Flood Hazard Dataset Build

backend/Data/flood-hazard/table_build.qmd

@@ -0,0 +1,87 @@
+---
+title: "Flood Hazard Table Build"
+format: html
+---
+
+## Load dataset
+
+```{python}
+import duckdb
+import pandas as pd
+
+con = duckdb.connect("flood.db")
+raw = con.execute("SELECT * FROM raw").df()
+raw.head()
+
+raw.isna().sum().sort_values(ascending=False)
+
+for col in raw.columns:
+    print("COLUMN:", col)
+    try:
+        print(raw[col].value_counts().head())
+    except Exception as e:
+        print("Could not compute value counts:", e)
+    print("\n")
+
+
+```
+
+## Null values
+
+```{python}
+for col in raw.columns:
+    print(col)
+    print(raw[col].value_counts(dropna=False).head(10))
+    print("\n")
+import numpy as np
+
+raw["VELOCITY"] = raw["VELOCITY"].replace(-9999, np.nan)
+
+
+```
+
+## Dropping columns
+
+```{python}
+
+
+drop_cols = [
+    "AR_SUBTRV",
+    "BFE_REVERT",
+    "DEP_REVERT"
+]
+
+drop_cols
+
+info_cols = [
+    "OBJECTID",
+    "AR_REVERT",
+    "DFIRM_ID",
+    "VERSION_ID",
+    "FLD_AR_ID",
+    "STUDY_TYP",
+    "FLD_ZONE",
+    "ZONE_SUBTYP",
+    "SFHA_TF",
+    "STATIC_BFE",
+    "V_DATUM",
+    "DEPTH",
+    "LEN_UNIT",
+    "VELOCITY",
+    "VEL_UNIT",
+    "DUAL_ZONE",
+    "SOURCE_CIT"
+]
+
+info_cols
+
+```
+
+```{python}
+geom_cols = ["OBJECTID", "geometry"]
+geom_cols
+
+final_cols = info_cols + ["geometry"]
+final_cols
+
+```

backend/build/flooding.py

@@ -0,0 +1,30 @@
+from build import BACKEND, CON, data_dir
+
+PARQUET = data_dir / "Flooding" / "vt-flood-hazard.parquet"
+proc_dir = BACKEND / "Data" / "_Processed" / "flood"
+SQL_DIR = BACKEND / "build" / "sql"
+
+
+def main():
+
+    # Load raw table from parquet
+    CON.execute(f"""
+        CREATE OR REPLACE VIEW raw AS
+        SELECT *
+        FROM read_parquet('{PARQUET}')
+    """)
+
+    # Run SQL files in order
+    for name in ["flood_info.sql", "flood_geom.sql", "flood_build.sql"]:
+        CON.execute((SQL_DIR / name).read_text())
+
+    for table in ["flood_hazard"]:
+        CON.execute(
+            f"COPY (SELECT * FROM {table}) TO '{proc_dir / 'Flooding' / f'{table}.parquet'}' "
+        )
+
+    print("Flood hazard tables built.")
+
+
+if __name__ == "__main__":
+    main()

backend/build/sql/flood_build.sql

@@ -0,0 +1,21 @@
+CREATE OR REPLACE TABLE hazard AS
+SELECT
+    i.id,
+    i.DFIRM_ID,
+    i.FLD_AR_ID,
+    i.STUDY_TYP,
+    i.FLD_ZONE,
+    i.ZONE_SUBTY,
+    i.SFHA_TF,
+    i.STATIC_BFE,
+    i.V_DATUM,
+    i.DEPTH,
+    i.LEN_UNIT,
+    i.VELOCITY,
+    i.VEL_UNIT,
+    i.AR_REVERT,
+    i.DUAL_ZONE,
+    i.SOURCE_CIT,
+    g.geometry AS geom
+FROM info AS i
+INNER JOIN geom AS g USING (id);

backend/build/sql/flood_geom.sql

@@ -0,0 +1,5 @@
+CREATE OR REPLACE TABLE geom AS
+SELECT
+    OBJECTID AS id,
+    geometry
+FROM raw;

backend/build/sql/flood_info.sql

@@ -0,0 +1,54 @@
+CREATE OR REPLACE TABLE info AS
+SELECT
+
+    -- Unique ID assigned by FEMA
+    OBJECTID AS id,
+
+    -- Indicates whether the flood zone is under heightened risk
+    AR_REVERT,
+
+    -- FEMA Digital Flood Insurance Rate Map identifier
+    DFIRM_ID,
+
+    -- Version number of the FEMA flood hazard dataset
+    VERSION_ID,
+
+    -- Unique flood area identifier used by FEMA
+    FLD_AR_ID,
+
+    -- Type of flood study conducted
+    STUDY_TYP,
+
+    -- FEMA flood zone designation
+    FLD_ZONE,
+
+    --Subtype of the flood zone providing additional classification detail
+    ZONE_SUBTY,
+
+    -- Indicates whether the area is a Special Flood Hazard Area
+    SFHA_TF,
+
+    -- Base Flood Elevation in feet
+    STATIC_BFE,
+
+    -- Vertical datum used for the BFE measurement
+    V_DATUM,
+
+    -- Flood depth in feet
+    DEPTH,
+
+    -- Unit of measurement for linear values
+    LEN_UNIT,
+
+    -- Flood velocity in feet per second
+    VELOCITY,
+
+    -- Unit of measurement for velocity values
+    VEL_UNIT,
+
+    -- Indicates whether the area is part of dual-zone classification
+    DUAL_ZONE,
+
+    -- Citation or reference for the source of the flood hazard data
+    SOURCE_CIT
+FROM raw;