Sign in to follow this  
Guest Giorgio

Resampling problems - missing data

Recommended Posts

Hi all.I'm working as a beginner on a mesh project for Cirenaica and Marmarica in North Africa, using the recently issued SRTM hgts. The software I'm using is Microdem to view the files, Blackart to clean them from unwanted data and Srtm_to_Bgl to produce the final bgl from the 20 hgt files - the result is quite good, but something unexpected is showing up:At the extreme northern part of the scenery (N32E21) data beyond 32.72 are not present in the bgl: the source hgt does comprehend them, but the final bgl does not, and does not end at N33 as it was supposed to. In tmfviewer as in the sim the bgl terminates with a clear cut at approximately N32.72 degrees?What is this up to?Any hint?maybe it's just something familiar to mesh designers, I hope!

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

Ciao Giorgio,The LOD grid has its own boundaries that vary from what you use as source. Richard Marklew's LODCalc (available in the AvSim library) will give you the boundaries of the LOD grid.The easiest solution to your problem is to include another hgt file extending your source to the North.Best regards.Luis

Share this post


Link to post
Share on other sites

Hi Luis,thanks for the assistance, I'm really a beginner in these things.I had imagined that including one more hgt file in the North could fix the issue, but actually there is no SRTM data for the area:Is it possible to build a fake hgt with zero elevation, and rename it according to the place it should occupy, and if so, how?thanks again,cheersGiorgioPS: How is your Paris 1789 project going along?This is something I would like to see in my sim!

Share this post


Link to post
Share on other sites

"there is no SRTM data for the area"There is, but you will have to use data for southern Europe, in the Eurasia data page, to extend your data coverage to the north. After you merge them, you may have to use Microdem to set missing values to sea level (or whatever the elevation of the Mediterranean is there).An alternative is to use a small subset of the GTOPO30 data for the area to extend the region. (rather than creating your own.) You will need e020n90.tar.gz, available at:ftp://edcftp.cr.usgs.gov/pub/data/gtopo30/global/* use Microdem to extract a subset for the area you need (no need to unzip it). No need to be exact either, graphical selection is adequate. Just a small piece of the Sea is enough, Microdem will fill the rest of the area with missing data values.* open the dem subset and save as DTED with 3 arcsec spacing* open your current dem and save it as DTED, same 3 arcsec spacing* merge the two, selecting the subset dem.dt1 first, then the original.dt1. This way the original data will be preserved.* set any missing data values in your new merged dem to the correct elevation (sea level?)This is a reasonable (but not perfect) way to fill large holes in the strm data as well, so it is a skill worth learning if you plan to do much work with that data. Better sources than the GTOPO30 data can be used, if available.I discuss your problem and have more mesh tips at www.fs-traveler.com.Steve

Share this post


Link to post
Share on other sites

Thanks Steve for the reply and for your site, where I have gathered much of the primary information to begin my attempts at mesh designing. Finally a clear set of procedures to merge different data! I was looking for such an end-user explanation to the issue, as my knowledge of what-is-into-the-box is almost none.This is the reason why, probably, I have so unorthodox ideas such as using another hgt file, previously razed to zero elevation, and rename it with the degrees of the needed file. It works!Thanks again,cheersGiorgio

Share this post


Link to post
Share on other sites

Giorgio, that is a great idea. I never thought of making a dummy hgt file to extend coverage. Brilliant!Probably you know that you can limit the coverage of the final bgl by including boundaries in the Destination part of the inf file. This will prevent your dummy hgt file from smashing the altitude in that part of the source.Thanks for asking about the Paris 1789 project - very kind of you. I am off on a Caribbean project for the moment, so have not had much time to work on it - it is all Holger's fault for providing great satellite images. But, as soon as I can, I shall put the finishing touches to it and let you know.Best regards.Luis

Share this post


Link to post
Share on other sites

Thanks for the kind words, Luis, as they confirm I'm slowly beginning to understand how it works.In fact I did'nt know that I could limit the coverage of the dummy file in the inf file, this will prove very helpful!Eager to see the final results of your Caribbean scenery,Cheers,Giorgio

Share this post


Link to post
Share on other sites

Nice work. Your approach is very practical, given your use of the srtm_to_bgl tool. And Luis' suggestion about editing the inf file is a good next step as you learn how to move from simply running a batch program to gaining more control over the process. A little imagination and a lot of experimenting are critical ingredients in learning to make good mesh. Steve

Share this post


Link to post
Share on other sites

But now I have a problem that imagination and experience alone cannot master, I'm afraid...Trying to deliver the whole set of tiles containing an area, at the boundary running along E20 (N30-N32) the coastal data of the tiles do not merge perfectly; the areas overlap for a stretch of several minutes. This is the result of ispection with Microdem, that confirms the result of the compiler.What shall I do?Could editing of the inf. file help?Thanks,Giorgio

Share this post


Link to post
Share on other sites

Hello Giorgio,Since nobody else has stepped in to answer, let me ask you for more details, please. It does not seem clear to me, so could you please explain what you did and perhaps show a screenshot of your result?Thanks.Best regards.Luis

Share this post


Link to post
Share on other sites

Thanks Luis for the offer; in fact I've realized that it was an issue caused by different sea-level despeckling parameters in Blackart. I had applied widely different parameters to distinct hgts, so the resulting bgl gave a visible linear rift crossing the landscape. With the same parameter applied to all confining bgls the rift disappeared. But a lot of the lowest altitudes too! Sorry for bothering anyone with a non-issue. But as CFS2 has no inbuilt flattens (what a blessing), if you apply low parameters to save altitude you get a sea full of bumps, if you try high parameters you clean your sea but also erase quite an amount of pretty slopes and hills. Now I've discovered that I can use text flattens up to seconds in the scenery.cfg , so this can cure (but how weary) most of the bumps.Thanks again,Giorgio

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this