Elevation & Terrain Data Scraper — altitude for any coordinates, no API key

Get the terrain elevation (altitude) of any latitude/longitude on Earth — and the depth of any point in the ocean — without an API key, without registration and without a Google Elevation API bill. Feed it a list of coordinates, a bounding box, or a route, and get back clean elevation data in metres and feet.

Powered by the free public OpenTopoData instance, which serves eleven verified digital elevation models (DEMs), from global 30 m SRTM coverage down to 8 m New Zealand LiDAR and European sea-floor bathymetry.

• No API key, no account, no OAuth. Just coordinates in, elevation out.
• Three modes: single points, bounding-box grids, and interpolated route profiles.
• Honest nulls. Outside a dataset's coverage you get elevation_m: null — never a fake 0.
• ~$0.50 per 1,000 elevation points on the Apify platform.

---

What you get

One dataset item per coordinate:

Field	Description
latitude	Latitude of the sampled point (decimal degrees)
longitude	Longitude of the sampled point (decimal degrees)
elevation_m	Elevation above sea level in metres. Negative for ocean depth. null when the dataset has no data there
elevation_ft	Same value in feet, rounded to 0.1. null when elevation_m is null
dataset	The DEM that answered (e.g. srtm30m)
in_coverage	true when the dataset actually returned a height for this coordinate
point_index	Position of the point in the request order — keeps grids and profiles ordered
path_distance_km	Cumulative great-circle distance from the first point. Path mode only, otherwise null
source	Always opentopodata
scraped_at	UTC ISO-8601 timestamp of the run

Sample output

[
  {
    "latitude": 47.421,
    "longitude": 10.9855,
    "elevation_m": 2937.0,
    "elevation_ft": 9635.8,
    "dataset": "srtm30m",
    "in_coverage": true,
    "point_index": 0,
    "path_distance_km": null,
    "source": "opentopodata",
    "scraped_at": "2026-07-28T18:10:03+00:00"
  },
  {
    "latitude": -30.0,
    "longitude": -40.0,
    "elevation_m": null,
    "elevation_ft": null,
    "dataset": "srtm30m",
    "in_coverage": false,
    "point_index": 1,
    "path_distance_km": null,
    "source": "opentopodata",
    "scraped_at": "2026-07-28T18:10:03+00:00"
  }
]

The second row is the South Atlantic. SRTM is a land elevation model, so it correctly reports null — not 0. Sea level and "no data" are different things, and this Actor never conflates them. If you need ocean depth, pick a bathymetry dataset such as gebco2020 or mapzen.

---

Three modes

1. points — an explicit coordinate list

Give it ["52.5200,13.4050", "47.4210,10.9855"] and it returns one item per pair. Ideal for enriching an existing list of addresses, sensor sites, store locations or GPS fixes.

2. grid — a bounding box + step

Set bboxSouth, bboxNorth, bboxWest, bboxEast and gridStep (in degrees; 0.01 ≈ 1.1 km of latitude). The Actor builds the lat/lon grid itself and caps it against maxItems before a single request goes out, so a careless bbox can never turn into a runaway run. Perfect for terrain rasters, viewshed inputs and slope analysis.

3. path — waypoints interpolated into a profile

Give two or more waypoints in coordinates and a pathSamples count. The Actor resamples the polyline into evenly spaced points by ground distance (haversine, not degrees) and adds a cumulative path_distance_km to every row — an elevation profile you can plot straight away.

---

Datasets and their coverage (all verified live)

Dataset	Resolution	Coverage
srtm30m	~30 m	Global land, 60°N–56°S (NASA SRTM GL1) — default
srtm90m	~90 m	Global land, 60°N–56°S (NASA SRTM GL3)
aster30m	~30 m	Global land, 83°N–83°S (ASTER GDEM v3)
mapzen	~30 m	Global land and ocean, near-global composite (Mapzen Terrarium)
gebco2020	~450 m	Global land and ocean bathymetry (GEBCO 2020)
etopo1	~1.8 km	Global land and ocean, coarse (NOAA ETOPO1)
eudem25m	25 m	Europe (EEA member states) only (Copernicus EU-DEM v1.1)
ned10m	~10 m	Continental USA, Hawaii, parts of Alaska only (USGS NED / 3DEP)
bkg200m	200 m	Germany only (BKG DGM200)
nzdem8m	8 m	New Zealand only (LINZ)
emod2018	varies	European seas only — bathymetry, negative depths (EMODnet 2018)

Rule of thumb: use a regional dataset when your points are inside its region (it is far more precise), and a global one otherwise. Querying a regional dataset outside its region returns elevation_m: null with in_coverage: false. That is correct behaviour, not a failure.

---

Rate limits — read this before planning a big run

This Actor talks to the free public OpenTopoData instance. Verified live against the running service (x-opentopodata-version: 1.9.0):

• 100 locations per request. Sending 101 returns HTTP 400 with "Too many locations provided (101), the limit is 100." — confirmed for both GET and POST. The Actor chunks your coordinates into batches of exactly 100.
• About 1 call per second. Three back-to-back calls produced an HTTP 429 on the third. The Actor waits ≥1.1 s between batches and backs off exponentially on 429.
• 1,000 calls per day, per the operator's published fair-use policy. Note: the public instance exposes no X-RateLimit-* or Retry-After headers, so the daily counter cannot be read from a response — you only find out by receiving a 429.

Practically, 1,000 calls × 100 locations = up to 100,000 elevation points per day on the free public instance. If you hit the quota, the run fails with an explicit rate-limit message rather than silently returning nothing.

Need more? OpenTopoData is open source and self-hostable. Running your own instance removes the 1 call/second and 1,000 calls/day limits entirely (you only pay for your own server). This Actor's request logic is unchanged against a self-hosted instance.

---

Use cases

• Route & hiking elevation profiles. Path mode turns a GPX-style waypoint list into a plottable profile with cumulative distance and altitude gain.
• Solar and wind siting. Grid mode gives you the terrain raster you need for shading, slope, aspect and wind-exposure modelling across a candidate site.
• Flood risk modelling. Sample elevations across a floodplain, compare heights against river gauge levels, and rank addresses by height above a reference datum.
• GIS and data pipelines. Enrich any table of coordinates — property listings, cell towers, agricultural plots, delivery stops — with a height column, with no key management.

---

Input example

{
  "mode": "path",
  "coordinates": ["47.4210,10.9855", "47.5000,11.1000"],
  "dataset": "srtm30m",
  "pathSamples": 100,
  "maxItems": 500
}

Grid example:

{
  "mode": "grid",
  "bboxSouth": 52.45,
  "bboxNorth": 52.55,
  "bboxWest": 13.30,
  "bboxEast": 13.45,
  "gridStep": 0.01,
  "dataset": "eudem25m",
  "maxItems": 500
}

---

Pricing

Pay-per-result: about $0.50 per 1,000 elevation points. A 500-point run costs roughly $0.25. Because up to 100 points ride in a single upstream call, even large grids finish quickly and cheaply.

---

Related Actors

• Geocoding Scraper — turn addresses into coordinates, then feed them straight into this Actor.
• Reverse Geocoding Scraper — turn coordinates back into addresses.
• OpenStreetMap Notes Scraper — community map notes for any area.
• Airports & Airlines Scraper — airport, runway and airline reference data.

---

FAQ

Do I need an API key? No. There is nothing to sign up for.

Why is my elevation null? The chosen dataset has no data at that coordinate — usually because it is a regional DEM and your point is outside its region, or a land-only DEM and your point is over water. Switch to a global dataset (srtm30m, aster30m) or a bathymetric one (gebco2020, mapzen).

Why not just return 0 for missing data? Because 0 m is a real, meaningful elevation (sea level). Substituting 0 for "unknown" silently corrupts every downstream flood, slope or profile calculation. This Actor keeps them strictly distinct via elevation_m: null and in_coverage: false.

How accurate is it? Depends on the DEM. SRTM 30 m has a vertical accuracy of roughly ±16 m; regional models like ned10m, eudem25m and nzdem8m are considerably better. Cross-checking two datasets for the same point is a cheap accuracy sanity check.

Can I get ocean depth? Yes — use gebco2020, mapzen or etopo1 globally, or emod2018 for European seas. Depths come back as negative elevation_m.

Data source and licensing. Elevation values come from public DEMs (NASA, USGS, ESA/Copernicus, GEBCO, BKG, LINZ, EMODnet) served via OpenTopoData. Check each dataset's own attribution requirements before redistributing.