PRISM and ENES: a European approach to Earth system modelling

Europe's widely distributed climate modelling expertise, now organized in the European Network for Earth System Modelling (ENES), is both a strength and a challenge. Recognizing this, the European Union's Program for Integrated Earth System Modelling (PRISM) infrastructure project aims at designing a flexible and friendly user environment to assemble, run and post-process Earth System models. PRISM was started in December 2001 with a duration of three years. This paper presents the major stages of PRISM, including: (1) the definition and promotion of scientific and technical standards to increase component modularity; (2) the development of an end-to-end software environment (graphical user interface, coupling and I/O system, diagnostics, visualization) to launch, monitor and analyse complex Earth system models built around state-of-art community component models (atmosphere, ocean, atmospheric chemistry, ocean bio-chemistry, sea-ice, land-surface); and (3) testing and quality standards to ensure high-performance computing performance on a variety of platforms. PRISM is emerging as a core strategic software infrastructure for building the European research area in Earth system sciences. Copyright (c) 2005 John Wiley & Sons, Ltd.

Calibration and use of a cesium-iodide prism in the infrared

With a cesium-iodide prism the long wavelength range of an infrared spectrometer may be extended to 55µ The use of such a prism, the choice of optical system, and the problems of stray radiation are all discussed. Accurate data are assembled for calibration in this region, and sample calibration traces are shown. A simple gas absorption cell is described for use at long wavelengths.

P-Prism: a computationally efficient approach to scaling up classification rule induction

Top Down Induction of Decision Trees (TDIDT) is the most commonly used method of constructing a model from a dataset in the form of classification rules to classify previously unseen data. Alternative algorithms have been developed such as the Prism algorithm. Prism constructs modular rules which produce qualitatively better rules than rules induced by TDIDT. However, along with the increasing size of databases, many existing rule learning algorithms have proved to be computational expensive on large datasets. To tackle the problem of scalability, parallel classification rule induction algorithms have been introduced. As TDIDT is the most popular classifier, even though there are strongly competitive alternative algorithms, most parallel approaches to inducing classification rules are based on TDIDT. In this paper we describe work on a distributed classifier that induces classification rules in a parallel manner based on Prism.

Random prism: an alternative to random forests

Ensemble learning techniques generate multiple classifiers, so called base classifiers, whose combined classification results are used in order to increase the overall classification accuracy. In most ensemble classifiers the base classifiers are based on the Top Down Induction of Decision Trees (TDIDT) approach. However, an alternative approach for the induction of rule based classifiers is the Prism family of algorithms. Prism algorithms produce modular classification rules that do not necessarily fit into a decision tree structure. Prism classification rulesets achieve a comparable and sometimes higher classification accuracy compared with decision tree classifiers, if the data is noisy and large. Yet Prism still suffers from overfitting on noisy and large datasets. In practice ensemble techniques tend to reduce the overfitting, however there exists no ensemble learner for modular classification rule inducers such as the Prism family of algorithms. This article describes the first development of an ensemble learner based on the Prism family of algorithms in order to enhance Prism’s classification accuracy by reducing overfitting.

Parallel random prism: a computationally efficient ensemble learner for classification

Generally classifiers tend to overfit if there is noise in the training data or there are missing values. Ensemble learning methods are often used to improve a classifier's classification accuracy. Most ensemble learning approaches aim to improve the classification accuracy of decision trees. However, alternative classifiers to decision trees exist. The recently developed Random Prism ensemble learner for classification aims to improve an alternative classification rule induction approach, the Prism family of algorithms, which addresses some of the limitations of decision trees. However, Random Prism suffers like any ensemble learner from a high computational overhead due to replication of the data and the induction of multiple base classifiers. Hence even modest sized datasets may impose a computational challenge to ensemble learners such as Random Prism. Parallelism is often used to scale up algorithms to deal with large datasets. This paper investigates parallelisation for Random Prism, implements a prototype and evaluates it empirically using a Hadoop computing cluster.

Random Prism: a noise-tolerant alternative to Random Forests

Ensemble learning can be used to increase the overall classification accuracy of a classifier by generating multiple base classifiers and combining their classification results. A frequently used family of base classifiers for ensemble learning are decision trees. However, alternative approaches can potentially be used, such as the Prism family of algorithms that also induces classification rules. Compared with decision trees, Prism algorithms generate modular classification rules that cannot necessarily be represented in the form of a decision tree. Prism algorithms produce a similar classification accuracy compared with decision trees. However, in some cases, for example, if there is noise in the training and test data, Prism algorithms can outperform decision trees by achieving a higher classification accuracy. However, Prism still tends to overfit on noisy data; hence, ensemble learners have been adopted in this work to reduce the overfitting. This paper describes the development of an ensemble learner using a member of the Prism family as the base classifier to reduce the overfitting of Prism algorithms on noisy datasets. The developed ensemble classifier is compared with a stand-alone Prism classifier in terms of classification accuracy and resistance to noise.

In-situ spatio-temporal measurements of the detailed azimuthal substructure of the substorm current wedge

The substorm current wedge (SCW) is a fundamental component of geomagnetic substorms. Models tend to describe the SCW as a simple line current flowing into the ionosphere towards dawn and out of the ionosphere towards dusk, linked by a westward electrojet. We use multi-spacecraft observations from perigee passes of the Cluster 1 and 4 spacecraft during a substorm on 15 Jan 2010, in conjunction with ground-based observations, to examine the spatial structuring and temporal variability of the SCW. At this time, the spacecraft travelled east-west azimuthally above the auroral region. We show that the SCW has significant azimuthal sub-structure on scales of 100~km at altitudes of 4,000-7,000~km. We identify 26 individual current sheets in the Cluster 4 data and 34 individual current sheets in the Cluster 1 data, with Cluster 1 passing through the SCW 120-240~s after Cluster 4 at 1,300-2,000~km higher altitude. Both spacecraft observed large-scale regions of net upward and downward field-aligned current, consistent with the large-scale characteristics of the SCW, although sheets of oppositely directed currents were observed within both regions. We show that the majority of these current sheets were closely aligned to a north-south direction, in contrast to the expected east-west orientation of the pre-onset aurora. Comparing our results with observations of the field-aligned current associated with bursty bulk flows (BBFs) we conclude that significant questions remain for the explanation of SCW structuring by BBF driven ``wedgelets". Our results therefore represent constraints on future modelling and theoretical frameworks on the generation of the SCW.

A scalable expressive ensemble learning using Random Prism: a MapReduce approach

The induction of classification rules from previously unseen examples is one of the most important data mining tasks in science as well as commercial applications. In order to reduce the influence of noise in the data, ensemble learners are often applied. However, most ensemble learners are based on decision tree classifiers which are affected by noise. The Random Prism classifier has recently been proposed as an alternative to the popular Random Forests classifier, which is based on decision trees. Random Prism is based on the Prism family of algorithms, which is more robust to noise. However, like most ensemble classification approaches, Random Prism also does not scale well on large training data. This paper presents a thorough discussion of Random Prism and a recently proposed parallel version of it called Parallel Random Prism. Parallel Random Prism is based on the MapReduce programming paradigm. The paper provides, for the first time, novel theoretical analysis of the proposed technique and in-depth experimental study that show that Parallel Random Prism scales well on a large number of training examples, a large number of data features and a large number of processors. Expressiveness of decision rules that our technique produces makes it a natural choice for Big Data applications where informed decision making increases the user’s trust in the system.

The salt wedge position in a bar-blocked estuary subject to pulsed inflows

A series of laboratory experiments were carried out to investigate the response of a bar-blocked, saltwedge estuary to the imposition of both steady freshwater inflows and transient inflows that simulate storm events in the catchment area or the regular water releases from upstream reservoirs. The trapped salt water forms a wedge within the estuary, which migrates downstream under the influence of the freshwater inflow. The experiments show that the wedge migration occurs in two stages, namely (i) an initial phase characterized by intense shear-induced mixing at the nose of the wedge, followed by (ii) a relatively quiescent phase with significantly reduced mixing in which the wedge migrates more slowly downstream.

Provided that the transition time t_T between these two regimes satisfies t_T>g′h⁴L/q³α, as was the case for all our experiments and is likely to be the case for most estuaries, then the transition occurs at time t_T=1.2(gα³L⁶/g′³q²)^1/6, where g′=gΔρ/ρ0 is the reduced gravity, g the acceleration due to gravity, Δρ the density excess of the saline water over the density ρ₀ of the freshwater, q the river inflow rate per unit width, and L and α are the length and bottom slope of the estuary, respectively.

A simple model, based on conversion of the kinetic energy of the freshwater inflow into potential energy to mix the salt layer, was developed to predict the displacement xw over time t of the saltwedge nose from its initial position. For continuous inflows subject to t<t_T, the model predicts the saltwedge displacement as x_w/h=1.1 (t/τ)^1/3, where the normalizing length and time scales are h=(q²/g)^1/3 and τ=g′α²h4L/q³, respectively. For continuous inflows subject to t>tT, the model predicts the displacement as xw/h=0.45N1/6(t/τ)1/6/α, where N=q²/g′h²L is a non-dimensional number for the problem. This model shows very good agreement with the experiments. For repeated, pulsed discharges subject to t<t_T, the saltwedge displacement is given by (xw/h)3−(x0/h)(xw/h)2=1.3t/τ, where x₀ is the initial displacement following one discharge event but prior to the next event. For pulsed discharges subject to t>t_T, the displacement is given by (xw/h)⁶−(x0/h)(xw/h)⁵=0.008N(t/τ)/α⁶. This model shows very good agreement with the experiments for the initial discharge event but does systematically underestimate the wedge position for the subsequent pulses. However, the positional error is less than 15%.

Aspects of the ecology of the Wedge-tailed Eagle, Aquila audax, in southern Victoria

This study investigated the nest site selection characteristics and diet of the Wedge-tailed Eagle, Aquila audax, in southern Victoria. It was found that sites where local topography afforded some protection from adverse weather and where the nest tree was live were most commonly selected. Rabbits were found to the major prey item of this eagle.

Social disadvantage, bone mineral density and vertebral wedge deformities in the Tasmanian Older Adult Cohort

Summary The relationship between social disadvantage and bone mineral density (BMD) is complex and remains unclear; furthermore, little is known of the relationship with vertebral deformities. We observed social disadvantage to be associated with BMD for females, independent of body mass index (BMI). A lower prevalence of vertebral deformities was observed for disadvantaged males.

Introduction The relationship between social disadvantage and BMD appears complex and remains unclear, and little is known about the association between social disadvantage and vertebral wedge deformities. We examined the relationship between social disadvantage, BMD and wedge deformities in older adults from the Tasmanian Older Adult Cohort.

Methods BMD and wedge deformities were measured by dual-energy X-ray absorptiometry and associations with extreme social disadvantage was examined in 1,074 randomly recruited population-based adults (51 % female). Socioeconomic status was assessed by Socio-economic Indexes for Areas values derived from residential addresses using Australian Bureau of Statistics 2001 census data. Lifestyle variables were collected by self-report. Regression models were adjusted for age, BMI, dietary calcium, serum vitamin D (25(OH)D), smoking, alcohol, physical inactivity, calcium/vitamin D supplements, glucocorticoids and hormone therapy (females only).

Results Compared with other males, socially disadvantaged males were older (65.9 years versus 61.9 years, p = 0.008) and consumed lower dietary calcium and alcohol (both p ≤ 0.03). Socially disadvantaged females had greater BMI (29.9 ± 5.9 versus 27.6 ± 5.3, p = 0.002) and consumed less alcohol (p = 0.003) compared with other females. Socially disadvantaged males had fewer wedge deformities compared with other males (33.3 % versus 45.4 %, p = 0.05). After adjustment, social disadvantage was negatively associated with hip BMD for females (p = 0.02), but not for males (p = 0.70), and showed a trend for fewer wedge deformities for males (p = 0.06) but no association for females (p = 0.85).

Conclusions Social disadvantage appears to be associated with BMD for females, independent of BMI and other osteoporosis risk factors. A lower prevalence of vertebral deformities was observed for males of extreme social disadvantage. Further research is required to elucidate potential mechanisms for these associations.