Problems with solar forcing in CESM 1.2.1

[andrew_kren@colorado_edu]

I have run two cases. The two cases are 1850 control runs from CESM 1.2.1 using a data ocean from a B1850 control run in an earlier CESM version. The ocean dataset is read correctly, however I am having issues with my solar response not coming out as clean in the upper atmosphere, as should not be the case. I am inputting my own spectral irradiance file and solar parms file using the last 4 solar cycles from the Lean data and repeating, specified by the nc files from NCAR. The model was run for 200 years. Attached are 3 png figures of QRS_TOT, showing the middle part of the run. What I am seeing is that in the upper atmosphere at 132, 129, and 125 km, QRS_TOT follows the variation in f107 very closely. I deseasoned QRS_TOT and multiplied f107 by a factor just so I can overplot them and see the variation. Lower down, in the other two figures, there is a shift between QRS_TOT and f017 starting between 122 and 118 km. The shift is larger and more evident in the third figure starting at 87 km, which is maybe the transition zone between when the parms file is used and the spectral irradiance? I have been debugging this issue for several months, and would be very appreciative is someone could help me with what may be going on? What would cause this shift or is it expected? Could someone look at my parms and spectral files I use? They are located on Pleiades under directory /nobackup/akren/input_data/solar_data_file_250yrs_1.nc and /nobackup/akren/input_data/solar_parms_file_250yrs_1.nc

I am not sure if this is an issue and if it is, if it is just due to a small error in my netcdf files, or something larger in the new cesm 1.2.1 version. Has anyone else run a varying solar cycle in cesm 1.2.1?