Adaptive AGN heating temperature jump

This merge request implements a model to adaptively calculate the AGN heating temperature jump dT, rather than using a prescribed constant value.

dT is calculated as min(max(dT_min, dT_mass, dT_crit, dT_local), dT_max), where:

• dT_min and dT_max are (constant) min/max values for dT (default: 1e7 K, 3e9 K)
• dT_mass = theta * (m_BH / m_ref)^beta (with theta, m_ref, and beta as free parameters, and m_BH the subgrid mass of the BH). This is intended as a proxy for a scaling with virial halo temperature, assuming that m_BH is proportional to M_halo. Beta is then fixed at 2/3. The default values for theta and m_ref are 3e8 K and 1e8 M_sun, respectively; tests in 25 Mpc boxes show little impact of the exact values.
• dT_crit is the critical temperature required to avoid numerical cooling losses according to Dalla Vecchia & Schaye (2012), multiplied by an arbitrary factor (default: 1).
• dT_local is the (SPH-smoothed) average gas temperature in the kernel, multiplied with an arbitrary factor (default: 0, i.e. do not take local temperature into account).

So far, only the EAGLE_ICs/EAGLE_25 parameter example is updated with the new parameters.

changed milestone to %Improved black hole model

added EAGLE SPH feature request labels

added 3 commits

• 634422bf - Clarify description of output AGN_delta_T
• 7b776638 - Incorporate constant AGN delta T into variable temperature calculation
• 4a836aff - Improve description of base line delta T for AGN, fix incorrect if-statement

Compare with previous version

resolved all threads

Thanks, I am just looking for a reference for

T_crit = 10^7.5  *  (n_H / (10 cm^-3))^(2/3)  *  (M_ngb / (10^6 Msun))^(1/3)

Also, do we want your EAGLE-25 parameters as defaults throughout all param files for now?

added 1 commit

• 89e82b09 - Added explanation comments for critical and background temperature in variable AGN dT model

Compare with previous version

Yes, I think we can consider the EAGLE-25 parameters as default for the time being. Is it straightforward for you to copy these to the the other cases? Otherwise I can do it as well.

I can do it. Thanks.

mentioned in merge request !1184 (merged)

added 2 commits

• 075d2d80 - Applied code formatting tool
• 5116afbb - Update all the relevant parameter files to use the new AGN temperature options

Compare with previous version

added 1 commit

• 39dc9c88 - Document the change in AGN feedback temperature model

Compare with previous version

Any last minute changes?

Ah, of course. Thanks!

No, I'm good to go. Do you think it makes sense to make the corresponding update in the Colibre code? Or would it be better to wait with this until Evgenii's updates are done?

Yes, but I guess when things have calmed down a bit on the front of merging other pieces of feedback-related code.