Identification of physical processes inherent in artificial neural network rainfall runoff models

The emergence of artificial neural network (ANN) technology has provided many promising results in the field of hydrology and water resources simulation. However, one of the major criticisms of ANN hydrologic models is that they do not consider/explain the underlying physical processes in a watershed, resulting in them being labelled as black‐box models. This paper discusses a research study conducted in order to examine whether or not the physical processes in a watershed are inherent in a trained ANN rainfall‐runoff model. The investigation is based on analysing definite statistical measures of strength of relationship between the disintegrated hidden neuron responses of an ANN model and its input variables, as well as various deterministic components of a conceptual rainfall‐runoff model. The approach is illustrated by presenting a case study for the Kentucky River watershed. The results suggest that the distributed structure of the ANN is able to capture certain physical behaviour of the rainfall‐runoff process. The results demonstrate that the hidden neurons in the ANN rainfall‐runoff model approximate various components of the hydrologic system, such as infiltration, base flow, and delayed and quick surface flow, etc., and represent the rising limb and different portions of the falling limb of a flow hydrograph. Copyright © 2004 John Wiley & Sons, Ltd.     

Transient phenomena in the energetic behind-the-limb solar flare of september 29, 1989

The powerful cosmic ray flare of Sept. 29, 1989 occurred behind the limb and was observed over a wide spectral range. The analysis of optical, radio, and other relevant data suggest two phases of energy release. After an impulsive phase a prolonged post eruption energy release occurred in an extended region of the corona following the eruption of a large coronal mass ejection (CME). This phase is responsible for numerous coronal and interplanetary phenomena including the ground-level increase of cosmic rays.     

Response of active tectonics on the alluvial Baghmati River, Himalayan foreland basin, eastern India

Active tectonics in a basin plays an important role in controlling a fluvial system through the change in channel slope. The Baghmati, an anabranching, foothills-fed river system, draining the plains of north Bihar in eastern India has responded to ongoing tectonic deformation in the basin. The relatively flat alluvial plains are traversed by several active subsurface faults, which divide the area in four tectonic blocks. Each tectonic block is characterized by association of fluvial anomalies viz. compressed meanders, knick point in longitudinal profiles, channel incision, anomalous sinuosity variations, sudden change in river flow direction, river flow against the local gradient and distribution of overbank flooding, lakes, and waterlogged area. Such fluvial anomalies have been identified on the repetitive satellite images and maps and interpreted through DEM and field observations to understand the nature of vertical movements in the area. The sub-surface faults in the Baghmati plains cut across the river channel and also run parallel which have allowed us to observe the effects of longitudinal and lateral tilting manifested in avulsions and morphological changes.     

Angular cross-correlation of galaxies: a probe of gravitational lensing by large-scale structure

The angular cross-correlation between two galaxy samples separated in redshift is shown to be a useful measure of weak lensing by large-scale structure. Angular correlations in faint galaxies arise as a result of spatial clustering of the galaxies as well as gravitational lensing by dark matter along the line of sight. The lensing contribution to the two-point autocorrelation function is typically small compared with the gravitational clustering. However, the cross-correlation between two galaxy samples is almost unaffected by gravitational clustering provided that their redshift distributions do not overlap. The cross-correlation is then induced by magnification bias resulting from lensing by large-scale structure. We compute the expected amplitude of the cross-correlation for popular theoretical models of structure formation. For two populations with mean redshifts of ≃0.3 and 1, we find a cross-correlation signal of ≃1 per cent on arcmin scales and ≃3 per cent on scales of a few arcsec. The dependence on the cosmological parameters Ω and Λ, the dark matter power spectrum and the bias factor of the foreground galaxy population is explored.     

Heat and mass transfer of a viscous heat generating fluid with Hall currents

A study of natural convection in hydrodynamic flows of a viscous heat generating fluid in the presence of Hall currents and variable suction has been carried out. The governing equations for the magnetohydrodynamic fluid flow and heat transfer are solved. The effects of Hall currentm and heat source parameter on the velocity and temperature distributions are discussed.     

The Non-Linear Stability of L4 in the Restricted Three-Body Problem when the Bigger Primary is a Triaxial Rigid Body

The non-linear stability of L ₄ in the restricted three-body problem has been studied when the bigger primary is a triaxial rigid body with its equatorial plane coincident with the plane of motion. It is found that L ₄ is stable in the range of linear stability except for three mass ratios:

Nonlinear coupled response of offshore tension leg platforms to regular wave forces

Among the compliant platforms, the tension leg platform (TLP) is a vertically moored structure with excess buoyancy. The TLP is designed to behave in the same way as any other moored structure in horizontal plane, at the same time inheriting the stiffness of a fixed platform in the vertical plane. Dynamic response analysis of a TLP to deterministic first order wave forces is presented, considering coupling between the degrees-of-freedom surge, sway, heave, roll, pitch and yaw. The analysis considers nonlinearities produced due to changes in cable tension and due to nonlinear hydrodynamic drag forces. The wave forces on the elements of the pontoon structure are calculated using Airy's wave theory and Morison's equation ignoring diffraction effects. The nonlinear equation of motion is solved in the time domain by Newmark's beta integration scheme. The effects of different parameters that influence the response of the TLP are then investigated.     

Second-order drift force response of offshore guyed towers

An iterative frequency domain method of analysis is presented for determining the response behaviour of Guyed Offshore Towers to low-frequency, second-order wave drift forces generated in a random sea environment. For the response analysis, the tower is idealized as a shear beam with a rotational spring at the bottom support. The guylines are replaced by a non-linear spring. The second-order drift force is considered to be proportional to the square of the wave elevation and is simulated using a drift force coefficient and the time history of a slowly varying wave envelope in random sea. The responses due to drift forces are obtained in frequency domain by incorporating the non-linearities produced due to non-linear guy lines. An example problem is solved under different random sea states to compare the response behaviour of the tower obtained by the second-order wave force, the first-order wave force and a combination of the two.     

Acoustic Mode Frequencies of the Sun During the Minimum Phase Between Solar Cycles 23 and 24

We investigate the spatial and temporal variations of the high-degree mode frequencies calculated over localized regions of the Sun during the extended minimum phase between solar cycles 23 and 24. The frequency shifts measured relative to the spatial average over the solar disk indicate that the correlation between the frequency shift and magnetic field strength during the low-activity phase is weak. The disk-averaged frequency shifts computed relative to a minimal activity period also reveal a moderate correlation with different activity indices, with a maximum linear correlation of about 72?%. From the investigation of the frequency shifts at different latitudinal bands, we do not find a consensus period for the onset of solar cycle 24. The frequency shifts corresponding to most of the latitudes in the northern hemisphere and 30° south of the equator indicate the minimum epoch to be February 2008, which is earlier than inferred from solar activity indices.