I was tying to reproduce the result Figure 6 of this 2019 paper by Paul Torrey (https://arxiv.org/abs/1711.05261), and I cannot get agreement with the data points as shown in the image
For clarity I did the following:
I iterated over all central galaxies in the simulation that have gas
For every galaxy, for all cells I computed the number of particles N_O = X_O m_cell / (16 m_p) and N_H = X_H m_cell / m_p where X_O, X_H are the relative fractions given by the "GFM_Metals" field, and then I computed log10(sum(N_O)/sum(N_H))+12
For the figure I took only central galaxies, binned in stellar mass. The values are median + standard deviation
I understand that only TNG100 are fully calibrated but this is unlikely to explain a ~0.7 dex difference ?
Thanks in advance,
Ivan
Dylan Nelson
17h
It is unlikely that the Torrey+19 paper used the actual oxygen information, i.e. used the gas cells themselves.
More likely, this line represents the gas metallicity, converted into the usual observer "units" of "log(O/H)+12". In this case, it is probably one of the SubhaloGasMetallicity* fields from the catalogs. Most likely, it is SubhaloGasMetallicitySfrWeighted. The weighting (and/or radial aperture restriction) is crucial here, for a comparison with data to make sense.
Hello,
I was tying to reproduce the result Figure 6 of this 2019 paper by Paul Torrey (https://arxiv.org/abs/1711.05261), and I cannot get agreement with the data points as shown in the image
For clarity I did the following:
I understand that only TNG100 are fully calibrated but this is unlikely to explain a ~0.7 dex difference ?
Thanks in advance,
Ivan
It is unlikely that the Torrey+19 paper used the actual oxygen information, i.e. used the gas cells themselves.
More likely, this line represents the gas metallicity, converted into the usual observer "units" of "log(O/H)+12". In this case, it is probably one of the
SubhaloGasMetallicity*fields from the catalogs. Most likely, it isSubhaloGasMetallicitySfrWeighted. The weighting (and/or radial aperture restriction) is crucial here, for a comparison with data to make sense.