Clinton:

If the data you get from the aeromagnetic and radiometric survey provide you the ellipsoidal height of ground points, you probably can build a regular mesh of height points (you must convert from ellipsoidal to geodetical by means of a geoid model EGM 2007). You can use ERDAS Imagine or SURFER to do interpolation and get a DEM. But as you say that the gap between the measurements is about 200 meters, I assume that the measurements are only on these points, then your DEM will have very low resolution. Remember that you have the USGS DTED 2 (30 meters) free DEM that will be better than 200 meters. On the other hand, if you have an 8 meter resolution DEM, this resolution is better than the average DEMs that you can get for free.

As Alexander says, you can do stereophotogrammetry if you have imagery or videos taken from the airplane, with an overlap of at least 60% . There are several software packages that provide you a DEM from a block of aerial photographs. Also by uing INSAR you can get DEMS from satellite radar imagery TERRASAR, COSMO SKYMED, etc , up to 2.5 meter resolution, but this expensive and it doesn´t work in selvatic or forested areas.

Obviously, LIDAR should be the better solution.

I hope this helps.

Regards

Gabriel

Dear Clinton,

It should be check that the Airborne geophysical survey also provide mapping solution by using, LiDar or photogrametry method. I have been work with several airborne gravity survey, and none of them provide mapping solution.

Regrads,

Brian

Thanks for your answer Gabriel. On the basis of your answer and that from Alexander and Brian, stereophotogrammetry sounds the most promising path. I imagine that processing stereophotos for 30,000 km2 would be rather challenging? I'd imagine we would need specialist software and computing capabilities?

Thanks again.

Clint

Yes Clinton. If you want to have a first approximation you can download PHOTOMOD , a classical photogrammetry software with capabilities to process UAV and satellite (like SPOT) stereo imagery. PHOTOMOD Lite version is free and has the full software capabilities but it is restricted to no more than 40 photographs that obviously is not enough for your area. It is restricted also in the size of the DEM you can generate (some few megabytes. You can also download the training videos to understand how to operate the software. The documentation is enough to understand, learn and start processing. There are five or six datasets with examples overall the world that allow you to do the whole processing for traditional photogrammetric cameras like ADS80, Zeiss RMK,...and more accesible, non photogrammetric cameras like Canon EOS 500. You have examples that include UAV processing.

ERDAS Imagine and ERDAS 2014 have also photogrammetric packages. I have no experience with open source free packages but surely you can search for it and you will find something useful for a first attrempt.

Rgarding the computing capabilities, for a big block of photos you need strong computing capabilities. And remember that you will need accurate ground control points (X,Y,Z) measured with GPS in differential mode. Groung control points is a key issue to get a precise DEM.

Best regards

Gabriel

Hi Gabriel,

Thanks once again for your response. I'm thinking of restricting the photogrammetry to lowland areas for which slope is

Clinton,

You can compute a DEM by subtracting the altimetry from the GPS Z data of the survey aircraft. It is a requirement of mag-rad airborne geophysical survey to collect altimetry data at the same rate as the geophysical data. Aeromagnetic is usually collected at 10 Hz and radiometric at 1 Hz. With typical survey speed of 270 km/h, so 75 m/s, this would give you one data point every 7.5m or 75m respectively. On multi-sensor survey, the higher data rate of aeromagnetic is usually applied to all the ancillary sensors, including the altimeter. However, across lines, you are limited by the survey line spacing, so 200m, from what you describe. So, you could then have, in theory, a spatial resolution of 7.5m along flight lines and 200m across flight lines.

While the spatial resolution is coarse, note that it is usually very accurate, since the GPS data of the survey aircraft is usually differentially corrected, and accuracy of a few cm can be expected. This is especially interesting when translated in the vertical accuracy of the dem that would be produced from the airborne geophysical survey data.

Please also note that radar altimetry will usually reflect at the ground surface, through vegetation, while laser will not penetrate vegetation and will represent the top of the canopy.

Richard

More to this...

It is also common to collect down-looking video or still photos during airborne geophysical surveys. This is usually to confirm the presence of ‘culture’ – power lines, ships, rails… - that could affect the magnetic signal and need to be corrected for. The quality of this imagery can vary a lot, and I am not sure about the overlap between images; it is probably good along lines, but may be limited across. It is usually not release with the data, but may have been archived by the organization sponsoring the geophysical survey.

I am wondering if it is possible to compile DEM with this imagery by using photogrammetric tools? If anybody endeavor into this, I would be very curious to know about any successes or limitations…

Richard

Hi Richard,

Thank you very much for your answers - very helpful. I'll keep you posted if we are able to progress things and get a useful output.

Regards,

Clint

Hey Clint,

I would contact Christopher Gomez at UC (http://www.canterbury.ac.nz/spark/researcher.aspx?researcherid=4353439) and see how he can help you. He is a bit of a photogrammetric whizkid. Not to mention a nice fella!

Cheers

Brendan

Thanks Brendan. Much appreciated.

Cheers

Clint

Actually Chris left UC a year or so ago! He's currently in Kobe in Japan.

Cheers

Kate

Thanks Kate. That may explain the lack of response! Regards Clint