3D vs. Conventional Volcanic Hazard Maps

Charles Preppernau
Oregon State University
#nacis2015

Transcript

1. 3D vs. Conventional Volcanic Hazard
   Maps: A User Study at Mount Hood
   Charles Preppernau
   Bernhard Jenny
   

   View Slide

2. Philip Maise, 2008
   Chris Newhall, 1984 Chris Newhall, 1991
   

   View Slide

3. “...so the lava flow will go down this river...”
   Research Problem
   

   View Slide

4. Research Question
   How can we use cartographic
   methods to improve volcanic hazard
   maps?
   

   View Slide

5. Modified from Scott et al. 1997
   Research Problem
   

   View Slide

6. Revised Research Questions
   Do 3D maps allow users to more accurately
   interpret terrain?
   Do isochrones give a better understanding
   of the speed of lahar hazards?
   Do these two cartographic methods improve
   user’s choice of evacuation routes?
   

   View Slide

7. Study Design
   

   View Slide

8. Study Design
   

   View Slide

9. 2D Maps
   

   View Slide

10. 2D Maps
    

    View Slide

11. 3D maps
    

    View Slide

12. 3D maps
    

    View Slide

13. Hypotheses
    H1: Users prefer 3D perspective maps over 2D contour maps for terrain
    representation, and prefer isochrones over point markers for travel time
    indicators.
    H2: Users more accurately judge relative elevation and slope on 3D maps
    than on 2D maps.
    H3: Users more accurately locate themselves on 3D maps than on 2D maps.
    H4: Users more accurately interpret lahar travel time and speed with isochrone
    maps than with point marker maps.
    H5: Users choose more successful evacuation routes with 3D maps or
    isochrones than with 2D maps or point markers.
    Study Design
    

    View Slide

14. Study Design
    

    View Slide

15. 1st trip: 34 participants  pilot survey
    2nd trip: 41 participants
    3rd trip: 39 participants
    = 80 participants total for final survey
    Study Design
    

    View Slide

16. H1: Do users prefer 3D maps and isochrones?
    H2: Do users more accurately read terrain on 3D maps?
    H3: Do users more accurately locate themselves on 3D maps?
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    User Preferences
    

    View Slide

17. User Preferences
    

    View Slide

18. Map
    Type
    Terrain
    Best
    Terrain
    Worst
    Time
    Best
    Time
    Worst
    Route
    Best
    Route
    Worst
    2D markers 3 47 4 51 4 44
    3D isochrones 42 2 38 3 41 4
    3D markers 31 1 13 15 18 9
    2D isochrones 0 26 21 7 13 19
    User Preferences
    

    View Slide

19. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    H3: Do users more accurately locate themselves on 3D maps?
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    User Preferences
    

    View Slide

20. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    H3: Do users more accurately locate themselves on 3D maps?
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Terrain Interpretation
    

    View Slide

21. Terrain Interpretation
    

    View Slide

22. Grouped 3D map scores are significantly greater than
    grouped 2D map scores (p value = 0.03)
    Terrain Interpretation
    

    View Slide

23. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope
    comparisons than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Terrain Interpretation
    

    View Slide

24. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope
    comparisons than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Self-location
    

    View Slide

25. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope
    comparisons than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    No, there was no difference in location error between map types.
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Self-location
    

    View Slide

26. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps
    were always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope
    comparisons than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    No, there was no difference in location error between map types.
    H4: Do users more reliably judge lahar arrival time and speed on isochrone
    maps?
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Lahar Travel Time Interpretation
    

    View Slide

27. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps were
    always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope comparisons
    than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    No, there was no difference in location error between map types.
    H4: Do users more reliably judge lahar arrival time and speed on isochrone maps?
    No. Users of marker maps performed just as well as users of isochrone
    maps.
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Lahar Travel Time Interpretation
    

    View Slide

28. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps were
    always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope comparisons
    than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    No, there was no difference in location error between map types.
    H4: Do users more reliably judge lahar arrival time and speed on isochrone maps?
    No. Users of marker maps performed just as well as users of isochrone
    maps.
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Evacuation Routes
    

    View Slide

29. View Slide

30. View Slide

31. Grouped 3D map scores are significantly lower than grouped 2D map
    scores (p value = 0.03)
    Evacuation Routes:
    

    View Slide

32. H1: Do users prefer 3D maps and isochrones?
    Yes, 3D isochrone maps were always liked most, 2D marker maps were
    always liked least.
    H2: Do users more accurately read terrain on 3D maps?
    Yes, users of 3D maps scored higher on elevation and slope comparisons
    than users of 2D maps.
    H3: Do users more accurately locate themselves on 3D maps?
    No, there was no difference in location error between map types.
    H4: Do users more reliably judge lahar arrival time and speed on isochrone maps?
    No. Users of marker maps performed just as well as users of isochrone
    maps.
    H5: Do users choose better escape routes on 3D or isochrone maps?
    Yes, users of 3D maps chose on-foot routes with lower critical speeds
    than users of 2D maps.
    Evacuation Routes:
    

    View Slide

33. 3D maps are more effective tools than 2D maps for
    conveying the dangers of lahars.
    Future research should focus on:
    Switching to open source software
    Automating the 3D map creation process
    Optimizing and testing new versions of the map
    design
    Conclusions:
    

    View Slide

34. Further work:
    www.3dvolcano.com
    

    View Slide

35. Thank you!
    

    View Slide

36. Research Problem
    

    View Slide

37. Previous Research
    

    View Slide

38. Methods:
    Study Site
    

    View Slide

39. Methods:
    User Study
    

    View Slide

40. Results
    

    View Slide

41. Conclusions
    

    View Slide

42. Acknowledgements
    

    View Slide

43. References
    Baxter, Peter J. 2000. “Impacts of Eruptions on Human Health.” In Encyclopedia of Volcanoes, 1035–43. Orlando: Academic
    Press.
    Blong, Russell J. 1984. Volcanic Hazards. A Sourcebook on the Effects of Eruptions. North Ryde, N.S.W.: Academic Press.
    Gardner, C. A., W. E. Scott, J. J. Major, and T. C. Pierson. 2000. “Mount Hood—history and Hazards of Oregon’s Most Recently
    Active Volcano.” US Geol. Surv. Survey Fact Sheet, no. 060-00.
    Haynes, Katharine, Jenni Barclay, and Nick Pidgeon. 2007. “Volcanic Hazard Communication Using Maps: An Evaluation of
    Their Effectiveness.” Bulletin of Volcanology 70 (2): 123–38.
    Jenny, Helen, Bernhard Jenny, and Lorenz Hurni. 2010. “Interactive Design of 3D Maps with Progressive Projection.” The
    Cartographic Journal 47 (3): 211–21.
    Marzocchi, Warner, Christopher Newhall, and Gordon Woo. 2012. “The Scientific Management of Volcanic Crises.” Journal of
    Volcanology and Geothermal Research 247: 181–89.
    Nakada, Setsuna. 2000. “Hazards from Pyroclastic Flows and Surges.” In Encyclopedia of Volcanoes, 945–95. Orlando:
    Academic Press.
    Newhall, C. G. 2000. “Volcano Warnings.” In Encyclopaedia of Volcanoes, 1185–97. Orlando: Academic Press.
    Newhall, C. G., and R. S. Punongbayan. 1996. “The Narrow Margin of Successful Volcanic-Risk Mitigation.” In Monitoring and
    Mitigation of Volcano Hazards, 807–38. Springer.
    Pierson, Thomas C. 1998. “An Empirical Method for Estimating Travel Times for Wet Volcanic Mass Flows.” Bulletin of
    Volcanology 60 (2): 98–109.
    Rodolfo, K. S. 2000. “The Hazard from Lahars and Jökulhlaups.” In Encyclopedia of Volcanoes, 973–95. Orlando: Academic
    Press.
    Schilling, S. P., S. Doelger, W. E. Scott, T. C. Pierson, J. E. Costa, C. A. Gardner, J. W. Vallance, and J. J. Major. 2008. Digital
    Data for Volcano Hazards of the Mount Hood Region, Oregon.
    Schobesberger, David, and Tom Patterson. 2007. “Evaluating the Effectiveness of 2D vs. 3D Trailhead Maps.” Mountain Mapping
    and Visualisation, 201.
    Scott, W. E., T. C. Pierson, S. P. Schilling, J. E. Costa, C. A. Gardner, J. W. Vallance, and J. J. Major. 1997. “Volcano Hazards in
    the Mount Hood Region.” Oregon: US Geological Survey Open-File Report 97: 89.
    Wood, Nathan J., and Mathew C. Schmidtlein. 2013. “Community Variations in Population Exposure to near-Field Tsunami
    Hazards as a Function of Pedestrian Travel Time to Safety.” Natural Hazards 65 (3): 1603–28.
    

    View Slide

44. View Slide

45. View Slide

46. 3D Line Features
    

    View Slide

47. 3D Line Features
    

    View Slide

48. Methods:
    The Lahar Travel Time Raster
    

    View Slide

49. Data from Pierson (1998)
    The Lahar Travel Time Raster
    

    View Slide

50. The Lahar Travel Time Raster
    

    View Slide

51. View Slide

52. View Slide

53. No significant differences between grouped 2D and 3D maps (p value = 0.57)
    Self-location
    

    View Slide

54. The Lahar Travel Time Raster
    

    View Slide

55. Lahar Travel Time Interpretation
    

    View Slide

56. No significant
    difference between
    groups or pairs
    (p values > 0.12)
    No significant
    difference
    Lahar Travel Time Interpretation
    

    View Slide

57. Evacuation Routes:
    Critical Speed
    

    View Slide

58. Evacuation Routes:
    Critical Speed
    

    View Slide

59. Modified from Haynes et al. (2007)
    Previous Research
    

    View Slide