Which GPX elements are read?

The tool reads latitude/longitude from track points (trkpt), route points (rtept), and waypoints (wpt). Points outside valid latitude/longitude ranges are ignored, and invalid XML is skipped safely.

How is the heat map computed?

Each coordinate is projected with the Web Mercator formula so the map keeps the right shape, then snapped into a pixel grid that counts overlapping points. A Gaussian-style blur spreads the density, and intensity is log-scaled and mapped to a blue-to-red colour ramp.

Why Web Mercator instead of plain latitude/longitude?

Plotting raw latitude and longitude stretches areas away from the equator. Web Mercator linearises latitude the same way web map tiles do, so your tracks keep correct proportions and angles at any location.

Do I need an internet connection or Strava account?

No. There are no map tiles, no API keys, and no account needed. The heat map is drawn purely from your own GPX point density, so it works completely offline.

Is my location data uploaded?

Never. Every file is read with the browser's local FileReader and processed on a canvas in your tab. Your routes and home location stay entirely on your device.

Why is one corner of my heatmap far brighter than everywhere else?

That is almost certainly your home. Because every activity starts and ends there, the start/end point accumulates far more overlap than any route. If you plan to share the image, crop or blur that cluster — it is the same signal that let researchers locate military bases from Strava's aggregate heatmap in 2018.

My heatmap shows one dot and lots of empty space — what happened?

A GPS spike: a single erroneous point in the wrong country or out at sea stretches the projection to include it, shrinking all your real tracks to a dot. Find and remove the bad point in that GPX file, or exclude the file, and the map will re-scale correctly.

What is the GPX Activity Heatmap?

Render a density heat map from multiple GPX files entirely in your browser. Track points are Web-Mercator projected, binned into a density grid, blurred, and painted as a warm-to-cool heat map showing your most-travelled routes. No tracks are uploaded; your activity data stays private. It runs free in your browser on Gera Tools, with nothing uploaded.

GPX Activity Heatmap

Name: GPX Activity Heatmap
Creator: Gera Tools
License: https://creativecommons.org/licenses/by/4.0/

The GPX Activity Heatmap builds a personal heat map of everywhere you’ve travelled from your own GPX files — no Strava subscription, no upload, no account. Drop in a season of runs or rides and instantly see which routes you favour, all rendered privately in your browser.

How the heatmap is rendered

The tool parses each GPX file’s XML and collects every coordinate from trkpt, rtept, and wpt elements. Those latitude/longitude pairs are then turned into a heat map in three stages:

Projection. Each point is projected with Web Mercator, which linearises latitude using y = ln(tan(π/4 + lat/2)). This is the same projection web maps use, so your tracks keep correct shape and proportion regardless of where on Earth they are.
Density binning. Projected points are scaled to fit the canvas and counted into a pixel grid, so places you’ve crossed many times accumulate higher values.
Blur and colour. A separable box blur (applied twice to approximate a Gaussian) spreads each point into a soft glow, intensity is log-scaled so even single-pass trails stay visible, and the value is mapped onto a blue → cyan → green → yellow → red ramp.

Reading the map

Cool blue areas are routes you’ve travelled once or twice. As overlap builds — your commute, a favourite loop, the streets near home — the colour warms through green and yellow to red at the busiest junctions. Loading more files makes the contrast sharper, since the colour scale adapts to the densest point in your whole dataset.

Notes and privacy

Because there are no background tiles, this is an abstract density view rather than a street map — but it is enough to recognise your regular routes and hotspots at a glance. Most importantly, your GPS data never leaves the browser: there are no servers, no tiles fetched, and no account, so you can safely visualise routes that reveal your home or workplace.

Privacy pitfalls a personal heatmap can create

A density map of your own tracks is exactly the data class that caused the 2018 Strava global-heatmap controversy, where aggregated activity revealed the layout and patrol routes of military bases. The same physics applies to an individual: the single hottest cluster on your map is almost always your home, because every run and ride starts and ends there. If you ever share or publish a heatmap image:

Crop or blur the start/end cluster. The brightest blob is your front door. Trim the render to exclude the first and last few hundred metres of every track, or exclude short “around the block” activities entirely.
Beware time metadata. GPX files carry timestamps. Even though this tool ignores them for the image, the files still reveal when you are routinely away from home — treat the raw GPX as sensitive.
Aggregate before sharing. A heatmap from hundreds of tracks is far less identifying than one from three; the more overlap, the more your incidental detours wash out and only genuine patterns remain.

This is precisely why the tool never uploads: keeping the computation in-browser means the one dataset that most reliably geolocates you never leaves your device.

Making a cleaner-looking heatmap

Load a whole season, not a handful. The colour scale normalises to the densest point, so with only a few tracks everything looks uniformly warm. Hundreds of overlapping tracks give the contrast that makes hotspots pop.
Split by activity type. Running and cycling tracks at different speeds produce different point densities (more GPS points where you move slowly), so a combined map can over-weight walking sections. Render runs and rides separately for a truer picture of each.
Watch for GPS spikes. A single erroneous point far offshore or in another country stretches the whole projection to fit it, shrinking your real routes to a dot. If the map looks empty with one lonely mark, a bad point is the cause — clean the offending GPX or exclude that file.
Denser sampling reads hotter. A watch logging every second paints a warmer trail than one logging every 10 seconds over the identical route. Consistent recording settings across files keep the comparison fair.

Where to export GPX files from

Almost every activity platform exports GPX, which you can then combine here:

Strava — open an activity, then Export GPX from the three-dot menu (bulk export of your whole history is available under Account → My Account → Download your data).
Garmin Connect — open an activity and choose Export to GPX.
Apple Health / Workouts — export via a third-party app that converts workout routes to GPX.
Komoot, Wahoo, Suunto, Coros — all offer a per-activity GPX download.

Drop a whole season of files in at once; the more overlapping tracks you load, the sharper the contrast, because the colour scale normalises to the densest point across your entire dataset.

Sources and references

GPX 1.1 schema (Topografix) — the GPS Exchange Format the tool parses (trkpt, rtept, wpt)
EPSG:3857 — Web Mercator projection — the projection used so tracks keep correct shape at any latitude

Maintained by the Gera Tools editorial team. Points are Web-Mercator projected, density-binned, blurred, and log-scaled onto a blue→red ramp entirely in your browser — no upload, no map tiles, no account. Last reviewed 2026-07-02.