Dead PFTs in PPE2_BHSon simulations

## Issue:
Decid. Temp trees and C4 grasses are dead at the end of the historical period in [`PPE2_BHSon` simulations](https://www.cgd.ucar.edu/tss/clm/diagnostics/clm5.1_dev/pft_level_plots/clm51_PPEn02ctsm51d021_2deg_GSWP3V1_leafbiomassesai_PPE_hist_pft_2dPDF_whittaker-1.png).
- [ ] **Confirm dead PFTs are occurring at the end of pAD in the `PPE2_BHSon` case** @olyson

## Notes
At the end of AD spinup decid. temperate trees and C4 grasses look OK in [`LMBIRCH`](https://www.cgd.ucar.edu/tss/clm/diagnostics/clm5.1_dev/pft_level_plots/clm51_lmbirchwkattgen01ctsm51d006_2deg_GSWP3V1_nocrop_PPE_1850AD_pft_2dPDF_whittaker-1.png), PPE2_BHSoff and [`PPE2_BHSon`](https://www.cgd.ucar.edu/tss/clm/diagnostics/clm5.1_dev/pft_level_plots/clm51_PPEn02ctsm51d021_2deg_GSWP3V1_leafbiomassesai_PPE_1850AD_pft_2dPDF_whittaker-1.png) simulations.
- [ ] **Continue the `PPE2_BHSoff` case through pAD to see if PFTs survive** @olyson

## Ideas:
- Assuming this can be isolated to `BHSon` simulations, are the dead PFTs reflecting N limitation caused by [cooler soil temperatures ](https://webext.cgd.ucar.edu/I20TR/clm51_PPEn02ctsm51d021_2deg_GSWP3V1_leafbiomassesai_PPE_hist/lnd/clm51_PPEn02ctsm51d021_2deg_GSWP3V1_leafbiomassesai_PPE_hist.1995_2014-clm51_lmbirchwkattgen01ctsm51d006_2deg_GSWP3V1_nocrop_PPE2_hist.1995_2014/set2/set2_ANN_TSOI_4.png)with `BHSon` that would slow N mineralization rates?
- The N sensitivity of decid. and C4 grass pft's is consistent with behavior of the model during CLM5 development & would not be immediately evident during AD spinup when decomposition and N mineralization rates are higher

### Ideas to try in parallel to explore this and other possibilities:
- [ ] Regional simulations in NE U.S. w/ Keith's code to replicate features of `BHSoff` & `BHSon`, @swensosc
- [ ] Fertilizing `BHSon` simulation in pAD by pointing to: stream_fldfilename_ndep = '/glade/work/wwieder/modified_inputs/fndep_clm_WACCM6_CMIP6piControl001_y21-50avg_1850monthly_0.95x1.25_c180802_fert05.nc' , @swensosc
- [ ] @oleson is it worth branching your global BHSon-pAD simulation with the extra nitrogen deposition?
- [ ] Look into parameter changes related to `ncosts` that may alleviate this issue, these can be tested in Sean's regional simulations? @wwieder @dlawrenncar