CReSIS OSU

1

• CReSIS
• The Ohio State University
• Ohio State University is responsible for four
  primary tasks
• Developing regional-scale geophysical data sets
  from satellite data
• Developing new techniques for measuring the
  physical properties of
• firn and ice
• Extending glaciological theory that incorporates
  new observations
• of the glacier bed provided by CReSIS
• Develop new, web-based courses in Polar Science
• OSU will also work closely with KU engineers to
  identify science requirements
• and engineering requirements that drive system
  development. OSU
• will also help design field experiments that
  validate system performance.
• Along with developing new courses in Polar
  Science, OSU is working
• With local industry to develop outreach
  opportunities

Analysis of Surface Velocity Fields

• Current Research
  Personnel
• K. Jezek, task leader and Cresis OSU P.I.
• E. Mosley-Thompson, Ice core analysis and
  interpretation
• C. J. van der Veen, Glacial theory, education
  and outreach
• L. Thompson, Climate from ice cores
• Carol Landis, Education
• D. Bromwich, Polar meteorology
• V. Zagorodnov, Firn sampling technology
• B. Csatho, DEM preparation, optical map
  products
• P.-N. Lin, ice core stable isotopes / chemistry
• K. Farness, SAR analysis
• 2 undergraduates funded by CReSIS
• 3 graduate students

• Schedule
• (Detailed schedules are
  provided with each task)
• Year 1 Objectives
• Data compilation
• Science requirements on radar and in situ
  technologies
• Web cast lectures to team members and visit to
  ECSU
• Year 1 Accomplishments
• Contributed to requirements documents
• Data sets made available on web
• Firn sampling instrumentation development
  started
• Multiple satellite data acquisitions initiated
• Paper submitted on distribution of melt under
  ice stream shear margins
• Two papers accepted on trimline and lichen
  mapping from spectral data
• Year 2 Objectives
• Engineering requirements refinement
• Prepare derived products (for example, surface
  velocity maps)

2
Analysis of Surface Velocity Field and
Velocity Gradients The Ohio State
University This task will measure and compile the
surface velocity fields for the polar ice
sheets. Velocity fields are already available
for portions of the Antarctic from the MAMM
project. Additional Radarsat InSAR data were
for interior Antarctica during AMM-1 and new
data about the perimeter have been collected
through 2005. Velocity gradients and derived
field will be analyzed using a proposed variant
of the conventional force budget technique.
The results will be used in comparison with
CReSIS imaging radar data to investigate
properties of the glacial bed. In addition, the
task will develop maps of Cresis study
sites using optical satellite imagery and
satellite altimeter systems. As part of this
task, the PI will visit ECSU to give
presentations on Polar Science and to interact
with students and staff. Two web-based Polar
Science Seminars will also be offered.
Analysis of Surface Velocity Fields

• Current Research
  Personnel
• K. Jezek, task leader and Cresis OSU P.I.
• B. Csatho, DEM preparation and site maps
  creation
• K. Farness, InSAR processing
• 2, CReSIS fundeded undergraduate students
  involved with
• feature retracking velocities and data
  rescue (K. Leibfacher, S. Westfall)
• 1 grad student (J. Wuite)

• Schedule
• Year 1 Objectives
• Assemble available SAR Data and available
  derived velocities
• Assemble available DEM Data (Ekholm, Bamber,
  IceSAT)
• Develop complete 3-d force theory
• Invited talk at ECSU
• Year 1 Accomplishments
• Two undergraduates involved in research
• SAR data products available on web site
• Envisat data acquisitions submitted and data
  being received
• TerraSAR X data acquisition request approved
• Radarsat Greenland data scheduled to be
  delivered in June
• Envisat upgrade to procesing software installed
• Several DEMs received, compared and integrated
• Net forces computed on several glaciers
• Paper submitted on melt beneath shear margins
• Year 2 Objectives
• Complete integrated 200 m DEM of Greenland
• Create SAR mosaics and begin InSAR processing
  (Envisat)

3
High Resolution Analysis of the Physical and
Chemical Properties of Snow and Firn using
Multiple Technologies The Ohio State
University Ice Core Paleoclimate Group
High resolution, in situ measurements of the
physical and chemical properties of firn and ice,
along with annual snow accumulation are essential
for correct interpretation of airborne and
satellite-borne remote sensing data. Density is
one of the most difficult properties to measure
and is best evaluated in situ. At least three
different tools will be constructed, tested, and
deployed to measure density with high vertical
resolution and increased precision over current
methods. Near-surface densities that change
rapidly with depth will be given special
attention. Ideally the speedograph will be
calibrated so that many shallow (lt 20 meter)
profiles can be measured quickly in a region.
This is critical as density can be highly
variable over small distances, especially in
regions where the snow facies are not dry and the
degree of melt and refreezing is laterally
variable.
Speedograph penetration is a function of density
Schedule

• Year 1 Objectives and Accomplishments Design,
  Fabrication, Testing
• ICAS (Ice Core Analysis System) for high
  resolution profiling of
• density, grain size, and electrical
  conductivity ECM
• design is complete and drawings are 35
  complete
• 40 of parts have been acquired or
  fabricated
• Fabrication of the Speedograph (in situ,
  continuous density profiling)
• design is complete and technical drawings
  are 20 complete
• 10 of parts have been acquired or
  fabricated
• - Conventional density measurements (for
  calibration)
• design is complete and technical drawings
  are 10 complete
• 10 of parts have been acquired or
  fabricated
• Year 2 Objectives Continue Fabrication and
  Testing
• Testing all three devices with firn cores on
  hand at OSU
• Density calibration and Inter-lab comparison
  (Japan or AWI)
• Year 3 Objectives field testing and equipment
  modification
• - Field test ICAS and Speedograph Greenland
• Make required refinements identified by field
  tests
• Analyze selected core sections by conventional
  methods to validate ICAS
• observations and confirm annual layer (net
  accumulation) interpretations

Research Personnel
Ellen Mosley-Thompson Victor Zagorodnov Lonnie G.
Thompson Ping-nan Lin 1 graduate student who
will analyze the physical properties of
firn / ice (calibration studies) anticipated
to start Sept 2006
4
To place ongoing changes on Greenland outlet
Glaciers in a broader historical perspective,
we will be measuring trimline elevations of
selected outlet glaciers and estimate the volume
of ice lost since the Little Ice Age
maximum. To gain better understanding of the
bed characteristics under fast-moving ice
streams, we will develop DEMs of paleo ice
streams on the Canadian shield, and compare these
with bed topography of modern active ice
streams.
Geomorphology mapping from satellite imagery and
aerial photographs

• Schedule
• Year 1 Objectives
• Produce 3D map of the trimline in Jakobshavn
  Isfjord based on existing aerial photographs
• Identify regions on the Canadian shield suitable
  for the study of paleo ice streams and collect
  satellite data images (Aster, IceSAT)
• Year 1 Accomplishments
• Create land-cover map from Landsat ETM and ASTER
  imagery, Jakobshavn Glacier
• Assess accuracy of land-cover maps by comparison
  with aerial photographs, geomorphologic maps and
  field observations (two publications in press)
• Create DEM from ASTER and assess its accuracy by
  aerial photogrammetry measurements, Jakobshavn
  Glacier
• Evaluate effect of subglacial topography on
  geothermal heat flux, paper in preparation
• Year 2 Objectives
• Compare Aster DEM with those derived from aerial
  photogrammetry
• Produce 3D map of the trimline of Jakobshavn and
  Kangerlussuaq glaciers, estimate volume loss
  since LIA
• Produce DEM of paleo ice stream
• Assess importance of geology on the occurrence of
  ice streams (sediment availability, erodibility
  of bedrock, geothermal heat flow)

• Research Personnel
• Kees van der Veen
• Bea Csatho
• Toni Schenk
• Student Personnel
• Kyung In Huh
• Yushin Ahn
• 1 potential graduate student

5
Our vision is to educate students about the
fundamental principles of earth science and the
unique role of the polar regions in earth
systems. Our objective is to train students who
will be able to critically and creatively apply
these principles in their chosen careers. CReSIS
outreach will be integrated into regular
BPRC activities that include annual visits by
primary and secondary school students from
Central Ohio. BPRC also hosts student groups
from local colleges and teacher organizations.
BPRC is working with McGraw Hill Company to
increase access to polar science information.
Education and Outreach

• 
  Schedule
• Year 1 Objectives
• ECSU presentations
• Continue discussions with McGraw Hill and
  production of K-12 material
• Proposal to establish UG/G track in cryospheric
  science at OSU
• Lonnie Thompson web-cast lecture
• Year 1 Accomplishments
• 2 undergraduate students hired into RSL
• ECSU/Haskell presentations
• Discussions with McGraw Hill
• Lonnie Thompson web-cast lecture
• Numerious visits by K-12 students, teachers,
  parents, college
• students and teachers
• Submitted a proposal to Batelle to host 3
  day-long climate change sessions
• (k-12, grad/undergrad and grad, general
  public)
• Submit proposal to GLOBE RFP (energy budget
  focus)
• Year 2 Objectives
• Lonnie Thompson participation in Dole Center
  Workshop
• Polar Science Seminar

• C. Landis
• All Senior OSU Faculty and Staff
• BPRC Administrative Staff

6
Students

• Jan Wuite Grad Student, glacier dynamics using
  feature retracking and InSAR velocities (NASA
  Fellow)
• Kyung In Huh Grad Student, glacier
  geomorphology (NASA Fellow)
• Karl Leibfacher senior velocity from AMM-1 and
  MAMM feature retracking (CReSIS funded)
• Stacey Westfall senior data rescue/reformatting
  of ERS/JERS SAR InSAR data (CReSIS funded)
• Two graduate student admitted for fall and
  nominated for OSU Presidential Fellowship
  (acceptance TBD).

AMM-1/MAMM Feature retracking
7
CReSIS Partners Talks

• Haskell Feb. 23 (Jezek)
• ECSU March 06 (Jezek)
• CRESiS Seminar presentation March 9, 2006. In
  preparation. (Zagorodnov V., E. Mosley-Thompson)
• Dole Center Workshop April 06 (
• L. Thompson)

8
(No Transcript)
9
CReSIS Related Publications
Mosley-Thompson, E., C. R. Readinger, P.
Craigmile, L. G. Thompson, and C. A. Calder.
2005. Regional sensitivity of Greenland
precipitation to NAO variability, Geophysical
Research Letters, 32, L24707, doi10.1029/2005GL02
4776. Raymond, C.F., G.A. Catania, N. Nereson,
and C.J. van der Veen, Bed radar reflectivity
across the north margin of Whillans Ice Stream
and implications for margin processes. Journal
of Glaciology, in press. Van der Veen, C.J.,
K.C. Jezek, and L. Stearns, Shear measurements
on three West Antarctic ice streams 1. Whillans
Ice Stream. Journal of Glaciology,
submitted. Van der Veen, C.J. and B. Csatho,
accepted. Spectral characteristics of Greenland
lichens, Geographie et Physique Quaternaire.
Csatho, B., C.J. van der Veen and C. Tremper,
accepted. Trimline mapping from multispectral
Landsat ETM imagery. Geographie et Physique
Quaternaire Zagorodnov V., O. Nagornov and L.G.
Thompson. 2006. Influence of air temperature on a
glaciers active layer temperature. Annals of
Glaciology (Accepted for publication)
Leibfacher, K., S. Mather, K. Farness, and K.
Jezek, 2006. Minimosaic Offset Investigation.
BPRC Technical Report, 196 p.
10
CReSISOSU Remote Sensing

• Progress Report
• 2006

Surface Flow Field Belgica Mts., Radarsat InSAR
11
Digital Elevation Models

• GEOSAT and ERS radar altimeter DEMS obtained from
  GSFC
• IceSAT DEMS obtained from GSFC
• Shape from shading DEM obtained from NSIDC
• Shape from shading has the fewest outliers and
  has been resampled to a 1 km grid
• DEMs will be merged to create the best available
  surface topography

12
Accuracy Assessment of Merged DEMs

• Available DEMs
• (A) Ekholm/Bamber DEM, created from ERS-1 and
  Geosat satellite radar altimetry, airborne laser
  altimetry (ATM, PARCA), photogrammetry, InfSAR,
  digitized maps (Bamber, J.L., S. Ekholm, W.B.
  Krabill, 2001. A new, high-resolution digital
  elevation model of Greenland fully validated with
  airborne laser data. JGR, 106(B4), 6733-6745)
• (B) Scambos DEM created from the Ekholm/Bamber
  DEM by adding details from AVHRR imagery by using
  shape from shading technique (Scambos, T.A. and
  T. Harran, An image-enhanced DEM of the Greenland
  ice sheet. Annals of Glaciology, 34, 291-298)
• New DEM
• (C) OSU-06 DEMs created from (A) and (B) grids by
  adding a correction grid derived from ICESat
  measurements

13
Rationale for Creating Merged Products from
Existing DEMs and ICESat

• Accuracy issues and processing uncertainties
  related to the ERS-1 and ERS-2 data have been
  identified. Therefore beging to develop merged
  products by combining existing DEMs with ICESat
  data and postpone the generation of a DEM from
  the point data sets (airborne and spaceborne
  lidar and satellite radar altimetry) until radar
  altimetry corrections are fully understood and
  accuracy studies are finished. So far all
  experiments suggest that the accuracy of the
  OSU-06 (C) product (existing DEMs with
  corrections derived from ICESat) will meet the
  criteria given for this project for most of the
  ice sheet except the marginal zone.

14
Ekholm/Bamber
Ekholm/Bamber DEM Statistics of difference
between ICESat elevations and DEM Number of
values 19904 Minimum -30.83 m Maximum 22.18
m Mean 0.18 m Standard deviation 6.67
m Accuracy agrees well with quoted accuracy of
DEM (Bamber et al., 2001)
Ekholm/Bamber with IceSAT Tracks
15
Residual between ICESat elevations and
Ekholm/Bamber DEM(obtained by Kriging
interpolation of residual)
16
Ekholm/Bamber DEM interpolated ICESat
residuals Interpolation Kriging No
filtering Statistics of difference between
ICESat elevations and DEM elevations after
interpolated residuals are added Number of
values 19904 Minimum -15.28 m Maximum 8.58 m
Mean -0.02 m Standard deviation 1.48 m
17
Ekholm/Bamber DEM interpolated ICESat
residuals Interpolation Kriging Filtering of
residual grid to remove artifacts using a 11 by
11 Gaussian filter Statistics of difference
between ICESat elevations and DEM elevations
after interpolated residuals are added Number of
values 19904 Minimum -25.00 m Maximum 13.92
m Mean 0.11 m Standard deviation 3.95 m This
is not real accuracy measure, since ICESat data
were used to create the improved DEM. Accuracy
assessment with ATM data is ongoing. Same applies
for subsequent examples
18
Scambos photoclinometry DEM Statistics of
difference between ICESat elevations and
DEM Number of values 19904 Minimum -20.32
m Maximum 17.60 m Mean -0.96 m Standard
deviation 5.81 m
19
Scambos photoclinometry DEM interpolated ICESat
residuals Interpolation Kriging No
filtering Statistics of difference between
ICESat elevations and DEM elevations after
interpolated residuals are added Number of
values 19904 Minimum -16.99 m Maximum 10.27
m Mean 0.02 m Standard deviation 1.36 m -
20
Scambos photoclinometry DEM interpolated ICESat
residuals Interpolation Kriging Filtering 11
by 11 Gaussian filter Statistics of difference
between ICESat elevations and DEM elevations
after interpolated residuals are added Number of
values 19904 Minimum -19.82 m Maximum 13.56
m Mean 0.02 m Standard deviation 3.42 m
21
Ekholm/Bamber DEM Photoclinometry
DEM Mean /- standard deviation of difference
between ICESat elevations and DEMs
Original Interpolated, smoothed ICESat
residual added Interpolated ICESat residual
added
0.18/-6.67 m 0.11/-3.95 m -0.02/-1.4
8 m
-0.96/-5.81 m 0.02/-3.42
m 0.02/-1.36 m
Recommended approach to developing OSU 06 DEM
22
SAR/InSAR Data

• MAMM Radarsat data on order from ASF
• ENVISAT data acquisitions planned and requests
  submitted
• Cycle 1 Envisat data being received at OSU
• TerraSAR-X acquisitions approved
• Modified FOCUS/RAMS software available to process
  ENVISAT data

23
Approved Envisat Acquisitions
24
TerraSAR-X Requested Coverage
25
Data Products

• Data products including geocoded MODIS images are
  available to the team at
• www-bprc.mps.ohio-state.edu/rsl

26
GLACIAL DYNAMICS CReSIS OSU Progress Report
2006 Spectral characteristics of Greenland
lichens and Trimline mapping from multi-spectral
Landsat imagery
B. Csatho C.J. van der Veen C. Tremper
In Press, Geographie et Physique Quaternaire
27
Classified Landsat Imagery
Ice
Melting ice, snow, firn
Supraglacial lakes, brash ice
Turbid water
Moderately turbid water
Slightly turbid water
Clear water, shadow
Lichen-dominated vegetation
Tundra-dominated vegetation
Dry sediment
Wet sediment
Trimzone, sediment, some vegetation
Trimzone, sediment, no vegetation
Debris-covered ice
28
Trimline location from Landsat classification and
ground GPS survey
29
Spectral properties of major landcovers
Lichen An association of a fungus and a photo
synthetic symbiont, resulting in a stable thallus
of specific structure
30
Bare rocks
Lichen-covered rocks
31
Intermittent retreat of Jakobshavns Isbræ since
the LIA maximum
32
Elevation on WGS-84
(meter)
3
5
0
Trimline Airborne laser
3
0
0
2
5
0
2
0
0
?
1
5
0
1
0
0
5
0
0
1
8
6
0
1
8
8
0
1
9
0
0
1
9
2
0
1
9
4
0
1
9
6
0
1
9
8
0
2
0
0
0
2
0
2
0
Date (year)
Surface elevation changes since the LIA dating
based on historical photographs, satellite
images, and lichenometry
33
Firn Core Chemistry and Physics
Progress Report 2006
34
Ice core processing schema
Split 60/40 with high speed horizontal band saw
Band saw design
Band saw in cold room
35
Ice core processing schema (continued)
Conventional analysis

• density (calibration)
• grain size (calibration)
• stable isotopic ratios
• major ions
• microparticles (dust)

Milling
Milling spindle-motor
36
Optical sensors sample section thickness
Two industrial proximity sensors Resolution
40 ?m
Snow-ice section 10 mm
Proximity sensors
37
Optical high resolution density sensors (ice core
longitudinal scan) - light
absorption and scatter (?1),
- light scatter (?2)
Ref PbS detector
Signal PbS detector
Combiner for laser beams
Positioning slides with spindle motor and laser
sensors tested prototype device
38
Speedograph
1. Hot point drill - power, 1.0 Kw -
length, 0.8 m - weight, 5 kg - diameter, 35
mm - penetration rate 9-12 m/h 2. Winch, 50
m, 3. Controller, 4. PC data acquisition
- depth - penetration rate - bit
pressure - tip power -
temperature 5. Power generator, 6. Shelter
Encoder well
Load cell
Slip ring
Winch front view
Winch side view
Hot point drill