Unfamiliar Ground

After eight years of war, the United States and its allies still lack the data necessary to navigate Afghanistan's unforgiving terrain.

BY CHRISTOPHER K. TUCKER | DECEMBER 16, 2009

Journalists and commentators covering Afghanistan and Pakistan have addressed everything from geopolitics to tribal dynamics, reconstruction aid, and, of course, terrorism. Even the casual observer has probably read stories that have addressed U.S. intelligence, surveillance, and reconnaissance efforts, most notably the use of Hellfire missile-armed Predator and Reaper drones. In covering these topics, pundits often pause to cite the area's difficult physical terrain, often mentioning it as a factor in the region's long history of repelling foreign armies. However, none of these analysts has yet written about the sad state of the United States' basic terrain data (specifically the Digital Terrain Elevation Data -- DTED) over this region, which has emerged as an obstacle in conducting operations over such complex terrain.

This is not a national security secret. It's just knowledge that simply doesn't often make it out of the jargon-laden community of geodetic science, remote sensing, and military terrain analysis. The general news consumer (and, in my experience, many in our national security leadership) assume that, in this age of Google Earth, U.S. forces have all the information they need. Unfortunately, these assumptions are both limiting U.S. mission effectiveness and getting people killed.

So, here's a short primer on this arcane, but vitally important subject.

Terrain data is collected by taking measurements at a certain "post spacing," or the distance between each data point, which varies depending on the sensor used and the altitude from which it is collected. For instance, in 2000, NASA and the National Imagery and Mapping Agency conducted the Shuttle Radar Topography Mission (SRTM), which collected 30-meter post spacing terrain data for everything that falls between northern Russia and southern Argentina. The measurements were taken from Space Shuttle Endeavour, which was orbiting the Earth at an altitude of 233 kilometers. This means that, every 30 meters or so, the U.S. defense and intelligence community has an elevation reading of what the terrain looked like in 2000. What the terrain looks like between these 30-meter intervals is anyone's guess.

Well, actually, in the case of the Afghanistan-Pakistan border region, local fighters know the answer intimately. U.S. and Pakistani military forces, however, do not, and so for the unsuspecting soldier, those 30 meters could hold a ravine, a structural impediment, a rugged natural obstacle, impassable vegetation, or some other kind of deadly cul-de-sac within a canyon complex. But U.S. soldiers usually don't know what to expect until it's too late.

There has been some higher-resolution terrain data collected, but very little at the 1-meter post spacing typically considered sufficient for military operations over urban and complex terrain. Furthermore, high-resolution color imagery is rarely collected simultaneously with terrain data, meaning that operators and analysts lack detailed visual knowledge of any "bump" that they observe.

The war effort also desperately needs for such terrain data collections to be unclassified, so they can be revealed to coalition partners and host-nation militaries. Anyone who has worked in this field understands how rare it is for the intelligence community to collect data it is willing to share. But the military terrain analysis community has done so for several years, despite working on a shoestring budget. What's more, the terrain analysis community has made this data broadly available to war fighters, as well as the development community

The United States has now been in Afghanistan for eight years. Despite spending hundreds of billions of dollars in the country, the United States still does not possess terrain data of the same quality as that routinely collected by local, state, and federal agencies within the United States. This is not for lack of commercially available terrain sensors or of contractors willing to fly in Afghanistan or Pakistan. It is because policymakers aren't aware of this critical need. The collection of high-resolution terrain data in Afghanistan is within the power of top U.S. and NATO commanders -- but it will require collaboration, policy reform, technical cooperation, and a substantial investment of resources.

As we continue to wrestle with Sir Mortimer Durand's dubious legacy -- the "Durand Line" that divides Afghanistan and Pakistan -- perhaps we can agree that the United States and its allies cannot succeed in their mission if they literally do not know where they are heading.

DAVID FURST/AFP/Getty Images

 

Christopher K. Tucker is a board member of the U.S. Geospatial Intelligence Foundation.

Facebook|Twitter|Reddit

SOLARGLOBALGREEN

6:22 PM ET

December 17, 2009

It's a problem of "too many

It's a problem of "too many chiefs and not enough indians". Politicians are so far removed from reality these days that information is stifled. It is shameful that at this stage of the game our troops do not have sufficient maps for the terrain they are fighting in.

Solar Global Green

 

MILAN

6:27 AM ET

December 31, 2009

chiefs and not enough

chiefs and not enough indians". Politicians are so far Must High School removed from high school diploma reality these days that Must High School information is stifled. It is shameful that at online high school this stage of the game our troops do not have sufficient maps for the terrain they are fighting in. online homeschooling - Must High School

 

MMELEN

3:47 AM ET

January 15, 2010

War also desperately in

War also desperately in collections of data is unclassified, so they may be disclosed to coalition partners and the armed forces of the host country. Anyone who has worked in this field also, how rare it is to collect intelligence information, it is ready to share. I saw this when I was selling mens swimwear in the area. The military community terrain modeling was made for several years while working on a tight budget. In addition, the analysis of the country has made information available to the public warriors, and the development community

 

TURNINGWINDS

5:47 AM ET

December 18, 2009

Peace not war

Why the government doesn't consider on having peace on Afghanistan. It's been 8 years and no improvement regarding peace in Afghanistan. Spending millions of dollars for combat why they don't want to spend it for helping Afghan people.

_______________
Turning Winds

 

GHORSTKJ

4:17 PM ET

December 18, 2009

The maps are available but...

This is an excellent article and I hope that it makes more folks take interest in this "arcane, but vitally important subject." The sad truth is that the maps and dizzying array of other "geospatial products" are available but unfortunately they often get caught up in the "intelligence network" and lesser quality, more generic, and outdated versions are left for the troops on the ground to choose to use (or ignore). The distribution of a Rand McNally-like Atlas of Afghanistan has proven to be a logistical challenge and the dissemination of a common digital “common operating picture (COP)” in a Garmin-like GPS device has proven to be a technical nightmare. I’ve found that smaller, lesser funded organizations like the Army Geospatial Center (AGC) or the Marine Corps Intelligence Activity (MCIA) have proven to be more responsive than the larger National Geospatial-Intelligence Agency (NGA), although even they have made great strides in the last several years in improving their direct support to both land-based services as well. All of them stand ready to support if soldiers and leaders know the right questions to ask and are persistent in getting the resources they need. Unfortunately, most either don’t have the time and interest or feel that someone down the line will simply find a way to make things happen.

With all of this said, great advancements have been made over the last decade to provide both maps and imagery in multiple data formats and scales (for those that take the time to find out and realize that some things can’t be provided overnight). Google has made providing these services online seem almost too easy to do when they bought a company named Keyhole along with its government sponsored software program with the same name and rebranded it as “Google Earth.” However, even more important and often a greater challenge in this type of counterinsurgency struggle is trying to understand the "human landscape" in which the population is the "key terrain." The mentioned unfamiliar and unforgiving? Try extracting vital nuances of information in varying languages and figuring out a way to capture that for others to learn from and understand the key "fault lines" amongst the people.

While we need to get the available hard/soft copy topographic information out to the troops in a more effective manner, we also need to be focusing on collecting and comprehending the vital layer of human activity on the surface of the Earth and how the people of Afghanistan interact with the land in which they live. I will look forward to reading more discussion here for those that are interested and want to figure out ways to improve the situation going forward rather than just look back at ways to point blame and accuse failure without really knowing what the ground truth is.

 

JLT

8:57 PM ET

December 18, 2009

Loss of Patience! Nothing will happen overnight!

Impatience is the greatest killer to operations. So is carefully thought problem analysis... The above diatribe clearly demonstrates a characteristic of the operational demand being before the actual quality of a requirement. In 2000, NASA conducted a detailed radar scan of over 80% of the Earth, we now have SRTM. This data set is the most positional accurate data set that we have, tied to one Reference System. If at any given time a satellite image positional accuracy is no better than +/- 250m in the horizontal and depending on the vertical exageration (steeper terrain), upwards to +/- 100m in the vertical... You might say what? Well, Google Earth may present visually accurate imagery, but has little if no positional accuracy.

So what can we do? We as a fighting force are trained to fight on 1:50,000 Topographic Line Maps. This is a tried and true positional accurate map. The question you must ask is, are their enough 1:50k TLM maps to be working on? If not then argue that point. Yet, often the 1:100k map is the only available coverage. Once we demand greater and greater map coverage than what the system can provide, we introduce artificial requirements.

Huh? To make more accurate digital map data either you have to land survey nearly every 1-5m locations? Or you have to introduce serious artificial gradients.

Currently the Global 1:50k TLM coverage is maintained by the Multinational Geospatial Coproduction Program with Germany and Norway taking the lead for Afghanistan 1:50k TLM, production, Controlled Imagery Base 1m, and DTED cell seamless production. They will complete production by end of 2010. This is a very labor intensive process which requires exact production specifications.

The US National Geospatial-Intelligence Agency went off the global 1:50k TLM standard a few years ago, because the MGCP program by international partners would provide better seamless coverage. Plus the Key Performance Parameters had been tied artificial national requirements.

Any better positionally accurate products is lunacy, or shows a near lack of appreciation of the actual requirement. What we must do is train our soldiers to work on 1:50K TLM maps as we did beforehand, and fought many, many wars on. And take the artificiality of Digital Maps and their serious misunderstandings. When I want a quick digital image, I'll overlay an orthorectified satellite or airborne image. Those that understand, know that the orthorectification process is not quick. We could probably at best provide only chips of imagery with +/- 100m accuracy.

But we must know and accept that positional accuracy is paramount. Without that, our best Blue Force Trackers will be absolutely useless!

Huh? If in doubt leave the geospatial-intelligence stewardship to those who know what they are dealing with, and how to best communicate to the Joint Force Commander. If in further doubt contact your nearest Geospatial-Intelligence Cell officer and have them push for the most current 1:50k, CIB1 and DTED data sets.