of 3
Volume 92
number 7
15 FebruArY 2011
pAges 53–60
Eos
, Vol. 92, No. 15, 12 April 2011
Volume 92
number 15
12 AprIl 2011
pAges 125–132
The Global Positioning System (GPS) is
an example of a Global Navigation Satellite
System (GNSS) that provides an essential
complement to other geophysical networks
because of its high precision, sensitivity to
the longest-period bands, ease of deploy
-
ment, and ability to measure displacement
and atmospheric properties over local to
global scales. Recent and ongoing tech
-
nical advances, combined with decreas
-
ing equipment and data acquisition costs,
portend rapid increases in accessibility of
data from expanding global geodetic net
-
works. Scientists and the public are begin
-
ning to have access to these high-rate, con
-
tinuous data streams and event-specific
information within seconds to minutes
rather than days to months. These data
provide the opportunity to observe Earth
system processes with greater accuracy
and detail, as they occur.
What Is “Real-Time” GPS?
GPS is a satellite system that broad
-
casts signals toward the Earth, allowing a
receiver to solve for its position when three
or more spacecraft are in view. It was origi
-
nally designed to provide accurate posi
-
tioning, navigation, and time anywhere on
Earth within seconds. Systems and analy
-
sis improvements in the 1980s and 1990s
increased the accuracy of positioning from
meters to centimeters, but such precision
usually required continuous data collection
for 24-hour or longer periods before mea
-
surements were produced. Solutions esti
-
mated more frequently than daily were con
-
sidered high rate.
More recently, position solution accu
-
racy and speed have advanced to the point
where centimeter-
precision coordinates
are available within seconds, and millime
-
ter precision is available for daily solutions.
Relative motions between stations thou
-
sands of kilometers apart can be resolved.
For some applications, position solution
rates of 100 samples per second have been
demonstrated. With the evolution of obser
-
vation technologies in mind, members of
the EarthScope Plate Boundary Observa
-
tory advisory committee recently authored
a white paper (see http://
unavco
.org/
research
_ science/
science
_ highlights/
2010/
RealTimeGPSWhitePaper2010
.pdf) that dis
-
cusses the expected future scientific value
of real-time GNSS information, particularly
with regard to GNSS positions that arrive
with high rate (e.g., 1 hertz or higher) and
low latency (e.g., seconds or less). Such
measurements are termed real-time GPS
(RTGPS).
The principal scientific benefit of RTGPS
data is realized when high-rate informa
-
tion improves temporal resolution in obser
-
vations of natural processes. RTGPS likely
will demonstrate an impact similar to that
of other high-
rate geophysical observations
(e.g., from seismological and meteorological
networks) for monitoring and understanding
earthquakes, seismic wave propagation, vol
-
canic eruptions, magmatic intrusions, move
-
ments of ice, landslides, and structure and
dynamics of the atmosphere. In many cases
the availability of low-latency data will sub
-
stantially enhance the processes and out
-
comes of the research itself. For example,
low latency ensures that high-rate data are
reliably transmitted to laboratories until the
moment catastrophic events destroy instru
-
ments or disable transmission lines. Imme
-
diate delivery can save precious near-field
measurements of the largest displacements
or atmospheric effects. Use of low-latency
data will enhance rapid scientific response
by improving targeting, by activation of new
data streams, or by changing instrument set
-
tings based on early results.
The availability of RTGPS information
will also have important impacts on how
scientists and societies prepare for and
cope with natural disasters. As a rule of
history, mitigating the effects of natural
disasters such as earthquakes, tsunamis,
Scientific Value of Real-Time
Global Positioning System Data
B
y
W. C. H
ammond
, B. a
. B
rooks
, r
. B
ürg
-
mann
, T. H
ea
Ton
, m
. J
aCkson
,
a. r
. L
oWry
,
and
s . a
nandakris
Hnan
PAGES 125 –126
Fig. 1. (a) EarthScope Plate Boundary Observatory (PBO) sites (red dots) plus other Global
Positioning System (GPS) sites (blue dots) in the western United States that have been, or soon
will be, upgraded to real-time streaming capabilities. (b) Global distribution of real-time GPS
sites of the Global Differential GPS (GDGPS) network (from http://
www
. gdgps
.net/, courtesy
of Y. Bar-Sever). (c) Time series of 5-hertz displacement, collected 70 kilo
meters from the 4 April
2010 El
Mayor- Cucapah
M
7.2 earthquake epicenter in Baja California (courtesy of K. Larson).
Coseismic offset of approximately 20 centimeters is clearly resolved in the north component
40 seconds after shaking begins.
Eos
, Vol. 92, No. 15, 12 April 2011
volcanic eruptions, and landslides requires
knowledge of the underlying Earth science
coupled with information about specific
events delivered and updated as quickly
as possible. Rapid detection and accu
-
rate characterization of events can make
a crucial difference during the minutes to
hours that follow. This point was clearly
made following disasters that occurred
during the past decade, including the cata
-
strophic 2004 Sumatra and 11 March 2011
Tohoku-oki earthquakes and tsunamis. In
both of these events, the initial seismic
notification of earthquake magnitude was
available in minutes but was more than an
order of magnitude smaller than the true
event size. GPS systems can address this
uncertainty in the rapid estimation of large
earthquake magnitudes.
Presently, about 240 GPS stations of the
EarthScope Plate Boundary Observatory
in the vicinity of the Cascadia subduction
zone are being upgraded to provide GPS
observations at 1 hertz and faster than
0.3-second latency. The NASA Jet Propul
-
sion Laboratory provides streams of posi
-
tions at 1 hertz for more than 120 glob
-
ally distributed stations with 5-second or
faster latency via its Global Differential
GPS system. The California Real Time Net
-
work (CRTN) at the Scripps Institution
of Oceanography distributes data and
positions from more than 150 stations in
California. Other sites and networks are
also being upgraded through various ini
-
tiatives (Figure 1). In light of these and
other developments, broad new realms of
processes can be studied with RTGPS.
Seismic Source, Event Characterization,
and Warning
Integration of RTGPS with seismic time
series will push forward the science of
broadest-
band seismology, meaning that
studies of “seismic” sources can be increas
-
ingly viewed as studies of Earth deforma
-
tion events that occur over a very wide
range of time scales, some of which may
not necessarily generate seismic waves.
The inclusion of RTGPS to extend mea
-
surements beyond typical seismic fre
-
quencies is essential to understanding the
complete spectrum of fault slip behaviors
associated with the earthquake cycle. For
instance, GPS-
measured static displace
-
ments and waveforms (Figure 1c) can pro
-
duce improved and rapidly available mod
-
els of earthquake slip, surface deforma
-
tion, and strong ground motion [e.g.,
Rolan
-
done et al.,
2006]. Such data are invalu
-
able for first-response efforts that require
knowledge of the areas of strongest ground
motion and surface rupture.
To supplement current seismic early warn
-
ing systems, RTGPS data will play a vital role
in early warning for large events with long
ruptures [
Böse and Heaton,
2010]. For these
events, an integrated RTGPS could provide
information about an ongoing earthquake
quickly enough to make a prediction of
shaking before it occurs. This type of noti
-
fication could give people a few seconds to
minutes to prepare themselves, machinery,
and infrastructure for shaking.
Tsunami Event Characterization
and Warning
Tsunami warning has particular require
-
ments for calculating accurate earthquake
magnitude, propagation direction, and ver
-
tical and horizontal motion of the seafloor.
The goal is to rapidly recognize that a tsu
-
nami event is occurring and improve pre
-
dictions of where the wave will rise on near
and distant coasts. Displacements at GPS
sites are used to constrain a fault slip model,
which predicts motion of the seafloor. An
adequate near-field network is preferred to
constrain the slip distribution. Had a GPS-
based warning system been in place for
the 2004 Sumatra earthquake, a more accu
-
rate magnitude could have been estimated
within 15 minutes [
Vigny et al.,
2005;
Blewitt
et al.,
20 0 6].
During the 11 March Tohoku-oki event in
Japan, a prototype GPS-based warning sys
-
tem, the NASA GPS Real-Time Earthquake
and Tsunami (GREAT) Alert, was running in
test mode. The project team is now evaluat
-
ing performance of that system, so details
are not yet available as this article goes to
press. However, once they are operational,
systems such as GREAT should help to shave
minutes off the time needed to recognize
what has occurred, thus supporting quicker
warnings that can save lives. The GREAT
system is currently being evaluated by the
Pacific Tsunami Warning Center as a tool
to complement existing tsunami warning
systems.
Volcanic and Magmatic Events
Volcanic activities, magma chamber
inflation/
deflation, dike intrusions, and effu
-
sive and/or explosive eruptions often pro
-
duce measurable surface deformations.
These deformations provide information
about processes inside magmatic plumb
-
ing systems and can vary rapidly in space
and time. Because these deformations can
precede hazardous eruptions by hours to
months, telemetered GPS networks com
-
bined with low-latency processing strategies
are in operational use in volcano observa
-
tories in the United States (Hawaii, the Cas
-
cades, and Yellowstone) as well as Japan
and Italy.
Additional hazard arises when steep
slopes of island volcanoes fail catastrophi
-
cally and generate a tsunami. Slope failures
on some island volcanoes involve poorly
understood transitions from slow-slip events
to abrupt failures. For example, a flank-
related slow-slip event at Kilauea was likely
triggered roughly 15–20 hours after a dike
intrusion in the east rift zone stressed the
flank [
Brooks et al.,
2008]. In these situa
-
tions, RTGPS networks can monitor motions
preceding potentially catastrophic events.
Cryosphere
In just the past few years, GPS has had a
remarkable impact on ongoing research relat
-
ing to glacier volume, flow, and evolution,
leading to improvements in measurements
of flow velocities, rates of surface snowfall,
and isostatic adjustment associated with gla
-
cial mass change. High-rate monitoring of the
cryosphere has had a transformational effect
on scientific understanding of dynamic gla
-
ciology. These measurements have shown
that glacier flow can change speed and direc
-
tion on time scales that were once thought
impossible: seasonal, fortnightly, daily, and
even subhourly [
Nettles et al.,
2008]. The pro
-
cesses associated with these changes are
poorly understood and not included in cur
-
rent models of ice sheet flow, resulting in poor
estimates of future glacial contributions to sea
level. RTGPS can contribute to a better under
-
standing of sea level by allowing researchers
to collect and analyze glacier flow data along
with conventional ocean and atmospheric
data.
Tropospheric Modeling
RTGPS measurements have the poten
-
tial to contribute to climate modeling and
weather forecasting through integrative mea
-
surement of atmospheric water vapor in
GPS signal delays and measurements of soil
moisture flux. First, microwave frequencies
used in GPS are particularly sensitive to the
presence of water vapor, and much effort
has been devoted to estimation of water
vapor along GPS signal propagation paths
[e.g.,
Braun et al.,
2001]. GPS water vapor
data will become more useful for weather
and climate applications as RTGPS net
-
works provide the data with low latency and
high reliability. Second, researchers have
extracted measurements of snow depth and
soil moisture from multipath interference
in the environment around GPS antennas
[
Larson et al.,
2008, 2009]. Measurements of
local moisture obtained with high-rate sam
-
pling could contribute to larger-scale quanti
-
fication of water fluxes.
Space Weather
Virtually all aspects of ionospheric
research use GPS observations, primarily
through measurements of total electron con
-
tent from the differential delays of two signal
frequencies. Higher sampling rates of RTGPS
will benefit studies of traveling ionospheric
disturbances and other wave phenomena,
including disturbances from earthquakes
and tsunamis, while lower latency will aid
in the development of operational forecast
-
ing for space weather, with significant impli
-
cations for global communications systems
and satellite maintenance.
Future Directions
Through rapid and widespread adop
-
tion of RTGPS, geodetic information will
Eos
, Vol. 92, No. 15, 12 April 2011
soon have latency, availability, and public
impact similar to meteorological and seis
-
mic information. This requires overcom
-
ing specific technical challenges, such
as improving data networks and develop
-
ing faster algorithms for coping with data
streams. It will also be necessary to over
-
come cultural challenges that inhibit the
exploration of the overlap of geodesy, seis
-
mology, and cryospheric and atmospheric
science. Despite these obstacles, GNSS
geodesy will experience a rapid evolution
as various communities critically evaluate
and use these data for research purposes,
leading to the development of accessible
and actionable public information prod
-
ucts. Ultimately, this evolution is essential
for improving understanding of high-impact
Earth system processes and for increasing
public engagement.
References
Blewitt, G., C. Kreemer, W. C. Hammond, H.-P. Plag,
S. Stein, and E. Okal (2006), Rapid determination
of earthquake magnitude using GPS for tsunami
warning systems,
Geophys. Res. Lett., 33,
L113 0 9,
doi:10.1029/
2006GL026145.
Böse, M., and T. H. Heaton (2010), Probabilistic
prediction of rupture length, slip and seismic
ground motions for an ongoing rupture: Implica
-
tions for early warning for large earthquakes,
Geophys. J. Int., 183
(2), 1014–1030, doi:10.1111/
j.1365-246X.2010.04774.x.
Braun, J., C. Rocken, and R. Ware (2001), Valida
-
tion of line-of-sight water vapor measurements
with GPS,
Radio Sci., 36
(3), 459–472, doi:10.1029/
2000RS002353.
Brooks, B. A., J. Foster, D. Sandwell, C. J. Wolfe,
P. Okubo, M. Poland, and D. Myer (2008),
Magmatically triggered slow slip at Kilauea
volcano, Hawaii,
Science, 321,
1177, doi:10.1126/
s c i e n c e .115 9 0 0 7.
Larson, K. M., E. E. Small, E. D. Gutmann, A. L.
Bilich, J. J. Braun, and V. U. Zavorotny (2008),
Use of GPS receivers as a soil moisture network
for water cycle studies,
Geophys. Res. Lett., 35,
L24405, doi:10.1029/
2008GL036013.
Larson, K. M., E. D. Gutmann, V. U. Zavorotny, J. J.
Braun, M. W. Williams, and F. G. Nievinski (2009),
Can we measure snow depth with GPS receiv
-
er s?,
Geophys. Res. Lett., 36,
L17502, doi:10.1029/
2009GL039430.
Nettles, M., et al. (2008), Step-wise changes in
glacier flow speed coincide with calving and
glacial earthquakes at Helheim Glacier, Green
-
land,
Geophys. Res. Lett., 35,
L24503, doi:10.1029/
2008GL036127.
Rolandone, F., D. Dreger, M. Murray, and R. Bürg
-
mann (2006), Coseismic slip distribution of the
2003
M
w
6.5 San Simeon earthquake, Califor
-
nia, determined from GPS measurements and
seismic waveform data,
Geophys. Res. Lett., 33,
L16315, doi:10.1029/
2006GL027079.
Vigny, C., et al., (2005), Insight into the 2004
Sumatra-
Andaman earthquake from GPS
measurements in southeast Asia,
Nature, 436,
201–206, doi:10.1038/
nature03937.
Author Information
William C. Hammond, Nevada Bureau of Mines
and Geology, University of Nevada, Reno; E-mail:
whammond@ unr
.edu; Benjamin A. Brooks, School
of Ocean and Earth Science and Technology, Uni-
versity of Hawai`i at Manoa, Honolulu; Roland Bürg-
mann, Department of Earth and Planetary Science,
University of California, Berkeley; Thomas Heaton,
Earthquake Engineering Research Laboratory, Cali-
fornia Institute of Technology, Pasadena; Michael
Jackson, UNAVCO, Inc., Boulder, Colo.; Anthony R.
Lowry, Department of Geology, Utah State University,
Logan; and Sridhar Anandakrishnan, Department of
Geosciences, Pennsylvania State University, Univer
-
sity Park
As Arctic temperatures increase, there is
growing concern about the melting of the
Greenland ice sheet, which reached a new
record during the summer of 2010. Under
-
standing the changing surface mass bal
-
ance of the Greenland ice sheet requires
appreciation of the close links among
changes in surface air temperature, sur
-
face melting, albedo, and snow accumula
-
tion. Increased melting accelerates surface
snow grain growth, leading to a decrease
in surface albedo, which then fosters fur
-
ther melt. In turn, winter accumulation
contributes to determining how much
snow is required before a dark (e.g., lower
albedo), bare ice surface is exposed in
spring (Figure 1).
A recent analysis of surface glaciologi
-
cal observations, remote sensing data, and
model output for 2010 points to new records
for surface melt, albedo, runoff, and the
number of days when bare ice was exposed
[
Tedesco et al.,
2011]. Large areas of the abla
-
tion zone in southern Greenland underwent
melting for up to 50 days longer than the
1979–2009 average of 48 days of melting per
year.
Warm conditions persisted through spring
and summer, with the positive albedo feed
-
back mechanism playing a major role in fur
-
ther enhancing melting and in leading to
large negative surface mass balance anoma
-
lies. Summer snowfall was below average,
maintaining the low albedo throughout the
melting season. Melt during August and Sep
-
tember was also exceptional, consistent
with low surface albedos and near-surface
temperature anomalies of up to +3ºC, yield
-
ing a long ablation season.
The unusually warm conditions over
the Greenland ice sheet in 2010 and
the reduced snowfall can partially be
explained by a negative North Atlantic
Oscillation index phase from late spring
through summer.
Monitoring the Greenland ice sheet
through modeling and observational tools
is crucial to understanding how recently
observed increasing surface temperatures
over the Arctic region are affecting sur
-
face mass balance and how this, in turn,
Record Summer Melt in Greenland in 2010
NEWS
Fig. 1. Surface temperatures up to 3°C higher than average during summer 2010 generated
earlier melting. When snow melts, grains tend to cluster (a process known as constructive meta-
morphism). This reduces the albedo, leading to increased absorbed solar radiation. New snow
can increase albedo again, but accumulation was low in 2010. Bare ice, on the contrary, is much
“darker” than snow, absorbing even more solar radiation and further increasing melting. Red
arrows indicate incoming radiation; blue arrows indicate outgoing radiation. Percentages indi-
cate the amount of reflected incoming solar radiation.
PAGE 126