H and He Reionization heating

Fixes #554 (closed). Implements #526 (closed). Adds a cosmological test evolving a uniform Universe with gas mean density to test the temperature evolution.

added 1 commit

• 3cef1157 - Adds makeIC.py

Compare with previous version

added 1 commit

• 87a827ae - Make the test case use primordial metallicities.

Compare with previous version

changed title from Reionization heating to H and He Reionization heating

added EAGLE SPH labels

changed the description

added 4 commits

• 8aba645a - Applied code formatting tool.
• f47787ac - Updated plotting script of the thermal history test to be agnostic about the cooling model in use.
• 400f2aaa - Also updated the signature of the cooling_update function in the other cooling models.
• 72c18dfd - Moved the energy injection for H reionization to a separate function.

Compare with previous version

added 2 commits

• dec85fa8 - Specify on the plot that the He reionization is Helium II.
• 6bbf6272 - Use the pre-existing energy injection routines for feedback when injecting…

Compare with previous version

added 1 commit

• 76187a09 - Also update the other hydro schemes for the new energy injection functions.

Compare with previous version

@jborrow can you check that you are happy with the energy injection functions in the different P-U schemes?

added 1 commit

• f79402e4 - Moved the new example into the 'Cooling' category and made it non EAGLE specific.

Compare with previous version

@dc-arri1 I have moved the example to a different directory and made it less EAGLE specific since it can be used for other schemes.

I have also changed the energy injection to use functions that exist (or will exist once another merge is done) for the feedback.

Let me know if you disagree with these.

added 1 commit

• 8046a28d - Added a README to the new example.

Compare with previous version

added 2 commits

• 8eee0a80 - Documentation fixes.
• b6c6b2cb - Document the amount of energy injected by H reionization in the RTD.

Compare with previous version

Looks good to me, as long as these are only ever used after the force loop and never in the hydro, as otherwise they'll mess up the unions.

Yes. They should only be used for feedback, so outside of the hydro loops.

mentioned in commit 9e45a0c3

merged

Looks good to me. The only issue would be in the case of Pressure-Entropy SPH, where we may want to treat reionization differently to feedback (in the way that I implemented in my version of the EAGLE code). @matthieu probably remembers many discussions about this. In any case it seems that feedback hasn't been implemented for Pressure-Entropy yet so perhaps we do not need to worry about this right now.

Yes, exactly.

Actually the function call now returns an error message telling the user to implement the function if you use P-E. The same issue applies to feedback so we will have to do something when/if we seriously consider PE.