3
Geodesy Requirements for Earth Science

Precise geodetic infrastructure enables ground- and space-based observations that are critical to a wide array of scientific disciplines, including seismology, geodynamics, climate science, hydrology, oceanography, meteorology, and space weather. Geodetic observations, for example, allow us to measure and monitor gradual changes in tectonic plate movement, sea level rise, glacial ice melting, and aquifer depletion. Similarly, the geodetic infrastructure provides the foundation for numerous applications with broad societal and commercial impact, from early warning systems for hazards to intelligent transportation systems (Figure 3.1). Over time, the precision and timeliness of these applications has improved, with operational applications now routinely working at accuracies only recently achievable, while the scientific applications are approaching the part-per-billion accuracy level and near-real-time operations. The scientific questions that are being asked about how the Earth system works and what can be predicted for the future continue to drive ever more stringent requirements for geodesy.

The demand for geodetic observations will continue to grow as we move from global to regional forecasts of climate change impacts, from manual to autonomous systems that require more precise positioning in real time, and from point positioning to geodetic imaging. Given the breadth of scientific and societal applications, this chapter is not intended to be comprehensive, but rather to highlight only a sampling of the benefits that high-precision geodetic infrastructure provides for Earth science and the nation.

SOLID EARTH DYNAMICS

Geodynamics

Plate Motion and Tectonics

Nothing on Earth’s surface is fixed, and enormous pieces of the crust are being ripped apart or forced into collisions with each other by the movement of the mantle below, causing earthquakes, volcanoes, and mountain building. Earth’s crust is currently considered to consist of approximately



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 37
3 Geodesy Requirements for Earth Science Precise geodetic infrastructure enables ground- and space-based observations that are critical to a wide array of scientific disciplines, including seismology, geodynamics, climate science, hydrol - ogy, oceanography, meteorology, and space weather. Geodetic observations, for example, allow us to measure and monitor gradual changes in tectonic plate movement, sea level rise, glacial ice melting, and aquifer depletion. Similarly, the geodetic infrastructure provides the foundation for numerous applications with broad societal and commercial impact, from early warning systems for hazards to intelligent transportation systems (Figure 3.1). Over time, the precision and timeliness of these applications has improved, with operational applications now routinely working at accura - cies only recently achievable, while the scientific applications are approaching the part-per-billion accuracy level and near-real-time operations. The scientific questions that are being asked about how the Earth system works and what can be predicted for the future continue to drive ever more stringent requirements for geodesy. The demand for geodetic observations will continue to grow as we move from global to regional forecasts of climate change impacts, from manual to autonomous systems that require more pre - cise positioning in real time, and from point positioning to geodetic imaging. Given the breadth of scientific and societal applications, this chapter is not intended to be comprehensive, but rather to highlight only a sampling of the benefits that high-precision geodetic infrastructure provides for Earth science and the nation. SOLID EARTH DyNAMICS Geodynamics Plate Motion and Tectonics Nothing on Earth’s surface is fixed, and enormous pieces of the crust are being ripped apart or forced into collisions with each other by the movement of the mantle below, causing earthquakes, volcanoes, and mountain building. Earth’s crust is currently considered to consist of approximately 

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE FIGuRE 3.1 This schematic plots the precision of current geodetic applications as a function of the required time interval. The most demanding applications at the shortest time intervals include GNSS/GPS seismology and tsunami warning systems. At the longest time intervals, the most demanding applications include sea level change and geodynamics. Note that the positioning scale is in powers of 10 and that range of geodetic applica- tions spans approximately nine orders of magnitude in the time scale. Consistency in connecting the longest to the shortest time scales requires an accurate and stable global terrestrial reference frame, which drives the most stringent requirements on the geodetic infrastructure. * Plate motion, plate deformation, mountain building, mass transport, ice-sheet changes (using loading mo - tion and gravity changes observed from space). ** Vertical surface motion from GNSS/GPS and InSAR for ground water management; water redistribution is monitored from space based on gravity measurements). *** Water vapor and other meteorological information from GNSS/GPS ground stations and radio occulta - tions in space. 52 “rigid” plates (14 major plates and approximately 38 minor plates) that slowly drift across the surface of the planet, changing speed and direction on million-year timescales (McKenzie and Parker, 1967). The edges of the plates undergo a variety of non-rigid and unsteady motions, which are classified according to the direction of relative plate motion across the boundary. Divergent boundaries form the mid-ocean ridges that encircle the planet like seams on a baseball. Here, plates spread apart and the void is filled from below by hot material. Convergent plate boundaries form the deep ocean trenches where the cooled plates subduct back into Earth’s mantle. The largest earth - quakes occur when these subducting plates slip past each other after sticking together for a period of 300–1,000 years (stick-slip behavior). Transform boundaries (which cause strike-slip motion) mainly occur in the deep oceans, although they occasionally cut across the continental areas. A

OCR for page 37
9 GEODESY REquiREMENTS FOR EARTh SCiENCE prominent example of a transform boundary is the San Andreas Fault, which undergoes strike-slip motion on 100–1,000 year time scales, resulting in destructive earthquakes. Mountainous areas on the continents, such as the Himalayas and the Alps, are formed by convergent motion (collisions) between continental plates. However, where an oceanic plate collides with a continental plate, such as the North American Cascades, major earthquakes and volcanism can be expected. Global-scale geodetic measurements of plate motions from Very Long Baseline Interferometry (VLBI) and Satellite Laser Ranging (SLR), based on less than a decade of data, show remarkable agreement (at the 3–5 millimeters per year level) with plate motions derived from the 1–3 million year average rates derived form the geological and geophysical data (Herring et al., 1986; Carter and Robertson, 1986). Some tectonic plate studies, however, drive a requirement for a higher level of accuracy. For example, the boundaries of the plates have narrow regions where shorter timescale plate-to-plate interactions are important. These include areas of high crustal strain (Figure 3.2), which result in destructive earthquakes and volcanoes. In addition, the plates do not behave exactly rigidly, and the measure of horizontal intraplate deformation could be associated with the thermal contraction of the cooling oceanic lithosphere. This plate shrinkage has recently been detected at the 3 millimeter per year level (Kumar and Gordon, 2009). Post-Glacial Rebound During the last Ice Age, vast ice sheets up to 4–5 kilometers thick lay over the Hudson Bay region in northern Canada and across much of Scandinavia. The amount of ice locked up in the ice sheets at the time was enough to cause global sea levels to lie 100–150 meters below their pres - ent levels (Peltier, 2004). The pressure from that ice load on Earth’s crust caused the underlying mantle to be depressed. When the ice melted, starting roughly 20,000 years ago and continuing until approximately 10,000 years ago for Canada and Scandinavia, Earth began to rebound. That rebound (also known as glacial isostatic adjustment) continues today because Earth is viscous, and it takes time for a viscous body to fully respond to the removal of a load. Observations of the rebound rate provide information about Earth’s viscosity profile, which plays a key role in determining the pattern and vigor of convection in Earth’s mantle that drives plate motion and causes earthquakes FIGuRE 3.2 Geologic, geodetic, and earthquake data help determine the zones on Earth where the crustal motion diverges from rigid plate motion. Areas of high strain, in red, experience increased earthquakes and volcanoes. SOURCE: Kreemer et al., 2000.

OCR for page 37
0 PRECiSE GEODETiC iNFRASTRuCTuRE and volcanic activity. The characteristics of the ancient ice sheets are also of interest because they provide insight into how the cryosphere has responded to dramatic changes in climate in the past. It is only through geodetic observations that the present-day rebound can be observed as it actu - ally happens. There are two types of relevant geodetic observations. One involves the measurement of surface uplift rates at points near the locations of the ancient ice sheets. Measurements of this type are now usually made with GNSS/GPS (see Lidberg et al., 2007; Sella et al., 2007), though VLBI observations have been used as well (Heki, 1996; James and Morgan, 1990; Mitrovica et al, 1993). The other type of geodetic observation is time-variable gravity. The ongoing rebound leads to changes in Earth’s mass distribution, which in turn cause Earth’s gravity field to evolve over time. The gravity field over northern Canada, for example, is steadily increasing in strength as mass in the underlying mantle flows in sideways from surrounding regions, pushing Earth’s surface upwards. Those gravity changes are best seen in satellite gravity data, such as from LAGEOS (Cox and Chao, 2002; Cheng and Tapley, 2004) and, especially, GRACE (see Box 3.1) (Paulson et al., 2007; Tamisiea et al., 2007). These space-based gravity measurement techniques use the geodetic infrastructure to determine an accurate reference against which to measure the small post-glacial rebound motions; the geodetic techniques also provide the data containing important geophysical signals. Earth Orientation (Length of Day, Polar Motion, and Nutation) The direction of Earth’s rotation axis and the rate of rotation about that axis vary with time (for general reviews, see Dehant, 2007; Gross, 2007; Lambeck, 1980). As described in Chapter 1, the change in the rate of rotation causes a small change in the length of day. The motion of the rotation axis itself is described by polar motion and nutation. Polar motion refers to motion of the axis relative to fixed points on Earth’s surface, while nutation refers to motion with respect to fixed objects in space. For the past several decades, the determination of Earth’s orientation has been based on observations from the global network of VLBI, GNSS/GPS, and SLR stations, and the results are critically dependent on reference frame accuracy. The orientation and rotation rate of the reference frame used when analyzing these geodetic measurements are directly related to the rotational parameters used for geophysical interpretation. Consequently, reference frame errors can map directly into errors in that interpretation. Earth Tides The tidal force from the sun and moon causes tides in the solid Earth, just as it does in the ocean. The periods are the same, and even the amplitudes are similar: tidal displacements in both the solid Earth and the open ocean are on the order of several tens of centimeters. Solid Earth tides are best observed using geodetic measurement techniques. Information on tidal deformation of Earth at global-scale wavelengths provides unique information on Earth’s structural parameters. Natural Hazards Volcanoes There are 170 volcanoes in the United States, of which at least 65 are active or potentially active. The U.S. Geological Survey (USGS) Volcano Hazards Program routinely monitors volcanoes using a variety of methods designed to detect and measure changes caused by the underground move - ment of magma (molten rock). Rising magma typically will: trigger earthquakes and other seismic events; cause swelling or subsidence of a volcano’s summit or flanks; and lead to the release of

OCR for page 37
1 GEODESY REquiREMENTS FOR EARTh SCiENCE BOX 3.1 The Gravity Recovery and Climate Experiment (GRACE) The Gravity Recovery and Climate Experiment (GRACE), a joint NASA/Deutsches Zentrum für Luft- und Raumfahrt (DLR) mission to map the time-variable and mean gravity field of Earth, was launched on March 17, 2002 (Tapley et al., 2004a,b). Using an extremely high-precision inter-satellite ranging system, subtle variations in Earth’s gravity field are detected through changes in the distance between the two satellites (since each satellite detects slightly different gravity effects due to their separation of approximately 200 kilometers). This has made it possible to observe seasonal, long-term, and climate- driven changes in Earth’s mass distribution at a resolution of several hundred kilometers. In addition, knowledge of the static component of Earth’s gravity field has been improved by orders of magnitude, providing a global geoid model accurate to the centimeter level at 200-kilometer resolution. The twin GRACE satellites using a dual one-way K-band (20–40 GHz) ranging system to observe changes in their relative distance to the few micron level. The scale is for illustration purposes only; the actual separation distance is kept close to 200 kilometers. SOURCE: John C. Ries, The University of Texas Center for Space Research. volcanic gases from the ground and vents. By monitoring these phenomena, scientists sometimes can anticipate an eruption days to weeks in advance or remotely detect explosive eruptions and lahars (a mixture of water and rock fragments that flow down the slopes of a volcano). Successfully monitoring and forecasting Mount Pinatubo’s cataclysmic eruption in 1991, for example, prevented property losses of more than $250 million (Newhall et al., 2005). GNSS/GPS receivers and Interferometric Synthetic Aperture Radar (InSAR) are the main geodetic tools used for monitoring volcanic surface deformation, as well as earthquake deforma - tion (discussed below). GNSS/GPS receivers can be set up at strategic locations around a site to stream measurements continuously. InSAR, on the other hand, can uniquely map and resolve sur- face deformation over a wide range of spatial scales that is not possible with GNSS/GPS (see Fig - ure 3.3). The research community in the United States currently relies on InSAR data collected by several radar satellite missions, including those flown by the European (ERS-2, Envisat), Canadian (Radarsat-1) and Japanese (ALOS) space agencies (see Table 4.1). Combining the high-spatial- resolution InSAR map with high-temporal-resolution GNSS/GPS point measurements provides the full four-dimensional picture of volcanic deformation. An unexpected result of using InSAR and GNSS/GPS systematically to study volcanoes in Alaska and the western United States is the

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE FIGuRE 3.3 An InSAR-defined area of uplift near the Three Sisters cluster of volcanoes in central Oregon, where each concentric circle of red corresponds to approximately 28 millimeters in deformation. This deforma- tion, which does not lie directly beneath any volcano, is in an area where the most recent eruption occurred 1,500 years ago. Uplift of the ground’s surface, which began in 1997, reached 15 centimeters at the center of the “bull’s eye” pattern in 2001. Subsequent GPS monitoring shows that uplift continues at a steady pace, suggesting that it is produced by upward movement of magma (intrusion). The InSAR pattern places the depth of intrusion at 6–7 kilometers. SOURCE: Wicks et al., 2002. Courtesy of the American Geophysical Union. discovery that volcanoes once thought to be dormant are actually undergoing gradual deformation and may eventually erupt. Because volcanic deformation tends to be a slow process taking decades to centuries between eruptions, InSAR and GNSS/GPS measurements must be tied to decades of data from a stable terrestrial reference frame with better than 5 millimeter precision. Earthquakes The sudden release of energy along a major earthquake fault is one of the most destructive forces of nature. Predicting earthquakes depends on our ability to understand the earthquake cycle, which requires geodetic measurements of Earth’s crustal motion between seismic events. At depths greater than 20 kilometers, the North American and Pacific plates slide freely past one another. Shallow fault zones, however, are colder and more brittle and undergo stick-slip behavior, in which the shallow surfaces of the fault remain locked for periods sometimes lasting hundreds of years because of friction and surface imperfections. Eventually, the tectonic stress exceeds the fault strength, causing the plates to slip past each other (a coseismic event). Recently, scientists discovered that slips can be either fast and destructive or slow and minimal. An example of short-term transient slow slip behavior at a subduction zone is the phenomenon of Episodic Tremor and Slip (ETS) (Rogers and Dragert, 2003; see Figure 3.4). ETS, as observed along the northern Cascadia Subduction Zone, has been defined empirically as repeated, transient ground motions at a plate margin. Prior to the discovery of ETS, geologists thought the Cascadia coastal margin was squeezed landward in a continuous, steady fashion. It is inferred that the deeper plate interface also undergoes a stick-slip behavior, but over a much shorter time-scale than the earthquake cycle. The relationship between ETS and regional earthquakes is not yet clear, but it is conceivable that an ETS episode could ultimately trigger a large earthquake.

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE FIGuRE 3.4 GNSS/GPS and seismic data as observed on the Cascadia Subduction Zone. The plot dem - onstrates an example of Episodic Tremor and Slip (Rogers and Dragert, 2003), a phenomenon of short-term transient slow slip behavior at a subduction zone. Each blue circle in the plot indicates the daily change of the east-west position of the Victoria GPS station relative to the interior of the North American Plate. The green line shows the long-term linear motion over the 14-year period. The red line shows that for roughly 15 months this deeper fault zone resisted slip but then slipped several centimeters over a period of weeks, resulting in a characteristic sloped saw-tooth time series for the longitude component of coastal GPS stations. SOURCE: Rogers and Dragert, 2003. Fast ruptures generate elastic waves that propagate outward and destroy buildings. Slow rup - tures, which could be detected only recently using GNSS/GPS receivers, can release the tectonic stress over periods of several days to weeks. Following an earthquake, the fault continues to slip, generating aftershocks. This postseismic deformation can last for tens of years. Some scientists also hypothesize that there is a short period of concentrated deformation just prior to a major earthquake, although this period of preseismic deformation is poorly documented. Modern space-based geodetic measurements, such as GNSS/GPS and InSAR, have recorded all but the preseismic elements of the earthquake cycle. With the new tools geodesy offers, however, scientists are beginning to understand the earthquake process and may someday be able to provide useful earthquake forecasts. Accurate and robust measurements of subtle secular and precursory deformation are the new frontiers in crustal deformation studies and are pivotal for natural hazards research. These measure - ments can be accomplished with GNSS/GPS, InSAR and Aircraft Laser Mapping (ALM). The Plate Boundary Observatory, part of the National Science Foundation’s EarthScope project, consists of 1,100 strategically distributed GPS receivers, as well as borehole seismometers, tiltmeters, and laser strainmeters installed along the western United States (see Figure 3.5). This geodetic observatory continuously monitors the strain field that results from the deformation of the active boundary zone

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE FIGuRE 3.5 Map showing the continuously operating GPS stations in the United States. Blue dots are Plate Boundary Observatory (PBO) sites installed by the NSF. Red dots are Continuously Operating Reference Stations (CORS), which are installed and operated by a variety of federal, state, and local agencies and some private companies and whose data is available through the National Geodetic Survey. Black dots are other sites that do not fall into either the PBO or CORS groups. Some of the PBO sites are also CORS sites. SOURCE: Courtesy of Thomas Herring, Massachusetts Institute of Technology, using Generic Mapping Tool (GMT developed by Paul Wessel and Walter Smith). Graphic generated on May 28, 2010, using NOAA/NGS CORS data acquired from CORS website: http://www.ngs.noaa.gov/CORS/sort_sites.shtmll. over multiple time scales. InSAR is a highly complementary and synergetic technique to the Plate Boundary Observatory’s GPS network, because it can generate continuous high-resolution maps of surface strain over large areas in any weather condition day or night, typically on monthly times - cales. Figure 3.6, based on InSAR data, illustrates the ground deformation resulting from the 7.9- magnitude Wechuan earthquake in China. Precise orbits and ionospheric delay corrections are needed to construct a seamless deformation map from multiple swaths. ALM offers a third geodetic method for estimating seismic hazards. The B4 project, for example, has mapped the entire San Andreas fault system at 1-meter spatial resolution and 0.1 meter vertical accuracy. Studies of the morphology of the fault zone, such as those combining offset stream channels with geochronology, provide critical information on the recent slip histories of specific faults. This information can be compared with the strain accumulation rate measurements from geodesy to estimate when the next major rupture is most likely to occur. In addition, mapping that is done before the next major rupture (hence, “B4”) will

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE FIGuRE 3.6 Ground deformation from ALOS L-band interferometry (each concentric color “fringe” cor- responds to approximately 10 centimeters of displacement) due to the 7.9-magnitude Wenchuan earthquake, which occurred on May 12, 2008, along the western edge of the Sichuan Basin in China. Shaking from the 270-kilometer-long rupture destroyed thousands of structures, killing nearly 70,000 people and leaving more than 4.8 million homeless. ALOS-derived ground deformation maps were available within a few days after the rupture in order to assess the extent of the damage zone, as well as to provide an estimate of regions of increased seismic risk. SOURCE: David Sandwell, University of California–San Diego.

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE provide the reference surface for assessing the very-small-scale deformation associated with a major rupture. The geodetic infrastructure for earthquake hazard monitoring with GNSS/GPS, InSAR and ALM would require better than 5-millimeter accuracy over 10 years. Landslides Landslides include a wide range of ground movements, such as rock falls, deep failure of slopes, and shallow debris flows (USGS, 2010). Although gravity acting on an over-steepened slope is the primary reason for a landslide, there are other contributing factors such as soil satura - tion by rainfall or snowmelt, earthquakes, volcanic eruptions, and stress induced by man-made structures. Understanding and mitigating landslide risk involves identifying areas of susceptibility (see Radbruch-Hall et al., 1982), determining which potential landslides are active, and deploying ground-based investigations in area of high risk and/or active slide areas. A promising method for identifying active landslide areas greater than about 200 meters is to use InSAR. Newer Japanese InSAR satellites operating at a longer wavelength (L-band) may make it possible to perform a global inventory of active landslides, even including those on vegetated surfaces. When the location of an active slide has been determined, ground-based methods, such as Light Detection and Ranging (LiDAR) mounted on tripods or aircraft, can be deployed to obtain “bare Earth” surveys (see Figure 3.7). The three-dimensional coordinates of the laser points can be used to determine the volume of the material involved in the landslide, as well as surface roughness and slopes of the slide and surrounding terrain. Floods Many natural processes, including hurricanes, weather systems, and snowmelt, can cause floods. Floods can also be caused by failure of levees and dams and inadequate drainage in urban areas. FIGuRE 3.7 These false-color geodetic images of a landslide near Flathead Lake, Montana, were made from airborne LiDAR data collected by the NSF National Center for Airborne Laser Mapping (NCALM). Laser shots passed through openings in the forest and reached the ground, allowing a filter to be used to remove the returns from the trees and reveal the landslide, which is not normally visible to the eye or a camera. The white scale bars are 500 meters in length. SOURCE: Ramesh Shrestha, National Center for Airborne Laser Mapping (NCALM), University of Houston; see also Carter et al., 2007.

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE On average, floods kill more than 100 people in the United States each year and cause billions of dollars in property damage (Figure 3.8). Land surface elevation defines the direction, velocity, and depth of flood flows, while subsidence measurements indicate how these values will change in the future. Elevations of individual streams and rivers traditionally have been mapped by land survey - ing. However, because Federal Emergency Management Agency (FEMA) floodplain mapping must cover nearly one million miles of the nation’s streams and shorelines, land surface elevation data are mostly derived from mapped sources, not from land surveying (NRC, 2007b). Land surface elevation information is combined with data from flood hydrology and hydraulic simulation models to define the base flood elevation, which indicates the extent of inundation. The creation of flood - plain maps is an important part of the National Flood Insurance Program, as these maps determine flood insurance requirements. FEMA is undertaking an ambitious five-year program to update and digitize the nation’s floodplain maps. Flat terrain in coastal zones and river flood plains are particularly flood-prone. According to the National Research Council (NRC) report Elevation Data for Floodplain Mapping (2007b), “…elevation data of at least 1-foot equivalent contour accuracy should be acquired in these very flat areas, rather than the 2-foot equivalent contour accuracy data that the FEMA floodplain map - ping standards presently require for flat areas.” Achieving this 1-foot accuracy actually requires two geodetic measurements—the terrain elevation and the geoid height (see Chapter 1). The NRC FIGuRE 3.8 Presidential disaster declarations related to flooding in the United States, shown by county: Green areas represent one declaration; yellow areas represent two declarations; orange areas represent three declarations; and red areas represent four or more declarations between June 1, 1965, and June 1, 2003. Map is not to scale. SOURCES: FEMA; Michael Baker Jr., Inc.; the National Atlas; and the USGS (from http://www.usgs.gov/hazards/floods/).

OCR for page 37
 FIGuRE 3.14 Major groundwater loss in the Sacramentro–San Joaquin River Basins in California (left) (Strassberg et al., 2009) and northern India (right) (Tiwari et al., 2009) revealed by the GRACE gravity mission and supplementary data. SOURCES: California groundwater iamge, NASA, http://www.nasa.gov/topics/earth/features/graceImg20091214.html (Left). India groundwater image, ScienceNOW, adapted from Tiwari et al., 2009 (Right).

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE –28.60 to –17.60 –17.59 to –13.54 Lake Pontchartrain –13.53 to –10.20 –10.19 to –8.90 –8.89 to –8.10 –8.09 to –7.50 International –7.49 to –7.00 Lake airport –6.99 to –6.60 Borgne –6.59 to –6.30 –6.29 to –6.00 –5.99 to –5.70 –5.69 to –5.50 –5.49 to –5.30 –5.29 to –5.10 –5.09 to –4.90 –4.89 to –4.70 –4.69 to –4.50 –4.49 to –4.30 –4.29 to –4.00 –3.99 to –3.70 M –3.69 to –3.40 RG O –3.39 to –3.10 –3.09 to –2.80 –2.79 to –2.40 –2.39 to –1.80 –1.79 to 10.30 FIGuRE 3.15 Map showing surface uplift rates (scale on right; units are millimeters per year) of New Orleans for 2002–2005, obtained from an InSAR analysis. Negative values for the uplift rates indicate subsidence. Most of New Orleans is subsiding relative to the global mean sea level, at an average rate of about 8 millimeters per year. Inset shows high subsidence rates between 2002 and 2005 on the MRGO levee, which later failed catastrophically during Hurricane Katrina. SOURCE: Dixon et al., 2006. River and Lake Levels In spite of the limitations of nadir-viewing radar altimetry (as opposed to side-looking radar imaging), monitoring surface water levels of rivers and lakes from space has a number of hydrologi - cal applications. These include studying the spatial and temporal effects of climate variability on surface waters, in particular over international river basins; improvement of models used in forecasts of hydrological variability; and water resource management (see Alsdorf and Lettenmaier, 2003; Cretaux and Birkett, 2006; Calmant and Seyler, 2006). Over floodplains, the combination of altim - etry-based water levels with radar or visible satellite imagery allows scientists to monitor changes in surface water volume, particularly during floods (see Frappart et al., 2006). Laser altimetry (for example, the Ice, Cloud, and Land Elevation Satellite, or ICESat, mission) is also being used. More than 15 years of radar altimetry measurements are now available for several thousand continental lakes, as well as for “virtual stations” on rivers (the intersection of the satellite ground track and the river) and floodplains. Typical height precision over lakes, where data from several altimetry missions can be combined, is a few centimeters. Over rivers, altimetry-based height precision is less accurate (in the range of 10–40 centimeters) for two reasons: (1) only those data that come from repeat passes over the same satellite track can be combined in a single analysis, so there is limited data available except along very large rivers, and (2) because of the large radar footprint, reflections from river banks perturb radar echoes (waveforms); moreover, unlike for oceans, there is no simple interpretation of river waveforms. A new concept of wide-swath radar interferometry has been recently proposed (Alsdorf et al., 2007) to monitor surface waters with unprecedented resolution (of approximately 100 square meters) and provide global coverage of worldwide rivers, lakes, and floodplains every few days. This mission, called SWOT (Surface Water Ocean Topography), was recommended by the NRC “Decadal Survey” and is listed among NASA’s future priority missions. SWOT will open a new

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE era in land hydrology, offering important new perspectives for studies on the terrestrial water cycle, flood prediction, and water resources. WEATHER Satellite imagery shown on television can give the impression that weather forecasts are based on these images, but forecasts actually come from physics-based models of the troposphere, the lowest 14 kilometers of the atmosphere. These models must continuously ingest measurements of the atmospheric state (pressure, temperature, humidity, and winds) at different altitudes around the planet to stay aligned with actual atmospheric conditions. The ability to predict both the severity and the temporal and spatial extent of weather changes, especially precipitation, is critical for public safety and agriculture, and governments around the world collaborate to collect the data used in forecasts. For decades, the input data for these models were provided by radiosondes, better known as weather balloons. Unfortunately, the spatial distribution of radiosonde sites is limited both by lack of coverage over oceans and by a significantly reduced number of sites in the southern hemisphere. Increasing the number of radiosonde launches per day is constrained by the cost of the instrumenta - tion, which cannot be reused. Geodesists focused on precise positioning use corrections to remove atmospheric effects, which are considered to be a source of noise in their measurements. Conversely, these same corrections may be used by meteorologists to better understand the atmosphere. Thus, improvements in the geodetic infrastructure benefit both fields of science. Ground-based Measurements Atmospheric refraction effects have long been recognized as an important error source in geodesy. Instead of relying upon uncertain models, space geodetic techniques such as GNSS/GPS estimate tropospheric variations along with the positioning parameters of interest. Eventually, it became possible to reverse the problem; assuming that the station position is well-determined, scientists could use the “nuisance” signal in the GNSS/GPS estimates to recover the time-varying behavior of the atmosphere (Ware et al., 2000). Unlike radiosondes, which measure the atmosphere conditions at multiple altitudes, ground-based GNSS/GPS measures only the “integrated” effect of the atmosphere, that is, how much the atmosphere delays the measurement in total. About 80 percent of the delay (known as the “dry troposphere”) can be predicted if surface pressure is measured. Once the dry troposphere delay is removed, one can recover the delay due to water vapor, which can be scaled to what is called precipitable water vapor (PWV). Measurements of PWV estimate moisture and latent heat transport models, which are critical for weather forecasts. To be useful for weather forecasting, however, the data must be available at close to real time. Since PWV varies both in space and time, globally distributed and frequently collected data are needed. For the current constellation of approximately 30 satellites, a single GNSS/GPS receiver will typically receive signals from 6 to 12 satellites. These data are primarily used to estimate PWV in the column of air above the GNSS/GPS receiver. But, in principle, having measurements from more than one direction means that GNSS/GPS has azimuthal, as well as vertical, sensitivity. The more satellites that are transmitting signals from a given direction, the greater the sensitivity will be. This also means that the combination of signals from multiple systems, such as Galileo, GLONASS, and COMPASS, will yield even more sensitive atmospheric monitoring capabilities than are achievable from the United States’ GPS alone. In addition, because GNSS/GPS receivers operate continuously, GNSS/GPS atmospheric sensing has extraordinary temporal sensitivity, which is particularly important for monitoring and predicting the behavior of strong weather events. Because the atmospheric delay is intrinsically related to how well position can be determined, GNSS/GPS tropospheric studies require the same infrastructure: accurate orbit determination, along

OCR for page 37
9 GEODESY REquiREMENTS FOR EARTh SCiENCE with a stable reference frame. Atmospheric applications require a co-located barometer to allow removal of the pressure, or “dry troposphere,” effect. Well-designed sites with low multipath errors are also valuable as they will produce more accurate estimates of PWV. For weather prediction, how- ever, there is the additional requirement that accurate orbits must be available in real time, whereas for climate studies, orbits can be made available in days to weeks. In order to achieve real-time accuracy, a reliable, globally distributed network of at least 50 real-time GNSS/GPS sites must be maintained. Rapid analysis of the data produced by this tracking network also must be supported (see Chapter 4). In order to improve the value of the large number of existing GPS receivers in the United States for meteorology, real-time telemetry capabilities would need to be expanded. Space-based Measurements While GNSS/GPS receivers have much greater temporal sensitivity than radiosondes launched twice per day, GNSS/GPS receivers have similar spatial limitations, in that they are (currently) limited to land and are present mostly in the northern hemisphere. For this reason, the atmospheric science community has actively sought measurements with better spatial density in the southern hemisphere and over the oceans. In the early 1990s, it was proposed that limb-sounding, a method that had been used to sense planetary atmospheres, could provide spatially dense measurements of Earth’s atmosphere. In a limb-sounding, or radio occultation, experiment, a radio signal is tracked from space rather than from Earth (Figure 3.16). This means that the horizon restriction is elimi - nated, and the signal can be refracted through the atmosphere for very long distances. The more GNSS/GPS and low-Earth orbit satellites are available, the more occultations that can be retrieved per day. After the success of a proof-of-concept satellite mission in the late 1990s (Kursinski et al., 1996), the six-satellite Constellation Observing System for Meteorology Ionosphere and Climate FIGuRE 3.16 Schematic of a GNSS/GPS radio occultation, or limb-sounding, measurement. Signals trans - mitted by a GNSS/GPS satellite are refracted by the atmosphere and received by the orbiter, generating a vertical profile of pressure and temperature through the atmosphere until the signal is eventually blocked (occulted) by Earth. SOURCE: Yunck, 2002.

OCR for page 37
0 PRECiSE GEODETiC iNFRASTRuCTuRE (COSMIC) was launched in 2006. Now operational, COSMIC provides global three-dimensional coverage of atmospheric temperature and water vapor from Earth’s surface to 40 kilometers (Anthes et al., 2008). In one day, COSMIC produces more than 1,000 well-distributed “soundings” in the southern hemisphere, compared to just over 100 southern hemisphere radiosondes above the continents. Furthermore, COSMIC is an all-weather system, and thus provides crucial data from the polar regions, observations for which are currently limited by weather restrictions. These new GNSS/GPS occultation data reduce uncertainties in both global and regional weather analysis and are routinely assimilated into both U.S. and European weather forecast models. In order to operate properly, however, they have requirements similar to those of ground-based GNSS/GPS networks, including accurate orbits and reference frame (which require a global GNSS/GPS tracking network providing near-real-time data). In addition, the velocity of the GNSS/GPS satellite relative to the low-Earth orbit satellite must be known to 0.1 millimeters per second, placing stringent demands on near-real-time orbit determination precision. SPACE WEATHER Above altitudes of 70–400 kilometers, the atmosphere is so thin that free electrons can exist. This ionized portion of the atmosphere contains a plasma—the ionosphere—in which the magni- tude of the ionization is controlled by solar activity. Characterizing changes in the ionosphere is important, because these changes—known as “space weather”—can have severe adverse effects on the increasingly sophisticated ground- and space-based geodetic systems of importance to govern - ments, corporations, and citizens. As described by Buonsanto (1999), the effects of space weather include “electric power brownouts and blackouts due to damaging currents induced in electric power grids, damage to satellites cause[d] by high energy particles, increased risk of radiation exposure by humans in space and in high-altitude aircraft, changes in atmospheric drag on satellites, errors in GPS and in VLF (Very Low Frequency) navigation systems, loss of HF (High Frequency) com - munications, and disruption of UHF (Ultra High Frequency) satellite links due to scintillations.” Before the advent of ground-based GNSS/GPS, researchers relied on limited and expensive in situ observations of the ionosphere. After the first GNSS/GPS satellites were launched, it was rec - ognized that GNSS/GPS could be used to monitor ionosphere variations (see Coster and Komjathy, 2008). The effect of the ionosphere on a radiometric signal is a time delay in the signal that is pro - portional to the level of ionization, described by the total electron content (TEC), along the signal path. The delay is also proportional to the radio frequency being used, so ionospheric scientists use the fact that the L1 signal (1575.42 MHz) has a different delay than the L2 (1227.60 MHz) to pro - duce global maps of TEC. Currently, these maps include GNSS/GPS measurements from more than 1,000 receivers (Komjathy et al., 2005). The NOAA Space Weather Prediction Center assimilates GPS data to model TEC over the United States (Fuller-Rowell, 2005). Figure 3.17 is an example of the TEC distribution over the United States during a period of relatively low solar activity. Global maps of TEC distribution are produced by the International GNSS Service and NASA JPL. However, ground-based TEC studies are intrinsically limited by the lack of GNSS/GPS receivers in the southern hemisphere and over the oceans. Missions such as COSMIC can produce 1,000–2,500 TEC profiles through the ionosphere, enabling a more accurate three-dimensional image of the ionosphere (see Anthes et al., 2008). Ionosphere maps can be used to calibrate single-frequency systems that are commonly used by surveyors, which cannot remove the ionosphere effect by themselves. They also can be used to study the temporal and spatial behavior of ionospheric TEC. Using GNSS/GPS for ionosphere mapping requires additional information that is typically not needed for positioning or troposphere studies. The hardware delays for the two signals on each GNSS/GPS satellite must be known very precisely, as well as hardware biases within the GNSS/GPS receivers. Because the estimation of

OCR for page 37
1 GEODESY REquiREMENTS FOR EARTh SCiENCE FIGuRE 3.17 Example of an ionosphere map produced by the NOAA Space Weather Prediction Center Au - gust 17, 2010. The unit for total electron content (TEC) is 1016 electrons per square meter (the total number of electrons in a tube with a cross-section area of one square meter, extending vertically from the surface through the ionosphere). SOURCE: NOAA, http://www.swpc.noaa.gov/ustec/. these biases is far more difficult in the presence of multipath error, the best way to improve knowl - edge of instrumental delays would be to improve multipath conditions at GNSS/GPS sites. This requires coordination on the national level to define the site requirements for infrastructure and to establish the leadership structures for implementing these requirements. The space weather com - munity also would benefit from more sites that telemeter data in real or near-real time. The NOAA Space Weather Forecasting network now uses data from only approximately 100 real-time sites. These sites can easily assimilate more data. The United States now has more than 2,000 GPS sites supported by a variety of government agencies; however, data from these sites are often transmitted only once per day. The Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS) system also can contribute to monitoring space weather through the launch of specially designed instruments that take advantage of worldwide DORIS transmitting beacons. The primary objective of the Scin - tillation and Tomography Receiver in Space (CITRIS) is to detect ionospheric irregularities from space at low latitudes. CITRIS, developed at the U.S. Naval Research Laboratory, differs from the normal DORIS receiver in that it is able to capture two-frequency transmissions at a sample rate of 200 hertz. With CITRIS flying on the U.S. Space Test Program satellite STPSat-1, two years of data were collected and processed to determine the fluctuations in ionospheric TEC and radio scintillations associated with equatorial irregularities (Bernhardt and Siefring, 2010). PRECISION SPACECRAFT NAVIGATION Precision Orbit Determination for Near-Earth Satellites Over the past several decades, the requirements for highly accurate determination of the orbits of near-Earth satellites have been driven by the evolution of the fields of satellite geodesy, includ - ing reference frame and gravity field determination, satellite radar and laser altimetry, and InSAR. The ability to use accurate range and range-rate measurements between an orbiting satellite and

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE tracking systems located on Earth’s surface has provided a dramatic improvement in the ability to monitor tectonic deformation and subsidence as well as monitor small but important changes in Earth’s rotation. The ability to use satellite altimeter measurements to obtain accurate, globally distributed observations of the ocean surface has opened a new era in oceanography. These same tracking measurements, along with satellite-to-satellite measurements, are providing unparalleled views of Earth’s gravity field and the gravity signals associated with temporal variations in the distribution of mass within Earth. These advances are intimately tied to the advances in precision orbit determination of Earth-orbiting satellites. Determining the orbit of near-Earth satellites involves four elements: (1) equations describing the motion of the satellite; (2) a numerical integration procedure for the equations of motion; (3) accurate observations of the satellite from the ground or other satellites; and (4) an estimation method combining the results of the first three elements to estimate the satellite’s position (Tapley and Ries, 2003). Continuous measurements of the three-dimensional position of a spacecraft are usually not available, but as long as the observations depend on the satellite’s motion, they contain informa- tion that helps to determine the orbit. The evolution of the satellite’s position and velocity must be consistent with both the physics of the mathematical model and the sequence of observations, which constrains the estimated orbit to a specific solution. As the tracking data and mathematical models for satellite dynamics have steadily improved, the accuracy of orbit determination also has improved. For example, whereas typical orbit accuracy for geodetic satellites was at the several-meter level in the 1970s, centimeter-level orbit accuracy is now achievable when high-precision tracking systems (GNSS/GPS, SLR, and/or DORIS) are used. This accuracy has been enhanced by the availability of dramatically improved Earth gravity models provided by the GRACE mission. interplanetary Navigation The primary aim of interplanetary spacecraft navigation is to position spacecraft relative to solar system bodies for precise fly-bys, orbit insertion, and surface landings. This differs significantly from near-Earth orbit determination because of the difficulty in obtaining tracking, the geometry of that tracking, and the long delays in receiving signals from distant spacecraft. The data used in interplanetary spacecraft tracking are generally radiometric measurements of range, range-rate, or the interferometric difference of signal arrival times, which are provided by very large radio anten - nas such as those of the Deep Space Network. Optical (image) measurements also have been used, but, in recent years, these are most often performed with the imaging systems on the spacecraft as opposed to terrestrial optical telescope measurements. Interplanetary laser ranging, using transpon - ders or one-way systems, are already being investigated, such as the current tests of one-way laser ranging to the Lunar Reconnaissance Orbit. Like near-Earth orbit determination, the basic analysis of spacecraft-tracking data requires relating the position of the spacecraft to the position of the tracking system, but rather than orbiting Earth, an interplanetary spacecraft is orbiting the sun or some other planet. The geometry of the basic measurement of range or range-rate is not sufficient to directly determine a spacecraft’s position, but the equations of motion for the spacecraft are very accurately known. Combined with precise knowledge of the tracking station’s position, a sufficient accumulation of tracking data constrains the possible motions of the spacecraft to a particular trajectory. For this, accurate coordinates as a function of time (ephemerides) for the solar system bodies also are needed. To keep improving the planetary ephemerides, tracking data collected on spacecraft near planets are incorporated with radar and optical measurements of the planets (Standish, 1998). Another important requirement is knowledge of Earth’s orientation in space, particularly Earth’s rotation angle (denoted as Universal Time 1, or UT1, a nomenclature leftover from when time was determined by Earth’s rotation rather than atomic clocks). Wind and water movements on Earth cause UT1 variations and cannot be pre -

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE dicted into the future with high accuracy. Determining Earth’s orientation with real-time accuracy would require a station position equivalent of 30 centimeters, which, for UT1, requires accuracies of 650 microseconds. Current spacecraft navigation systems obtain 24-hour prediction accuracies, or 125 microseconds for UT1. TIMING AND TIME TRANSFER The high-accuracy timing methods at the heart of geodetic techniques are useful for precise measurements of time and frequency. The ability to synchronize distant clocks accurately is an oper- ation commonly referred to as “time transfer.” Time synchronization requirements for such everyday functions as bank transfers, transportation, television broadcasting, and power grid regulation are on the order of milliseconds or less. GNSS/GPS signals are the most popular and economical way to achieve clock synchronization for most of these commercial applications. Scientific applications have yet more demanding time synchronization requirements, at the nanosecond level (10–9 seconds) or better. “Precise time-transfer” is typically only needed to synchronize clocks with comparable precisions (for example, atomic clocks). Because no single clock can be expected to be absolutely stable or reliable, the international definition of time for Earth (International Atomic Time or TAI) is based on an ensemble of approximately 300 atomic clocks in various laboratories around the world. Since these clocks must be compared constantly to produce the official time system, the precision of the time-transfer method is just as important as building the clocks themselves (Arias, 2005). The United States has demonstrated its long-standing commitment to the definition of TAI by contributing more than half of the clocks used to define the international timescale. Before the advent of GNSS/GPS, the most precise method for time-transfer was TWSTFT— two-way satellite time and frequency transfer. TWSTFT used laboratory-based transponders and expensive commercial satellites with a precision of approximately 0.5–1.0 nanoseconds (Hanson, 1989). With GNSS/GPS now freely available, methods developed by the geodetic community have improved precision by a factor of 10 (Bauch et al., 2006; Larson et al., 2000). Meanwhile, kilohertz time transfer is being tested on the Jason-2 mission with the T2L2 experiment, where picosecond- level stability over several minutes may be achievable. In addition to the international timescale TAI, a few countries operate their own primary frequency standards (Arias, 2005). NIST-F1, a cesium fountain frequency standard operated by the National Institute of Standards and Technology, serves as the United States’ primary frequency standard, with an uncertainty of 5 × 10 -16 (such a clock does not lose or gain more than one second in 60 million years). Such extraordinarily accurate clocks place the most demanding requirements for time transfer capabilities on the geodetic community and infrastructure. DECADAL MISSIONS The NRC’s Earth Science and Applications From Space: National imperatives for the Next Decade and Beyond (known as the “Decadal Survey,” NRC, 2007a) recommended a number of efforts and missions to address a wide range of scientific and societal challenges, from scientific questions related to melting ice sheets and sea level change to the occurrence of extreme events like earthquakes and volcanic eruptions. Specifically, the Decadal Survey recommended three space geodetic missions: ICESat-2, DESDynI, and GRACE-II. The committee agrees with these recommendations as being of high priority to the nation. Moreover, there is an important synergy between these missions and the current geodetic infrastructure that increases the potential for scientific discovery and other societal benefits associated with these Decadal Survey missions.

OCR for page 37
 PRECiSE GEODETiC iNFRASTRuCTuRE As noted in the Decadal Survey, sea level will change in part due to the thermal expansion (or decrease in water density) of the oceans as a result of a global-scale increase in temperature com - bined with the addition of water volume from melting mountain glaciers and ice sheets. Of these factors, changes in the volume of ice sheets in response to climate change is the least understood. The laser altimeter on ICESat-2 would quantify polar ice sheet contributions to recent sea level change and illuminate the linkages to climate conditions. It also would quantify regional signatures of ice sheet changes to assess the sources of that change and to improve predictive models, as well as estimate sea ice thickness to examine how ice, the ocean, and the atmosphere exchange energy, mass, and moisture. Massive urbanization and the extensive societal and economic infrastructure that has developed in coastal areas over the past century make the precise monitoring of sea level critical. In addition, ICESat-2 would measure vegetation canopy height as a basis for estimating large-scale land biomass (the amount of living matter in a given area) and biomass changes. Land biomass stores a significant amount of carbon. Measurement of canopy height will allow scientists to better assess the effects of climate and land management on vegetation and to improve under- standing of the global carbon budget. The DESDynI (Deformation, Ecosystem Structure and Dynamics of Ice) mission would employ an L-band Synthetic Aperture Radar (SAR) and multiple-beam LiDAR to monitor surface defor- mation and terrestrial biomass structure. Observing surface changes is critical for estimating the likelihood of earthquakes, volcanic eruptions, and landslides, and for predicting the response of ice masses to climate change and the impact of that response on sea level. Monitoring the size and distribution of vegetation will enable characterization of the effects of changing climate and land use on species’ habitats and the global carbon budget. The GRACE-II gravity monitoring mission, which proposes to use lasers instead of microwaves for satellite-to-satellite tracking, is expected to provide hydrological measurements down to scales approaching 100 kilometers or better (see Box 3.1). However, to avoid an undesirable gap in the gravity monitoring missions between GRACE and GRACE-II (proposed for the 2016–2020 time frame), a GRACE follow-on mission is now planned. These measurements will allow scientists to track large-scale water movement over the entire globe in order to better understand Earth’s hydro - logical cycle, provide inputs to meteorological models, detect changes in aquifers for improved groundwater management, and assess changes in ice sheet volume and distribution to improve predictions of sea level change. These Decadal Survey missions are highly complementary and are essential to fully exploit - ing existing geodetic observation systems. For example, uncertainty in the models for how Earth’s crust accommodates changing ice load (post-glacial rebound or glacial isostatic adjustment, GIA) currently hampers the determination of changes in the Antarctic and Greenland mass balance using satellite gravity measurements. However, by combining satellite gravity measurements with satel - lite altimetry, it is possible to distinguish between GIA and ice mass change, because GIA-induced changes (which involve relatively dense rock) will produce a different combination of surface and gravity change than those produced by variations in ice alone. Over vegetative areas, the laser altimeter and microwave SAR instruments will monitor surface deformation at different frequencies to more precisely separate biomass changes from surface topography changes. Between the three missions, changes in mass and surface topography for nearly all of Earth’s land area will monitored for seasonal and long-term changes. SuMMARy The operational applications and scientific research made possible by an accurate and eas - ily-accessible global geodetic infrastructure are limited only by the imagination of the user com - munity. The examples presented in this chapter were selected to provide some sense of the variety

OCR for page 37
 GEODESY REquiREMENTS FOR EARTh SCiENCE of applications in various stages of development and routine use. As the extent, density, accuracy, and accessibility of the geodetic infrastructure continues to improve, the scientific applications will continue to grow, enabling new services to be developed and new challenges to be solved. In addition to existing systems, some “Decadal Survey” missions—DesDynI, GRACE-II, and ICESat-II—also have the potential for great direct public benefit. They will each measure, in real time, an important component of the climate system. This not only will enable scientists to gather valuable data about climate, but also will provide the public with images that will enable them to visualize the dynamic Earth system and to understand the connection and interaction among the global water cycle, climate, and the solid Earth. The Decadal Survey missions also will benefit greatly from the global geodetic infrastructure. Indeed, these missions have the potential to become part of the infrastructure by providing unique geodetic information.

OCR for page 37