A Single Column Model Evaluation of Mixing Length Formulations and Constraints for the sa-TKE-EDMF Planetary Boundary Layer Parameterization

Parameterizing boundary layer turbulence is a critical component of numerical weather prediction and the representation of turbulent mixing of momentum, heat, and other tracers. The components that make up a boundary layer scheme can vary considerably, with each scheme having a combination of processes that are physically represented along with tuning parameters that optimize performance. Isolating a component of a PBL scheme to examine its impact is essential for understanding the evolution of boundary layer profiles and their impact on the mean structure. In this study we conduct three experiments with the scale-aware TKE eddy-diffusivity mass-flux (sa-TKE-EDMF) scheme: 1) releasing the upper limit constraints placed on mixing lengths, 2) incrementally adjusting the tuning coefficient related to wind shear in the modified Bougeault and Lacarrere (BouLac) mixing length formulation, and 3) replacing the current mixing length formulations with those used in the MYNN scheme. A diagnostic approach is adopted to characterize the bulk representation of turbulence within the residual layer and boundary layer in order to understand the importance of different terms in the TKE budget as well as to assess how the balance of terms changes between mixing length formulations. Although our study does not seek to determine the best formulation, it was found that strong imbalances led to considerably different profile structures both in terms of the resolved and subgrid fields. Experiments where this balance was preserved showed a minor impact on the mean structure regardless of the turbulence generated. Overall, it was found that changes to mixing length formulations and/or constraints had stronger impacts during the day while remaining partially insensitive during the evening.

Psychometric properties of the Revised Children's Manifest Anxiety Scale-Second Edition in Peruvian students.

The aim of this study was to analyze the psychometric properties of the Revised Children’s Manifest Anxiety Scale–Second Edition (RCMAS-2) among Peruvian students. The sample consisted of 472 participants aged between 7 and 18 years, of whom 250 were female (53%). Likewise, 191 were enrolled from third to sixth grade of primary school (40.5%), and 281 were registered from first to fifth grade of secondary school (59.5%). The results of the study indicated that the RCMAS-2 scores had adequate levels of reliability for all its dimensions (ordinal alpha > .70). On the other hand, a four-factor structure (Physiological anxiety, Worry/Social anxiety, Defensiveness I, and Defensiveness II) was found to be invariant to gender and schooling level. Also, convergent and discriminant validity evidence was provided. Finally, a moderate difference in Defensiveness II according to the schooling level through the latent mean structure analysis was found. Taking into account the results, it was concluded that the RCMAS-2 scores have evidence of reliability, validity, and equity for its use in Peruvian regular elementary school students.

How to Estimate Absolute-Error Components in Structural Equation Models of Generalizability Theory

Structural equation modeling (SEM) has been proposed to estimate generalizability theory (GT) variance components, primarily focusing on estimating relative error to calculate generalizability coefficients. Proposals for estimating absolute-error components have given the impression that a separate SEM must be fitted to a transposed data matrix. This paper uses real and simulated data to demonstrate how a single SEM can be specified to estimate absolute error (and thus dependability) by placing appropriate constraints on the mean structure, as well as thresholds (when used for ordinal measures). Using the R packages lavaan and gtheory, different estimators are compared for normal and discrete measurements. Limitations of SEM for GT are demonstrated using multirater data from a planned missing-data design, and an important remaining area for future development is discussed.

Zonal scale and temporal variability of the Asian monsoon anticyclone in an idealised numerical model

The upper-level monsoon anticyclone is studied in a 3-D dry dynamical model as the response of a background circulation without any imposed zonal structure to a steady imposed zonally confined heat source. The characteristics of the background circulation are determined by thermal relaxation towards a simple meridionally varying state, which gives rise to baroclinic instability if meridional gradients are sufficiently large. This model configuration allows study of the dependence of the monsoon anticyclone response on characteristics of both the imposed heating and the background state, in particular including interactions between the anticyclone and the active dynamics on its poleward side in the form of the jet and baroclinic eddies. As characteristics of forcing and background state are varied a range of different behaviours emerges, many of which strongly resemble phenomena and features associated with the monsoon anticyclone as observed in re-analysis data. For a resting background state the time-mean anticyclone is highly extended in longitude to the west of the forcing region. When the active mid-latitude dynamics is included the zonal extent of the time-mean anticyclone is limited, without any need for the explicit upper-level momentum dissipation which is often included in simple theoretical models, but difficult to justify physically. We further describe in detail the spontaneous emergence of temporal variability in the form of westward eddy shedding from the monsoon anticyclone for varying strength of the imposed heating. By varying the strength of the background mid-latitude dynamics we observe a transition of the system from a state with periodic westward eddy shedding to a state dominated by eastward shedding. The details of the time-mean structure and temporal evolution depend on the structure of the background flow and for certain flows the monsoon anticyclone shows signs of both westward and eastward shedding.

Coastal gradients in False Bay, south of Cape Town: what insights can be gained from mesoscale reanalysis?

Mesoscale datasets are used to study coastal gradients in the marine climate and oceanography in False Bay, south of Cape Town. Building on past work, satellite and ocean–atmosphere reanalyses are used to gain new insights into the mean structure, circulation and meteorological features. HYCOM v3 hindcasts represent a coastward reduction of mixing that enhances stratification and productivity inshore. The mean summer currents are westward 0.4 m s−1 along the shelf edge and weakly clockwise within False Bay. The marine climate is dominated by southeasterly winds that accelerate over the mountains south of Cape Town and fan out producing dry weather. Virtual buoy time series in December 2012–February 2013 exhibit weather-pulsed upwelling in early summer interspersed with quiescent spells in late summer. Intercomparisons between model, satellite and station data build confidence that coupled reanalyses yield opportunities to study air–sea interactions in coastal zones with complex topography. The 0.083∘ HYCOM reanalysis has 16 data points in False Bay, just adequate to resolve the coastal gradient and its impacts on ocean productivity.

A Penalized h-Likelihood Variable Selection Algorithm for Generalized Linear Regression Models with Random Effects

Reinforcement learning is one of the paradigms and methodologies of machine learning developed in the computational intelligence community. Reinforcement learning algorithms present a major challenge in complex dynamics recently. In the perspective of variable selection, we often come across situations where too many variables are included in the full model at the initial stage of modeling. Due to a high-dimensional and intractable integral of longitudinal data, likelihood inference is computationally challenging. It can be computationally difficult such as very slow convergence or even nonconvergence, for the computationally intensive methods. Recently, hierarchical likelihood (h-likelihood) plays an important role in inferences for models having unobservable or unobserved random variables. This paper focuses linear models with random effects in the mean structure and proposes a penalized h-likelihood algorithm which incorporates variable selection procedures in the setting of mean modeling via h-likelihood. The penalized h-likelihood method avoids the messy integration for the random effects and is computationally efficient. Furthermore, it demonstrates good performance in relevant-variable selection. Throughout theoretical analysis and simulations, it is confirmed that the penalized h-likelihood algorithm produces good fixed effect estimation results and can identify zero regression coefficients in modeling the mean structure.

The Deep Western Boundary Current and Adjacent Interior Circulation at 24°–30°N: Mean Structure and Mesoscale Variability

The mean North Atlantic Deep Water (NADW, 1000 < z < 5000 m) circulation and deep western boundary current (DWBC) variability offshore of Abaco, Bahamas, at 26.5°N are investigated from nearly two decades of velocity and hydrographic observations, and outputs from a 30-yr-long eddy-resolving global simulation. Observations at 26.5°N and Argo-derived geostrophic velocities show the presence of a mean Abaco Gyre spanning the NADW layer, consisting of a closed cyclonic circulation between approximately 24° and 30°N and 72° and 77°W. The southward-flowing portion of this gyre (the DWBC) is constrained to within ~150 km of the western boundary with a mean transport of ~30 Sv (1 Sv ≡ 106 m3 s−1). Offshore of the DWBC, the data show a consistent northward recirculation with net transports varying from 6.5 to 16 Sv. Current meter records spanning 2008–17 supported by the numerical simulation indicate that the DWBC transport variability is dominated by two distinct types of fluctuations: 1) periods of 250–280 days that occur regularly throughout the time series and 2) energetic oscillations with periods between 400 and 700 days that occur sporadically every 5–6 years and force the DWBC to meander far offshore for several months. The shorter-period variations are related to DWBC meandering caused by eddies propagating southward along the continental slope at 24°–30°N, while the longer-period oscillations appear to be related to large anticyclonic eddies that slowly propagate northwestward counter to the DWBC flow between ~20° and 26.5°N. Observational and theoretical evidence suggest that these two types of variability might be generated, respectively, by DWBC instability processes and Rossby waves reflecting from the western boundary.

Coastal gradients south of Cape Town: what insights can be gained from mesoscale reanalysis?

Mesoscale datasets are used to study coastal gradients in the marine climate and oceanography south of Cape Town. Building on past work, satellite and ocean/atmosphere reanalysis are used to gain new insights on the mean structure, circulation and meteorological features. HYCOM v3 hindcasts represent a coastward reduction of mixing that enhances stratification and productivity inshore. The mean summer currents are westward –.4 m/s along the shelf edge and weakly clockwise within False Bay. The marine climate is dominated by southeasterly winds that accelerate over the mountains south of Cape Town and fan out producing dry weather. Virtual buoy time series in Dec 2012–Feb 2013 exhibit weather-pulsed upwelling in early summer interspersed with quiescent spells in late summer. Intercomparisons between model, satellite and station data build confidence that coupled reanalyses yield opportunities to study air-sea interactions in coastal zones with complex topography. The 0.083° HYCOM reanalysis has 16 data points in the embayment south of Cape Town, just adequate to resolve the coastal gradient and its impacts on ocean productivity.

Recognition of Online Handwritten Bangla Characters Using Supervised and Unsupervised Learning Approaches

This paper explores the domain of online handwritten Bangla character recognition by stroke-based approach. The component strokes of a character sample are recognized firstly and then characters are constructed from the recognized strokes. In the current experiment, strokes are recognized by both supervised and unsupervised approaches. To estimate the features, images of all the component strokes are superimposed. A mean structure has been generated from this superimposed image. Euclidian distances between pixel points of a stroke sample and mean stroke structure are considered as features. For unsupervised approach, K-means clustering algorithm has been used whereas six popular classifiers have been used for supervised approach. The proposed feature vector has been evaluated on 10,000-character database and achieved 90.69% and 97.22% stroke recognition accuracy in unsupervised (using K-means clustering) and supervised way (using MLP [multilayer perceptron] classifier). This paper also discusses about merit and demerits of unsupervised and supervised classification approaches.