feature: Introducing configurable option to simplify GADM shapes

config.default.yaml

@@ -106,6 +106,7 @@ cluster_options:
 
 build_shape_options:
   gadm_layer_id: 1 # GADM level area used for the gadm_shapes. Codes are country-dependent but roughly: 0: country, 1: region/county-like, 2: municipality-like
+  simplify_gadm: false # When true, shape polygons are simplified else no
   update_file: false # When true, all the input files are downloaded again and replace the existing files
   out_logging: true # When true, logging is printed to console
   year: 2020 # reference year used to derive shapes, info on population and info on GDP

scripts/build_shapes.py

@@ -362,6 +362,7 @@ def eez(
     distance=0.01,
     minarea=0.01,
     tolerance=0.01,
+    simplify_gadm=True,
 ):
     """
     Creates offshore shapes by buffer smooth countryshape (=offset country
@@ -386,22 +387,24 @@ def eez(
         }
     ).set_index("name")
 
-    ret_df = ret_df.geometry.map(
-        lambda x: _simplify_polys(x, minarea=minarea, tolerance=tolerance)
-    )
+    if simplify_gadm:
+        ret_df = ret_df.geometry.map(
+            lambda x: _simplify_polys(x, minarea=minarea, tolerance=tolerance)
+        )
 
-    ret_df = ret_df.apply(lambda x: make_valid(x))
+        ret_df = ret_df.apply(lambda x: make_valid(x))
 
     country_shapes_with_buffer = country_shapes.buffer(distance)
     ret_df_new = ret_df.difference(country_shapes_with_buffer)
 
-    # repeat to simplify after the buffer correction
-    ret_df_new = ret_df_new.map(
-        lambda x: (
-            x if x is None else _simplify_polys(x, minarea=minarea, tolerance=tolerance)
+    if simplify_gadm:
+        # repeat to simplify after the buffer correction
+        ret_df_new = ret_df_new.map(
+            lambda x: (
+                x if x is None else _simplify_polys(x, minarea=minarea, tolerance=tolerance)
+            )
         )
-    )
-    ret_df_new = ret_df_new.apply(lambda x: x if x is None else make_valid(x))
+        ret_df_new = ret_df_new.apply(lambda x: x if x is None else make_valid(x))
 
     # Drops empty geometry
     ret_df = ret_df_new.dropna()
@@ -1256,6 +1259,7 @@ def gadm(
     out_logging=False,
     year=2020,
     nprocesses=None,
+    simplify_gadm=True,
 ):
     if out_logging:
         logger.info("Stage 3 of 5: Creation GADM GeoDataFrame")
@@ -1305,7 +1309,9 @@ def gadm(
         lambda x: x if x.find(".") == 0 else "." + x
     )
     df_gadm.set_index("GADM_ID", inplace=True)
-    df_gadm["geometry"] = df_gadm["geometry"].map(_simplify_polys)
+
+    if simplify_gadm:
+        df_gadm["geometry"] = df_gadm["geometry"].map(_simplify_polys)
     df_gadm.geometry = df_gadm.geometry.apply(
         lambda r: make_valid(r) if not r.is_valid else r
     )
@@ -1338,6 +1344,7 @@ def gadm(
     contended_flag = snakemake.params.build_shape_options["contended_flag"]
     worldpop_method = snakemake.params.build_shape_options["worldpop_method"]
     gdp_method = snakemake.params.build_shape_options["gdp_method"]
+    simplify_gadm = snakemake.params.build_shape_options['simplify_gadm']
 
     country_shapes = countries(
         countries_list,
@@ -1349,7 +1356,7 @@ def gadm(
     country_shapes.to_file(snakemake.output.country_shapes)
 
     offshore_shapes = eez(
-        countries_list, geo_crs, country_shapes, EEZ_gpkg, out_logging
+        countries_list, geo_crs, country_shapes, EEZ_gpkg, out_logging, simplify_gadm
     )
 
     offshore_shapes.reset_index().to_file(snakemake.output.offshore_shapes)
@@ -1371,5 +1378,6 @@ def gadm(
         out_logging,
         year,
         nprocesses=nprocesses,
+        simplify_gadm=simplify_gadm,
     )
     save_to_geojson(gadm_shapes, out.gadm_shapes)