Harm J. J. Jonker

Size Statistics of Cumulus Cloud Populations in Large-Eddy Simulations

Abstract Cloud size distributions of shallow cumulus cloud populations are calculated using the large-eddy simulation (LES) approach. A range of different cases is simulated, and the results are compared to observations of real cloud populations. Accordingly, the same algorithm is applied as in observational studies using high-altitude photography or remote sensing. The cloud size density of the simulated cloud populations is described well by a power law at the smaller sizes. This scaling covers roughly one order of magnitude of cloud sizes, with a power-law exponent of −1.70, which is comparable to exponents found in observational studies. A sensitivity test for the resolution suggests that the scaling continues at sizes smaller than the standard grid spacing. In contrast, on the other end, the scaling region is bounded by a distinct scale break. When the cloud size is nondimensionalized by the scale break size, the cloud size densities of all cases collapse. This corroborates the idea of a universal de...

Entrainment process of carbon dioxide in the atmospheric boundary layer

Received 2 March 2004; revised 7 June 2004; accepted 23 June 2004; published 23 September 2004. [1] Aircraft and surface measurements of turbulent thermodynamic variables and carbon dioxide (CO2) were taken above a grassland in a convective atmospheric boundary layer. The observations were analyzed to assess the importance of the entrainment process for the distribution and evolution of carbon dioxide in the boundary layer. From the observations we were able to estimate the vertical profiles of the fluxes, the correlation coefficients, and the skewness of the virtual potential temperature, the specific humidity, and the carbon dioxide. These profiles indicate that important entrainment events occurred during the observed period. The data were also used to estimate the budgets for heat, moisture, and carbon dioxide. By studying this observational data we find that the entrainment of air parcels containing lower concentrations of water vapor and carbon dioxide significantly dries and dilutes the concentration of these two constituents in the boundary layer. This process is particularly important in the morning hours which are characterized by a rapidly growing boundary layer. The observations show that the CO2 concentration in the boundary layer is reduced much more effectively by the ventilation with entrained air than by CO2 uptake by the vegetation. We quantify this effect by calculating the ratio of the entrainment flux of CO2 to the surface flux of CO2(bc = � (wc)e/(wc)o). A value of bc equal to 2.9 is estimated at around 1300 UTC from the vertical profile of the carbon dioxide flux. We corroborate this observational evidence by reproducing the observed situation using a mixed layer model. The mixed layer model also yields the variation in time of bc. During the morning the ventilation process is more important than the CO2 uptake by the vegetation (bc > 1), whereas in the afternoon the assimilation by grass at the surface becomes the dominant process (bc < 1). This research points out the relevance of the entrainment process on the budget of carbon dioxide in the lower troposphere and the relevance of boundary layer dynamics in controlling the diurnal variation of carbon dioxide. INDEX TERMS: 0315 Atmospheric Composition and Structure: Biosphere/atmosphere interactions; 0322 Atmospheric Composition and Structure: Constituent sources and sinks; 0368 Atmospheric Composition and Structure: Troposphere— constituent transport and chemistry; KEYWORDS: entrainment carbon dioxide, mixed layer model Citation: de Arellano, J. V.-G., B. Gioli, F. Miglietta, H. J. J. Jonker, H. K. Baltink, R. W. A. Hutjes, and A. A. M. Holtslag (2004), Entrainment process of carbon dioxide in the atmospheric boundary layer, J. Geophys. Res., 109, D18110,

Comparison of scaling parameters from spatial and temporal distributions of cloud properties

The most common method to evaluate the quality of cloud parameter retrievals from satellite data is comparison with time series of ground-based measurements. For highly variable cloud parameters, such as liquid water path (LWP), however, a direct comparison of spatial and temporal distributions has limited value. It is questionable which period in the time series is representative of an area in the satellite image. A necessary but not sufficient boundary condition for statistical analysis is that the variability of both subsets is similar. In this paper we study the variability in terms of scaling properties of both spatial distributions and timeseries of LWP and focus on the relation between them. The time series are obtained from ground-based microwave radiometer measurements at 1 Hz. The spatial distributions are derived from Advanced Very High Resolution Radiometer 0.6 μm radiances. It is shown that the scaling properties of both distributions are similar for a stratocumulus field, which exhibits scale invariance, and for a cumulus field for which scales of preference were identified. A double logarithmic representation was appropriate to obtain the spectral exponent. A log-linear representation was appropriate to identify a scale of preference. In the cumulus case it was possible to calculate a feasible conversion factor to map variability in time to variability in space. This enables an estimation of the spatial variability at very high resolutions.

Feature tracking in VR for cumulus cloud life-cycle studies

Feature tracking in large data sets is traditionally an off-line, batch processing operation while virtual reality typically focuses on highly interactive tasks and applications. This paper presents an approach that uses a combination of off-line preprocessing and interactive visualization in VR to simplify and speed up the identification of interesting features for further study. We couch the discussion in terms of our collaborative research on using virtual reality for cumulus cloud life-cycle studies, where selecting suitable clouds for study is simple for the skilled observer but difficult to formalize. The preprocessing involves identifying individual clouds within the data set through a 4D connected components algorithm, and then saving isosurface, bounding box, and volume information. This information is then interactively visualized in our VR Cloud Explorer with various tools and information displays to identify the most interesting clouds. In a small pilot study, reasonable performance, both in the preprocessing phase and the visualization phase, has been measured.

A model of the human smooth pursuit system based on an unsupervised adaptive controller

A first attempt was made, on the basis of Pavels proposal (1990), to integrate a predictive mechanism into a model of the human smooth pursuit system. The predictive mechanism contained an adaptive filter based on the LMS algorithm. Smooth pursuit simulations were made of a large variety of target movements. The model provides a fairly good qualitative and mostly also a fairly good quantitative description of human tracking of the various stimuli. However, when the model was applied to the tracking of sinusoidal target movements with frequencies higher than about 1 Hz, it performed even better than the human smooth pursuit system.

Self-Organizing Neural Mechanisms Possibly Responsible for Muscle Coordination

In this chapter we will discuss certain mechanisms that may play an important role in the organization and the learning of motor control. These mechanisms are based on the generally accepted notion that the strength of synaptic connections between neurons in the central nervous system can be modified under the influence of synchronous activity of these neurons. Thus, neural signals not only contain current information, they also contribute to the long-term organization of neural structures.

Unsupervised dynamic learning in layered neural networks

The authors consider a stochastic two-layer neural network of binary neurons in which the connections between the layers are updated according to the Hebb rule, whereas the lateral connections in the output layer are modified according to an anti-Hebb rule. In equilibrium the output overlap is found to be a linear transformation of the input overlap. Next they extend the model by considering learning as a dynamic process, which means that synaptic efficacies as well as neuronal states may vary in time. Despite the coupling of these two variables, they show that in this particular model the behaviour can be well analysed. It turns out that the network filters the information available at the input in such a way that important components of the input data can pass through, whereas components with a low information content are suppressed.

On the function and development of spatial structure in layered neural networks

The authors study the relation between spatial structure and information processing properties of layered Ising spin neural networks with lateral interactions. The interactions between layers are given by the Hebb rule, the interactions within layers by the so-called anti-Hebb rule. Secondly they study the development of spatial structure in such networks as the result of an unsupervized learning process (now both neurons and interactions are dynamical variables). By calculating the spectrum of the output covariance matrix as a function of the spectrum of the input covariance matrix, they show how the spatial characteristics of the input signals are reflected in the information processing properties of the equilibrated system.

NUMERICAL SIMULATION OF FLOW AND HEAT TRANSFER IN MULTIPLE IMPINGING JETS

This paper focuses on the effect of nozzle arrangement on the heat transfer of multiple impinging jets. Two different geometrical arrangements with an equal number of nozzles are studied, a square set-up (3 × 3 regularly spaced jets) and a circular set-up (eight jets surrounding a central jet). The predictions with two turbulence models are compared, the standard k – ∈ model with wall functions and the υ 2 ¯ − f model ( Durbin (1991) ) with integration to the wall. Quite surprisingly we find for one of the arrangements – the square set-up – an asymmetric flow field, even though the geometry, as well as the initial and boundary conditions are perfectly symmetric. Both turbulence models predict this peculiar broken symmetry. The other arrangement – the circular set-up – behaves more in line of expectation: both turbulence models predict a flow field that reflects all the symmetries of the physical situation. Interestingly, the differences between the two arrangements are quite small, but the corresponding flow fields clearly exhibit marked differences. Although both turbulence models agree qualitatively quite well on the mean features of the flow field for the two different set-ups, there are significant quantitative differences between the models regarding the turbulent kinetic energy and heat transfer. In particular the peak values of both the kinetic energy and the Nusselt number are quite different. Finally, we make a detailed comparison between the resulting heat transfer of the different arrangements. In the circular set-up the individual jets have a similar heat transfer, except for the central jet which is stabilised by the outer jets. In the square set-up, the difference between the jets is much more pronounced: due to the aforementioned asymmetry, some jets impinge vigorously, with a high heat transfer at the expense of the other, deflected, jets.

Learning in neural networks by eliminating frustrated bonds

The authors study neural network models in which the initial interaction matrix elements are drawn from an arbitrary probability distribution and in which patterns are subsequently stored by eliminating the frustrated bonds, generalizing a proposal by Kinzel (1985). They show that the optimal choice for the a priori distribution corresponds to choosing uniform ferromagnetic initial interactions. For the optimal model they study analytically the dynamical behaviour, the equilibrium properties, the sizes of the domains of attraction and a number of information-theoretical performance measures.