add EAMxx support for cmorization

e3sm_to_cmip/cmor_handlers/_formulas.py

@@ -466,20 +466,49 @@ def pr(ds: xr.Dataset) -> xr.DataArray:
 
     High frequency version:
     pr = PRECT * 1000.0
+
+    EAMxx version:
+    pr = precip_liq_surf_mass_flux + precip_ice_surf_mass_flux
+    or:
+    pr = precip_total_surf_mass_flux
     """
     if all(key in ds.data_vars for key in ["PRECC", "PRECL"]):
         result = (ds["PRECC"] + ds["PRECL"]) * 1000.0
     elif "PRECT" in ds:
         result = ds["PRECT"] * 1000.0
+    elif all(
+        key in ds.data_vars
+        for key in ["precip_liq_surf_mass_flux", "precip_ice_surf_mass_flux"]
+    ):
+        result = ds["precip_liq_surf_mass_flux"] + ds["precip_ice_surf_mass_flux"]
+    elif "precip_total_surf_mass_flux" in ds:
+        result = ds["precip_total_surf_mass_flux"]
     else:
         raise KeyError(
             "No formula could be applied for 'pr'. Check the handler entry for 'pr' "
-            "and input file(s) contain either 'PRECC' and 'PRECL', or 'PRECT."
+            "and input file(s) contain either 'PRECC' and 'PRECL', 'PRECT', "
+            "'precip_liq_surf_mass_flux' and 'precip_ice_surf_mass_flux', "
+            "or 'precip_total_surf_mass_flux'."
         )
 
     return result
 
 
+def clwvi(ds: xr.Dataset) -> xr.DataArray:
+    """
+    CMIP6 clwvi = condensed water path (liquid + ice).
+
+    EAM version:
+    clwvi = TGCLDCWP  (total condensed = liquid + ice, single variable)
+
+    EAMxx version:
+    clwvi = LiqWaterPath + IceWaterPath
+    In EAMxx, LiqWaterPath is liquid-only and IceWaterPath is ice-only;
+    they must be summed to match the CMIP6 definition.
+    """
+    return ds["LiqWaterPath"] + ds["IceWaterPath"]
+
+
 def prsn(ds: xr.Dataset) -> xr.DataArray:
     """
     prsn = (PRECSC  + PRECSL) * 1000.0

e3sm_to_cmip/cmor_handlers/handlers.yaml

@@ -1040,3 +1040,280 @@
   formula: null
   positive: null
   levels: { name: plev19, units: Pa, e3sm_axis_name: plev }
+# ============================================================
+# EAMxx variable handlers (issue #339)
+# Maps EAMxx output variable names to CMIP6 variables.
+# These entries coexist with EAM entries above; the handler
+# system selects whichever raw_variables are present in input.
+# ============================================================
+# --- Precipitation and column water ---
+- name: pr
+  units: kg m-2 s-1
+  raw_variables: [precip_liq_surf_mass_flux, precip_ice_surf_mass_flux]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: precip_liq_surf_mass_flux + precip_ice_surf_mass_flux
+  positive: null
+  levels: null
+- name: prw
+  units: kg m-2
+  raw_variables: [VapWaterPath]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- Surface temperature and humidity ---
+- name: ts
+  units: K
+  raw_variables: [surf_radiative_T]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+- name: tas
+  units: K
+  raw_variables: [T_2m]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+- name: huss
+  units: "1"
+  raw_variables: [qv_2m]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- Surface energy and momentum fluxes ---
+- name: hfls
+  units: W m-2
+  raw_variables: [surface_upward_latent_heat_flux]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: hfss
+  units: W m-2
+  raw_variables: [surf_sens_flux]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: evspsbl
+  units: kg m-2 s-1
+  raw_variables: [surf_evap]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+- name: tauu
+  units: Pa
+  raw_variables: [surf_mom_flux_U]
+  table: CMIP6_Amon.json
+  unit_conversion: "-1"
+  formula: null
+  positive: down
+  levels: null
+- name: tauv
+  units: Pa
+  raw_variables: [surf_mom_flux_V]
+  table: CMIP6_Amon.json
+  unit_conversion: "-1"
+  formula: null
+  positive: down
+  levels: null
+# --- 10m wind speed ---
+- name: sfcWind
+  units: m s-1
+  raw_variables: [wind_speed_10m]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- Sea-level pressure ---
+- name: psl
+  units: Pa
+  raw_variables: [SeaLevelPressure]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- TOA radiation ---
+- name: rsdt
+  units: W m-2
+  raw_variables: [SW_flux_dn_at_model_top]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: down
+  levels: null
+- name: rsut
+  units: W m-2
+  raw_variables: [SW_flux_up_at_model_top]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rsutcs
+  units: W m-2
+  raw_variables: [SW_clrsky_flux_up_at_model_top]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rlut
+  units: W m-2
+  raw_variables: [LW_flux_up_at_model_top]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rlutcs
+  units: W m-2
+  raw_variables: [LW_clrsky_flux_up_at_model_top]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+# --- Surface radiation ---
+- name: rsds
+  units: W m-2
+  raw_variables: [SW_flux_dn_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: down
+  levels: null
+- name: rsus
+  units: W m-2
+  raw_variables: [SW_flux_up_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rsdscs
+  units: W m-2
+  raw_variables: [SW_clrsky_flux_dn_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: down
+  levels: null
+- name: rsuscs
+  units: W m-2
+  raw_variables: [SW_clrsky_flux_up_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rlds
+  units: W m-2
+  raw_variables: [LW_flux_dn_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: down
+  levels: null
+- name: rlus
+  units: W m-2
+  raw_variables: [LW_flux_up_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: up
+  levels: null
+- name: rldscs
+  units: W m-2
+  raw_variables: [LW_clrsky_flux_dn_at_model_bot]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: down
+  levels: null
+# --- Cloud water paths ---
+# CMIP6 clwvi = "condensed water path" = liquid + ice (per CMIP6_Amon.json comment).
+# In EAMxx, LiqWaterPath is liquid-only (equiv. EAM TGCLDLWP), and
+# IceWaterPath is ice-only (equiv. EAM TGCLDIWP). Their sum gives the
+# total condensed water path required by CMIP6 clwvi.
+# By contrast, EAM uses TGCLDCWP (total = liquid + ice) directly for clwvi.
+- name: clwvi
+  units: kg m-2
+  raw_variables: [LiqWaterPath, IceWaterPath]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: LiqWaterPath + IceWaterPath
+  positive: null
+  levels: null
+- name: clivi
+  units: kg m-2
+  raw_variables: [IceWaterPath]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- Aerosol optical depth ---
+- name: od550aer
+  units: "1"
+  raw_variables: [AerosolOpticalDepth550nm]
+  table: CMIP6_AERmon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: null
+# --- 3D pressure-level variables (requires data interpolated to plev) ---
+- name: ta
+  units: K
+  raw_variables: [T_mid]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: { name: plev19, units: Pa, e3sm_axis_name: plev }
+- name: hus
+  units: "1"
+  raw_variables: [qv]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: { name: plev19, units: Pa, e3sm_axis_name: plev }
+- name: hur
+  units: "%"
+  raw_variables: [RelativeHumidity]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: { name: plev19, units: Pa, e3sm_axis_name: plev }
+- name: wap
+  units: Pa s-1
+  raw_variables: [omega]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: { name: plev19, units: Pa, e3sm_axis_name: plev }
+- name: zg
+  units: m
+  raw_variables: [z_mid]
+  table: CMIP6_Amon.json
+  unit_conversion: null
+  formula: null
+  positive: null
+  levels: { name: plev19, units: Pa, e3sm_axis_name: plev }

e3sm_to_cmip/cmor_handlers/utils.py

@@ -4,7 +4,6 @@
 from importlib.machinery import SourceFileLoader
 from typing import Any, Literal, get_args
 
-import pandas as pd
 import yaml
 
 from e3sm_to_cmip import (
@@ -256,23 +255,21 @@ def _get_handlers_from_yaml() -> dict[str, list[dict[str, Any]]]:
     with open(HANDLER_DEFINITIONS_PATH, "r") as infile:
         handlers_file = yaml.load(infile, yaml.SafeLoader)
 
-    df_in = pd.DataFrame.from_dict(handlers_file)
-
     handlers = defaultdict(list)
-    for row in df_in.itertuples():
+    for entry in handlers_file:
         var_handler = VarHandler(
-            name=row.name,  # type: ignore
-            units=row.units,  # type: ignore
-            raw_variables=row.raw_variables,  # type: ignore
-            table=row.table,  # type: ignore
-            formula=row.formula,  # type: ignore
-            unit_conversion=row.unit_conversion,  # type: ignore
-            positive=row.positive,  # type: ignore
-            levels=row.levels,  # type: ignore
+            name=entry["name"],
+            units=entry["units"],
+            raw_variables=entry["raw_variables"],
+            table=entry["table"],
+            formula=entry.get("formula"),
+            unit_conversion=entry.get("unit_conversion"),
+            positive=entry.get("positive"),
+            levels=entry.get("levels"),
         ).to_dict()
-        handlers[row.name].append(var_handler)
+        handlers[entry["name"]].append(var_handler)
 
-    return dict(handlers)  # type: ignore
+    return dict(handlers)
 
 
 def _get_handlers_from_modules(path: str) -> dict[str, list[dict[str, Any]]]: