Use a dynamically determined grid scale for a region based on size

tests/test_plots.py

@@ -90,12 +90,12 @@ def assert_images_match(
     Compares two images using diffimg and asserts the difference is within tolerance.
     Provides informative assertion messages.
     """
-    assert (
-        generated_path.exists()
-    ), f"Generated image file does not exist: {generated_path}"
-    assert (
-        expected_path.exists()
-    ), f"Expected image file does not exist: {expected_path}"
+    assert generated_path.exists(), (
+        f"Generated image file does not exist: {generated_path}"
+    )
+    assert expected_path.exists(), (
+        f"Expected image file does not exist: {expected_path}"
+    )
 
     diff_ratio = diffimg.diff(expected_path, generated_path)
 

visualisation/sources/plot_rise.py

@@ -8,6 +8,7 @@
 from pygmt_helper import plotting
 from qcore import cli
 from source_modelling import srf
+from visualisation import utils
 
 app = typer.Typer()
 
@@ -53,12 +54,13 @@ def plot_rise(
     fig = plotting.gen_region_fig(
         title, projection=f"M{width}c", region=region, map_data=None
     )
-
+    dx = srf_data.header.iloc[0]["len"] / srf_data.header.iloc[0]["nstk"]
+    grid_scale = min(utils.grid_scale_for_region(region), dx * 1000)
     for i, segment_points in enumerate(srf_data.segments):
         cur_grid = plotting.create_grid(
             segment_points,
             "trise",
-            grid_spacing="5e/5e",
+            grid_spacing=f"{grid_scale}e/{grid_scale}e",
             region=(
                 segment_points["lon"].min(),
                 segment_points["lon"].max(),
@@ -82,7 +84,7 @@ def plot_rise(
         time_grid = plotting.create_grid(
             segment_points,
             "tinit",
-            grid_spacing="5e/5e",
+            grid_spacing=f"{grid_scale}e/{grid_scale}e",
             region=(
                 segment_points["lon"].min(),
                 segment_points["lon"].max(),

visualisation/sources/plot_srf.py

@@ -119,6 +119,7 @@ def show_slip(
     )
     dx = srf_data.header.iloc[0]["len"] / srf_data.header.iloc[0]["nstk"]
     subtitle = f"Slip: {slip_stats}, dx = {dx:.2f} km, {len(srf_data.header)} planes"
+    grid_scale = min(utils.grid_scale_for_region(region), dx * 1000)
     for (_, segment), segment_points in zip(
         srf_data.header.iterrows(), srf_data.segments
     ):
@@ -129,7 +130,7 @@ def show_slip(
         cur_grid = plotting.create_grid(
             segment_points,
             "slip",
-            grid_spacing="5e/5e",
+            grid_spacing=f"{grid_scale}e/{grid_scale}e",
             region=(
                 segment_points["lon"].min(),
                 segment_points["lon"].max(),
@@ -156,7 +157,7 @@ def show_slip(
         time_grid = plotting.create_grid(
             segment_points,
             "tinit",
-            grid_spacing="5e/5e",
+            grid_spacing=f"{grid_scale}e/{grid_scale}e",
             region=(
                 segment_points["lon"].min(),
                 segment_points["lon"].max(),

visualisation/sources/plot_srf_distribution.py

@@ -1,5 +1,6 @@
 #!/usr/bin/env python3
 """Plot SRF distributions."""
+
 from pathlib import Path
 from typing import Annotated, Optional
 
@@ -46,7 +47,7 @@ def plot_srf_distribution(
     ax.set_xlabel("Slip (cm)")
     ax.set_title(
         title
-        or f'Slip PDF for {srf_ffp.stem} ({utils.format_description(srf_data.points["slip"], compact=True)})'
+        or f"Slip PDF for {srf_ffp.stem} ({utils.format_description(srf_data.points['slip'], compact=True)})"
     )
 
     plt.savefig(plot_png, dpi=dpi)

visualisation/sources/plot_stoch.py

@@ -82,7 +82,7 @@ def plot_stoch(
             ax.text(
                 k * dx + dx / 2,
                 (j * dy + dy / 2),
-                f"{int(slip[j,  k])}",
+                f"{int(slip[j, k])}",
                 ha="center",
                 va="center",
                 color=colour,

visualisation/utils.py

@@ -292,3 +292,27 @@ def bounding_region_for(
         min_latitude - latitude_pad,
         max_latitude + latitude_pad,
     )
+
+
+def grid_scale_for_region(region: tuple[float, float, float, float]) -> int:
+    """Compute a suitable grid scale for a pygmt region.
+
+    Parameters
+    ----------
+    region : tuple[float, float, float, float]
+        The pygmt region you will plot a grid in.
+
+    Returns
+    -------
+    int
+        A value (in metres) represent for `plotting.create_grid` to
+        use when plotting the lat-lon grid. Scale is based on the
+        maximum extent in the lat or lon direction for the figure in
+        kilometres. Works out that 10km = 25m, 100km = 250m, with a
+        minimum resolution of 5m.
+    """
+    min_lon, max_lon, min_lat, max_lat = region
+    lat_km = (max_lat - min_lat) * 111
+    lon_km = (max_lon - min_lon) * 111 * np.cos(np.radians((min_lat + max_lat) / 2))
+    maximum_extent = max(lat_km, lon_km)
+    return int(round(max(5, 2.5 * maximum_extent)))