Operation IceBridge
Phase 3 DRAFT White Paper
Land Ice Team
December 15, 2011
1.
Introduction
NASA established the Operation IceBridge program (OIB) with a mandate to fulfill the
following programmatic goals:




Make airborne altimetry measurements over the ice sheets and sea ice to extend and
improve the record of observations begun by ICESat.
Link the measurements made by historical airborne laser altimeters, ICESat, ICESat-2,
and CryoSat-2 to allow accurate comparison and production of a long-term, ice altimetry
record.
Monitor key, rapidly changing areas of ice in the Arctic and Antarctic to maintain a long
term observation record.
Provide key observational data to improve our understanding of ice dynamics, and better
constrain predictive models of sea level rise and sea ice cover conditions.
In additional, OIB has the following technical goal:

Adapt existing instruments for airborne remote sensing of ice by unmanned aerial
systems such as NASA’s Global Hawk.
Based on a review of the programmatic goals and an assessment of the specific science OIB can
address, the OIB Land Ice Team further identified several science questions that would help
direct the development of measurement requirements and accuracies mission planning,
instrument suites, data processing and data product delivery
(http://bprc.osu.edu/rsl/IST/documents/Ice%20Bridge%20Level-1_3_06_11_append.docx).
These are


Where are glaciers continuing to thin and where may they be slowing/ thickening (G1)
What are the major forces and mechanisms causing the ice sheets to lose mass and
change velocity, and how are these processes changing over time? (G2)
o How do the ice sheet/glacier surface topography, bed topography, bed geology,
ice shelves/tongues, and grounding line configurations effect ice dynamics?
o How far inland are the effects of coastal thinning transmitted and by what
physical processes?
o How far downstream do changing processes near the ice divide (changes in snow
accumulation, divide migration) effect glacier flow
o What is the important scale for measuring geophysical parameters so as to
substantially improve modeling fidelity?
1


o Where is the subglacial water produced and where is it going?
How do ocean, sea ice, ice sheet interactions influence ice sheet behavior (G2)
o How does the bathymetry beneath Arctic fjords and Antarctic ice shelves
influence ocean/ice sheet interactions and ice sheet/glacier flow dynamics?
What are yearly snow accumulation/melt rates over the ice sheets? (G1)
o How do changing accumulation rates (and hence near surface densities and firn
structure) impact altimetry measurements
o What are the surface-melt flow-patterns and how do they change with time?
OIB has now been operating successfully for over 3 years. In that time, the program has
demonstrated substantial abilities to conduct science missions across Greenland, Antarctica,
Arctic ice caps and Alaskan Glaciers. The resulting data sets are enriched with a variety of
measurements ranging from photon counting lidar elevations to UHF radar sounding data to
airborne gravity anomalies. A robust suite of geophysical parameters is routinely derived from
the basic data sets (Figure 1). Data and derived products are delivered to the science community
in an open and very timely fashion. Indeed a hallmark of the activity is the quick availability of
the data to any interested investigator.
This white paper briefly reviews the development of OIB. The white paper goes on to establish a
longer term vision for OIB that leads to a hand-off of altimetric measurements to ICESat-2.
Figure 1. The richness of the OIB data set is illustrated in this figure of Pine Island Glacier
showing OIB surface altimetry overlain on Radarsat-1 SAR imagery. The middle layer ice
bottom topography derived from depth sounding radar measurements. The lower level is the sea
floor topography constructed from gravity measurements. The estimated grounding line of the
glacier is shown on the basal and surface topographies (courtesy K. Jezek)
2
2.
OIB Phase 1
NASA recognized there would be a gap between ICESat and ICESat-2 that could be partially
filled with airborne data. During phase 1, which started in mid-2008, NASA identified available
aircraft and integrated equipment wherein some of the later were still in late development stages.
Based on the science need, the logistics capability, and the instrument suite, the report titled 'An
analysis and summary of options for collecting ICESat-like data from aircraft
(http://bprc.osu.edu/rsl/IST/documents/IceSAT_Gapfiller_final.pdf) was distributed in early
2009. The report envisioned a diverse instrument suite, identified general areas for data
collection, recognized the need for a data management plan, and anticipated a gradual expansion
of the aircraft fleet to include unmanned vehicles. Phase 1 very successfully seeded the OIB
activity by demonstrating that successful, spatially extensive data collections could be carried out
in both the Arctic and Antarctic (figure 2 and 3).
Figure 2. Pre OIB coverage (1993-2008 Left) and combined coverage (1993-2011 Right)
illustrating that OIB is achieving more than the previous 15 years (courtesy E. Rignot)
3
Figure 3. Antarctic coverage 2009 (courtesy M. Studinger)
3.
OIB Phase 2
Broader and more formal program management was initiated during Phase 2 (the summer of
2010). NASA conducted a proposal review for various instruments and later in the fall of 2010,
established the OIB science team. Based on the technologies demonstrated in Phase 1, the
science team contributed two, important new facets to OIB. First, the science team developed a
set of science requirements that were grounded in primary science questions and substantiated by
estimates of the measurement capabilities as demonstrated in Phase 1. Relying on the science
requirements, the science team proceeded to develop acquisition campaigns that could be
directly traced to the science requirements. After several cycles of planning, the science team
work has contributed to a more complete science definition for OIB and that our organized effort
provides a smoother and more efficient interface between science and implementation. In fact
the team has tried to adopt and refine a specific acquisition strategy as discussed in the next
section.
2.1 Phase 2 Acquisition Strategy
The OIB acquisition strategy aims to build layers of observations that optimize single
observations of static variables (such as basal topography) and construct repeat coverages that
efficiently lead into extending surface dh/dt time series shortly after launch of ICESat-2. The
strategy has three main elements:
4
 Use depth sounding radar, low elevation lidar, gravity, accumulation radar, and DMS to
establish baseline data sets.
o Collect data along closely spaced flight lines (5 km or less) roughly parallel to
elevation contours near the coast and which extend seaward of the ice margin.
o Collect data along more widely separated flight lines (10-20 km) roughly
parallel to elevation contours inland of about the 400 m/yr surface velocity
contour (speed number is just a placeholder right now).
o Collect data on flight lines along center lines of most Greenland and Antarctic
glaciers and which selectively extend across the ice divide. In general it is
preferable to collect coincident radar and altimeter data early in the project.
 Using high elevation lidar and DMS for extending coverage and for repeat elevation
measurements
o Fill regions between low elevation flight lines
o Repeat acquisitions in succeeding years for dh/dt
 Use both the extended and base data sets to construct a spatial and temporal coverage
sufficient for improving ICEsat cross track slope errors, tying ICEsat and OIB data to
the Cryosat data set, connecting all earlier data sets to the ICEsat-2 data set. The
coverage should be sufficient to allow an assessment of regional ice sheet elevation
change shortly after the start of the ICEsat-2 mission.
High elevation coverage strategies are illustrated in figure 5 which shows B-200 LVIS coverage
over Greenland. The notion behind the coverage is to fill in boxes around the ice sheet that form
a baseline elevation data set for bridging ICESat and ICESat-2 observations. Selected ICESat
repeat tracks are also included in the ensemble of observations.
5
Figure 5. B200 LVIS coverage in 2011 (courtesy J. B. Blair and M. Hofton)
6
2.2 Science Contributions
2.2.1 Long term records
Airborne platforms can be used to make measurements at specific points that might be
inaccessible by satellites operating in a fixed orbital geometry. This capability is especially
important for extending the records of earlier in situ observations. Figure 6 shows a 29 year
record of surface elevation change (10 cm/yr increasing rate of surface elevation) for sites in
South Central Greenland and originally established by The Ohio State University. The
combination of Doppler satellite elevations, in situ GPS observations, ATM overflights and slope
corrected ICESat data provide a convincing measure of the trend. Moreover, when combined
with other geophysical data can be used to demonstrate that thickening in this western sector of
the ice sheet is driven by increasing accumulation which falls on a land terminating flow band
(western and central cluster). Thinning east of the divide (Dye-3) is probably driven by changes
in stresses at the ocean terminating terminus.
Figure 6. 29 year record of elevation change in south central Greenland. In situ Doppler satellite
and GPS measurements (red), ATM measurements (blue) and slope corrected ICESat
measurements green at OSU Clusters. Central Cluster site 2001 located at 46.7 W, 65.1 N.
(courtesy K.Jezek)
4.
OIB Phase 3
Key logistical, instrumental, and science elements of a sustainable Phase 2 activity are now in
place. Moreover, additional instruments and valuable aircraft assets have been added to the OIB
portfolio subsequent to Phase 1. An ambitious set of science requirements is in place that is
intrinsically extensible to larger geographic regions with more temporal sampling. With those
developments and with planning experience in hand, the program can turn its attention to several
emerging issues. These are summarized in the following sections.
7
4.1 Phase 3 Measurement Strategy
OIB is meant to provide data to improve our understanding of the mass evolution of the ice
sheets and major mountain glaciers, their vulnerability to climate change and their contribution
to sea-level. In particular, the data will help characterize and understand the processes governing
change. OIB will uniquely improve our understanding in specific targeted local areas at spatial
resolutions that satellite data cannot capture, and to improve the larger spatial scale (but smaller
temporal scale) observations acquired by satellite data (e.g. I-1, I-2 and GRACE and GRACE-2).
Of course, OIB is also collecting data during a gap in satellite laser altimetry.
Therefore OIB during Phase 3 should:




Carefully vet data sets to be sure that the desired geophysical products (for example
bottom topography) are derivable from the ensemble of observations
continue to targets "hot spots" identified by satellite and other data to provide very
detailed spatial sampling for characterization and modeling, and to improve satellite
measurement model algorithms and performance assessment. This would be focused
area mapping or dense grid and flowline mapping of "hot spots", and Icesat line
coverage.
provides large scale spatial sampling to provide a fundamental datum to link past and
future satellite missions. This would be large area gridding, which provides sufficient
sampling for current and future predicted change and to acquire enough collocated
satellite/airborne data.
utilize recently developed tools to automatically create optimized flight lines.
4.2 Increasing spatial and temporal sampling
Aircraft resources envisioned in Phase 1 limit spatial and temporal coverage. These issues have
now been partly mitigated by the good cooperation with other groups including University of
Texas and University of Alaska which provide complementary data sets. However vast areas of
the Antarctic remain inaccessible and temporal sampling of the Arctic is woefully inadequate.
Consequently, there are unresolved issues regarding logistics that must be resolved before we
can extend the reach of OIB. Spatially, this means configuring aircraft for operations to map the
Antarctic grounding line. Achieving this goal requires suitable aircraft along with agreements
between international partners to support extended missions. Temporally, this mandates several
observations over the course of the year are needed to understand short term variability in outlet
glacier elevation in Greenland. Achieving more frequently temporal coverage without overly
stressing instrument and flight crews dictates a transition to unmanned vehicles.
Continued diversification of the available aircraft should be considered to potentially include
vehicles that fly lower and slower. Accurate inversion of gravity data collected over the complex
terrain of Greenland may require denser sampling and more accurate measurements of
8
gravitational acceleration. A key task for the science team will be to assess the current data set
and to make recommendations on necessary characteristics for future platforms.
4.3 Advanced Technologies
OIB should be flexible to exploit advanced technologies such as radar tomography. This
approach is now demonstrated to yield coincident cross track estimates of surface and basal
topography and radar reflectivity. These are deemed critical for detailed studies of select
glaciers (figure). Here instruments and mission profiles have to be optimized for the more site
specific applications of this revolutionary technique.
Figure 7. Tomographically derived surface and basal topography (left) for Russel Glacier
(right). Ice thickness is about 1000 m. (Courtesy of X. Wu and J. Paden).
Similarly, OIB should be quick to integrate technologies. For example, sub-ice-shelf bathymetry
from gravity (Figure 8) can be merged with radar ice thickness on grounded ice to provide a
seamless view of the subglacial terrain. Gravity and radar may also provide important
complementary information where the physical situation proves challenging for one or the other
of the measurement techniques.
9
Figure 8. Gravity derived, sub-glacial bathymetry for Pine Island Glacier. Image courtesy of J.
Cochran, E. Rignot, and M. Studinger (Fogt, 2011).
4.2 An end-to-end view of OIB
To answer the science questions posed at the outset, OIB must evolve to an end to end system
wherein data planning, data collection, and analysis flow smoothly into models with a quick (6
month) turn around. This evolution is important for two reasons. Establishing the flow ensures
that the science team is making good choices for future aircraft deployments along with assuring
that the collected data are vetted for scientific utility. As importantly, this mission-oriented flow
will mean that OIB focuses attention on answering key science questions. This last point is
fundamentally important. Said differently, OIB will be most successful when viewed as a
mission with all of its resources working in a coordinated way to solve key questions. While
individual, exploratory research will remain an element of OIB, the more encompassing
approach to OIB research will become the dominant theme.
10
5.
OIB Phase 4
The last OIB phase has several goals which primarily focus on the hand-off to ICESat-2. Prior to
ICESat-2 launch and once the orbit geometry has been fixed, OIB will conduct measurement
campaigns that optimize coverage beneath ICESat-2 measurement footprints. Along with
establishing a data set to vet ICESat-2 elevations close to the launch time frame, the data will be
crucial for validating ICESat-2 multibeam slope corrections.
After launch, OIB will continue underflights to establish an overlap data set sufficient to connect
the ICESat, OIB, and ICESat-2 data sets. Work will be required in the Arctic and Antarctic
suggesting about a 1 year period of joint operations.
Once the overlap period is over, there is a strong case to be made for maintaining a modified
OIB operation based on the complementary characteristics of airborne and spaceborne data
instruments. These include:
1) The navigational flexibility of airborne instruments is crucial for maintaining the most
accurate time series over fixed sites and for spatially interpolating satellite data. Aircraft provide
an invaluable tool for refining trends and understanding processes. Even in cases where the early
observations may have large errors, differences in a geophysical parameter increased with time
may well outweigh the instrumental uncertainties. Future remeasurements of Antarctic traverse
lines from the 1950’s and 60’s may soon prove fruitful.
2) Combinations of in situ, airborne and satellite data greatly increase the interpretability of the
data sets and increase the confidence in the interpretation of those data. Multiple, independent
estimates of a parameter (for example free air gravity and altimetry) also increase confidence.
3) A rich suite of measurements is required to capture the complete ensemble of glaciological
parameters necessary to interpret the behavior of the ice sheet. Presently, there are several
essential measurements that can only be made from aircraft including subglacial topography and
ocean bathymetry as well as near surface accumulation rates.
4) Airborne measurements provide a unique flexibility to acquire data over short term or
unexpected phenomena. Figure 9 illustrates a recently acquired and initially unplanned set of
measurements across a crack developing in the Pine Island Glacier. Data of this sort are
invaluable in understanding crack formation and assessing the likelyhood that the phenomena is
a routine, episodic calving event or a harbinger of more drastic ice shelf retreat.
11
Figure 9. Satellite and OIB data showing a developing rift on Pine Island Glacier, October,
2011. Upper image is a TerraSAR-X image (courtesy Dana Floricioiu, DLR). The middle scene
shows DMS details of the rift. The lower scene shows ATM surface elevations along the rift.
Depth sounding radar data were also acquired along and across the rift. (composite courtesy of
M. Studinger).
6.
References
Fogt, R. L., Ed., 2011: Antarctica in State of the Climate in 2010, Bull. Amer. Meteor. Soc., 92
(6), S161-171.
Koenig, L., S. Martin, M. Studinger, and J. Sonntag (2010), Polar airborne observations fill gap
in satellite data, Eos Trans. AGU, 91(38), 333–334.
12
13