Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning

The evaluation of surface water resources is a necessary input to solving water management problems. Neural network models have been trained to predict monthly runoff for the Tirso basin, located in Sardinia (Italy) at the S. Chiara section. Monthly time series data were available for 69 years and are characterized by non-stationarity and seasonal irregularity, which is typical of a Mediterranean weather regime. This paper investigates the effects of data preprocessing on model performance using continuous and discrete wavelet transforms and data partitioning. The results showed that networks trained with pre-processed data performed better than networks trained on undecomposed, noisy raw signals. In particular, the best results were obtained using the data partitioning technique.     

Nassarius corniculus (Olivi, 1792) (Caenogastropoda: Nassariidae): a model of environmental complexity of Italian brackish and marine habitats

Strong environmental selection may give rise to complex species and/or camouflage the presence of sibling species in brackish waters. The prosobranch gastropod Nassarius corniculus (Olivi, 1792) is a common scavenger widely distributed in Mediterranean brackish and marine areas. We examined the morphological and molecular features of N. corniculus populations and related these features to the environmental complexity of brackish habitats. Morphological data revealed differences in radular features between specimens collected from two Mediterranean lagoons (Lake Faro, Messina, Sicily and Lake Fusaro, Gulf of Gaeta). Analysis of with the mitochondrial marker mt 16S rRNA revealed genetic variability within the Lake Faro population but did not show the different geographic structure between the two lagoons indicated by morphological data. We also predicted the gene flow of N. corniculus by comparing populations of two marine areas (Faro Sea and Ischia).     

Precipitable water evaluations from infrared sun‐photometric measurements analyzed using the atmospheric hygrometry technique

Measurements of direct solar irradiance were taken employing 4 different sun‐photometers at near infrared wavelengths, suitable for use in atmospheric hygrometry. This technique utilising a set of spectral ratios, in and out of selected water vapour absorption bands, was applied to the measurements to obtain accurate evaluations of precipitable water. For all the hygrometric ratios given by the 4 sun‐photometers used at the 3 stations of Sagres, Monchique and Mt. Foia, during the CLEARCOLUMN experiment, we determined the calibration curves by correcting them for the Rayleigh scattering effects and, then, plotting the natural logarithms of such corrected ratios versus the square root of the water vapour mass present along the atmospheric slant path. The regression lines drawn for the various scatter diagrams were estimated to give evaluations of precipitable water with an uncertainty of less than 5%, 3% and 10% at the 3 stations, respectively. The calibration curves of the sun‐photometer located at the Sagres station were determined using the precipitable water evaluations obtained from the local radiosounding measurements taken on 5 clear‐sky days. Those of the sun‐photometers used at the Monchique and Mt. Foia stations were instead determined through intercomparison between subsets of measurements simultaneously taken with various instruments at Sagres and Mt. Foia. Using these calibration curves, we examined all the field measurements determining the time‐patterns of precipitable water at the 3 stations. During the period from 16 June to 25 July 1997, precipitable water was found to vary between 1.1 and 3.7 g cm⁻² at the Sagres station (with an accuracy within ±13%), between 1.0 and 2.8 g cm⁻² at Monchique (±11%) and between 0.8 and 3.0 g cm⁻² at the top of Mt. Foia (±26%).     

Analytical development of the lunisolar disturbing function and the critical inclination secular resonance

We provide a detailed derivation of the analytical expansion of the lunar and solar disturbing functions. Although there exist several papers on this topic, many derivations contain mistakes in the final expansion or rather (just) in the proof, thereby necessitating a recasting and correction of the original derivation. In this work, we provide a self-consistent and definite form of the lunisolar expansion. We start with Kaula’s expansion of the disturbing function in terms of the equatorial elements of both the perturbed and perturbing bodies. Then we give a detailed proof of Lane’s expansion, in which the elements of the Moon are referred to the ecliptic plane. Using this approach the inclination of the Moon becomes nearly constant, while the argument of perihelion, the longitude of the ascending node, and the mean anomaly vary linearly with time. We make a comparison between the different expansions and we profit from such discussion to point out some mistakes in the existing literature, which might compromise the correctness of the results. As an application, we analyze the long-term motion of the highly elliptical and critically-inclined Molniya orbits subject to quadrupolar gravitational interactions. The analytical expansions presented herein are very powerful with respect to dynamical studies based on Cartesian equations, because they quickly allow for a more holistic and intuitively understandable picture of the dynamics.     

Four Classical Methods for Determining Planetary Elliptic Elements: A Comparison

The discovery of the asteroid Ceres by Piazzi in 1801 motivated the development of a mathematical technique proposed by Gauss, (Theory of the Motion of the Heavenly Bodies Moving about the Sun in Conic Sections, 1963) which allows to recover the orbit of a celestial body starting from a minimum of three observations. Here we compare the method proposed by Gauss (Theory of the Motion of the Heavenly Bodies Moving about the Sun in Conic Sections, New York, 1963) with the techniques (based on three observations) developed by Laplace (Collected Works 10, 93–146, 1780) and by Mossotti (Memoria Postuma, 1866). We also consider another method developed by Mossotti (Nuova analisi del problema di determinare le orbite dei corpi celesti, 1816–1818), based on four observations. We provide a theoretical and numerical comparison among the different procedures. As an application, we consider the computation of the orbit of the asteroid Juno.     

Estimate of the Transition Value of Librational Invariant Curves

We investigate the break-down threshold of librational invariant curves. As a model problem, we consider a variant of a mapping introduced by M. Hénon, which well describes the dynamics of librational motions surrounding a stable invariant point. We verify in concrete examples the applicability of Greene's method, by computing the instability transition values of a sequence of periodic orbits approaching an invariant curve with fixed noble frequency. However, this method requires the knowledge of the location of the periodic orbits within a very good approximation. This task appears to be difficult to realize for a libration regime, due to the different topology of the phase space. To compute the break-down threshold, we tried an alternative method very easy to implement, based on the computation of the fast Lyapunov indicators and frequency analysis. Such technique does not require the knowledge of the periodic orbits, but again, it appears very difficult to have a precision better than Greene's method for the computation of the critical parameter.     

Some properties of the dumbbell satellite attitude dynamics

The dumbbell satellite is a simple structure consisting of two point masses connected by a massless rod. We assume that it moves around the planet whose gravity field is approximated by the field of the attracting center. The distance between the point masses is assumed to be much smaller than the distance between the satellite’s center of mass and the attracting center, so that we can neglect the influence of the attitude dynamics on the motion of the center of mass and treat it as an unperturbed Keplerian one. Our aim is to study the satellite’s attitude dynamics. When the center of mass moves on a circular orbit, one can find a stable relative equilibrium in which the satellite is permanently elongated along the line joining the center of mass with the attracting center (the so called local vertical). In case of elliptic orbits, there are no stable equilibrium positions even for small values of the eccentricity. However, planar periodic motions are determined, where the satellite oscillates around the local vertical in such a way that the point masses do not leave the orbital plane. We prove analytically that these planar periodic motions are unstable with respect to out-of-plane perturbations (a result known from numerical investigations cf. Beletsky and Levin Adv Astronaut Sci 83, 1993). We provide also both analytical and numerical evidences of the existence of stable spatial periodic motions.     

Measures of basins of attraction in spin-orbit dynamics

We consider a dissipative spin-orbit model where it is assumed that the orbit of the satellite is Keplerian, the obliquity is zero, and the dissipative effects depend linearly on the relative angular velocity. The measure of the basins of attraction associated to periodic and quasi-periodic attractors is numerically investigated. The results depend on the interaction among the physically relevant parameters, namely, the orbital eccentricity, the equatorial oblateness and the dissipative constant. In particular, it appears that, for astronomically relevant parameter values, for low eccentricities (as in the Moon’s case) about 96% of the initial data belong to the basin of attraction of the 1/1 spin-orbit resonance; for larger values of the eccentricities higher order spin-orbit resonances and quasi-periodic attractors become dominant providing a mechanism for explaining the observed state of Mercury into the 3/2 resonance.     

Non‐impact origin of the crater‐like structures in the Gilf Kebir area (Egypt): Implications for the geology of eastern Sahara

Abstract— Several small crater‐like structures occur in Gilf Kebir region (SW Egypt). It has been previously suggested that they could be the result of meteoritic impacts. Here we outline the results of our geological and geophysical survey in the area. The proposed impact origin for these structures is not supported by our observations and analyses, and we suggest an alternative interpretation. The crater‐like structures in Gilf Kebir area are likely related to endogenic processes typical of hydrothermal vent complexes in volcanic areas which may reflect the emplacement of subvolcanic intrusives.