roms-tools 1.6.0__py3-none-any.whl → 1.6.2__py3-none-any.whl

roms_tools/__init__.py

@@ -1,5 +1,5 @@
 from importlib.metadata import version as _version
-
+import logging  # noqa: F811
 
 try:
     __version__ = _version("roms_tools")
@@ -13,3 +13,6 @@ from roms_tools.setup.tides import TidalForcing  # noqa: F401
 from roms_tools.setup.surface_forcing import SurfaceForcing  # noqa: F401
 from roms_tools.setup.initial_conditions import InitialConditions  # noqa: F401
 from roms_tools.setup.boundary_forcing import BoundaryForcing  # noqa: F401
+
+# Configure logging when the package is imported
+logging.basicConfig(level=logging.INFO, format="%(levelname)s - %(message)s")

roms_tools/_version.py

@@ -1,2 +1,2 @@
 # Do not change! Do not track in version control!
-__version__ = "1.6.0"
+__version__ = "1.6.2"

roms_tools/setup/boundary_forcing.py

@@ -1,9 +1,10 @@
 import xarray as xr
 import numpy as np
 import pandas as pd
+from scipy.ndimage import label
+import logging
 import yaml
 import importlib.metadata
-import warnings
 from typing import Dict, Union, List
 from dataclasses import dataclass, field, asdict
 from roms_tools.setup.grid import Grid
@@ -197,6 +198,13 @@ class BoundaryForcing:
                             ].isel(**bdry_coords[location][direction])
 
                 if not self.apply_2d_horizontal_fill:
+                    self._validate_1d_fill(
+                        processed_fields,
+                        variable_info,
+                        bdry_coords,
+                        direction,
+                        bdry_data.dim_names["depth"],
+                    )
                     processed_fields = apply_1d_horizontal_fill(processed_fields)
 
                 # vertical regridding
@@ -329,38 +337,46 @@ class BoundaryForcing:
                     "is_vector": False,
                     "vector_pair": None,
                     "is_3d": False,
+                    "validate": True,
                 },
-                "temp": default_info,
-                "salt": default_info,
+                "temp": {**default_info, "validate": True},
+                "salt": {**default_info, "validate": False},
                 "u": {
                     "location": "u",
                     "is_vector": True,
                     "vector_pair": "v",
                     "is_3d": True,
+                    "validate": True,
                 },
                 "v": {
                     "location": "v",
                     "is_vector": True,
                     "vector_pair": "u",
                     "is_3d": True,
+                    "validate": True,
                 },
                 "ubar": {
                     "location": "u",
                     "is_vector": True,
                     "vector_pair": "vbar",
                     "is_3d": False,
+                    "validate": False,
                 },
                 "vbar": {
                     "location": "v",
                     "is_vector": True,
                     "vector_pair": "ubar",
                     "is_3d": False,
+                    "validate": False,
                 },
             }
         elif self.type == "bgc":
             variable_info = {}
             for var_name in data.var_names.keys():
-                variable_info[var_name] = default_info
+                if var_name == "ALK":
+                    variable_info[var_name] = {**default_info, "validate": True}
+                else:
+                    variable_info[var_name] = {**default_info, "validate": False}
 
         return variable_info
 
@@ -506,9 +522,78 @@ class BoundaryForcing:
 
         return ds
 
+    def _validate_1d_fill(
+        self, processed_fields, variable_info, bdry_coords, direction, depth_dim
+    ):
+        """Check if any boundary is divided by land and issue a warning if so,
+        suggesting the use of 2D horizontal fill for safer regridding.
+
+        Parameters
+        ----------
+        processed_fields : dict
+            A dictionary where keys are variable names and values are `xarray.DataArray`
+            objects representing the processed data for each variable.
+
+        variable_info : dict
+            A dictionary containing metadata about each variable (e.g., location,
+            whether it's a 3D variable, etc.). Used to retrieve information for
+            validating each variable.
+
+        bdry_coords : dict
+            A dictionary containing boundary coordinates for different directions (north, south,
+            east, west), used to slice the boundary-specific data for each variable.
+
+        direction : str
+            The boundary direction being processed (e.g., "north", "south", "east", or "west").
+
+        depth_dim : str
+            The dimension representing depth (e.g., 'z', 'depth', etc.), used when slicing 3D
+            data for a specific depth level.
+
+        Returns
+        -------
+        None
+            If a boundary is divided by land, a warning is issued. No return value is provided.
+        """
+
+        for var_name in processed_fields.keys():
+            # Only validate variables based on "validate" flag if use_dask is False
+            if not self.use_dask or variable_info[var_name]["validate"]:
+                location = variable_info[var_name]["location"]
+
+                # Select the appropriate mask based on variable location
+                if location == "rho":
+                    mask = self.grid.ds.mask_rho
+                elif location == "u":
+                    mask = self.grid.ds.mask_u
+                elif location == "v":
+                    mask = self.grid.ds.mask_v
+
+                mask = mask.isel(**bdry_coords[location][direction])
+
+                if variable_info[var_name]["is_3d"]:
+                    da = processed_fields[var_name].isel({depth_dim: 0, "time": 0})
+                else:
+                    da = processed_fields[var_name].isel({"time": 0})
+
+                wet_nans = xr.where(da.where(mask).isnull(), 1, 0)
+                # Apply label to find connected components of wet NaNs
+                labeled_array, num_features = label(wet_nans)
+                left_margin = labeled_array[0]
+                right_margin = labeled_array[-1]
+                if left_margin != 0:
+                    num_features = num_features - 1
+                if right_margin != 0:
+                    num_features = num_features - 1
+                if num_features > 0:
+                    logging.warning(
+                        f"For {var_name}, the {direction}ern boundary is divided by land. It would be safer (but slower) to use `apply_2d_horizontal_fill = True`."
+                    )
+
     def _validate(self, ds, variable_info, bdry_coords):
-        """Validate the dataset for NaN values at the first time step based on the fill
-        method used.
+        """Validate the dataset for NaN values at the first time step (bry_time=0) for
+        specified variables. If NaN values are found at wet points, this function raises
+        an error.
 
         Parameters
         ----------
@@ -530,14 +615,11 @@ class BoundaryForcing:
         Notes
         -----
         Validation is performed on the initial boundary time step (`bry_time=0`) for each
-        variable in the dataset. If the `apply_2d_horizontal_fill` attribute is set to False,
-        a warning is issued instead of a strict NaN check, as the data may not be reliably validated.
-        Conversely, if `apply_2d_horizontal_fill` is True, a strict NaN check is performed, raising
-        a ValueError if any NaN values are detected.
+        variable in the dataset.
         """
-        if self.apply_2d_horizontal_fill:
-            # Strict NaN check with ValueError makes sense to be applied
-            for var_name in variable_info:
+        for var_name in variable_info:
+            # only validate variables based on "validate" flag if use_dask is false
+            if not self.use_dask or variable_info[var_name]["validate"]:
                 location = variable_info[var_name]["location"]
 
                 # Select the appropriate mask based on variable location
@@ -547,38 +629,29 @@ class BoundaryForcing:
                     mask = self.grid.ds.mask_u
                 elif location == "v":
                     mask = self.grid.ds.mask_v
-                else:
-                    continue  # Skip if location is not recognized
 
                 for direction in ["south", "east", "north", "west"]:
                     if self.boundaries[direction]:
                         bdry_var_name = f"{var_name}_{direction}"
 
                         # Check for NaN values at the first time step using the nan_check function
+                        if self.apply_2d_horizontal_fill:
+                            error_message = None
+                        else:
+                            error_message = (
+                                f"{bdry_var_name} consists entirely of NaNs after regridding. "
+                                f"This may be due to the {direction}ern boundary being on land in the "
+                                f"{self.source['name']} data, which could have a coarser resolution than the ROMS domain. "
+                                f"Try setting `apply_2d_horizontal_fill = True` to resolve this issue."
+                            )
+
                         nan_check(
                             ds[bdry_var_name].isel(bry_time=0),
                             mask.isel(**bdry_coords[location][direction]),
+                            error_message=error_message,
                         )
-        else:
-            # Can't apply strict NaN check because land values haven't been filled before regridding step; instead warn user
-            for direction in ["south", "east", "north", "west"]:
-                if self.boundaries[direction]:
-                    for var_name in variable_info:
-                        bdry_var_name = f"{var_name}_{direction}"
-                        if ds[bdry_var_name].isel(bry_time=0).isnull().any().values:
-                            warnings.warn(
-                                f"NaN values detected in regridded variables along the {direction}ern boundary. This may indicate that the entire boundary is on land in the source data, or that the source data does not cover this boundary.",
-                                UserWarning,
-                            )
-                            # Break after the first warning for this direction to avoid duplicates
-                            break
 
-    def plot(
-        self,
-        var_name,
-        time=0,
-        layer_contours=False,
-    ) -> None:
+    def plot(self, var_name, time=0, layer_contours=False, ax=None) -> None:
         """Plot the boundary forcing field for a given time-slice.
 
         Parameters
@@ -634,6 +707,8 @@ class BoundaryForcing:
             If True, contour lines representing the boundaries between vertical layers will
             be added to the plot. For clarity, the number of layer
             contours displayed is limited to a maximum of 10. Default is False.
+        ax : matplotlib.axes.Axes, optional
+            The axes to plot on. If None, a new figure is created.
 
         Returns
         -------
@@ -649,7 +724,14 @@ class BoundaryForcing:
         if var_name not in self.ds:
             raise ValueError(f"Variable '{var_name}' is not found in dataset.")
 
-        field = self.ds[var_name].isel(bry_time=time).load()
+        field = self.ds[var_name].isel(bry_time=time)
+
+        if self.use_dask:
+            from dask.diagnostics import ProgressBar
+
+            with ProgressBar():
+                field = field.load()
+
         title = field.long_name
 
         if "s_rho" in field.dims:
@@ -693,30 +775,36 @@ class BoundaryForcing:
                 interface_depth = None
 
             _section_plot(
-                field, interface_depth=interface_depth, title=title, kwargs=kwargs
+                field,
+                interface_depth=interface_depth,
+                title=title,
+                kwargs=kwargs,
+                ax=ax,
             )
         else:
-            _line_plot(field, title=title)
+            _line_plot(field, title=title, ax=ax)
 
     def save(
-        self, filepath: Union[str, Path], np_eta: int = None, np_xi: int = None
+        self,
+        filepath: Union[str, Path],
+        np_eta: int = None,
+        np_xi: int = None,
+        group: bool = False,
     ) -> None:
-        """Save the boundary forcing fields to netCDF4 files.
-
-        This method saves the dataset by grouping it into subsets based on the data frequency. The subsets are then written
-        to one or more netCDF4 files. The filenames of the output files reflect the temporal coverage of the data.
+        """Save the boundary forcing fields to one or more netCDF4 files.
 
-        There are two modes of saving the dataset:
+        This method saves the dataset either as a single file or as multiple files depending on the partitioning and grouping options.
+        The dataset can be saved in two modes:
 
-          1. **Single File Mode (default)**:
+        1. **Single File Mode (default)**:
+            - If both `np_eta` and `np_xi` are `None`, the entire dataset is saved as a single netCDF4 file.
+            - The file is named based on the `filepath`, with `.nc` automatically appended.
 
-            If both `np_eta` and `np_xi` are `None`, the entire dataset, divided by temporal subsets, is saved as a single netCDF4 file
-            with the base filename specified by `filepath.nc`.
+        2. **Partitioned Mode**:
+            - If either `np_eta` or `np_xi` is specified, the dataset is partitioned into spatial tiles along the `eta` and `xi` axes.
+            - Each tile is saved as a separate netCDF4 file, and filenames are modified with an index (e.g., `"filepath_YYYYMM.0.nc"`, `"filepath_YYYYMM.1.nc"`).
 
-          2. **Partitioned Mode**:
-
-            - If either `np_eta` or `np_xi` is specified, the dataset is divided into spatial tiles along the eta-axis and xi-axis.
-            - Each spatial tile is saved as a separate netCDF4 file.
+        Additionally, if `group` is set to `True`, the dataset is first grouped into temporal subsets, resulting in multiple grouped files before partitioning and saving.
 
         Parameters
         ----------
@@ -728,6 +816,8 @@ class BoundaryForcing:
             The number of partitions along the `eta` direction. If `None`, no spatial partitioning is performed.
         np_xi : int, optional
             The number of partitions along the `xi` direction. If `None`, no spatial partitioning is performed.
+        group: bool, optional
+            If `True`, groups the dataset into multiple files based on temporal data frequency. Defaults to `False`.
 
         Returns
         -------
@@ -742,7 +832,18 @@ class BoundaryForcing:
         if filepath.suffix == ".nc":
             filepath = filepath.with_suffix("")
 
-        dataset_list, output_filenames = group_dataset(self.ds.load(), str(filepath))
+        if self.use_dask:
+            from dask.diagnostics import ProgressBar
+
+            with ProgressBar():
+                self.ds.load()
+
+        if group:
+            dataset_list, output_filenames = group_dataset(self.ds, str(filepath))
+        else:
+            dataset_list = [self.ds]
+            output_filenames = [str(filepath)]
+
         saved_filenames = save_datasets(
             dataset_list, output_filenames, np_eta=np_eta, np_xi=np_xi
         )
@@ -783,6 +884,7 @@ class BoundaryForcing:
                 "boundaries": self.boundaries,
                 "source": self.source,
                 "type": self.type,
+                "apply_2d_horizontal_fill": self.apply_2d_horizontal_fill,
                 "model_reference_date": self.model_reference_date.isoformat(),
             }
         }
@@ -796,7 +898,7 @@ class BoundaryForcing:
             # Write header
             file.write(header)
             # Write YAML data
-            yaml.dump(yaml_data, file, default_flow_style=False)
+            yaml.dump(yaml_data, file, default_flow_style=False, sort_keys=False)
 
     @classmethod
     def from_yaml(
@@ -930,6 +1032,7 @@ def apply_1d_horizontal_fill(processed_fields: dict) -> dict:
             raise ValueError(
                 f"No valid horizontal dimension found for variable '{var_name}'."
             )
+
         # Forward and backward fill in the horizontal direction
         filled = one_dim_fill(
             processed_fields[var_name], selected_horizontal_dim, direction="forward"

roms_tools/setup/datasets.py

@@ -6,7 +6,7 @@ from datetime import datetime, timedelta
 import numpy as np
 from typing import Dict, Optional, Union, List
 from pathlib import Path
-import warnings
+import logging
 from roms_tools.setup.utils import (
     assign_dates_to_climatology,
     interpolate_from_climatology,
@@ -410,7 +410,7 @@ class Dataset:
                             ds[time_dim].where(before_start, drop=True).max()
                         )
                     else:
-                        warnings.warn("No records found at or before the start_time.")
+                        logging.warning("No records found at or before the start_time.")
                         closest_before_start = ds[time_dim].min()
 
                     # Identify records after or at end_time
@@ -420,7 +420,7 @@ class Dataset:
                             ds[time_dim].where(after_end, drop=True).min()
                         )
                     else:
-                        warnings.warn("No records found at or after the end_time.")
+                        logging.warning("No records found at or after the end_time.")
                         closest_after_end = ds[time_dim].max()
 
                     # Select records within the time range and add the closest before/after
@@ -451,11 +451,11 @@ class Dataset:
                     if ds.sizes[time_dim] > 1:
                         # Pick the time closest to self.start_time
                         ds = ds.isel({time_dim: 0})
-                    print(
+                    logging.info(
                         f"Selected time entry closest to the specified start_time ({self.start_time}) within the range [{self.start_time}, {self.start_time + timedelta(hours=24)}]: {ds[time_dim].values}"
                     )
         else:
-            warnings.warn(
+            logging.warning(
                 "Dataset does not contain any time information. Please check if the time dimension "
                 "is correctly named or if the dataset includes time data."
             )

roms_tools/setup/grid.py

@@ -13,7 +13,7 @@ from roms_tools.setup.plot import _plot, _section_plot, _profile_plot, _line_plo
 from roms_tools.setup.utils import interpolate_from_rho_to_u, interpolate_from_rho_to_v
 from roms_tools.setup.vertical_coordinate import sigma_stretch, compute_depth
 from roms_tools.setup.utils import extract_single_value, save_datasets
-import warnings
+import logging
 from pathlib import Path
 
 RADIUS_OF_EARTH = 6371315.0  # in m
@@ -350,11 +350,7 @@ class Grid:
             _plot(self.ds, straddle=self.straddle)
 
     def plot_vertical_coordinate(
-        self,
-        varname="layer_depth_rho",
-        s=None,
-        eta=None,
-        xi=None,
+        self, varname="layer_depth_rho", s=None, eta=None, xi=None, ax=None
     ) -> None:
         """Plot the vertical coordinate system for a given eta-, xi-, or s-slice.
 
@@ -376,6 +372,8 @@ class Grid:
             The eta-index to plot. Default is None.
         xi : int, optional
             The xi-index to plot. Default is None.
+        ax : matplotlib.axes.Axes, optional
+            The axes to plot on. If None, a new figure is created. Note that this argument does not work for horizontal plots that display the eta- and xi-dimensions at the same time.
 
         Returns
         -------
@@ -477,12 +475,13 @@ class Grid:
                     interface_depth=interface_depth,
                     title=title,
                     kwargs=kwargs,
+                    ax=ax,
                 )
             else:
                 if "s_rho" in field.dims or "s_w" in field.dims:
-                    _profile_plot(field, title=title)
+                    _profile_plot(field, title=title, ax=ax)
                 else:
-                    _line_plot(field, title=title)
+                    _line_plot(field, title=title, ax=ax)
 
     def save(
         self, filepath: Union[str, Path], np_eta: int = None, np_xi: int = None
@@ -679,7 +678,7 @@ class Grid:
             # Write header
             file.write(header)
             # Write YAML data
-            yaml.dump(yaml_data, file, default_flow_style=False)
+            yaml.dump(yaml_data, file, default_flow_style=False, sort_keys=False)
 
     @classmethod
     def from_yaml(cls, filepath: Union[str, Path]) -> "Grid":
@@ -728,9 +727,8 @@ class Grid:
                 roms_tools_version_current = "unknown"
 
             if roms_tools_version_header != roms_tools_version_current:
-                warnings.warn(
+                logging.warning(
                     f"Current roms-tools version ({roms_tools_version_current}) does not match the version in the YAML header ({roms_tools_version_header}).",
-                    UserWarning,
                 )
 
         if grid_data is None: