[1pt] PR: Tool to evaluate crosswalk accuracy

Dockerfile

@@ -54,6 +54,7 @@ ARG depDir=/dependencies
 ENV inputsDir=$dataDir/inputs
 ENV outputsDir=/outputs
 ENV srcDir=$projectDir/src
+ENV toolsDir=$projectDir/tools
 ENV workDir=/fim_temp
 ENV taudemDir=$depDir/taudem/bin
 ENV taudemDir2=$depDir/taudem_accelerated_flowDirections/taudem/build/bin

docs/CHANGELOG.md

@@ -1,6 +1,28 @@
 All notable changes to this project will be documented in this file.
 We follow the [Semantic Versioning 2.0.0](http://semver.org/) format.
 
+## v4.4.8.4 - 2024-01-12 - [PR#1061](https://github.com/NOAA-OWP/inundation-mapping/pull/1061)
+
+Adds a post-processing tool to compare crosswalked (conflated) `feature_id`s between NWM stream network to DEM-derived reaches. The tool is run if the `-x` flag is added to `fim_pipeline.sh`. Results are computed for branch 0 and saved in a summary file in the HUC output folder.
+
+### Additions
+
+- `tools/evaluate_crosswalk.py`: evaluates crosswalk accuracy using two methods:
+    - intersections: the number of intersections between streamlines
+    - network (or tree): compares the feature_ids of the immediate upstream segments
+
+### Changes
+
+- `Dockerfile`: added `toolsDir` environment variable
+- `fim_pipeline.sh`: added `-x` flag to run crosswalk evaluation tool
+- `fim_post_processing.sh`: changed hardcoded `/foss_fim/tools` to `toolsDir` environment variable
+- `fim_pre_processing.sh`: added `evaluateCrosswalk` environment variable
+- `src/`
+    - `add_crosswalk.py`: fix bug
+    - `delineate_hydros_and_produce_HAND.sh`: added a call to `verify_crosswalk.py` if evaluateCrosswalk is True.
+
+<br/><br/>
+
 ## v4.4.8.3 - 2024-01-05 - [PR#1059](https://github.com/NOAA-OWP/inundation-mapping/pull/1059)
 
 Fixes erroneous branch inundation in levee-protected areas.

fim_pipeline.sh

@@ -48,6 +48,7 @@ usage()
                             will be skipped.
       -isaws            : If this param is included, AWS objects will be used where possible
                         - Note: This feature is not yet implemented.
+      -x                : If this param is included, the crosswalk will be evaluated.
 
 
     Running 'fim_pipeline.sh' is a quicker process than running all three scripts independently; however,

fim_post_processing.sh

@@ -210,7 +210,7 @@ echo
 echo -e $startDiv"Combining crosswalk tables"
 # aggregate outputs
 Tstart
-python3 /foss_fim/tools/combine_crosswalk_tables.py \
+python3 $toolsDir/combine_crosswalk_tables.py \
     -d $outputDestDir \
     -o $outputDestDir/crosswalk_table.csv
 Tcount

fim_pre_processing.sh

@@ -101,6 +101,9 @@ in
     -isaws)
         isAWS=1
         ;;
+    -x)
+        evaluateCrosswalk=1
+        ;;
     *) ;;
     esac
     shift
@@ -129,6 +132,7 @@ if [ "$jobBranchLimit" = "" ]; then jobBranchLimit=1; fi
 if [ -z "$overwrite" ]; then overwrite=0; fi
 if [ -z "$skipcal" ]; then skipcal=0; fi
 if [ -z "$isAWS" ]; then isAWS=0; fi
+if [ -z "$evaluateCrosswalk" ]; then evaluateCrosswalk=0; fi
 
 # validate and set defaults for the deny lists
 if [ "$deny_unit_list" = "" ]
@@ -234,6 +238,7 @@ echo "export deny_branch_zero_list=$deny_branch_zero_list" >> $args_file
 echo "export has_deny_branch_zero_override=$has_deny_branch_zero_override" >> $args_file
 echo "export isAWS=$isAWS" >> $args_file
 echo "export skipcal=$skipcal" >> $args_file
+echo "export evaluateCrosswalk=$evaluateCrosswalk" >> $args_file
 
 echo "--- Pre-processing is complete"
 

src/add_crosswalk.py

@@ -258,7 +258,11 @@ def add_crosswalk(
             else:
                 update_id = output_flows.loc[output_flows.HydroID == short_id]['HydroID'].item()
 
-            str_order = output_flows.loc[output_flows.HydroID == short_id]['order_'].item()
+            output_order = output_flows.loc[output_flows.HydroID == short_id]['order_']
+            if len(output_order) == 1:
+                str_order = output_order.item()
+            else:
+                str_order = output_order.max()
             sml_segs = pd.concat(
                 [
                     sml_segs,

src/delineate_hydros_and_produce_HAND.sh

@@ -17,7 +17,7 @@ T_total_start
 
 ## MASK LEVEE-PROTECTED AREAS FROM DEM ##
 if [ "$mask_leveed_area_toggle" = "True" ] && [ -f $tempHucDataDir/LeveeProtectedAreas_subset.gpkg ]; then
-    echo -e $startDiv"Mask levee-protected areas from DEM (*Overwrite dem_meters.tif output) $hucNumber $branch_zero_id"
+    echo -e $startDiv"Mask levee-protected areas from DEM (*Overwrite dem_meters.tif output) $hucNumber $current_branch_id"
     date -u
     Tstart
     python3 $srcDir/mask_dem.py \
@@ -303,3 +303,18 @@ python3 $srcDir/add_crosswalk.py \
     -e $min_catchment_area \
     -g $min_stream_length
 Tcount
+
+## EVALUATE CROSSWALK ##
+if [ "$current_branch_id" = "$branch_zero_id" ] && [ "$evaluateCrosswalk" = "1" ] ; then
+    echo -e $startDiv"Evaluate crosswalk $hucNumber $current_branch_id"
+    date -u
+    Tstart
+    python3 $toolsDir/evaluate_crosswalk.py \
+        -a $tempCurrentBranchDataDir/demDerived_reaches_split_filtered_addedAttributes_crosswalked_$current_branch_id.gpkg \
+        -b $b_arg \
+        -c $tempHucDataDir/crosswalk_evaluation_$current_branch_id.csv \
+        -d $tempHucDataDir/nwm_headwater_points_subset.gpkg \
+        -u $hucNumber \
+        -z $current_branch_id
+    Tcount
+fi