Search

Gravity inversion is a commonly applied data analysis technique in the field of geophysics. While machine learning methods have previously been explored for the problem of gravity inversion, these are deterministic approaches returning a single solution deemed most appropriate by the algorithm. The method presented here takes a different approach, where gravity inversion is reformulated as a Bayesian parameter inference problem. Samples from the posterior probability distribution of source model parameters are obtained via the implementation of a generative neural network architecture known as Normalising Flows. Due to its probabilistic nature, this framework provides the user with a range of source parameters and uncertainties instead of a single solution, and is inherently robust against instrumental noise. The performance of the Normalising Flow is compared to that of an established Bayesian method called Nested Sampling. It is shown that the new method returns results with comparable accuracy 200 times faster than standard sampling methods, which makes Normalising Flows a suitable method for real-time inversion in the field. When applied to data sets with high dimensionality, standard sampling methods can become impractical due to long computation times. It is shown that inversion using Normalising Flows remains tractable even at 512 dimensions and once the network is trained, the results can be obtained in $O(10)$ seconds., Comment: 14 pages, 6 figures, to be published in the Computers & Geosciences Journal

In the present paper, we use Si IV 1393.755 \AA\ spectral line observed by the Interface Region Imaging Spectrograph (IRIS) in the quiet-Sun to determine physical nature of the solar transition region (TR) oscillations. We analyze the properties of these oscillations using wavelet tools (e.g., power, cross-power, coherence, and phase difference) along with the stringent noise model (i.e., power-law + constant). We estimate the period of the intensity and Doppler velocity oscillations at each chosen location in the quiet-Sun (QS) and quantify the distribution of the statistically significant power and associated periods in one bright and two dark regions. In the bright TR region, the mean periods in intensity and velocity are 7 min, and 8 min respectively. In the dark region, the mean periods in intensity and velocity are 7 min, and 5.4 min respectively. We also estimate the phase difference between the intensity and Doppler velocity oscillations at each location. The statistical distribution of phase difference is estimated, which peaks at -119\degree $\pm$ 13\degree, 33\degree $\pm$ 10\degree, 102\degree $\pm$ 10\degree\ in the bright region, while at -153\degree $\pm$ 13\degree, 6\degree $\pm$ 20\degree, 151\degree $\pm$ 10\degree\ in the dark region. The statistical distribution reveals that the oscillations are caused by propagating slow magnetoacoustic waves encountered with the TR. Some of these locations may also be associated with the standing slow waves. Even, in the given time domain, several locations exhibit presence of both propagating and standing oscillations at different frequencies., Comment: 16 pages, 11 figures

This paper describes a temperature sensitivity or thermal sag measurement of a geometric anti-spring based micro-electromechanical system (MEMS) gravimeter (Wee-g). The Wee-g MEMS gravimeter is currently fabricated on a (100) silicon wafer using standard micro-nano fabrication techniques. The thermal behavior of silicon indicates that the Young's modulus of silicon decreases with an increase in temperature (around 63 ppm/K). This leads to a softening of the silicon material resulting in the proof mass (PM) displacing (or sagging) under the influence of an increasing temperature. It results in the change on the measured gravity and it is expressed as a temperature sensitivity in terms of change in gravity per degree temperature. The temperature sensitivity for the silicon based MEMS gravimeter is found to be 60.14-64.87 muGal/mK, 61.76 muGal/mK and 62.76 muGal/mK for experimental, finite element analysis (FEA) simulation and analytical calculations respectively. It suggests that temperature sensitivity is depended on material properties used to fabricate the MEMS devices. In this paper the experimental measurements of thermal sag are presented, along with analytical calculations and simulations of the effect using FEA (finite element analysis). The bespoke optical measurement system to quantify the thermal sag is also described. The results presented are an essential step towards the development of temperature insensitive MEMS gravimeters., Comment: 6 pages, 3 figures

We analyse the phase shifts of standing, slow magnetohydrodynamic (MHD) waves in solar coronal loops using a linear MHD model taking into account the role of thermal conductivity, compressive viscosity, radiative losses, and heating-cooling misbalance. We estimate the phase shifts in time and space of density and temperature perturbations with respect to velocity perturbations and also calculate the phase difference between density and temperature perturbations. The overall significance of compressive viscosity is found to be negligible for most of the loops considered in the study. For loops with high background density and/or low background temperature, the role of radiative losses (with heating-cooling misbalance) is found to be more significant. Also the effect of heating-cooling misbalance with a temperature- and density-dependent heating function is found to be more significant in the case of longer loop lengths ($L=500$\, Mm). We derived a general expression for the polytropic index [$\gamma_{\rm eff}$] and found that under linear MHD the effect of compressive viscosity on polytropic index is negligible. The radiative losses with constant heating lead to a monotonic increase of $\gamma_{\rm eff}$ with increasing density whereas the consideration of an assumed heating function [$H(\rho,T) \propto \rho^{a}T^{b}$, where $a=-0.5$ and $b=-3$] makes the $\gamma_{\rm eff}$ peak at a certain loop density. We also explored the role of different heating functions by varying the free parameters $a$ and $b$ for a fixed loop of $\rho_0 = 10^{-11}$\, kg $\text{m}^{-3}$, $T_0 = 6.3$\, MK and loop length $L= 180$\, Mm. We find that the consideration of different heating functions [$H(\rho,T)$] leads to a significant variation in the phase difference between density and temperature perturbations; however, the polytropic index remains close to a value of 1.66., Comment: Solar Physics (in press)

We study phase shifts of the propagating slow waves in coronal loops invoking the effects of thermal conductivity, compressive viscosity, radiative losses and heating-cooling imbalance. We derive a general dispersion relation and solve it to determine phase shifts of density and temperature perturbations relative to the velocity and their dependence on equilibrium parameters ($\rho_0$, $T_0$). We estimate phase difference ($\Delta \phi$) between density and temperature perturbations and its dependence on $\rho_0$ and $T_0$. The effect of viscosity on the phase shifts was found negligible. The role of radiative losses along with h/c imbalance for chosen specific heating function ($H(\rho, T) \propto \rho^{-0.5} T^{-3}$) in determining phase shifts, is found to be significant for the high density and low temperature loops. The h/c imbalance can increase the phase difference ($\Delta \phi \approx 140^\circ$) for low temperature loops compared to the constant heating case ($\Delta \phi \approx 30^\circ$). We derive a general expression for the polytropic index. We find that in the presence of thermal conduction alone, the polytropic index remains close to its classical value for all the considered $\rho_0$ and $T_0$. However, it reduces to a value $1.2$ when loop density is decreased by an order of magnitude compared to its normal coronal value. We find that the inclusion of radiative losses, with or without h/c imbalance, cannot explain the observed polytropic index. The thermal ratio ($d$) needs to be enhanced by an order of magnitude, in order to explain its observed value $1.1 \pm 0.02$ in the solar loops. We also explore the role of different heating functions for typical coronal parameters and found that although the polytropic index remains close to its classical value, the phase difference is highly dependent on the form of heating function (The abstract is restructured for arxiv)., Comment: 32 Pages; 8 Figures; 1 Table; The final paper (arxiv version) as appeared in "Solar Physics" Journal after the editorial and proof correction (Prasad et al., Sol Phys., 296, 105 (2021))

A single-axis Microelectromechanical system gravimeter has recently been developed at the University of Glasgow. The sensitivity and stability of this device was demonstrated by measuring the Earth tides. The success of this device was enabled in part by its extremely low resonant frequency. This low frequency was achieved with a geometric anti-spring design, fabricated using well-established photolithography and dry etch techniques. Analytical models can be used to calculate the results of these non-linear oscillating systems, but the power of finite element analysis has not been fully utilised to explore the parameter space before now. In this article, the results of previous analytical solutions are replicated using finite element models, before applying the same techniques to optimise the design of the gravimeter. These computer models provide the ability to investigate the effect of the fabrication material of the device: anisotropic <100> crystalline silicon. This is a parameter that is difficult to investigate analytically, but finite element modelling is demonstrated here to provide accurate predictions of real gravimeter behaviour by taking anisotropy into account. The finite element models are then used to demonstrate the design of a three-axis gravimeter enabling the gravity tensor to be measured - a significantly more powerful surveying tool than the original single-axis device.

In the present paper, we derive a new dispersion relation for slow magnetoacoustic waves invoking the effect of thermal conductivity, compressive viscosity, radiation and unknown heating term along with the consideration of heating cooling imbalance from linearized MHD equations. We solve the general dispersion relation to understand role of compressive viscosity and thermal conductivity in damping of the slow waves in coronal loops with and without heating cooling imbalance. We have analyzed wave damping for the range of loop length $L$=50-500 Mm, temperature $T$=5-30 MK, and density $\rho$=10$^{-11}$-10$^{-9}$ kg m$^{-3}$. It was found that inclusion of compressive viscosity along with thermal conductivity significantly enhances the damping of fundamental mode oscillations in shorter (e.g., $L$=50 Mm) and super-hot ($T>$10 MK) loops. However, role of the viscosity in damping is insignificant in longer (e.g., $L$=500 Mm) and hot loops (T$\leq$10 MK) where, instead, thermal conductivity along with the presence of heating cooling imbalance plays a dominant role. For the shorter loops at the super-hot regime of the temperature, increment in loop density substantially enhances damping of the fundamental modes due to thermal conductivity when the viscosity is absent, however, when the compressive viscosity is added the increase in density substantially weakens damping. Thermal conductivity alone is found to play a dominant role in longer loops at lower temperatures (T$\leq$10 MK), while compressive viscosity dominates in damping at super-hot temperatures ($T>$10 MK) in shorter loops. The predicted scaling law between damping time ($\tau$) and wave period ($P$) is found to better match to observed SUMER oscillations when heating cooling imbalance is taken into account in addition to thermal conductivity and compressive viscosity for the damping of the fundamental slow mode oscillations., Comment: 40 Pages; 13 Figures; 01 Table; Solar Physics (in press)