Working with Tangible GIS

GeoISEM Home




GMSlab 2001

Helena Mitasova, Lubos Mitas NCSU in collaboration with
Carlo Ratti, SENSEable City laboratory, MIT and VISSTA group, ECE NCSU

Mitasova, H., Mitas, L., Ratti, C., Ishii, H., Alonso J., and Harmon, R.S., 2006,
Real-time Human Interaction With Landscape Models Using a Tangible Geospatial Modeling Environment,
IEEE Computer Graphics & Applications, Special Issue - GeoVisualization, 26(4), pp 55-63.[pdf]

Tangible GIS in action: Water flow simulation, Modifying terrain and analysis Fort Bragg by Brent Fogleman

Setup January 2013

Kinect experiments

Jockey's Ridge model
Elevation model, slope, sharper version, flooded, flow test

Sand Model:
Initial pile of sand, Sand pile modified based on projected contours, DEM interpolated with modified parameters, Flow accumulation over the kinect-based DEM

Setup December 2007:

a b

DEM+cover            overland flow           soil detachment         net erosion/deposition
real world model

plasticine models:

road breach and checkdam

buildings I, water from buildings only, water from everywhere

Galery of designs by VISSTA Lab visitors
design for bioretention: elevation, water and sediment flow, solar irradiation

buildings II and III

lagoon and landscape design

buildings IV and off-road vehicle tracks, 3x z-exager

tec design

The first experiments

The first set up

Animations: overland flow
over smoothed DEM derived from real-world data
over DEM from TanGIS
over DEM from TanGIS with modified surface roughness and buildings

Future system linked to real time monitoring

Terrain and wireless signal analysis for web cam monitoring


Old solid model elevation difference: 72mm representing 76ft, 6x vertical exageration, horizontal scale 1:2000,
DEM with 315x284 rows/cols at 6ft resolution and model 630x560(fix) rows/cols at 1mm resolution
New model: 50mm, 3x vertical exageration, horizontal scale 1:1200, 0.4x0.4m,
DEM 480x480m, 1m resolution, derived from lidar

Real world test area : airphotos(2005)
a b
observed and simulated flow
a b c more photos

Model scanned at MIT and lidar-based DEMs
a b c d
a) Scanned model interpolated at 1mm resolution, high tension/low smoothing
b) Scanned model interpolated at 1mm resolution, low tension/high smoothing
c) Lidar-based DEM (2001) interpolated at 2m(6ft) resolution, high tension/low smoothing
d) Lidar-based DEM (2001) interpolated at 2m(6ft) resolution, low tension/high smoothing

Landscape properties from GIS data

landuse soils

aspect slope erdep

solar Jan Mar Jun


Ready to use GRASS6 database with dynamic water flow model results in mapset lwhome

Elevation raster data in GRASS ascii format(larger area than above)


Tangible GIS: New Human-Computer Interface For Geospatial Modeling [ppt]

This project is supported by Army Research Office

Map to VISSTA lab (pdf 2.5MB)