⚙️ How-To Dams & reservoirs Maps & visualisation Notebook By · Updated

How to Map Dams and High-Hazard Dams by County

On this page

Problem statement

Some counties hold hundreds of dams, most of them small earthen dams on farm ponds and stock tanks; others have a handful of large ones. A county map of all dams shows where people have dammed small streams. A map of high-hazard dams shows where a failure would put lives at risk. Both maps start from the same join: the inventory's county column holds a county name, not a code, so the dams are placed in counties by their coordinates instead. This page counts dams per county both ways and maps them.

Quick answer

Join the dam points to the county polygons spatially, then count all dams and high-hazard dams per county:

dams = gpd.sjoin(nid[["nidId", "publicHazardId", "geometry"]], counties[["GEOID", "NAME", "STUSPS", "geometry"]].to_crs(nid.crs), predicate="within")
per = dams.groupby(["GEOID", "NAME", "STUSPS"]).agg(dams=("nidId", "size"), high=("publicHazardId", lambda s: (s == "High").sum())).reset_index()
print(per.nlargest(5, "dams").to_string(index=False))
print(per.nlargest(5, "high").to_string(index=False))
print(f"dams placed in a county: {len(dams):,} of {len(nid):,}; counties with a dam: {len(per):,}")
GEOID      NAME STUSPS  dams  high
25027 Worcester     MA   388   103
37183      Wake     NC   363   207
20087 Jefferson     KS   338     5
12105      Polk     FL   325    19
19193  Woodbury     IA   310    11
GEOID        NAME STUSPS  dams  high
37183        Wake     NC   363   207
25027   Worcester     MA   388   103
29099   Jefferson     MO   136   102
37119 Mecklenburg     NC   156    89
29221  Washington     MO   119    88
dams placed in a county: 92,595 of 92,626; counties with a dam: 3,041

Worcester County, Massachusetts, has the most dams in the inventory, 388, followed by Wake County, North Carolina, Jefferson County, Kansas, and Polk County, Florida. The high-hazard ranking is different. Wake County leads with 207 high-hazard dams, more than half its total, and Worcester is second with 103. Jefferson County, Kansas, has 338 dams but only 5 high hazard. 3,041 counties have at least one dam.

County map of the lower 48 shaded by number of dams, darkest in Montana and the Dakotas, the southern plains, the Southeast and New England. County map of the lower 48 shaded by number of dams, darkest in Montana and the Dakotas, the southern plains, the Southeast and New England.
Dams per county.

The join, checked

1. Why not join on the county column

names = nid.countyState.str.replace(" County", "", regex=False)
key = counties.NAME + ", " + counties.STATE_NAME
print(f"rows whose county name matches a 2024 county by name: {names.isin(key).mean():.1%}")
print(nid.loc[~names.isin(key), "countyState"].value_counts().head(4).to_string())
rows whose county name matches a 2024 county by name: 98.4%
countyState
Fairfield, Connecticut     220
Litchfield, Connecticut    220
New Haven, Connecticut     217
Hartford, Connecticut      171

A name join would work for 98.4% of rows. The failures are mostly Connecticut, whose dams still carry the eight historical counties, which the Census Bureau's 2024 boundaries replace with planning regions. A spatial join sidesteps names, spellings and boundary changes.

2. Dams outside every county

out = nid[~nid.nidId.isin(dams.nidId)]
print(f"outside the county polygons: {len(out)}; by state: {out.state.value_counts().head(4).to_dict()}")
outside the county polygons: 29; by state: {'Texas': 8, 'Virginia': 7, 'Michigan': 3, 'Wisconsin': 2}

31 rows, under 29 distinct ids, fall outside every county polygon. The 1:20 million county outlines are simplified, so a dam right at a shoreline or a state line can land just outside them. For exact counts use the 1:500,000 boundaries, or assign the stragglers to the nearest county with sjoin_nearest.

County map shaded by number of high-hazard dams, darkest in the Carolinas' Piedmont, Missouri, southern New England and parts of California and Arizona. County map shaded by number of high-hazard dams, darkest in the Carolinas' Piedmont, Missouri, southern New England and parts of California and Arizona.
High-hazard dams per county.

High-hazard density

The data used on this page

The US Army Corps of Engineers' National Inventory of Dams (NID), the whole inventory as a GeoPackage from its API, downloaded on 19 October 2026 and saved on this site, because the inventory is updated as states and agencies send revisions. Heights are in feet, storage in acre-feet and areas in acres, as the NID gives them.

from pathlib import Path
import zipfile

import geopandas as gpd
import numpy as np
import pandas as pd
import requests

Path("data").mkdir(exist_ok=True)
# US Army Corps of Engineers, National Inventory of Dams: the whole inventory as a GeoPackage, and the Census
# Bureau's 2024 cartographic boundary counties. Saved 19 October 2026, because the inventory is updated continuously.
# Live: https://nid.sec.usace.army.mil/api/nation/gpkg
ARCHIVE = "usace-nid-2026-10-19.zip"
if not Path(f"data/{ARCHIVE}").exists():
    Path(f"data/{ARCHIVE}").write_bytes(requests.get(f"https://data.spatialworkflow.io/{ARCHIVE}", timeout=900).content)
Z = zipfile.ZipFile(f"data/{ARCHIVE}")
if not Path("data/nid.gpkg").exists():
    Z.extract("nid.gpkg", "data")
NUMERIC = ["damHeight", "hydraulicHeight", "structuralHeight", "nidHeight", "damLength", "volume", "yearCompleted",
           "nidStorage", "maxStorage", "normalStorage", "surfaceArea", "drainageArea", "maxDischarge"]
nid = gpd.read_file("data/nid.gpkg")
nid[NUMERIC] = nid[NUMERIC].apply(pd.to_numeric, errors="coerce")       # the GeoPackage stores them as text
Path("data/cb_2024_us_county_20m.zip").write_bytes(Z.read("cb_2024_us_county_20m.zip"))
counties = gpd.read_file("data/cb_2024_us_county_20m.zip")
print(f"National Inventory of Dams: {len(nid):,} rows, {nid.nidId.nunique():,} NID ids; counties: {len(counties):,}")
National Inventory of Dams: 92,626 rows, 91,840 NID ids; counties: 3,222

Example 1: high-hazard dams per 1,000 km²

per = per.merge(counties[["GEOID", "ALAND"]], on="GEOID")
per["high_per_1000km2"] = per.high / per.ALAND * 1e9
print(per[per.high >= 10].nlargest(5, "high_per_1000km2")[["NAME", "STUSPS", "high", "high_per_1000km2"]].round(1).to_string(index=False))
       NAME STUSPS  high  high_per_1000km2
       Wake     NC   207              95.8
    Passaic     NJ    46              95.4
   Rockwall     TX    26              78.9
     Warren     MO    78              70.3
Mecklenburg     NC    89              65.6

Among counties with at least ten high-hazard dams, Wake County again has the most per area, 95.8 per 1,000 km², closely followed by Passaic County, New Jersey. Rockwall in Texas, Warren in Missouri and Mecklenburg, home of Charlotte, follow. Most of them are fast-growing suburban counties, where houses now stand below dams that once had only fields downstream.

Horizontal bars of high-hazard dams per 1,000 square kilometres for twelve counties, led by Wake, North Carolina, at 95.8 and Passaic, New Jersey, at 95.4. Horizontal bars of high-hazard dams per 1,000 square kilometres for twelve counties, led by Wake, North Carolina, at 95.8 and Passaic, New Jersey, at 95.4.
Counties most densely covered by high-hazard dams.

Example 2: the choropleth

m = counties.merge(per[["GEOID", "dams", "high"]], on="GEOID", how="left")
m = m[~m.STATEFP.isin(["02", "15", "60", "66", "69", "72", "78"])].to_crs(5070)
ax = m.plot(column="dams", scheme="FisherJenks", k=6, cmap="YlGnBu", legend=True, missing_kwds={"color": "#eeeeee"}, figsize=(11, 6))
ax.set_axis_off()
print(f"lower 48 counties with no dam in the inventory: {m.dams.isna().sum()}")
lower 48 counties with no dam in the inventory: 116

On the choropleth, the darkest counties are in the plains of Montana and the Dakotas, from north Texas to Kansas, across the Southeast and in New England. Only 116 counties in the lower 48 have no dam in the inventory at all. The high-hazard map looks different: its darkest counties are in the Piedmont of the Carolinas, the Ozarks of Missouri, southern New England and some large counties of California and Arizona.