Which DEM data is better for elevation mapping, ASTER or SRTM satellite data?

Aster because it is available freely with 30 m resolution. I do not think SRTM is of 30 m for every regions. 

SRTM is better than ASTER, it is available freely with 30 m spatial resolution at Earthexplorer for wholw globe.

http://earthexplorer.usgs.gov/

What about quality features other than resolution and availability? Errors due to undetected canyons/deep narrow valleys in mountains and variations in vegetation canopy cover elevations exceeding those of the terrain in plains? And how easy is it to correct or repair SRTM versus Aster for such errors.

Dear Sir ASTER is a 15 m resolution data and SRTM is 3 arc second (90m) so certainly ASTER is better, we have developed set methodoology and was quite successful. Please let me knwo if any help is required. You can download some papers on thsi aspect and report if you wish.

I'd say that from working with it in forested areas, ASTER has more small errors in topography than the latest SRTM DEM version.  It would be nice to use one to correct the other.  So I'd say download both and compare for your region.  Neither is perfect--often the SRTM DEM is a DSM, not a DTM, for example.  

It depends on the study that you plan for. SRTM has better positionnal accuracy, but it presents less details because of its mesh size. GDEM presents more details, but it contains the elevation of the canopy. There is an interesting new article which can help, you find it in the following link:

https://www.researchgate.net/publication/268157579_Effect_of_Scale_on_the_Correlation_Between_Topography_and_Canopy_Elevations_in_an_Airborne_InSAR_Product_Over_AMAZONIA

 I think SRTM is much better than DEM, because it presents more details and now there is  new version of SRTM 1-arc available at  http://earthexplorer.usgs.gov/ for all the world.

There is no absolute answer. The main difference between SRTM and ASTER GDEM is the mesh size since SRTM is generally available with a 90 m mesh, but the mesh size is not the resolution. Note that both DTMs have a bias due to vegetation (elevation is given on the top of canopy). There are so many criteria (position bias and accuracy, resolution, slope accuracy, drainage network positional accuracy and topological consistency etc. as well as a variety of fractal or spectral indicators) that it is impossible to give a general answer. I agree with Matthew E Fagan : download both DTMs and test them in your study area, after defining the quality criteria you are interested in.  The attached paper suggests several criteria for the comparison of two elevation data bases.

I've had pretty bad experiences with ASTER for landscape archaeology and erosion modelling: large artefacts (mainly peaks) particularly in flat terrain which are very difficult to remove through filtering.

I eventually settled with 90m SRTM data, but recently 30m resolution data has been made available for much of Europe, Asia, Africa, the Americas, Oceania and Australia. It's not free from artefacts either, but if you're familiar with GRASS GIS or GDAL there's a fantastic "mdenoise" module/utility which does a great job of removing noise while preserving sharp features like ridge lines and valleys.

GRASS r.denoise module: 
http://grass.osgeo.org/grass64/manuals/addons/r.denoise.html

Denoising algorithm: 
http://personalpages.manchester.ac.uk/staff/neil.mitchell/mdenoise/

USGS/NASA SRTM press releases:
http://www.jpl.nasa.gov/news/news.php?release=2014-321
http://www.usgs.gov/newsroom/article_pf.asp?ID=4068
http://www.usgs.gov/newsroom/article_pf.asp?ID=4109

Check out ACE2, it is a SRTM grid augmented with airborne and other local data. It does not feature nan values over land and has the same resolution of SRTM. They also provide standard deviation grids to check the quality.

http://tethys.eaprs.cse.dmu.ac.uk/ACE2

Thanks a lot  Robbi Bishop for fruitful explanation.

 Dear Saifuddin,

Please send me some paper in my email:ahmadali.syed@gmail.com   

Also send me your paper on Aravalli mapping of Thrusts may be 2000 publication year. 

Laurent Polidori and Mathew.E Fagan thanks for suggesting use of both and find which one best suitable for my study.

ASTER is better then SRTM according resolution.
if you r working for Indian continent then may go for Cartosat DEM with better resolution following link (http://bhuvan.nrsc.gov.in/data/download/index.php)

It depends on the region that you are interested in and the issue, but, in general terms I vote for SRTM (but, notice that it is like a world picture 10+ years old). I tested SRTM versión 4.1 for PERU.

As several answers voted for SRTM, I would like to add another vote for SRTM base on our study conducted in forest environment. In the study, we compared the performance of ASTER GDEM2 and SRTM.

It depends on the size of the study area. If you want to make a study of regional scale SRTM is for sure better. But if you want to study of local scale them ASTER is better.

SRTM is better

Depends on scale, location and physical landscape characteristics of your study area. But generally STRM is better. 

STRM is the best for elevation.

Getting in really late on this, but it really depends what you want to do with the data... STRM is best if what you're looking for is only great resolution in the vertical scale. ASTER's plus point is 1-arc minute coverage of basically everywhere. If you're doing very local scale studies then you might consider ASTER for its better horizontal resolution (unless you're in a position where you have 1-arc second STRM coverage).

SRTM is better for topographical mapping