Update to S2S forecast processing script

 - Update to the CFSv2 processing python script to
    account for year-crosssover for January, which
    has pentad-date members in prior December+year.

 - Minor unit update for one s2splots verification
    plotting script (no impact on any processing
    of files or LIS runs).

lis/utils/usaf/s2s/s2s_modules/bcsd_fcst/bcsd_library/process_forecast_data.py

@@ -234,10 +234,11 @@ def _driver():
     ens_num = int(args['ens_num'])
     print(f"[INFO] {fcst_init['monthday']} {year}")
 
+    fcst_init["year"] = year
     if fcst_init['monthday'] == "jan01":
-        fcst_init["year"] = year - 1
+        fcst_init["year_cfsv2"] = year - 1
     else:
-        fcst_init["year"] = year
+        fcst_init["year_cfsv2"] = year
 
     mmm = fcst_init['monthday'].split("0")[0].capitalize()
     dt1 = datetime.strptime('{} 1 {}'.format(mmm,fcst_init["year"]), '%b %d %Y')
@@ -246,8 +247,8 @@ def _driver():
     dt2 = np.datetime64(dt2.strftime('%Y-%m-%d'))
 
     monthday = args['all_monthdays'][ens_num - 1]
-    temp_name = f"cfsv2.{fcst_init['year']}{monthday}.nc"
-    fcst_init['date'] = f"{fcst_init['year']}{monthday}"
+    temp_name = f"cfsv2.{fcst_init['year_cfsv2']}{monthday}.nc"
+    fcst_init['date'] = f"{fcst_init['year_cfsv2']}{monthday}"
     fcst_init['month'] = int(monthday[0:2])
     fcst_init['day'] = int(monthday[2:4])
     fcst_init['hour'] = args['all_ensmembers'][ens_num - 1]
@@ -272,23 +273,23 @@ def _driver():
     if not os.path.exists(outdirs['outdir_monthly']):
         os.makedirs(outdirs['outdir_monthly'])
 
-    cfsv2 = []
+    cfsv2 = []         
     for varname in ["prate", "pressfc", "tmp2m", "dlwsfc", "dswsfc",
                     "q2m", "wnd10m"]:
         print(f"[INFO] {varname}")
         subdir, file_pfx, file_sfx = \
-            _set_input_file_info(fcst_init['year'],
+            _set_input_file_info(fcst_init['year_cfsv2'],
                                  fcst_init['month'],
                                  varname)
         indir = f"{args['forcedir']}/{subdir}/"
         if subdir == "Oper_TS" and not os.path.exists(indir):
             indir = f"{args['forcedir']}/"
-
-        indir += f"{fcst_init['year']}/{fcst_init['date']}"
+            
+        indir += f"{fcst_init['year_cfsv2']}/{fcst_init['date']}"
 
         # Convert GRIB file to netCDF and handle missing/corrupted data
         cfsv2.append(read_wgrib (indir, file_pfx, fcst_init['timestring'], file_sfx, outdirs['outdir_6hourly'], temp_name, varname, args['patchdir']))
-
+        
     cfsv2 = xr.merge (cfsv2, compat='override')
     reg_precip = _regrid_precip(cfsv2, args)
     _migrate_to_monthly_files(cfsv2.sel (step = (cfsv2['valid_time']  >= dt1) &

lis/utils/usaf/s2s/s2s_modules/s2splots/plot_anom_verify.py

@@ -72,7 +72,7 @@ def compute_median (anom, lead):
     ''' compute median anomaly of the specified lad '''
     da_slice=[]
     for da in anom:
-        da_slice.append(da.isel(lead=lead))
+        da_slice.append(da.isel(time=lead))
     da_conc = xr.concat(da_slice, dim = 'ens')
     return da_conc.median(dim = 'ens')
 
@@ -349,6 +349,8 @@ def write_clim(config_file, month, flabel):
 
         nc_med = compute_median (anoms, lead)
         plot_arr[0,:] = nc_med.values *convf
+        if var == 'Precip':
+           plot_arr[0,:] = plot_arr[0,:]/ndays 
 
         # plotting
         style_color = plot_utils.load_table(load_table)