Source author record

P. M. Visser

P. M. Visser appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

astro-ph.EP astro-ph.IM Machine Learning physics.comp-ph

Catalog footprint

What is connected

4works

4topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Collision detection for N-body Kepler systems

In a Keplerian system, a large number of bodies orbit a central mass. Accretion disks, protoplanetary disks, asteroid belts, and planetary rings are examples. Simulations of these systems require algorithms that are computationally efficient. The inclusion of collisions in the simulations is challenging but important. We intend to calculate the time of collision of two astronomical bodies in intersecting Kepler orbits as a function of the orbital elements. The aim is to use the solution in an analytic propagator ($N$-body simulation) that jumps from one collision event to the next. We outline an algorithm that maintains a list of possible collision pairs ordered chronologically. At each step (the soonest event on the list), only the particles created in the collision can cause new collision possibilities. We estimate the collision rate, the length of the list, and the average change in this length at an event, and study the efficiency of the method used. We find that the collision-time problem is equivalent to finding the grid point between two parallel lines that is closest to the origin. The solution is based on the continued fraction of the ratio of orbital periods. Due to the large jumps in time, the algorithm can beat tree codes (octree and $k$-d tree codes can efficiently detect collisions) for specific systems such as the Solar System with $N<10^8$. However, the gravitational interactions between particles can only be treated as gravitational scattering or as a secular perturbation, at the cost of reducing the time-step or at the cost of accuracy. While simulations of this size with high-fidelity propagators can already span vast timescales, the high efficiency of the collision detection allows many runs from one initial state or a large sample set, so that one can study statistics.

preprint2022arXiv

Exoplanet Cartography using Convolutional Neural Networks

In the near-future, dedicated telescopes observe Earth-like exoplanets in reflected light, allowing their characterization. Because of the huge distances, every exoplanet will be a single pixel, but temporal variations in its spectral flux hold information about the planet's surface and atmosphere. We test convolutional neural networks for retrieving a planet's rotation axis, surface and cloud map from simulated single-pixel flux and polarization observations. We investigate the assumption that the planets reflect Lambertian in the retrieval while their actual reflection is bidirectional, and of including polarization in retrievals. We simulate observations along a planet's orbit using a radiative transfer algorithm that includes polarization and bidirectional reflection by vegetation, desert, oceans, water clouds, and Rayleigh scattering in 6 spectral bands from 400 to 800 nm, at various photon noise levels. The surface-types and cloud patterns of the facets covering a model planet are based on probability distributions. Our networks are trained with simulated observations of millions of planets before retrieving maps of test planets. The neural networks can constrain rotation axes with a mean squared error (MSE) as small as 0.0097, depending on the orbital inclination. On a bidirectionally reflecting planet, 92% of ocean and 85% of vegetation, desert, and cloud facets are correctly retrieved, in the absence of noise. With realistic noise, it should still be possible to retrieve the main map features with a dedicated telescope. Except for face-on orbits, a network trained with Lambertian reflecting planets, yields significant retrieval errors when given observations of bidirectionally reflecting planets, in particular, brightness artefacts around a planet's pole. Including polarization improves retrieving the rotation axis and the accuracy of the retrieval of ocean and cloud facets.

preprint2019arXiv

Fourier series for eclipses on exoplanet binaries

A double planet system or planet binary undergoes eclipses that modify the reflective light curve. In the time domain, the eclipse events are fast and weak. This would make their signal difficult to find and recognize in the phase light curve, even for small inclinations when eclipses happen frequently. However, due to the quasiperiodic nature of the phenomenon, the Fourier transform of the direct reflection signal consists of a double sum of sharp peaks. These peaks can be resolved for large close binaries and sufficiently long observation times with a star coronagraph. Eclipses modulate the phase curve, having an orbital period $2π/ω$, with a contribution from the relative motion in the binary plane of a period $2π/Ω$. This leads to a spectral structure with basis frequencies $ω$ and $Ω$. We aim to characterize these spectra. We studied the regime of short eclipses that occur when the planet radii are small compared to the planet separation. We derived formulas for the peak amplitudes applicable to homogeneous (Lambertian) planet binaries in circular orbit with small inclination. The effects of an eclipse and of double reflection appear as first- and second-order contributions (in planet radius over separation) in the reflection signal respectively. Small peaks appear as observable side bands in the spectrum. Identical structures around $mΩ$ are characteristic of short-duration eclipses. Deceasing side bands could indicate double reflection between companions. Fourier analysis of the light curve of non-transiting planets can be used to find planets and their moons. Difficulties in interpreting the structures arise for small planet separation and when there are several moons in mean-motion resonance.

preprint2015arXiv

Fourier spectra from exoplanets with polar caps and ocean glint

The weak orbital-phase dependent reflection signal of an exoplanet contains information on the planet surface, such as the distribution of continents and oceans on terrestrial planets. This light curve is usually studied in the time domain, but because the signal from a stationary surface is (quasi)periodic, analysis of the Fourier series may provide an alternative, complementary approach. We study Fourier spectra from reflected light curves for geometrically simple configurations. Depending on its atmospheric properties, a rotating planet in the habitable zone could have circular polar ice caps. Tidally locked planets, on the other hand, may have symmetric circular oceans facing the star. These cases are interesting because the high-albedo contrast at the sharp edges of the ice-sheets and the glint from the host star in the ocean may produce recognizable light curves with orbital periodicity, which could also be interpreted in the Fourier domain. We derive a simple general expression for the Fourier coefficients of a quasiperiodic light curve in terms of the albedo map of a Lambertian planet surface. Analytic expressions for light curves and their spectra are calculated for idealized situations, and dependence of spectral peaks on the key parameters inclination, obliquity, and cap size is studied.