures.