Am 19. August um 14 Uhr gibt Dr. Jenny Wagner (Institute of Astronomy and Astrophysics, Academia Sinica) spannende Einblicke in ihr neues Verfahren Lensing of ’69. Statt komplexer Modellannahmen nutzt sie direkte Beobachtungsdaten, um die Verteilung unsichtbarer Materie in Galaxien und Galaxienhaufen präziser und ressourcenschonender zu rekonstruieren.

Abstract: Facing an under-constrained modelling problem, we often compensate the lack of data by adding models. The latter are based on symmetry assumptions or other physically reasonable principles and heuristics.
In this talk, I will show in how far the problem to reconstruct a cosmic mass density like a galaxy or a galaxy cluster is under-constrained when we are only using the so-called strong gravitational lensing effect, i.e. observing the light deflections that the high mass density causes according to Einstein's field equations. Adding model assumptions about the light-deflecting mass density, we need to scan a high-dimensional and vast parameter space. As optimum solutions, we often find several, very different models that describe the light-deflecting mass distribution equally well. Which of these models comes closest to reality given the observations? How can we decide this, given that most of the mass in such gravitational lenses is supposed to be undetectable dark matter? Last but not least, is there an efficient, resource-saving way to reconstruct a light-deflecting mass distribution?

To answer these questions, I introduce "Lensing of '69", a new approach that does not fit global models to a sparse set of data, but rather infers local lens properties at the data points. They are directly and uniquely determined from observed, extended multiple images without any lens model. Since these local lens properties are the maximum information that all lens models agree upon, they are the ideal starting point to extrapolate lens reconstructions to regions without multiple images. Hybrids of local lensing properties by "Lensing of '69" and model-based extensions or complementary data thus offer the possibility to reduce systematic biases in lens reconstructions, and provide the results much faster than conventional lensing algorithms.
In summary, combining the power of the mathematical foundations of the gravitational lensing formalism with physically well-motivated extrapolations, huge amount of upcoming survey data can be efficiently evaluated in a way that is immediately understandable by humans.