EPM integration time greatly exceeds convergence time

[chinahg]

The default integration time for the EPM ( timeFinal in Integrate.cpp line 209) is set to 2000 seconds. This timeframe of ~33 minutes is much longer than the generally accepted timeframe for contrail jet and vortex sinking (~3 min). [1]

APCEMM/Code.v05-00/src/EPM/Models/Original/Integrate.cpp

Line 209 in 7a2ec32

double timeFinal = 2.0E+03;

Additionally, the differential equations being solved generally reach a quasi-steady state by around 300s.

I would suggest this parameter be added as a user input for improved transparency to the user. For diagnostic reasons I would also suggest the epm-output.nc file include the time resolved arrays for the reported parameters instead of just scalar values at t=2000s.

[1] Contrail cirrus, in Cirrus, edited by D. K. Lynch, K. Sassen, D. O. Starr and G. L. Stephens, pp. 233, Oxford Univ.

[chinahg]

Update: The EPM output which is fed into the MPM is chosen at the integration time of t=180s. The choice of 180s is not backed by any current documentation/citations. The EPM integration between 180s and 2000s is essentially thrown out and should be removed.

[sdeastham]

Fully agreed! @chinahg : would you be able to generate a pull request implementing that change, limiting integration to 180 seconds? It sounds like it would hopefully be straight-forward.