7th LISA PE telecon 14.02.2008 ================================================ Partial list of participants: Curt Cutler Alicia M Sintes sintes#aei.mpg.de B.S.Sathyaprakash K G Arun Han Wang Jing Zeng Emanuele Berti Marta Volonteri Miquel Trias Neil Cornish John Baker James Ira Thorpe Ed Porter ========================================================= 1-EMRI comparison Gair (not present) posted some results on the wiki. Ed and Jonathan now have results that can be directly compared to those of Neil, using the start time parameters rather than parameters at plunge. Results posted for the training set 1.3.2. They will post more results soon. The widths of the distributions obtained in all of the parameters are quite similar to those obtained by Neil, but for some parameters the shape of the distributions do not match at all. Neil points out that it is hard to make comparisons depending on the approximations on the waveform performed. Ed plans to use better waveforms using Besel functions. ========================================================= 2- binary black hole population models and results Marta and Emanuele have provided Curt a file with the first set of parameters (case isotropic distributions of spins). They plan to provide ~12 files containing 120 different realizations of merger trees. Cutler has produce some m_files to convert that data into the format than can be read in by Cornish code. Curt plans to run that code on this first set. He also plans to send Marta histograms of her files to make sure he is understanding them properly. The next step is to run on the supercomputer and produce results. The results produced will be in form of a task-force report for the lisa project. First using Neil's code (currently available to the project) for the different LISA's configurations. Marta and Emanuele will discuss the most sensible cases to run e.g., efficient, inefficient, or chaotic accretion. Will try to avoid doing too many assumptions about nature and try to cover the whole spectrum of possibilities. Neil asks to consider other extreme cases. Marta hesitates to produce all possible cases since some are unphysical. An unphysical case is for example when we ignore accretion completely. This is just to see what LISA can see. Among the cases to be studied, it will be included: - isotropic efficient accretion - equatorial chaotic accretion. They will test extreme but physical cases. Kicks are not included in the data. So results should be taken as an upper limit. One should be careful since there is a huge number of assumptions in these cases. John Baker asks for some extreme cases and if the Tamara Bogdanovic alignment scenario is right, also about the spin-orbital align cases (no precession) as a subset of the equatorial case. Marta will keep discussing with them and also depending on the impact on parameter estimation. Keep talking offline. Cutler decides to use Neil's code since this one includes both higher harmonics and spin precession. Sathya and others point out that the results obtained depend on the waveform and how this is treated or cut, as it has been seen doing the SMBH comparison. All groups involved in SMBH comparison agree in all the cases of using the same low and high frequency cut off, but disagree with Neil's code otherwise because how the waveform is attenuated in Neil's code. The Goal is now to produce a report with results. The document should also mention how the different codes have been verified with each other, how they work and there should be a plan of results (e.g., different lisa arm-lenght) At next telecon there will be a discussion about the possibility of producing a paper and if this report/paper should be edited/checked under cvs or similar. Also results should be discussed together. ========================================================= 3-SMBH status of the various check and comparisons: TS have worked very closely with AISSV and have checked that both SNR and errors coincide for both RWF and FWF, only there were some little discrepancies in the last test case but it has been fixed in the meantime (the case with both frequency cut-offs) Neil's code also agrees with these two. So the exercise of the comparison is finalized. Regarding waveform comparison TS used a h_plus and h_cross with the opposite sign of Cornish et al. Thus the waveforms did not match perfectly. After changing this sign, it is possible to find the best match phi_c between Cornish and the rest and with this value then SNR and Errors match up to the 3rd decimal!!!! Note that a change of sign in the case of harmonics are used is no longer equivalent to a change of phi_c. Notice that no codes have been modified at all. All changes were related to just make sure we were all using the same parameters, instrumental noise, cut-offs, initial lisa position, phi_c etc. This comparison exercise validates the results presented last summer by Arun et al. and Trias et al. regarding the impact of adding further harmonics. ========================================================= 4- Topics for future discussions: - how to proceed with the task-force report. join discussion of results, codes to be used, cases on which to run - Possibility of writing a *group* paper about these results, authorship. - Continuation of this task-force, since 2 of the 3 main gals have been achieved, and how or on which topics ========================================================= Next telecon: February 28th 8:15 PST, 11:15 EST, 16:15 UTC, 17:15 CET Please send comments or correction to sintes (@) aei.mpg.de