Pablo's Mission Planning Website

Home Links Tips Downloads MPSSF Tips

Greetings Mission Planners,

About half of you didn't receive the image last week of Shea Stadium with the DTED Level 2 Grid.  Queertrons.  To see the image (or see any recent tip) you can go to http://www.mission-planning.com.  Last week's tip is here.  

Hopefully you've updated to the most recent DAFIF that came into effect last week.  If not you should download the data from NGA's DAFIF Page.    No June EChum yet, but I'm sure it'll be up in the next few days.  

A TaskView 3.3.1 problem was found during test that effects the older version you've got.  If you:

1. Load an ATO and ACO
2. Reload the ACO, or load an ACO Change 
3. Then create a PFPS Route from an ATO mission 

you can end up with an ACO assigned altitude for a segment instead of the altitude assigned in the ATO mission.  If you don't create routes from TaskView then this doesn't effect you.  If you create a route check each segment's altitude against the ATO's assignment.  You don't have to print out the ATO, just check against the text in TaskView (only the TaskView created route altitude that's in error).  For more details see the MPSSF's announcement.

Target Acquisition Weapons Software (TAWS) 3.2 for PFPS 3.2 is available to AF, ANG, AFRC and associated contractor types from the MPSSF.  ACC/DOW will push TAWS to their users but others will have to order via the CPIN process.  For more details and the CPIN go to the MPSSF's announcement

Mission Planning Tip: Digital Terrain Elevation Data Part 2

Over the years DTED has been produced several different ways.  The earliest DTED was "Cartographic" or Carto DTED.  Cartographic DTED is derived from existing charts.  Imagine taking a chart and drawing dots to mark each DTED post.  Now look at each dot, determine it's elevation and enter it in a computer.  Now imagine doing this for eight hours, day after day.  Is that dot closer to the 2100ft contour or the 2200ft contour?  Not surprisingly Carto DTED has a well deserved reputation for varying quality. 

Next NGA began producing photogrammetric DTED using satellite imagery.  If you've got images taken from two different orbits (and you know the satellite location) you can calculate the elevation by identify the same point in both images then solving the simultaneous equations.  Much of the process can be automated, but you need the satellite imagery.  Areas that are perennially cloud covered are difficult to image and therefore difficult to produce photogrammetric DTED for.

NGA also could take existing Digital Elevation Models (DEM's) produced by other agencies and convert them into MILSPEC DTED data.  You can convert data at a very low cost, but are at the mercy of the quality of the data producer.

Finally NGA got fed up with all of the above.  The only surefire way to generate a uniform DTED dataset  is to use the same sensor and same processing and that meant using a sensor in space.  In 2000 NGA and NASA flew the Shuttle Radar Topography Mission (SRTM) to gather elevation data on a worldwide scale.  More on SRTM data to follow.

DTED's accuracy varies, but the DTED Specification sets minimum performance levels.  To meet spec a DTED Level 1 cell should have less than 90% confidence of a linear error of less than 30 Meters and a horizontal error of less than 50 Meters.  DTED Level 2 is more restrictive with a 90% confidence of a linear error of less than 18 Meters and a horizontal error of less than 23 Meters.  Why a looser specification for Level 1 DTED?  My suspicion is that much of the older Level 1 DTED wouldn't meet the increased accuracy of the newer Level 2 DTED.  There is almost no Carto Level 2 DTED, and Carto DTED is the most prone to error due to the source data and production method.

What does a "90% confidence" mean?  Essentially nine of ten error values should be less than that value, but there is a potential for the tenth post to have an error greater than that number.  While it would be nice to say "no post in this cell has an error more than xx meters, the reality is that error estimates are just that - estimates.  Without a team of surveyors it's impossible to say that no posts have a more extreme error.  Fortunately there is another part of the DTED spec that provides a bit more protection.  There has to be a 90% confidence of less than a 15 meter relative error between two adjacent posts in mountainous terrain.  High terrain covers multiple DTED posts so it should be difficult for a mountain to hide.

When we think of DTED we think of words like "vertical" and "elevation".  Why does DTED have a horizontal error estimate?  In reality DTED is a model of the earth's surface made up of points that exist in three dimensions, Latitude, Longitude and Elevation.  You can have a highly accurate post elevation but if the point where the elevation was measured wasn't at the actual Lat/Long you'll have a different kind of error.  In aviation we don't typically care about errors like this (who cares if the mountain is displaced 100 Meters) but this kind of error becomes more and more important the closer you operate to the earth's surface and the higher the resolution of data that you use.

More processing goes into the data before it becomes DTED.  Bodies of water are flattened so they all have the same elevation.  This makes it easier for software to identify lakes and color them appropriately, or just help make it clear what's water and what's land.  All bodies of water with an axis greater than 1,200 Meters for Level 1 DTED or greater than 600 Meters for Level 2 DTED must be flattened.  The USGS chart below shows Morgan Lake in New Mexico:

This is the SRTM DTED covering the same area:

All oceans and sea elevations are set to zero and in FalconView the area is colored blue.  All areas that are declared "land" have their elevation set to at least one meter (the minimum elevation DTED supports).  Islands with an axis greater than 600 Meters for DTED Level 1 or 300 Meters for DTED Level 2 must appear in DTED.  Smaller islands with a vertical relief of at least 15 Meters must also be included.  DTED supports negative as well as positive elevations so it works in Death Valley.

Of course the above specifications are minimums.  Nothing prevents a smaller island from being depicted or a smaller lake from being flattened.  Since the SRTM Level 1 DTED will be downsampled from the Level 2 Data any processing done on the original will be apparent on the lower resolution dataset as well.

A great feature in FalconView is your ability to right click on a map and select "Get map info..."  - a feature that will be much improved in FalconView 4.0.  Really I should say you can right click on most maps and get info because unfortunately there's no way to see the DTED's map info.  DTED comes with a substantial amount of metadata and the curious and brave can view it by opening any DTED cell with a normal text editor like Notepad.  In the case of SRTM cell N36.DT2 in the W109 directory you'll see:

UHL11090000W0360000N001000100005U G19 062 360136010 DSIUOOPUBLIC SALE/NO RESTRICTION DTED2G19 062 99A000000000000USCNIMA PRF89020B000005E96WGS84SRTM 0002 360000.0N1090000.0W360000N1090000W370000N1090000W370000N1080000W360000N1080000W0000000.0001000103601360199 ACC00130005NA 0005 X

I've highlighted a few pieces of information.  First you'll see the cell's place in the world, i.e. "109 00 00W 36 00 00N".  While it's the opposite of the way I'd write it, you can easily see what cell you're looking at.  This can be important if you've got DTED file but don't know what folder (the Longitude value) it goes in.  0005 means the 90% CEP vertical error is 5 METERS - SRTM data can be highly accurate.  The U means the data is unclassified and the distribution instructions show it's available for public sale w/o restriction.  99 means that 99% of the data posts have values and 1% are voids, i.e. don't have any value - more on this later.  The zeros are the maintenance date, merge date and the type maintenance performed. Since this is the first Level 2 cell over this area the values are all zeros.  PRF89020B is our old friend the DTED specification.  E96 means the vertical datum is EGM 96 and WGS84 means the horizontal datum is WGS 84.  The data source is SRTM and the source date is 0002, or February 2000 when the SRTM mission was launched to commemorate MPUC2K or perhaps vice versa.

The image below shows the Northwest quadrant of that DTED cell with some VMap Level 0 data to help provide some SA:

Next time I'll talk more about the SRTM mission and some of it's unique characteristics.  Thanks to Festus who taught me (much to my amazement) how to view the DTED Metadata.

Paul