Update parallel code

Update variable and file names, remove unused stuff

Top-down/parallelization/README.md

@@ -57,12 +57,14 @@ Input data layers must be named and placed according to some rules, as follows:
 
 - Create a working directory and set it as DATA_DIR (e.g. ./data) in `common.py`.
   Inside the workdir, the data folders must follow this structure:
+
 ```
     DATA_DIR/ADM: Administrative boundaries as '.gpkg' (multiple levels)
     DATA_DIR/HZD: Hazard layers as '.tif' using CRS 4326
     DATA_DIR/EXP: Exposure layers as '.tif' using CRS 4326
 ```
 - To name datasets, use [`ISO3166_a3`](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3) country code followed by specific data identifier - which is not fixed, you need to edit it in `runAnalysis.py`.
+
 ```
     CCDR_tools/ADM/SEN_ADM.gpkg
     CCDR_tools/HZD/SEN_FL_RP10.tif
@@ -73,29 +75,30 @@ Input data layers must be named and placed according to some rules, as follows:
 ## Setting parameters
 
 Edit the `main.py` file to specify:
-- **country (`country_dd`)**: [`ISO3166_a3`](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3) country code
-- **hazard type (`haz_cat_dd`)**: `'FL'` for floods; `'HS'` for heat stress; `'DR'` for drought; `'LS'` for landslide
+- **country (`country`)**: [`ISO3166_a3`](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3) country code
+- **hazard type (`haz_cat`)**: `'FL'` for floods; `'HS'` for heat stress; `'DR'` for drought; `'LS'` for landslide
 - **return periods (`return_periods`)**: list of return period scenarios as in the data, e.g. `[5, 10, 20, 50, 75, 100, 200, 250, 500, 1000]`
-- **exposure categories (`exp_cat_dd_list`)**: list of exposure categories: `['pop', 'builtup', 'agri']`
-  - exposure categories file name (`exp_cat_dd_list`): list  of same length of `exp_cat_dd_list` with file names for exposure categories: `['WPOP20', 'WSF19', 'ESA20_agri']`
-    If 'None', the default `['WPOP20', 'WSF19', 'ESA20_agri']` applies
-- **analysis approach (`analysis_app_dd`)**: `['Classes', 'Function']`
+- **exposure categories (`exp_cat_list`)**: list of exposure categories: `['POP', 'BU', 'AGR']`
+  - exposure categories file name (`exp_cat_list`): list  of same length of `exp_cat_list` with file names for exposure categories, e.g.: `['GHS', 'WSF19', 'ESA20']`
+    If 'None', the default `['POP', 'BU', 'AGR']` applies
+- **analysis approach (`analysis_app`)**: `['Classes', 'Function']`
   - If `'Function'`, you can set minimum hazard threshold value (`min_haz_slider`). Hazard value below this threshold will be ignored
   - If `'Classes'`,  you can set the number and value of thresholds to consider to split hazard intensity values into bins (`class_edges`)
-- **admin level (`adm_dd`)**: specify which boundary level to use for results summary (must exist in the `ISOa3`_ADM.gpkg file)
+- **admin level (`adm`)**: specify which boundary level to use for results summary (must exist in the `ISOa3`_ADM.gpkg file)
 - **save check (`save_check_raster`)**: specify if you want to export intermediate rasters (increases processing time) `[True, False]`
 
-Example of `main.py` running flood analysis (`haz_cat_dd`) over Cambodia (`country_dd`) for 10 return periods (`return_periods`) over three exposure categories (`exp_cat_dd_list`) using hazard classes according to thresholds (`class_edges`); results summarised at ADM3 level (`adm_dd`). Do not save intermediate rasters (`save_check_raster`).
+Example of `main.py` running flood analysis (`haz_cat`) over Cambodia (`country`) for 10 return periods (`return_periods`) over three exposure categories (`exp_cat_list`) using hazard classes according to thresholds (`class_edges`); results summarised at ADM3 level (`adm`). Do not save intermediate rasters (`save_check_raster`).
+
 ```
     # Defining the initial parameters
-    country_dd         = 'KHM'
-    haz_cat_dd         = 'FL'
+    country            = 'KHM'
+    haz_cat            = 'FL'
     return_periods     = [5, 10, 20, 50, 75, 100, 200, 250, 500, 1000]
     min_haz_slider     = 0.05
-    exp_cat_dd_list    = ['pop', 'builtup', 'agri']
-    exp_nam_dd_list    = ['WPOP20', 'WSF19', 'ESA20_agri']
-    adm_dd             = 'ADM3'
-    analysis_app_dd    = 'Classes'
+    exp_cat_list       = ['POP', 'BU', 'AGR']
+    exp_nam_list       = ['GHS', 'WSF19', 'ESA20']
+    adm                = 'ADM3'
+    analysis_app       = 'Classes'
     class_edges        = [0.05, 0.25, 0.50, 1.00, 2.00]
     save_check_raster  = False
 ```

Top-down/parallelization/main.py

@@ -7,29 +7,29 @@
 # Defining the main function
 def main():
     # Defining the initial parameters
-    country_dd         = 'KHM'
-    haz_cat_dd         = 'FL' #'FL' for floods; 'HS' for heat stress; 'DR' for drought; 'LS' for landslide
+    country         = 'BGD'			  # ISO3166-a3 code
+    haz_cat         = 'FL' 			  #'FL' for floods; 'HS' for heat stress; 'DR' for drought; 'LS' for landslide
     return_periods     = [5, 10, 20, 50, 75, 100, 200, 250, 500, 1000] # FL [5, 10, 20, 50, 75, 100, 200, 250, 500, 1000] # TC [10, 20, 50, 100, 200, 500] # add here as will
-    min_haz_slider     = 0.05 # FL 0.05 # TC 25.0 # ASI 0.01
-    exp_cat_dd_list    = ['pop', 'builtup', 'agri'] # ['pop', 'builtup', 'agri']
-    exp_nam_dd_list    = ['WPOP20', 'WSF19', 'ESA20_agri']
-                         # if None, the default applies: 'pop':'WPOP20', 'builtup':'WSF19', 'agri':'ESA20_agri', 'cstk'.'CSTK19' - If not None, expect a list of same length of exp_cat_dd_list
-    adm_dd             = 'ADM3' #['ADM1', 'ADM2', 'ADM3']
-    analysis_app_dd    = 'Classes' #['Classes', 'Function']
+    min_haz_slider     = 0.05 			  # FL 0.05 # TC 25.0 # ASI 0.01
+    exp_cat_list    = ['pop', 'BU', 'AGR']	  # ['pop', 'builtup', 'agri']
+    exp_nam_list    = ['POP', 'BU', 'AGR']	  # if None, the default applies: 'Population':'POP', 'Built-up':'BU', 'Agricultural land':'AGR' - If not None, expect a list of same length of exp_cat_list
+    adm             = 'ADM3' 			  #['ADM1', 'ADM2', 'ADM3']
+    analysis_app    = 'Function'		  #['Classes', 'Function']
     class_edges        = [0.05, 0.25, 0.50, 1.00, 2.00] # FL [0.05, 0.25, 0.50, 1.00, 2.00] # TC [17.0, 32.0, 42.0, 49.0, 58.0, 70.0] # DR_ASI [0.01, 0.10, 0.25, 0.40, 0.55, 0.70, 0.85]
     save_check_raster  = False
 
     import time
     start_time = time.time()
 
     # Running the analysis
-    if exp_nam_dd_list is not None and len(exp_nam_dd_list) != len(exp_cat_dd_list): sys.exit("ERROR: Parameter 'exp_nam_dd_list' should either be 'None' or have the same length as 'exp_cat_dd_list'")
-    # For every exp_cat_dd
-    for i in range(len(exp_cat_dd_list)):
+    if exp_nam_list is not None and len(exp_nam_list) != len(exp_cat_list): sys.exit("ERROR: Parameter 'exp_nam_list' should either be 'None' or have the same length as 'exp_cat_list'")
+    # For every exp_cat
+    for i in range(len(exp_cat_list)):
         # Defining the list variable to pass to run_analysis
-        exp_cat_dd = exp_cat_dd_list[i]
-        exp_nam_dd = exp_nam_dd_list[i]
-        run_analysis(country_dd, haz_cat_dd, return_periods, min_haz_slider, exp_cat_dd, exp_nam_dd, adm_dd, analysis_app_dd, class_edges, save_check_raster)
+        exp_cat = exp_cat_list[i]
+        exp_nam = exp_nam_list[i]
+        run_analysis(country, haz_cat, return_periods, min_haz_slider,
+                       exp_cat, exp_nam, adm, analysis_app, class_edges, save_check_raster)
 
     print("--- %s seconds ---" % (time.time() - start_time))
 

Top-down/parallelization/runAnalysis.py

@@ -25,7 +25,6 @@ def zonal_stats_parallel(args):
 # Defining the main function to run the analysis
 def run_analysis(country: str, haz_cat: str, valid_RPs: list[int],
                  min_haz_threshold: float, exp_cat: str, exp_nam: str, adm_name: str,
-                 time_horizon: list[int], rcp_scenario: list[str],
                  analysis_type: str, class_edges: list[float], 
                  save_check_raster: bool):
     """
@@ -73,28 +72,24 @@ def run_analysis(country: str, haz_cat: str, valid_RPs: list[int],
 
     # Checking which kind of exposed category is being considered...
     # If the exposed category is population...
-    if exp_cat == 'pop':
+    if exp_cat == 'POP':
         damage_factor = mortality_factor
-        exp_ras = f"{exp_folder}/{country}_WPOP20.tif"
+        exp_ras = f"{exp_folder}/{country}_POP.tif"
     # If the exposed category is builtup area...
-    elif exp_cat == 'builtup':
+    elif exp_cat == 'BU':
         damage_factor = damage_factor_builtup
-        exp_ras = f"{exp_folder}/{country}_WSF19.tif"
+        exp_ras = f"{exp_folder}/{country}_BU.tif"
     # If the exposed category is agriculture...
-    elif exp_cat == 'agri':
+    elif exp_cat == 'AGR':
         damage_factor = damage_factor_agri
-        exp_ras = f"{exp_folder}/{country}_ESA20_agri.tif"
-    # If the exposed category is capital stock...
-    elif exp_cat == 'cstk':
-        damage_factor = damage_factor_builtup
-        exp_ras = f"{exp_folder}/{country}_CSTK19.tif"
+        exp_ras = f"{exp_folder}/{country}_AGR.tif"
     # If the exposed category is missing, then give an error
     else:
         exp_ras = None
         ValueError(f"Missing or unknown data layer {exp_cat}")
     # If user-specified exposure file name is passed, then
     if exp_nam is not None: 
-        exp_cat = str(exp_cat+'_'+exp_nam)
+        exp_cat = str(exp_cat)
         exp_ras = f"{exp_folder}/{country}_{exp_nam}.tif"
 
     # Running the analysis