Pitfalls and Possibilities in the Use of Maps to Visualize Earth Data and Understand Earth Processes

1
Pitfalls and Possibilities in the Use of Maps
to Visualize Earth Data and Understand Earth
Processes

• Kim Kastens
• Lamont-Doherty Earth Observatory of Columbia
  University
• Presented at workshop on Use of Visualization in
  Geosciences
• Carleton College, February 27, 2004

2
Pitfalls Preview

• Many students are weak in projective spatial
  skills bad news for map projections
• Students prior map skills training may have
  lacked any connection with the real world
• When maps and profiles are combined to explore a
  data volume, likelihood of nonpenetrative errors
• Clearly articulating and modelling an approach to
  map-based data exploration that works for
  professionals may be insufficient for novices
• Understanding is not the same as believing and
  acting upon.

3
Pitfall 1
Projective spatial skill includes the ability to
anticipate the shape of a projection (or shadow)
cast by an object held up to a light.
(from Downs Liben, 1991)
Good answers show understanding of variation as a
function of thickness of the shape.
Good answers show understanding of variation as a
function of the angle of presentation of shape
relative to the light and screen.
4
Mistaken answers include upside down
shadows incomplete shadows concave
shadows of convex objects asymmetrical
shadows of symmetrical objects inconsistent
treatment of thickness of object
(from Downs Liben, 1991)
5
This projective spatial skill is important in
teaching about maps because map projections are
often illustrated by having students envision a
light inside the globe that projects shapes from
the surface of the globe onto a sheet of paper
held tangent to the Earths surface.
(http//www.cfr.washington.edu/research.urbaneco/G
IS_Tutorial/UEGIS-tutorial-notes.htm)
6
Many college students do very poorly on Downs
Libens shadow projection test. How well do you
suppose these students will understand standard
explanations of map projections? What can we
do to help them?
7
Not to mention other, more obscure, spatial
representations used in Geosciences, which rely
heavily on projective spatial skills.
from Hobbs, Bruce E., Winthrop D. Means, and
Paul F. Williams (1976) An Outline of Structural
Geology. New York John Wiley Sons, Inc.
8
Pitfall 2
Frontispiece from Geography for Life The
National Geography Standards
What is odd about this picture?
9
Frontispiece from Geography for Life The
National Geography Standards
It doesnt show a school or a school activity!
10
Thats because the crucial and complex skill of
translating between the real world and the
representation (the map) is usually not taught in
school.
11
The City Park is between which streets? A
and B Streets B and C Streets A and C
Streets How many stores are on the map? 0
1 2 3 4 5
For many elementary school map skills activities
there is no actual represented space. Or even
if the represented space does exist, the
questions can be answered entirely in the frame
of reference of the map, without thinking about
the represented space.
12
Its like learning to read musical notation
without connecting it to the heard sounds.
?
equals
13
(diagram after Liben, 1997)
14
(after Liben, 1997)
15
Many elementary school map skills curriculum
materials involve tasks that can be solved
entirely in the frame of reference of the map,
without ever thinking about the represented space
(the real world).
16
(No Transcript)
17
Pitfall 3
Many important aspects of the Earth system are
inherently three-dimensional, not flat like a map
nor even a draped surface like the Earths
surface.. .. for example the ocean and its
spatial distribution of salinity and temperature.
18
So we combine maps with profiles/sections to
allow students to explore a three-dimensional
data volume to discover 3-D structure and infer
flows and process.
19
It looks to me, from watching students try the
ocean salinity exercise, that many are making
nonpenetrative errors in the sense of Kali
Orion (1996)
20
In ocean salinity exploration students can draw
any N-S or E-W profile they want or maps at any
water depth . . But they can only see one map
or profile at a time. How to help learners grasp
internal structure of 3-D volume?
21
Pitfall 4
(from Mayer, 2002, based on software by W.
Prothero)
22
First Training Approach Look at sketches of
possible geological features Pictorial
Training
(after Mayer et al, 2002)
23
Second training approach Strategic Scaffolding
(after Mayer et al, 2002)
24
Correct answers out of 5 M2.36
SD1.52 M3.25 SD1.41 M2.90
SD1.78 M3.39 SD1.41
Control (no aids)
Both aids
25
Analyzing and clearly articulating the strategies
used by experts.. . Isnt necessarily
going to be the strategy that is most effective
for novices.
26
What happens when we try to communicate outside
our cozy circle of geoscientists and geoscience
students?
Pitfall 5
What geologists are trained to think about.
27
What society wants to know about
28
Forecast maps combine two challenging skills or
understandings spatial thinking
probability How well does the intended audience
of decision-makers and policy-makers understand
such maps?
(from International Research Institute for
Climate Prediction http//iri.columbia.edu/foreca
st/net_asmt)
29
Suppose that you were given this forecast map in
January 2003. Based on this map, how would you
answer the following question Which area will
receive a greater amount of total precipitation
for this forecast period, Southern California or
Washington State?
(from International Research Institute for
Climate Prediction http//iri.columbia.edu/foreca
st/net_asmt)
30
Participants 47 students in the Masters program
in Environmental Science and Policy, School of
International and Public Affairs, Columbia
University
(from Ishikawa, Barnston, Kastens, Louchouarn and
Ropelewski , in prep)
31
Forecast
Subsequently Observed
(from International Research Institute for
Climate Prediction http//iri.columbia.edu/foreca
st/net_asmt)
32
how would you characterize the correspondence
between the forecast and the observation?
(from Ishikawa, Barnston, Kastens, Louchouarn and
Ropelewski , in prep)
33
how would you characterize the correspondence
between the forecast and the observation?
would you recommend that such forecasts be
used to make decisions about what crops to plant?
(from Ishikawa, Barnston, Kastens, Louchouarn and
Ropelewski , in prep)
34
Understanding data (from map or
otherwise). .is different from believing
data. is different from acting upon data.
(from Ishikawa, Barnston, Kastens, Louchouarn and
Ropelewski , in prep)
35
Pitfalls Review

• Many students are weak in projective spatial
  skills
• Students early map skills training often lacks
  any connection with the real world
• When maps and profiles are combined to explore a
  data volume, likelihood of nonpenetrative errors
• Clearly articulating and modelling an approach to
  map-based data exploration that works for
  professionals may be insufficient for novices
• Understanding is not the same as believing and
  acting upon.