feature: Feature to add a tolerance value 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_tolerance: 0.01 # Default value is 0.01, higher the value more is the simplification of the GADM shapes
   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

doc/configtables/build_shape_options.csv

@@ -1,9 +1,10 @@
-,Unit,Values,Description
-gadm_layer_id,, "{0, 1, 2}", "GADM level area used for the gadm_shapes. Codes are country-dependent but roughly: 0: country, 1: region/county-like, 2: municipality-like."
-update_file, bool, "{True, False}", "True: all input files are downloaded again and replace the existing files."
-out_logging, bool, "{True, False}", "True: Logging is printed in the console."
-year,, "past year; e.g. YYYY", "Reference year used to derive shapes, info on population and info on GDP."
-nprocesses, int,, "Number of processes to be used in build_shapes."
-worldpop_method,, "{""standard"", ""api"", false}", "Specifies how population is added to every shape: ""standard"" pulls from web 1kmx1km raster; ""api"" pulls from API 100mx100m raster; false (not ""false"") no population addition to shape. This is useful when generating only cutout."
-gdp_method,, "{""standard"", false}", "Specifies how GDP is added to every shape: ""standard"" pulls from web 1x1km raster; false (not ""false"") no gdp addition to shape. This is useful when generating only cutout."
-contended_flag,, "{""set_by_country"", ""drop""}", "Specifies what to do with contended countries: ""set_by_country"" assigns the contended areas to the countries according to the GADM database; ""drop"" drops the contended areas from the model."
+,Unit,Values,Description,,,,
+gadm_layer_id,,"""{0",1,"2}""","""GADM level area used for the gadm_shapes. Codes are country-dependent but roughly: 0: country",1: region/county-like,"2: municipality-like."""
+simplify_tolerance,float,>= 0,Specifies tolerance limit when simplifying GADM shapes. A higher value indicates more simplification,,,,
+update_file, bool," ""{True"," False}"""," ""True: all input files are downloaded again and replace the existing files.""",,,
+out_logging, bool," ""{True"," False}"""," ""True: Logging is printed in the console.""",,,
+year,," ""past year; e.g. YYYY"""," ""Reference year used to derive shapes"," info on population and info on GDP.""",,,
+nprocesses, int,," ""Number of processes to be used in build_shapes.""",,,,
+worldpop_method,," ""{""standard"""," ""api"""," false}""","""Specifies how population is added to every shape: ""standard"" pulls from web 1kmx1km raster; ""api"" pulls from API 100mx100m raster; false (not ""false"") no population addition to shape. This is useful when generating only cutout.""",,
+gdp_method,," ""{""standard"""," false}""","""Specifies how GDP is added to every shape: ""standard"" pulls from web 1x1km raster; false (not ""false"") no gdp addition to shape. This is useful when generating only cutout.""",,,
+contended_flag,," ""{""set_by_country"""," ""drop""}"""," ""Specifies what to do with contended countries: ""set_by_country"" assigns the contended areas to the countries according to the GADM database; ""drop"" drops the contended areas from the model.""",,,

scripts/build_shapes.py

@@ -261,7 +261,7 @@ def _simplify_polys(polys, minarea=0.01, tolerance=0.01, filterremote=False):
     return polys.simplify(tolerance=tolerance)
 
 
-def countries(countries, geo_crs, contended_flag, update=False, out_logging=False):
+def countries(countries, geo_crs, contended_flag, update=False, out_logging=False, tolerance=0.01):
     "Create country shapes"
 
     if out_logging:
@@ -282,7 +282,8 @@ def countries(countries, geo_crs, contended_flag, update=False, out_logging=Fals
     df_countries.rename(columns={"GID_0": "name"}, inplace=True)
 
     # set index and simplify polygons
-    ret_df = df_countries.set_index("name")["geometry"].map(_simplify_polys)
+    ret_df = df_countries.set_index("name")["geometry"].map(lambda x: _simplify_polys(x,tolerance=tolerance))
+
     # there may be "holes" in the countries geometry which cause troubles along the workflow
     # e.g. that is the case for enclaves like Dahagram–Angarpota for IN/BD
     ret_df = ret_df.make_valid()
@@ -1256,6 +1257,7 @@ def gadm(
     out_logging=False,
     year=2020,
     nprocesses=None,
+    tolerance=0.01,
 ):
     if out_logging:
         logger.info("Stage 3 of 5: Creation GADM GeoDataFrame")
@@ -1305,7 +1307,7 @@ 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)
+    df_gadm["geometry"] = df_gadm["geometry"].map(lambda x: _simplify_polys(x,tolerance=tolerance))
     df_gadm.geometry = df_gadm.geometry.apply(
         lambda r: make_valid(r) if not r.is_valid else r
     )
@@ -1338,18 +1340,20 @@ 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"]
+    tolerance = snakemake.params.build_shape_options['simplify_tolerance']
 
     country_shapes = countries(
         countries_list,
         geo_crs,
         contended_flag,
         update,
         out_logging,
+        tolerance=tolerance,
     )
     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, tolerance=tolerance
     )
 
     offshore_shapes.reset_index().to_file(snakemake.output.offshore_shapes)
@@ -1371,5 +1375,6 @@ def gadm(
         out_logging,
         year,
         nprocesses=nprocesses,
+        tolerance=tolerance,
     )
     save_to_geojson(gadm_shapes, out.gadm_shapes)