Conversation
The original GST paper (http://doi.org/10.1007/s11467-013-0319-7) gives a parameterization of the CR flux in several populations based on p, He, C, O and Fe mass groups (Table 3). I see that this doesn't match the 5-component model that is used in some working groups (that's why I didn't touch the 5-component model) but for the 4-component version it would be appropriate to use the correct Z values and combine C and O.
for more information, see https://pre-commit.ci
Updating the 4-component GST model to the 4-population version, having C and O combined with the correct Z and Fe combined with the super heavy component
for more information, see https://pre-commit.ci
Test Results 5 files ±0 5 suites ±0 1m 58s ⏱️ -1s For more details on these failures, see this check. Results for commit fbbb347. ± Comparison against base commit b79c131. |
|
I am hesitant to change the value of models that already exist and would prefer to create a new model with a new name and mark the old model as deprecated. If the 5-comp version is similarly incorrect a correct version should also be created and the old version marked as depreciated. Since he was the original author of the 5comp model i have asked @The-Ludwig to coordinate on the correct solution. Also not the pre-commit is failing because there was a file |
I am super interested in making this clearer and easier to use correctly, but I implented the GSF 5-component version, not the GST 5-component version. Although I would like to take credit for it, the 5-component GST model was in this repository from the initial commit and is thus likely carried over from icetray and predates my time in IceCube. |
|
Yes it appears i read the blame log incorrectly the original commit is from 2015 https://code.icecube.wisc.edu/projects/icecube/changeset/139164/IceCube |
3 participants
The original GST paper (http://doi.org/10.1007/s11467-013-0319-7) gives a parameterization of the CR flux in several populations based on p, He, C, O and Fe and heavier mass groups (Table 3). I see that this doesn't match the 5-component model that is used in some working groups (that's why I didn't touch the 5-component model) but for the 4-component version it would be appropriate to use the correct Z values and combine C and O. Also, it is necessary to include the heavier components with Z>50 into the iron group. According to the author (Serap), the 4-population version is to be preferred, that's why I swtiched to it.
I am aware that this creates a difference between
GlobalFitGSTandGlobalFitGST_ITbut the way it has been defined so far is simply incorrect.