updowninjections

Data release supporting:

• Parameter estimation of binary black holes in the endpoint of the up-down instability. Viola De Renzis, Davide Gerosa, Matthew Mould, Riccardo Buscicchio, Lorenzo Zanga. arXiv:2304.13063.

Credits

You are welcome to use this dataset in your research. We kindly ask you to cite the paper above. If you want to cite this data release specifically, the DOI code is: .

Data

Data need to be downloaded from the github release page. The total size is ~56GB. To download the entire dataset use our download_all.py script.

We provide two types of data products, named *.json and Backpropagation_*.dat

• The *.json files are the raw bilby outputs, see here for instructions. Each of these files contains a nested set of dictionaries with all the information of the injection (true values, waveform arguments, priors, posteriors, log evidence etc...).
• The Backpropagation_*.dat files contain the posterior samples of the tilt angles $\theta_1$ and $\theta_2$ that are numerically evolved from $20$ Hz to $0$ Hz. The first (second) column is the posterior distribution of $\theta_1$ ($\theta_2$). The posterior distributions for the tilt angles at $20$ Hz are contained in the *.json files.

Readme

The data provided refer to the results of the paper as follows:

• Data to reproduce Fig. 2 and Fig. 3 of the paper are listed as ud_rX_SNR_YYY.json where ud stands for up-down,X ranges from 1 to 8 and refers to the 8 series of injections shown in Fig. 3 performed with different values of the orbital separation r and with an YYY value of the $\mathrm{SNR}$. The chosen values of the $\mathrm{SNR}$ for the 8 series are $\mathrm{SNR}=150,100,80,60,40,20$. The files that reproduce the injection set of Fig. 2 (and the orange scatter points of Fig.3) are ud_r1_SNR_YYY, with YYY$=150,100,80,60,40,20$.

• The files to reproduce the posterior distribution at $20$ Hz in Fig. 4 are ud_r1_SNR_150.json (left panel) and ud_r6_SNR_150.json (right panel). The posterior distribution of the tilt angles $\theta_1$ and $\theta_2$ at $0$ Hz are contained in Backpropagation_r1_SNR_150.dat (left panel) and Backpropagation_r6_SNR_150.dat (right panel).

• The files for the 151 injections used in Fig. 5 of the paper are named as updown_X.json, where X ranges between 0 and 150. For all these runs, injection and recovery are done with the IMRPhenomXPHM waveform model using the standard uninformative priors (see the paper). The injection performed without higher modes are called updown_X_noHM.json where $X=39,61,99,104,113,125,138$.