diff --git a/doc/water/index.rst b/doc/water/index.rst
index f8a4a8e1b9..a5ad94f021 100644
--- a/doc/water/index.rst
+++ b/doc/water/index.rst
@@ -96,6 +96,9 @@ Build and run
 .. automodule:: message_ix_models.model.water.build
    :members:
 
+.. automodule:: message_ix_models.model.water.config
+   :members:
+
 Data preparation
 ----------------
 
diff --git a/doc/whatsnew.rst b/doc/whatsnew.rst
index ec052a8413..c42d62c50a 100644
--- a/doc/whatsnew.rst
+++ b/doc/whatsnew.rst
@@ -6,6 +6,7 @@ Next release
 
 - Add IAMC code list :class:`~.iamc.structure.CL_SCENARIO_DIAGNOSTIC` (:pull:`501`).
 - New module :ref:`tools-newclimate` (:pull:`499`).
+- Add :class:`.model.water.Config` to collect water module settings (:pull:`509`).
 - Add :doc:`/api/model-bmt` (:pull:`433`).
 
   - Add
diff --git a/message_ix_models/model/water/__init__.py b/message_ix_models/model/water/__init__.py
index 46583ecf73..9bdedaf45f 100644
--- a/message_ix_models/model/water/__init__.py
+++ b/message_ix_models/model/water/__init__.py
@@ -1,4 +1,5 @@
+from .config import Config
 from .data import demands, water_supply
 from .utils import read_config
 
-__all__ = ["demands", "read_config", "water_supply"]
+__all__ = ["Config", "demands", "read_config", "water_supply"]
diff --git a/message_ix_models/model/water/build.py b/message_ix_models/model/water/build.py
index 795b65a397..ebb2989b05 100644
--- a/message_ix_models/model/water/build.py
+++ b/message_ix_models/model/water/build.py
@@ -10,6 +10,7 @@
 from message_ix_models.model.structure import get_codes
 from message_ix_models.util import broadcast, package_data_path
 
+from .config import Config
 from .utils import filter_basins_by_region, read_config
 
 log = logging.getLogger(__name__)
@@ -184,9 +185,10 @@ def cat_tec_cooling_calib(
             - 'tec': Name of the cooling technology.
         - regions_df: A list of unique region nodes from the scenario.
     """
+    cfg = Config.from_context(context)
     FILE1 = (
         "cooltech_cost_and_shares_"
-        + (f"ssp_msg_{context.regions}" if context.type_reg == "global" else "country")
+        + (f"ssp_msg_{context.regions}" if cfg.type_reg == "global" else "country")
         + ".csv"
     )
     path1 = package_data_path("water", "ppl_cooling_tech", FILE1)
@@ -288,7 +290,9 @@ def get_spec(context: Context) -> Mapping[str, ScenarioInfo]:
         # Elements to add
         add.set[set_name].extend(config.get("add", []))
 
-    if context.nexus_set == "nexus":
+    cfg = Config.from_context(context)
+
+    if cfg.nexus_set == "nexus":
         # Merge technology.yaml with set.yaml
         context["water set"]["nexus"]["technology"]["add"] = context[
             "water technology"
@@ -311,7 +315,7 @@ def get_spec(context: Context) -> Mapping[str, ScenarioInfo]:
 
         # Share commodity for groundwater
         results = {}
-        df_node = context.all_nodes
+        df_node = cfg.all_nodes
         n = len(df_node.values)
 
         d = {
@@ -550,6 +554,7 @@ def map_basin(context: Context) -> Mapping[str, ScenarioInfo]:
 
     # define an empty dictionary
     results = {}
+    cfg = Config.from_context(context)
     # read csv file for basin names and region mapping
     # reading basin_delineation
     FILE = f"basins_by_region_simpl_{context.regions}.csv"
@@ -564,8 +569,8 @@ def map_basin(context: Context) -> Mapping[str, ScenarioInfo]:
     df["node"] = "B" + df["BCU_name"].astype(str)
     df["mode"] = "M" + df["BCU_name"].astype(str)
     df["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df["REGION"].astype(str)
     )
 
@@ -580,9 +585,9 @@ def map_basin(context: Context) -> Mapping[str, ScenarioInfo]:
 
     results["map_node"] = nodes
 
-    context.all_nodes = df["node"]
+    cfg.all_nodes = df["node"]
     # Store the filtered basin names for use in other functions
-    context.valid_basins = set(df["BCU_name"].astype(str))
+    cfg.valid_basins = set(df["BCU_name"].astype(str))
 
     for set_name, config in results.items():
         # Sets to add
@@ -604,7 +609,7 @@ def main(context: Context, scenario, **options):
 
     log.info("Set up MESSAGEix-Nexus")
 
-    if context.nexus_set == "nexus":
+    if Config.from_context(context).nexus_set == "nexus":
         # Add water balance
         spec = map_basin(context)
 
diff --git a/message_ix_models/model/water/cli.py b/message_ix_models/model/water/cli.py
index d85b3446fb..b0d6709d3f 100644
--- a/message_ix_models/model/water/cli.py
+++ b/message_ix_models/model/water/cli.py
@@ -8,6 +8,8 @@
 from message_ix_models.model.structure import get_codes
 from message_ix_models.util.click import common_params, scenario_param
 
+from .config import Config
+
 if TYPE_CHECKING:
     from message_ix import Scenario
 log = logging.getLogger(__name__)
@@ -39,6 +41,8 @@ def water_ini(context: "Context", regions, time):
 
     from .utils import read_config
 
+    config = Config.from_context(context)
+
     # Ensure water model configuration is loaded
     read_config(context)
     if not context.scenario_info:
@@ -61,9 +65,9 @@ def water_ini(context: "Context", regions, time):
         regions = "R12"
     # add an attribute to distinguish country models
     if regions in ["R11", "R12", "R14", "R32", "RCP"]:
-        context.type_reg = "global"
+        config.type_reg = "global"
     else:
-        context.type_reg = "country"
+        config.type_reg = "country"
     context.regions = regions
 
     # create a mapping ISO code :
@@ -71,10 +75,10 @@ def water_ini(context: "Context", regions, time):
     # only needed for 1-country models
     n_codes = get_codes(f"node/{context.regions}")
     nodes = list(map(str, n_codes[n_codes.index(Code(id="World"))].child))
-    if context.type_reg == "country":
+    if config.type_reg == "country":
         map_ISO_c = {context.regions: nodes[0]}
-        context.map_ISO_c = map_ISO_c
-        log.info(f"mapping {context.map_ISO_c[context.regions]}")
+        config.map_ISO_c = map_ISO_c
+        log.info(f"mapping {config.map_ISO_c[context.regions]}")
 
     # deinfe the timestep
     if not time:
@@ -82,13 +86,13 @@ def water_ini(context: "Context", regions, time):
         time = sc_ref.set("time")
         sub_time = list(time[time != "year"])
         if len(sub_time) == 0:
-            context.time = ["year"]
+            config.time = ["year"]
 
         else:
-            context.time = sub_time
+            config.time = sub_time
     else:
-        context.time = [time]
-    log.info(f"Using the following time-step for the water module: {context.time}")
+        config.time = [time]
+    log.info(f"Using the following time-step for the water module: {config.time}")
 
     # setting the time information in context
 
@@ -163,12 +167,11 @@ def nexus_cli(
     water balance linking different water demands to supply.
     """
     # Set basin filtering configuration on context
-    context.reduced_basin = reduced_basin
-    if filter_list:
-        context.filter_list = list(filter_list)
-    if num_basins is not None:
-        context.num_basins = num_basins
-    context.basin_selection = basin_selection
+    config = Config.from_context(context)
+    config.reduced_basin = reduced_basin
+    config.filter_list = list(filter_list or [])
+    config.num_basins = num_basins
+    config.basin_selection = basin_selection
 
     nexus(context, regions, rcps, sdgs, rels, macro)
 
@@ -193,17 +196,18 @@ def nexus(context: "Context", regions, rcps, sdgs, rels, macro=False):
         Specifies the reliability of hydrological data ['low','mid','high']
     """
     # add input information to the class context
-    context.nexus_set = "nexus"
+    config = Config.from_context(context)
+    config.nexus_set = "nexus"
     if not context.regions:
         context.regions = regions
 
-    context.RCP = rcps
-    context.SDG = sdgs
-    context.REL = rels
+    config.RCP = rcps
+    config.SDG = sdgs
+    config.REL = rels
 
     log.info(
-        f"SSP assumption is {context.ssp}. SDG is {context.SDG}. "
-        f"RCP is {context.RCP}. REL is {context.REL}."
+        f"SSP assumption is {context.ssp}. SDG is {config.SDG}. "
+        f"RCP is {config.RCP}. REL is {config.REL}."
     )
 
     from .build import main as build
@@ -298,12 +302,13 @@ def cooling(
         Specifies the climate scenario used ['no_climate','6p0','2p6']
 
     """
-    context.nexus_set = "cooling"
-    context.RCP = rcps
-    context.REL = rels
+    config = Config.from_context(context)
+    config.nexus_set = "cooling"
+    config.RCP = rcps
+    config.REL = rels
 
     log.info(
-        f"SSP assumption is {context.ssp}. RCP is {context.RCP}. REL is {context.REL}."
+        f"SSP assumption is {context.ssp}. RCP is {config.RCP}. REL is {config.REL}."
     )
 
     from .build import main as build
diff --git a/message_ix_models/model/water/config.py b/message_ix_models/model/water/config.py
new file mode 100644
index 0000000000..251bd7e089
--- /dev/null
+++ b/message_ix_models/model/water/config.py
@@ -0,0 +1,65 @@
+from dataclasses import dataclass, field
+from typing import TYPE_CHECKING, Any, Literal
+
+from message_ix_models.util.config import ConfigHelper
+
+if TYPE_CHECKING:
+    from message_ix_models import Context
+
+
+@dataclass
+class Config(ConfigHelper):
+    """Settings for :mod:`message_ix_models.model.water`."""
+
+    #: Water build variant.
+    nexus_set: Literal["cooling", "nexus"] = "nexus"
+
+    #: Climate scenario used for water availability and cooling data.
+    RCP: str = "no_climate"
+
+    #: Water SDG policy setting.
+    SDG: str = "baseline"
+
+    #: Hydrological-data reliability setting.
+    REL: str = "low"
+
+    #: Time slices handled by the water module.
+    time: list[str] = field(default_factory=lambda: ["year"])
+
+    #: Whether :attr:`regions` names a global node codelist or a single country.
+    type_reg: Literal["country", "global"] = "global"
+
+    #: Single-country model mapping from region code to ISO code.
+    map_ISO_c: dict[str, str] = field(default_factory=dict)
+
+    #: Enable basin filtering.
+    reduced_basin: bool = False
+
+    #: Basins to add to the automatic reduced-basin selection.
+    filter_list: list[str] = field(default_factory=list)
+
+    #: Number of basins per region to keep when :attr:`reduced_basin` is enabled.
+    num_basins: int | None = None
+
+    #: Automatic reduced-basin selection method.
+    basin_selection: Literal["first_k", "stress"] = "first_k"
+
+    #: Basin names retained after optional filtering.
+    valid_basins: set[str] = field(default_factory=set)
+
+    #: Water-module basin node labels, populated during structural mapping.
+    all_nodes: Any = None
+
+    @classmethod
+    def from_context(cls, context: "Context") -> "Config":
+        """Return the shared water configuration for `context`.
+
+        Create `context.water` if missing, or convert a mapping to `Config`.
+        Repeated calls return the same `Config` instance.
+        """
+        if "water" not in context:
+            context["water"] = cls()
+        elif isinstance(context["water"], dict):
+            context["water"] = cls.from_dict(context["water"])
+
+        return context["water"]
diff --git a/message_ix_models/model/water/data/__init__.py b/message_ix_models/model/water/data/__init__.py
index d118c174a5..9463aad062 100644
--- a/message_ix_models/model/water/data/__init__.py
+++ b/message_ix_models/model/water/data/__init__.py
@@ -9,6 +9,7 @@
 import pandas as pd
 
 from message_ix_models import ScenarioInfo
+from message_ix_models.model.water.config import Config
 from message_ix_models.util import add_par_data
 
 from .demands import add_irrigation_demand, add_sectoral_demands, add_water_availability
@@ -57,12 +58,13 @@ def add_data(scenario, context: "Context", dry_run=False):
 
     info = ScenarioInfo(scenario)
     context["water build info"] = info
+    cfg = Config.from_context(context)
 
     data_funcs: list[DataFunc] = (
         [add_water_supply, cool_tech, non_cooling_tec]
-        if context.nexus_set == "cooling"
+        if cfg.nexus_set == "cooling"
         else DATA_FUNCTIONS
-        if context.type_reg == "global"
+        if cfg.type_reg == "global"
         else DATA_FUNCTIONS_COUNTRY
     )
 
diff --git a/message_ix_models/model/water/data/demands.py b/message_ix_models/model/water/data/demands.py
index 2bf9a1c140..6d4c42a065 100644
--- a/message_ix_models/model/water/data/demands.py
+++ b/message_ix_models/model/water/data/demands.py
@@ -9,6 +9,7 @@
 import xarray as xr
 from message_ix import make_df
 
+from message_ix_models.model.water.config import Config
 from message_ix_models.model.water.utils import KM3_TO_MCM
 from message_ix_models.util import broadcast, package_data_path
 
@@ -178,12 +179,13 @@ def add_sectoral_demands(context: "Context") -> dict[str, pd.DataFrame]:
     results = {}
 
     # Reference to the water configuration
+    cfg = Config.from_context(context)
     info = context["water build info"]
     year_vtgs = tuple(range(2010, info.Y[0], 5))
 
     # defines path to read in demand data
     region = f"{context.regions}"
-    sub_time = context.time
+    sub_time = cfg.time
     path = package_data_path("water", "demands", "harmonized", region, ".")
     # make sure all of the csvs have format, otherwise it might not work
     list_of_csvs = list(path.glob("ssp2_regional_*.csv"))
@@ -220,14 +222,14 @@ def add_sectoral_demands(context: "Context") -> dict[str, pd.DataFrame]:
     df_dmds["time"] = "year"
 
     # Filter to only include basins that exist after basin filtering
-    df_dmds = df_dmds[df_dmds["node"].isin(context.valid_basins)]
+    df_dmds = df_dmds[df_dmds["node"].isin(cfg.valid_basins)]
 
     # Write final interpolated values as csv
     # df2_f.to_csv('final_interpolated_values.csv')
 
     # if we are using sub-annual timesteps we replace the rural and municipal
     # withdrawals and return flows with monthly data and also add industrial
-    if "year" not in context.time:
+    if "year" not in cfg.time:
         PATH = package_data_path(
             "water", "demands", "harmonized", region, "ssp2_m_water_demands.csv"
         )
@@ -245,7 +247,7 @@ def add_sectoral_demands(context: "Context") -> dict[str, pd.DataFrame]:
         df_m.columns = pd.Index(["year", "node", "variable", "value", "time"])
 
         # Filter monthly data to only include valid basins
-        df_m = df_m[df_m["node"].isin(context.valid_basins)]
+        df_m = df_m[df_m["node"].isin(cfg.valid_basins)]
 
         # remove yearly parts from df_dms
         df_dmds = df_dmds[
@@ -307,9 +309,9 @@ def add_sectoral_demands(context: "Context") -> dict[str, pd.DataFrame]:
         ]
     )
 
-    if context.SDG != "baseline":
+    if cfg.SDG != "baseline":
         # only if SDG exactly equal to SDG, otherwise other policies are possible
-        pol_scen = context.SDG
+        pol_scen = cfg.SDG
         if pol_scen == "SDG":
             # reading basin mapping to countries
             FILE2 = f"basins_country_{context.regions}.csv"
@@ -769,6 +771,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
     """
 
     # Reference to the water configuration
+    cfg = Config.from_context(context)
     info = context["water build info"]
     # reading sample for assiging basins
     PATH = package_data_path(
@@ -777,15 +780,15 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
     df_x = pd.read_csv(PATH)
 
     # Filter to only include valid basins
-    df_x = df_x[df_x["BCU_name"].isin(context.valid_basins)]
+    df_x = df_x[df_x["BCU_name"].isin(cfg.valid_basins)]
 
-    if "year" in context.time:
+    if "year" in cfg.time:
         # Adding freshwater supply constraints
         # Reading data, the data is spatially and temprally aggregated from GHMs
         path1 = package_data_path(
             "water",
             "availability",
-            f"qtot_5y_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qtot_5y_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
         # Read rcp 2.6 data
         df_sw = pd.read_csv(path1)
@@ -794,7 +797,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
         # Filter rows to valid basins using index positions from full list
         full_basin_df = pd.read_csv(PATH)
         valid_indices = full_basin_df[
-            full_basin_df["BCU_name"].isin(context.valid_basins)
+            full_basin_df["BCU_name"].isin(cfg.valid_basins)
         ].index
         df_sw = df_sw.iloc[valid_indices]  # Keep only rows for valid basins
         df_sw.reset_index(drop=True, inplace=True)
@@ -817,7 +820,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
         path1 = package_data_path(
             "water",
             "availability",
-            f"qr_5y_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qr_5y_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
 
         # Read groundwater data
@@ -847,7 +850,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
         path1 = package_data_path(
             "water",
             "availability",
-            f"qtot_5y_m_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qtot_5y_m_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
         df_sw = pd.read_csv(path1)
         df_sw.drop(["Unnamed: 0"], axis=1, inplace=True)
@@ -855,7 +858,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
         # Filter rows to valid basins
         full_basin_df = pd.read_csv(PATH)
         valid_indices = full_basin_df[
-            full_basin_df["BCU_name"].isin(context.valid_basins)
+            full_basin_df["BCU_name"].isin(cfg.valid_basins)
         ].index
         df_sw = df_sw.iloc[valid_indices]
         df_sw.reset_index(drop=True, inplace=True)
@@ -878,7 +881,7 @@ def read_water_availability(context: "Context") -> Sequence[pd.DataFrame]:
         path1 = package_data_path(
             "water",
             "availability",
-            f"qr_5y_m_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qr_5y_m_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
         df_gw = pd.read_csv(path1)
         df_gw.drop(["Unnamed: 0"], axis=1, inplace=True)
diff --git a/message_ix_models/model/water/data/infrastructure.py b/message_ix_models/model/water/data/infrastructure.py
index 169e398eed..d8051348ec 100644
--- a/message_ix_models/model/water/data/infrastructure.py
+++ b/message_ix_models/model/water/data/infrastructure.py
@@ -9,6 +9,7 @@
 from message_ix import make_df
 
 from message_ix_models import Context, ScenarioInfo
+from message_ix_models.model.water.config import Config
 from message_ix_models.model.water.utils import (
     ANNUAL_CAPACITY_FACTOR,
     KM3_TO_MCM,
@@ -273,11 +274,12 @@ def add_infrastructure_techs(context: "Context") -> dict[str, pd.DataFrame]:
         ``context["water build info"]``, plus the additional year 2010.
     """
     # Reference to the water configuration
+    cfg = Config.from_context(context)
     info = context["water build info"]
 
     # define an empty dictionary
     results = {}
-    sub_time = pd.Series(context.time)
+    sub_time = pd.Series(cfg.time)
     # load the scenario from context
     scen = context.get_scenario()
 
@@ -293,14 +295,14 @@ def add_infrastructure_techs(context: "Context") -> dict[str, pd.DataFrame]:
     df_node = pd.read_csv(PATH)
 
     # Filter to only valid basins (already filtered in map_basin)
-    df_node = df_node[df_node["BCU_name"].isin(context.valid_basins)]
+    df_node = df_node[df_node["BCU_name"].isin(cfg.valid_basins)]
 
     # Assigning proper nomenclature
     df_node["node"] = "B" + df_node["BCU_name"].astype(str)
     df_node["mode"] = "M" + df_node["BCU_name"].astype(str)
     df_node["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_node["REGION"].astype(str)
     )
 
@@ -322,7 +324,7 @@ def add_infrastructure_techs(context: "Context") -> dict[str, pd.DataFrame]:
     df_elec = df[df["incmd"] == "electr"].reset_index()
 
     inp_df = start_creating_input_dataframe(
-        sdg=context.SDG,
+        sdg=cfg.SDG,
         df_node=df_node,
         df_non_elec=df_non_elec,
         df_dist=df_dist,
@@ -383,9 +385,7 @@ def add_infrastructure_techs(context: "Context") -> dict[str, pd.DataFrame]:
         )
 
     # Process distribution outputs using helper function
-    dist_out = _make_dist_output(
-        df_out_dist, scenario_info, df_node, sub_time, context.SDG
-    )
+    dist_out = _make_dist_output(df_out_dist, scenario_info, df_node, sub_time, cfg.SDG)
     out_df = pd.concat([out_df, dist_out]) if not dist_out.empty else out_df
 
     results["output"] = out_df
@@ -496,7 +496,7 @@ def add_infrastructure_techs(context: "Context") -> dict[str, pd.DataFrame]:
     df_var = df_inv[~df_inv["tec"].isin(techs)]
     df_var_dist = df_inv[df_inv["tec"].isin(techs)]
 
-    if context.SDG != "baseline":
+    if cfg.SDG != "baseline":
         for index, rows in df_var.iterrows():
             # Check if this is a dummy technology
             use_same_year = is_dummy_technology(rows)
@@ -646,6 +646,7 @@ def prepare_input_dataframe(
     df_elec: pd.DataFrame,
 ) -> defaultdict[Any, list]:
     result_dc = defaultdict(list)
+    cfg = Config.from_context(context)
     # Unit 1 KWh/m^3 = 10^3 GWh/Km^3 = 1 GWh/MCM,
     # Parkinson et al.
     # which is the only explanation as to how the model solved.
@@ -654,7 +655,7 @@ def prepare_input_dataframe(
             # Check if this is a dummy technology (for distribution techs)
             use_same_year = is_dummy_technology(rows)
 
-            if context.SDG != "baseline":
+            if cfg.SDG != "baseline":
                 inp = make_df(
                     "input",
                     technology=rows["tec"],
@@ -774,7 +775,8 @@ def add_desalination(context: "Context") -> dict[str, pd.DataFrame]:
     """
     # define an empty dictionary
     results = {}
-    sub_time = pd.Series(context.time)
+    cfg = Config.from_context(context)
+    sub_time = pd.Series(cfg.time)
     # Reference to the water configuration
     info = context["water build info"]
 
@@ -801,7 +803,7 @@ def add_desalination(context: "Context") -> dict[str, pd.DataFrame]:
     df_desal = pd.read_csv(path)
     df_hist = pd.read_csv(path2)
     df_proj = pd.read_csv(path3)
-    df_proj = df_proj[df_proj["rcp"] == f"{context.RCP}"]
+    df_proj = df_proj[df_proj["rcp"] == f"{cfg.RCP}"]
     df_proj = df_proj[~(df_proj["year"] == 2065) & ~(df_proj["year"] == 2075)]
     df_proj.reset_index(inplace=True, drop=True)
     df_proj = df_proj[df_proj["year"].isin(info.Y)]
@@ -813,20 +815,20 @@ def add_desalination(context: "Context") -> dict[str, pd.DataFrame]:
     df_node = pd.read_csv(PATH)
 
     # Filter to only valid basins (already filtered in map_basin)
-    df_node = df_node[df_node["BCU_name"].isin(context.valid_basins)]
+    df_node = df_node[df_node["BCU_name"].isin(cfg.valid_basins)]
 
     # Assigning proper nomenclature
     df_node["node"] = "B" + df_node["BCU_name"].astype(str)
     df_node["mode"] = "M" + df_node["BCU_name"].astype(str)
     df_node["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_node["REGION"].astype(str)
     )
 
     # Filter to basins that exist after filtering
-    df_hist = df_hist[df_hist["BCU_name"].isin(context.valid_basins)]
-    df_proj = df_proj[df_proj["BCU_name"].isin(context.valid_basins)]
+    df_hist = df_hist[df_hist["BCU_name"].isin(cfg.valid_basins)]
+    df_proj = df_proj[df_proj["BCU_name"].isin(cfg.valid_basins)]
 
     # output dataframe linking to desal tech types
     out_df = (
diff --git a/message_ix_models/model/water/data/irrigation.py b/message_ix_models/model/water/data/irrigation.py
index 5626aa2537..48c2f69a18 100644
--- a/message_ix_models/model/water/data/irrigation.py
+++ b/message_ix_models/model/water/data/irrigation.py
@@ -4,6 +4,7 @@
 from message_ix import make_df
 
 from message_ix_models import Context
+from message_ix_models.model.water.config import Config
 from message_ix_models.util import broadcast, package_data_path
 
 
@@ -27,6 +28,7 @@ def add_irr_structure(context: "Context") -> dict[str, pd.DataFrame]:
 
     # define an empty dictionary
     results = {}
+    cfg = Config.from_context(context)
 
     # reading basin_delineation
     FILE2 = f"basins_by_region_simpl_{context.regions}.csv"
@@ -34,14 +36,14 @@ def add_irr_structure(context: "Context") -> dict[str, pd.DataFrame]:
     df_node = pd.read_csv(PATH)
 
     # Filter to only include valid basins
-    df_node = df_node[df_node["BCU_name"].isin(context.valid_basins)]
+    df_node = df_node[df_node["BCU_name"].isin(cfg.valid_basins)]
 
     # Assigning proper nomenclature
     df_node["node"] = "B" + df_node["BCU_name"].astype(str)
     df_node["mode"] = "M" + df_node["BCU_name"].astype(str)
     df_node["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_node["REGION"].astype(str)
     )
 
diff --git a/message_ix_models/model/water/data/water_for_ppl.py b/message_ix_models/model/water/data/water_for_ppl.py
index a63438c9df..8cb84c9014 100644
--- a/message_ix_models/model/water/data/water_for_ppl.py
+++ b/message_ix_models/model/water/data/water_for_ppl.py
@@ -9,6 +9,7 @@
 from message_ix import make_df
 
 from message_ix_models import Context
+from message_ix_models.model.water.config import Config
 from message_ix_models.model.water.data.water_supply import map_basin_region_wat
 from message_ix_models.model.water.utils import get_vintage_and_active_years
 from message_ix_models.util import (
@@ -43,11 +44,12 @@ def _load_scenario_and_cooling_data(
     scen : Scenario object
     cost_share_df : DataFrame with cost and share data
     """
+    cfg = Config.from_context(context)
     # File paths
     tech_perf_path = package_data_path(
         "water", "ppl_cooling_tech", "tech_water_performance_ssp_msg.csv"
     )
-    suffix = "ssp_msg_" + context.regions if context.type_reg == "global" else "country"
+    suffix = "ssp_msg_" + context.regions if cfg.type_reg == "global" else "country"
     cost_share_file = f"cooltech_cost_and_shares_{suffix}.csv"
     cost_share_path = package_data_path("water", "ppl_cooling_tech", cost_share_file)
     basin_path = package_data_path(
@@ -59,8 +61,8 @@ def _load_scenario_and_cooling_data(
     df_node["node"] = "B" + df_node["BCU_name"].astype(str)
     df_node["mode"] = "M" + df_node["BCU_name"].astype(str)
     df_node["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_node["REGION"].astype(str)
     )
     node_region = df_node["region"].unique()
@@ -152,7 +154,8 @@ def _compute_cooling_rates(input_cool: pd.DataFrame) -> pd.DataFrame:
 
 def _make_input_params(input_cool: pd.DataFrame, context: "Context") -> pd.DataFrame:
     """Generate input parameter DataFrame."""
-    commodity = "surfacewater" if context.nexus_set == "nexus" else "freshwater"
+    cfg = Config.from_context(context)
+    commodity = "surfacewater" if cfg.nexus_set == "nexus" else "freshwater"
 
     # Electricity input for parasitic demand
     electr = input_cool[input_cool["parasitic_electricity_demand_fraction"] > 0].copy()
@@ -325,18 +328,19 @@ def _make_technical_lifetime(
 
 def _make_capacity_factor(inp: pd.DataFrame, context: "Context") -> pd.DataFrame:
     """Generate capacity_factor parameter with optional climate impacts."""
+    cfg = Config.from_context(context)
     cap_fact = make_matched_dfs(inp, capacity_factor=1)["capacity_factor"]
     cap_fact["unit"] = "-"  # capacity_factor is dimensionless
     cap_fact = cap_fact.drop_duplicates().reset_index(drop=True)
 
-    if context.RCP == "no_climate":
+    if cfg.RCP == "no_climate":
         return cap_fact
 
     # Apply climate impacts on freshwater cooling
     impact_path = package_data_path(
         "water",
         "ppl_cooling_tech",
-        f"power_plant_cooling_impact_MESSAGE_{context.regions}_{context.RCP}.csv",
+        f"power_plant_cooling_impact_MESSAGE_{context.regions}_{cfg.RCP}.csv",
     )
     df_impact = pd.read_csv(impact_path)
 
@@ -569,7 +573,7 @@ def _add_nexus_params(
     results: dict, context: "Context", node_region: list, info
 ) -> None:
     """Add basin-region distribution for nexus mode."""
-    if context.nexus_set != "nexus":
+    if Config.from_context(context).nexus_set != "nexus":
         return
 
     df_sw = map_basin_region_wat(context)
@@ -797,7 +801,11 @@ def non_cooling_tec(context: "Context", scenario=None) -> dict[str, pd.DataFrame
 
     merged = output_data.merge(non_cool, on="technology", how="right").dropna()
 
-    commodity = "surfacewater" if context.nexus_set == "nexus" else "freshwater"
+    commodity = (
+        "surfacewater"
+        if Config.from_context(context).nexus_set == "nexus"
+        else "freshwater"
+    )
 
     inp = make_df(
         "input",
diff --git a/message_ix_models/model/water/data/water_supply.py b/message_ix_models/model/water/data/water_supply.py
index 21e82ec0ad..707c084e71 100644
--- a/message_ix_models/model/water/data/water_supply.py
+++ b/message_ix_models/model/water/data/water_supply.py
@@ -5,6 +5,7 @@
 from message_ix import make_df
 
 from message_ix_models import Context, ScenarioInfo
+from message_ix_models.model.water.config import Config
 from message_ix_models.model.water.data.demands import read_water_availability
 from message_ix_models.model.water.utils import (
     ANNUAL_CAPACITY_FACTOR,
@@ -36,16 +37,17 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
     -------
         data : pandas.DataFrame
     """
+    cfg = Config.from_context(context)
     info = context["water build info"]
 
-    if "year" in context.time:
+    if "year" in cfg.time:
         PATH = package_data_path(
             "water", "delineation", f"basins_by_region_simpl_{context.regions}.csv"
         )
         df_x_full = pd.read_csv(PATH)
 
         # Get positional indices of valid basins from the unfiltered list
-        valid_mask = df_x_full["BCU_name"].isin(context.valid_basins)
+        valid_mask = df_x_full["BCU_name"].isin(cfg.valid_basins)
         valid_indices = df_x_full[valid_mask].index
         df_x = df_x_full[valid_mask].reset_index(drop=True)
 
@@ -54,7 +56,7 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
         path1 = package_data_path(
             "water",
             "availability",
-            f"qtot_5y_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qtot_5y_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
 
         df_sw = pd.read_csv(path1)
@@ -64,8 +66,8 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
         df_sw = df_sw.iloc[valid_indices].reset_index(drop=True)
         df_sw["BCU_name"] = df_x["BCU_name"]
         df_sw["MSGREG"] = (
-            context.map_ISO_c[context.regions]
-            if context.type_reg == "country"
+            cfg.map_ISO_c[context.regions]
+            if cfg.type_reg == "country"
             else f"{context.regions}_" + df_sw["BCU_name"].str.split("|").str[-1]
         )
 
@@ -93,7 +95,7 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
         path3 = package_data_path(
             "water",
             "availability",
-            f"qtot_5y_m_{context.RCP}_{context.REL}_{context.regions}.csv",
+            f"qtot_5y_m_{cfg.RCP}_{cfg.REL}_{context.regions}.csv",
         )
         df_sw = pd.read_csv(path3)
 
@@ -104,7 +106,7 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
         df_x_full = pd.read_csv(PATH)
 
         # Get positional indices of valid basins from the unfiltered list
-        valid_mask = df_x_full["BCU_name"].isin(context.valid_basins)
+        valid_mask = df_x_full["BCU_name"].isin(cfg.valid_basins)
         valid_indices = df_x_full[valid_mask].index
         df_x = df_x_full[valid_mask].reset_index(drop=True)
 
@@ -113,8 +115,8 @@ def map_basin_region_wat(context: "Context") -> pd.DataFrame:
         df_sw["BCU_name"] = df_x["BCU_name"]
 
         df_sw["MSGREG"] = (
-            context.map_ISO_c[context.regions]
-            if context.type_reg == "country"
+            cfg.map_ISO_c[context.regions]
+            if cfg.type_reg == "country"
             else f"{context.regions}_" + df_sw["BCU_name"].str.split("|").str[-1]
         )
 
@@ -161,6 +163,7 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
     results = {}
 
     # Reference to the water configuration
+    cfg = Config.from_context(context)
     info = context["water build info"]
     # load the scenario from context
     scen = context.get_scenario()
@@ -168,7 +171,7 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
     fut_year = info.Y
     year_wat = (*range(2010, info.Y[0] + 1, 5), *info.Y)
     last_vtg_yr = info.Y[0] - 5
-    sub_time = context.time
+    sub_time = cfg.time
     scen_info = ScenarioInfo(scen)
 
     print(" future year = ", fut_year)
@@ -187,8 +190,8 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
     df_node["node"] = "B" + df_node["BCU_name"].astype(str)
     df_node["mode"] = "M" + df_node["BCU_name"].astype(str)
     df_node["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_node["REGION"].astype(str)
     )
 
@@ -200,8 +203,8 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
     PATH1 = package_data_path("water", "availability", FILE1)
     df_gwt = pd.read_csv(PATH1)
     df_gwt["region"] = (
-        context.map_ISO_c[context.regions]
-        if context.type_reg == "country"
+        cfg.map_ISO_c[context.regions]
+        if cfg.type_reg == "country"
         else f"{context.regions}_" + df_gwt["REGION"].astype(str)
     )
 
@@ -211,11 +214,11 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
     df_hist = pd.read_csv(PATH2)
 
     # Filter to only include valid basins (nexus mode only)
-    if context.nexus_set == "nexus":
-        df_hist = df_hist[df_hist["BCU_name"].isin(context.valid_basins)]
+    if cfg.nexus_set == "nexus":
+        df_hist = df_hist[df_hist["BCU_name"].isin(cfg.valid_basins)]
         df_hist["BCU_name"] = "B" + df_hist["BCU_name"].astype(str)
 
-    if context.nexus_set == "cooling":
+    if cfg.nexus_set == "cooling":
         # Add output df  for surfacewater supply for regions
         output_df = (
             make_df(
@@ -324,7 +327,7 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
 
         results["input"] = pd.concat([inp_sw, inp_gw], ignore_index=True)
 
-    elif context.nexus_set == "nexus":
+    elif cfg.nexus_set == "nexus":
         # input data frame  for slack technology balancing equality with demands
         inp = (
             make_df(
@@ -559,7 +562,7 @@ def add_water_supply(context: "Context") -> dict[str, pd.DataFrame]:
             ]
         )
 
-        if context.type_reg == "global":
+        if cfg.type_reg == "global":
             inp.loc[
                 (inp["technology"].str.contains("extract_gw_fossil"))
                 & (inp["year_act"] == 2020)
@@ -995,6 +998,7 @@ def add_e_flow(context: "Context") -> dict[str, pd.DataFrame]:
     """
     # define an empty dictionary
     results = {}
+    cfg = Config.from_context(context)
 
     info = context["water build info"]
 
@@ -1008,10 +1012,9 @@ def add_e_flow(context: "Context") -> dict[str, pd.DataFrame]:
     )
     df_x_full = pd.read_csv(PATH)
     # Index positions of valid basins in the full CSV (for positional CSV filtering)
-    valid_indices = df_x_full[df_x_full["BCU_name"].isin(context.valid_basins)].index
-    df_x = df_x_full[df_x_full["BCU_name"].isin(context.valid_basins)].reset_index(
-        drop=True
-    )
+    valid_mask = df_x_full["BCU_name"].isin(cfg.valid_basins)
+    valid_indices = df_x_full[valid_mask].index
+    df_x = df_x_full[valid_mask].reset_index(drop=True)
 
     dmd_df = make_df(
         "demand",
@@ -1026,12 +1029,12 @@ def add_e_flow(context: "Context") -> dict[str, pd.DataFrame]:
     dmd_df = dmd_df[dmd_df["year"] >= 2025].reset_index(drop=True)
     dmd_df["value"] = dmd_df["value"].apply(lambda x: x if x >= 0 else 0)
 
-    if "year" in context.time:
+    if "year" in cfg.time:
         # Reading data, the data is spatially and temporally aggregated from GHMs
         path1 = package_data_path(
             "water",
             "availability",
-            f"e-flow_{context.RCP}_{context.regions}.csv",
+            f"e-flow_{cfg.RCP}_{context.regions}.csv",
         )
         df_env = pd.read_csv(path1)
         df_env.drop(["Unnamed: 0"], axis=1, inplace=True)
@@ -1054,7 +1057,7 @@ def add_e_flow(context: "Context") -> dict[str, pd.DataFrame]:
         path1 = package_data_path(
             "water",
             "availability",
-            f"e-flow_5y_m_{context.RCP}_{context.regions}.csv",
+            f"e-flow_5y_m_{cfg.RCP}_{context.regions}.csv",
         )
         df_env = pd.read_csv(path1)
         df_env.drop(["Unnamed: 0"], axis=1, inplace=True)
@@ -1074,7 +1077,7 @@ def add_e_flow(context: "Context") -> dict[str, pd.DataFrame]:
         df_env = df_env[df_env["year"].isin(info.Y)]
 
     # Return a processed dataframe for env flow calculations
-    if context.SDG != "baseline":
+    if cfg.SDG != "baseline":
         # dataframe to put constraints on env flows
         eflow_df = make_df(
             "bound_activity_lo",
diff --git a/message_ix_models/model/water/utils.py b/message_ix_models/model/water/utils.py
index 624b316bc0..6eb49a6b37 100644
--- a/message_ix_models/model/water/utils.py
+++ b/message_ix_models/model/water/utils.py
@@ -14,6 +14,8 @@
 from message_ix_models.model.structure import get_codes
 from message_ix_models.util import load_package_data, package_data_path
 
+from .config import Config
+
 log = logging.getLogger(__name__)
 
 if TYPE_CHECKING:
@@ -117,19 +119,18 @@ def filter_basins_by_region(
     if not context:
         context = Context.get_instance(-1)
 
-    # Check if reduced basin filtering is enabled
-    reduced_basin = getattr(context, "reduced_basin", False)
+    cfg = Config.from_context(context)
 
-    if not reduced_basin:
+    if not cfg.reduced_basin:
         # No filtering, return original dataframe
         log.info("Basin filtering disabled, returning all basins")
         return df_basins
 
     # Basin filtering is enabled — run automatic selection, then augment with
     # filter_list if provided.
-    filter_list = getattr(context, "filter_list", None)
-    num_basins = getattr(context, "num_basins", None)
-    basin_selection = getattr(context, "basin_selection", "first_k")
+    filter_list = cfg.filter_list
+    num_basins = cfg.num_basins
+    basin_selection = cfg.basin_selection
 
     if num_basins is None:
         log.info(f"num_basins not set, using default n_per_region={n_per_region}")
diff --git a/message_ix_models/tests/model/water/conftest.py b/message_ix_models/tests/model/water/conftest.py
index cb1d5253c7..5f16f783b2 100644
--- a/message_ix_models/tests/model/water/conftest.py
+++ b/message_ix_models/tests/model/water/conftest.py
@@ -6,6 +6,7 @@
 
 from message_ix_models import ScenarioInfo
 from message_ix_models.model.structure import get_codes
+from message_ix_models.model.water.config import Config
 from message_ix_models.util import package_data_path
 
 REGION_CONFIG = {
@@ -52,7 +53,7 @@ def setup_valid_basins(context, regions="R12"):
     df_filtered = filter_basins_by_region(df_basins, context)
 
     # Set valid_basins as set of basin names
-    context.valid_basins = set(df_filtered["BCU_name"].astype(str))
+    Config.from_context(context).valid_basins = set(df_filtered["BCU_name"].astype(str))
 
     return context
 
@@ -70,29 +71,31 @@ def water_context(test_context, request):
 
     # Apply defaults
     test_context.regions = params.get("regions", "R11")
-    test_context.type_reg = params.get("type_reg", "global")
-    test_context.time = params.get("time", "year")
-    test_context.nexus_set = params.get("nexus_set", "nexus")
+    cfg = Config.from_context(test_context)
+    cfg.type_reg = params.get("type_reg", "global")
+    cfg.time = params.get("time", "year")
+    cfg.nexus_set = params.get("nexus_set", "nexus")
 
     # Optional attributes
     for attr in [
         "RCP",
         "REL",
         "SDG",
-        "ssp",
         "reduced_basin",
         "basin_selection",
         "num_basins",
         "filter_list",
     ]:
         if attr in params:
-            setattr(test_context, attr, params[attr])
+            setattr(cfg, attr, params[attr])
+    if "ssp" in params:
+        test_context.ssp = params["ssp"]
 
     # Node mapping for country models
-    if test_context.type_reg == "country":
+    if cfg.type_reg == "country":
         nodes = get_codes(f"node/{test_context.regions}")
         nodes = list(map(str, nodes[nodes.index("World")].child))
-        test_context.map_ISO_c = {test_context.regions: nodes[0]}
+        cfg.map_ISO_c = {test_context.regions: nodes[0]}
 
     # Set up valid_basins for basin filtering
     setup_valid_basins(test_context, regions=test_context.regions)
diff --git a/message_ix_models/tests/model/water/test_build.py b/message_ix_models/tests/model/water/test_build.py
index 92e8fcafc4..6204f904d4 100644
--- a/message_ix_models/tests/model/water/test_build.py
+++ b/message_ix_models/tests/model/water/test_build.py
@@ -5,6 +5,7 @@
 from message_ix_models.model.structure import get_codes
 from message_ix_models.model.water.build import cat_tec_cooling, get_spec, map_basin
 from message_ix_models.model.water.build import main as build
+from message_ix_models.model.water.config import Config
 
 
 @pytest.mark.xfail(reason="Temporary, for #106")
@@ -16,15 +17,16 @@ def test_build(request, test_context):
     # TODO If all water functions require these keys, set this up in a central location
     # or via default value
     # Ensure test_context has all necessary keys for build()
-    test_context.nexus_set = "nexus"
-    test_context.type_reg = "global"
-    test_context.time = "year"
+    cfg = Config.from_context(test_context)
+    cfg.nexus_set = "nexus"
+    cfg.type_reg = "global"
+    cfg.time = ["year"]
     nodes = get_codes(f"node/{test_context.regions}")
     nodes = list(map(str, nodes[nodes.index("World")].child))
     map_ISO_c = {test_context.regions: nodes[0]}
-    test_context.map_ISO_c = map_ISO_c
-    test_context.RCP = "6p0"
-    test_context.REL = "med"
+    cfg.map_ISO_c = map_ISO_c
+    cfg.RCP = "6p0"
+    cfg.REL = "med"
     test_context["water build info"] = ScenarioInfo(scenario_obj=scenario)
 
     # Code runs on the bare RES
@@ -80,9 +82,10 @@ def parametrize_for_cat_tec(request, context):
 @pytest.mark.parametrize("nexus_set", ["nexus", "cooling"])
 def test_get_spec(request, test_context, nexus_set):
     # Ensure test_context has all necessary keys for get_spec()
-    test_context.nexus_set = nexus_set
+    cfg = Config.from_context(test_context)
+    cfg.nexus_set = nexus_set
     test_context.model.regions = "R12"
-    test_context.type_reg = "global"
+    cfg.type_reg = "global"
 
     test_context = parametrize_for_cat_tec(request, test_context)
 
diff --git a/message_ix_models/tests/model/water/test_config.py b/message_ix_models/tests/model/water/test_config.py
new file mode 100644
index 0000000000..50285f101b
--- /dev/null
+++ b/message_ix_models/tests/model/water/test_config.py
@@ -0,0 +1,12 @@
+from message_ix_models.model.water.config import Config
+
+
+def test_config_from_context(test_context) -> None:
+    test_context.water = {"RCP": "2p6", "SDG": "SDG", "REL": "med"}
+
+    cfg = Config.from_context(test_context)
+
+    assert isinstance(cfg, Config)
+    assert cfg.RCP == "2p6"
+    assert cfg.SDG == "SDG"
+    assert cfg.REL == "med"
diff --git a/message_ix_models/tests/model/water/test_report.py b/message_ix_models/tests/model/water/test_report.py
index 3bfb7d001d..3e0794416e 100644
--- a/message_ix_models/tests/model/water/test_report.py
+++ b/message_ix_models/tests/model/water/test_report.py
@@ -8,6 +8,7 @@
 
 from message_ix_models import ScenarioInfo
 from message_ix_models.model.structure import get_codes
+from message_ix_models.model.water.config import Config
 from message_ix_models.model.water.report import (
     ScenarioMetadata,
     aggregate_totals,
@@ -23,8 +24,9 @@
 def test_report_full(test_context: Any, request: pytest.FixtureRequest) -> None:
     # FIXME You probably want this to be part of a common setup rather than writing
     # something like this for every test
-    test_context.time = "year"
-    test_context.type_reg = "global"
+    cfg = Config.from_context(test_context)
+    cfg.time = ["year"]
+    cfg.type_reg = "global"
     test_context.regions = "R12"
     codes = get_codes(f"node/{test_context.regions}")
     world_code = [n for n in codes if str(n) == "World"][0]
diff --git a/message_ix_models/tests/model/water/test_utils.py b/message_ix_models/tests/model/water/test_utils.py
index 0bd9e9c881..366eb3e954 100644
--- a/message_ix_models/tests/model/water/test_utils.py
+++ b/message_ix_models/tests/model/water/test_utils.py
@@ -125,14 +125,16 @@ def test_n_per_region_respected(self, stress_df):
 @pytest.mark.parametrize("basin_selection", ["first_k", "stress"])
 def test_filter_list_additive_on_automatic(test_context, basin_selection):
     """filter_list augments automatic selection rather than replacing it."""
+    from message_ix_models.model.water.config import Config
     from message_ix_models.util import package_data_path
 
     df_basins = pd.read_csv(
         package_data_path("water", "delineation", "basins_by_region_simpl_R12.csv")
     )
 
-    test_context.reduced_basin = True
-    test_context.basin_selection = basin_selection
+    cfg = Config.from_context(test_context)
+    cfg.reduced_basin = True
+    cfg.basin_selection = basin_selection
     test_context.regions = "R12"
     test_context.ssp = "SSP2"
 
@@ -145,7 +147,7 @@ def test_filter_list_additive_on_automatic(test_context, basin_selection):
     extra_basins = list(all_basins - auto_basins)[:3]
     assert len(extra_basins) > 0, "Need basins outside automatic set"
 
-    test_context.filter_list = extra_basins
+    cfg.filter_list = extra_basins
     combined = filter_basins_by_region(df_basins, test_context, n_per_region=1)
     combined_basins = set(combined["BCU_name"])
 
diff --git a/message_ix_models/tests/util/test_click.py b/message_ix_models/tests/util/test_click.py
index c638217b73..372951abd0 100644
--- a/message_ix_models/tests/util/test_click.py
+++ b/message_ix_models/tests/util/test_click.py
@@ -36,10 +36,10 @@ def func(ctx, rep_out_path):
     assert result.output.startswith(f"{expected}\n")
 
 
-def test_regions(mix_models_cli):
-    """--regions=… used on both group and a command within the group.
+def test_regions(mix_models_cli) -> None:
+    """--regions=… redundantly declared on a group and its subcommand.
 
-    If the option is not provided to the inner command, the value given to the outer
+    If the option is not provided to the subcommand, the value given to the parent
     group should persist.
     """
 
@@ -64,6 +64,26 @@ def inner(context, regions):
     assert "ZMB" == result.output.strip()
 
 
+def test_scenario_param_preserves_outer_value(mix_models_cli) -> None:
+    """An unset subcommand --ssp does not overwrite the parent command's value."""
+
+    @click.group()
+    @scenario_param("--ssp", default="SSP2")
+    def outer():
+        pass
+
+    @outer.command()
+    @scenario_param("--ssp")
+    @click.pass_obj
+    def inner(context):
+        print(context.ssp)
+
+    with temporary_command(cli_test_group, outer):
+        result = mix_models_cli.assert_exit_0(["_test", "outer", "--ssp=SSP1", "inner"])
+
+    assert "SSP1" == result.output.strip()
+
+
 @pytest.mark.parametrize(
     "args, command, expected",
     [
diff --git a/message_ix_models/util/click.py b/message_ix_models/util/click.py
index a1bee15695..41a1a88fac 100644
--- a/message_ix_models/util/click.py
+++ b/message_ix_models/util/click.py
@@ -211,11 +211,17 @@ def store_context(context: click.Context | Context, param, value):
     Use this for parameters that are not used directly in a @click.command() function,
     but need to be carried by the Context for later use.
     """
-    setattr(
-        context.obj if isinstance(context, click.Context) else context,
-        param.name,
-        value,
-    )
+    target = context.obj if isinstance(context, click.Context) else context
+
+    # In nested commands, an omitted inner option should not clear a value already set
+    # by an outer command.
+    if value is None and (
+        (isinstance(target, Context) and param.name in target)
+        or (not isinstance(target, Context) and hasattr(target, param.name))
+    ):
+        return value
+
+    setattr(target, param.name, value)
     return value