Viewshed Creation: From Digital 
Terrain Model to Digital Surface 
Terrain Model to Digital Surface
           Model
     ...
Viewshed Creation: From Digital 
Terrain Model to Digital Surface 
Terrain Model to Digital Surface
           Model
     ...
Viewshed Creation: From Digital 
Terrain Model to Digital Surface 
Terrain Model to Digital Surface
           Model
     ...
History of Viewsheds
           History of Viewsheds
• What is a 
  viewshed
• Historical uses
  – Military
    Military 
History of Viewsheds
           History of Viewsheds
• What is a 
  viewshed
• Historical uses
  – Military
    Military 
Common uses
                Common uses
• Modern uses of viewsheds
  – Homeland security
          l d
  – Military
  – Co...
Problem
• Intervisibility studies are limited by the dearth 
       g                     (      )
  of Digital Surface Mo...
Project Goals
               Project Goals
• Overcome limitations of DTM based viewshed
  Overcome limitations of DTM base...
Previous Work
              Previous Work
• Swanson (2003) – Different GIS applications
  Swanson (2003)  Different GIS ap...
Previous Work
              Previous Work
• Swanson (2003) – Different GIS applications
  Swanson (2003)  Different GIS ap...
Previous Work
               Previous Work
• Guth (2009) – Proposed feasibility of using
       (2009)  Proposed feasibili...
Previous Work
              Previous Work
• Fisher (1991) Beaulieu (2007) Anile Furno
  Fisher (1991), Beaulieu (2007), An...
Previous Work
              Previous Work
• Fisher (1991) Beaulieu (2007) Anile Furno
  Fisher (1991), Beaulieu (2007), An...
Scope of problem
                      p     p
• Viewshed at large and 
  small scale
   – L l f d il f diff
     Level of...
Scope of problem
                      p     p
• Viewshed at large and 
  small scale
   – L l f d il f diff
     Level of...
DEM Datasources
             DEM Datasources
• NED – DTM‐Free, quality differs based on how 
                 ee, qua ty d...
DTM 
              DTM         DSM
• ArcGIS Model Builder
         Model Builder 
• USGS National Elevation Dataset (NED) ...
Methodology
DSM Data NLCD 2001
         DSM Data‐NLCD 2001
• 29 landcover categories
  29 landcover categories 
• Resolution of 302 me...
DSM Data Land Fire
            DSM Data‐Land Fire
• Wildland fire 
     da d e
  management
• Based off NLCD
• No man‐made...
DSM Data NBCD
            DSM Data‐NBCD
• Based on SRTM NED Landfire and NLCD‐2001
  Based on SRTM, NED, Landfire and NLCD...
DSM Data IFSAR
            DSM Data‐IFSAR
• Resolution of 52 meters
  Resolution of 5
• DTM
DSM Data LIDAR
            DSM Data‐LIDAR
• 0 7 meter ground sample distance
  0.7 meter ground sample distance
• Point cl...
Software Used
               Software Used
• Quick Terrain Modeler
  Quick Terrain Modeler 
• Arcview
  – S ti l A l t
   ...
In‐Situ observation
• Digitize using latest NAIP as basemap
     Convert to raster  used to 
  compare with finished views...
Locations for studies
          Locations for studies
• Rio Grande Valley Texas
  Rio Grande Valley, Texas
  – mean slope ...
Locations for studies
          Locations for studies
• Aroostook County Maine
  Aroostook County, Maine 
  – mean slope a...
Limitations
• The 302 meters resolution of NLCD Landfire
  The 30 meters resolution of NLCD, Landfire, 
  and NBCD limits ...
Preliminary Findings
            Preliminary Findings
• USGS DTM vs in‐situ
  USGS DTM vs. in situ    • Lidar DSM vs in si...
Tasks to do
                   Tasks to do
•   Obtain all the data
    Obtain all the data
•   Verify Projection, reprojec...
Predictions/Conclusions
• 30 meter data (NLCD NBCD Landfire) will yield
  30 meter data (NLCD, NBCD, Landfire) will yield ...
Questions
Possible Future considerations
      Possible Future considerations
•   How does the method used to collect the DEM affect...
Upcoming SlideShare
Loading in …5
×

Viewshed Creation: From Digital Terrain Model to Digital Surface Model

3,713 views

Published on

Presentation for school of my proposed Master's project.

Published in: Education, Technology, Sports
0 Comments
3 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
3,713
On SlideShare
0
From Embeds
0
Number of Embeds
21
Actions
Shares
0
Downloads
0
Comments
0
Likes
3
Embeds 0
No embeds

No notes for slide

Viewshed Creation: From Digital Terrain Model to Digital Surface Model

  1. 1. Viewshed Creation: From Digital  Terrain Model to Digital Surface  Terrain Model to Digital Surface Model Edward Ashton Edward Ashton
  2. 2. Viewshed Creation: From Digital  Terrain Model to Digital Surface  Terrain Model to Digital Surface Model Edward Ashton Edward Ashton
  3. 3. Viewshed Creation: From Digital  Terrain Model to Digital Surface  Terrain Model to Digital Surface Model Edward Ashton Edward Ashton
  4. 4. History of Viewsheds History of Viewsheds • What is a  viewshed • Historical uses – Military Military 
  5. 5. History of Viewsheds History of Viewsheds • What is a  viewshed • Historical uses – Military Military 
  6. 6. Common uses Common uses • Modern uses of viewsheds – Homeland security l d – Military – Commercial – Land Use Planning
  7. 7. Problem • Intervisibility studies are limited by the dearth  g ( ) of Digital Surface Model (DSM) elevation data.   Most publically available DEMs are Digital  ( Terrain Models (DTM).  ) DTM DSM
  8. 8. Project Goals Project Goals • Overcome limitations of DTM based viewshed Overcome limitations of DTM based viewshed • Create DSM from DTM using ancillary data  sources • Measure effectiveness of process • Use publically available DEMs for DTM – To include IFSAR and LIDAR where available
  9. 9. Previous Work Previous Work • Swanson (2003) – Different GIS applications Swanson (2003)  Different GIS applications  viewshed creation compared to in‐situ  observation to verify findings observation to verify findings
  10. 10. Previous Work Previous Work • Swanson (2003) – Different GIS applications Swanson (2003)  Different GIS applications  viewshed creation compared to in‐situ  observation to verify findings observation to verify findings
  11. 11. Previous Work Previous Work • Guth (2009) – Proposed feasibility of using (2009)  Proposed feasibility of using  vegetation to increase accuracy of DTM
  12. 12. Previous Work Previous Work • Fisher (1991) Beaulieu (2007) Anile Furno Fisher (1991), Beaulieu (2007), Anile, Furno,  Gallo, Massolo (2003) ‐ Fuzzy Viewsheds
  13. 13. Previous Work Previous Work • Fisher (1991) Beaulieu (2007) Anile Furno Fisher (1991), Beaulieu (2007), Anile, Furno,  Gallo, Massolo (2003) ‐ Fuzzy Viewsheds
  14. 14. Scope of problem p p • Viewshed at large and  small scale – L l f d il f diff Level of detail for different  applications – Small scale – Earth  Curvature, projection C t j ti – Large scale ‐ Detail – How much detail does  scale need?  Camera  l d? placement vs radar site 
  15. 15. Scope of problem p p • Viewshed at large and  small scale – L l f d il f diff Level of detail for different  applications – Small scale – Earth  Curvature, projection C t j ti – Large scale ‐ Detail – How much detail does  scale need?  Camera  l d? placement vs radar site 
  16. 16. DEM Datasources DEM Datasources • NED – DTM‐Free, quality differs based on how  ee, qua ty d e s based o o collected (1” ,1/3” and 1/9”)  • SRTM – DSM‐Free, covers world, 1” and 3” • LIDAR – DTM & DSM‐Free (limited availability) – Very expensive, detailed, artifacts, large amount  of data • IFSAR – DSM‐Expensive, artifacts in DSM • Limitations and pros of each type
  17. 17. DTM  DTM DSM • ArcGIS Model Builder Model Builder  • USGS National Elevation Dataset (NED) +  vegetation information and man‐made  vegetation information and man made structures  • N i National Land Cover Database, Landfire, and  lL dC D b L dfi d National Biomass and Carbon Dataset.   • Compare with in‐situ observations, LIDAR and  IFSAR DSM viewsheds. 
  18. 18. Methodology
  19. 19. DSM Data NLCD 2001 DSM Data‐NLCD 2001 • 29 landcover categories 29 landcover categories  • Resolution of 302 meters
  20. 20. DSM Data Land Fire DSM Data‐Land Fire • Wildland fire  da d e management • Based off NLCD • No man‐made features • Resolution of 302 meters • Vegetation Heights in  bins, not absolute  values
  21. 21. DSM Data NBCD DSM Data‐NBCD • Based on SRTM NED Landfire and NLCD‐2001 Based on SRTM, NED, Landfire and NLCD 2001  Resolution of 302 meters • Estimate weighted canopy height biomass Estimate weighted canopy height, biomass,  and carbon stock
  22. 22. DSM Data IFSAR DSM Data‐IFSAR • Resolution of 52 meters Resolution of 5 • DTM
  23. 23. DSM Data LIDAR DSM Data‐LIDAR • 0 7 meter ground sample distance 0.7 meter ground sample distance • Point cloud DSM&DTM
  24. 24. Software Used Software Used • Quick Terrain Modeler Quick Terrain Modeler  • Arcview – S ti l A l t Spatial Analyst – Military Analyst
  25. 25. In‐Situ observation • Digitize using latest NAIP as basemap Convert to raster  used to  compare with finished viewsheds i h fi i h d i h d Rules ‐possible locations  of vehicles or  humans   ‐Vegetation = fuzzy
  26. 26. Locations for studies Locations for studies • Rio Grande Valley Texas Rio Grande Valley, Texas – mean slope under 1% – 130 square miles 130 square miles  – 68% cultivated crops and  pasture  pasture – 18% developed space  – 14% undeveloped area 14% undeveloped area. 
  27. 27. Locations for studies Locations for studies • Aroostook County Maine Aroostook County, Maine  – mean slope about 10% – 120 square miles 120 square miles – 44% cultivated crops and  pasture – 6% developed area  – 43% forest 43% forest  – 7% undeveloped area
  28. 28. Limitations • The 302 meters resolution of NLCD Landfire The 30 meters resolution of NLCD, Landfire,  and NBCD limits the DSMs that can be created • Data was collected circa 2000 Data was collected circa 2000 • LIDAR data collected 2005‐2007 • IFSAR data collected ?? – 5 meter • NAIP imagery collected 2007 for Maine sites,  g y , 2008 for Texas Sites
  29. 29. Preliminary Findings Preliminary Findings • USGS DTM vs in‐situ USGS DTM vs. in situ  • Lidar DSM vs in situ DSM vs in‐situ  observation observation
  30. 30. Tasks to do Tasks to do • Obtain all the data Obtain all the data • Verify Projection, reproject data • Go through methodology steps G h h h d l • Compare results • Repeat process for Maine sites
  31. 31. Predictions/Conclusions • 30 meter data (NLCD NBCD Landfire) will yield 30 meter data (NLCD, NBCD, Landfire) will yield  results that while easy to create will be  unsuitable for large scale viewshed in some  g areas. • IFSAR better results but still issues with artifacting IFSAR better results but still issues with artifacting • LIDAR best results but massive amounts of  storage space and very expensive to collect and  storage space and very expensive to collect and maintain. • Geographic patterns of failure and success Geographic patterns of failure and success
  32. 32. Questions
  33. 33. Possible Future considerations Possible Future considerations • How does the method used to collect the DEM affect the viewshed? GPM II,  manual profiling, DLG2DEM, DCASS and LT4X manual profiling DLG2DEM DCASS and LT4X • How does environment affect the results?  Study done in limited terrains  • How can a DTM be turned into a DSM with minimal effort?  Explore using land use  data, automated photogrammetric techniques, object extraction from other  imagery types.  imagery types • Testing in areas that does not have excessive vegetation (ie trees) and no man  made structures.  Does this yield similar results between a manual viewshed and  using a DEM? • The manual viewshed made assumptions about what was visible and what was  The manual viewshed made assumptions about what was visible and what was not.  Those areas that a human or vehicle could be located and were visible to the  analyst were marked as visible and everywhere else was non visible.  Thus the tops  of trees, the facades or buildings were marked non‐visible even though the analyst  could see them.  How does this affect the results?  How can the results be  normalized to take this into consideration? • Different GIS applications compute viewsheds differently.  This study explored  ArcGIS only, how do other applications compare.

×