Skip to content

Bank Erosion Input Data Models#

The Bank Erosion Input Data Models module provides data structures for representing input data in the D-FAST Bank Erosion software.

Overview#

The Bank Erosion Input Data Models module contains classes that represent various aspects of bank erosion input data, such as river data and simulation data. These data models are used by the Bank Erosion module to process and analyze bank erosion.

Components#

The Bank Erosion Input Data Models module consists of the following components:

Data Models#

dfastbe.bank_erosion.data_models.inputs #

BankLinesResultsError #

Bases: Exception

Custom exception for BankLine results errors.

Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
268
269
270
271
class BankLinesResultsError(Exception):
    """Custom exception for BankLine results errors."""

    pass

ErosionRiverData #

Bases: BaseRiverData

Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
class ErosionRiverData(BaseRiverData):

    def __init__(self, config_file: ConfigFile):
        super().__init__(config_file)
        self.bank_dir = self._get_bank_line_dir()
        self.output_dir = config_file.get_output_dir("erosion")
        self.debug = config_file.debug
        # set plotting flags
        self.plot_flags = config_file.get_plotting_flags(config_file.root_dir)
        # get filter settings for bank levels and flow velocities along banks
        self.zb_dx = config_file.get_float("Erosion", "BedFilterDist", 0.0, positive=True)
        self.vel_dx = config_file.get_float("Erosion", "VelFilterDist", 0.0, positive=True)
        log_text("get_levels")
        self.num_discharge_levels = config_file.get_int("Erosion", "NLevel")
        self.output_intervals = config_file.get_float("Erosion", "OutputInterval", 1.0)
        self.bank_lines = config_file.read_bank_lines(str(self.bank_dir))
        self.river_axis = self._read_river_axis()
        self.erosion_time = self.config_file.get_int("Erosion", "TErosion", positive=True)

    def simulation_data(self) -> ErosionSimulationData:
        """Simulation Data."""
        ref_level = self.config_file.get_int("Erosion", "RefLevel") - 1
        # read simulation data (get_sim_data)
        sim_file = self.config_file.get_sim_file("Erosion", str(ref_level + 1))
        log_text("-")
        log_text("read_simdata", data={"file": sim_file})
        log_text("-")
        simulation_data = ErosionSimulationData.read(sim_file)

        return simulation_data

    def _get_bank_output_dir(self) -> Path:
        bank_output_dir = self.config_file.get_str("General", "BankDir")
        log_text("bank_dir_out", data={"dir": bank_output_dir})
        if os.path.exists(bank_output_dir):
            log_text("overwrite_dir", data={"dir": bank_output_dir})
        else:
            os.makedirs(bank_output_dir)

        return Path(bank_output_dir)

    def _get_bank_line_dir(self) -> Path:
        bank_dir = self.config_file.get_str("General", "BankDir")
        log_text("bank_dir_in", data={"dir": bank_dir})
        bank_dir = Path(bank_dir)
        if not bank_dir.exists():
            log_text("missing_dir", data={"dir": bank_dir})
            raise BankLinesResultsError(
                f"Required bank line directory:{bank_dir} does not exist. please use the banklines command to run the "
                "bankline detection tool first it."
            )
        else:
            return bank_dir

    def _read_river_axis(self) -> LineString:
        """Get the river axis from the analysis settings."""
        river_axis_file = self.config_file.get_str("Erosion", "RiverAxis")
        log_text("read_river_axis", data={"file": river_axis_file})
        river_axis = XYCModel.read(river_axis_file)
        return river_axis

simulation_data() -> ErosionSimulationData #

Simulation Data.

Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
225
226
227
228
229
230
231
232
233
234
235
def simulation_data(self) -> ErosionSimulationData:
    """Simulation Data."""
    ref_level = self.config_file.get_int("Erosion", "RefLevel") - 1
    # read simulation data (get_sim_data)
    sim_file = self.config_file.get_sim_file("Erosion", str(ref_level + 1))
    log_text("-")
    log_text("read_simdata", data={"file": sim_file})
    log_text("-")
    simulation_data = ErosionSimulationData.read(sim_file)

    return simulation_data

ErosionSimulationData #

Bases: BaseSimulationData

Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
class ErosionSimulationData(BaseSimulationData):

    def compute_mesh_topology(self) -> MeshData:
        """Derive secondary topology arrays from the face-node connectivity of the mesh.

        This function computes the edge-node, edge-face, and face-edge connectivity arrays,
        as well as the boundary edges of the mesh, based on the face-node connectivity provided
        in the simulation data.

        Returns:
            MeshData: a dataclass containing the following attributes:
                - `x_face_coords`: x-coordinates of face nodes
                - `y_face_coords`: y-coordinates of face nodes
                - `x_edge_coords`: x-coordinates of edge nodes
                - `y_edge_coords`: y-coordinates of edge nodes
                - `face_node`: the node indices for each of the mesh faces.
                - `n_nodes`: number of nodes per face
                - `edge_node`: the node indices for each of the mesh edges.
                - `edge_face_connectivity`: the face indices for each of the mesh edge
                - `face_edge_connectivity`: the edge indices for each of the mesh face
                - `boundary_edge_nrs`: indices of boundary edges

        Raises:
            KeyError:
                If required keys (e.g., `face_node`, `nnodes`, `x_node`, `y_node`) are missing from the `sim` object.

        Notes:
            - The function identifies unique edges by sorting and comparing node indices.
            - Boundary edges are identified as edges that belong to only one face.
            - The function assumes that the mesh is well-formed, with consistent face-node connectivity.
        """

        # get a sorted list of edge node connections (shared edges occur twice)
        # face_nr contains the face index to which the edge belongs
        n_faces = self.face_node.shape[0]
        n_edges = sum(self.n_nodes)
        edge_node = np.zeros((n_edges, 2), dtype=int)
        face_nr = np.zeros((n_edges,), dtype=int)
        i = 0
        for face_i in range(n_faces):
            num_edges = self.n_nodes[face_i]  # note: nEdges = nNodes
            for edge_i in range(num_edges):
                if edge_i == 0:
                    edge_node[i, 1] = self.face_node[face_i, num_edges - 1]
                else:
                    edge_node[i, 1] = self.face_node[face_i, edge_i - 1]
                edge_node[i, 0] = self.face_node[face_i, edge_i]
                face_nr[i] = face_i
                i = i + 1
        edge_node.sort(axis=1)
        i2 = np.argsort(edge_node[:, 1], kind="stable")
        i1 = np.argsort(edge_node[i2, 0], kind="stable")
        i12 = i2[i1]
        edge_node = edge_node[i12, :]
        face_nr = face_nr[i12]

        # detect which edges are equal to the previous edge, and get a list of all unique edges
        numpy_true = np.array([True])
        equal_to_previous = np.concatenate(
            (~numpy_true, (np.diff(edge_node, axis=0) == 0).all(axis=1))
        )
        unique_edge = ~equal_to_previous
        n_unique_edges = np.sum(unique_edge)
        # reduce the edge node connections to only the unique edges
        edge_node = edge_node[unique_edge, :]

        # number the edges
        edge_nr = np.zeros(n_edges, dtype=int)
        edge_nr[unique_edge] = np.arange(n_unique_edges, dtype=int)
        edge_nr[equal_to_previous] = edge_nr[
            np.concatenate((equal_to_previous[1:], equal_to_previous[:1]))
        ]

        # if two consecutive edges are unique, the first one occurs only once and represents a boundary edge
        is_boundary_edge = unique_edge & np.concatenate((unique_edge[1:], numpy_true))
        boundary_edge_nrs = edge_nr[is_boundary_edge]

        # go back to the original face order
        edge_nr_in_face_order = np.zeros(n_edges, dtype=int)
        edge_nr_in_face_order[i12] = edge_nr
        # create the face edge connectivity array
        face_edge_connectivity = np.zeros(self.face_node.shape, dtype=int)

        i = 0
        for face_i in range(n_faces):
            num_edges = self.n_nodes[face_i]  # note: num_edges = n_nodes
            for edge_i in range(num_edges):
                face_edge_connectivity[face_i, edge_i] = edge_nr_in_face_order[i]
                i = i + 1

        # determine the edge face connectivity
        edge_face = -np.ones((n_unique_edges, 2), dtype=int)
        edge_face[edge_nr[unique_edge], 0] = face_nr[unique_edge]
        edge_face[edge_nr[equal_to_previous], 1] = face_nr[equal_to_previous]

        x_face_coords = self.apply_masked_indexing(
            self.x_node, self.face_node
        )
        y_face_coords = self.apply_masked_indexing(
            self.y_node, self.face_node
        )
        x_edge_coords = self.x_node[edge_node]
        y_edge_coords = self.y_node[edge_node]

        return MeshData(
            x_face_coords=x_face_coords,
            y_face_coords=y_face_coords,
            x_edge_coords=x_edge_coords,
            y_edge_coords=y_edge_coords,
            face_node=self.face_node,
            n_nodes=self.n_nodes,
            edge_node=edge_node,
            edge_face_connectivity=edge_face,
            face_edge_connectivity=face_edge_connectivity,
            boundary_edge_nrs=boundary_edge_nrs,
        )

    @staticmethod
    def apply_masked_indexing(
        x0: np.array, idx: np.ma.masked_array
    ) -> np.ma.masked_array:
        """
        Index one array by another transferring the mask.

        Args:
            x0 : np.ndarray
                A linear array.
            idx : np.ma.masked_array
                An index array with possibly masked indices.

        returns:
            x1: np.ma.masked_array
                An array with same shape as idx, with mask.
        """
        idx_safe = idx.copy()
        idx_safe.data[np.ma.getmask(idx)] = 0
        x1 = np.ma.masked_where(np.ma.getmask(idx), x0[idx_safe])
        return x1

    def calculate_bank_velocity(self, single_bank: "SingleBank", vel_dx) -> np.ndarray:
        from dfastbe.bank_erosion.utils import moving_avg
        bank_face_indices = single_bank.bank_face_indices
        vel_bank = (
                np.abs(
                    self.velocity_x_face[bank_face_indices] * single_bank.dx
                    + self.velocity_y_face[bank_face_indices] * single_bank.dy
                )
                / single_bank.segment_length
        )

        if vel_dx > 0.0:
            vel_bank = moving_avg(
                single_bank.bank_chainage_midpoints, vel_bank, vel_dx
            )

        return vel_bank

    def calculate_bank_height(self, single_bank: SingleBank, zb_dx):
        # bank height = maximum bed elevation per cell
        from dfastbe.bank_erosion.utils import moving_avg
        bank_index = single_bank.bank_face_indices
        if self.bed_elevation_location == "node":
            zb_nodes = self.bed_elevation_values
            zb_all = self.apply_masked_indexing(
                zb_nodes, self.face_node[bank_index, :]
            )
            zb_bank = zb_all.max(axis=1)
            if zb_dx > 0.0:
                zb_bank = moving_avg(
                    single_bank.bank_chainage_midpoints, zb_bank, zb_dx,
                )
        else:
            # don't know ... need to check neighbouring cells ...
            zb_bank = None

        return zb_bank

    def get_fairway_data(self, fairway_face_indices):
        # get fairway face indices
        fairway_face_indices = fairway_face_indices

        # get water depth along the fair-way
        water_depth_fairway = self.water_depth_face[fairway_face_indices]
        water_level = self.water_level_face[fairway_face_indices]
        chez_face = self.chezy_face[fairway_face_indices]
        chezy = 0 * chez_face + chez_face.mean()

        data = {
            "water_depth": water_depth_fairway,
            "water_level": water_level,
            "chezy": chezy,
        }
        return data

apply_masked_indexing(x0: np.array, idx: np.ma.masked_array) -> np.ma.masked_array staticmethod #

Index one array by another transferring the mask.

Parameters:

Name Type Description Default
x0

np.ndarray A linear array.

 required
idx

np.ma.masked_array An index array with possibly masked indices.

 required

Returns:

Name Type Description
x1 masked_array

np.ma.masked_array An array with same shape as idx, with mask.
Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
@staticmethod
def apply_masked_indexing(
    x0: np.array, idx: np.ma.masked_array
) -> np.ma.masked_array:
    """
    Index one array by another transferring the mask.

    Args:
        x0 : np.ndarray
            A linear array.
        idx : np.ma.masked_array
            An index array with possibly masked indices.

    returns:
        x1: np.ma.masked_array
            An array with same shape as idx, with mask.
    """
    idx_safe = idx.copy()
    idx_safe.data[np.ma.getmask(idx)] = 0
    x1 = np.ma.masked_where(np.ma.getmask(idx), x0[idx_safe])
    return x1

compute_mesh_topology() -> MeshData #

Derive secondary topology arrays from the face-node connectivity of the mesh.

This function computes the edge-node, edge-face, and face-edge connectivity arrays, as well as the boundary edges of the mesh, based on the face-node connectivity provided in the simulation data.

Returns:

Name Type Description
MeshData MeshData

a dataclass containing the following attributes: - x_face_coords: x-coordinates of face nodes - y_face_coords: y-coordinates of face nodes - x_edge_coords: x-coordinates of edge nodes - y_edge_coords: y-coordinates of edge nodes - face_node: the node indices for each of the mesh faces. - n_nodes: number of nodes per face - edge_node: the node indices for each of the mesh edges. - edge_face_connectivity: the face indices for each of the mesh edge - face_edge_connectivity: the edge indices for each of the mesh face - boundary_edge_nrs: indices of boundary edges

Raises:

Type Description
KeyError

If required keys (e.g., face_node, nnodes, x_node, y_node) are missing from the sim object.
Notes
  • The function identifies unique edges by sorting and comparing node indices.
  • Boundary edges are identified as edges that belong to only one face.
  • The function assumes that the mesh is well-formed, with consistent face-node connectivity.
Source code in src/dfastbe/bank_erosion/data_models/inputs.py 
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
def compute_mesh_topology(self) -> MeshData:
    """Derive secondary topology arrays from the face-node connectivity of the mesh.

    This function computes the edge-node, edge-face, and face-edge connectivity arrays,
    as well as the boundary edges of the mesh, based on the face-node connectivity provided
    in the simulation data.

    Returns:
        MeshData: a dataclass containing the following attributes:
            - `x_face_coords`: x-coordinates of face nodes
            - `y_face_coords`: y-coordinates of face nodes
            - `x_edge_coords`: x-coordinates of edge nodes
            - `y_edge_coords`: y-coordinates of edge nodes
            - `face_node`: the node indices for each of the mesh faces.
            - `n_nodes`: number of nodes per face
            - `edge_node`: the node indices for each of the mesh edges.
            - `edge_face_connectivity`: the face indices for each of the mesh edge
            - `face_edge_connectivity`: the edge indices for each of the mesh face
            - `boundary_edge_nrs`: indices of boundary edges

    Raises:
        KeyError:
            If required keys (e.g., `face_node`, `nnodes`, `x_node`, `y_node`) are missing from the `sim` object.

    Notes:
        - The function identifies unique edges by sorting and comparing node indices.
        - Boundary edges are identified as edges that belong to only one face.
        - The function assumes that the mesh is well-formed, with consistent face-node connectivity.
    """

    # get a sorted list of edge node connections (shared edges occur twice)
    # face_nr contains the face index to which the edge belongs
    n_faces = self.face_node.shape[0]
    n_edges = sum(self.n_nodes)
    edge_node = np.zeros((n_edges, 2), dtype=int)
    face_nr = np.zeros((n_edges,), dtype=int)
    i = 0
    for face_i in range(n_faces):
        num_edges = self.n_nodes[face_i]  # note: nEdges = nNodes
        for edge_i in range(num_edges):
            if edge_i == 0:
                edge_node[i, 1] = self.face_node[face_i, num_edges - 1]
            else:
                edge_node[i, 1] = self.face_node[face_i, edge_i - 1]
            edge_node[i, 0] = self.face_node[face_i, edge_i]
            face_nr[i] = face_i
            i = i + 1
    edge_node.sort(axis=1)
    i2 = np.argsort(edge_node[:, 1], kind="stable")
    i1 = np.argsort(edge_node[i2, 0], kind="stable")
    i12 = i2[i1]
    edge_node = edge_node[i12, :]
    face_nr = face_nr[i12]

    # detect which edges are equal to the previous edge, and get a list of all unique edges
    numpy_true = np.array([True])
    equal_to_previous = np.concatenate(
        (~numpy_true, (np.diff(edge_node, axis=0) == 0).all(axis=1))
    )
    unique_edge = ~equal_to_previous
    n_unique_edges = np.sum(unique_edge)
    # reduce the edge node connections to only the unique edges
    edge_node = edge_node[unique_edge, :]

    # number the edges
    edge_nr = np.zeros(n_edges, dtype=int)
    edge_nr[unique_edge] = np.arange(n_unique_edges, dtype=int)
    edge_nr[equal_to_previous] = edge_nr[
        np.concatenate((equal_to_previous[1:], equal_to_previous[:1]))
    ]

    # if two consecutive edges are unique, the first one occurs only once and represents a boundary edge
    is_boundary_edge = unique_edge & np.concatenate((unique_edge[1:], numpy_true))
    boundary_edge_nrs = edge_nr[is_boundary_edge]

    # go back to the original face order
    edge_nr_in_face_order = np.zeros(n_edges, dtype=int)
    edge_nr_in_face_order[i12] = edge_nr
    # create the face edge connectivity array
    face_edge_connectivity = np.zeros(self.face_node.shape, dtype=int)

    i = 0
    for face_i in range(n_faces):
        num_edges = self.n_nodes[face_i]  # note: num_edges = n_nodes
        for edge_i in range(num_edges):
            face_edge_connectivity[face_i, edge_i] = edge_nr_in_face_order[i]
            i = i + 1

    # determine the edge face connectivity
    edge_face = -np.ones((n_unique_edges, 2), dtype=int)
    edge_face[edge_nr[unique_edge], 0] = face_nr[unique_edge]
    edge_face[edge_nr[equal_to_previous], 1] = face_nr[equal_to_previous]

    x_face_coords = self.apply_masked_indexing(
        self.x_node, self.face_node
    )
    y_face_coords = self.apply_masked_indexing(
        self.y_node, self.face_node
    )
    x_edge_coords = self.x_node[edge_node]
    y_edge_coords = self.y_node[edge_node]

    return MeshData(
        x_face_coords=x_face_coords,
        y_face_coords=y_face_coords,
        x_edge_coords=x_edge_coords,
        y_edge_coords=y_edge_coords,
        face_node=self.face_node,
        n_nodes=self.n_nodes,
        edge_node=edge_node,
        edge_face_connectivity=edge_face,
        face_edge_connectivity=face_edge_connectivity,
        boundary_edge_nrs=boundary_edge_nrs,
    )

The data models component provides classes for representing various types of input data related to bank erosion, such as:

  • ErosionSimulationData: Represents simulation data for erosion calculations, including mesh topology, bank velocity, and bank height
  • ErosionRiverData: Represents river data for erosion calculations, including river center line, bank lines, and simulation data
  • BankLinesResultsError: An exception class for bank lines results errors

Usage Example#

from dfastbe.bank_erosion.data_models.inputs import ErosionRiverData, ErosionSimulationData
from dfastbe.io.config import ConfigFile

# Load configuration file
config_file = ConfigFile.read("config.cfg")

# Create river data object
river_data = ErosionRiverData(config_file)

# Access river data properties
print(f"River center line: {river_data.river_center_line}")
print(f"Bank lines: {river_data.bank_lines}")

# Get simulation data
simulation_data = river_data.simulation_data()

# Compute mesh topology
mesh_data = simulation_data.compute_mesh_topology()

For more details on the specific classes and their properties, refer to the API reference below.