Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/research/LVmeeting/magsys_pres/ruzicka.pdf Дата изменения: Sun Aug 5 03:30:22 2007 Дата индексирования: Tue Oct 2 17:18:20 2012 Кодировка: Поисковые слова: emission

Magellanic Clouds in interaction
evolutionary search for good models
Adam Rzicka1, Jan Palous1, Christian Theis2
1Astronomical

Institute, Prague; 2Institut fЭr Astronomie, Wien


Idea
investigate the LMC-SMC-Milky Way (MW) interaction by modeling the distribution of HI related to the Magellanic System (BrЭns et al. 2005)


To be more specific...
set up a parameter and initial condition space of the interaction following the available observations of the Magellanic Clouds use genetic algorithms (GA) and a fast restricted N-body model to... ...perform a detailed and complete search of the entire parameter space to study influence of the parameters on the interaction proper motion of the Clouds


Motivation
approx. 20 models of the Magellanic System carried out either no or insufficient answers exist to the following questions Are the interaction scenarios introduced so far unique? How much does the HI distribution tell about the LMC/SMC motion? Is a long term LMC-SMC gravitational binding necessary? Is the SMC the only source of matter for the large-scale structures? Why is there no stellar content in the Magellanic Stream?


Parameter space
interaction is determined by approx. 20 parameters including total LMC and SMC masses + parameters of mass distribution structure of the LMC and SMC particle setup distribution of dark matter in the MW halo initial conditions of LMC and SMC motion
no assumptions on the LMC/SMC motion but just trust observations...


Proper motion
different methods and different values in Jones et al. 1994 (J94), Kroupa et al. 1994 (PPM), Kroupa&Bastian 1997 (HIP) Kallivayalil et al. 2006A, Kallivayalil et al. 2006B


Numerical model
3D restricted N-body model of the LMC-SMC-MW interaction including Newtonian law of gravity flattened axisymmetric logarithmic MW potential

122 z2 ( R, z ) = VD ln R + d 2 + 2 + const. 2 q
spherical potential of both LMC, SMC Dynamical friction in the MW halo (Binney 1977) run from T = -4Gyr no gas but just a tidal model


Genetic algorithm search
measure of a model quality is a func. in a 20-D space and we call it fitness func. (0 < FF < 1) 110 GA runs performed to map the FF the fits always of FF > 0.4, i.e. ... ...every fit contained major HI structures trailing stream (Magellanic Stream) leading tail (Leading Arm) in total only 106 parameter combinations had to be tested to search the entire 20-dimensional parameter space simple testing of every parameter on even a very sparse grid of 10 nodes/dim. would mean probing 1020 parameter combinations...


It's all about proper motion...


It's all about proper motion...
select a high-quality model (FF 0.5) and look around

Tidal models don't work...?


PM versus reproduction of HI
VyLMC - V
y SMC

FF landscape for a good GA fit


Why is the LMC/SMC velocity so critical?


BrЭns et al. 2005


Conclusions
no model of FF > 0.53 was identified insufficient volume of the parameter space simplification of physical processes distribution of HI is extremely sensitive to the LMC/SMC PM tidal models don't seem to work for the HST PM... ...or the HST PM don't work for tidal models...

BUT
the Clouds are so unique ­ fast moving guests make sense other physics might help... specific parameter combinations might allow for great models surrounded by rubbish, so... ...exclude some parameters and make life easier for the searching algorithms


Magellanic Stream was created for non-spherical MW halos q 1.0 prevent the system from close (R < 10 kpc) LMCSMC encounters HI redistribution started and continued by MWLMC/MW-SMC tidal stripping offer a natural explanation of the missing star problem tidal models don't seem to work for the HST PM...