How does Basecamp deal with elevation data?

If your unit has a barometer sensor it records elevation based on the barometer reading (the calibration set-up can affect results). If your unit doesn’t have a barometer sensor then it records elevation based on GPS elevation and a world geoid model. This gives a mean sea-level elevation. I’m not sure if you can ever see the GPS elevation on a Garmin. I believe you can see the geoid corrected GPS elevation on the satellite page.

Garmin tracks record a horizontal position (in decimal lat/lon), elevation (in meters) and time (date+time). All other information is calculated by the program and is not stored. If you alter the track, the information is recalculated. The data will be displayed in whatever units you have chosen but are stored as above.

For example, total distance sums all horizontal leg lengths which are calculated from adjacent horizontal coordinates. Total ascent and descent are sign based sums of the elevation change between two adjacent points.

Yes, if you draw a track on a map with a DEM (digital elevation model), it will save elevation data for each track point based on the DEM.

I’m not sure what you mean by consistency (precision or accuracy?). But each has its own problems. A digital elevation model assumes a constant gradient between contour lines. Consequently all the little hills and valleys between contours will are not measured. Elevation based only on barometer reading assumes a constant barometric reading at a fixed elevation. Elevation based on the GPS elevation will be affected by the array of visible satellites. Garmin sensor units have an option to correct the barometer calibration based on the GPS elevation, but I have no idea how this works or how well it works.

Any questions?

If your unit has a barometer sensor it records elevation based on the barometer reading (the calibration set-up can affect results). If your unit doesn’t have a barometer sensor then it records elevation based on GPS elevation and a world geoid model. This gives a mean sea-level elevation. I’m not sure if you can ever see the GPS elevation on a Garmin. I believe you can see the geoid corrected GPS elevation on the satellite page.

Garmin tracks record a horizontal position (in decimal lat/lon), elevation (in meters) and time (date+time). All other information is calculated by the program and is not stored. If you alter the track, the information is recalculated. The data will be displayed in whatever units you have chosen but are stored as above.

For example, total distance sums all horizontal leg lengths which are calculated from adjacent horizontal coordinates. Total ascent and descent are sign based sums of the elevation change between two adjacent points.

Yes, if you draw a track on a map with a DEM (digital elevation model), it will save elevation data for each track point based on the DEM.

I’m not sure what you mean by consistency (precision or accuracy?). But each has its own problems. A digital elevation model assumes a constant gradient between contour lines. Consequently all the little hills and valleys between contours will are not measured. Elevation based only on barometer reading assumes a constant barometric reading at a fixed elevation. Elevation based on the GPS elevation will be affected by the array of visible satellites. Garmin sensor units have an option to correct the barometer calibration based on the GPS elevation, but I have no idea how this works or how well it works.

Any questions?

Fantastic, thanks very much indeed - that's really clear. "Consistency" was a poor communication sorry; I am co-ordinating the submissions of several contributors who are planning hikes using GPS traces heavily edited in Basecamp....and need a consistent method of measuring length/ascent & descent in order to make judgments about their relative difficulty. I'm less concerned about accuracy (within reason!).
I understand about the relative merits of barometric vs GPS gained elevation data, but it seems to me that tracks plotted from DEM using Basecamp will give me what I'm looking for.
I must admit I'm surprised tho by what you say about DEM assuming a constant gradient between contour lines. I had assumed from the stats on elevation changes in Basecamp (and other mapping programmes I've used) that the DEM is hidden from view - the values returned by the programme are usually way higher than those produced by simply counting how many contour lines a track passes through, so I'm not sure I've understood 100%. I'll have a play around and see what happens. Many thanks again.

I confess that I over simplified things a bit and chose my words poorly concerning DEM data. The point was that the data will not catch every undulation in the real terrain. How much of an issue this would be depends on the terrain, spacing of the DEM data, length of track and degree of accuracy required. For your purposes I would guess that using the DEM data would be fine. It might be interesting to compare the graphs of a recorded track with a tracing of the track.

I'll do that, thank you. Do you know where DEM"s come from? Are there lots of different providers used by digital map makers? I'm wondering eg whether the elevation values produced by a track plotted in Topo Espana are comparable with those produced in a Topo Italy map. They're both Garmin branded products.
In the end it's only of academic interest, I've got a protocol now and I'm very happy about it!

I’m afraid I know the answer to that question. My guess would be that they would first try to get the data from the national agency in each country responsible for surveying/mapping.

If national agencies of both countries supply paper topo maps of similar scale I suspect they will eventually, if not already, have DEMs of similar density. If the map scales of the Garmin products are different (2x or more), the DEM densities may be different.

I think the newer US 100k topo maps have DEM data. If someone has both 100k and 24k products in a mountainous area, it might be interesting to compare graphs of tracks drawn on the maps (you’ll have to draw the track twice, once for each map). Try to find a spot where there is an obvious difference in the contours. Any takers?