Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userTargeted Likelihood-Free Inference of Dark Matter Substructure in Strongly-Lensed Galaxies
87
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The analysis of optical images of galaxy-galaxy strong gravitational lensing systems can provide important information about the distribution of dark matter at small scales. However, the modeling and statistical analysis of these images is extraordinarily complex, bringing together source image and main lens reconstruction, hyper-parameter optimization, and the marginalization over small-scale structure realizations. We present here a new analysis pipeline that tackles these diverse challenges by bringing together many recent machine learning developments in one coherent approach, including variational inference, Gaussian processes, differentiable probabilistic programming, and neural likelihood-to-evidence ratio estimation. Our pipeline enables: (a) fast reconstruction of the source image and lens mass distribution, (b) variational estimation of uncertainties, (c) efficient optimization of source regularization and other hyperparameters, and (d) marginalization over stochastic model components like the distribution of substructure. We present here preliminary results that demonstrate the validity of our approach.
rate research
Read More
We investigate how strong lensing of dusty, star-forming galaxies by foreground galaxies can be used as a probe of dark matter halo substructure. We find that spatially resolved spectroscopy of lensed sources allows dramatic improvements to measurements of lens parameters. In particular we find that modeling of the full, three-dimensional (angular position and radial velocity) data can significantly facilitate substructure detection, increasing the sensitivity of observables to lower mass subhalos. We carry out simulations of lensed dusty sources observed by early ALMA (Cycle 1) and use a Fisher matrix analysis to study the parameter degeneracies and mass detection limits of this method. We find that, even with conservative assumptions, it is possible to detect galactic dark matter subhalos of ~ 10^8 M_{odot} with high significance in most lensed DSFGs. Specifically, we find that in typical DSFG lenses, there is a ~ 55 % probability of detecting a substructure with M>10^8 M_{odot} with more than 5 sigma detection significance in each lens, if the abundance of substructure is consistent with previous lensing results. The full ALMA array, with its significantly enhanced sensitivity and resolution, should improve these estimates considerably. Given the sample of ~100 lenses provided by surveys like the South Pole Telescope, our understanding of dark matter substructure in typical galaxy halos is poised to improve dramatically over the next few years.
Based on the strongly lensed gravitational waves (GWs) from compact binary coalescence, we propose a new strategy to examine the fluid shear viscosity of dark matter (DM) in the gravitational wave domain, i.e., whether a GW experiences the damping effect when it propagates in DM fluid with nonzero shear viscosity. By assuming that the dark matter self-scatterings are efficient enough for the hydrodynamic description to be valid, our results demonstrate that future ground-based Einstein Telescope (ET) and satellite GW observatory (Big Bang Observer; BBO) may succeed in detecting any dark matter self-interactions at the scales of galaxies and clusters.
We consider three extensions of the Navarro, Frenk and White (NFW) profile and investigate the intrinsic degeneracies among the density profile parameters on the gravitational lensing effect of satellite galaxies on highly magnified Einstein rings. In particular, we find that the gravitational imaging technique can be used to exclude specific regions of the considered parameter space, and therefore, models that predict a large number of satellites in those regions. By comparing the lensing degeneracy with the intrinsic density profile degeneracies, we show that theoretical predictions based on fits that are dominated by the density profile at larger radii may significantly over- or underestimate the number of satellites that are detectable with gravitational lensing. Finally, using the previously reported detection of a satellite in the gravitational lens system JVAS B1938+666 as an example, we derive for this detected satellite values of r_max and v_max that are, for each considered profile, consistent within 1sigma with the parameters found for the luminous dwarf satellites of the Milky Way and with a mass density slope gamma < 1.6. We also find that the mass of the satellite within the Einstein radius as measured using gravitational lensing is stable against assumptions on the substructure profile. In the future thanks to the increased angular resolution of very long baseline interferometry at radio wavelengths and of the E-ELT in the optical we will be able to set tighter constraints on the number of allowed substructure profiles.
Obtaining accurately calibrated redshift distributions of photometric samples is one of the great challenges in photometric surveys like LSST, Euclid, HSC, KiDS, and DES. We combine the redshift information from the galaxy photometry with constraints from two-point functions, utilizing cross-correlations with spatially overlapping spectroscopic samples. Our likelihood framework is designed to integrate directly into a typical large-scale structure and weak lensing analysis based on two-point functions. We discuss efficient and accurate inference techniques that allow us to scale the method to the large samples of galaxies to be expected in LSST. We consider statistical challenges like the parametrization of redshift systematics, discuss and evaluate techniques to regularize the sample redshift distributions, and investigate techniques that can help to detect and calibrate sources of systematic error using posterior predictive checks. We evaluate and forecast photometric redshift performance using data from the CosmoDC2 simulations, within which we mimic a DESI-like spectroscopic calibration sample for cross-correlations. Using a combination of spatial cross-correlations and photometry, we show that we can provide calibration of the mean of the sample redshift distribution to an accuracy of at least $0.002(1+z)$, consistent with the LSST-Y1 science requirements for weak lensing and large-scale structure probes.
We study the application of machine learning techniques for the detection of the astrometric signature of dark matter substructure. In this proof of principle a population of dark matter subhalos in the Milky Way will act as lenses for sources of extragalactic origin such as quasars. We train ResNet-18, a state-of-the-art convolutional neural network to classify angular velocity maps of a population of quasars into lensed and no lensed classes. We show that an SKA -like survey with extended operational baseline can be used to probe the substructure content of the Milky Way.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
