Timestep when calling the function plume_temporal_evolution in Cocipgrid

[Frank251826]

The timestep here directly takes the parameter "dt_integration". For the first contrail, dt are usually not as long as dt_integration. I am wondering if this is something we can modify a little bit.

Start at line 1674 to 1684 in cocip_grid:

sigma_yy_t2, sigma_zz_t2, sigma_yz_t2 = contrail_properties.plume_temporal_evolution(
    width_t1=width_t1,
    depth_t1=depth_t1,
    sigma_yz_t1=sigma_yz_t1,
    dsn_dz_t1=dsn_dz_t1,
    diffuse_h_t1=diffuse_h_t1,
    diffuse_v_t1=diffuse_v_t1,
    seg_ratio=seg_ratio_t12,
    dt=params["dt_integration"],
    max_depth=params["max_depth"],
)

Refer to the same usage in Cocip, dt is defined as contrail_2['time']-contrail_1['time'], which is reasonable.
A lot appreciate if anyone could provide some help.

[zebengberg]

Great catch! What if we change dt=params["dt_integration"] to dt=next_contrail["time"] - curr_contrail["time"]? Would that bring the two implementations into harmony?

[zebengberg]

@Frank251826 Please let me know what you think: #245

[Frank251826]

I did exactly the same and it works perfectly, which brings same result as Cocip.