Mapping Alaska—America’s Last Frontier
A multi-year initiative underway to digitally map Alaska will increase public safety and community preparedness throughout the nation's largest state.
By Col. David F. Maune, Ph.D., CP, CFM, PSM, PS, GS, SP, USA (Ret.)
With insufficient maps and no statewide digital orthophotos, Alaskans have long been challenged to manage the state’s natural resources and infrastructure. Military and civilian pilots also face perilous conditions in navigating through mountainous areas.
Now, a statewide digital mapping initiative is using interferometric synthetic aperture radar (IFSAR) to create new maps and enhance safety and preparedness throughout America’s Last Frontier.
SCOPE OF THE PROBLEM
Alaska has historically proven so difficult to map that some experts believe the Moon and Mars have been mapped more accurately. Digital orthophotos, common elsewhere, have never been produced statewide because of the persistent cloud cover and the inaccuracy of the Digital Terrain Model (DTM) required for orthorectification. The U.S. Geological Survey’s (USGS) 1:63,360-scale topographic quadrangles of Alaska have long been obsolete and were never produced to National Map Accuracy Standards. When the maps were produced in the 1950s from stereo photogrammetry, survey control was essentially non-existent. Digital cameras, GPS and photogrammetric block triangulation had yet to be invented. Photogrammetrists had to bridge one stereo model at a time for each flight strip, then mosaic the individual strips together. This is the equivalent of having survey control in Virginia, flying to Missouri to take aerial photos without airborne GPS or ground control, returning to Virginia to produce maps with analog technology, and then finding that the Gateway Arch in St. Louis is mapped a mile in the wrong location.
Alaska has fewer total miles of paved roads than many counties in the continental United States. As a result, residents depend upon aviation and thousands of small airstrips to access remote villages. However, flying in small aircraft across the state can be life-threatening due to the poor quality of existing maps. Pilots must often navigate using instrument flight rules because darkness, clouds, or fog prevent the use of visual flight rules. Some USGS topographic maps depict mountains more than a mile from their actual locations. This has resulted in a consistently high rate of Controlled Flight into Terrain accidents.
Outdated or nonexistent maps have challenged the operations of government agencies as they manage a state spanning more than 600,000-mi² and with vast natural resources. The mapping insufficiencies have limited industry and economic development initiatives. Climate change researchers also have struggled with inadequate map data as they address the impacts of global warming, shoreline erosion, earthquakes, floods, and changes in permafrost conditions.
SEEING THROUGH CLOUDS
In 2008, Dewberry began consulting with state representatives seeking solutions to the challenge of creating new maps for Alaska, given the historic lack of success in producing cloud-free digital orthophotos. The firm advised that the state needed to determine its requirement for Digital Elevation Models (DEMs) and that accurate models needed to be produced first, before any imagery could be orthorectified. Dewberry then outlined a plan—in a document known as the “Alaska DEM Whitepaper”—that recommended using IFSAR to address the statewide need for DEMs with 20-ft contour accuracy.
Alaska has fewer total miles of paved roads than many counties in the continental United States. As a result, residents depend upon aviation and thousands of small airstrips to access remote villages. However, flying in small aircraft across the state can be life-threatening due to the poor quality of existing maps.
IFSAR is the only remote sensing technology that can map through clouds, fog, smoke and haze. IFSAR directly produces Ortho-rectified Radar Images (ORIs) that clearly delineate water boundaries, and Digital Surface Models (DSMs) of top reflective surfaces.
Bare-earth DTMs are required for orthorectification of aerial or satellite imagery on which features are displaced by terrain relief (topographic variations). IFSAR DTMs can be derived from IFSAR DSMs using manual editing techniques and algorithms that model the height and density of vegetation—or they can be derived by employing a separate IFSAR wavelength that penetrates vegetation.
While high-accuracy DTMs can be produced with 1-ft contour accuracy using photogrammetry or LiDAR, neither of these technologies would work statewide in Alaska due to the weather limitations. Other technologies also would be unaffordable even if the skies were totally clear. IFSAR is much more affordable—though it, too, has limitations. It is a mid-accuracy technology that can produce DTMs with 20-ft contour accuracy, yet only for slopes up to 20⁰ and where vegetation is not too dense to see the ground from side-looking IFSAR.
In addition to penetrating dry snow, IFSAR’s phenomenology includes "layover," which is an error that can result when the slope of the terrain is greater than the angle the incident radiation makes with respect to vertical. Layover discrepancies can smooth over the top of steep, narrow mountaintops such as Denali (formerly Mount McKinley), which has a slope of nearly 80⁰ and a ridge only a few meters wide. Photogrammetry, LiDAR and IFSAR are all limited in mapping accurate elevations of snow-capped mountains where only GPS can survey elevations reliably.
Through a Geospatial Products and Services Contract, USGS contracted with Dewberry to produce IFSAR ORIs, DSMs and DTMs for Alaska’s Statewide Digital Mapping Initiative. Dewberry subcontracted with Intermap Technologies and Fugro EarthData—each operating in different terrain types—for IFSAR data acquisition and the production of high-resolution ORIs as well as hydro-enforced DSMs and DTMs with 5-m elevation post spacing. With different IFSAR systems, Intermap used X-band IFSAR to map the larger areas while Fugro used both X-band and P-band IFSAR to map the most difficult terrain.
Dewberry then completed a pilot project that demonstrated how to map through clouds; pan-sharpen satellite imagery with 5-m pixels to produce orthophotos with sub-meter pixel resolution; and produce National Hydrography Dataset and transportation feature updates that correctly fit the DTMs and digital orthophotos. Dewberry also demonstrated how to drape satellite imagery over the DTM to view the terrain from any perspective. By the end of 2015 the team had acquired IFSAR data for more than 60 percent of Alaska, including such areas as the Copper River region and Glacier Bay National Park and Preserve. The IFSAR data acquisition is now part of the national 3D Elevation Program that is being coordinated by USGS. Subject to the availability of funding, the mapping work is expected to be completed within the next three years.
To ensure accuracy requirements are met, quality control checkpoints were subcontracted and surveyed by JOA Surveys, consistent with guidelines from the American Society for Photogrammetry and Remote Sensing. All IFSAR datasets passed accuracy tests, including checkpoints within Denali National Park.
DETERMINING DENALI’S ELEVATION
While IFSAR has proven to be the most logical choice for the Statewide Digital Mapping Initiative, the technology may have unwittingly smoothed the summit of Denali, the highest peak in North America, during image capturing.
In 2010, IFSAR yielded an elevation of 20,237-ft, potentially a significant change from the 20,320-ft elevation established by Bradford Washburn’s landmark trigonometric leveling survey done in 1953. However, Dewberry specialists recognized that IFSAR’s limitations on sharp peaks necessitated further investigation to determine Denali’s true elevation. This would require direct examination.
Anchorage, Alaska-based E-Terra LLC took a fly-by photograph of the Denali summit and compared it with a satellite image draped over Fugro’s IFSAR DTM. The two images looked remarkably similar—but it was impossible to tell if the summit ridge was shortened by IFSAR layover or snow penetration.
Following the 2015 Alaska Surveying and Mapping Conference, which was held last February in Anchorage, Dewberry proposed a Denali GPS survey expedition led by a crew from CompassData and the University of Alaska Fairbanks (UAF). USGS and the National Geodetic Survey recognized the importance of an accurate elevation for Denali and supported the project by providing funding. UAF also provided in-kind support.
In June 2015, the CompassData/UAF team reached the summit of Denali and surveyed the elevation with GPS for nearly 18 hours. The crew computed a new elevation of 20,310-ft for the top of the snow; and USGS announced the revised elevation on Sept. 2, 2015. Still, the depth of the ice on the Denali summit remains unknown and there were likely some variations between 2010, when the IFSAR was acquired, and 2015, when the GPS survey was performed. Terrain averaging and layover are unlikely to explain completely the apparent height discrepancy between the 2015 GPS survey elevation and the 2010 IFSAR result. Multiple IFSAR estimates of the summit were made from widely varying azimuth angles and these independent measurements are in excellent agreement.
All the measurements were made at fairly large look angles, which was necessitated by Denali’s relative height compared to the IFSAR acquisition height of approximately 39,000-ft. At large look-angles, IFSAR lever-arm effects are increased. But the diversity of the look-angles among the multiple measurements that were acquired do not show the presence of a systematic look-angle error. The exact cause for the discrepancy likely will remain an open issue for future surveys and continued debate.
SAFETY AND PREPAREDNESS
All sensors have limitations. Neither LiDAR nor photogrammetry can map through clouds and fog. Both also are much more expensive than IFSAR. An unlimited budget and a combination of LiDAR and photogrammetry would take decades longer to produce and likely end up in a patchwork of quality. With IFSAR, Alaska will have consistent quality data for the entire state delivered in a timely manner; and IFSAR is ideal for the mapping of hydrographic features important in Alaska.
It appears that IFSAR remains the best technology for mid-accuracy mapping of America’s Last Frontier, with outstanding progress made to date. The new maps have already proven beneficial to pilots, first responders and government agencies charged with enhancing community preparedness and public safety throughout the nation’s largest state.
Knowing all other states have been mapped and remapped many times, Sen. Lisa Murkowski (R-Alaska) may have summarized it best at a ceremony in August 2015 to commemorate exceeding the halfway mark for elevation acquisition over Alaska: “What other state,” she said, “would have 250 people assemble to celebrate the fact that their state is now over 50 percent mapped?”