Present-Day Deformation Across the Basin and Range Province , Western United States

carbon cycle is established and that the waxing and waning of the terrestrial biosphere, possibly related to subtle climate variations and early human land use, are the most important factors controlling atmospheric CO2 concentrations over the last 10,000 years. During further glaciation in MIS 5.4, CO2 concentrations remain constant, although temperatures strongly decline. We suggest that this reflects the combination of the increased oceanic uptake of CO2 expected for colder climate conditions and CO2 release caused by the net decline of the terrestrial biosphere during the glaciation and possibly by respiration of organic carbon deposited on increasingly exposed shelf areas. These processes, however, should terminate (with some delay) after the lowest temperatures are reached in MIS 5.4 and ice volume is at its maximum at 111 ky B.P. (22). In agreement with this hypothesis, CO2 concentrations start to decrease in the Vostok record at about 111 ky B.P. Another possibility to explain this delayed response of CO2 to the cooling during MIS 5.4 would be an inhibited uptake of CO2 by the ocean. In any case, about 5°C lower temperatures on the Antarctic ice sheet during MIS 5.4 (17) are difficult to reconcile with the full interglacial CO2 forcing encountered at the beginning of this cold period and again question the straightforward application of the past CO2-climate relation to the recent anthropogenic warming. Another scenario is encountered during MIS 7, in which no prolonged warm period is observed. Although temperatures at the end of termination III are comparable to those at the end of termination II and CO2 concentrations are even slightly higher, a much shorter lag in the decrease of CO2 relative to the Antarctic temperature decrease in MIS 7.4 is found. Comparison with the SPECMAP record (23) shows that during the preceding interglacial MIS 7.5, ice volume was much larger than during the Holocene and the Eemian warm periods. Accordingly, the buildup of the terrestrial biosphere during MIS 7.5 is expected to be much less and sea level changes smaller, leading to a smaller net release of CO2 into the atmosphere during the following glaciation, which is not able to fully counterbalance the CO2 uptake by the ocean.

cene (9) shows that no equilibrium in the carbon cycle is established and that the waxing and waning of the terrestrial biosphere, possibly related to subtle climate variations and early human land use, are the most important factors controlling atmospheric CO 2 concentrations over the last 10,000 years.
During further glaciation in MIS 5.4, CO 2 concentrations remain constant, although temperatures strongly decline. We suggest that this reflects the combination of the increased oceanic uptake of CO 2 expected for colder climate conditions and CO 2 release caused by the net decline of the terrestrial biosphere during the glaciation and possibly by respiration of organic carbon deposited on increasingly exposed shelf areas. These processes, however, should terminate (with some delay) after the lowest temperatures are reached in MIS 5.4 and ice volume is at its maximum at 111 ky B.P. (22). In agreement with this hypothesis, CO 2 concentrations start to decrease in the Vostok record at about 111 ky B.P. Another possibility to explain this delayed response of CO 2 to the cooling during MIS 5.4 would be an inhibited uptake of CO 2 by the ocean. In any case, about 5°C lower temperatures on the Antarctic ice sheet during MIS 5.4 (17) are difficult to reconcile with the full interglacial CO 2 forcing encountered at the beginning of this cold period and again question the straightforward application of the past CO 2 -climate relation to the recent anthropogenic warming.
Another scenario is encountered during MIS 7, in which no prolonged warm period is observed. Although temperatures at the end of termination III are comparable to those at the end of termination II and CO 2 concentrations are even slightly higher, a much shorter lag in the decrease of CO 2 relative to the Antarctic temperature decrease in MIS 7.4 is found. Comparison with the SPECMAP record (23) shows that during the preceding interglacial MIS 7.5, ice volume was much larger than during the Holocene and the Eemian warm periods. Accordingly, the buildup of the terrestrial biosphere during MIS 7.5 is expected to be much less and sea level changes smaller, leading to a smaller net release of CO 2 into the atmosphere during the following glaciation, which is not able to fully counterbalance the CO 2 uptake by the ocean. in situ production of CO 2 in the ice, possibly related to carbonate or organic species reactions (or both), and have strongly compromised the validity of the determined CO 2 concentrations. However, Antarctic ice cores are (if at all) much less affected by this effect because of the very low abundance of reactive carbon species dissolved in Antarctic ice. 4. C. D. Keeling, T. P. Whorf, M. Wahlen, J. van der Pflicht, Nature 375, 666 (1995 The distribution of deformation within the Basin and Range province was determined from 1992, 1996, and 1998 surveys of a dense, 800-kilometeraperture, Global Positioning System network. Internal deformation generally follows the pattern of Holocene fault distribution and is concentrated near the western extremity of the province, with lesser amounts focused near the eastern boundary. Little net deformation occurs across the central 500 kilometers of the network in western Utah and eastern Nevada. Concentration of deformation adjacent to the rigid Sierra Nevada block indicates that external plate-driving forces play an important role in driving deformation, modulating the extensional stress field generated by internal buoyancy forces that are due to lateral density gradients and topography near the province boundaries. The northern Basin and Range province is an actively deforming intracontinental plateau lying between the stable blocks of the Sierra Nevada and the Colorado Plateau (Fig. 1).
The province has extended (increased in area) by about a factor of 2 over the past ϳ20 million years (1,2), and extension continues with ongoing seismic activity and slip along numerous faults distributed across a zone ϳ800 km wide (3)(4)(5). Constraints on the internal deformation of the province are limited. Geologic studies delineate regions of Holocene and late Quaternary fault slip (3,4). Space geodetic measurements broadly define movements across the province (6-8), and local surveys map concentrated deformation in several seismically active zones (9)(10)(11). The detailed pattern is important because it defines the current seismic hazard, with regions of high velocity gradient having more frequent damaging earthquakes than regions of low gradient. In addition, the spatial pattern constrains the fundamental processes driving active continental deformation, suggesting that external plate motions are more important than internal buoyancy forces in deforming the province.
Here we show the detailed velocity field mapped from a dense Global Positioning System (GPS) network that spans the Basin and Range. The GPS network consists of 63 stations, most of which were occupied on 2 or more days during surveys in 1992, 1996, and 1998 (12). The velocity of each station relative to stable North America was determined ( Fig. 1), and velocity magnitude and vector orientations were calculated (Fig. 2).
Several first-order features are immediately apparent from Figs. 1 and 2. First, deformation is strongly concentrated in two regions: the westernmost ϳ200 km and easternmost ϳ100 km of the network, with little internal deformation of the intervening ϳ500 km of the central Basin and Range. Locally high velocity gradients ( Fig with existing geologic, seismic, and space geodetic data. Reconaissance geologic mapping (3,4) and seismicity compilations (5) show evidence for Holocene fault slip and historical seismic activity in central Utah and western Nevada but for pre-Holocene slip  and low seismicity levels in the central Basin and Range. Widely spaced VLBI (Very Long Baseline Interferometry) and continuous GPS station data are consistent with our results (13).
The distribution of deformation across western Nevada suggests that the 8 to 12 mm/year of ϳ310°-oriented relative motion across the eastern California shear zone (7,(14)(15)(16), which lies south of our network near longitude 118°W, is partitioned between two fault zones. Average velocities of 2.8 Ϯ 0.5 mm/year between 114.9°and 117.7°W increase to 6.5 Ϯ 0.7 mm/year between 118°a nd 119.2°W and to 12.5 Ϯ 1.5 mm/year between 119.9°and 120.2°W. Thus, 2.8 Ϯ 0.5 mm/year of relative motion occurs across the Wasatch and related faults in central Utah, 3.7 Ϯ 0.8 mm/year of relative northwestward motion occurs across the CNSZ, and an additional 6.0 Ϯ 1.6 mm/year is accomodated within the SNTZ. The latter value is within the range of the ϳ3 to 6 mm/year of 300°-oriented motion inferred across faults in northwestern California and central Oregon (17), which suggests that much of this deformation may be accommodated through western Nevada.
Velocity vectors within the Basin and Range show the superposed effects of extensional stresses due to lateral density gradients in the lithosphere and tractions exerted by the relative motions of the bounding stable blocks. The average trend in velocity vector orientations across the province (Fig. 2B) is close to 310°, which is the direction of relative motion of the Sierra Nevada microplate with respect to stable North America (6,7,13), immediately suggesting the influence of this motion on internal deformation of the province. However, local variations in vector orientations provide clues that internal driving forces also affect the deformation.
The ϳ295°orientation of velocities in central Utah would seem to suggest deformation due largely to the motion of the Colorado Plateau (essentially stable North America) relative to the eastern Great Basin. However, the ϳ1-km increase in elevation and 15-km increase in crustal thickness across the Basin-Range/Colorado Plateau transition zone is expected to produce extensional stresses perpendicular to the Wasatch fault zone in central Utah (18). Geodetic measurements across the Wasatch zone are consistent with this stress field orientation. Velocity vectors and extensional strains are nearly normal to the local N20°E trends of the faults across our network ( Fig. 1) and to the north-south-striking Wasatch zone near Ogden, 200 km farther north (9,10). These orientations are also consistent with least principal stress orientations inferred from various stress indicators near the Wasatch front (19).
Between 118°and 120°W, the orientation of velocities is within Ϯ15°of the vector defining the relative motion of the Sierra Nevada block with respect to stable North America. This orientation, along with the high-velocity gradients across the region, suggest that Pacific plate-coupled motion of the Sierra Nevada microplate is responsible for much of the deformation of western Nevada. However, the large component of normal faulting present in this region suggests the perturbing influence of extensional stresses caused by buoyant, low-density, upper mantle beneath the Great Basin (20, 21). The local 295°orientation of velocity vectors across the Sierra Nevada-bounding Genoa fault, a pure dip-slip north-south-striking normal fault near 120°W, may be due to the perturbing effects of stresses generated by topographic gradients across this transition zone (18). These stresses would tend to rotate velocity vectors toward the normal to the Genoa fault in the elevated Sierra Nevada and away from this direction in the lower lying Basin and Range, as observed.
The velocity field measured across active faults provides estimates of fault slip rate and constraints on the mechanism of elastic strain buildup in the adjacent crustal blocks (22). An elastic half-space dislocation model with a normal fault dipping 60°that does not slip between the surface and some fixed depth (H ) but slides freely at a constant slip velocity below that depth yields the horizontal velocity expected due to elastic strain accumulation across the fault (23) (Fig. 3). The locked zone depth is taken to be the depth to which seismic fault slip or small earthquake hypocenters extend, which is 10 to 20 km in the Basin and Range. The high-velocity gradient in the model is ϳ3 H wide, or 30 to 60 km for locking depths that are appropriate here. For the area covered by our network, the expected pattern of horizontal velocity across an individual fault is thus well represented by a smoothed step and local peak or trough, with the net offset equal to the horizontal component of the fault slip rate. Strain accumulation across a series of widely spaced faults should resemble an irregular staircase, with steps being the zones of elastic strain accumulation and flats representing the intervening undeformed blocks.
The observed pattern of deformation across most of Nevada is similar to these expectations, with high-velocity gradients near 118°and 119.6°W and nearly constant velocities elsewhere. Figure 4 shows that bur-  ied faults beneath the CNSZ and SNTZ can explain the main features of the data. Both normal and right-lateral strike-slip faulting occur across each zone, so the modeled fault slip rate is the resultant of these two components. Because a number of active faults are exposed at the surface in each zone, the models are undoubtedly oversimplified. For example, the SNTZ contains both strike-and dip-slip faults, and the width of the deforming zone suggests that several subparallel faults locked to ϳ10 to 15 km would match the data at least as well as a single fault locked to 30 km.
Movements in central Utah and eastern Nevada are more complex than those shown by the simple buried fault slip model. Horizontal velocities ( Fig. 2A) increase near the Wasatch fault zone as expected but then abruptly decrease west of it. Velocity subsequently increases near the Drum Mountain faults, decreases to the west, and finally reaches a stable value of ϳ3 mm/year west of 113.3°W. There is suggestive evidence that a local velocity decrease similar to those shown in central Utah occurs near the Schell Creek Range [at 114.6°( Fig. 2A)]. The changes there are small, about 2 mm/year, and are supported by very few observations. However, it may be noteworthy that this is the only mapped Holocene fault that crosses our network between the Utah-Nevada border and the central Nevada seismic zone (4). Models like those in Fig. 3 match general features of the Utah and eastern Nevada data (24) but cannot reproduce the near-fault variations, which may be due to changes in slip along the fault strike (25).
Our results suggest that forces exerted on the Basin and Range by motions of the bounding plates are more important than internal buoyancy forces in driving large-scale deformation. Mappings of the gravitational potential energy that is available to cause deformation (21) show high gradients in eastern Nevada that do not correlate with strain rate gradients obtained by us. It may be that the stresses implied by these mappings are supported by the local strength of the lithosphere, with little resulting deformation. In contrast, the concentration of deformation near the western edge of the province indicates an important role for boundary forces due to motion of the Sierra Nevada microplate. Crustal thickness variations near the province boundaries may be important in causing deformation near the Wasatch fault and in complementing plate drag forces near the Sierra Nevada. However, the effects of all driving forces may be substantially modulated by lateral variations in lithospheric rheology, which is the relation between applied forces and the resulting displacements. Rheology is very sensitive to temperature and rock type (26), so we might expect that the varied structural and thermal history of the western United States would generate strong lateral variations in strength, channeling deformation into zones that are intrinsically weaker than their surroundings. 12 Res. Lett. 17, 2113 (1990). 17. S. K. Pezzopane and R. J. Weldon, Tectonics 12, 1140 (1993). 18. Isostatically balanced lateral differences in topography result from lateral density gradients that generate horizontal forces capable of causing deformation. For a two-dimensional structure across which the difference in topographic elevation is h and isostatic compensation occurs entirely within the crust, the force per unit length, F, normal to the structure is given by where c is crustal density, m is mantle density, g is the gravitational acceleration, and y cco is the crustal thickness of the low-lying region. If y cco ϭ 35 km, m ϭ 3300 kg m -3 , c ϭ 2750 kg m -3 , and h ϭ 1 km, then F ϭ 2 ϫ 10 12 N m -1 , which is comparable to many plate-driving and resisting forces.  37 (1996)] calculate that the driving force contribution of buoyant upper mantle in the northern Basin and Range (termed gravitational potential energy by them) averages 1.5 ϫ 10 12 N m -1 , which is comparable to the effects of province margin topography given in (18). However, these authors also point out significant uncertainties in the computed force gradients due to poor constraints on the exact upper mantle density structure and its lateral variations.  Fig. 3 is for a west-dipping fault. If the fault has an eastward dip, the pattern is reversed, with all local peaks replaced by troughs and vice versa (turn Fig. 3 upside down to visualize this). For dips shallower than 60°, the local peaks and troughs become progressively less prominent for the same slip distribution as that in Fig. 3 (9,10). 25. Some of the local velocity decreases seen at fault crossings of our network may be due to a sampling bias. Both coseismic fault slip and topographic height are expected to be a maximum near the center of each range and to decrease toward its ends. The interseismic strain accumulation rate (the velocity gradient normal to the fault) should generally mimic this pattern. Because highway engineers site roadways through the lowest available topographic gradient, most of our range-crossing stations are located either near the ends of active ranges or between adjacent ones. This bias will produce local alongstrike velocity gradients and velocity minima near the ends of active ranges, which is qualitatively consis-tent with patterns seen in our data across central Hubble Space Telescope imaging observations of two nearby brown dwarfs, DENIS-P J1228.2-1547 and Kelu 1, made with the near-infrared camera and multiobject spectrometer (NICMOS), show that the DENIS object is resolved into two components of nearly equal brightness with a projected separation of 0.275 arc second (5 astronomical units for a distance of 18 parsecs). This binary system will be able to provide the first dynamical measurement of the masses of two brown dwarfs in only a few years. Upper limits to the mass of any unseen companion in Kelu 1 yield a planet of 7 Jupiter masses aged 0.5 ϫ 10 9 years, which would have been detected at a separation larger than about 4 astronomical units. This example demonstrates that giant planets could be detected by direct imaging if they exist in Jupiter-like orbits around nearby young brown dwarfs.
Brown dwarfs (BDs) are "failed stars"; that is, they are not massive enough to sustain stable hydrogen burning but are sufficiently massive to start deuterium burning (1). Brown dwarfs are more like giant planets than stars in that their luminosity and temperature drop continuously with time, and ultimately they become extremely cool and faint. The borderline between stars and BDs is estimated to be at about 0.075 solar mass (M J ) for solar metallicity (2). The deuterium burning limit is at a mass of about 13 M J , where M J denotes a Jupiter mass (ϳ0.001 M J ) (3). We adopt this mass to separate BDs from planets in order to avoid the problems of a definition based on poorly understood formation mechanisms (4). For many years, BDs have eluded firm detection, but since 1995 several objects have been shown to be unambiguously substellar (5,6 ). The evidence for BDs is based on observations of lithium (7 ), luminosity, and surface temperatures (8). However, no direct mass measurement of a brown dwarf has been obtained to date. We present here the first object for which this can be done in the near future.
The first free-floating BD discovered in the solar neigborhood was Kelu 1. It was found in a proper motion survey (9). The second nearby free-floating BD was discovered by the Deep Near-Infrared Survey (DENIS). The DENIS and 2MASS surveys are ongoing and have the aim of yielding a complete sky coverage in the near-infrared I, J, and KЈ bands (10). The analysis of only 220 square degrees (about 1% of the planned DENIS survey) provided three objects (11) with I-J colors redder than GD 165B, which was the coolest BD candidate known before the discovery of Gl 229B (12). The surface temperatures of Kelu 1 and the DENIS objects are not obviously low enough to establish a substellar status, because young BDs and very low-mass stars can have the same effective temperature. The necessary distinctive substellar signature came from the spectroscopic detection of lithium in the atmospheres of Kelu 1 and DENIS-P J1228.2-1547 (hereafter abbreviated as DENIS 1228-15). (13). On the other hand, in very lowmass stars, the temperature and pressure at the bottom of the convection zone are high enough so that lithium gets rapidly destroyed through proton capture. The combination of high lithium abundance with low surface temperature implies that Kelu 1 and DENIS 1228-15 must have masses lower than 0.065 M J and ages younger than 10 9 years (1 Gy) (13).
We observed DENIS 1228-15 and Kelu 1 with NICMOS camera 1 (NIC1) on the Hubble Space Telescope (HST). The NIC1 data of DENIS 1228-15 were obtained on 2 June 1998 in multiple-accumulate mode with filters F110M, F145M, and F165M (14). This instrumental configuration provides an optimal combination of throughput and spatial resolution. Our observations of Kelu 1 were obtained using the same configuration on 14 August 1998. The NIC1 images of DENIS 1228-15 resolved two components of similar brightness (Fig. 1). To estimate the parameters of this binary system, we used an iterative approach: We modeled the data assuming two point sources and using both model and observed point spread functions (PSFs). The positions and the brightness ratios of the two point sources were free parameters, and the iterations continued until the residuals were similar to the noise. We obtained a separation of 0.275 Ϯ 0.002 arc sec and a position angle of 41.0 Ϯ 0.2°. The apparent F110M, F145M, and F165M magnitudes (respectively) on the HST-Vega system (15) are as follows: DENIS 1228-15 A (15.69, 14.96, and 13.98); DENIS 1228-15 B (15.89, 15.12, and 14.13); and Kelu 1 (14.13, 13.23, and 12.37). The standard deviation of these magnitudes is less than 0.01 magnitude, but the systematic errors can be up to 0.1 magnitude. Our F165M magnitude of 12.37 for Kelu 1 is in agreement with the published H magnitude of 12.32 (9). The B/A flux ratio of the DENIS 1228-15 system increases toward longer wavelength (0.83 for F110M, 0.86 for F145M, and 0.87 for F165M), indicating that DENIS 1228-15 B is slightly cooler than DENIS 1228-15 A. An independent confirmation of the binary nature of DENIS 1228-15 comes from public HST/NIC3 observations with filter F187N obtained on 24 June 1998 for another program by Hugh Jones and Todd Henry. The scale of NIC3 of 0.2 arc sec/pixel undersamples the PSF (theoretical full width at half maximum of 0.16 arc sec at 1.87 mm). Within the uncertainties due to the undersampling, the fitted values for the NIC3 data are in agreement with the results derived from the NIC1 data.
The trigonometric parallaxes of DENIS 1228-15 and Kelu 1 are not yet known, although they can be obtained with groundbased telescopes (16). The distance to Kelu 1