|
 
On
October
29th, 1999, members of the
Apoala research group
introduced a prototype model of a same time-different place collaborative
environment at the WebGIS
conference held in State College,
PA.

Exploratory
analysis of complex, multi-dimensional space-time data sets
demands new and innovative tools. As an extension from the Apoala
Project, individuals in the GeoVISTA Center and the
Center for Academic Computing are working together to explore
the potential of same-time-different place collaboration among scientists
at remote locations as they explore complex spatiotemporal data.
Many
potential scientific, educational, and decision-making applications
for Geographic Visualization (geovisualization) involve small groups
working together on a problem solution, but existing tools are designed
for use by individuals. We developed a prototype geovisualization
environment focused on same time-different place collaboration. Such
environments have potential application to regional and local planning/decision-making,
scientific research by distributed interdisciplinary teams, and web-based
education.

Screen
Capture of the Collaborative Environment
Our
prototype, consisting of a series of linked desktops, is designed
to facilitate collaboration among users who are exploring time series
of climatic data for a large drainage basin. The
exploration involves
interaction with shared dynamic (animated and interactive) displays.
The prototype
is constructed from a set of Java/Java3D
tools. These include VisAD, a DEM viewing module that
works
with VisAD, and our own extensions for data queries and networking.
In relation to query, we focused on temporal query tools designed
to help users explore both linear and cyclic components of the ata.
The
prototype collaborative geovisualization environment allows multiple
users to view and manipulate the changing climatic data simultaneously
and thus
to share knowledge as they identify drainage basin scale patterns
and processes. Though the Java language operates across platforms
and operating systems, dealing with different levels of hardware performance
and Internet access speeds represent future challenges for implementing
ideas explored here. The data used in the demonstration are extracted
from a much larger climate data set for the Susquehanna River Basin
of Pennsylvania, New York, and Maryland -- specifically daily maximum
temperature, minimum temperature and precipitation extending from
1983-1993.
Types
of Collaborative GIS/geovisualization
- same
place - same time
- planning
meeting; Group SDSS
- same
place - different time group
- different
place - same time
- different
place - different time
Our
prototype is a system of linked desktops using Java/Java3D (VisAD
+ DEMViewer + our extensions). VisAD and the DEM Viewer are explained
below.

VisAD
– Java (2D/3D) class library for interactive and collaborative visualization
and analysis of numerical data… primary author: Bill Hibbard, Space
Science and Engineering Center University of Wisconsin, Madison
www.ssec.wisc.edu/~billh/visad.html

DEMViewer
is a digital elevation model viewer for ArcGrid ASCII export files.
It is written in Java and uses VisAD and Java3D
DEMViewer
was developed by Ugo Taddei, Department of Geoinformatics, Geohydrology
and Modelling, Institute of Geography, University of Jena, Germany.
www.geogr.uni-jena.de/~p6taug/demviewer/demv.html

In
order to create a collaborative environment, it was necessary to build
a TalkServer allowing communication among different computers.
TalkServer
is a JAVA application for communicating user initiated events among
networked collaborative applications. TalkServer was developed at
the Visualization Group of Penn
State's Center for Academic Computing. Hadi
Abdo is the primary author of TalkServer. TalkServer listens on
a predetermined port of a server for new connections from client applications.
For each new socket connection detected, TalkServer creates a TalkServerThread
(TST) to communicate with the connected client application. When
a TST receives subsequent messages from its client application that
indicate changes that will effect other clients in the collaborative
session, the messages are relayed to the TalkServer. TalkServer then
requests that all TSTs update their corresponding clients accordingly.

- Dr.
Alan MacEachren,
Director, GeoVISTA Center, Professor
and Faculty Fellow CAC
- George
Otto, Manager
- Visualization Group, CAC, PSU
- Dr.Brent
Yarnal, Professor Dept of Geography,
PSU
- Jack
Gundrum,
Visualization Group, CAC, PSU
- Isaac
Brewer, research assistant Dept of Geography, PSU
-
Hadi Abdo- Visualization Group, CAC,
PSU
- Amy
Griffin, research assistant
Dept of Geography, PSU
- Jeremy
Mennis, research assistant
Dept of Geography, PSU
- Daniel
Haug,
GeoVISTA staff scientist Dept of
Geography, PSU
- Masahiro
Takatsuka, GeoVISTA
staff scientist Dept of Geography, PSU

|

Animated
View Window

The
animated view window allows collaborating users to manipulate the
3-D cube in all directions. The program also allows the users to zoom
in and out from all angles. The example above displays precipitation
distribution across the Susquehanna River Basin, with green indicating
precipitation.
Cyclic
Legend

The
cyclic legend was developed to allow users to query the data for cyclical
phenomena. In the above example, the user has chosen to animate only
the summer months (June, July, August, and September) between 1983
and 1992.
Controls

The
Animation Control box allows the user to start the animation, or step
through with each click of the mouse. The text box allows the user
to specify the speed of the animation. (333 ms/frame is equivalent
to three frames per second and 1000 ms/frame represents 1 frame/second).
We are currently investigating ways to make the animation speed selector
more user friendly.
The
linked desktops are sychronized when the GO button is pressed. The
computers will stay relatively close to the same speed, however, delays
can occur due to network speed as well as differences in the relative
CPU speed required for animation of the data displayed on each client's
monitor.
Vertical
exageration is controled by the bar on a scale of 1-10. In the future,
we hope to allow the user to specify the boundaries of this scale.
The color selector allows the user to select all possible color combinations
by manipulating the red, green, and blue curves in the spectrum. The
numbers contained on the legend represent the range of values taken
from the data loaded into the program (in this case the maximum temperature
range is from 51 degrees F to 99 degrees F.
The Animation by Day selector shows the user which file is being displayed
in the viewable portion. The user can move the slider to select individual
days. There are 31 separate tick marks on this slider bar representing
the 31 days of the month.

- Link
the environment to an object-oriented database that supports more
flexible queries
- Make
the time selector a dynamic legend
- Develop
more flexible and complete interaction -- e.g., color selection
- Add
temporal averaging tools experiment with more than two collaborators
- Compare
same & different place
collaboration
- Transfer
the prototype into IDesk environment

The
visualized data comprise the full extent of the Susquehanna River
Basin (SRB), which covers over fifty percent of the land area of Pennsylvania,
part of west-central New York, and a portion of Maryland on the Chesapeake
Bay. The extent ranges from about 39.5° N to 43° N latitude and 75.5°
W to 79° W longitude. The physiographic regions of the SRB stretch
from the Piedmont in the southeast corner to the Appalachian Plateau
in the upper half of the basin. Nestled between these two physiographic
regions and accounting for the majority of the lower half of the basin's
areal extent is the Ridge and Valley province.

The
raw data, provided by the National Climatic Data Center in Asheville,
North Carolina, span a period from 1948 to 1993. Between 250 and 300
data recording stations collected the raw daily maximum temperature,
minimum temperature, and precipitation measurements, with an average
of 150-200 stations actually contained within the boundaries of the
Susquehanna River Basin on any given day. First, unreliable data were
eliminated, and the temperature data were corrected for localized
topographic variation by standardizing the values to Mean Sea Level.
The data were then interpolated into a 4km cell based grid. Finally,
a Digital Elevation Model was used to restore the elevation values
to the temperature data sets.
|