Comparison of Green and Lowtran Radiation Schemes with a Discrete Ordinate Method UV Model

Abstract— Using the same input parameters for the calculations, the Green and Lowtran codes for calculating UV irradiances were compared to the discrete ordinate method (DOM) model by Stamnes et al , which was used as a reference. The comparisons were performed at 305 and 380 nm for different ozone concentrations, aerosol optical depths and aerosol absorption characteristics. No obvious dependencies on optical depth, single scattering albedo or column ozone were found for the ratio of the Green and the Lowtran code to the DOM model. At 380 nm the Green model agrees with DOM within 10%, whereas the Lowtran code shows discrepancies of ±25%. At 305 nm the Green model shows 10% higher values than the DOM model at low zenith angles and up to 80% lower values for zenith angles between 60 and 80°. The Lowtran code shows 60% higher values than DOM at small zenith angles and 60% lower values at large zenith angles. When the spectra from each model were weighted with the erythemal action spectrum the Green model overestimated the DOM results by less than 10% for zenith angles less than 50°. Discrepancies between DOM and Lowtran models exceeded 10% except for a small range of zenith angles.     

Experimental two-photon,three-dimensional entanglement for quantum communication

Orbital angular momentum entangled photons emitted by a down-conversion source are in higher dimensional entangled states. Here we report the experimental confirmation by demonstrating a violation of a generalized Clauser-Horne-Shimony-Holt-type Bell inequality in three dimensions by more than 18 standard deviations. Higher dimensional entangled states allow the realization of new types of quantum communication protocols. They also provide a more secure quantum cryptography scheme. Therefore our experimental results are likely to have applications in future quantum communication technology.     

Concentration of higher dimensional entanglement: qutrits of photon orbital angular momentum

Enhancement of entanglement is necessary for most quantum communication protocols many of which are defined in Hilbert spaces larger than 2. In this work we present the experimental realization of entanglement concentration of orbital angular momentum entangled photons. We investigate the specific case of three dimensions and the possibility of generating different entangled states out of an initial state. The results presented here are of importance for pure states as well as for mixed states.     

Two-dimensional compressible viscous sink flows

A theoretical investigation of two-dimensional viscous compressible sink flow is presented. An analytic solution in parametric form is obtained for specific values of the ratio of specific heats. For the more general case, a number of asymptotic solutions were also obtained for different flow regimes including hypersonic, transonic, acoustic and low Reynolds number flow.     

On the polynomial approximation of boundary-layer flow profiles

The behaviour of the polynomial approximation to the boundary layer velocity profile is investigated. Various orders of polynomials and 4 different schemes of “reasonable” boundary conditions are examined for applicability as approximate solutions to the Blasius flow over a flat plate. A variational formulation, based upon the local potential is used to obtain the solution. It is found that the best and most consistent results are obtained when a symmetric distribution of auxiliary boundary conditions on the wall and outer edge of the boundary layer is used. The 6th order polynomial of this type, for example, already gives a wall friction factor within 0.5% of the exact solution.     

Editorial for ADAC issue 3 of volume 14 (2020)

Advances in Data Analysis and Classification -     

A comparative study on large scale kernelized support vector machines

Kernelized support vector machines (SVMs) belong to the most widely used classification methods. However, in contrast to linear SVMs, the computation time required to train such a machine becomes a bottleneck when facing large data sets. In order to mitigate this shortcoming of kernel SVMs, many approximate training algorithms were developed. While most of these methods claim to be much faster than the state-of-the-art solver LIBSVM, a thorough comparative study is missing. We aim to fill this gap. We choose several well-known approximate SVM solvers and compare their performance on a number of large benchmark data sets. Our focus is to analyze the trade-off between prediction error and runtime for different learning and accuracy parameter settings. This includes simple subsampling of the data, the poor-man’s approach to handling large scale problems. We employ model-based multi-objective optimization, which allows us to tune the parameters of learning machine and solver over the full range of accuracy/runtime trade-offs. We analyze (differences between) solvers by studying and comparing the Pareto fronts formed by the two objectives classification error and training time. Unsurprisingly, given more runtime most solvers are able to find more accurate solutions, i.e., achieve a higher prediction accuracy. It turns out that LIBSVM with subsampling of the data is a strong baseline. Some solvers systematically outperform others, which allows us to give concrete recommendations of when to use which solver.     

Benchmarking local classification methods

In recent years in the fields of statistics and machine learning an increasing amount of so called local classification methods has been developed. Local approaches to classification are not new, but have lately become popular. Well-known examples are the $k$ nearest neighbors method and classification trees. However, in most publications on this topic the term “local” is used without further explanation of its particular meaning. Only little is known about the properties of local methods and the types of classification problems for which they may be beneficial. We explain the basic principles and introduce the most important variants of local methods. To our knowledge there are very few extensive studies in the literature that compare several types of local methods and global methods across many data sets. In order to assess their performance we conduct a benchmark study on real-world and synthetic tasks. We cluster data sets and considered learning algorithms with regard to the obtained performance structures and try to relate our theoretical considerations and intuitions to these results. We also address some general issues of benchmark studies and cover some pitfalls, extensions and improvements.     

Sensitivity of erythemally effective UV irradiance and daily exposure to temporal variability in total ozone

The provision of information to the public about current levels of the erythemally effective UV radiation is an important issue in health care. The quality of promoted values is therefore of special importance. The atmospheric parameter which affects the erythemally effective UV radiation under clear sky most is the total ozone content of the atmosphere. In this paper we examined the sensitivity of the erythemally effective irradiance and daily radiant exposure to the temporal variability of total ozone on time scales from 1 to 15 days. The results show that the sensitivity is highest for the first 24 h. Larger time scales do not exhibit a similar influence. Total ozone measurements of the previous day may already cause uncertainties higher than 0.5 UV index (UVI) independent of the geolocation. For comparison, a temporal persistence of 15 days may cause uncertainties of 1.2 UVI at 50°N, 1 UVI at 30°S and less than 1 UVI at the equator. The results of this study allow finding the necessary temporal resolution of total ozone values when a certain accuracy for the UVI or for the purpose of sun protection is required. The results are compared with those of two preceding studies where we quantified the influence of measurement uncertainties and spatial total ozone variability to the erythemally effective irradiance at noon and to the daily dose. We conclude that temporal variability of total ozone is the most critical issue, but also measurement uncertainties do have a noticeable influence on the erythemally effective radiation.