1.Regarding the gas temperature, I noticed that there is a floor at approximately (Log) 3.9–4 K. Would this affect studies of the cold-gas physics? In other words, if I aim to investigate the properties of cold gas, would it be necessary to apply additional treatment to this component?
2.Since wind particles are decoupled, does this imply that measurements of inflows or outflows within small radii may not be very accurate? My concern is that, in such regions, the wind particles have not yet released their energy, whereas in reality the gas in this region should still be influenced by stellar feedback.
3.How can tracers be accessed in JupyterLab? Given that I only have 10 GB of memory, I am unable to use loadSubset method to call all of PartType3. In addition, is there a detailed article or documentation that provides a comprehensive explanation of tracers?
Thank you for your help and best wishes!
Zhang
怡远 张
18 Sep
Hi Nelson! I have one more question that I hope you could help. I noticed that the TNG website provides examples for data download, but the wget method appears to retrieve the entire snapshot. If I only want to download specific fields for certain particle types—for instance, downloading only the InternalEnergy field for gas, while retrieving all fields for black holes—what would be the correct way to do this?
Dylan Nelson
18 Sep
(1) If you want to study gas at ~1e4K, or "total cold gas" at <~ 1e4K, then it is reasonable to take all cold gas (e.g. log(T) < 4.5) as a proxy for this. Note you should include all star-forming gas in this accounting, since by mass it is cold. If your question requires a more careful modeling, then you would need to do this.
(2) Yes you should look at outflows with caution at distances of ~kpc i.e. near the galaxy itself. If you are so close, you must carefully think if you want to include wind particles in e.g. a calculation of mass outflow rate. (Typically I would say yes: wind particles represent a flux of gas mass). As you say, however, the TNG model intentionally does not resolve the interaction of SN-driven outflows and the ISM, so you cannot study such small-scale interactions.
(3) You can load subsets of PartType3 at a time, e.g. see the recently added examples/chunked_load.py function. However, it will be difficult to do a global ID matching - you would have to write such an algorithm, that worked on chunked data. One other alternative would be to use the tracer_tracks supplementary catalog, if it is relevant.
(4) See "Task 11" under the API documentation. You can download a specific field from the snapshot files in this way.
怡远 张
1 Oct
Hi Nelson! Thank you for your helpful guidance！
I noticed that you mentioned considering star-forming gas when analyzing the cold component. Does this gas also reside within PartType0? In another word, do I need to apply any special treatment when reading such data?
2.In addition, after downloading the data using Task11 method, some files were not successfully retrieved due to network issues. How should I verify the completeness of the dataset and re-download the missing files?
I would be greatful for your assistance !
Dylan Nelson
1 Oct
(1) Yes, star-forming gas is simply PartType0 with StarFormationRate > 0, so you should be careful when you consider its temperature.
(2) If the files are missing, you can re-try for missing files. If the files are corrupt, you could use h5check or a similar command to determine which files need to be requested again.
Hi Nelson !
I would like to ask a few questions:
1.Regarding the gas temperature, I noticed that there is a floor at approximately (Log) 3.9–4 K. Would this affect studies of the cold-gas physics? In other words, if I aim to investigate the properties of cold gas, would it be necessary to apply additional treatment to this component?
2.Since wind particles are decoupled, does this imply that measurements of inflows or outflows within small radii may not be very accurate? My concern is that, in such regions, the wind particles have not yet released their energy, whereas in reality the gas in this region should still be influenced by stellar feedback.
3.How can tracers be accessed in JupyterLab? Given that I only have 10 GB of memory, I am unable to use loadSubset method to call all of PartType3. In addition, is there a detailed article or documentation that provides a comprehensive explanation of tracers?
Thank you for your help and best wishes!
Zhang
Hi Nelson! I have one more question that I hope you could help. I noticed that the TNG website provides examples for data download, but the wget method appears to retrieve the entire snapshot. If I only want to download specific fields for certain particle types—for instance, downloading only the InternalEnergy field for gas, while retrieving all fields for black holes—what would be the correct way to do this?
(1) If you want to study gas at ~1e4K, or "total cold gas" at <~ 1e4K, then it is reasonable to take all cold gas (e.g. log(T) < 4.5) as a proxy for this. Note you should include all star-forming gas in this accounting, since by mass it is cold. If your question requires a more careful modeling, then you would need to do this.
(2) Yes you should look at outflows with caution at distances of ~kpc i.e. near the galaxy itself. If you are so close, you must carefully think if you want to include wind particles in e.g. a calculation of mass outflow rate. (Typically I would say yes: wind particles represent a flux of gas mass). As you say, however, the TNG model intentionally does not resolve the interaction of SN-driven outflows and the ISM, so you cannot study such small-scale interactions.
(3) You can load subsets of PartType3 at a time, e.g. see the recently added
examples/chunked_load.pyfunction. However, it will be difficult to do a global ID matching - you would have to write such an algorithm, that worked on chunked data. One other alternative would be to use thetracer_trackssupplementary catalog, if it is relevant.(4) See "Task 11" under the API documentation. You can download a specific field from the snapshot files in this way.
Hi Nelson! Thank you for your helpful guidance！
2.In addition, after downloading the data using Task11 method, some files were not successfully retrieved due to network issues. How should I verify the completeness of the dataset and re-download the missing files?
I would be greatful for your assistance !
(1) Yes, star-forming gas is simply PartType0 with
StarFormationRate > 0, so you should be careful when you consider its temperature.(2) If the files are missing, you can re-try for missing files. If the files are corrupt, you could use
h5checkor a similar command to determine which files need to be requested again.