Sample GPX file: 50 km road cycling loop in Sonoma County

The route follows plausible roads in the Dry Creek Valley and Russian River area but the trackpoints don't trace specific paved routes. Coordinates are realistic for the named region but the file is synthetic — don't use it as a navigation reference. Local cycling clubs and Strava routes are better sources for actual rideable Sonoma routes.

Grade-aware pace coupling — slower on the climbs (12-18 km/h on the steeper sections), faster on descents (35-40 km/h on the long downs), normal on flats (~25 km/h). This matches a real cycling recording where the rider's effort is roughly constant but the speed varies with terrain. Apps that compute moving-time average speed should produce a value close to 25 km/h for this file.

Probably some accidental matches, since Strava's segments are dense in California wine country. The file is synthetic so any match is coincidental. Useful for testing your app's segment-matching logic against real Strava data; not useful as training data for the actual roads named in the route description.

No — this file has no power data. The Garmin TrackPointExtension namespace is absent. For a file with sensor data including power-equivalent fields, use the extensions-test.gpx sample which includes heart rate, cadence, and temperature; for true power data, you'll need a recording from a real power-meter sensor since synthesized power isn't included in any of the published samples.

Smart recording at 1-second cadence with grade-aware sampling. About one trackpoint every 21 metres on average, denser on the climbs (slower speed = more points per metre) and sparser on the descents. This shape is typical of real Wahoo or Garmin recordings; if your parser handles it cleanly, it'll handle most real cycling files.