You are on page 1of 157

THE HORIZONTAL VELOCITY FIELD IN SOUTHERN CALIFORNIA FROM A

COMBINATION OF TERRESTRIAL AND SPACE-GEODETIC DATA

by
DANAN DONG
Submitted to the Department of Earth, Atmospheric and
Planetary Sciences in partial fulfillment of the
requirements for the degree of
DOCTOR OF PHILOSOPHY
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY

September, 1993
@Massachusetts Institute of Technology 1993. All rights reserved.

Signature of author
Department of Earth, Atmosplh4c and Planetary Sciences
July 23, 1993
Certified by
Professor Bradford H. Hager
Thesis supervisor

Certified by
Associate Professor Thoids A. Herring
Thesis supervisor
Certified by
Principal Research Sc ntist Robert W. King
Thesis supervisor

Accepted by

H. Jordan, Department Head
omas H. Jordan, Department Head
Professor ~lIiomas

MASSA*

AUG

_IT

-

THE HORIZONTAL VELOCITY FIELD IN SOUTHERN CALIFORNIA FROM A
COMBINATION OF TERRESTRIAL AND SPACE-GEODETIC DATA

by

DANAN DONG

Submitted to the Department of Earth, Atmospheric and
Planetary Sciences on June 24, 1993 in partial fulfillment of
the requirements for the Degree of Doctor of Philosophy

ABSTRACT
The rapid development of space-geodesy has given an impetus to research on
regional deformation, as well as to advancement in the methodology of geodetic
analysis. Taking advantage of the 3-D vector measurement and the high accuracy of
space-geodetic surveying over a large area, we develop the methodology to realize the
combination of space-geodetic and terrestrial survey data. Our methodology has
several important advantages:
(1) We adopt as our mathematical framework four-dimensional integrated
geodesy, which embodies the intrinsic relation between Earth's shape and its gravity
field. All formulations are expressed in a geocentric frame. Hence this methodology is
rigorous and easy to connect with space-geodetic data.
(2) The methodology can handle various types of geodetic data.
(3) Our methodology uses quasi-observations obtained from loosely constrained
intermediate solutions from subsets of the data, to which we apply general constraints
at the final step. Hence it is efficient and flexible.
(4) We do not assume a spatially uniform velocity gradient. Hence this method
is easy to apply to large areas.
(5) With our method, all three elements of a time-dependent reference frame site coordinates, velocities, and episodic displacements - are estimated simultaneously.
(6) We establish the criteria to check the compatibility of data entering the
solution. The criteria consider the correlation between the solutions before and after
combination.

First. Geological Survey .7). Ms = 6.0). King. MIT Robert W. U. Principal Research Scientist. we can trace the transition from compression in the Santa Barbara Channel and the Ventura Basin to simple shear near the San Andreas fault. Using this combination. Jordan. Finally. and another 5 mm/yr within a wedge-shaped area of the Ventura basin. Geophysicist. S. Cecil and Ida Green Professor of Earth Sciences. Our analysis also reveals several unsolved problems. Herring.6). Homestead Valley (03/15/79. Ms = 6.2. MIT Michael Lisowski. Thesis Committe Bradford H.The second objective of our research is to combine VLBI. ML = 6. Joshua Tree (04/22/92. During the transition. mainly in the Salton Trough area.2). North Palm Springs (07/08/86. Robert B. and Planetary Sciences. The physical reason is still unknown and should be further investigated. ML = 6.6). the data from observations performed in 1970's by the California Division of Mines and Geology are not compatible with the USGS data after 1980. Hager. Third. ML = 5. two important byproducts stem from our analysis: (1) Updated coordinates for the trilateration sites. and Landers and Big Bear (06/28/92. in the San Luis trilateration network. the observed fault-normal velocity in the Southern Coast Ranges between Parkfield and the coast is not significantly different from zero. 6. in the region of the Big Bend. and terrestrial survey data to obtain the horizontal velocity field in southern California. Our residual velocity field indicates that the current multi-dislocation model needs to be refined in its transition between the creeping zone and the locking zone in Parkfield area. Second.2). ivuii Thomas A. Ms = 6. Ms = 7. Atmospheric. Mexicali (10/14/79. there is some incompatibility between USGS trilateration data and part of the VLBI/GPS solution. ML = 5. Westmoreland (04/26/81.5.6). the current VLBI/GPS solution is still weak in the southern part of our network and east of the San Andreas fault. (2) Estimates of coseismic site displacements from 7 earthquakes: Superstition Hills (11/24/87. MIT Thomas H. Using both data sets. Shrock Professor of Earth. GPS.6) and Victoria (06/09/80. about 5 mm/yr of fault-normal velocity is absorbed in the off-shore area. 6. Associate Professor of Geophysics.

and for bringing Brad Hager and Tom Herring to our Department. resisting all kinds of pressure from China. but you cannot. Brad taught me the tectonics. when thousands of Chinese students went abroad to study advanced science and technology. Tom taught me the rigor of science. "white" was the synonym for capitalist. The response from my agency. To get such equal rights. I returned to the University and restarted my study. my only dream was to survive. During the following Pandemonium period. Brad Hager. however. I sincerely thank Tom for his continuous concern. degree. I expressed the same desire. Then Tom Jordan and Bob King.ACKNOWLEDGMENTS When I got the admission notice from MIT five years ago. Bob King and Yehuda Bock led me in climbing the mountain of GPS techniques and the GAMIT software." Study should be an equal right of human beings. In 1985. National Oceanic and Atmospheric . I am grateful for their grants from the National Science Foundation (EAR-86-18513 and EAR-89-05560). I had to struggle for another three years. When I opened a can of worms. I have exhausted all my English vocabulary and cannot find an appropriate word to express my special thanks to Bob. National Aeronautics and Space Administration (NAS-5-538 and NAS-5-737). It was at one of Tom's parties that I learned the difference between the hotness of cactus wine and red peppers. it is. just for the equal right to study. but also the morals and ethics.D. Finally. compared to "red"-the synonym for communist) and was expelled from the University without being allowed to ask why. Bob taught me not only the science. It was surprising that my volley of questions never irritated them. and Bob King guided me in accomplishing this thesis. was surprisingly negative: "Everybody can pursue his Ph. I was internally labeled as a "white specialist" (in China. His comments always sharpened my mind and enlarged my horizon. We need you to work here. When I was a freshman at Beijing University in 1964. Tom Herring. after fifteen years exiled in a remote desert area. deprived of all rights as a human being. I am deeply grateful to Bob and Yehuda for their generous support and help. however. accepted me as their graduate student. I had to struggle for more than twenty years. Tom Jordan and his graduate students have been like a big family. I could not believe it: "Is it real?" Now I can convince myself: "Yes. Unfortunately." This time it was not political persecution. Air Force Office of Scientific Research (AFOSR-90-0339). his keen instinct saved me from going astray.

and to Don Grant and Hadley Johnson. 72. Trans. I thank all of my English teachers: Bob King. Daxin Wang. Mark Murray. Mike sacrificed several days to help me to track down these ruffians.. and Madam Ye for her understanding my desire to study. Duncan Agnew. 1991]. Amer. Whenever they showed up. Richard Snay.Administration (NA90AA-D-AC481). I hope my work will reward her. and Dazhou Wang. and Zhengkang Shen for long discussions on the broad scope of sciences. She was sent to labor camp for several months. just for her pleading justice for me. Thanks to my officemates Steve Shapiro and Lana Panasyuk. They bring harmony into our office life. Without their data. fun showed up. It was they who encouraged me against decadence during my hardest time. Kurt Feigl and I learned GPS and started GPS field survey together. I also thank Dawei Zheng. Zhengkang Shen. Lana never forgot to say "Dobroy vecherom" (Good night). Damu Wang. . Dave Jackson. Thanks to Burch Oral for his magic sh family of shell scripts. She sacrificed herself to support my study. and Rob Reilinger for their generous support. I am greatly indebted to Mike Lisowski for his e-mail lessons on the EDM technique. I thank all of my student peers including my volleyball teammates. to Bob Ge for his story about the rupture at Landers. who sent me their theses to provide detailed information. 441. She tolerated my 12-hour work per day and 7-day work per week for 5 years. Andrea Donnellan. As companions. I sincerely thank my mother Yihui Dong. I was deeply impressed by Mark's patience. Thanks to my sister and brothers: Jin Wang. Kurt's magic functions made my computer work much more efficiently. my velocity field would be a runaway horse. Geophys. it would be very hard to understand what I am saying. and Southern California Earthquake Center. Please accept my promise: I will never do it again. My wife Daxin Wu deserves all of my praises. Michael Cline. to Rick Bennett for his results unearthed from the Imperial Valley. Thanks to Ben Chao. It is they who make my life colorful and joyful. my former advisor. Tonie vanDam. and Steve Shapiro. Steve (and his extremely nice father) showed me American humor. The maps and the figures of the thesis were generated using the Generic Mapping Tools (GMT) software. Ming Fang. for his overall concern. Will Prescott. generously made public domain by Paul Wessel and Walter Smith [EOS. which supported my research. Without their help. Un. Special thanks to Mike Lisowski. either in the office or in the field. Mark Murray. Mark Murray. While I was wrestling with the range busts. Kurt Feigl.

D. A. Seismology Bureau of China.4. A. 1988 Zheng Dawei.1993 Visiting scientist. Res. 1991 Danan Dong and Yehuda Bock. 10. W. 1973 .4. 28. Shanghai Observatory. and R. D. of Geophysics. 18.. S. 1984 The second Science and Technique Award. Atmospheric and Planetary Sciences. Atmospheric and Planetary Sciences. King. Beijing University. 94.1988 Research Associate. Current and future accuracy of Earth orientation measurements. Sub-milliarcsecond determination of pole position using Global Positioning System data. 1987 . MIT Dept. No. F. 10. 1987 .. A. Ashtech Incorporated. Proceedings of the AGU Chapman Conference on Geodetic VLBI: Monitoring Global Change. 11. NOAA Technical Report NOS 137 NGS 49. Herring. 3949-3966. 1893-1896. Dong. Dong. Libration of the Earth's rotation. Global Positioning System Network Analysis with Phase Ambiguity Resolution applied to Crustal Deformation Studies in California. Res. Lett. 18. 1988 . MIT Dept. Dong Danan. 1979 . and T. summer 1992 Visiting Scientist. 306-324.. 1982-1984 B.1982 Teacher of Mathematics in high school. Dept. 1991 T. Academia Sinica. Acta Astronomica Sinica. Chao. 1984 .. 1991 T. Lin Jizeng. A study on the inversion of rupture process using the compound model method. of Earth. Geophys. Marple Algorithm of Autoregressive Spectrum Estimate and its Application to Analyzing Astrometrical Data. Herring and D. N.1987 M. Acta Seismologica Sinica. Dong Danan. Shanghai Observatory. S. J. Shanghai. 337-351. Lett. 1989 PUBLICATIONS B. 364-373. Herring.1979 HONORS The second Excellent Essays for the young. Geophys. Dong.1989 Lin Banghui. 1984 The fourth Science and Technique Award. No. Res.CURRICULUM VITAE EXPERIENCE Research Assistant. Geophys. 2007-2010. of Earth. Hu Xiaoxing..

1985 Dong Danan. 1985 Dong Danan. Dong Danan.4.1. 1986 Zheng Dawei.736-750. Chen Youfen. A Collection of Papers of International Symposium on Continental Seismisity and Earthquake Prediction. Dong Danan. No. 27. Annals of ShanghaiObservatoryAcademia sinica. 55-63. 27. Proceedings of the International Conference on Earth Rotation and the Terrestrial Reference Frame. Dong Danan. 1986 Zhao Ming. Dong Danan. research on the fine structure in polar motion with multi stage filter. 27. Zheng Dawei. Annals of Shanghai Observatory Academia Sinica. Chen Youfen. Dong Danan. Acta Astronomica Sinica. Simulation Test and Discussion on Q estimation of the Chandler Wobble by Spectral Analyses. Acta Astronomica Sinica. 13-25. 41-50. 16-22. On the 30-year Scale Libration of the Earth's Pole. No. Computation of Displacement of Station Position caused by Tides using Spherical Harmonic Method. 1984 Lin Banghui. Measurement of the Quality Factor Q of the Chandler Wobble. 7. 352-359. A study of orthogonal and oblique rupture process with application to the 1966 Xingtai earthquakes.Dong Danan. End Effects of the Vondrak Filter. 7. Realization of Narrow-band Filtering of the Polar Motion Data with Multi-stage Filter. No. 1984 . Acta Astronomica Sinica. Columbus. 50-62. Annals of Shanghai Observatory Academia Sinica. 1986 Zheng Dawei. 368-376. Dong Danan. 1985 Yang Zhigen. Ohio. A new research for the secular drift of the Earth's pole. 1986 Zhao Ming. Annals of Shanghai Observatory Academia Sinica. 6.4.

......................................... Terrestrial survey data ....... ................................................................................................ References ..................TABLE OF CONTENTS Abstract ......................... 3 3................ 15 16 17 19 20 20 20 21 23 24 26 28 CHAPTER 3: RESULTS Introduction ........ ........................................... ................................................................................................... Conclusions.... East of the SAF ....................................... 5 7 9 CHAITER 1: INTRODUCTION Introduction .......... Choosing the observables ........... ................................................ ............ Acknowledgments............................................................................................................................................................................................ Curriculum Vitae .......................... Quasi-observation approach.................... ................. Southern section .......... ................................... Table of Contents......................................................................................................... ... .................................................................................. .... .................... Description of the derived velocity field ................................................................ Brief review of the methods currently available ........................................................................................................................... Estimation procedure...... Big Bend region ............ Outer coordinate solutions for each trilateration subnetwork ..................... Analysis of the trilateration data ...... Third step: impose general constraints to obtain robust solution ............................ Conclusions .......... ....... General observation equations in geocentric Cartesian frame ........................... .................................................................... Dealing with a large number of stations ................................. Southern Coast Ranges.................. References 31 32 32 33 34 34 36 36 37 39 39 41 41 42 42 45 ..................................... ....................................................................... Combination of the VG and EDM data .................... Second step: obtain the combined solution under a uniform frame.......................................... .... 13 CHAFIER 2: METHODOLOGY Introduction ......... ............................................................................. ............................................................. Minimum-constraint soluiton .. Geodetic observation in southern California ................................................. Checking compatibility of the data sets ....................... .......................... VLBI/GPS data........................ ............ First step: get a loosely constrained solution .......................

.... .......................... Appendix 7: Contribution of nominal lengths to the estimate of velocity .....74 3e: Anza ................................. 77 Figure 6: Combined solution using constraints in all common sites...................................... Appendix 5: Solution changes in the case of adding new data and new param eters ......................................................... 84 Figure 9b: Residuals of the combined solution relative to MPNS ....................... 80 7c: Remove constraint at Asbestos ............... Figure 3: Scatters 3a: San Luis subnetwork........... Site displacement and velocity....................................................................................... ..............................APPENDICES Appendix 1: Coordinate frames .............................. 76 Figure 5: Residuals of the minimum constrained solution ..................................................................................... 79 7b: Remove constraint at Monu_res ........................................................ .. Baseline and baseline rate vectors ............................ 68 69 Figure 2: Outer coordinate solutions.Mexicali subnetwork .................................................. Episodic site displacement vector ........................ 72 3d: San Gabriel ..................................... 73 ..... Appendix 8: Sensitivity testing................ Appendix 9: Error model................... Appendix 2: Specific linearized observation equations ............................................................ Horizontal and vertical angles and mark-to-mark distance ....................... 70 3b: Carrizo subnetwork .................................................................... .............Tehachapi subnetwork ................ Astronomical longitude A........... 3f: Salton .......................................................... 85 ........................................................78 Figure 7: Tests of combined solution in southern section 7a: Remove constraint at Niguel.............................................................. 71 3c: Los Padres ...........Joshua subnetwork ....................... Appendix 3: Reparameterization .......................... Appendix 11: Parkfield dilemma ......... 82 83 Figure 8b: Combined solution relative to SAF ............................. ........................................Tehachapi subnetwork .......... Appendix 13: List of all baseline length observations ............................................. Figure 9a: Residuals of the combined solution relative to Pacific plate............................. 49 50 50 51 51 51 52 53 56 56 57 58 59 61 62 63 64 67 FIGURES Figure 1: VG and EDM networks in southern California ................. Appendix 10: Procedures of updating coordinates........................................... 81 Figure 8a: Combined solution relative to North American Plate ....................... ....... ................................. Appendix 4: Comparison between Gauss-Markov model and other approaches .................................................................. 75 Figure 4: Minimum constrained solution for EDM data ........................................ Appendix 6: Procedures for updating poorly known coordinates in a terrestrial network .............................. Appendix 12: Coseismic site displacements .. latitude Dand gravity g ...

............................... . 93 14b: 04/26/81 Westmoreland earthquake (Ml = 5..7)................ 96 14e: 07/08/86 North Palm Springs earthquake (Ms = 6.......... 95 14d: 03/15/79 Homestead Valley earthquake (Ms = 5...... 6............... 155 8b: formal uncertainties of VG solution scaled by a factor of 3.........147 Table 3: USGS EDM sites used in this analysis.... 90 Figure 13 Comparison of outer coordinate solutions in San Luis network 13a: Using USGS data 1980-1991.. 89 12b: Rotation rates (Delaunay triangle representation) ......... ..... and blunders in eastern part of the Mexicali subnetwork 1981-1991 .............. 91 13b: Using CDMG data 1971-1979 and USGS data 1980-1984...... 153 Table 7: Velocity constraints on the combined solution....1..............145 Table 1: Accuracies of terrestrial measurements ....... Table 6: Constraints on the northern and southern sections...........................0) ...................... analysis in this used sites Table 4: VLBI/GPS 153 Table 5: USGS EDM subnetworks ....... 146 Table 2: Accuracies of VLBI/GPS measurements.. 92 Figure 14: Coseismic site displacements 14a: 11/24/87 Superstition Hills earthquake (Ms = 6.................. .6) .............................................2....... 88 12: Strain rates and rotation rates in southern section 12a: Strain rates (Delaunay triangle representation) ............. 87 11: Residuals of the combined solution relative to Red Hill ...... steam well extraction............6) and 06/09/80 Victoria (ML= 6...................Figure Figure Figure Figure 10a: Velocity of the combined solution relative to MPNS . ............ 97 14f: 04/22/92 Joshua Tree earthquake (Ms = 6..................100 Figure 16: Time series of the EDM observations . ..................56 8c: add constraint on site SNP2 with scaling factor of 3...................6)............................................................ 86 10b: Residuals of the combined solution relative to MPNS ............................................1).................................146 .....................6) ......................157 ........2) earthquakes ......................... 6..94 14c: 10/14/79 Mexicali (ML= 6..................... 98 14g: 06/28/92 Landers/Big Bear earthquakes (Ms = 7......... 99 Figure 15: Aftershock...........5.............. 154 Table 8: Compatibility test using constraints at all effective common sites 8a: formal uncertainties of VG solution scaled by a factor of 2 ... 152 ................. 101 Figure 17: VLBI/GPS and EDM network updated...................

12- .

thermal convection. Although the crustal movements no longer obey simple rigid motion in these regions. and monitoring disasters [Snay and Holdahl. there have been increasing efforts to recover the horizontal velocity field in California. fault interactions.. the deformation appears to be constant. to derive a more homogeneous solution with better spatial resolution. Caltech. The deformations caused by the other sources. 1993]. investigators from Scripps. Snay et al. are not as well constrained.. except for short intervals around catastrophic events.1984. axial rotation. Some. using either terrestrial survey data [Lisowski et al. are comparatively well-understood. In recent years. . tectonically induced stresses. Ward . recovering the interseismic velocity field would characterize the main features of regional deformation.. ocean loading. and to develop a methodology of combining different geodetic data in a way that optimally extracts the information appropriate for modeling deformation. Combining Global Positioning System (GPS) data from 1986-1992 with Very Long Baseline Interferometry (VLBI) data from 1984-1991. detecting abnormal temporal deformation. such as solid tides. In this thesis I further combine VLBI and GPS results with terrestrial survey data for this region. UCLA and MIT have derived the horizontal velocity field of central and southern California with a precision of better than 2 mm/yr [Feigl et al. Hence. documenting historical motion. and rigid plate motion. magma intrusions and earthquakes. Another important application of a regional velocity field is that it establishes the base for a dynamic geodetic reference system. The aims of this work are to take advantage of more of the geodetic information currently available. such as ice loading. 1991]. such as earthquakes. volcanic eruptions. mostly located at plate boundaries. 1990]. which provides a powerful tool for predicting position as a function of time.1987] or space-geodetic data [Clark et al. 1991. polar motion. In actively deforming regions. These regions serve as the windows through which we gain insight into how the crust responds to various tectonic forces. crustal movement demonstrates more complicated patterns than simply rigid-plate motion.CHAPTER 1 INTRODUCTION Temporal deformations of the Earth's crust are induced by various sources.

Chapter 3 will demonstrate the derived velocity field and its geophysical implications. .14 Chapter 2 will focus on the methodology. Appendices will describe the details in specified areas.

designated FONDA (FOward modeling Network Deformation Analysis). especially to vertical deformation. Our formalization is based on the work of Hein [1986] and Collier et al. A general mathematical model of deformation analysis should embody the intrinsic correlation between the Earth's shape and its gravitational potential. we impose general constraints to get a more robust estimate. 1990]. 3) process efficiently and flexibly through general constraints and reparameterizations. Third. Second.. we combine the individual loosely constrained estimates into a uniform solution. 2) estimate velocity directly without the assumption of a spatially uniform velocity gradient. The GPS phase observations were analyzed using the GAMIT software [King and Bock. .updated terrestrial network coordinates from the original poorly defined coordinates and earthquake-induced coseismic site displacements. was developed to process terrestrial survey data and to combine terrestrial and VLBI/GPS results. [1988]. site velocities. 1992. Site coordinate adjustments. we obtain loosely constrained estimates of geodetic parameters from space-geodetic or terrestrial observations. First. We perform our analysis in three steps. A new software package. as well as episodic site displacements are expressed explicitly as the estimated parameters. Herring et al. The VLBI group delay observations were analyzed using the CALC/SOLVK software developed at the Goddard Space Flight Center (GSFC) and the Harvard Smithsonian Center for Astrophysics (CfA) [Ma et al. 1993]. The Earth's external gravity field is sensitive to crustal deformation. with some modifications. 1993]. This methodology allows us to 1) perform simultaneous reduction with various types of geodetic data. and 4) obtain simultaneously two important byproducts . We have implemented our analysis scheme using three preexisting software packages and a new one to incorporate terrestrial observations. This approach is powerful because it recognizes that most geodetic observations are functions of both position and gravity.CHAPTER 2 METHODOLOGY We adopt four-dimensional integrated geodesy as our mathematical framework. The combination of VLBI and GPS estimates is accomplished using the GLOBK software developed at CfA and MIT [Herring..

and assumes no prior model. The extended Frank's method permits multiple observation epochs and no longer requires identical networks. Frank's method [Frank. This method requires homogeneous data and an identically repeated network configurations. [1987] recognized the limitation of this approach and suggested that future improvements should "adjust California to a single nodel" anc eplace current expressions for secular motion with individual station velocity parameters". The advantage of these methods is that the estimated parameters are uniquely determined. For noisy data. Even by integrating the velocity gradient from a site with a well-constrained velocity. But this method still cannot handle directly heterogeneous data. all these methods are most suitable for a small area. external constraints are required to eliminate the rank deficiency of such terrestrial survey data. where the assumption of a spatially uniform velocity gradient is valid. 1987] is the most widely used implementation of the simultaneous reduction algorithm. these methods provide spatially averaged results to enhance the signal-to-noise ratio. in order to apply these methods. In the new era of coexisting multiple geodetic techniques. the translation and rotation of the entire network cannot be resolved. A large or tectonically complicated area must be divided into smaller districts. it is now rarely used. 1988] applies the velocity field derived . 1979] represents the minimum-norm generalized inverse of a singular normal matrix. which estimates the velocity gradient with the assumption that the gradient is constant over the area network occupied. An alternative approach is to remove the assumption of a spatially uniform velocity gradient and to estimate station velocity directly. we only get the spatially averaged velocity field. 1981] adds a network rotation that minimizes velocities along a specified direction. The "outer coordinate solution" [Prescott. Therefore. the "model coordinate solution" [Segall and Mathews. Thus. Bibby [1982] developed the simultaneous reduction method. Prescott [1976] extended Frank's method to estimate angular shear strain rate 71 and Y2. assuming that the strain is uniform in both space and time.Brief review of the methods currently available A new methodology usually develops in response to new measurement techniques. But it is still not possible to derive the velocity field directly. based on a priori knowledge. The DYNAP program developed at NGS [Drew and Snay. 1966] of calculating angular shear strain 71 and 72 was applied to repeated triangulation observations. assuming strain is uniform in space. When repeated triangulation data with different networks became available. The "inner coordinate solution" [Brunner. If only trilateration or triangulation data are used. Snay et al.

the ambiguity of space geodesy is on a continental or global scale. It is also difficult to judge the model accuracy directly. used conveniently for terrestrial observations. whose time-dependence is described by the parameters a. practically efficient. these methods are still applied to small areas. They are especially useful when we want to calibrate or quantify the model parameters. the solution is not unique and depends strongly on the model adopted. Several studies [Snay and Drew 1989. Space-geodetic observations still have their own rank deficiency. ellipsoidal geodetic. Furthermore. General observation equations in a geocentric Cartesian frame The observations from different techniques are conventionally expressed in one of several possible frames: geocentric Cartesian. the space-geodetic solution can be considered unambiguous on a regional scale.from a geophysical model to constrain the solution. . in a large or tectonically complicated area. Nevertheless. but also to link observations of separated terrestrial networks to get a uniform solution.b. Combining terrestrial survey data with space geodetic data makes it possible not only to derive an unambiguous velocity field. a globally distributed GPS network solution cannot resolve the longitudinal rotation of the whole network. However. it is very difficult to specify a single dominant direction to apply outer coordinate constraints or to construct a model velocity field.t). our work attempts to develop a theoretically rigorous.t). If accompanied by a global network. and usefully flexible methodology for combining heterogeneous data sets. we adopt the geocentric Cartesian frame as the internal coordinate system in which to construct the observation equations. The geodetic measurement 1(t) can be expressed as 1 (t) = F(X(a.t). h(t)) (1) where X(a. The transformation between the Cartesian and the geodetic or topocentric frames. Considering the potential of space geodesy. Space geodesy has now arrived in the geodetic arena and can estimate 3dimensional positions with unprecedented accuracy. is given in Appendix 1. Building on the previous studies. For example. Grant 1990] have shown that such a combinational technique is feasible.t) is the position vector. both of these solutions usually match well the known local tectonic features. As a result. or topocentric planimetric.W(X(a. Hence.

W(X(a. Thus the potential terms are used only to correct the raw observations. Readers interested in using 4-dimensional integrated geodesy to estimate potential parameters together with other parameters simultaneously are referred to Milbert and Dewhurst [1992]. so that a is simply the site velocity V. AXo is the adjustment of a priori position. The linearized observation equation is 1= A (AXo+Au(ao. F denotes a non-linear functional which operates upon X. AF A= -P= x DF aU +-- aw ax OF aAX aX aa C-- B= + DF aU AX aW aX a Q Q- aF w F xr (3) aW ab aF ah In the absence of an episodic event.bo. Site acceleration and high-order time-derivative terms are neglected. such as rotation.t) is the gravitational potential.bo. we usually assume that crustal movement is uniform in time to the first order. and e is the noise term. Time-dependent disturbing potentials n are not needed to estimate a horizontal velocity field. Aa. (2) becomes &6=A (AXo + (t . translation. W.t) is the correction of displacement with a priori model parameter ao. Ah is the correction of the a priori auxiliary parameter h o .t))+PAa+QAb+CAh+e (2) where 8 represents the residual of observed minus calculated. which can be decomposed into a reference potential U and a time-dependent disturbing potential x. Based on these simplifications. Ab are corrections of the a priori model parameters a o and bo. and satellite orbits.t) is the disturbing potential defined by a priori parameters. The parameters b describe the temporal variation of the disturbing potential.t))+B(AU(Xo)+x(Xo. based on the a priori model. Additional parameters can be added when significant time-dependent signals are detected in the residuals.t). AU(Xo) is the correction to the reference potential at a priori position Xo. Au(ao. n(Xo. and h to produce the scalar value of I (t).to) AV) + B AU(Xo) + P AV + C Aho + E (4) . h(t) are the auxiliary parameters.b.

In the quasi-observation approach. we address the importance of using quasi-observations. and their covariance matrix. data from a single GPS observation session consist typically of phase measurements of 6 satellites from 20 stations at 1000 epochs. There are two approaches to estimating AXo and AV: i) estimate the position and velocity directly from the raw observations. only the parameters related to the sites in southern California and their covariance matrix are necessary. employed by both GLOBK and FONDA. Here the term "combination" represents not only the combination of data from different techniques. we need to keep only the estimated station coordinates. For example. and the parameters relating to sites outside southern California need not be used directly to obtain a rigorous combination solution (see Appendix 3). easily stored in about 0. Then using only the resultant baseline length rates as quasiobservations. In data combination. Such an approach is .Appendix 2 lists the linearized observation equations for various measurements. This advantage of the quasi-observation approach cannot be easily realized by processing the raw data directly. we can obtain reliable site velocity estimates. Computation time and memory space savings are the primary advantages of the quasi-observation approach. The parameters of the satellite orbits. Computation time for the combination is typically reduced by about two orders of magnitude. If we use the GLOBK output as quasi-observations to combine with the terrestrial survey data. satellite orbits. The files required to store the residuals and the partial derivatives with respect to site coordinates and satellite parameters occupy about 20 Mb of storage. we can compress the data into baseline length rates and baseline lengths at weighted mid epochs. The final estimates of the parameters from the two approaches are equivalent (see Appendix 3).5 Mb. the quasi-observation approach is more efficient and flexible. Earth rotation parameters. VLBI radio-source positions. at which the nominal baseline lengths and baseline length rates are uncorrelated. ii) estimate intermediate solutions from the raw observations and use these intermediate solutions as quasi-observations to estimate AX and AV. Quasi-observation approach Before introducing the trilogy of our analysis. but also the combination of measurements from different sessions using the same technique. When the a priori coordinates for trilateration sites are poorly determined.

not just the position adjustments. we use GAMIT to analyze GPS phase data to obtain loosely constrained estimates of site coordinates and satellite orbits. we use CALC/SOLVK to analyze the VLBI group delay data to obtain loosely constrained estimates of site coordinates and source positions. with covariance matrix C1 812 = A2 8X + E2. site velocities. and earth rotation parameters. 1991. On the other hand. Loose constraints are necessary to prevent the normal matrix from becoming singular. these constraints should be loose enough that they do not bias the parameter estimates derived in later steps of the analysis. Our philosophy is to keep the loose constraints during most of the analysis. and all observation uncertainties are purely Gaussian. Such an approach maintains a uniform fra:r e and minimizes the effects from overconstraining [Feiglet al. Second step: combining the data under a uniform referenceframe Suppose the observation equations for two subsets of the data are 611 = A1 6X + el. GLOBK treats these solutions as quasi-observations to obtain a homogeneous estimate of site coordinates at a reference epoch. 1991]. 1991].insensitive to the poorly determined a priori coordinates and hence has been widely used to derive the velocity estimates in previous studies [Lisowski et al. due mostly to the inconsistency of fiducial stations and insufficient orbital modeling [Murray. Similarly. and to impose tight constraints at the final stage. Using loose constraints is necessary to obtain a homogeneous solution in the combination. with covariance matrix C2 (5) (6) Here X represents all parameters. Estimation procedures First step: the loosely constrained solutions For each daily observation. We assume that the observations in the two subsets are uncorrelated. Directly using tightly constrained single-session solutions degrades the long-term repeatability. multi-session satellite orbits. Then the least squares solution of tlhe combined system is given by X1+2 = C1 +2 (AT C 1 811 + Al C-1 812) (7) where C1+2 = (AT C1 Al + A C-21 A2 + CX) . Larson and Agnew. 1993].1 (8) .

Suppose that the original adjustments are XL with covariance CXL and misfit XL. Third step: impose general constraints to obtain a robust solution In the final step. The misfit is measured by "the weighted sum of squared residuals" (X2) X2+2 = (811-A1 BXI+2 ) C11(811-A1 8X 1 +2 ) + (81 2 -A 2 8X1+ 2 ) C21(812-A 2 8X 1+ 2 ) (9) The above model is usually referred to as a Gauss-Markov model. For GPS. In general. we might adjust the velocity translation of the whole network at the final stage to stabilize the solution. The translation of the frame can be specified later by fixing one site position and velocity or forcing the sum of the adjustments to the coordinates and velocities of several sites be zero. The constraint observation equation is lc = A.XL (14) . This latter procedure is used in GLOBK and is analogous to an inner coordinate solution. we update the site coordinates of these networks to align them with the uniform frame (see Appendix 6). we compare this formulation with the "model coordinate" approach. we impose general constraints on the solution through conditional equations. A uniform terrestrial reference frame is necessary for combining solutions.is the covariance matrix of the combined solution X 1 +2 and Cx is the covariance matrix of the a priori values of the parameters X. X with covariance Cc (10) Sequential least squares gives the updated solution: Xc = XL + CXL AT (Cc + AcCXLAT)'Alc (11) 1 Cxc = CXL .CXL AT (Co + AcCXLA) Ac CXL (12) X = XZL+ AXT C'L AX + AI~ C 1 A1l (13) where Alc= le .Ac XL AX = Xc . In GLOBK. the reference frames implied by different techniques do not coincide. the translation is constrained initially by the weak constraints. For VLBI. we solve only the rotation angles explicitly. there is no translational rank deficiency since the orbital dynamics is sensitive to the position and velocity of the Earth's center of mass. which has almost no sensitivity to translation because of the nearly infinite distant quasar observation. and no explicit translation is allowed. In Appendix 4. When the GPS network is not strong enough. Since the original coordinates of terrestrial survey networks are very poor.

y. so that they move together. The savings in time and space are obvious.(vi+vi)= kv ((uj+uj) . as if they were on the same rigid block. we can manipulate various constraints without reprocessing the original observation data. x(x.. We can bind velocities of several closely located sites.v. We can also bind the velocities at terrestrial survey sites to the velocities at the nearby GPS/VLBI sites. When the a priori coordinates of several sites have common errors.(ui+6ui)) where kx. Second.. We have incorporated a variety of constraint equations in both GLOBK and FONDA. The same will be true for other constraint equations. 3) Assign the orientation of the velocity vector: vi+Bvi = kv (ui+8ui) (17) where kv is a specified constant..w) denote the coordinate and velocity respectively.(xi+Bxi)) (18) (vj+vj) . the constraint can be imposed separately for each component. kv are constants. 1) Assign specified values to the coordinate adjustment or velocity for one or more sites: (15) 5Vi= V spec 8xi= X spec. Specifying values to zero is equivalent to fixing the a priori coordinate or velocity for this site. First.to)) x.to)) 2 = (dxij + dvij(t .There are several advantages in using a sequential least squares approach to impose the constraint. 2) Tie the adjustments or total values of the coordinates or velocities: X i = BXj = SX k =. Unrealistic constraints can be identified through abnormal X2 increments.z (dxij + dxij+(dvij+ 8dvij)(t .y. thereby resolving the translation and rotation of the whole network. In order to provide flexibility in the analysis. 4) Assign baseline or baseline-velocity orientation: (yj8yj) . for example a "main" mark and several "reference" marks. The velocity tie is the most useful one.. most of these constraints are applied in a geodetic or topocentric frame.y.z) and v(u.z (19) . In the following equations. Although we have written the equations in the form of a vector.. Vi = (16) Vj = 8Vk =.(yi+Syi)= kx ((xj-+x) . the coordinate tie may be used. 5) Freeze the baseline length: I x. the reasonableness of the constraints can be easily assessed by the increments of X2 from (13).

In the constraints of (20) and (21). outer.6) Fix the center of a sub-network or fix the motion of the sub-network center: k k i=l (20) (vi + 8vi) = 0 S8x i = 0. A site with large uncertainty will affect the solutions and the covariance matrix of all other sites in the inner coordinate solution and hence should be eliminated in advance. 9) Inner. compatibility checking is necessary to avoid misleading results due to systematic errors and blunders. the uncertainties depend on which site is chosen to be fixed. every parameter has the same weight.v iXj -X i 8 vi Vk + Vk-Vi Xk-Xi Vi (22) Formula (22) actually includes 4 independent constraints for horizontal velocities. i=l 7) Fix horizontal or vertical rotation of a sub-network to zero: k . In the inner coordinate solution. and the geometrical distance to the center is the dominant factor. dy. It should be noted that these constrained solutions of no-network-translation and nonetwork-rotation are different from the inner or outer coordinate solutions. (dzi(ui + Bui) . and model coordinate constraints. The details of these constraints can be found in Segall and Mathews [1988]. If the original solution is derived from only trilateration or triangulation data.dyi(ui + uij) = 0 i=l k . (dyi(wi + 8wi) .dxi(wi + 8wi))= 0 i=l Here dx.dzi(vi + 8vi))= 0 i=1 (21) k . . dz denote the distance from site i to the center of the sub-network. 8) Link velocity gradient: Rotating the coordinate frame by a specified angle. and the parameters with smaller uncertainties will be less changed. Here our discussion focuses on the compatibility of velocity solutions. in this frame Vj+Vj . (dx (vi + vi) . the parameter uncertainties affect the constrained solution. Checking compatibility of the data sets During the combination.

the covariance matrix of differences of the parameter estimates has the relation: C . In this extended case. If the solution difference -X exceeds the 95% confidence level. this requires 30 megabytes of memory. but additional parameters are also added.15) shows that adding additional parameters enlarges the covariance of common parameters for the same data.For the same data but estimating additional parameters. In the case of incompatible data. Therefore external compatibility checking based on prior geophysical information is also useful. we give the proof (see Appendix 5) that formula (23) is still valid provided that the new data are uncorrelated with the previous data. Hence the above compatibility test is unable to detect errors of network translation and rotation even though these errors exist in the space geodetic solution. we have attempted to identify the sites with significant incompatibility and to avoid using them to impose constraints. small errors of rotation in the space geodetic solution will be propagated and enlarged in the far sites of the terrestrial survey network. In combining terrestrial survey data with space geodetic data. Our methodology requires 6 parameters for each site plus episodic parameters and auxiliary parameters. the covariance matrix of the differences of the common parameters is given by Equation 15 of Appendix 3. we rescale the original covariance matrix to incorporate unknown systematic deviations. If the first step does not improve the solution enough. adding new data reduces the estimated covariance. Terrestrial survey data have no sensitivity to the translation and rotation of the whole network. Equation (23) is often used to check the compatibility of new data with the original data or to test the sensitivity of parameter estimates to subsets of data. In a typical solution for southern California involving 400 stations. whereas for the same parameters. For . Dealing with a large number of stations The main shortcoming of our methodology is the computer resources necessary to store and use the large matrix required. If the common sites are concentrated in a small area of the terrestrial survey network. = C -C (23) Here i and x denote the estimated parameters with and without new data added respectively. Davis et al. not only are new data added. Our methodology uses (23) to check the compatibility between terrestrial survey data and space geodetic data. the terrestrial survey data are not compatible with the space geodetic data. [1985] have proved that for the same parameters with the addition of new data. The comparison between (23) and (3.

A Helmert blocking approach [Schwarz and Wade. The Helmert blocking method is rigorous but requires complicated indexing and bookkeeping. Since that EDM data are observed in separated networks. which are previously solved implicitly. allowing a large network to be handled with a sparse normal matrix. At the top level is the combined solution with only the parameters related to VLBI/GPS sites estimated explicitly. We have not yet fully realized this method in our software since we have not yet experienced the pressure of processing a large number of sites. Thus. Such a procedure is repeated until the last data set at the top of the block pyramid is used. In combining the data from several lower blocks. At the bottom level. our block pyramid has only two levels. The shortcoming of this simplified Helmert blocking approach is that it does not work where the EDM network inclodes no VLBI/GPS sites. Then this method backwards estimates explicitly all parameters of each block. Each higher level block compresses the data from several lower level blocks. We use the Helmert blocking approach to combine the VLBI/GPS quasi-observations with the USGS EDM trilateration data. At the last step the estimated parameters related to GPS/VLBI sites are ^1 and their covariance matrix is C^x. Thus we do not need additional indexing and bookkeeping. We use the back solution approach described byHerring. which then appear in only one block. the parameters that are related to the EDM sites are estimated implicitly except at the sites collocated with VLBI/GPS sites. all data should be arranged into multi-level blocks in advance. which are the most accurate terrestrial surveys. [1983] to recover the parameters of the EDM network. Our philosophy is to realize the processing using a simpler method. The remaining common parameters are estimated explicitly and form the data at the higher level.000 sites. this method implicitly estimates all the parameters.our ambition of processing up to 10. even a supercomputer is not large enough. and the VLBI/GPS quasi-observations are distributed in all blocks and are considered as common parameters. we combine the VLBI/GPS solution with the trilateration data in each network sequentially. In each step. the EDM data in each network form a block. To perform Helmert blocking. 1990] would reduce the size of space. derived from the normal equations: where N 11+CZ N 12 x N2 1 N22 X2 = Bi+C B2 i (24) .

taking into account the correlation between the solutions before and after combination. For the less accurate triangulation data. The method presented in this chapter provides a useful tool for estimating the horizontal velocity field from a combination of terrestrial and spacegeodetic survey data. Within a few years. N12 . and episodic displacement . Given this rich data set. Thus we can divide a whole region into several blocks. N21 . We try only to obtain the velocities of triangulation sites relative to GPS/VLBI/EDM sites. (3) Constraints can be applied in a general and flexible manner.xl and Cx denote the estimates and covariance matrix of xl from the previous step. our basic assumption is that these data cannot change the combined GPS/VLBI/EDM results significantly. our methodology should be further developed to incorporate a large number of sites.N 2 1 . The coefficients of the normal sub-matrices N11. Our methodology offers several important features: (1) The theoretical frame is rigorous and can easily incorporate leveling and gravity observations. N2 2 and the righthand terms B 1 . The back solutions are then (25) X2 = N B 2 -E k1 and C 2 = N 2 + E C^x ET x2 where E =N. (4) The compatibility criteria are relatively rigorous. used as quasi-observations. Within each block there are enough GPS/VLBI/EDM sites to control the movement of the whole block. velocity. space-geodetic data will be available for California and other regions with great spatial and temporal resolution. we can obtain the estimates and covariance matrices related to the sites of each EDM subnetwork. How to handle time-dependent deformations in an optimal way will be an important extension of our current methodology.are estimated simultaneously.site coordinate. As long as the velocities estimated from the entire GPS/VLBI/EDM network are selfconsistant. Using formula (25) repeatedly. (2) All three elements of a time-dependent reference frame . the velocities at triangulation sites should be self-consistant also. B2 are constructed from the data of the EDM subnetwork. and represent the parameters related to the sites of one EDM subnetwork. Conclusions The developments of modern space geodesy have motivated the development of a rigorous and flexible methodology for combining terrestrial survey data with space geodetic data. Although our current simplifications do not .

27 inhibit our determination of a horizontal velocity field for southern California. the rapid development of space-geodesy will spur us to remove these simplifications in the near future. .

1993 Frank.. Geophys. C. M. H. A. J. H. Larsen. T. Discussion on displacement analysis: detection of crustal deformation. H. J. Ziony. Lecture Notes in Earth Sciences.. GLOBK: Global Kalman filter VLBI and GPS analysis program. 1990 Gu. Geodesy by radio interferometry: effects of atmospheric modeling errors on estimates of baseline length. Preliminary slip-rate table and map of late Quaternary faults of California. S. Jordan.. A. Matti. Sarna-Wojicki. Theses. 1986 Herring. 1986 Hein.. MIT. and Y. 161-174. Res. and I. Z. Webb. Lajoie. Eissfeller. and W. U. Jackson.. Res. A. 91. 20. T. K. B. 1984-1992. R.. G. 505-548. A. Donnellan. D. L. Sharp. H. 1985 Drew. Seismol. Rymer. K. S. Radio Sci.. J. Prescott. D. D. 1992 . D. 1982 Clark. M. J. Perkins. C. A. R. On a four-dimensional integrated geodesy. 1593-1607. 62. A.. Harwood. J. Res. A. Snay. V. A. 12561-12581. D. Integrated geodesy state-of-the-arc 1986 reference text. G. C. Bull. Am. J. Bull. Larson. H. Geol. Shen.. et al.REFERENCES Bibby. Surv. R. R. Harms. Tinsley III. submitted to J. R. and F. Thesis. I. Deduction of earth strains from survey data. Agnew. Combination of terrestrial and GPS data for Earth deformation studies. D. G. 71-91. F. K. Herring. W. T.. D. J. Soc.. D. T. Davis.. Vol. Dong. Geod. Geodesy by radio astronomy: the application of kalman filtering to very long baseline interferometry. Bock. Ph. 82. DYNAP: software for estimating crustal deformation from geodetic data. W. Sunkel.. Measurement of the velocity field of central and southern California. M. Unisurv report S-32. Sims.. M. King. 7439-7446. Bock. M. and J. Tectonophysics. Y. J. Shapiro. Hager. Ph. 95. Open file report 84-106. B. Murray. Hein and H. C. I. L.. unpublished documentation. Landau. M. L. and R.. T. Precision and accuracy of intercontinental distance determinations using radio interferometry. W. J. J. K.. Geophys. Geophys. Massachusetts Institute of Technology. A. B. 35-42. M. 56. Tectonophysics. Documentation for the MIT GPS analysis software: GAMIT. 1984 Collier. 1993 King. A.. 1983 Herring. 7.. 1966 Grant. 1989 Feigl. 331-343. J. W. 162.P. 1990 Herring. Massachusetts Institute of Technology. 1988 Davis. Lienkaemper. D. S. Unbiased estimate of strain from triangulation data using the method of simultaneous reduction. K. Springer-Verlag. Edited by H.

S. J. J. C... Geophys. M. Agnew. Geophys.King. Monogr. Res. A. School of Surveying. Soc. Application of the Global Positioning System to crustal deformation measurement 1. and Holdahl. Recent Plate Motions and Crustal Deformation... W. and D. Crustal Dynamics Project Data Analysis-1991 VLBI geodetic results.. D. 1976 Prescott. Drew. 1991 Lisowski. C. J. 104552. California. M. E. Res. Res. 6067-6072. J. J. 1-1. 64. W. A. Geod. 14954-14966. The determination of displacement field from geodetic data along a strike slip fault. The North American Datum of 1983: project methodology and execution. Ph. vol. Springer-Verlag. 225-236. The velocity field along the San Andreas fault in central and Southern California. 1991 Ma.. Res. Australia. 66. R. R.. 6001-6008 1973 Schwarz. Annapolis.. and D... Geophys. 28-62. 97. 1991 Lisowski. supplement. New South Wales. precision and accuracy. C. Prescott. Collins. M. and A. J. 1991 Prescott. P... High PrecisionNavigation -Integration of Navigationaland Geodetic Methods. University of New South Wales. Res. 1989 . D. and E. Mattews. Masters. Caprette. J. The Yellowstone-Hebgen lake geoid obtained through the integrated geodesy approach. An extension of Frank's method for obtaining crustal shear strains from survey data. Reviews of Geophysics. C. 545-558. Geophys. Global Positioning System measurement of crustal deformation in Central California. MIT. R. G. Surveying with GPS. T. H. H. Rizos. 1990 Segall. C. . R.. Geophys. W. 1991 Snay. 78. Prescott. AGU Chapman Conference on Time Dependent Positioning: Modeling Crustal Deformation. and J. C. S. and W. R. 96. 162-171. Ser. Bull. G... W. NASA Tech. 1847-1853. Wade. Savage and W.New York. Precision of geodolite distance measurements for determining fault movements. 1981 Savage. R. Res. 93. Displacement calculation from geodetic data and the testing of geophysical deformation models. M. H. Bull. A. 9. 1988 Snay. Mem. Dewhurst. H. V. Seismol. Theses. and M. J. J. 16547-16565. K. and W.. 1992 Murray. 83698389. The emergence of dynamical geodetic reference systems. 96. 1985 Larson. Stolz.. Ruan. B. 1992 Milbert. Geophys. 86. Combining GPS and classical geodetic surveys for crustal deformation in the Imperial Valley. Am.. H. Maryland.

M. Timmerman. S. 1987 Ward. L. N. NOAA Technical Report NOS 125 NGS 42. Pacific-North America plate motions: New results from very long baseline interferometry. A. 95. Cline and E.30 Snay. W.. 21965-21982. Res. R. 1990 . Geophys. J.. Project REDEAM: Models for Historical Horizontal Deformation..

however. find significant fault-normal velocities. caused predominantly by strike-slip faults. In Southern California.CHAPTER 3 RESULTS Introduction Southern California is located in the boundary zone between the North American and Pacific plates. Geological and geodetic evidence indicates that the slip rate of the San Andreas fault (SAF) accounts for less than three quarters of the velocity predicted by the NUVEL-1 model [Demets et al. Trilateration observations show that the deformation in a 100 km wide band near the main segments of the San Andreas fault is dominated by right-lateral shear.. Lisowski et al. with only a few sites across the SAF. 1991]. 1993] (hereafter F93). Most VG sites used by F93 are concentrated in the southern California borderland.. Thus the current VG solution cannot "see" clearly the deformation transition from coastal areas to the SAF. 1984]. The combination of terrestrial survey data with space geodetic data not only enhances the temporal and spatial coverage of the velocity . based on a combination of VLBI and GPS (hereafter VG) data... and can be modeled successfully by fault-slip motion on multiple dislocations [Lisowski et al. which cannot be explained by a multi-fault dislocation model [Feigl et al. 1987]. 1990]. is attributed to extension in the Basin and Range province and to the integrated rate of deformation in California [Minster and Jordan.. The remaining motion. 1991.. Recent results. and hence cannot "see" the deformation far from the SAF. Quantifying the distribution of deformation in Southern California will provide important constraints on the geodynamical processes in this region. the main patterns of deformation are characterized by the horizontal velocity field. Feigl et al. 1987. termed the "San Andreas discrepancy" [Minster and Jordan. 1993]. where the crust shows more complicated behavior than simple rigid plate motion. Mapping this field directly from geodetic data has become feasible only in recent years [Snay et al. Several questions arise: (1) Is the velocity field from terrestrial survey data compatible with the field derived from space geodetic data? (2) What is the distribution of the fault-normal velocity? (3) Can the deformation in the inland region of southern California absorb all of the San Andreas discrepancy? The trilateration networks have narrow apertures spanning short distances across the SAF.

such as block rotation [Jackson and Molnar.. Regularly repeated length observations along " .. we added trilateration data surveyed by the California Division of Mines and Geology (CDMG) prior to 1979. survey networks were extended into the interior of California. Table 1 lists the accuracies of the terrestrial survey data. Most of the trilateration data used in this chapter were obtained by the USGS between 1971 and 1991 (Table 1 and Figure 1). Geological Survey (USGS) since 1970. 1991]. Nearly an order of magnitude improvement in accuracy was achieved in 1969 when the CDMG and USGS began using aircraft to measure the temperature and humidity along the line of sight. when they were largely supplanted by electronic distance measurements (EDM).solution. but also provides the ability to preserve a homogeneous solution in a large area and to find the locations of the missing fault-normal velocity.fault segments in California were performed by the California Department ui "Yater Resources (CDWR) from 1959 to 1969. The rotational tectonic features. After 1930. we attempt to derive the horizontal velocity field in southern California from both VG and terrestrial survey data. (3) some incompatibility between space-geodetic data and terrestrial survey data still exists. 1990]. For the San Luis network. Furthermore. can therefore be determined directly. Most nineteenth-century surveys were performed along the coast to support navigation. 1992. to which terrestrial survey data are insensitive. We also include data from USGS surveys in the Joshua network made just after the Joshua Tree earthquake (April 22. the California Division of Mines and Geoti. Geodetic observations in southern California Terrestrialsurvey data The history of terrestrial survey data in southern California can be traced from 1850 [Snay et al. The causes of this incompatibility are not yet clear and can be resolved only with additional data. 1987. . (CDMG) from 1969 to 1979. Motivated by the benefits of combination. Direction were the dominant measurements until the 1950's. S. Hager et al. the vector measurements of space-geodesy can resolve the rotation rate. partly in the Ventura basin area.6). using the methodology developed in Chapter 2. Our results indicate that (1) deformation in the off-shore region contributes to the San Andreas fault discrepancy. providing better calibration of atmospheric refractivity. Ms=6.5.1992.1) and again after the Landers and Big Bear earthquakes (June 28.. (2) the fault-normal velocities are absorbed partly in the off-shore region. and the U. and partly east of the SAF. Ms=6. Ms=7.

Among the 30 USGS trilateration networks in southern California. There are 94 VLBI sites and 147 GPS sites represented. Los Padres. 1990]. 1990]. such as the Landers and Palmdale networks. the first VLBI experiments were performed using both dual-frequency observations and the Mark II recording system. Our VG data include VLBI experiments from 1984 to 1991. In order to reduce the computational burden. VLBIIGPS data In 1982. San Jacinto and Elsinore faults from the southern end of the Big Bend to the Mexicali area. and the global GPS experiments from 1991 to 1992. We have also omitted the Mojave network because it was surveyed only once.. we have omitted from our analysis the Garlock. Table 3 lists all of the trilateration sites used in this analysis. we have omitted 16 small-aperture networks.The earlier trilateration and triangulation measurements are also useful because of their long time span and dense spatial coverage. San Gabriel networks. since our current analysis is focused solely on broadscale features of the velocity field. The accuracies of the VG data in southern California are listed in Table 2. Coso. During the first few years. and one site (BLKB7269) is in the Salton Trough. Mexicali networks. Three VG sites share the same benchmarks as the USGS trilateration sites: . The permanent GPS Geodetic Array (PGGA) in southern California began operating in 1990 to continuously monitor the deformation and to provide a homogeneous reference series [Bock & Leppard. with most of the California sites concentrated in the western Transverse Ranges and the southern California borderlands (Figure 1). we have not yet included these data in our analysis. The first extensive GPS survey in southern California was made in 1986 [Dixon et al. Carrizo. in 1982. Tehachapi. and most of the Monitor network. the distribution of sites is sparse. Joshua. Finally. Salton. the satellite constellation and the network of global tracking stations were limited. and Barstow networks because they are isolated from other networks and because there is no VG site within these three networks to provide control. East of the SAF. A regular program of observations at about 15 sites began in 1984. The southern section straddles the SA. DEAD7267) are in the Mojave region. California GPS experiments from 1986 to 1992. Only one site (FIBR_GPS) is in the Great Valley. We group the remaining 10 networks into two separate sections. however. including total 144 sites of the Anza. and a small part of Monitor network. GOLDVENU. The location of the sites are shown in Figure 1. The northern section of 179 sites spans the SAF from Parkfield to Pearblossom and includes the San Luis. three sites (MOJA.

In Appendix 7. when the original coordinates are poor. Analysis of the trilateration data Choosing the observables Previous analyses of the trilateration data have used baseline length rates as quasi-observables to derive the velocity field [Lisowski et al.g. 2) After conducting statistical tests and comparing different solutions. if the the site coordinates are well constrained. assessment of the condition required for compressed nominal baseline lengths to make a significant contribution to the velocity solution. but requires external knowledge of the coordinates. Therefore. from VG. and MPNS_GPS with Mt_pinos. .. We take the VG solution and its covariance matrix as quasiobservations and make the following modifications: 1) We remove three subsets of sites: i) the sites outside of southern California because they are beyond the scope of our research. the few sites collocated by VG in our current data limit the accuracies of updated horizontal coordinates to the 1-3 meter level. and solution quality are described in F93. The main advantage of this approach is that the baseline length rates are insensitive to site coordinates and contain most of the velocity information. In the end. The software used for our analysis is described in Chapter 2.. 1991]. data analysis. and (2) compressed nominal baseline lengths at weighted mid epochs (See Appendix 5). as did F93. we use the scaled covariance matrix. ii) the sites with fully correlated velocity solutions since they are redundant. however. All of the details concerning the data distribution. we give a quantitativ. e. SNP2_GPS with Santapau. and iii) the sites with poorly determined velocities because they cannot provide reliable constraints. which is not sufficient to provide a significant contribution to the velocity estimates. 43 sites in southern California (including OVRO) enter our analysis (Table 4). F93 concluded that the formal standard deviations should be doubled to reflect the realistic uncertainties in the solution. But this approach discards two types of data: (1) baseline length with observations at only one -3och. The length-observable approach utilizes all data.DEAD7267 with Sand Hill. Although we were able to update the coordinates and to estimate the velocities simultaneously (Appendix 10). These two types of data can provide meaningful velocity information.

Outer coordinate solutionsfor each trilaterationsubnetwork
Prioro to combining terrestrial and space-geodetic data, we divided the USGS
trilateration data into 6 subnetworks (Table 5) and applied outer coordinate
constraints [Prescott, 1981] to get the velocity field relative to the center of each
The purpose of this analysis is threefold: (1) to scrutinize the local
deformation pattern and its relation to local faults; (2) to compare the data quality of
each subnetwork; (3) to diagnose outliers. The minimum-velocity directions for the
outer coordinate solutions were chosen to be perpendicular to the local segment of the
SAF. Based on our tests, the velocity solutions had no significant changes if the
subnetwork.

minimum-velocity direction changes were within ± 50. We assessed the data quality
of each subnetwork by the scatter in the length observations (Appendix 9). In order
to avoid contamination from coseismic displacements, we estimated episodic
displacement parameters for 7 recent earthquakes (Appendix 12). We have not,
however, evaluated offsets due to errors in the eccentric ties at some of the sites [M.
Lisowski, personal communication, 1993]. Our future research should examine the
records of the local eccentric ties and correct the errors.
Our outer coordinate solutions of subnetworks are very similar to the results of
Lisowski et al. [1991], so elaborating the solutions and their tectonic implications
appears redundant. Figures 2 and 3 show the velocity fields and scatters
corresponding to each subnetwork. We draw the following general conclusions:
(1) Simple shear is the dominant feature in all 6 subnetworks.
(2) No fault-normal velocities were found that are significantly different from zero.
However, this conclusion should be further investigated for two reasons: i) Since the
outer coordinate solution eliminates the translation and constrains the rotation of
subnetwork, it is impossible to detect a common fault-normal velocity of an
subnetwork. ii) For many sites far from the SAF the velocity uncertainties are
so we cannot rule out the possibility that a fault-normal component of velocity
at these sites.
(3) In the San Luis - Parkfield subnetwork, the USGS trilateration data from 1980
show different trends from the earlier CDMG data (Appendix 11). Thus the more
recent data do not support Harris and Segall 's [1987] model of uniform fault-normal

whole
entire
large,
exists

compression, which was based mostly on the CDMG data. Since neither the CDMG
nor USGS data sets can be rejected, a conclusive statement awaits the analysis of
recent GPS measurements in this region.
(4) Our scatter analysis indicates that the length measurements are not very
homogeneous. The divergences from the scatter model curves are large. Using

Equation (9.1) in Appendix 9, we estimate the coefficients of the error model for the
six subnetworks. Our estimated constant terms (3-6 mm) are larger than the term (3
mm) of Savage and Prescott [1973], while our estimated length-dependent terms are
consistent with theirs (0.2 ppm). Recently, Johnson [1993] also rechecked the errors
of the USGS trilateration data in the Anza, Joshua, and Salton networks, using a
different approach. He reached the same conclusion as we did. Since the difference
between our estimate and the estimate of Savage and Prescott is not too great and
since most of the difference can be eliminated if we downweight a few apparent
abnormal scatters with normalized rms larger than 1.0 (Figure 3), in this analysis we
have adopted the error model of Savage and Prescott without modification.
Minimum-constraintsolution
Two questions cannot be resolved by the outer coordinate solution: 1) What is
the motion of the whole subnetwork? 2) Is there any inconsistency between the outer
coordinate solutions of adjacent subnetworks? In a second analysis of the trilateration
data alone, we attempted to obtain the velocity solution for the entire northern and
southern sections using minimum constraints to control the rigid-body movement of a
whole section. Our purpose is twofold: 1) to look at the relative movement between
subnetworks, and 2) to explore the maximum capability of the current USGS
trilateration data set to provide a homogeneous velocity solution. For a large
network, there is usually no single dominant velocity direction for all sites. Hence the
outer coordinate constraint is difficult to impose. In our analysis we fixed the velocity
at one site and constrained one velocity orientation at another site. We chose sites to
be close to the SAF since the near-fault sites most likely move parallel to the fault.
In the current USGS trilateration network, the connections between
subnetworks are much weaker than the connections within subnetworks. In the
,"
southern section, the Anza-Joshua subnetwork and the Salton-Mexicali suhb
-,rk
are connected through only two sites, Mecca and Laquinta. The connections in the
northern section are even weaker than those in the southern section. Two
quadrilateral structures exist in the Los Padres - Tehachapi subnetwork. Further
north, the Los Padres - Tehachapi subnetwork and the Carrizo subnetwork are
connected by the nearly collinear sites Mckttrck, Crocker, and Caliente, making the
Carrizo subnetwork very vulnerable to small rotational perturbations. The far northern
San Luis - Parkfield subnetwork is connected to the southern part of the northern
section only through site Simmler. In order to strengthen the connections between the
northern networks, we added to the trilateration data the baseline rate vectors

between five sites (Black Hill, Almond, Red Hill, Gould, and FIBR) whose relative
positions were measured in 1989 and 1992 by GPS (Z. Shen, Southern California
Earthquake Center (SCEC), personal communication, 1993). For these GPS-derived
baseline rates, we used uncertainties of about 7 mm/yr, over 10 times the
uncertainties of the EDM baseline length rates in this region. Our constraints are
listed in Table 6 and the solutions are plotted in Figure 4.
In order to examine the fine structures of the velocity field, which are hidden by
the dominant feature of the simple shear, we remove a model velocity field, created
from a multi-fault dislocation model [F93]. This model consists of the SAF, the San
Jacinto fault, the Elsinore fault and the Garlock fault. The SAF is divided into seven
segments The details of the model are described by F93. The residual velocity field
is shown in Figure 5.
From the minimum constraint solution for the EDM data, we conclude the
following:
(1) Simple shear is still the dominant pattern over both the northern and southern
sections.
(2) The sizes of the error ellipses increase rapidly from the center to the edge of a
section, due to the weak connections between subnetworks. At the edge sites, the
error ellipses are so large that all information for these sites is obscured by the errors.
Therefore the possibility of using trilateration data only to obtain a homogeneous
solution for a large network is very limited.
(3) There is differential motion between subnetworks. The apparent west-east
velocity divergence between the San Gabriel and the Los Padres subnetworks could
be explained by the north-south compression in the Ventura basin region [Donnellan
et al., 1993].
Combination of the VG and EDM data
Besides the three VG sites that are collocated with EDM sites, there are six
VG sites located less than 500 meters, and three VG sites located less than 2 km
from nearby EDM sites. Assuming that the velocity varies little over a distance of 2
km, we can set the velocities at these EDM sites to match the velocities at the
nearby VG sites. For an area with a strain rate of 10-7/yr, such an assumption
introduces 0.2 mm/yr error for each velocity component in a 2-km distance. With more
GPS sites in Southern California being measured in recent years, we will eventually
discard the velocity ties between sites more than 500 m apart. All of the constraints

6±7. and (2) rescaling the covariance matrix of the VG solution and the EDM observation to account for the unmodeled systematic deviation. The resultant apparent vortex (Figure 7a) is hard to interpret by the current dislocation model. In the northern section.5±7. In the southern section. The resultant relative velocity field is also similar to the EDM-only solution (Figure 4). To single out which site is most responsible for the incompatibility. (23) of Chapter 2).that we tested in the combination are listed in Table 7. we remove the velocity links between JPLM_GPS and Jpll_rml and between WHIT_GPS and Whitaker because the velocities at Jpll_rml and WHIT_GPS are poorly determined and seem inconsistent with the velocities at nearby sites. a common velocity is subtracted from all velocities. (3) Remove the constraint from the site PINY_uS (Figure 7c). The resultant velocity field is similar to the EDM only solution (Figure 4). and the resultant velocity field using the velocity ties from the 12 VG sites is shown in Figure 6.1±3. The velocity at the site DEAD7267 is changed by 2. both Figures 7b and 7c are the velocities relative to site Asbestos (close to PINY_GPS). If we use the common velocity (see Figure 6) as a reference. there is a significant incompatibility between the velocities at VG sites NIGU_GPS. We have tried to reduce the incompatibility by (1) identifying those velocity-constrained sites with the greatest apparent incompatibility and removing the velocity links from these sites. PINY_GPS and MONP7274. The velocity link between SNP2_GPS and Santapau has also been removed because there are GPS observations for only three epochs. we tested three options: (1) Remove the constraint from the site NIGU_GPS (Figure 7a). It seems that option (3) gives the least change in the velocities of the VG sites in the southern section between the solutions before and after combination. Therefore Figure 6 can be considered as the velocity relative to the SAF. the velocities of the VG sites in the northern section (except the VG sites in the Ventura basin and the off-shore islands) are almost parallel to the local SAF and close to a symmetric distribution across the SAF. We find that after subtracting this common velocity from the F93 solution.3 mm/yr in nc -I-. Figure 7b needs to have added about 1 mm/yr east velocity component to the velocity field of the whole southern section. and Figure 7c needs to have added . In Figure 6.2 mm/yr in north..2 mm/yr in west and 6. This combined velocity solution fails to pass the compatibility criteria at the 95% confidence level (see Eqn. (2) Remove the constraint from the site MONP7274 (Figure 7b). i'jth.the compatible range.6±2.7 mm/yr in west and 14. However. The velocity at the site DEAD7267 is changed by 0. and the east velocity component of the combined solution at Santapau cannot pass our compatibility criterion.

we rescale the formal uncertainties of the VG solution by a factor of three instead of the factor of two suggested by F93 (see Table 8). we choose to remove the velocity link between MONP7274 and Monu_res. Thus we consider the movements at these three islands to be predominantly Pacific Plate motion. It is clear that removing the velocity link at Monu_res leads to a poorly determined velocity field in the Salton-Mexicali subnetwork. to analyze the residual velocity field. i. Thus. Nevertheless. Description of the derived velocity field The most striking feature of the VG-derived velocity field is the fault-normal velocities. the residuals at three offshore island sites (TWIN. from the observed velocity field. we must remove the dominant feature. Since removing these velocity links still does not help enough. This means that the velocity at MONP7274 is less in harmony with the velocities in the northern section than the velocity at PINY_GPS. and to resolving the conflict of the velocity constraints at PINY_GPS and MONP7274. it is easier to discern the origins for the velocity residuals if we put the reference point at the stable interior of the Pacific plate. using the Pacific frame introduces two error sources: the error from the multi-dislocation model and the error from the NUVEL-1 model.about 1 mm/yr west and 8 mm/yr south.e. a model velocity field describing the simple shear associated with the SAF system. Big Bend region Viewed relative to the Pacific Plate (Figure 9a). The resultant velocity field and residuals are shown in Figures 8 and 9. This approach requires us to transfer the observed velocity field from the North American frame to the Pacific frame using the NUVEL-1 model. One could ask why such an obvious pattern has not been seen in the results obtained using USGS trilateration data? To answer this question. The . How to judge the two cases (Figures 7b. as did F93. Where to put the reference point for the velocity residuals should also be considered. BLUF. For now. If the reference point is chosen in the stable interior of the North American plate. and BRSH) are nearly zero. we adopt the Pacific frame. the velocity residuals will include contributions both east and west of the SAF. Adding GPS data from the Salton Trough and Riverside County (STRC) surveys into the VG solution is critical to overcoming the poorly determined velocity field in Salton-Mexicali subnetwork. 7c) is still an open question. Since our data cover primarily the regions west of the SAF and the vicinity of the SAF.

we further zoom the velocity field and residual field relative to the site MPNS (Figure 10a. Figure 10a shows that the fault-normal velocities are confined to a wedge-shaped area. some deformation in the off-shore region must be incorporated to account for the velocity residuals at coastal sites. the velocity at MPNS changes by 2. [1993] for the shortening rate in the Ventura basin. Also. In the Ventura basin itself. SOLI. 10b). Note that the velocity residuals at four coastal sites (PVER. [1990]). and LACU) are very similar (Figure 9a). and the Palos Verdes fault and the NewportInglewood fault near the coast from the Los Angeles basin to the border. :rn . Averaging the velocity residuals at these four sites and projecting this velocity vector onto the direction N17 0 E (perpendicular to the strike of the SAF in the Big Bend as given by Eberhardt-Phillipset al.L . velocity at nearby site MUNS [Donnellan et al. the significant fault-normal . we obtain an estimate of 5±1 mm/yr for the fault-normal velocity. The resultant velocity at MPNS appears more consistent with the velocities at other nearfault sites (Figure 10a). CATO. and 0. But our estimate represents the crustal shortening rate from the coast to the SAF. which is mostly attributed to the off-shore deformation.6 mm/yr in the east with the uncertainty reduced by 1..residuals at all coastal sites cannot be removed by simply modifying the parameters of the multi-dislocation model. For the western Transverse Ranges and Ventura basin region. the velocity at MPNS is determined by observation at only two epochs and seems inc ngrt.. CATO. Thus we still use the average velocity at the four coastal sites (PVER. In the VG-alone solution. Thus the area north of the Ventura basin makes nearly zero contribution to the crustal shortening (see Figure 10a). nearly all of the velocity difference between MPNS and MUNS can be explained by the multi-dislocation model (Figure 10b). In order to better understand the residual velocities onshore. and LACU) to estimate the rate of crustal shortening in the N17 0 E direction.1 mm/yr in north with the uncertainty reduced by 1..7 mm/yr. we replot the field relative to a site (MPNS) near the SAF at the northern end of the Big Bend (Figure 9b). Therefore.. bounded to the northeast by the San Gabriel fault and to the northwest by the Santa Ynez fault. When we add the EDM data. The resultant rate of 5±1 mm/yr is very close to the estimate of Donnellan et al. We find that the velocities at the sites near the SAF in the Big Bend area are basically parallel to the fault (Figure 10a).1993]. SOLI.5 mm/yr. Modifying the parameters of the multi-dislocation model will change the fault-parallel velocities but will have little impact on the fault-normal velocities. the fault-normal velocities can be explained by the multi-dislocation model. West of the Ventura basin. The faults most likely to contribute to the off-shore deformation are the Santa Cruz island fault in the Santa Barbara channel.

East of the SAF Information about the deformation between the SAF and the East California Shear Zone (ECSZ) [Travis and Dokka. . 1990. the Parkfield dilemma has not yet resolved. In order to better represent the near-fault field deformation.1993]. A similar wedge-shaped region is the Los Angeles basin area. Due to insufficient data coverage between east of SAF and ECSZ. Savage et al. should be used. Davis et al. The current model uses a step-increase in locking depth from 1 km to 25 km at 35 0 58'N to approximate the transition from a locked to a creeping zone (see F93). There is some indication that the locking depth may be shallower. and Sung and Jackson [1989]. 1992]. Sung and Jackson [1988].. such as the "smooth transitional" model or "locked" model. as discussed by Harris and Segall [1987]. Southern Coast Ranges In the Parkfield segment (north from 35 0 45'N). 1991. a more realistic model. the cluster of large residuals indicates that our multi-dislocation model suffers from oversimplification (Figure 11). Segall and Harris [1989]. 2) As described in Appendix 11.. the residuals are basically fault-parallel. 1990] is very limited.. At the current stage. The apparent east-west extension in the San Gabriel mountains and the region north of the Santa Ynez fault indicates that the vertical movement beneath the Ventura basin area does not absorb all of the extra material from the north-south compression. In the segment of the SAF from the northern end of the Big Bend to Parkfield. Hence the material in the San Gabriel mountains is pushed eastward. we cannot discern if the deformation is uniformly distributed in this region or is concentrated in the ECSZ area. southeast of the Ventura basin area. as F93 inferred by comparing the velocity residuals at MADC and FIBR. Comparing the velocity residual of MOJA with the velocity residuals of the sites east of SAF in the Big Bend area. which also undergoes north-south compression [Shen. the difference represents the accumulated deformation from east of the SAF to MOJA.residuals indicate that there must be other mechanisms to account for the crustal shortening [Donnellan et al. we do not attempt to modify the multi-dislocation model for two reasons: 1) The residual uncertainties in the Parkfield region are too large. and the material in the region north of the Santa Ynez fault is pushed westward. Can we also see the east-west extension along both flanks of the wedge? We cannot answer this question without incorporating additional space-geodetic data and terrestrial survey data.

The seismicity time sequences indicate active interaction between the cross-faults and the main faults.geodetic measurements over a large area and the spatial density of trilateration measurements to derive the horizontal velocity field for southern California. they proposed a right-lateral shear plane at N400 W and a locking depth of 10 km across this area. ..blems with the current data sets. The strength of the combination for most of the region can be seen by comparing Figure 9b with the minimum constraint solution shown in Figure 5. Most sites which provide the velocity constraints are west of the SAF. In such a tectonically active area. Therefore our combined velocity solution is poorly constrained east of the SAF. The minimum constraint solution (Figure 4) shows that significant deformation occurs in a narrow corridor from the SAF to the San Jacinto fault. Conclusions We have taken advantage of the high accuracy . a series of parallel left-lateral faults trend northeast. Lisowski et al. Larsen and Reilinger [1992] also noticed a narrow deformation zone in the Imperial Valley.Southern section For the southern section. To explain the observed 1988-1989 GPS station displacements in the valley. In this area. Our results can be described from three perspectives. Mexicali subnetworks in the southern section and the San Luis subnetwork in the northern section are also not well constrained. Abnormal uplift in this area had been reported [Reilinger. 1985] but not resolved. The Joshua. Such a "deformation corridor" is clearly seen in strain rate and rotation rate figures (Figure 12). the combination solution reveals several important r-. nearly orthogonal to the main right-lateral faults [see Figure 1 of Hudnut et al. [1991] interpreted the steep velocity gradient at the San Jacinto fault by a shallow locking depth of 5-6 km near Anza or by a broad fault zone with low rigidity. First. (1) The current VG site distribution cannot adequately constrain the trilateration data. Salton. 1989]. The combination field allows us to explore the crustal deformation at the transition zone of the plate boundary from a wide-angle view. a denser GPS network is likely to allow us to observe directly any block rotation. and probably extends to the Imperial fault. the VG data are insufficient to provide a useful constraint. Johnson [1993] found the deformation corridor is dominated by the San Jacinto fault instead of the SAF.-.

Although the weakness of our current data set impedes us from deriving a strong horizontal velocity field in southern California. (2) Estimated coseismic site displacements (Appendix 12). (2) Our current multi-dislocation model should be modified. The updated coordinates are not yet good enough to directly improve the velocity estimates. The observed fault-normal velocity from the combined data set in this region is not significantly different from zero. Third. In the Parkfield area. the step-function transition model should be replaced by the smooth transition model or more sophisticated model. we can still reach some useful conclusions about deformation in southern California. (1) Through removing a model velocity field from a buried multi-dislocation [F93]. These problems indicate that there are still some unsolved systematic differences between terrestrial and space-geodetic data. Hence inverting for fault slip parameters using the trilateration data in this region should be done with great caution. both of which are very strong sites in the VG solution. The CDMG data from 1970 to 1979 are inconsistent with the USGS data from 1980 to 1990. our approach also produces two important byproducts: (1) Updated coordinates for the USGS trilateration sites (Table 2). and between GPS and VLBI. which can be used to compare with the site displacements predicted by the seismic slip model. Second. we are in a good position to realize such a dream for four reasons: (1) We have developed a rigorous and flexible methodology to handle the combination. (2) All software to perform the combination is ready and eager to work. the SAF probably changes its orientation and joins the Imperial fault along the Brawley seismic zone. in spite of the weakness in the current combination. . the configuration of the San Luis subnetwork is weak for deriving a velocity solution (see Appendix 8).(2) Our combination shows a significant conflict between the velocity solutions at PINY_GPS and MONP7274. (3) In the San Luis subnetwork. We find that the narrow aperture of the EDM network may be the reason why the trilateration data cannot see the fault-normal velocity. There are also conflicts between the EDM and VG solution for some sites whose velocities are less reliably determined. In the southern end. we explore the distribution of the fault-normal velocities relative to the Big Bend segment of the SAF. Furthermore. but they facilitate further research in combining the trilateration data and the space-geodetic data.

. data from a much denser network of GPS sites will be available (Figure 17).44 (3) We have acquired experience in combining terrestrial and space-geodetic data. (4) Within a few years.

K. 40-56. Global Positioning System An Overview. Hager. Herring. 40-56. Role of Eastern California Shear Zone in accommodating Pacific-North America plate motion. Geophys. Geophys. T. L. 425-478.. 71. segment of the San Andreas faul. Un. T. Amer. A. W. K. 101. H. J. J. B. Strange. Bock. D.. L. T. Res. Z. Skrumeda. and P.. J. L. H. A. Intnl. N. Agnew. Himwich. P. R. Geophys. and F. 17.. Y. R. Geophys. Geodetic measurement of shortening across the Ventura basin. W. Travis. 1323-1326. King. H. Eos Trans. Donnellan. J. M. Measurement of the velocity field of central and southern California. and M. submitted to J. Herring. Crustal strain near the Big Bend of the San Andreas fault: analysis of the Los Padres-Tehachapi trilateration networks. King. K. California. E. 1993 Eberhardt-Phillips. 1987 Herring. L. Geophys. B. H. Murray. W. Zoback. Segall. G. Gordon. Blewitt. R. R. 1990 Dokka. 1992 Demets.. Res. J.. 1051-1056. H. D. Murray. D. Detection of a locked zone at depth on the Parkfield. J.. Lisowski.. Springer.. New York. & Leppard. GLOBK: Global Kalman filter VLBI and GPS analysis program. and W. submitted. 1990 Donnellan. A. 1990 Feigl. (eds). Dong. 351382.. 12561-12581.. southern California. Agnew. Hager. Geophys. D. Annu. Stein. Lett. D. Shen. J. 7945-7962. and S. I. 1991 Harris. Jackson. D. A. Hager. unpublished documentation. S. GPS measurements of regional deformation in California: some constraints on performance.. 1990 Davis. Jordan. R. Larsen. 1993 . Res. 95. 1990 Herring. Res. and I. C. T.. and W. M. Y. and C. and T. D. K. 1139-1153. 95. Geophys... Prescott. A. and M. Res. A. Space-geodetic determination of crustal deformation in the Los Angeles Basin.REFERENCES: Bock. Webb. F. California. 1984-1992. Larson. B. submitted to J. Measurement of crustal deformation using the Global Positioning System. M. C. Argus.. G. Rev. R. Kroker. 1993 Hager.. Larson. H.. 92. King. H. 19. Geodesy by radio astronomy: the application of kalman filtering to very long baseline interferometry. Shapiro. W. Res. Sci. 1990 Dixon. R. A. D. H. MIT. EarthPlanet. Current plate motions. Davis. B. Geophys. T. Science.

Pacheco. 2119-2133. Geophys. Geophys.. E. C. Global Positioning System Measurements of deformations associated with the 1987 Superstition Hills earthquake: evidence for conjugate faulting. 97.. and B. EOS Trans. Geophys. California. O.. Geodetic Measurements near Parkfield. Seeber. Mexico. 97. C.. Johnson. Techniques and studies in crustal deformation. Hager. 95... Soc. August 10-12. R. Vector constraints on quantenary deformation of the western United States east and west of the san Andreas fault.. Res. and Y.. 8865-8876. 4885-4902. R. and W. Mexico. 957. 1992 Larsen. J. 1990 Johnson. R. Paleontol. S. California. AGU. Bock. earthquake swarm (abstract). Pac. Reilinger. 1-16. V. K. Sect. Rice. M. 1987 King. P. H.D. 2747-2766. Res. Cross-fault triggering in the November 1987 Superstition Hills earthquake sequence. Scripps Institution of Oceanography. Savage and W. 1992 Larsen.Mineral. 1992 Larsen. Thesis. C. and W. 98. Preliminary study of the Westmorland. W. Neugebauer. 1959-1984. Massachusetts Institute of Technology. K. Mexico. J.. Reilinger. J. 11533-11551. 1989 Hutton. 1987 Lisowski. Res.2 Victoria. San Diago. H. 1991 Minster. J. H. The velocity field along the San Andreas fault in central and Southern California. Guadalajora. Geophys. Strain accumulation in the Santa Barbara channel: 1970-1988. N. Geophys. 38. 62. Econ. Prescott. Geophys.. 1993 King. California: 1986-1989. Crustal deformation in great California earthquake cycles. and P. 1982 Lisowski. J.. California. H. Res. Global Positioning System Measurements of strain accumulation across the Imperial Valley. 96. J. Bachman. S. Ph. Jalisco. L. B.. and J.. Strange. Tectonics and Sedimentation along CaliforniaMargin. Prescott. Lett. J. 16. 1981 Jackson. earthquake. Segall. Jordan.. and R. Res. 22073-22087. G. H.. J. Deformation near the epicenter of the 9 June 1980 ML = 6. edited by J. M. 1993 Li. 83698389.. Documentation for the MIT GPS analysis software: GAMIT.. B. K. and J.. 1984 . W. J. Proceedings of the Fourth Symposium of the Cerro Prieto geothermal field.. Baja California. Crouch and S. D. University of California. southern California.. S. H. Res. 199-202. R. L. J. Active faulting and block rotations in the western Transverse Ranges. Prescott. Agnew. Geophys. E. Res.Hudnut. 92. 92.. Molnar. and C. and T. and W.

Drew.. Lisowski. Bull. Geophys. H. High PrecisionNavigation -Integration of Navigationaland Geodetic Methods. Murray. earthquake (Ms = 7. 1991 Snay. Lett. A. Prescott and M. Res. and T.. Lett. Geophys. 1973 Savage.. R.. 1985 Savage. P. D. 1986 Savage. J. Lisowski. thesis. Geophys. The determination of displacement field from geodetic data along a strike slip fault. J. 91. 28-62. D. and R. 623-626.. J. 86. H.. 225-236. W.-Y. H. H. J. The North American Datum of 1983: project methodology and execution.. Lisowski. and E. J. 2113-2116. and A. 64.Minster.5). Deformation along the San Andreas fault in 1982-1986 as indicated by frequent Geodolite measurements.. Strain accumulation in southern California. M. R. H. Lett.. B. Jordan. C. California.. M. and W. Deformation from 1973 through 1991 in the epicentral area of the 1992 Landers. J. California. Geophys. Lisowski. Geophys. W. 1987 Murray. 16. neotectonics. Los Angeles. 1987 Savage. Harris.New York. R. M. 12. 1981 Reilinger. Res.. Precision of geodolite distance measurements for determining fault movements. inferred from terrestrial geodesy and the Global Positioning System. Geophys. B. Geod. Res. M. W. 1990 Segall. H.. J. J. C. 78. Coseismic displacements: 1992 Landers. Gross. Vector constraints on western U. Geophys. C. 1993 Schwarz. H. 6067-6072. Geophys.. 6001-6008. Lett. Prescott . K. Gu. Res. C. A. 1990 Savage. Regional tectonic deformation in southern California. J. Res. 7455-7473. Imperial Valley. 20. Ph. K. 92 47984804. Z.. 92. Jackson. S. and W. Prescott and G. Combining GPS and classical geodetic surveys for crustal deformation in the Imperial Valley. and plate motions. California. C. Savage. California' by L. 4785-4797.. Comment on: 'Geodetic evidence for seismic potential at Parkfield.. Geophys. Springer-Verlag. Prescott. 17. 1993 Prescott. J. H.. 1989 . A strain anomaly near the southern end of the San Andreas fault. Sung and D. California. earthquake. 1989 Shen. eastern California. Res. University of California. J. 101-104. C. Res. submitted to J. and M. 1973-1984. Geophys. An apparent shear zone trending north-northwest across the Mojave desert into Owens Valley.. Res. Res.. Res. and W. J. 561-564. C. Wade. R. deformation from space geodesy.

. Jackson. Y... 15. L. and E. and D.. Tectonics. Project REDEAM: Models for Historical Horizontal Deformation.. A kinematic model of southern California. Jackson. Geophys. S. Lett. R. Pacific-North America plate motions: New results from very long baseline interferometry. 1987 Sung. L. Cline and E. 1989 Ward. Geodetic evidence of the seismic potential at Parkfield. 95. 1988 Sung. 105-107. A. D.. California. R. 21965-21982. Geophys. and D. 16.48 Snay. Geophys. Timmerman. N. Lett.. 33-48. Res. J. NOAA Technical Report NOS 125 NGS 42. 1990 Weldon. 5. Res. Reply. M. Res. D. L. 820-823. Humphreys.. Y. 1986 . W.

The conversion of (9.Appendices Appendix 1: Coordinate frames (a) Geocentric Cartesian coordinate (X. A) and geodetic ((P. The vector in a local topocentric coordinate frame is transformed to a geocentric Cartesian frame via a time-variable rotation matrix R(t): (1. (b) Geodetic coordinates (0.sin Q(t) cos A(t) . [1988]. 0 cos A(t) . the X-axis is towards mean Greenwich zero meridian.xi(t) is defined in a local topocentric frame at site i.xj(t) . X. and geodetic height h(t) as its components. geodetic longitude X(t) with positive east. (c) Local topocentric coordinates (x. X)coordinates . The differences between astronomic ((D. h) to (X. Y. 1989] and different from that of Collier et al.1) Z(t) = (N(1-e 2 ) + h(t)) sin (p(t) where N is the radius of curvature in the prime vertical and e is the first eccentricity of the ellipsoid. and the z-axis coincides with local zenith. z) The x-axis is directed to local astronomical east. h) This coordinate system includes geodetic latitude 0(t) with positive north. Such a definition leads to a right-hand system which is the same that used by DYNAP [Drew and Snay.sin A(t) R(t) = . Z) We use the IERS Terrestrial Reference System (ITRS). the y-axis is directed to local astronomical north.2) dxij(t) = R(t) dXU(t) where the bold character represents vector. The Z-axis is aligned with the pole of the ITRS. Z) is: X(t) = (N + h(t)) cos (p(t) cos X(t) Y(t) = (N + h(t)) cos (p(t) sin X(t) (1. y. X. dXij(t) = Xj(t) . and the Y-axis points east 900. Y.sin 1(t) sin A(t) cos D(t) cos A(t) cos CD(t) sin A(t) cos D(t) sin D(t) where Q4(t) is astronomic latitude at site i.Xi(t) dxy(t). A(t) is astronomic longitude at site i.

Yo. rl is the east-west component of deflection. One reason is that there is no detailed velocity contour of southern California at present. ox. Appendix 2: Specific linearized observation equations In this section. rx is the translation. Another reason is that the value of this term is usually very small if the coordinate adjustments are at meter level or less. with the downward vertical deviation towards the west being positive. [1988]. O. 1) Site displacement and velocity If the observables are in a geocentric Cartesian frame. 0 -zo Yo =zo 0 -xo -yo xo 0 13) (2. Tx.4) are called deflections of the vertical. [1985] and Herring [1983] respectively. Derivations of the terrestrial observation equations are similar to the work by Collier et al.-(p= . . the term representing the velocity gradient is not used. Observation equations for carrier beat phase and group delay can be found in King et al. o)x is the rotation angle. the equations are aX = Lx 8Xo Vo +LX 10. we present the linearized observation equations for the quasiobservations (site position and velocity) and for terrestrial observations.3) xo. o) is the rotation angle rate. Currently. 4 is the north-south component of deflection.1) Og0 8V = Lv 8Xo 8Vo w 0 g cov x v where L = 13 + axo (t-to) ( 133 t-to) L = axo 13 is the (3 x 3) identity matrix. coy. A-k= (1. rv is the translation rate.2) (2.g 0 00 ITxv (2. with the downward vertical deviation towards the south being positive. are common for all sites. zo are the a priori coordinates.

latitude i.5) becomes a Bg =- 0 go -1 0 0 0 Ro M Lx 8Xo (2. 2) Baseline and baseline rate vectors: If the observed vector is in a geocentric Cartesian frame.If the observables are in a local topocentric frame. (dVij) = Lv 8Vo . so that (2. The linearized observation equations of triangulation. and M =- ax 2 is the Marussi tensor of second-order derivatives of U [Hein. and the topographic gravity anomaly for g [Hein. D and g are used.4) 8Voj . If the observed vector is in local frame. Lx is replaced by L.6). defined in Appendix 1 (1. (Appendix 1). then the AUo term in (2.5) where go is the apriori gravity.8 (2. 3) Astronomic longitude A. and the vertical angle is 3(t). (dXij)= Lx . the distance is 1(t).6) 1 1V If the quasi-observable rates of A. 0ois the apriori geodetic latitude. We prefer to absorb the deflections of the vertical and the topographic gravity anomaly correction into the residuals. 4) Horizontal and vertical angles and mark-to-mark distances From local site i to site j. in (2.3) as a function of the astronomic latitude (I)o and longitude Ao.8Vo. If the disturbing potential AUo is dominated by known mass anomalies. 1986]. 1986]. and trilateration are expressed as: .5) represents the deflections of the vertical for A and D. the azimuth is a(t). 13 is again replaced by Ro in Lx and I . and gravity g: =I A g 0 (gocosgo)0 go 1 0L 0 o ML 0 8Xo 8voI a(AU(X)) X /1 (2. . I3 is replaced by the rotation matrix Ro. leveling.8Vo.

but the latter must construct a full instead of a diagonal covariance matrix to describe the correlations among observations [Prescott. with Lx replaced by L.9) 0 and the zero subscript indicates values calculated from a priori coordinates and velocities.10) if t < tk < to if t>tk.tk) 8 Vo k if t > tk > to Where tk is the occurrence epoch of the k-th event. For direction observations. go lo sinao sinpo 0 ao locoso 0 0 0 (2. This approach is equivalent to the angle difference method. all the above equations should be modified by adding the episodic site displacement terms.8) sin o0 cos 0o K= KRo MLx 8X01 8Vo. such as coseismic displacements.tk>to (2. FONDA adopts auxiliary parameters to account for the unknown arbitrary azimuth of the first "pointing" in each "zero". In the case of known episodic events. The observation equations of quasi-observable rates of a.cosao sinoo) lsio (2. and 68k is the site displacement vector from the k-th event.7).. the residual terms of (2.SlocospoP 8X ( 06 81 where 0 0) 0 lo- 0 0 0 1 aX SX ST Ro LxX 8VoI-SVol (o -sin ao sin Do sin ao cos Do S = -sin ao -cos ao sin 1o cos ao cos 3o cos 0 l10 (tanto cospo . it is possible to resolve them simultaneously with other parameters.11) Xo + I rk(t. 3 and I are similar to (2.tk<tO or t<tk.tk) = -1 0 1 6 k (2.7) include the terms for the deflection of the vertical. 1976].7) (2. As mentioned before.. We note that the estimated . If the displacements occupy only a limited area of the whole network. are estimated as step functions in site position. The time-dependent positions can then be expressed as 8X(t) = Lx rk(t. 5) Episodic site displacement vector Episodic deformations.

4) It is easy to prove that if we construct a quasi-observation equation =B x with covariance matrix C-y (3. For example.3) and (3. in trilateration observations if one site has only one distance observation around the earthquake epoch.y)' A Cil 1 (3. Appendix 3: Reparameterization Let x(mx) and y(my) be two different parameterizations of the same observations.7) The solution is: = (BT C B B Cy Cx = (BT CI B) = (3.6) Research on the general relations between the two solutions is beyond the scope of this thesis.4) = (3. Our estimated episodic displacement field is independent of the fault model.8) (3. Another advantage is that this approach can estimate the site displacement caused by other sources.5) = (AT C11 Ay) (3. l(n) I = Ax x + Ex with covariance C l 1= Ay y + Ey with covariance CI (3.episodic site displacement vector does not always represent the actual displacement of the benchmark.3) Cx =(AT Ci Ax)"1 (3. We choose not to estimate fault-slip parameters directly in order to avoid biasing our velocity estimates through an inadequacy in the fault-slip model. 1) Use y and Cy as quasi-observationsto obtain the final solution (3. If a whole network is shifted by an earthquake.4) S= (A Ci1 A.2) The least square estimates are well-known: = (AT CiI Ax)' A T Ci 1 (3.3) (3. the estimated site displacements from trilateration or triangulation data will contain the ambiguities of network translation and rotation.1) (3. such as benchmark instability. Here we discuss only the applications of reparameterization related to our methodology. Therefore the geodetically estimated coseismic displacement field provides a useful comparison to the seismically derived fault-slip model.9) . the estimated k is the projection of the benchmark displacement vector onto the baseline vector.

After some manipulations.15) If the parameters Xa actually do not exist..^ = Q12 Q212 Q21 (3. = A Bx.The great benefit is that we do not need to keep and reprocess the original observation data.10) To distinguish from (3..ate the variance of x and lead to a biased estimate Runless the expectation of xa is zero. we obtain the relations x = + Q12 Q22 ia (3. adding stochastic parameters.8) and (3. it is straightforward to get Q_. to get the optimal solution. Q 12 =.X (3.12) where C = N11. polar motion. for example. UT1. 2) Use y and Cy as quasi-observationsand add extra parametersXa to x The quasi-observation equation is y =(Bx Bx) ( (3..15)). and x2 station coordinates and velocities. .5) and its covariance matrix (3.11).N 2 1 Ni 11 N12)" = (3. We keep only the intermediate solution (3. quasar position. the omission of xa will undere. we can choose to test different final parameter sets x.16) From (3. i is still an unbiased estimate of x..N 11 N 12 Q 22 = 2T1 N22 = AT C 1 Ax. and terrestrial survey data as . The typical example is to consider spacegeodetic data as the first data set.13) CZ- From (3.9).17) That means that even if the additional parameters Xa do not exist.16) E( ) = E 'R) + Q12 Q22 (Q21 A CI + Q22 A T. On the other hand. Ax. etc.14) Then C = C + CQ'. with xl satellite orbit.11) and using (3.C1i) E(l) = + Q12 Q1(Q2 1 N11 +Q 22 N21 )X = X (3. = N 12 = AT C I Ax. Furthermore. 3) Two data sets related to two sets of parameters with only part of the parameters in common Assume the first data set is related to parameters xl and x2 and the second data set is related to parameters x2 and x3. Q.10). Q22 = (N 22 .6). the expectation of the observation is E(1) = Ax K (3.11) CQ = Cx + Q12 Q212 Q21 (3. But its covariance is enlarged (see (3.. if the additional parameters xa do exist. we use R and Cx to denote the estimated subsolutions of (3.

the second data set with x3 new station coordinates and velocities. We still treat the
solution of space-geodetic data as quasi-observations. In this case the quasiobservation equations are

0

%

with covariance matrix C = C11

2

E2

The combined normal equations are

N2E2

N111
EIN21

A32

0

x2

(3.18)

2

(3.19)

N)x+N1

0)

E1N 22E 2 +A 22 A2 3

C12

=

B2 +ETN 2 1~1+EIN 22'
B3

A33

Here N = C- 1.
Solving xl implicitly and using the matrix partition formula, we obtain

ETCi2E 2+A22 A23
A33

A32

B2+ETCi2 2

2
3

(3.20)

B3

This important result shows that if we do not care about the estimate of xl, we can
use only the common parameters x2 and their covariance submatrix C22 as the quasiobservations. The result is as rigorous as using all parameters.
4) Use part of the solutions to estimate another set of parameters
Assuming the solutions include parameters xl and x2, we attempt to use x2 to
estimate another parameter set u2. A typical example is to estimate fault slips from
episodic site displacement solutions.

(1 )(I
x2)

E
0(

2

X)

with covariance matrix C= C11
C21

C12)

(3.21)

C22

As has been proved above, if we use x2 and its covariance submatrix C22 as the
quasi-observations, we can obtain the same estimates by directly estimating xi and
u2 from the raw data. But this time we do not update the solutions of xl, since we
know that the episodic site displacements more accurately describe the real episodic
deformation field than the fault slip parameters. We want to leave the model errors in
the residuals instead of distorting the estimate of xl. Here the quasi-observation
approach gives us the flexibility to obtain the optimal choice.

Appendix 4: Comparison between
approaches

the Gauss-Markov model and other

If we consider the velocity derived from space geodesy as a model velocity,
both the Gauss-Markov model and model coordinate method can be summarized as
minimizing
S(811-A1 8X)T Cij (811-A 1 8X) + (812 -A2 8X)T C2 (812-A2 8X)

(4.1)

where 11 represents the terrestrial observation with covariance C 1 and design matrix
A 1 , and 12 represents the space-geodetic solution with covariance C2 and design
matrix A2.
The Lagranian multiplier X is equivalent to an externally imposed relative
weighting factor between 11 and 12. The choice of X is rather philosophical, and reflects
the balance between believing the formal uncertainties of the estimated velocities and
imposing the prior model on the final solution. The Gauss-Markov model represents
the case of X = 1, which means the estimated formal covariance matrices are accepted.
The model coordinate method corresponds to the case of X = o,, placing maximum
weight on fitting the terrestrial survey data. Finally X = 0 represents the case of fixing
the velocities to the model i.e., space-geodetically derived velocities. For our
analysis, we choose the Gauss-Markov model to perform the combination. Because
the space-geodetically derived velocities are from the real data, not from a "pure"
model, neither X = oo nor . = 0 correctly represents the statistical character of the
space-geodetically derived velocities.
Appendix 5: Solution changes in the case of adding new data and new
parameters
Assume the original data 11 relate to the parameters x, and the new data 12
relate to parameters x and y. x1 are the original estimate and x +2, Y1+2 are estimates
from data 11+12.
11 = A1 x + E with covariance C11
(5.1)
(5.2)

12 = B 1 x + B2 y + e2 with covariance C12
The covariance matrices are

, =(A T Ci A)

1

ATCiAI+BTICiB

C(^
1 )=

BICB

Cl., = Q11

(5.3)
1

BTC B 2 -1

N11

BCiJB 2

N2 1 N 22

N12

1

Q11

Q12

Q21 Q22

(5.4)
(5.5)

The solution change of parameters x can be derived as
1+2 - X= (Q 11 -(ATCiA

1

) ATC1

11 + (Q1 BT+Q 12

) C12 12

(5.6)

Since the two data sets are uncorrelated,
,,X111

T

1

T 1-1

- 1 T 1-1

-1

fV2-

= (Q11N11 + Q12 N21 - I) Qll - Qll + Cx, + (Q1 N1 2 + Q12 N 22 ) Q21
From the well-known relations,
Q11NI1 + Q12 N2 1 = I

and

Q 11N 12 + Q12 N22 = 0

We derive the very simple result
CX1+2-r

= Cx1 - CX1+2

(5.7)

Appendix 6: Procedures for updating poorly known coordinates in a
terrestrial network
In order to update the coordinates of a terrestrial survey network, there must
be at least three non-collinear sites that have also been observed by space-geodetic
techniques. Theoretically, two space-geodetic sites can control the translation and
rotation of the network provided there is no figure defect. The third site is necessary
to avoid divergence during the iteration. If this condition is satisfied, then a
bootstrapping strategy may be applied to update the poorly known terrestrial
coordinate system. For simplicity, we use a trilateration network as an example. We
adopt the following procedures:
1) Perform a transformation to align the original trilateration coordinate system to
the space-geodetic reference frame. The parameters are estimated from the
coordinate differences of the common sites. Currently, we estimate only the
translation and rotation, and neglect the scale factor. Due to an error in the velocity of
light used to convert the EDM measurements to ranges, they are too long by 0.14
ppm [M. Lisowski, personal communication, 1993], causing about 4 mm error for a 30
km baseline. We have not applied this (relatively small) correction to our estimates
of site coordinates.
2) Compress the trilateration data into baseline lengths related to weighted mid
epochs and baseline length rates. Use baseline length rates as quasi-observations to
estimate site velocities.
3) Using the compressed data, tight outlier identification criteria, and tightly
constrained coordinates at the common sites, estimate and update the remaining site
coordinates.

for which the length and its rate are uncorrelated.1) CV . the estimated coordinates and velocities are in general correlated. Their covariance matrix is C and the reference epoch is to. We prefer to obtain updated coordinates that are approximately uncorrelated with velocities. C= (6.5) Contribution of nominal lengths to the estimate of velocity A time series of distance measurements can be compressed into one baseline length rate and one length at a weighted mid epoch. In this case. We use the weighted average estimate CXxV.2) V The epoch offset At is selected so that the off-diagonal terms of C' are close to zero. Then to = to + At X'=X+VAt.4) C'=JCJT (6.2) (6.4) Repeat step 3) with updated coordinates.x. + At =- + 1 1 1 Cxx. To realize this. This procedure continues until no more data fail to pass the criteria. 5) Using the initial raw data and realistic constraints on the coordinates and velocities of all sites. Where C' = J C Jr. There are limits on the ability of terrestrial survey data to update coordinates. Appendix 7: (6. trilateration data is not sensitive to height change. In the final step. Assume that the estimated coordinates and velocities for one site are X and V.. - At (X 0 1 V 1 = JX V (6.3) (6.V. For example. the heights should be tightly constrained in steps 3) and 4) and estimated only in the last step to get limited improvement. Cxy CX. the linearized observation equations are . we modify the reference epoch for each site coordinate. The uncorrelated coordinate is X' and the modified reference epoch is to. Cx. estimate coordinates and velocities simultaneously.xy Cx. gradually loosening the outlier identification criteria. Using all nominal lengths and their rates to estimate site coordinate adjustments and site velocities.

2. A22 = The maximum contribution of L to V (ax -. A12 = At. CL = (1 mm/year)2. the compressed baseline lengths L can reduce the formal uncertainty by about 30%. All.3) reveals the following: 1. Appendix 8: Sensitivity testing . Consider the simplest case of 1-d.L A A2 X with C CL 0 CL A0 8V A22 (7.0. ov is the apriori uncertainty on V. At < 5 years.1) with The normal equations are: ATICL Aii1+ AT2 Ci-A 1 AT2C'A AI CLA12 X CA22 8V 2 +A 2 ATC L A 2CLAL+A22C (7. Hence poor a priori coordinates should be updated before estimating velocities.2) where ox is the apriori uncertainty on X. For most baselines.3) 2CA22+ +A C CL+A lo2xA11) 'A12] Equation (7. oy = 200 mm/year At is the difference between the reference epoch and the weighted mid epoch. Using only rates is equivalent to the case of ox -*oo. the accuracies of a priori coordinates should reach the centimeter level. Given the scalar values: 1. (7. 16 + (At)2 be determined by The formal uncertainty of the estimated velocity from original 1 (considering only the length rates) to 1-bt 16 + (At) ) 5. The current updated USGS trilateration site coordinates (ax) are considered as accurate as 1 to 10 meters. Using the above example with a. Solving for 8X implicitly. The maximum contribution of L to V is the case of ox -. L and L are the compressed baseline length and baseline length rate respectively. 4. CL = (4 mm)2 .2) becomes: 8V = A22 C AL+AT 2(CL+ATlloxA11)-AL (7. To get a meaningful contribution to V from L. it can bias significantly the velocity estimate. = 1 cm and At = 5 year. 3. Therefore the contribution of L to V is very small.0) can 8V = 16 AL 16 AL + + At At AL AL. If the misfit of L (AL) is too large.

so .1) - In this special case. where x and s are adjusted and unadjusted parameters respectively. we do not discuss the theory of optimal network design. = (AX Cf Ax + x " Usually we choose the parameter model as its a priori value xm = xo. we use a simplified methodology to estimate the sensitivity of a given network. equation (8. The difference 8s between the true value of s and the a priori value. there is also no sensitivity. s). if the baselines are nearly perpendicular to the fault. the solution is sensitive to small perturbations. all velocities are expected to be parallel to the y-axis. because cos 0 = 0. s o. If vy has a constant gradient in the x direction. Rather.L(xo. The observable is the baseline length rate dL/dt.Network configuration affects the sensitivity of the adjustments to the signal to be detected. However.Svx2 ) + cos 0 (t-to) (8vyl . If the baselines are perpendicular to the fault (0 = 900). Let the signal be the velocity field induced by a dislocation model of an infinitely long strike-slip fault in a half-space.2) The computed linearized observation equation is l(t) = L(x. there is no sensitivity since all measurements are always zero.3) with observation covariance C1 and model covariance Cxo. If the baselines are parallel to the fault orientation (0 = 00). the optimal baseline orientation is 0 = 450. Take the fault orientation to be along the y-axis in a local 2-D reference frame. In this Appendix. The baseline makes an angle of 0 to the yaxis. The estimated adjustments are C= Q (A Cf1 + Cx (Xm.8vy 2 ) (8.X)) (8. The linearized observation equation can be simplified to aL aL dL tL dL L(t-to) Vx1 + (t-to) 8vx2+-x(t-to) 8Vy 1 + (t-tO) 8Vy 2 dt axl ax 2 ayl aY2 = sin 0 (t-to) (8vxi .4) then becomes . 8s = s . s. To quantify the above statement.4) with (8. to) = Ax(t) Sx + ex(t) (8. represents the systematic error.so (8.5) C. For a narrow network nearly perpendicular to a strike-slip fault. t) . the observable L is described as L = L(x. and all measurements are still zero to first order. the solution is sensitive to small changes in the observations even if some diagonal baselines in the network have the optimal angle of 0 = 450.

L(xo. s. observation equation should be It(t) = L(x.6) is biased. it is important to check whether the different networks have different data quality due to different topography. or operators.Li(to). In this chapter.. For each baseline. Appendix 9: Error model The commonly accepted model for the errors in EDM observations has the form (9.(8.xt = S 8s (8. we use (8. This kind of systematic errors affects all measurements in a particular survey but is related randomly to the systematic errors of other surveys due to frequent recalibrations. t) .5 mm and b = 0. The data quality of each network is assessed by the scatters about the best-fit linear trend of the baseline length observations.1) . the scatter is defined as I si = wj (Li(tj) .Li (tj . Systematic errors occur in individual surveys due to improper calibration of the meteorological probes.9) and (8.2) .= a2 + b 2 L2 where L is the baseline length.2 ppm [Savage and Prescott 1973].9) where (8.7) Therefore. a = 3 mm and b = 0. the true solution should be Xi-t = C' (AT C. to) = l(t) + As(t) 8s The true linearized (8. the estimate (8.6) i = C A C' 1 Due to the systematic error. 1986].10) to test the perturbation from each isolated systematic change of baseline length rate.to) n -n j=1 (9.8) (8. Since our analysis attempts to estimate the velocity field from multiple trilateration networks simultaneously.1 + AT C1 1 As 8s) The perturbation from systematic error can be written px = r . Therefore the systematic component of errors can be treated as random errors in the analysis of multiple survey data.C AT C 1 As is known as the sensitivity matrix. which includes the systematic errors with a = 0. s. instrumentation. For USGS EDM data with line-of-sight atmospheric calibration.10) S =.14 ppm [Savage et al.

California. Note that these scatters reflect not only the quality of the observations. but also the appropriateness of the linear trend.3) where L is the baseline length. providing that all residuals are independent and follow a normal distribution. we used several collocated VG site coordinates from the solution of F93. [1989] modeled the differences of baseline lengths between the two techniques using the equation AL = a + b L (9. we obtain a series of scatters from the baselines with three or more observations. any timevarying deformation will be treated as noise. While updating coordinates.1 ± 0. j is the observation index for this baseline. Performing a least squares adjustment to these scatters by fitting the error model (9. Li(to). The systematic differences between VG and EDM data should be discussed. Li are the estimated length at epoch to and the length rate.6 ± 0. Montana. i is the baseline index.1 ppm) do not significantly differ from zero. Analysis of these coordinates gives an accuracy of about a decimeter. Therefore. the result is assigned as the uncertainty for the scatter.5 mm) and b (0. Appendix 10: Procedures for updating coordinates To update the coordinates of the USGS trilateration sites. Their results show that both a (0. w= J The squares of the residuals obey the X2 distribution with degrees of freedom n-2. We also used coordinates determined for several collocated GPS sites from GPS observations made by USGS in the Parkfield area and by SCEC in the Gorman region. Therefore the upper and lower bounds of each scatter are obtained from the X2 distribution with a specified confidence level a. respectively. Thus we neglect the effects of the systematic differences between VG and EDM data in this analysis. Thus for each network. Choosing a=0. we . we get the estimated values of a and b for each network. The relative horizontal positions of these VG sites have accuracies better than 2 cm. Davis et al.68 and averaging the distances from the upper and lower bounds. Based on the GPS and EDM measurements performed at Loma Prieta.1). and Hebgen Lake.where Li(t) is the observed baseline length.

both aspects affect an inversion for a fault slip model. Omitting the CDWR data from 1959 to 1969 does not change the solution significantly. decreasing X2 by 25%. Bench-Bonnie. It is the different data sets that cause the discrepancy. After checking part of the original survey log sheets. This discrepancy between our solution and HS's needs to be explored. We use the Geodolite data from 1980 to 1984 and the CDMG survey data between 1971 and 1979 to derive an outer coordinate velocity solution with a minimum velocity direction of N500 E (Figure 13b). using the CDWR measurements from 1959 to 1969.7 mm/yr of shortening for the 80km wide network. All four baselines show . because we have not performed an exhaustive search. representing about 6. we identified and corrected 12 range busts. This conclusion can be further confirmed by looking at the observation data series directly (see Appendix 13). since we nearly reproduce the HS solution (Figure 9 of HS). the updated horizontal coordinates can reach 1-3 meter accuracy for the sites with observations of multiple baselines. 100-ft. Harris and Segall [1987] (hereafter as HS) found that there was a normal strain perpendicular to the SAF. to have 10-ft. These errors are known as "coarse range busts" and it is possible. although the error ellipses indicate that such a dilation is not significantly different from zero. It is surprising that such a convergent pattern does not show up in the same network when using the USGS data from 1980 to 1991 (Figure 13a). and the USGS Geodolite data from 1980 to 1984. Because this discrepancy exists not only in the fault-normal velocity component. the CDMG observations in the 1970's.also encountered the problem with errors in the integer-wavelength ambiguities in EDM measurements. The vertical coordinates keep their original accuracy because baseline length observables have little sensitivity to vertical changes. and 1000-ft range busts. as small as 10 feet and as large as 1100 feet. Figure 13a implies that some dilation occurs along both sides of the SAF. However. although uncommon. we are still suspicious of hidden range busts in the corrected data set. the derived site velocity field fit their model velocity field better. Assuming that there are no undiscovered range busts. Instead. Including such a normal strain. The baselines Almond-Bench. and Bonnie-Cotton are the main sources for the divergence. Kenger-Mason. but also in the fault-parallel velocity component.1 ± 1. Appendix 11: Parkfield dilemma In the San Luis network. different trends before and after 1980.

For the three common sites.Which data set should be adopted in this analysis. Is it time-dependent deformation? Other groups have also noticed this discrepancy.02 grad/yr strain normal to the SAF from the combination of GPS observations with much earlier triangulation data. From this analysis.2 event occurred along the northeast-trending Elmore Rar . the change of instruments could introduce a discontinuity in the measured baseline lengths. and neither can be rejected on the basis of its internal statistics. but 0. or neither? Both the CDMG data set and the USGS data set are strong and self-consistent. However. Time-dependent deformation is a possible interpretation. For now. we present our estimates using the USGS trilateration data. we adopt both data sets with a constant-velocity model in our analysis. Furthermore. it is premature to reach this conclusion just based on these data. the USGS data set. We also cannot attribute the discrepancy to the change of measuring instruments or survey agency since not all observations show such a dramatic change. Additional GPS data for this area are critical to resolving the dilemma.03 ± 0. we also estimated episodic site displacements related to other origins: aftershocks and steam well extractions from 1981 to 1991 in the eastern part of the Mexicali network (Figure 14). and Alamo). but not their rate of change.02 grad/yr compression from the HS trilateration data set. and apparent offsets from the eccentric tie deductions at the site Hopper of the San Luis network in 1981 and 1986.. has coseismic displacement. Larsen et al. Dixie. [1992] estimated coseismic displacements using GPS measurements made in 1986 and 1988. suggesting a rotational or translational offset. : iit:wiV hours later. we obtained estimates of displacements induced by 7 earthquakes (Figure 14). There are four possibilities to account for this offset: (i) We missed some sites which had significant coseismic displacements. (iii) Site OCTI. (ii) All EDM sites have been rotated or translated by the Superstition Hills earthquake. The 1987 Superstition Hills earthquake consisted of two rmain shocks: an Ms=6. an Ms=6. (iv) The estimate of . both. In addition. Shen [1991] estimated -0. which was used as the reference site by Larsen et al. In Figurel4a. there are differences of 5-10 cm. Appendix 12: Coseismic site displacements We estimated the coseismic site displacements simultaneously with site coordinate adjustments and site velocities (Chapter 2).6 event was triggered along the northwest-trending Superstition Hills fault.08 ± 0. and we also compare ours with their GPS estimates at three collocated sites (Kane. the CDMG data set.

Lisowski et al. they adopted satellite orbits calculated by broadcast ephemerides. all ambiguities were fixed to their nearest integers. Ms = 6..6. June 9. for the 1986 data. 1982]. and sites 17 and 3 were most affected by the Victoria earthquake and its aftershocks [Lisowski and Prescott. It is unlikely that we missed any significantly affected site since the EDM measurements in the Salton subnetwork were strong. Prieto. but for the 1988 data. For the 1986 data. North Palm Springs.6 . [1991] presumed that the anomalous velocity at Alamo in their map was associated with the Westmoreland earthquake. Ms = . April 26. the trilateration data recorded the superimposed effects of the two strike-slip earthquakes (Figure 14c). For the 1988 data. and the offset between them will remain.0. Sites Puerta. We estimate the coseismic displacement at Alamo (Figure 14b) and our results (Figure 4) confirm their statement. and the survey after the Superstition Hills earthquake covered a large area. We do not see any other significant discontinuities in the length series (see Appendix 12). they adjusted the orbits of satellites using tracking data from three CIGNET fiducial sites in North America. Joshua Tree. 1986. 1981) was nucleated by a left-lateral slip along a northeast-trending fault [Hutton and Johnson.Larsen et al. There were several differences between their analysis of the 1986 and 1988 GPS data sets. The Westmoreland earthquake (ML=5. July 8. Finally. has some systematic errors. in order to discern the origin of the offset. We suspect that systematic errors from Larsen et al. which had an accuracy of about 1 ppm. October 14. March 15.2. This earthquake induced baseline discontinuities related to site Alamo [Savage et al.'s estimate are the main source for the offset. 1979. 1981]. 24. 1992. All of the remaining four earthquakes (Homestead Valley. 1980) occurred along the Cerro Prieto fault southeast of the Mexicali subnetwork. so did the site OCTI. but for the 1988 data. the coordinates of one site in California (Mojave) and two sites in eastern North America (Westford. all ambiguities were estimated. Because no survey was conducted in this area between these two earthquakes. April 22. Therefore both the estimates will shift. Florida) were fixed. Ms = 5. Possibilities (ii) and (iii) do not help since if all of the EDM sites had common coseismic displacements. It may worthwhile to reprocess these GPS data with a uniform scheme (see Chapter 2). each one of which could cause systematic errors. Bnp 10065. Massachusetts. and 10 were most affected by the Mexicali earthquake. 1979) ruptured in the eastern vicinity of the Mexicali network. The Mexicali earthquake (ML= 6 . 8. and Richmond. 1986]. 9.7. The Victoria earthquake (ML=6. several California sites were fixed to adjust coordinates at other sites. For the 1986 data.

In the case of poor data coverage. Therefore our error ellipses of estimated coseimic site displacements are larger. The 10-year interval seems too long to identify the geophysical events in a tectonically active region.1. Except Edom. Pax_ncer. aftershock effects. Ms = 7. The estimated coseismic site displacements show that if the data cover a large area both before and after the earthquake. 2 of Murray et al. and Warren). Landers and Big Bear.6) occurred in the Joshua network. 1992. All GPS-derived site discontinuities are relative to the site Resort. only at site Sandhill do we obtain a significant estimate of coseismic displacements for the Homestead Valley earthquake (Figure 14d). possible survey blunders. or from the outliers of the trilateration survey made after the Landers earthquake. and steam extraction in this geothermal area. This is not surprising because we use most of their data and impose the same external constraints. We leave the interpretation of these displacements for future study.. June 28. Hence the estimated coseismic displacements may have a common offset from the coseismic displacement at the site Resort. After 1981. there is one exception: we triple the formal errors of the GPS results. there was only one survey in 1991 in the Mexicali network. our approach obtains satisfactory results (see the coseismic site displacements induced by Superstition Hills earthquake (Figure 14a) and Landers earthquake (Figure 14g). Beacon.6. these sites have only one or two baseline observation. or both. Only four sites (Edom. It is not yet clear whether this inconsistency comes from the two-epoch GPS surveys. The coseismic site displacements from the North Palm Springs earthquake are the most poorly determined. Significant discontinuities in the baseline lengths are seen in the eastern part of the Mexicali subnetwork between the 1981 and 1991 surveys (Figure 15). We find that if we impose the GPS constraints with their formal errors. and Stubbe) are significantly affected by this earthquake. 1993). We use the USGS GPS survey results from before and after the Landers earthquake to set up an external control on the movement of the entire network. the velocity field of the Joshua network is distorted significantly. [1993] (see our Figure 14g and Fig. Because the northern part of the Joshua network was not established before the Homestead Valley earthquake.5. The estimated coseismic displacements are significant at the 95% confidence level at only four sites (Edom 2. However. Dome. Ms = 6. . We nearly reproduce the result of Murray et al. The large error ellipses (Figure 14e) reflect the weakness. Thus the estimated site displacements are a mixture of secular motin. The Landers earthquake displored the entire Joshua subnetwork. The estimated coseismic site displacements from the Joshua Tree earthquake are poorly constrained (Figure 14f). Inspncer.

Dixie to Fish. 1986]. and Dixie to Off_229). we double the formal errors of the measurements in 1983 and 1984 related to site Dixie (Carri_sa to Dixie. However. we present the plots of all baseline length data used in this analysis (Figure 16).. The solid lines in the plots represent the least squares fits for all of the data from each line independently. Outliers have been identified and removed from the measurement series based on a three-sigma (standard deviation off the linear trend) criterion [Savage et al.our estimated site displacements do not fully reflect the coseismic displacements. It seems inconsistent with previous observations. we double the formal error of the Keys-Sandhill baseline observation in 1988. But we cannot attribute the jumps to an episodic site displacement at Dixie. . we leave these data and estimate episodic site displacements. For several series of measurements. if the anomalies are associated with a known earthquake. 1993]. together with other unknowns to absorb the coseismic positional offsets. we downweight the data: (1) In the Carrizo subnetwork. [1987] reported a systematic error in the measurements of the Monitor subnetwork from 1984 due to the change of survey group. There were apparent jumps in these baseline data series at 1983. When there is coseismic displacement. In this case. Appendix 13: List of all baseline length observations In this Appendix. so we did not treat it in a special way. (3) In the Joshua subnetwork. the fit statistics do not reflect the real quality of the baseline observations. Dixie to Off_225. The dashed lines are the postfit from our analysis. We do not find a significant difference between these series and others. The statistics of the fit are printed at the top of each plot. we double the formal errors of all 1977 observations.. Dixie to Sup. Savage et al. (2) In the Salton subnetwork. the better way is probably to combine our approach and with an a priori rupture model to remove the coseismic offsets induced by small earthquakes [Savage et al.

. the Santa Maria Fold and Trust Belt (SMFTB). the western Transverse Ranges (WTR). the Santa Barbara Channel (SBC). The sites of the Permanent GPS Geodetic Array (PGGA) are marked with circles. - -\ .F:' 34"N" l ------. The names of the main VG sites are labeled with four-character codes.of. Map of southern California showing the VLBI/GPS (VG) stations (triangles) and USGS trilateration networks (named with italics and linked with solid lines).o. the Elsinore (ELS).s sothr 340N -B Ci i R S t. Osot E E..37"N -1 Fg36rN \ VND Map I. the southern Coast Ranges (SCR). San Jacinto (SJC). SHU-ANZA A LKB Twit# 33 N 50 km "" "u" 101. Major faults include the San Andreas (SAF).. and the southern Borderlands (SBL).... .. the Ventura Basin (VB). 1. The main tectonic domains are labeled with italics: the eastern California shear zone (ECSZ). SALTO-MEXIC MON 32°N 122*W 121"W 120W 119W 118W 117*W 116\ i1- Figure. and the Garlock (GAR).

San Gabriel . The subnetworks and their minimum velocity directions to constrain the outer coordinate solutions are listed in Table 5. I 120*W I 119"W I 118*W I I 117 0W 116 0W Outer coordinate solutions of the six USGS trilateration subnetworks (from north to south: San Luis . The ellipses represent one-sigma errors (i.. Los Padres .San Luis. . " TWIN 33° 32"N 50 km BLUF SOLJ 10 mm/yr I 121*W Figure 2.. 39% confidence in two-dimension).. Anza .Tehachapi. I 115"W ..Mexicali). Salton .Tehachapi. The six outer coordinate solutions are independent and are separated by heavy dashed lines. ...36N ..Parkfield. Carrizo . . All tectonic features shown in this figure are the same as in Figure 1.e.Joshua..

70
20 -

n=107 a= 4.84mm b= 0.25 ppm

15 -

10 -T

0

5

10

formal(mm)
Figure 3a. 'Scatters of the trilateration data of the USGS San Luis - Parkfield trilateration
network. The formal errors are calculated from the model 6 2 = a2 + b2 L2 with a = 3 mm,
b = 0.2 ppm [Savage and Prescott, 1973]. The scatters are estimated by the procedures
described in Appendix 9. The estimated coefficients of a and b are printed at the top of the
figure. The straight line of unity slope represents agreement with assumed model.

20

n= 36 a= 3.07 mm b= 0.22 ppm

15 -

4 10

5 -

0-_
0

I-5

t
10

formal(mm)
Figure 3b. Scatters of the trilateration data of the USGS Carrizo - San Luis trilateration
network. The formal errors and best-fit coefficients are as described in Figure 3a.

n= 47 a= 4.23 mm b= 0.20 ppm

2U-

15 -

5-

0

0

5

10

formal(mm)
Figure 3c. Scatters of the trilateration data of the USGS Los Padres - Tehachapi trilateration network. The formal errors and best-fit coefficients are as described in Figure 3a.

70mm b= 0. Scatters of the trilateration data of the USGS San Gabriel . The formal errors and best-fit coefficients are as described in Figure 3a.73 20 - n= 78 a= 5.Tehachapi trilateration network. .17ppm 15 - 10 - 1 5 0~ 0 5 10 formal(mm) Figure 3d.

64mm b= 0.22 ppm 15 I 5 it- 10 0 00 5 10 formal(mm) Figure 3e. . The formal errors and best-fit coefficients are as described in Figure 3a.n = 131 a= 3.Joshua trilateration network. Scatters of the trilateration data of the USGS Anza .

Scatters of the trilateration data of the USGS Salton .35mm b= 0.20 - n=122 a= 3. The formal errors and best-fit coefficients are as described in Figure 3a.Mexicali trilateration network.22ppm 15 - I 5- 0 0 _ 5 10 formal(mm) Figure 3f. .

W1190 26'). Constrained velocity solutions of the northern and southern sections. the velocities are relative to site Asbestos (N33 0 38'. and are separated by a heavy dashed line in this plot. The error ellipses represent 39% confidence level. In the southern section. the velocities are relative to the site Pattiway (N370 58'. W1160 28').360N j NFIBR 35"N - MOJA l 34-N --PVER "I TWIN " I S 33"N 32"N - 10 mm/yr 1210W 120'W 119"W 118 0W 1170W 116°W Figure 4. In the northern section. The two solutions are independent. All imposed constraints are listed in Table 6. 115°W .

Same as Figure 4 but with the multi-dislocation model of Feigl et al. [1993] subtracted from the velocities for all sites. 115"W .36°N - 350N - 34 0N - 33 0N - 32"N 121'W 120"W 119W 118"W 117W 116"W Figure 5.

115*W .. . 340N 32"N - ------.. Velocity field of the combined solution with the velocity constraints on common sites listed in Table 7..36"N common removed 35"N - C VND ... In this plot.. Vn = 20..2 mm/yr. .. The error ellipses represent 39% confidence level... The estimated velocity field is relative to the North America plate. CE 10 mm/yr 121"W I I I 120 0W 119W 118*W I 117" I 116W Figure 6. This commonly subtracted velocity is plotted with a gray arrow at the northeastern corner of the figure..8 mm/yr) is subtracted from the velocities at all sites. a common velocity (Ve = -18.

. Velocity fields of the combined solution in the southern section with the velocity constraint between Niguel and NIGU_GPS removed.EGU EE ICH 34"N GL 10 mmyrDA 118W M Cv I 117*W 116W 9 IXI2 115W Figure 7a. W1160 28'). The velocities are relative to site Asbestos (N33 0 38'.

W1160 28'). 33"N 2 TORO p 947 AVID 10 mm/yr 118"W 0 117"W 116"W 115W Figure 7b. . Velocity fields of the combined solution in the southern section with the velocity constraint between Monures and MONP7274 removed. The velocities are relative to site Asbestos (N330 38'.35N - 34 0N IGU'.

. The velocities are relative to site Asbestos (N330 38'. W1160 28').350N MEE ICH 34"N - 33"N OA10 (N33 mm/yrLSI Asbestos B (I AVID 10 mmlyr 118"W 117- 1160W 115W Figure 7c. Velocity fields of the combined solution in the southern section with the velocity constraint between Asbestos and PINYGPS removed.

115W .3460N 33"N 32"N - 50 IB- o k 20 mmlyr 121W 120W 119W 118W 117W 116'W Figure 8a. The error ellipses represent 39% confidence level. Velocity field relative to the North American plate of the combined solution with the velocity constraints listed in Table 7 (final status: used).

Vn = 20. 1150W . Same as Figure 8a but with a common velocity (Ve = -18. 34*N - CEN ou 33"N-o"o 32"N - 10 mm/yr I 121W 120W II I I 119*W 118*W 117W 116W Figure 8b. This commonly subtracted velocity is plotted with gray arrow at the northeastern comer of the figure.8 mm/yr) subtracted from the velocities at all sites.2 mm/yr.36N Scommonremoved 35 0N V34ND .

The error ellipses represent 39% confidence. Velocity residuals with respect to the Pacific Plate of the combined solution with the multi-dislocation model of Feiglet al [1993] removed.360N - LH 35"N IBR ) LOS? VI 33"N 320N - 50 Ian L 10 mm/yr 121 0W 120 0W 119W 118"W 117W 116W Figure 9a. 115"W .

I 115"W . 34"N 33'N 32°N - 50 . Vn = -7. The velocity residuals are shown relative to site MPNS_GPS near the SAF.36"N commonremoved 35N L.". Same as Figure 9a but with a common velocity residual relative to Pacific plate (Ve = 0.1 mm/yr) subtracted from the velocities at all sites. This commonly subtracted velocity residual is plotted with a gray arrow at the northeastern corner of the figure.km U 10 mm/yr I 121W 120OW 119W 118"W I 117W I 116 "W Figure 9b. :-.5 mm/yr... .

. The error ellipses represent 39% confidence level. 33"N 10 mm/yr I 120*W 119W I 118*W Figure 10a.35N PA IH . relative to site MPNS. . O LOVE RS .. Enlargement of the combined velocity field for the Big Bend area.

. Same as Figure 10a but with the dislocation model of Feigl et al. [1993] removed.35"N - 34"N 33"N 120"W 119*W 118W Figure 10b.

Residual velocities relative to site Red Hill in the San Luis network. .36"N 35"N 121*W 120"W Figure 11.

225 JACU R1 CILA DIAB /i 3 x 10-7 yr 118°W 98 117W 116'W D ' / MAYO 115W Figure 12a. . The inward pointing arrows represent compression. In each Delaunay triangle. or if the sum of the magnitudes of the two principal strain rates is larger than 2x10 -6.CREO S 34"N L RANG• L C BUTT i RO SNIGU SSALV la 33"N - % I . outward pointing arrows represent extension. Principal axes of the horizontal strain rate tensors in each Delaunay triangle of the southern section. calculated from the constrained velocity solution of Figure 4. if both principal strain rates are smaller in magnitude than their uncertainties. the axes are not plotted.35N MAUM 1.

the rotation rate is not plotted.4°/Ma) 118"W 225 AP x - x~ // '_ 24_ 7 M / AY O 117*W 1160W 115"W Figure 12b.35N SEGU 34"N / RANG \ 33"N " ~~~ M LAL E BE \ b k MMAYOfans NIu of Figure 4. The fans display clockwise or anticlockwise rotations from north. In each Delaunay triangle. calculated from the constrained velocity solution of Figure 4. The from the constrained velocity solution ~ E ~ _-- BUT Ay clockwise or- MONU . if the uncertainty on the rotation rate exceeds 10'/Ma. JAl 2 CILA DIAB 2xl0"7/yr (11. . Rotation rates in each Delaunay triangle of the southern section.

Outer coordinate solution for the San Luis subnetwork using the USGS EDM data from 1980 to 1991. The error ellipses represent 39% confidence level.50 10mmyr 350N 1 121*W 120W Figure 13a. . MINE TABL MID PITT ENG MONT BUCK BONN PF1 S FLAT \ MOOS . The sites are the same as used by Harrisand Segall [1987]. .WILD c ENC HEAR " OPP TWIS VILL PROS RR ESS MCDO SIMM IRW .36"N CHOL .

EN EA ED "LMO TWIS VILL OS AAR ( \CHIC T S BL MCDO SIMM IRW AVI 50 10 mmyr 350N 121W 120"' Figure 13b. Same as Figure 13a but using the CDMG data from the 1970's and the USGS from only 1980 to 1984.360N N CHOL MONT BUCK MINE MID TABL PITT ENG PF1 'ONN ATSO. . r TOWE LD .

Estimated coseismic site displacements from the combined solution (Figure 8) for the 11/24/87 Superstition Hills earthquake sequence (Ms = 6. . and Alamo) represent the 1986-1988 displacements estimated by Larsen et al. 6. The gray arrows at three sites (Kane.2. The error ellipses represent 95% confidence.6). Dixie. [1992] but with the secular motion from our velocity solution removed.33N - 3VFISH U XIE OFF 229 OFF_225 CENTINEL 10 cm 116*W Figure 14a. The scaling arrows are shown in the lower left corner of the plot.

~autionFigur ?' for the 04/26/81 Westmoreland earthquake (ML = 5.7).225 CENTINEL 10 mm 116 0W Figure 14b. Estimated coseismic site displacements from the combinea .330N I I DIXIE OFF 229 OFF. . The scaling arrows are shown in the lower left corner of the plot. The error ellipses represent 95% confidence.

Estimated coseismic site displacements from the combined solution (Figure 8) for the 10/14/79 Mexicali earthquake (ML = 6. The error ellipses represent 95% confidence.6) and 06/09/80 Victoria earthquake (ML = 6.2). The scaling arrows are shown in the lower left corner of the plot.32" 30'N 8 36 (3PUERTA DAVID 10 cm 1150 30W 115 00'W Figure 14c. .

340 30'N CREOLE MAUMEE SSEGUNDO M EK RICH ANDHILL EMESQ VALM . Estimated coseismic site displacements from the combined solution (Figure 8) for the 03/15/79 Homestead Valley earthquake (Ms = 5. The error ellipses represent 95% confidence.6). AXCER KEYS 4 . The scaling arrows are shown in the lower left corner of the plot. .P 10 mm 34* 00'N 116" 30'W 116" 00W Figure 14d.U29.

340 OO'N INSP BERDOO LAQUI 10 mm 116" 30W 116 OO'W Figure 14e. . Estimated coseismic site displacements from the combined solution (Figure 8) for the 07/08/86 North Palm Springs earthquake (Ms = 6. The scaling arrows are shown in the lower left corner of the plot. The error ellipses represent 95% confidence.0).

Estimated coseismic site displacements from the combined solution (Figure 8) for the 04/22/92 Joshua Tree earthquake (Ms = 6.1). . The scaling arrows are shown in the lower left comer of the plot. The star symbol represents the epicenter. The error ellipses represent 95% confidence.34 00'N SP tEAC 2 cm 116" 30W 116"00 Figure 14f.

6) earthquakes..5) and Big Bear (Ms= 6. The error ellipses represent the 95% confidence The scaling arrows are shown in the lower left corner of the plot. [1993]. The surface trace of the rupture (heavy dash line) is from Bock et al. Estimated coseismic site displacements from the combined solution (Figure 8) for the 06/28/92 Landers (Ms = 7.------ r=--~------=----- 340 30'N 1 1. REOLE UMEE EGUNDO [EEK ANDHILL 14H ALM S A CER 34" 00'N NS iOM 1 -ERDOO 30 cm 1160 30W 116 00"W Figure 14g..04/22/92.III I~ I I ---------. .

. Estimated site displacements in the Mexicali network from 1981 to 1991. The error ellipses represent 95%.confidence.100 32" 30'N 8 P1 RIETU 65 10 PUERTA 24 DAVID 10 cm 115" 30'W 115 .oW Figure 15.

considered independently. the dashed lines represent the estimated values from our combined (VG+EDM) analysis. . Time series of the EDM observations. shown by discontinuities in the dashed lines. The error bars represent one-sigma formal uncertainties. shown 6 per page. Any discontinuities in the dashed lines represent our attempt to account for breaks due to destruction of sites or changes in instrumentation. are for those lines with coseismic displacements. shown 15 per page. The first group of plots. The solid lines represent the best weighted-least-squares fit to each time series. Note the change of scale. are for those lines experiencing no co-seismic displacements.101 Figures 16. The second group of sites.

ov*Vm 003tE0 *Os W0ot-um C9t -(WWj9 toI (AMww)A- *05s *Os- soip po P t w I0wtam-(ww~eis~-( nuwA 0s % 'in- 0 05 *Sw 00 000O woUZti ZO *(wu ot*C .(AWiUJ -(uWW~ 1E-.I 0 00 -Siiti 000 ('Wiuo 074 -(AtUt)A *osa~mlogip0*et -(wXiLta.(wuj)x q~ Oft tgrt-(htozstO goo0Wmait *09 -(4nw)A -A l a-(ww)a ggp9.(A~i Msa4Ww - Uwp *Os gEE61LZ I Ij.4ANIUw AtUV -(Asjowsc~s .0as" Z -(Wtul jtt .3661 S261 0861 SL61 0L6T 066T S961 0861 SW6 0L61 0661 £861 0861 SL61 0LW 09O- -Os- 0 0 - F9.(4nw)Ak WIl . -(WOgVust(A viat466t :(muw) Aru-saips *Os- l 05 1t -$a t -(lUlt 6tt It Itt Wo0gLgg -(whg M~-Mgp(AX ewottAa- I Os gap$ la-mii *Os- :.(~t -Lop ~ siatwAs -I' I 0 0 05- Os- Vp -(WW~ qto.6MB( _ 7 9UitI rszm~ -(ww~.a7twi 09O twi 0 0 Os- Os- Fj 05 Os wo asmG cB -(tw~ u 0.st -(AMuiu)A 9ggt-*M~t -(wXgW~w-(A~j ogesjpw -...L 6L0 -IUL Z (iu " 6Lt fi J-A~ ojw - uos~~. -(MUW)A 06 0 sw ESt -(WwlZO I m0s m0 t -(whiK~ -(AX -(4OWU4A -Sops - pip *Os- 0 Os Ito -sutm we~ (ww~got' *Os ±41 w)A- Os- 0 OLz -A-W~Lg-t .

-oum i i StI 1um gm* 69*9 -(MUWw)AJPslJO. .05- 0o L.t -(WW)B ill 901I SULIJ iLGtM9 -WNgKO6-(4- 11111LJ11-. 09id -- jgon 0 0O -O -0 oo -sumj NZ -(ww~s LVII.(w~ .:MUW)A t*Z-9ILLI -(w)h OLWMI(AX DIPeq -.Somiq I III liii 1111111 1111 05O [5- lii W 0 -Sum ULE -(Ww~s6L 0 -(AAUW)A 9tSS0L691 -(w)i aigaW61-(A)I ?AWps- i Liiii 09O sauaq iiiiiii -- Z90-SUUJ99C -(WwU)StEO -(4AUW)A_ A . ..ilI__I I AAUW)A Amwfil. win On 11 0o I -(WW)B toz -(Muw)A- IIII t-A II 09O I 6i- Et * s- I Au) IGA . 0861 5L61 0L61 0661 S861 0861 SL61 OL61 0661 £861 0861 SL61 0L61 .09- I. 1 .-(A4uw)A -(w)I gtc099i-A l oiacm .1 ..L at 0 -.0661 . £861 I .NJB IP9Ma I -(w) ML6L6M-A ZAePOIS ow C0T .7"mO WO -SWJ CI E (WW)S Dt? -(w16SISeI-A tt9099"Z& 0s ICt -(wg OL V.-jpoin tII *Os- -(AdA- tea -swuL 90£ w tg t6091.1.--I V9L6t9IZZ -(w)i £9LWZ6-(A1 -IIDU ooa -sumj 000 (ww~s mo-A0 wA eiei X GiOLAOs ZZ'KM -(W)l EGZ6 swomijq - -O 05O -sum.000 -(Lws 199 -QA~wwA .61Z -(wI 61gL61-A -- -s -O -O 0O 9L6L90£ I I I I I I -(i W III 09O sepo -si"=o-& 096iA I Ii I I I Ii I i 96 0 -(ws SIS0 -um ES 9i9a-aqu -(W)i gamWi(A -QAAW)A amp - 09O gOppq ii1..

(AuW)A OZ'1-SwMIk09 -(ws ge v..-~Amow wO1 .-CLotS -09 0 L110 -swAJ c£c £E9zMSO __09 ZSO-WAJ Co -(ww~s ts -(wiiis t~e -(Auw)A -(w)l jcczqt-(AX -tidpm .-pused *09- 105 0O 0o 09O 090 -SWAJ U E1 -(Ww~ M~ I C69969% -(W)I M-Mt96(Aht ThPo--Atu 09- 09Os E~S LM91 Z -(ww~ 610 -QdW)A_ SIM6L90 I -(W)t 01fZ961I (A 80-MPqW - -0 9L0 -SWIJ E-(WW~geq0 artgZK*1 -(w)j W96 -Ah tPAI3 05 -(4JJW)A -Aot~m .(s ) -0Os I 09- I -0 0 O 0 go tSw-- 919 -(ww~s .A) I --l I I iiI (QW)A ojumeA I ! I - I tt' -SWI 6* 9 -(ww)B Eva -(Auw)A a S-sAJ 9ft SL61(Ah 9012 LOLtI -ii .(w t -(A)t - - 09 -(AIww)A asodem c00I6 Os- -O 99 0 -WAS 09 C -(WW)S09 -(A~uWw)A_ 0199Z9901 -(w)t IL90961(Aht -p~Io5.(ww~ot.-(AAU a% (Ahtuo-oqw 1Z*LL 019£t -(w)i LIisB 09O -pno6 L9O0 SWJJ16 Z -(Ww~s9 t £99096L I (wht gogag -(Ak ZSO -SUAI 61 -(WW)SG* t _-41JW)A S99Ltt99 . (AUWw)A 619 OOP -(w)I LZVG 1-(Ah -seABJ--ter 09O ZV0SWAI 9S*1 -(WW) 089VE996 o.J6014311 -(A$uw)A 09-05 0o -0 09O- W. 910 0ta -(ww) go -(whI 50098 L.(W)l 6*9-Zg6l(Ah I~ -PSIAmd .Aei ze I swam £9 .(MUW)A -Ae~jr4.0661 5861 0861 5L61 0L6T 0661 5861 0861 SL61 0L61 0661 S861 0861 SL61 0LW -- I -0 L Os g6o Suma &.o-o..A Z90 -sw~i U13 (WW)S 9L-Z.

A) *I)A 6£9 msi. 0961 SL61 OL61 0L610861 £861 £61 0861 £86 561 0661 SL61 0L61 06 0661 8106 0861 S861 L106 OL61 SL61 l.4w).4A). WcOMI.sk SI ~ - L~& ~ - *0£ IO - 0£ 0£ a s-z (ww~g61*0..4w).4wweLeLI (MUW)A c121 OL696IZ .A l4~udJu-Jma -0£ IIIn--K =191111111 - tLI(AdWw)A ZOOSULUZS -(Ww)s99 t06 9601 .4ww~gga w6r6L6£ .4Wwws6Zc SULJ SE MI09601 *(whjL9*G6W4A).(A4uw)A ISO-Guu iZz ms6ge9 -0£ I I I I -(wh .souirm t II I I I I I 1I1 1 1 1 1 1 1 1 1 I 1- -0 -0£ oro -ww tot . gP96L61.a~s~si.AmW)A -4w).4uwwo91-0 .44UW) 611Z£960E .-. q9Z19614A o~kIe .4w). -1ww~ Iptt.4ww~lS*pg .4A COd-Smiw *0£ !Juw)okA u woIa-ow 1111111111111 111111111 09- 0£- - *0- I S09 III II I clI IWJJuLL-L .ggt'O~j(Ak 8mpipxx--sL 0(u)A .seuirhw S0£ M666t 066sos-4ww~sto 09.4w)..4 09 0 suuu 11j.4A)o-wuA--ooes o L-0 S0l .4w).qlqtgsi. gtolvow.4w).- I - T T Ii ~i -09 -sum 6£9 .4MUW) W96. wm - *0£IltI jtillf111111 it weoswm -(wwrztgoA 0iw) -og -0 -0 *0£ to o -suij toz -4ww~s690o-. gjjgOUI.- W) - -0£ ~~g-o~gg1 . 66r19t-*L£Z 0 tt (ww -(w).4A) ol uwA-soud-jw et1 -.4&tw)A CI6S'L060C .561 S961 6.

-(A~ww)A -(w)I 6uL961(A)I L9t -(ww~s690 -(AMuw)A QLO" SwAu t9~ -0£ .oFsid -0 Li I I 1 1 1 ' l I I - Z9 -S.(ww~s06 t.to49 96699£90 ZO0 -Swju e-jiim-aammz I 1 -1 I 1 1 1 1 1 1 1 1 1 1 I I-L - -0£ .-(A uw)A 0L61 0661 £861 Jugs 0861 SL6T 0L61 0661 £861 jugsi 0861 £L61 0L61 0 0 0 09- 0£- -s 09 09 0£- 0 91*t -SW1 9C9 -(ww)B e'S -VsnwM) 99L I.eo0z -(Lw)I 6oCM jgg(AX owswiq .-Aw) £1lu9av01 -(wk~m09-A4 !%d-wpm £956*L0£E -W) 6L1 1961(A~ rudozm.(W) 9g10961(A of-Wpd-ojuiud -QAAW)A -(W)I SLE IS6 -(A oF-zzel .:.oqpiou -0 -09- I T 1: T I I II I I I I I T1'T -0£ Z9£E £L0 -SULAZ (ww~s91&0--(4MW)A 991 & -09 - WI I1 09 96 9 .(wh g9tm v-(4 IIu) ~aw - 0 0 SEI 'SwiL 699 ..404PJOU -(tw)I 969 096-(A 69 1 -SWI 6(*9 0919LISI -(ws 060.(wws go00._"IIw)A -(w)I 61g096t-(A)I _UeNj-opOU 901 .661 Lt £861 -Suuu ttL jeoi( 0861 -(ww) SL61 162.(ww~s6£ v -(MUW)A £tZZEt09 .1. 11961 -(W)I vit*gs8-(A or-wwA-oig*pd £9 (Ww~s 6t 0 'sumA t Wt9V W 1. -(4lw)A V1011-M91 -(W)I 196*Ml-5 1( 0£ - ~t-4ow izowpO ow 'e paz 0 09- 0£ gI 05 -0 r09 -09 gIl.(ws Eva.ow *nq 911 sum1 069 -(wws Ito0 .Qtww)A woq .Z 96 0 Swju 10£e .(ws LE I.(AuW)A 6=5L019a *(wh o(1mj1-(AA owuCs -ow OW~ t95go SULJUOL 6 I(w~ tq c11 - .

(AX owisuaui.d -(W)g099~EN t(Aka WE 99 I 96'0 -Os GWJ 6E*9 *(WW)B%'Z .-(A4WW)A- 09O O~-W -UY.(W)gggNggeL-(AXowIsulu .ouw~o~ -SWJJSig -(ww~olm Itit f(A'U-M WO 0 Ii 05 141VICKP II t j 0OOT O 09 1. OS - OS- - -Ub..otu a= 09O I -(WW)BES11--(4MW)A %*I -um 9L L0uns9z -(wkg m"M9I(4 ow-tfpsw.-(A4UwM SL I -SWJJEl9 -(LwwU~ -(uJ)tZLQ996(Ah ufPOLU SIK1 9IPM lIP4 -Os _ J9~ 0 -o - -O 0O Z60 -swAJ60og -(Wwu~s 99.p m -(WW~ S .-(4uWW)A -(W)l gjajg(A -Dosi-PL. 05 II -O -O 09 9t0o swx 6g1I.(AUW)A &-(A~jnpvs .-(AuWw)A 9ESLLt It I -(w)i SgLL95 1-(Ah owud .9 -(wW~l g601 -(AjWW)A -oUWeuau 1 99L6*9961 -(w) i99&(A~ OS 906..)661 5861 0861 SL61 OL61 0661 S961 0961 SL6 OW6 066T 9861 0861 5L61 OL61 ..owAB 91 's"i 9L6 -41110 K-1 WIWM -(WN W~t~IMM :1* 091S -(~wlw) f-w Ow.(Wua)...(Aw 99*0 "MMLL'C I990m IJ-I i if 11 09O ll -O I 06 1 -3UWItf9 -(WW~e69 1.006 Pw Os- Os- zS I 96C6 C16L -(wU)tuLtIPLe (A ii EcI -SWA 96./A..es £S90-SWAI 99et (ww~s9 19..6 -sWAUIP 6 -(ws 90 6 (Mnw)A 61tWEE19 ..-(uW)A__ 0bi oB1 -(waikmcls6t. -Swmal&t gm-06010 *(ww~eSZ 0!LM zgsgg-( )t -(wkg W3-jtx. 0O wi.4wU)A L A .Irmmud -(wht gmemU as6o gm90 LOT .....ow AW6 09O ~gm ggwg 111ltI 05- -Os- 0 05 001 OS 913 S9WAJ L99 . ge' -(AouW)A iLuLZ £05 (UIht9IL96. (ww~s6q0.

f 09 wi l 05- -0 Z9 I -SffJU99 6 =(Ww)sZL 0 -(/UWw)A -ifUU--mod-eq 99OLELtiZ =(Uw)iZLt996t(A)I *Os 0 09 09Q6 -(ww) S '.-IuMdsP iodsip Onfp -WIf ga'-( OE - WAug IM -f~ 06' tw *OS M 0 01 OS- 09- 09 fstt -" st9 -(WuiUa 19960S91 -(0) L69r9614 If I I I 1 f I muM III.-mCdvq -0 09- - It~ZSWAJ 0Lj9££L££ 6LLI -(w)s *L f -(VW)A -(wj)f tgWS61(Ahl jumifp--modf-pq ZEf -SwJ I - OS 9t 6 -(Ww~s IL~t-(Mw)A Ii I t cg mo Is Ia#A I t9 I -ui £orZZ96C 9 0S 1-(ww)satt-Aw) -(w) egl. .561 9861 0861 SL61 0L61 0661 5861 0861 SL61 0L61 0661 5861 JugsX 0861 SL6T 0L61 09- MY~ "I-- 0 0S It I -SWJU tZL -(ujm).(4UWw 991 -SUMA -g(w)f t11 696 (AA jeou~md -9DI99g 9ff -SUMa g QN6I9Lt9V Dpo -(WW)S 6Lt-*) -(w)f LL6906 MA Poerwt .*.-(4luW)A-(AX O-WULUM mj&~g Lq wI-(N w L 801 .-(AAJW)A LL*S9Z£OI =(w)l MEf961(A1 Md-~ 09O g I -suu 09 9 -(w L~s6 -(AAMW)A aW Laq t -(w)l 990 9% 1-AX oOup DfMOU 1MIA I 09 gI us 66 9 -(uws t t ."Uwm g99 -(WWiS099 caIm 1f0S-m*f -(w go M~-AX 09 LlI (4MUW)A4 *ir. 22 -(AjW4A 66 CUM0 -(Wh~ £99Z981(Af twfd- 06'0.- Iof4f gumo 66L(w)s La 1*9= 0 tL 0 =SUJU 9ZC '(wwl Q I£999 LZ9t -(WhfZLV*90t(Ahf LuJ W3 -sum LAL6 -(tWws LI Z -(AuwW)A4 6166690t -(whg6ELZg6I-(Ah esPsa.9(A-i fLvodip- Xv Luw I k-OS- ur I I -0 LOI -SWAJ 9V -(WUJ) ZL V.

-(AAUW) MU*0o -(Wkg1L6qwt6I(A~ j&UUm -767dwuo t-gs(A iwfbod .(AX empon .(W)l ON Z96I-At cOOV-W- 7nP 601 .-(4U)A -(Whg99M95t(AX J&-WSM - JSOUd 0£- - 03 Wit -uu .-(A4uw)A -mttd .gL(wwL-(4uw)u £6600 I.-(Aiww)A 92~9 96LtZ .96Z -(w) atL ZUB I(AAteuPSG .(whg 096MlI(A l&lliJ~d-76dWumo -0 -0 S~ WIj -SWum it 69I -I~ -(WW~l W-* "( W)A M981t =t Luck -0£ 6c 1 -uS .tom6rail 901 -su Ge p(ww~o 9*1v ( W)A_ V1060909 -(w)i ZL9Z9I4( Um~itM .(w)s 1 ZVsuI(4 paFJUWd -0 D5£ *0£ og t SUAU nL *(wsw gI .J~med *t 09101T -0 I-_ IE 66 1-U JI1 L (Ww~s Z . -u I99L (WWu~SL10.(W)l gg-gl-(Agt 4 MI.0.(Wh Ztg'ZJI(Aha jdmwZm-v3SLpd 6L09I-JwkiZ £60 -swu99 -(ww) 6C1.(AjW)A -0£ 1919 69W= ."(4uw~o 6tiLoist9I .Iud 0 -0£ 0 11ItWA MR 966069M6 -CL(Ww~s Me.100 rati -0£ £009 9VL6 .vuuopd -- 0F -0 0£ du £10I 9tL0Z "(wli ULS96 I(Ah sstaied - 6duwo AU) 3WA1JCIS9 -(WW)l S1.* 9COVS~ -(w)i jssq~e -(A4i es~s.0661 £861 0861 SL61 0L61 - 0661 £861 0861 SL61 09- - 0L61 - 0661 £861 jugsX 0861 SL61 0L61 os- 09- -0 -0 acI WaIu 999 Uiww~ 9C. - EVI 19 I-u ggUU999 -(WW)SI Z -(AAUW)A MW0 190L11-(W)t 096 9% 1I(Aht isouod-&~ -09 9C60tM6 .(WNgIuWusI.

u (mm).portrml v(mmy) 50 1985 1990 1970 1975 1980 year 1985 1990 1970 1975 1980 year 1985 199' .1982.1-986.744 1(m). 338 s(mm).00 I 1 8691.195661 0 47 mm). 133 -jplllrml v(mnmy). 123656455 int_- 50 I I 1 01 s(mm) v(mnvy) 0 7 22 ms I- cn -Jv i l l i I glnnr . 138260661 50 v(mmy).1963306 K(m).) -015 (mm). 140055517 v(mmy).49 s(mm).assis t(y).2305 611 nrwn 176 I I I -50 1970 t(y).700 (m).19687817(m). 311 dugo 50 dugo 1 50 50 .1. 133423753 v(mmy). v(my-).196 489 1(m). 0 00 I 1111111 gkincar-pa2r 50 0- II I 111111111 2 t(y). - -50 -50 -Jv II I I I II I I I I 11111 I sslse 1y).199538 (m). 512 nms. -1. 0- 0 0- dugo - . 50 I II II I 1(y). 6 05 rrms. 1. 8690.-388 (mm). 13349.1391 rms.442 I(m).25 mtgleamn-palw_ 50 0 0 -50 -50 309g6 i v(mVI . -0 06 s(mm). 69169370 v(mmrVy) -136 s(mm) 0 2 nrms. 4 78 nrms. 0 39 mm) -0 49 -flnt v(mmy 98373206 t(y)-1985.19 (mm) 1 71 nrrms.114 -50 -50 II mglea .portrml 50 I I I I I I fint 50 186 i Ii ilIII Iiii i I I flint 0 -50 -50 I 0 24 gn_ncer . 1 II TI I I I I ff 7 46 rms. 0.1985 413 1(m). 0. 4 56 nrms. 93553627 1 38 nrms.0.42 50 0- Il I Il l l Tl l -handle I l ll (y).1 (y).6429 0 95 (mm). 191 li IIIII I I gin nacr-peatfco l(y)1986 710 I(m).1427 0 00 nrms.00 0- -50 0 I t I I " " I I II I I I I I miglen-iselde 1(y). 138335454 v(mmy) -0 42 s(mm) 0 06 nnrms-0 01 50 0- cn -50 -50 -Jr II II II I gln_ncs-pak_rm3 v(mnVy).45 s(mm). 2.1956097 I(m).1986 709 I(m). 125205131 v(mny). I(m) 8705. 000 nrms.710 I(m). 50 I I I I I 0 97 nrms I 9 -pllnrml 1(y)-1 7851 1(m)51533045 4m-mm). 1400&1034 v(mmly. 000 0 (y).uss y). 0. 09 (mm). II III (m).po oo (y).48 50 0 1975 1980 year - i miga.--- ~--~ 110 -.1985.

88 1970 50 50 0- 1980 year 1985 1990 1970 . 5( 0 III 1f may . 10781.2.057 I(m). 16607 9649 v(mmf: 3.blr v(mny) ~ PT F t(y). 118 may Ii IIiii 50 - 5 im - v(mn).6 rme. 19829.241712472 Sv(mnVy)2. 0.1985038 v(mn~v 0.129830469 193 s(mm). 209519312 -40 (mm)w 403 nrms.63 I II II I I IIl I I I II II I I -warm_ spr 1(y)-1980.5292 s(mn"W .442 a(nm) 10.13 00- -50111111i1iili11111 rwy/ .9757 4 73 nms.195504 (m) 1971.29 pot ) 5C sisolie 1(y)1980. I I iiI II I II II I III II I II II III I I(y)-1982. 303187039 -3. 154702242 v(mnmy).pconer 1(y). 095 v(mny).7622 o-1 -.-034 9(mm).1960.459 I(m).556 nrms.0 54 (mm) - 1iiI -50 -j 1 -50 I IlI 0- i may 11111 111111 11 0 pe clncer- Sv(mnlW)- 5 11 1 50 -50 II - rl 0 i -L - III ll p-plansa 1(y)-1950.4 08 nrm.70 50 - 0- 0- -50 - -50 - T -5 Il 5( 11 1I II I 1111111111 I 1 arway . 180 s(mm)- 0 -51 0 III I I II may_ I II I III II .096 ''II -50 - -50 1975 IIIl1 I T 0- -50 -tank t(y)-1986. 157 50 I .bm n 1(y). 7 04 nrms.58 50 1 0 159824041 9 41 rms.6207 (mmVy)=-1 20 mm).713 I(m).2 nrms.6 70 nrm. (y)-192692 1(m).23 rims.20 mm).52 mm).do* Ify)-.1166 hauser 6 02 nrms. 0 III -p lm s v(mm") 50 1975 1980 year 1985 1990 1970 1975 1980 year 1985 1990 . 24126.1984207 I(m).v(mm.7607 rway 9.66 o(mm*) 7. 2 15 v(mnmy).14 rms. 763 nrms- paclno 1. 065 ~ i- - -jr- -lITlL h t)s .796 rms.1.22 s(mm).14775.slsioe(y).1.052 (m).169 0 -50 -50 II I1111111 may -portrml v(mmy)- 50 11 -l (y).185435(m).111 -port rml 1(y).dd2no1 1(y)-1985318 I(m).302423153 0 19 s(mm).1979 909 I(m).1983 090 (m). 0 15 s(mm). 50 - 1(m).103 50 - l aw l i I l I I - i T T .666 1(m)99424584 122 smm). 152 v(mnmy). parker 50 13047.0.005I(m).93 s(mm).19B4213 I(m).3 29 nrms. 1.0.

9 -(ww~s qvt e~ori-(Lu w~ggmge-M pwwp -/W)Ass% - sm-p 0 05- 09- -OS 96 V -(Wwws 9L0O . . 0861 .(4 OS ZZJ L -(wws 911.-(W)A ZM9L9gzl cmg . .psioP -0 It7 titt~tt an V9.(MUW)A 991 09 J I -0 * *UL-- 0 ssq.61 0661 5861 0861 SL61 0L61 0661 *I . I I 1 -"1t_ I I 1_I (Ww)i awt -um tot oomml99 -(wki mmlU ItL _ I I I I 1I I I I I ~iw)A- Mm" I I 001 -4 I I I s -09- -0 -05- I--- 09 - I 09'0 -SUIU LEE-WW$Z 09Gl -(w)Nggm-A ±4uW)A ~ 4sc.61 0L61 .01-1A~ -Pww uiu- ge61 -SWAI 19.9 Ittw~ I IW) 05 05O 05-(Ww)n ZI9C61 SIAJ g1l1 qo*2 -(wht gggg96j-A 05 961IWJ VOL -(WW)B06 0 :±4uw)A 9gmensi.SWIIjR t66g I (W)i066t t.9eIttq -(wkt901tm61(AX 611t 6W11 919 .(AA 660CLMt Ioedswd-uaueq -0 <WW~sg qLo I SWLJIS 6z 6966EILEI -(W) 619%IA W)A _ siq- 09O usimq Sil . -(L iai. 5861 I .0661 5861 0861 5L61 OL. I . OS0 -0 05 gVprt p6lua II I 1111 -(wwg~u1iA I I ~R5-s 111 IIII 05 -05 j II.09- i1 Iuuq 9-wt I oI 0~g6l=(A I -0' It -ii 1.91 .(Ww~ 91V miWW)A $IMP.w)I E~Zt0m9 OS -(4W)A -ULJS 90 1 -(WW)S 1LZ MLePILP-hamoq -(W)l66W9MI. SL.

6312 . 25 I(mm) 12. ~I -50 - " 1980 50- 0- -50 - I II I1II1 I(m) 13379. . 1638. 0.95 50 0- 0- II -50 - bench 50 - .720 K(m).25 11111 1 lrll ! benchi_.04 nrms 0.1984. I 1111111 II year 1985 1990 hopper 1970 - wit 1 I II 111I 11 I IIIII 1~y 95024 1(m)."(mm).505 533 (mm).96 50 - 0 0- -50 -- 50 .1 25 -hatchrml (y). 24984.07 rnms. 76445446 1(y)-1983. 17641.194565 (m).1987550 I(m). 331 (mm) 3 25 nrms. (m) 95328630 1. 4.mdon~m v(mnW) I 11111111 I 1 y). 840 nrms. 1 0- bench 170255563 885 nrms.198305 . 1282.81 nims. 1.22 s(mm) 3. 3.59 s(mm). 688 rims.93 rms.1144 v(mmy) 040 (mm). v(mI-cn 1.32 1 1 50 hoppo e -dhi v(mm).moose v(mmy" y.1985321 K(m).02 u(mm).6413 (mn" -2. . 100 9.1984.masn (y). 1. .72 50 - 1 50 1 K(m). 1.136 1(m) 23684.99 s(mm) 943 srms. 0.red 1(y).08 s(mm). 13.2209 2. 1844&9671 mMV" -. 1. 170163298 v(mmly).55 rms. 78846954 v(mmV-) 4. -50 IIIlll II11 bend 50 - l171 11 11 .18.1963505 1(m).2334.51 nm.8124 5.68 0- -50 bench 50 11 1 If I i -wWid Sv(mnW-- I I I II )=1984125(m).463 -0mnw .06 50- S fl111 I bench 50 - 1ffl ill I If 1 i .573 1(m). 2. 0.55 (mm). s .*'n . 2.owers (y). 0.29 0*-5 .15 f~ .3404 )a1985.807 s(mm).49 -herst v(mmy bench B2~ 0- 0- -50 - -100 - -50 - -501 111il v(miiqp . 111I11 bendch 50 - 4.22 rms.40 nrm.mason y). -164 (mm).1982533 -hatd (m). 0- I I I I I I I I I I I I I I Tz III 50 - II II chides2 .77 rrms.10 1I11I i l11I I1 l l l I drch2-red hi 1(y). 12719.1 1975 1980 year 1985 1990 1970 1970 1975 1975 I ' '' 1980 year I 1985 ' ' ' 1990 .49 nrme 2.304 722 S(mm).113 bench 50 - y).78 0- -50 - 1970 III I ~ - I I l 11111111 -50 - IIl i fi l 0- -50I l I -50 -50 - --.198346 (m).27 -50 I1111111111 I I hoppeW _ . 4.8071 "-mni). 1. v(mmlVV 661 s(mm).32 0 - 1975 .kerr W "1986.

(AJWw)A_ 09096EL01 -("L)l t9~996 -(Ahi -.-(A~i -m6uwi4-muoq emjuuoq -0S (Ww)@L6-0 .00w- ppo L£0G..itL . . I .(wus ito ..at -e t71 I . .1 0L61 0661 £861 Iig 0961 5L61 0L61 0661 S861 J13x 0861 5L61 0L61 AA Os- -09- -o -0 0 0O K o -sm Got .-AAW)A .(Ww LZ0.tiioq Os- -Os- -0Os -0 O0 9NI W)A 3 emoq -Os Zi I -SWAJ ILL -(Wws LVZ 10996=~Z -(w)t GLL91I( 09O -(Aluw4A UtVA-iPO&OAJ ago 'WAJutos .ShA Il L 9 -(WN OtC961.( 4 UW) 09 Aj cc I -SwLIt 96 s -(ww)g0& !i t)A - I LI I I III It I II 09- *Os- 9 OS 1SWJ 00 9 -(WW)$COo.-(WAA~ 9998BZqg -(w)t CIO£96.(wht9 ggtA I" - Pg6 09 - it1'mLO4-W~ 6t- I(4) OsI 09T0010 -0 - 0 isIM-9 -SW wig sgGWZ.)661 I.-(AUWw)A_ -(wI ONEVsi-/4 SiPO4 -9$Uwq 99*Vt~t 0Wt9t6P oo-swiu ggi (wws6zo -(4~ Z06VAt I to I SWJ Lt C (WW)S~5 Z6 091961L -(wht ggcg(Aht -Aw) umd-eSKwKAq -WW)A -oow -- suuoq *Os- - 001-lii -0Os TT --- - teiSWJU -L(WW)S Wg9 -(W)I1609% -(Ah -(AUWw)A -uow- 69 1 -SWJ 9E -(Www 06(tggol (w qot~q~-(A)I _umw 1619 LS69t M 0 -sWmi VL C .(A)i *ZLIPZ0966 . SL61 ..(ww~gg 29* -Mw)ZWE£9t-MA os cgg6 :W)A 6. I 5861 I . 0861 1-.

II 1 t -40.-eA- IW)Aw -Os- - h L90 SGWxjtr WOM916* -O .s C19 .. 111 11 tg-w19 0 -(wk~ tigmt6-AI 0s !WW 8uw)A PIA - iiWA I -OOV.(4uw)~ 99a owz1 -(w)i 9L cw-(A)I i9969u uasuw.(AA~sp momt- m~umi 1 -- -s -O -O -O F .-(4uW)A_ 05 plOD '*(W)IZM'CW-(AA s~MO#- L09LOLk OS 0- 41.A) s 1. -09- 09O 05- - -O -O 9 / (WW~tarot -41MW)A -gg (W)I Mpg~-(A fJPW - t -SWJU gi Mgg9% I III III III I II I -(wws Co-a -(AowW)A 00 -auuup1 -04e0.(ww~sg01z -(4UW)A_ £s W 0691 990 'Sum VOE -(WU~ LZO .jo5uI O* -(W gg61.m wLt (Wxgols'EI4 09O inBL34 I -09 KO 0Su WJJZ -(ws 6L I (AIIW)A_ qcaoj~q -(w)I ojgEc.(ww~sEGO-(4uuwft -.d - 9601'9=Z -(W)I 969£9%1(A 16 9 ~OEN M*()o -(wI Z -SIIU gX EV I I(A)t -(AJUW)A sJOmoi- po 911 .suuJWg .0661 9861 0261 5L61 0L61 0661 5861 0861 5L61 0L61 0661 5861 ivox 0861 5L61 0L61 Os- -- -Op -O -Os ElO 0O LOI -sum99CAI wl § -001 I AUW)A_ 99 1-wAj6 qSLv9mO ma.(AUW)A_ lid ULGtq -(w)I o~Wg-(A p10 IILFBLF (WI SZ9 ON11111111 09- 09OOgE I e-w)I ZV 6-Coab -AAW) - PP5 I 0 tpo I -SWAJ £E -(WW)$ LEO.(ww~g~.(W)IOW96 1(h-' -O -OL ma~ 0 -0 O 0 060o1 I 09O 05.

27 s(mm). II v(mm")...1984 830 I(m). v(mnVy) 0 39 o(mm). 50 break- 27902329 3 08 rms..116 break -flattp_ t(y). -18.19 706 (m).1.1983. 371 rrms.1987. 99569422 1188 smm) 509 nrms 1 41 50 -50 I 50 Itop_ . 1.1956089 1(m)..401 (mm) 4 15 nrm. 0. 99977823 smm).16 s(mm).1987085 I(m). 3.102 %. 909 nrms. 111528057 -169 s(mm).087 I(m).2731 .. 7248.065 4710 7639 479 nrms.smmi v(mmy). 120040945 v(mm/). 2888.085 I(m) 21521. t(y). 13717...met_ t y).198 108 I(m). 1975 I.chdame v(mmly).77 mm.pl v(mmi/. 081 448 1(y). 187388113 v(mmE. 152 v(mniW) 1. 2. break- 50 - 50 - J ' i l l I l l If il I -50 111111111111 I! T breask --tale (y).3269 v(mmyf 0.580 I(m).94 nrms. 579 r 0- 50 - i I I i garo -montos (y). 0- -50 - -50 .1987 086 I(m). 10723460 666 s(mm).17 III I II I ll II 1 1 9 g9@o 50 - i': -50- -50 1975 1980 year 1985 1990 I I Il i i III .089 I(m). t(y).418 T -50 II I ptt v(mmn)* T IIf--s I -f 0- -50 (y). 4291 5321 -162 s(mm). 354 nrms.3624 v(mnVy).pf1 II I I I I -50 ll I I 0- -50 I1 I I I 0- -50 - 50 I I 50 - 0- -50 t(y).45 s(mm) 365 nrms 0 95 1 0 -50 0- -50 - -50 II II II I I111 I 11111 I I 8 50 gasro_ .1986 073 I(m). 5 70 nrmns 1 65 Ii Ili II I I1I I I II I I diolame -montoms 1(y). 50 - I1 ill [II I 1- t(y)g1987089 I(m).45 s(mm)..65144822 2.06 IlI 1 ll771 1 TIII 1 -T1 buchkdecc-gasuo v(mnmy). 17.12 50 111111111 II buc'l.5314 v(mmPW) 1245 s(mm).1985. 1970 II 1970 I.1987 07 (m).1987.85 50 I I I I 50 - II 1 I I I I I I I I cholne . 248 v(mm) - buckdce .smheac 1(y).shae y). 11606.7940 0 49 (mm). cholnme-gso 1 50 -50 chdolam -shade v(mmP. I 1 1 I I I I 1(y).1305 nrms.1 7. 2 43 nrms. 1980 year .24 nrms..612 rrma 1.1987085 I(m). t(y). 1985 I ' 1990 ' 1970 1975 1980 year 1985 199( .. -239 s(mm). 0 93 50 - 0- .544 I(m).

(wNqcJ1(A) Vppw .s=9 .(Wws 9L91t -(4UW)A 9M~ O1.001 9901 -(IUI)9 *091 60-6k -(MuwW)A WO UUUWl-(W)BCIy -(AMWw)A 99K. Ts -0 iI -S31 itVVMo -0 W m1m 09.61 0L61 0661 9861 0861 SL61 OW6 0661 S861 0861 I1 - Os- t -(ww~o ste9 (w) grngsi-(Ak *09 :Mu)A m8601g -aLwimo .SWJJtW9 -(Ww)$LCt. .(whl s90L961-(AI usotua .(uW)A - Os -0 69*su IIa 11111I s6a It It (ww~esq' 09O .(uaw).iw Ueuu 11111il if it111111ll 11 1 111 05- *Os- 00v-05- -0 -0 05O . 5L61 1 1 .lAM01MOW L '(4~ L90996I~ Iii 3-tOStaS 61£L -(WWs COV -(w)l LC 9%t-(AI (AAwW)A UWd-W-eLNW 05O LITI .wpo lt1111 111 I1 I I t f 11111 - I1 t t Itt11 I It 09- 0O LLo -SumJ199 -(WW)Bg ? -(AMuw)A -(W)i MENts-(Ax -Pqjh .9 .I 0661 9961 Ju~s 0861 SL. t I-0s 41ow .uPxw -0 go. .(4uw)A 11 11 05O- O 0 -SI jsW I?9 -(Uww)Sg M ~ gg~-(w)itgt~i-Aq Itt 111111111 (~ wj 1111 -00 -~~ -0Os 99t "SUM C9*0t -(WW)S -(Aww)A *LaqlOW. 0L61 .%8P* 60 1 .-QAAW)A 6199.t ..-picw6 itgr-(AX mmw~ iiugqE= 1111 -s 0 OVLt5 -(wNg tftjgg-(AX -unuu .-SUmU ae.-(WW~l tt -(4AUW)A L0697Z9=O-(W)I girsE I-(AM £991101.01uem~ O0tILMi go011t096a I IL I astiem 0161 -(Ww)sLt'l e-(wxg -( £69699i-i 09O MW)A - cmintj -tu-Lw 1 L9*t 41uA 99'6 .

-(W)A -(W)I 690*9961(A1 &d - J9E 1 So1 -SWA1 1I tZ E -IAU) I I 0 GLEVOLSV.A e*iow -Ojloo *0 000o"swzo000 -(WwU~ to -(4uW)A _ L991L699 -(W)I 9iV M1e-(AA 700-114 .(UW 90L6C699 (wMISIIM g 9g1 -(AAUW)A oggl.ajqoo w-sc -(ww~s 910.(w)l WI6O9LA -0 09 -09 . JW)A -(w)l gsW0g6 t-(A)a gqe - -001 qupd OfrV9K= =(w)i zz tgs 1-(Abi -od *09 .K a osems0u~t*% (w 9M9t( 4m No uwiuSrq-(w~s z o - 09 -ww insA~ L0o *05-Ub.s3.-s if fIIili ill [I if *05 J 111 OS05 -0 6SW0 cci.qOLM 09O- * IY 0 -0 0O 000 -WAS 000 -(WwJ) t9g9O-wjOCSg6£-(AI j -(MUW)A -d-pitus Coo0 -Swj oo 0-r. gM -(Wwws 19 0.-(AAUW)A_ EVCL9?S -(W)g IZO961-(Ahi iieAM3JqDD LO0t -WAJCZ C *09 *ON001- I. I -0 OI t SU 6a 9 (WWI I IC I I I I I -0 09 ESt -SWM 9CL-9% 9 -(WW)SCQgZ.0661 5861 jugs 0861 SL61 0L61 0661 S861 0861 SL61 0L61 0661 S861 0861 5L61 0L61 090 - *05 99Oaz 0-.(A..05 L 90 "SWJU69Z -(wWWS VIL -(AAUW)A -p iq-Oo -(w)l 0z99MI-AM 99LtLt6L LIPtI 5WM V10 1 -(Ww)s CS 99692101C -(uJ4i6L 96-(A -(uW)A )AP-Z-wUOPw p~ gzgt -(ww)s 999t (W)l ELZZ6I-A -(AtwW)A_ oGPBi-- S3590 *05 001 *091 811 .wzat 96(wh 6was -(WW)S t(A3 M-- -(4AAW)A usojmw - zto __~ ILaiU) .

205491467 -236 s(mm) 5..soledal_ 1(y). 5s I lill 23875.19#1962 (m).9921 I(. 1 11 0 -50 0 -50- -100 -150 -50 - -50 I I I I I I I I 50 wmue .119 50 50 avenue-t-frmunt 1y).-i -50 -50 - --.57 s(mm). 52 nrms.82 0 prlon -saw c v(mm) tttt II I t t(y). 1543.saw 0c ). 1 16 0 wenu 50 I I I I I I rtrd_ (y). 50 - _(mnjfi (y)198048 Kim)17539.0.36626 469 nmrm 0.1985214 I(m). I 1970 1975 1980 year 1985 1990 1970 1975 1980 year 1985 1990 1970 1975 1980 year 1985 1990 .28 rnm 1. 153 I I I I I I I I I IIT1 emannt.pelona. 24892. -396 s(mm).10 S(mm).25 - manu _lema -.57 s(mm) 5 12 rms.230120405 -1. 126496353 v(mn ) -.79 nrms.10rrms.46 50 0 50 0 -50 I IlII oaosdld . 27003. - 0 II - i - 'E I I I I I I 18887.1. 327159967 v(mmy). V(mnmv) -2.7016 5.03 s(mm) 606 rims.1985238 I(m).1985204 1(m). 0 37 50 0 -50 -50 l I I I i i I I II f rmant-lhumb 1).48 0I.1954. I(m)-33259.5.huntr v(mV --" II III I I I | 1 I 1(y). -1. Il l I I IT T 1 I II (y).90 s(mm).thumb v(mn I i I IIIII t(y's190518 1(m). 174223109 v(mfmy).23 s(mm).4672 0.'--.1.75 rrms.. 4.17 I 50 I i I I I wmue.-soledad_ v(mny).98 50 0 f 1i |iI I TIT I TTI I-- I -50 I 50 - bdid - ld 1y).3788 1.66vms-1.194.64 3.'----. 1.1t (mm). 2 69 nrms. 0.39 Smm)..1277 (mmf).1985230 (m). 5.926 I(m).994 (m) -1.4899 v(m-m 29* (mm) 6.04 0 -50 IIlI IlIIII III l y -50 50 - l 0 50 I formant .6.756 (m).10 -50 - I i l i l iI II l l i II I I I III fairmont .10489 I(m).3326 v(mnVy).103 s(mm). 7 16 rims.pelona 1(y)1985213 I(m).8890 nrms 1. 1823. t(y)-1979.93 s(mm). 346 (mm). 1960531 v(mn)r.1950513 l(m). 0 97 -I 0 -I -50 -50 IlI 50 I wenue II III .2 rms.r> labu -Itma lty).0. 21404.59 nrms..

..- It 99 9 .-(4II W)A -(A4WW)A 09O 090-SWJ91Z .. I ..A-L I .ewo=P -1 0 L60 .-soiod gIC I basmsI - e9-W~ IEI-UJ EM0 =SWAJ96g-(Ww -0 IiW) [oo l-(4uw I -uo%- 1I II I I 1-l I I II *05 VL I .(WW)990.Q(RIJW)AI GZii scza -(W)I 9tL 096i-(Ahi Thos-as - -SWJ 999 -(ww~r 699 . 990 96k-<A t iuoiei-eGiuoV o 09OS I I I - " -I 1 1 -i I - 1 1I 05- 09Os -0 Zt -SW. - 8106 0861 9961 - . I I .. .-.-ssisjfl (W) CEOL96L. I .969 -(UW)SLZ9.u oz gs-A -Im& I 09 .SWia 9Me 001t I I ft I.IM I II I I II I I I I 09O I 11L I 05- -- -0 0 0O GL1tSWJU Zt 6 -(wws C9 t.- - I .-I ..-(4UW)A _ 0z 1 . I os- OsT 0 l--T oz -swiutg K .SUU 99LOtMt ££ -(LW)s 09It(AuJW)A sssub-eomp -(W)i SLSOMS-(A1 -Os- -05 -0 V91SU t 1~ut19 -(ww CI s -0 99 1 -SWAJSL 9 -(ws LZ .(ww t6o-(~ w)A -- o = i .0661 S861 0861 SL61 OL61 I ..I 0661 0661 586 0861 561 0L610861 5861 5L61 0L61 06 0661 6.Mm .(ws 09I.-(A4uw)A LS I W3Ju 099 999199L6 -(wh tc6 1ig1-(AX iw. .-(AqU~w)A 999tL6WL =(w)l 99Z1961(A1 zwuu~iei-ssu6 ~ I~ I I fI LzIt sWJ vg9 -(wws6t I -(AAUw)A t60oWA -sij Lictggm£ LM OOZZ~ =(W)lLK6096t=(A~ _Wa-SIUOP 0s -(Lws zs -(A4uW)A =(w).. I . L106 OL61 SL61 I I . ..(ws LVE -(MaW)A OmlOd .(4Aw)A £9L993 -Ml) OE9'o9 1-(Abl s"10d.I .

o *swm u taL IX09 (ww)s psIA 99o t .i I 05O- II II I I l i l t I I ~9- IJ -o I-I *91 s- GL -iww)s* o- OWz*'LZIg . 0o 0 L OS tgi -Iwm tre .tuuWClp 99tosumg Stwot g(0g6L-(At oo -(Aieyin izi .(wht Ift IlI I i lI III II tl II lfIIIf I I I I I I I I I I I I -OS.(4MW) zm"4%it .(w)I M9096t(AM -(ouiw)A ~.(A .(Aww)A tli ett it - 0661 9861 max~ 0861 5L61 0L61 0661 £861 0861 SL61 0L61 wo0 um~ LZ z 096sEL~ .apnbd oaovmz O 0 09O '(wN 9grge &-(A) -Pp"=c .iwkm.05o - -PIqco.au EO -(uuw~s96 .4wtu~ £0oL6mia -(WAg goosu.suuu wE 0996Sw)I LEVZ95(A4 -Os OS- -(4UW)A louolei.

I I -I I I- Os-60-WJ199 -uwM~ VMMgo .(Muw)A (ww)B ino - s 9wg99 1 -(UWwBSC 0 796(A I 1I I3 I -(4UW)A %W3ds*sl.1Uf " 1 9961 . L . SL61 4 .- 0861 .0661 I S86T I 0861 . spi -u"~ Zz i .(Ah L I K 3 Os- - lll - 4-- 0 O 0 6a 01A -Suu I -(ww~sIt . .I 1 -I 1 I I I -OS- (WW)B 0661 i U -- -o I61 -sm 5W1 -(4uw)A 190- -0 53J 6Z I' 39£90 (W)I 999-(AiM to0 -lUul.-(4uw)A *933699 -(W)l teggal6P(A Inm t - slm u 05 09O 93 1U311Ol -(wS OL V -(4WW) %*drulI-6u1 -(u))i LI99£60(A 96IL"Mt9 9iSL0131 un3 (ww~sgzo !SW) £0630691 .L -s -0 09O psMAw4 . Log- -0 910 05O -u- .1 9L61 0661 9861 0861 SL61 I. I .09uq I.-(ww 903B 106900t6 '(Wk~913VS"( £961Ah 16-a goI I1 I I I I I I I3 3 I I 13I I .(qO~ VvyWpo- 83-A~ am9s961 M(WNwa~ L I1 II I I I I I I I I III I 09O 1wsqa I I I.1wuasoq I -Os- -O 9L*0 W311A1 LLZ ELK Il -I. .(WiotilL696()I 3 I III -Os -(4tuw~ok Tso-Oua 333311 III 0s I I IJ I I I I I I -0 I 09 so9I-sum so99 -(iw~ atj t-(u ZZ 9 1 18" -0s IIJWl t 1 9t-(A~t 669938093 (wMea LI 3I3 I 3 I -(41W)A u4-draV .(W) 1169t61jA.(Wtu~g Coe .(ww *(AuwW)A -(wwt 9L0 09T -0 199 -ull ~ I9SWA3063c -(ww)B Ito 01luzL1 "(wh g Lst-(Aj :t--uaa s - 093 IW1 13tZ9-(WWlU t0Z- I I I I 0 I 33 I 09 -(AUW)A nis 8s )09MIS93 -(WhL 99j6j.1IMM I I I I I 3 zoo 09O -sum G6930I I II 3 hs -(WW)B so* 3i . .

-s -0 Zvi emisc 9-(ww~s V.(AUW)A *1961*99=9 -(whi jW6nt6-(AX smau.(ww~sawa-(wkgq j.so-jopinq -05 I -- -0 -(AowW)A seiaew-joptpe -Os Ij .6t9Lj-(At I I I I I I I f I I I I I I I I -Us.(kw ie~t'I61M owqam-*sRl I I] I 00o0-SwiUoo -(ww~o I I I I I I I I I I I I I I I I I 0 -0 ub- -Os- 09O -(Mw)A am9616 -(wNg 1g96L1.(AX 9i1zat 4w)A tui-eq ea E~Si 0 - - 09O I I I I I I I I I I I I I I I I I I I Lj.(Aww)A 6WAISUULo 1 -(WW)S90 0 -(4iitu N66L£1£ -(iiSggg£191-(A~ buOF-us- s"e fi se" 09O e~s ~st -(w) £966 -(Ah dctpw-oi-JDMII~q 09O 1 -- i -O -0 001L -Swiui99 9 -(ws IL IMo09. 5L61 .-L - 09O =moiw-ee*q I Ii I I I I I f I I I 1- L 1 1 1 1 I 1 I1 I I' ' i' f I 1 1 1 -- 9 E 096 SUo -(Wt)g -Os- seq) -o goig ggt -(whgto &961{(A) o~etI l l 1 1 1 I1 I 1 I I1 I 1t 1 -0 9t10 -SWAJ09 E -(WW) 010..(Wg £0Z-Iq(AI LnFu-ipug -SUIlE*L 09O 01*0-SUIlt WE9 -(ws 10 0 -(44w10 3599t969 ..A ZIz-swjj Leg ..11.I tt 9 -(WIIJ5at I . 0861 I.0661 S861 0861 SL61 0661 S861 0861 5L6T 0661 I 6mA- 9861 .opgsqq -I I I I I I t 1 1 I 006 I j II I I I -I I I I I I (Aiw)A -(AX -uaeu -jopinsq I I I I I I I 9Z 9 -(ww~s 100 11IWZ961.tt -(w~og Lo01 WJJt19t~ -(Ww~sIn -(AAUIW)A Zq9£ 9£991 -(w)i t£60961-(Ahl sb~/pd .(IUJW)A . -O -0 LOk -SWJ. -- I4=-T--II 901tSWAJ 09O 9L10ISWAJit C -(WW)$09 0 -(AUWw)A 99£ 19091t-(w)l 197zg6 I (Ahl usLie-il-somMqs 05O LU . (w~g.1.

.(WA -(wu)IL96 0981t(A1 D600ew.0661 5861 0861 SL.OBPiOO 05O 601 -swAI 99 C -(WW)S tZ W20 62V6 -(wIML961g-MA I I .(AX -uop .(Auw)A 919066M? -(w)I g2t OW1.(xw)I Ip (w 0861 0O I 9D t Mm Qg9 -(uiwstoA iS90L99 . 05- -09- -0 001 901-s .I -(WW)gS -(u0 s euo9 1.uoumku" I *05- " Os- aS60-s~ UOPA [ Os Os- *05 L90o-lu oeE-(uWw~ goo.() OOgI() osq~ I I I I !JUW)A &Ug~ BM(~W .seffluOW tv I .e6pipo -- -ub.(W) LKS6L6M. -0 c swAi9 *ss (wwsL9 z-Au) Z9699 t961 -(w)l OO7M H WtA)1 u- 990 -swAI a2 * -(uw 6s I-.(Ah *A - 69 -OI -amp I 001 aDJt -su a1 -(ww)s zs- (m 0661 5861 ) -09 LoaIA I I - SL61 -0 to I wxjLt 9 -(ww~so1 92k ca .s6Fipoo .-(MAw_ 5ULU £P1tWA LMo9tgi -(WkI6LZ'tg6l(A~ UOSIMff.SWJ 99 -(Www 96'.seinuoul 09O 99 1 .Sui L9 9 -(Wws69OCIP066M60I -(W)l 696 196M(A -(AAUW)A -0M -.-(44Luw)A -LopA .owAm 9 (w)igoogst(Ah II IIIIII IIII IIIII II" Os0.A -uows .sflP~po 09O -0 09O gSM 09O .-(4AUW)A wpod . I I ii I 99-0 -s-i gt -(ww~s cs -(AAUW)A uooew-e6PPMo 6M 6 -(wJ)I696096 HAX -0 09O t'N 9COZ .26111mw 09 -- -0 -0 13 1691 '(4) 691 WS6HA~ ZsUaWW .61 0661 9861 0861 5L61 I I I I I L I (Mut)A pos- I I I.&3VIA gL g69Mi (ww~st10. 09- i iT 0 ZV90-s-A 9 I7Z9m IQ0 V(Ww~ 6E60-.-(/WW)A OZ96I 109Z -(w)l ELC098 M-Ahl 03600.

I I III -(WW) I' (MUW)A -(wkI ggWOgM1.6[ 05- -o 00*0 -WWJ COO -(~WW 91*0. 0861 I SL61 0661 S861 0861 SL6T I 0661 S861 0861 5L.01w .( w)A 19 MgAX o0.wkui-" I I II I I I I I p I O5- -Os- -0 -0 OLO0-suu Oct -(ww~s 9 Wir L969&'(Wk~ OIZOgeS- W) 09O OJIw . I.0661 S861 I.(W)A :!mc~u I I I or I.J- 09O 91 1 sWim O.(Ah wri9. jWuAg9wogsj(AX I I ii I Ii I ao I.(wXg Lwm 1(AX &*Wqcpi.Llruq Ozto OagI .(W) IL I I I I I I I I I f I I f II I I -o OsWUA1 OCS . 9 -(ww~s zsz .-(AiiW)A 611 M-ZMiI -(&uNg gg0q i-(AX i*Wpmp . III .01 [ I I I I -(ww~sc I.-(A"Ww)A Jet -(w) WZ'1961(A1 11 UJOI.09 -(wW K .03913-Ai -(4AU)A JIB6LN~ 09O 99o.in*puq I SWJJ ECL -(WW)5660 I I If -IN lWqI -(4IUW)A -0 -Attiibe-iiom Os- -Os- 09O I - COO -m nume1O 0 0 -(WW)s W O _ :iuw)A ~ 9 9mJ-(wws to 0.qcplop - 0 T 9Ct-SIW CDL (WW)S L91 -(411wM GtL-L I a.Lupqap L 1 1 1 1 1 1 l 1 1 1f 1 1 1 SIC1 'SUUU CZ ILMEM~L I .-UXSS wa - I I =L6061. . (UW)A -(W)I WgS 1-(Ah si6A4pd- (U) am~~~o 73p16 .(AM -.wva L6I se1Js -(uw~ st I. (WW) 9t-f 9Z't -WAJ Z1 t-(WmUt t9*I 91PZ996L9-(W)I 11.-*MR~ -(W)f 1906961 ILLt ttf If 9Z1 .U-j"1. 0L99*L6MI..lu6pmp I I iW-puwp 09 I -Os- -0 I *9O0 WAI 61 V -(WW)$ O96~9L -(~f~gOg5- III I ii I I 0o1--._-Cu)A eO69-MO -(k w mtpg-gs.I -(4iuw)A [ L -(WN990*go61s(AX $6671woiL-1111rP I I I I I j I -I I I I I 05 I.

I I.(ww)B 190.co LI tS-1' -09 WK 99L t I -1 I1 L 9* V -(wuw)sSL*t -(4U1W)A -(w)l __ os t-(A~s 966 wot- k 05 OAS I 09- -- -0 Ogo -SWAJ t9C -0 -0 -(AAUWwA IPWP£6991 .-(4WUW)A__ U'l0 eoowbtMVita I "(uj Wqi M09S-AX I I I II I I I I I I I I I spi I I -0 -09 M9 tSWAIU 066ZOtLI 999 -(Ww)s 6V*0. . 60~WJJ ~9 9009£0 -(4uw)A ."(AAjW)A_ -(W)g09L0961"(AA * OnApG- WgO-SWJJPtLt "(WU) C' . "(IIww)A 9096 L99t "(W)lW£9QL6t=(A~ eejMPW-JSJEIS 901I SWAJ CCL "(1W)s £9100ZP*LZ91Z "(W)i 91966t"(A1 05O J&LW)A UPI-JOUSI -05I III I I I I I I I I f_ I L I .0661 5861 0861 SL61 0661 .(Wi 161-096&-(A -usiu- . LAJ100 .(w)l otg~96l"(A)I -09 L6tI9t=0 .0 EtI 901(AI (A$AJW)A SJOw - ~ 09 _ eljqmp 9Z1 .-(hiUW)A OJDIW- 0906M I~1"(w)l £wggJI-(AI I I III I I I I I I I I I I J&Pwfl I I -09 V-qA4WUw)A 0Z0t SWMUIq 9 -(LWu~g roO KU6 11091 -(W)l Zse969I'MA aUOSpeu-i I I I I I -(ww)s .-(AAUW) 009991001 -(w)i Vtg96 I(A)I -uosIu - cooSWJ~Ju at 919C0990 -(wu)l 9Iflop -uwW).J 050 -05- 05O 96u00t1 -(w) 9t10961)sIo.9"'00991 -(w)l tt6 9%61-(A)h -emw . 0861 I I .-~) -(4)SWO961-A~ Lu we - - 0 5o Os - S -0 -0 I ZIS0 -ULJ 6* E -(BMW)SV9 0. -- I -0 I 691 "(BMW).-IILWS-J03 a9 0 -LEZ -(ww)s tL . II L I 9861 .siump -0 OP-I 4SwjJ 9E9 -(wws LZO. (uj)l zoo'o06 1-(A)I -m~ 060oWJ -sj " - QISLq~t St acc(ww)s CZ I. I 5L61 0661 5861 0861 5L61 .(4UW)A IPAO LW-ene* Ll i II -I I 1 - I 1 1 l 1 1 05- -0 000 -w ooJo6 -(ww)s g(o -(4~uW)A 1.

uflig00 I II.(4uW)A Iog t -(wj) MM'oos -(A~ dcqpo- 05O Uowf I it Iii tilt I III II Os- -0 cc L -sum ocs9 (ww~s69o0-(AM W)A C01.065999 -(w)i gtkcjI-(A s3oA4Pd--umin -0 LTS -(ww~s6og9(MUW)s WLJU 901 ams wmsz -(W)ioat tgs M-AXInwOtpo---~ 190 -WAJ ZV -(IWWS91t0 -(A.6t9I -(w)h LUjM-(AX rUn_.(W)II966ogo-(Ak 09O owe -O 99o -umA ac -(ww~sc~ . 10 - iPl LZI ..-suag L996oozt 0s 00-0 wmu00 o -(ww~o to t-(uw)A ee1LL. g LOWOLIS&Z -(wkg 0995 M -jftWw)A 04- L0PAUw 0s Oroe~ I -MAIBet--inotVQ-( () ui Ill I ~ii ~ ~II. ~ 1 l V2Z III t il(l ==*4 aq) g"tveg .(W)4£91 Z0 -(Al L*Uz -5-mflj 4WW)A 'bgon !M_ JOOias-mOIu 05O smz - I.0661 9861 0861 5L61 0661 9861 0861 5L61 0661 5861 mox~ 0861 5L61 09- 05- -Os- -0 -0 I -05 06*1 -sWJJ t to -(ww).-mwo I I I l I iI It Il I I I 1 I I 09O I Il -0 Wt -SWAILWO -(WW)SZZ'.! -05 I 05o- 09- 09 -0 -Os- 9t 0-SujgIqt -(Wwwsgo C -(AAIJW)A QEt9V6V9t -(w)b tqgstst-(Aht Lenflms-isotuoq I I I 1_ I I I I I ~I III -- -0 -0 09O I-.9 -(ww~s 190 :JLW)A SmowZMW90I . Ise1 -sumL 16.SWIJ 90 -(WW~l96 onvatlG -(WkmaI -0 gs o -swULM itg -(ww~s qz t.(W)x Lg96O901(AX msow -vntowi 05O I~p 0 -s -0 t (ww~s9OI4 !1w)A -(w)gmwt-( 04.-(MUW)A ILI I I I I I I I I I I I I I I If I I I LYI .IW)A GM56M9I.-(AjW)A . -(W)t Z91*o6-(AI t ..

-(AMUW)A ME6C66L9Z-(w)i vW 19B-(Aht o~uqms -. I .en~tu 0 05 =(WwsL90 ttIt -WJu Kt EtI -(AiJW) 09- -(Wwws t60.(w)I ilL ooe -(A)L sW.uktmi. .±ApnW)A 96L6tCLt -w)j t/G60961(AXi . I SL61 0661 5861 juas 0861 SL61 .0661 9861 0861 SL6T 0661 I .(w I I I I II 1I I I III I *ZZtZ~ -(w)i 91L0961(AX dc~12mo-sfloAbod 05O I11 I I I I 09- 090 ILO 0W-- 09 V =(ww@6ED.(W)g99gL6t-)l s6*ipd -OeMuew OOS9=Z9 t s. .9L61-(A)I -UOIU- ~IS4 -_ -05 O . 0861 .9Z 9 -Quuis IL10 -(A~iiw)A 1660$6991 -(W)I Zg.*pd -- I I I I I I I I tI I I " I -09 05II I I I I I IIII I I I I I I I I I I 09- -09- -0 0 09 wo 9Z -swAJESSI -(WW)Stg. £861 .-(4uW)A -UOSgOIJ.911L 910 -SWJJ ZL -(W) gto.-BMW I6-A $0806U991-(W) tM I 0 0 09- 09- St I 05 -(60oQWw)A 96 0 -SWAI0L1() L=1S006 -(w)l 9 96tJ-(A).-.. - OS- -- I -0 sc0 3U1A1ELE -0 4ww~osa t.sfoAod -09 00-0 -*'A 000 -(WW)B094 -(ANW) ool-~g~( swaia-sloAgod otavism .-(AW)A WJU 619 -(LUWJUJOB -(4AUW)A -op 09- -05- -(wht OL 096I-(A~ASBu-OCL- 19Z egm SWJJ t6S o SS -WA sunwu 05 96 =(w)s )A~ gl-(Aj~mA Of 0 -0 0 9l 6100 L0ooz .u.- -(Www SWJi01 I I I I I I I I I I I I I I tI I -s TF 1- -0 01 05 (A4UJW)A 96o0 .QWW)A ot 4.(W)I619 IL961-(Ah I L I t% o r05 -Su Q(ww)A Z -(ww~s 99~ ot 05 exoow s6APd - -(w)hgotsm96 7. .

WW)A__ ODKOLE91 -(wAg0609L61-(AX -amend-- 05O I I I I I I I -1 I I I II I III Og- -0 LLO SWAI a0£ -(ww~ czLM6 L901 -(WXi9LI6L61() -05 I 9 05O I -09.(uaw)s uot1UU~ a-(4uw)Awp SI' uimJ 639 -(WW~l66Q :AMWW)A __k0 W tg 1A 0IZ' I= *IW 0661 £861 9t I OUmAE96 0861 SL61 -(ws 19*1 -(MUW)A -9610 -Os- -0 -0 st -sum asg -(wiu~ LsIt -!Atw)A_ M 09 ga g iv-Ou Iu tIiI I I - I99VVSM8 -(WXiLggs~-A~l 1 L 1 -1 L L j 1 1 1 L 1 mW-MIS 1 L 09O 1- -0 9c lSU I I I I I .46L61-(A aoewoto-sfiApd 811 6Z1 .0661 5861 0861 SL61 ---.I 0661 S861 0861 SL61 -0 0I L 05- Os- wom-sum a'(h It .(w)I 69J1 t-(Ah iJs- 05O unIms Ico -s -- 7 0- -0 I G91 "umA LV6 -(Muw)A -(Iww) Zt- 05 90 GWAOI 19 (WW)B£11 -(AiUUwM __I 9s1v~E1 -(w)I 1c9 1961(A4 osiwotei. I -0 00'0 -SWJ £00 -(WW~g £0900 -QWW)A a0ot*9S6g -(W)I £goUo1-(Aha -oiae t L1IIIIIIIII I I I I -(4nu)A _OWAu-0 t I I I I I I 05O I I I 5 I -- -0 0o -0 091S1 681t-SUmU019 -(WW)$91 L -Qv'IJW)A ! yW)A 901 swx' tgg *(WW~s 9ttWM6KOGI -(W)I vt 1 085(Ah Joe .(WW)S I I I 90 I -(4lwW)A I I I I L -0 000 -MM* 00 0 -(WW)$ 19 -(/.ao)RIuS -0Os I *LI 6zl-A WtI Iv 05O 998 gg-(u~ AUW)A 9W~6- 9LE0£0801 .-pm -0Os 09O sum 199 -(ww~sgo v -(MUWw)A 9ZL1OM -(W) 1.

&qst-(AX oboom . I -(kww)A 6*6.-4mw 09 ge.1.96 1-44 -jobw .CKT -(wN aWI96L-(AA ow-wnwf--97op LP- os- 09 09- 0 0 g9o -swiu GZE -(ww)s 96*0 -(MUW)A =C6=Z -(4 91.-LPOW -1010 05 0 0 0 Mr- 09- OS. os it I -u- got -iww)s o9 a -(AAUW)A _ 1000*91KU -(LuX9Z6616I-(AX WOMPP- L OS Z9 0 -SULBJeg t -(WW)l ES'o -( W)A LLL6I.(ww~sIr (Mw 0#LV*OL I -(Wk 3 1-f09 (A"A -Wsn -ThtpW tz I -suiju It c -(ww)s tra -(AAuw)A LLO17U6* I -(W)l tWUS t-(A)t -ucws .I9U4uG3 SIM MLE 690 -Uuu 999 -(ww)s go c L9992S ILZ -(w)I ELI 196 vio -(AAMW)A -9-91g) .PUFM t6 0 -suuu La L -(ww)s qt c -(AAUW)A 09 -(WN EL 1 196 t-(A -GIWIP .-pmp 05 96 o swiu a t 9M 2 Im 1 -(ww)s sq.IGUWW -Suuu -(Ww)s 991 '(WLU)A S%960mz -(wh ZLL-1961-(A -0101.196L-O)l -WCUO-- 09 ZI I.I -swAj 96 s -(ww)s tt I -(AAUW)A foovem I -(wh 6L 11.0661 S861 0861 SL61 0661 5861 0861 5L61 0 0661 5861 iuoA 0861 5L61 0 0 05- 05- I-. I'Suuu 9W9 '(Ww)s 16 9 -(AAUW)A ES LOOZ -(W)l SW WSMAU OW-81xv .puwm os 111111111111ILLL os- os- 09- 0 09 I"-. -UM G6 t -(WW)g Zt 0 VLLZ96Z9l -(w)j6LttqBL-(Ah i i i i i i 05- i I I I I I -(A"W)A -oKpip--97vpD i I i i i i i LOI 'Suuu 911L -(Ww)s L9-E -( u1*1961-M LE91 em W)A 9 quo -0101- 09 i - os- 09- 0 G**I -SWJU 90 L om-Ema -W 11 -(WW)S 99'? -(MUW)A nrlost-M -6&-go-putueo 09 SC'q-SWJUES6j IPW2M -(WW)SZq(I- -( 05 W)A -(WNMEM'(A -SM-40 .puww . Sf1t "m* 095 .

- I I II I I I I I I I I -(ww)A 4DOPP- I I 9861 0861 5L61 Yh -0 956996.(AUWw)A 1." -0 It I -sumS NZ (ww~s (a tV0P9IKz -(wx sm6gi-st I I I I II I I I I I 6091~-T- I I I I I m I I 05O I I I I LE I sum1 VL6G -(ws6 I I I I _ I I I I I I 001 !JU) 1 -1 1 1 1 1 I -- -- -O -0 96 0 -SUJ ED~ c(ww~s t 1tf tPOOCI -(Whi6IL-MI-M -_AiWW)A 10174ld. I 5861 0861 .9DM -o LSSC ISIOP -(w) 119 ( 09.-0=1 -0 09O ga-UIAJ LIS -(WUJB LQC -(4WW)A 91111%aI .I 0661 S86T 0861 SL61 .(kww)A KLC6066Z -(w)N69 I0961. ..w~ -OS I 0661 L-.s 09O msI I I I I I I I I I I I 3- I I I II I I I I I I I I I -(ww)5 19o- gZZ -SUM' So6 1119 It It - I I -(.=BwP AMvP I II I I I~~~~~" I II III I I I I I I III I I I -0 O09 MW 09O I lIl 09- -Os- -0 69 1 -SWA 00Z -(WU1)S9'J .1-.06g9L -(W)I 91&'VL6I..(X &wPeOW.-0 6t Z -uLA1 ZZLI .(Ww~s£96V..-oid .(AI -WjO - PIMP .(A~ww)A -wpa-4 -- 09O jArw 1 I 1 I1I 1 -001- -- -- -0 -0 . .L . 9 -(whi V9LMZ6 (ww~so09 =(A4ww)A 6LI1f96 f. 0661 . I . UW)A I~U 1 19 MA IRW - -Os I I I 3 7 05- -O -0 -0O 00 1 3UAJ ZC L -(wsg 09O -(MAUW)A_ 066LLL126 .1.(WXg 961 096 VOXAI2MIPP - OL0 -SWAJLF3 .(A)i -WMo - -- 001 05O 690 -SW1uO 69 -(ww~s go a G6 t6M -(WU 61 I6 I(A 9909PLv I -(wNgVLI I 1-(A~ -wsnd .(W) 66L 1961 A -0101 - OPP LLM15C69 -(4 66C196tCA~ VWid--OcWP £60o 3wi.-(AiWW)A 12goK -(w)j g9£I&gsj-(A& -omu-WAow 9L1t -Wi 601 -(WW)s6L-Z -(kWW)A o OW?. I . 5L61 .(ww)s L9'9.6.. ..

(AbA -UGws-Wd=o I i i i I i I i f I i I I I I I I I 1 1 4ww~s is -(wh I If I 1 III 1 0 -O T ge t 9Oo -UM Sl got uOa-j -(MiiwI)A E~ -(%uk-(WwU)s m~ 196-(AX .(wh=Z~~ki(A~ I -0OS _ smeq *Aps - II l -0 -0 Mt6 MfSL t(~ 9W&L6t(A) ZE-ww~pg .uMM F Os 09o-0 Os s .- .(W)h III I I - - - I'(A4 LWOWU I I .L6qg9tt Ia ff IIL -(AtW)A goo .(Aa etom-omum3O If lIfIIIIt 1If 1 1 11 1 .(uih .I - 0661 9861 0861 SL61 0661 5861 0861 5L61 0661 9861 0861 SL61 Ks- F0 -0 -0 aZ !AU)%(WS IwW Wi I asoSWufSg .Sum~i 99 qww~s at 9 ww)A -OS gmmm~fD q(Wh~scozt-(A~ UOMiw" I I iIIII t f- I ..(WW)BgoI -(4hUW)A 0. 09- I 6t j -SW 919 -(WW)60 a-.(4aiv)i qZaj -(AX adoo--so tog I - 1 -O zso -S" I ago-swmu SE (wwso ( IisgtOWU -(w)j =Z (Muw)A% OWIQ-spq -Os 090 -W5m 99 a(ww)s Z.- -oew 1t I aopbiq - 11 1 1 -Os 1 h . I I 1 1 1 A 1 % 1 f ! I -Os- rs -0s T -0 ~OWoo I - - III - - - . so t -(4 uw)A 99fr9~~~~f6LI~~ x~o Lt -0 -0 ~ _ 09~66gI(~ Uoi~cetj . vmccmqvW -(wNgetigi.*w-d.(A~ -AW)A 3VAPM - IO - J iI fI L IlI i IfII I I If I I I I I I 05O- -I -0 LlE6 (Wfff OvL :JIW)A -0 05O 980o -(ww.Z66(AX uws -- 09O pqw .. "(&-WA E69O9LL~g-(Wh 6LgGLef.LU to 6103W1 0 iWqpp - s-gq w z -~um (ww)sg 09 SC - 6S tSZ -(Whg VL96L6t.

-(wI cootwmAi tv9 -1nu9 0g ..-(AtUU)A gOoU-I ~4nw)A LUOtOA-6S~JB g99 -05- I 11111)1 I III I I I I 11111~~ I 1 I~tI 1 I I I I1 to... Os- 0 I..(kWLw)A -Os s-mjcs I 1I 1I I I I I I I 1I I I 05- IIt -Os- I II 1 111 1 6v II -0 "A I9"900qu) -09 &*6 M 05 M(w) toz~-A -OPUBJ6 .II3IdiSo -W - I .I 066-1 5861 0861 SL61 0661 S861 0261 SL61 0661 S861 0861 SL61 -09- I -0 -05- -0 -0 tLo SwUU eC eoi~ujm~t -(w'* ~~pg~j . z-I -I~ Q(Aw)A 6m~196 I-AM t9O* t010 UJ OLQ:WAW)A 0LO)'I.-SWAS 9t't IF Lm ()1% 11(W)9K 096 (AI wooew .oqgoc 9v t(A~uWW)A gNtg'tLOC -(w)I 6L9196t-(AX -IWOIIM010 0 - -s -0 -0 05 60 (ww~s zs~ tz I..(w)~ gwogsi'( -(wtu .jAI) 161 aDV 001 t'I 9BtwoW..olo~o a I. s5uuum9 -1w "i6W)A uapd. 05- -0s ZLII -umA5qL . & wgt ~ LI I -Iww ss(A I co II - fiuospM.qtIn I I I I 09- -Os- -O -0 gm We .(Iuw *(wvowt~ Bo6n t 9g ua)A -A) LVI .5@ULIU -Iw~ zog PL9M96 t 14w) 99W1.O~u~a _ ImoSIm- OW'Ootel -(wI te616t(AX opo--UM~ flfl 05O .98( *rA .

mWA w)1 GsILcIstil -(whgtwtmt.w) t"riu-(4 sz 0-n 09O at$ III III)B MAWIIII 0-ll LJ II I- I L L III I I -O u-at(m ~i-s"m Ti 7 -0 -05 Ut6SO- (wg tro mtA MIA aso as av (ww~o s(.e -u~n Wlq . .W(WL)St V . *Os.(o j tt -Os 3PIULN 60 0 L3POKVZ (hE .066w"1 S861 0261 9L61 0661 9261 0861 SL61 0661 9861 0861 SL61 "I-- *0-0 - 09OPL a EI."rum -0 -0 W9 -~m*r (ww~ o.-(AAUW)A ZMiitZ "(wNg 9PZW-A unsum.AV. 11111111 09O SS& pm1111maN 111 111 -- to0I -PWANitL -(ww~s wo.j6 . I f )II 11 IIJ 1 l 11 111 - 111 -0 -0 nct -mv gov -(mw~ U-0 ..(uwW)A .~~) usa ym (w u. 0919*66UI -(wA Wto-A 3Sifnw -*a (AAw)A 3QWc-vSuR* tpI-W 11m vmom~Ogl -Wg 0MIN(A - -Os. -O~ tw@mt-(wk __ dPm Xf m JIPT -4aaw Cra (AAw)A .4uw)A mein.( Jww - vwm*s ow -001- 00- 05 -Os - -0 -T - Os- gso -nm szww WI.t -(a -0 S660 -- VJ 9g 4wwl EL V.

29565.957 (m).-561 r(mm).-1. 24145.soda 30639.39 s(mm).39 s(mm).soda t(y). year 50 1i111111 1111 in _ .65 nrms. .-.oda V).902 nrms.~-- --- -i===E. 135 sarlon 50 -soda _y)190 425 (m).soda v(mny'1 0- -50 I 50 buff 50 - 1985 1990 1975 1980 year .1735 Ymnv. 2.1951 021 (m). -X. 14.1980. 10. 18592 7100 4 5 s(mm).22 rms.47 nrm 165 (y)-1980314 (m)- v(m"y 378 s(mm)- 50 1 0- -50 Il pdm- I I I I I I II I TiI1 .1981.443 nrms.~-~.24 nrms. 270390241 v(mnmy). 12.127 50 -50 1980 111111 'I -50- 1975 (y)-1980376 1(m).058 1(m).9947 11.soda l(y). 223 I 0 I I I I I I I I I 1 cooidge .0 93 -50 I 50 - oct0tl . 1 54 0- .30 -50 I. 1985 I 1990 -50 1975 1980 1985 year 1990 I .-cr--mFt. 322974337 v(mr-0 54mm).4873 v(mnWmW1.

puent.ayor_ v(mrnVy) 17.1981. 29.1 80 (m) ..30 s(mm) 46.795 I(m) 34819.0155 v(mmly). 17 50 t(y).prieto v( 50 1 I I 1 I I I I 1 i I 50 I I I I I I I i I .182 1m)= 28622.15.0720 7. t(y)=1981.1.prieto_ t(y)._.178 I(m). 1025 - 0 0 -50 To- -100 IL -50 -150 I (Irlrllllll 150 -. -12.78 nrms. 15538.1981.prieto 1(y)=1981.mayor_ t(y).24674 6545 4.76 s(mm). 26S67 nrms.05 nrms.73 s(mm).0413 v(mmly).178 I(m).67 s(mm).1981.: -200 -200 1970 1975 1980 year 1985 1990 1970 1975 1980 year 1985 1990 m . 24.796 I(m). 6.136 . 66.12.63 nrms. 4.03 s(mm).114.89 i - 0 0-50 -50 - -100 x -150 -100- -200 -150i| I I I lI I I I L I i I I I i I I -davidv(m ry 3 . 3.) -.01 s(mm).5.9679 v(mmly)u -1.90 nrms.1981.37 100 100 I -- 50 0 -50 -50 -100 -100 -150 -150 .4336 -19.-8. 6..183 I(m)= 14967.55 nrms. 1221 nrms.1981. -20.1584 -15. v(mmny).76 s(mm).572 I(m).1. 27651.62 50 - -"- 0- 0 -50 - -50 1 I 1 I 17 1 1 .48 s(mm).12 3- 150 3 l l t .46 24 t(y)1981. 9304._ t(y).79 24 50 | 50 - I I I I i I i Ii I i t(y).25 nrms.dav&id v(mmny).

8058 42.1981.1979.443 1(m)= 22869.8.37 I carrtisa.7979 1. 21.19 s(mm).dixle v(mm y)- -.803 I(m).03 . 50 150 I I I I I I I I I I I I I t(y)=1980.85 s(mm).22 s(mm).9760 -2.10 nrrms 8.33 nrms. 1. 11495.926 1(m). I l I I I I I 18384. 4157.03 nrms.2.36 nrms.19673.79 s(mm).33415. 14. v(mm/y).22.30 i 100 I 50 0 - ff_ x I-T 1 TL TZ -50 1970 -50 1975 1980 year 1985 1990 1970 I I 1975 1980 year 1985 1990 I .81 nrms. t(y). 1. 1.8938 -sup v(mm/y)n -10.99 50 - 100 50 r--~ b3.91 I I bnp10065 .3424 6.tEr~I-f----p-I 0- 0-50 1 11 1 1 1 carrisa.137 200 50 150 " I 0- t(y)-1981.80 nrms.80 -4.180 1(m).969 I(m). 9 t(y).8286 -7.up t(y). t(y)=1981.1980.4559 20.20886.37 nrms.15. 9 I I I I I I I I I 1 I I I I 50 0 0 -50 -50 I I alamo I I I I I I I I I I I I I 1 t(y)=1979.I I( 1 I -50 11 - 1 1 1 t(y)=1980.512 1(m).05 s(mm).prieto v(mmly).4. 2.444 I(m). v(mm/y).1665 0.52 -prleto v(mm/y)- 36 - 100 '-T -50- 50 -100- 0 -I -50 -150 I I I I I I I I I .72 s(mm).96 s(mm).70 s(mm).54 I I I i dixie 150 50 I I I I .sup v(mm/y). 7.34569.180 I(m).40 nrmns 1.fierro v(mmly).3.prieto.

.46 v(mm/y)= -2.1980.72 s(mm). . 25. I Il . 50 50 - 0 0- -50 - -50 - I I S 1 i II I I I I I I i i /"II i I i i I I t(y)..53 nrms.4..---. 1980 year .57 s(mm).1980.off_225_ t(y)-1982. kane i .21761.1. i I .219 I(m)m . 17138.fIsh dixie -. -100 -150 0 -200 -250 - -300 - -50 -350 .2352 nrms. l l | i I l I t(y)m1980.46 nrrs.6363 -2.93 I I 150 - I i II I I lI I l I I i I 1 t(y)=1980. . -50 50 -100 -150 -:__-__ I ii-200 -50 I I I I I I 50 x-3:x I 0 - I II II 1I I 1 1 I I I 1 I I I I I I I t(y)=1981.kane_ v(mm/y)- fish I I I I I I -150 -200 -250 -300 -50 -100 -150 -350 -4o0 -200 -450 l .41 5v34522 n .sup _ v(mmly).soda v(mm/y)..sup v(mm/y)= -6.61 nms.133 I(m).~ v(mryn) 50 0 -50 -100 - - 1 1 I I I t(y). 1985 . 1970 1975 1980 year 1985 1990 1970 .82 nrms 12.36 fish . 1975 .138 t(y)71981.3855 v(mmly).78 dixie 273349601 9.643 I(m).60.5323 v(mm/y). 3.off 229_ t(y)-1982. .09 dixie 150 - I. 14853.445 I(m).14.178 I(m).439 I(m)= 15686. l. 14.4204 .47 .94 nrmns 5.81 nrms.4534 -6. 5. 1990 .11816.50 s(mm).86 s(mm)= 23. . 89. .643 I(m).442 I(m).26276.3051 -.5.2a * ka_.14 s(mm). . _. 35.17170. i I I I I I I I I I fr 50 0 100 ------. .35 s(mm)= 56. -19.73 nrms. 28.84 soda 100- 50 T 50 - - r.31 s(mm). I |.

-150- .36 29.1. v(mmly).38 beacojo . 1990 1970 1975 1980 year 1985 1990 . . .13468.5.661 I(m). .Inpncer t(y)-1984. 1970 1975 .3446 v(mmly).17804.59 nrms. 6.6580 v(mmly).10 nrms.65 nrms.0.69 s(mm).37 s(mm). 14.09 s(mm).1982.03 II I -3.28 s(mm).edom_Jo t(y).89 s(mm).00 I I 11111 I 1 I I I II berdojo . 1 l 14048.valmtecc t(y). 1985 .36 nrms.6.. .queen_ t(y)-1984.90 s(mm).8224 v(mmy).1982.358 I(m)..4870 0.1. -1.87 s(mm). 1 i 16169.3.11 nrns. 50 SI I I II I beaco jo .32 nrms.97 150 - 50 - I I I I I t(y). .754 I(m)m 31697.7552 1.13.3.stubbe_ t(y).31 50 50 - I 0- 0- T -50 - -50 1 1 I I I II I i I I I I I I 1 I 50 - 0 0- -50 -50 I I I I I I I I I I I I I I I 1 beacoo .14919.392 I(m).1983. 50 - -2.1975. I 10050 - 0- -I-c 0-50 - -50 I I Ii i I I l I I I I berdoJo.57 50 - I I I l l i l l l l l l l creole_-maumejo t(y).139 29palms .dome.712 I(m). 16175. .03 s(mm). 1980 year .dome_eoc t(y).9445 v(mmly).701 (m).4137 v(mmly)m -1.072 I(m)- v(mm/y).0.644 I(m).1985. 8.1984.palms -.2. 13. v(mm/y). 11. 12861.00 nrms.82 nrns. .4406 45. 1.11 i--- 00- -50 o-loOL -100 -50 .2.

58 s(mm)._o-inspncer t(y)=1984.39 150 50 - I I 1 I I I I I I I I I I I I edom_Jo -. 4. 15. 7.22999.01 nrms.edomjo t(y)=1975.90 s(mm)= 73.warren_ t(y)=1984.48 s(mm)m 15.633 I(m)= 10770. 1.1984. 200 - I I I t(y).951 I(m)= 19481.6943 v(mmly). 52.071 I(m)= 10777.44 150 100- 0- 50 ---------- 0 -50 - I 7 -50 - 1970 . 0..381 I(m).69 nrms 2.92 s(mm).17.8573 64.70 nrms.24 s(mm)= 27.ecc -.03 creole 50 I dome -.90 nrms. 3.06 nrms.sandhill t(y)=1986.laquuJo t(y)=1982. I I I I 23619. 11.701 I(m).140 -.3010 v(mmly)u -4.36 50 0 0- -50 -100 -- -50 - -150 I I I I I I I I I ! i I ] I I I edom.2876 v(mmly). 11.675 I(m)= 20631.00 50 - 50 0 -50 -50 -50 I l I l I l l I I I I I I l SII I edomjo -.681 I(m).edomjo t(y)=1983.6364 v(mmly)= -7.703 I(m)= 27931.40 s(mm).48 50 - I I I I I I I I I I III Inspncer -.10.laquijo t(y)-1984.76 nrms.7168 v(mrm/y) -4.0976 v(mmly)= -------- _--- 0. 3.21 s(mm).18 nrms.21469.00 nrms 0.warren v(mnmy).10 dome.33 - 100 50 - 1-- - 0- 0-50 - -50 l i I I I I I 1 l lI I I I I I inspncer. 6.' ' I 1975 I 1980 year 1985 ' ' ' ' I 1990 1970 1975 1980 year 1985 1990 .80 s(mm). 0.5779 v(mmly).

64 keys.92 50 -.357 I(m).392 I(m)w 13838.queen_.0.36 s(mm)= 18. 1980 year .warren t(y). 5.93 nrms.-.358 I(m)- -.48 s(mm)= 3.5849 v(mm/y).queen 20.15 50 | I I keys.198 .0.14 nrms. t(y)1984..14218.1984.1.64 nrns..1984.1984.9.8784 v mm/y) -7.I-' rr_------If -50 - -50 I I I I I I I I I I I I I I 4 mesquite-valmtecc t(y).391 I(m)w 13583.7136 v(mmly)= -1.09 nrmsw 5.keys_ t(y). .22. I 1985 1990 .- 0- --.0557 v(mmry).63 nrms.sandhill t(y)-1982. 2. I I I I I I I II I I . v(mm/y)= 50 50 I--0- 0 -50 -50- - I I I I I 1 I 1 1 I 1 I 1 I I I I I I keys_ .141 inspncer. .91 s(mm).46 ' I I I I I I I I I I mesquite.05 . 20780.356 I(m)..89 50 50 - --.90 s(mm).94 nrmsn 4. 3.f I- 0- 0- 1 -50 -50I 1970 1975 1980 year 1985 1990 -. 20208.--- 1970 1975 .13 s(mm). -1. 0..0426 v(mmly).1984.75. 111111 I I 50 - 50 0 I I _t .------- 0 -50 -100 -150 -50- -200 I i l l I I I I I I I I I 1111 I I I I 1i I I I I 1 I I 1 11 mesquite .7645 v(mm/y)= -2.segundo_ t(y).74 s(mm).3582 t(y).74 nrm 14.mesquite t(y)-1984.355 I(m).1174 v(mmly).28 keys_ 9280.355 I(m)u 15712.895 I(m).87 s(mm)= 1325 nrms.98 s(mm)= 10.16306. 2.71 -0.

edom. bolt 18630. 1. bm603_39 -.- .2.41 nrms= 0.833 I(m).142 -.3536 -0. 8. I I .73 nrms.view t(y). 6.3. 20583.0709 1. 2 v(mm/y)- I 1 50 - 50 - T I ! 0- I 0- -50 - -50 I I I I I I I I I edom_2_ .1305 9.56 s(mm). I I I I I I I I I I 50 - 0 0 -50 -50 1970 .3185 v(mrmy)m 3. 0.88 t(y)=1987.41 50 50 -I 0 -50 -50 S I I I I i 1 I I I I I 1 I SI I I I I I I I i I I I I I I -. 19798.view v(mm/y). 1975 . 1970 1975 1980 year 1985 1990 I .89 50 -a I .16 nrms.9169 -.18181.4.47 edom_2 4 50 0 0-T e -50 -50 II I I I I I II II I I I I I I I S11111 I 1111 i I I I I I 4 4 t(y)=198 .48 s(mm). 0.26 I(m).12 beacon_2 -.767 I(m)..210 I(m).45 gap t(y)-1984.4613 -3.24 v(mmly)V 1. 1.21 s(mm).12365. 5.889 I(m).view 10. 0.31 s(mm). I I 1980 year 1985 1990 I I I I I I I I I I I I I I I I I edom 2_-.15 s(mm)= t(y).191 I(m).31 s(mm).28 nrms 1.1985.18538.55 nrms.00 nrms.resort v(mm/y).00 beatrm3 .1985.edom2 v(mm/y)- t(y)=1986. 7.72 s(mm). 18176.979 I(m)= 23386.73 nrmns 1.2465 v(mm/y)= -1.tram t(y)=1988.146 I(m).36 s(mm).7869 t(y)=1982.22 nrms.view v(mm/y). 4.

28034. queen I I I I I 1 1(m)= 21863.8717 -7.1985..01 s(mm)...1984.391 v(mrmy). v(mm/y).3119 . 102.866 I(m).28827..61 50 - I .4302 v(mmly)= -1.1984.181 I(m)..1984.1026 3.78 nrms.05 s(mm).392 I(m)..624 I(m)m 13715. -6.. - 0 - -50 - -100- -50 -150 - 1970 1975 1980 year 1985 1990 1970 1975 1980 year 1985 1990 . I I I I I I I I I I I I I I I I I segundo_-valmtec t(y).61 s(mm)...49 nrnm 6J2 --. 0.sandhill t(y)=1984.2184 5.4946 v(mmly)= 10. -1. -i--I .86 s(mm)...392 I(m)- -2.1982. 17537..15245.sandhill t(y). t(y)-1984.96.05 s(mm)= 17.05 nrms= 6. 14735.372 I(m).88 50 250 1 I1 I -- 0 200 -50 150 100 -100 50 -150 - 0 ----------------- -50 -200 I 50- sandhill -.1694 47. 9.:i - --.143 meeks pax_ncer. T 0- I----------- --- ...07 I I I queen 50 I v I I I II I III I I .90 -.23 19548.390 I(m).4421 v(mmly).rich v(mmly)= 50 50- 0- S0- -50 - -50 - I I I I I I I I I I -.43 nrms= 20.4.sandhill t(y).valmtecc t(y). -150 -200 -250 -50 - t -300 I I I I I I rich_ 300 -100 I i I I I I I I 1i 1I I i I .segundo v(mmly).1.36 nrms..71 nrms.90 s(mm)= 28. 14.59 t(y).79 nrms.78 s(mm). 93.18 s(mm)= 6.- .1984.24 1 I1 1 1 I I 1 I I 1 I I I I I smndhill -valmtecc t(y). 50 0.69 nrmsm 1.-- I 0 50 - -50 i - T .

1975 . - -- -- .dch__ t(y)=1985.61 v( 72.04 nrms 58.866 J(M.53 . .____~_.91 nrms-180. - .1597 -22.pax_ncer t(y)=1985.17249:9-186 v(mmly)--120.682. 1980 year .rm4 42.76 200 I I I I I I I i I I I 1 I 7 1 1 v(mmny). v(mmly)= -40.3068 35.30 s(mm).626 I(m)- )-20-23.241 I(m)... 1990 -600 0 197( " ' I 1975 ' ' ' ' I ' ' 1980 year I r 1985 1990 I I . -600 -800 -1000 800 1I -1000 -1200 -1400 0 I I I I I I I -.78 s(mm).rich-.1984.60 nrmns-272.._ 245P2710 0 .meeks v(mm/y)- 200 I I I I I I I I I r 23118.pax_ncer t(y)-1983. -200 200- pax_ncer.43 s( 1244. - - - .616 I(m)= 14309. - --- .621 I(m)= I I I 23999. 11002. .35 s(mm)= 166.__ _ _ _ _ .15 0 -200 -400 -600 -200 -8 0 0 -1000 -1200 -1400 d -1600 -1800 -2000 -2200 -2400 -2600 -2800 400 -400 I I I I I I I I I I l I I I maumejo .231 m) 381.19958.80 keys I I1 I 0 II I II II -200 -400 -600 -800 -1000 I -400 I -1200 -1400 600 r I SI I maummneo I I I I I I I I 1985. ..353 I(m). I -600 -- 200 i i i I I I I I I i I i I I i I -= maumejo -1984. -4. . 1985 - . 61.82 nrms.18 s(mm). .58 nrms.218.0766 160014001200 1000 pax_ncer.87 0 - -200 600400- 1970 I 400 1800 - 0-200 -400 I I v(mm/y) 600 200 I meek 0 -200 -400-600 -800 - 800 I -400 I I.50 400 200 .79 -63 -.- .6424 v(mmly).00 nrms 66. 21.rich creole t v(mmry) 200 0 -200 1986. 749. . . 200 2366.30 nfftis-.-.1982 .952 I(m).warren t(y). -17.~___ ~ 144 400 -. 349.3152 t(y) 1985.31. .624 -1(m)-.

114 . personal communication. Wernicke. All the tectonic features and plotting conventions are the same as in Figure 1. The gray sites at the southeast corner are part of the Salton Trough and Riverside County (STRC) network (R. Most of the new GPS sites are from SCEC and Caltrans GPS experiments. The gray triangles represent the recent GPS sites. personal communication. Map of southern California showing the VLBI/GPS (VG) stations (triangles) and USGS trilateration network (linked with solid lines). . 1993. and those in the northeast corner are part of the Death Valley network (B. A.FIBR 35N - "LOSA AA SBC 335N "0 MC M km SOL A A BU 101 AMON AAk 32"N 122W km &MN So 121Vw 120W 119'W 118W 117W l 116W 115w Figure 17. 1993). Reilinger.145 37 0N -A 36"N .

California Department of Water Resources e.2 arc seconds Snay [1990] 4-th order direction 1-st order azimuth 3. National Geodetic Survey c.4mm+Oppb east reference up 6.0 arc seconds Snay [1990] 1..2 ppm 10 mm + 1.146 Table 1 : accuracy reference 1-st order direction 0. VLBI 1984-1991 7..1993 .. Electronic Distance Measurement b. There are offsets between CDWR and CDMG measurements due to the changes of instrument and the procedures for correcting atmospheric refraction effects [King et al.4mr- -12ppb Feigu~ at.0 ppm 6 mm + 0.7 arc seconds Snay [1990] 3-rd order direction 1. [1987] Savage. Table 2: data GPS span north 1986-1992 3. Hewlett-Packard g.0 ppm Snay et al.5mm+2 ppb 38. [1987] Lisowski [personal communication.1mm-1 ppb 5. 1987]. United States Geological Survey d.0 ppm 3 mm + 0. Prescott [1973] King et al.9mm+15ppb 21. 1993] a. California Division of Mines and Geology f.6 arc seconds Snay [1990] 2-nd order direction 0. [1987] King et al.7 ppm 3 mm + 2.4 arc seconds 10 mm + 1.1993 ppb Feigl et al.6L.0 ppm Snay [1990] Snay [1990] span measurement Taped distances EDMa (NGSb) EDM (USGSc) 1972-1992 EDM (CDWRd) g 1959-1969 EDM (CDMGe) 1970-1979 HP' (USGS) 15 mm + 1.

77835 N35 1 59.1156 4 .0290 1983.7462 -0 .6105 0.76932 2 2.0917 0.33954 N35 N34 448 13.00040 1983.5042 4.2006 0.08161 W118 48 56.3129 4.4292 3.022 6 0.0959 1.92406 W119 1.0222 0.0000 0.11203 W119 2 2.000 0.0000 1983.634 7 0.0210C 0.04428 N34 49 N34 37 52.70619 N34 24 35.6792 4.020 1 0.73038 W118 1.2978 4.9938 909.000 0 1983.2002 4 .000 .19671 14.000 0.72961 W118 588 8.0182 0.000 1789.1188 0.72927 N34 2 2 51.0206 1.5781 4.0204 0.0152 0.000 0 1992.0148 0.63994 W119 2.3789 0.8389 -c 4.9312 1225.36311 W117 56 18.0244 0.1225 3 . * site apache frazrm2 frazier kitchen_ madulce2 monte mtpinos nordhoff paa-1 patti_lo full-name lospa net lospa net lospa net lospa net lospa net lospa net lospa net lospa net lospa net lospa net pelatlo lospanet pinosecc lospa net poison_ lospanet lospa net police lospanet reyes santapau lospa net lospanet seco sulpher_ lospa net sulphrm2 lospa net tecuy_lo lospa net tejon_41 lospanet temblo lospa net tembl lo lospanet lospa net thorn wh2 r lo lospanet lo lospanet yam tehac net avenue tehac net bald black bu tehac net tehac net bull tehac net denis diorite tehac net doub rml tehac-net fairmont tehac net gneiss_ tehac net tehac net leona lit rock tehac net pa]uela_ tehac net pelona_ tehac_net tehacnet police rlt rid tehac net saw ecc tehac net sawmill tehac net soledad tehac net tecuyrmi tehac net te]32rm2 tehac_net te]on_32 tehac_net te]on 41 tehac_net thumb tehac net tehac net tom tom rm3 tehac net weed rml tehac net wheeler2 tehac net alderaux sanga net ant_aux_ sanganet badpow_ sanga net sanga_net bull sanga net camp_9 dispoint sanganet sanga_net duqo sanga_net flint gln_ncer sanganet sanqa_net handle hau eccl sanqa net sanqa_net hauser jpllrml sanqa_net litte]on sanga net may__ sanganet mtqleasn sanga net nbaldaux sanga net pacunk_ sanqa net paclncer sanga net pac2ncer sanga net pacifico sanga net parkaux2 sanganet park_rm3 sanga_net parker_ sanga_net pelon_sa sanqa_net plconcer sanqanet sanqa_net port lonc tude latitude N34 45 25.27059 W118 4 8 9.9937 1958.29996 W118 1 6 36.0182 0.0.08161 W118 4 8 56.0241 0 .000 0.47653 W118 8 36.79341 N34 26 25.0786 1497.9936 0.000 0 1983.7781 4.4345 .0201 0.3203 0.2.4178 -40.0192 0.0.0184 0.48607 8 43.28572 W118 49 20.4833 0.775 2 0.0290 1983.03093 N34 36 23.2923 .81899 N34 54 10.1099 3 .0000 0.000 0.327 5 0.95584 W118 29 12.86752 W119 2 5 13.98116 N34 1 2 17.0000 1992.0000 1983.3477 1446.4207 4.1392 0.95040 W118 58 53.01883 W119 1 6 53.28931 N34 56 4.000 2137.3239 -0.0000 1983.0196 0.45801 N34 4 8 2.6484 1450.00002 1983.13934 W118 2 5 46.0223 0.69263 W119 2 6 29.000 0 1983.022 6 0.1774 0.0000 0.5010 4.4510 16.000 0 1983.6176 18.0000 1983.2 51.61230 W119 1~9 0.9933 797.0000 0.2398 4.0294 0.0005 -( 0.0000 1983.0214 0.000 6. 6 23.000 10.0208 0.8605 0 .000 0.1915 4.0282 686.3305 -0.0305 0.021:3 0.9788 4.4488 -0.0315 0 .0000 1983.5950 16.44537 W118 3.5600 13.000 0.000 0.29922 N34 2:1 30.021:5 0.84206 W119 1 2 14.0000 1983.000 0.000 0 1983.0260 0.0303 0 4.9456 0 .9532 2137.67636 N34 22 58.0000 1983. 9?30 291.7793 -0.7561 0.3022 -0.0083 0.1089 0.2672 3.1000 1052.1546 0.0175 9.141 3 14.0000 0.01770 W119 7 39.000 0.000 0 1983.9928 1225.000 1245.4339 4.000.1546 2.0223 0.8944 3.0302 797.4721 3.1939 0. Su = up.0256 0.4957 2076.1912 0.9908 0.0000 1983.0000 1983.0000 0 1983.78824 N34 24 35.0000 0.000 13.619 4 1.021 5 0.0000 1983.147 Table 3: USGS EDM site used in this analysis * frame: WGS84 unit: m/year * u = V(east) v = V(north) w = V(up).000 0 1983.0196 0.43244 0.3605 -0.2006 0.93964 W117 4 5 52.9109 12.89344 W118 54 8.2259 0.76401 W119 N34 41 27.1000 2403.0238 0.000 6.0282 0.0998 1446.0705 0.5719 1441.7793 .0000 1983.1530 0.000 1648.25343 N34 411 35.2185 0.0000 0.30178 N34 46 30.43015 N34 55 16.0197 0.02141 0.0000 1983.0000 0.5959 4.0231 0.000 0.0100 1476.3935 .9927 0.36705 unit: meter v w u epoch Sn S' Se height 0.1150 .6098 18.000 0 1983.00004 1983.0226 1789.1118 4.72667 N34 37 31.0c30 1169.4738 1546.64696 N34 227 35.0201 0.0133 0.2933 -0.3607 -04.0991 932.000 0.8394 .53176 N34 29 53.9937 74.0245 0.04194 W118 11 19.0000 1983.000 0.6040 4.0160 0.0089 0.0291 0.000 10.0000 1983.8372 4.000C 0 1983.0000 1983.462 9 0.000 29.0000 1983.8245 2260.0206 0.000(0 1983.64678 W117 4 0 30.4235 -0.025 3 0.000 0 1983.87700 W118 2 4.0144 0.76137 N35 1 3 22.0177 0.0179 0.0.0822 2.0000 00.000 0.4956 44.9933 4.0136 0.69928 W118 17 40.286 8 0.1316 04.078 9 64.6805 0.0198 0.0000 0.91141 N34 3 2 51.0000 1983.000 19.42802 W119 3 3 17.43419 0 51.9705 1759.51892 W118 21 21.0 1983.1106 0.000 10.33391 W118 33 41.6130 .62599 N34 14 47.0224 0 .0140 0.30280 W118 36 54.6153 18.323 4 0.78244 7 25.0.0412 18.5719 1441.000 619.3611 -0.9402 0.000 0 1983. N34 22 20.0202 0.0742 -0.02261 W119 3 3 18.79210 W117 5 3 55.0231 0.0624 4.000 0.9937 4.0.00040 1983.0917 2.7215 807.3791 .000 5.000 0.17056 W119 4 31.4244 0.1349 0.81899 N34 119 50.43958 N34 57 35.95040 W118 58 53.0231 1983.0000 1983.000 0.9002 4.9588 2141.9938 1705.0000 0.611 8 0.000 0.000 0.0201 0.1349 0.9935 0.7605 -0.0182 0.4595 2.0199 0.4605 0.4679 -( 0.021 6 0.0000 0.1679 -0.9335 1646.0214 0.26642 W118 19 27.9937 991.0000 1983.0234 0.000 0 1983.30220 5 56.000 0.55058 W118 2 5 5.000 0 1983.0000 0.000 0 1983.0117 0.0.06896 W118 1 0 15.0226 0.73482 W117 4 5 52.00040 1983.900 1f 8.9937 2137.1560 0.81361 N34 13 39.0000 1983.6790 -0.0000 04.0234 0.84526 N34 223 10.000 0 1983.9928 1224.7286 1190.9545 -4 0.77896 N34 2 2 54.45679 5 42.1000 -0.01569 N35 6 45.4478 -0.84838 6.4488 4.0183 0.6904 .8970 -0.0111 0.07655 W118 233 12.000 1.55053 1 5.2910 -0.000 0.0000 0.1246 0.01611 0.9938 .05922 6 18.0297 0.5981 -C0.0000 1983.0198 0.0190 0.5370 19.4519 4.000 1I 983.0005 -0.000 0 1983.667 8 0.0186 840.87044 N34 22 54.7173 -C0.0939 0.01317 1983.81593 W117 4 5 45.000 10.000 542.85883 W118 12 52.0182 0.000 0.4649 2144.86265 W118 1 9 47.000 0.0219 0.16366 N34 37 51.000 19.108 9 0.0202 0.1231 0.51573 W119 3:5 28.5692 10.29782 W118 N34 3.1204 0.000 0.1466 4 .0101 0.0833 3.0135 0.7732 -0.0955 0.000 2831.0215 0.0926 0.0000 0. 4 19.000 0.87300 0.4144 17.000 0 1983.0.0175 0.0000 1983.58283 N34 2 3 12.8502 1792.000 1.165 1 0.000 0 1983.0449 0 .63516 N34 2 1: 8.66270 N34 227 35.0000 1983.000 0.075 7 18.4114 3.39881 N34 44 56.89266 N34 48 43.0177 0.2391 0.37498 W118 5 5 56.075 6 18.00040 1983.000 0.3552 -0.83236 9 4.4159 1895.4545 44.000 0 1983.0148 0.0217 0 .9937 0 .84302 W118 1 1 48.6612 -0.C297 0.000 0 1983.92070 W118 54 16.020 1 0.49014 W118 N34 1 4 54.000 0.93049 N34 411 35.18533 W118 1 2 55.0000 2410.08149 N34 548 57.44161 N34 45 14.0280 -0.82555 N35 7 14.0000 0.1605 -0.00000 1983.1965 0.6812 4.2177 4.6983 -0.0000 1983.0100 2662.1534 0.48728 W119 0 51.2076 -0.000 119.0252 0.1939 0.0000 2003.62711 W119 N34 24 35.5082 0.3673 0.0224 0.9938 757.020:1 0.0136 0.72622 W118 2 5 8.000 19.2319 3.3405 -0.9933 797.0000 1983.079 9 0.5622 -0.1912 0.4005 0.66490 N34 57 33.0165 0.000 0.0742 .2219 0.0162 0.0000 1983.2087 4.000 1765.0.0.01351 W118 1 3 7.56892 N34 2 2 58.51892 W118 2 1 21.000 0 1983.0.6605 0.0000 0.000 0.9904 -C0.82011 N35 7 25.9585 -0.74730 N35 0 38.83852 N34 50 32.0200 0.3492 2.9:33 1101.1095 0.5605 0.9971 2042.000 0.4393 960.4241 2830.16366 N34 346 23.83701 N34 540 32.6101 0.1349 0.3279 0.8919 2.1974 0.1000 0.1405 .0000 0.7883 -0.000 0.3545 0.9426 4.7848 0.0162 0.0200 0.0196 0.0224 0.3609 0.53902 N34 46 29.9921 1547.57132 8 54.3808 1181.0162 0.04242 W118 1 3 7.0000 1983.000 0.29141 W118 2 4.5583 13.1121 -0 .000 0.0542 -0.000 0.0000 0.3679 0.62243 N35 16 20.31202 W118 42 11.0.658 8 0.0180 C0.00000 1983.9937 441.8279 0.0181 0.079 9 0.1821 -0.151 2 0.0181 0.75213 W119 248 0.4478 -(0.021 5 0.0167 0.0000 1983.0219 0. Se = east.91081 N34 4 9 6.3935 -0.124 1 0.41672 W118 1L3 12.31278 N35 6 46.900 13.0000 1983.0000 0.4225 1050.000 0 1983. * uncertainty of coordinates: Sn = north.9912 898.9350 4.0833 3.04428 134 2 1 11.9936 624.000 0 1983.0000 0.0000 1983.0000 0.35890 W118 33 41.2977 4.26217 W119 0 34.0000 1983.1::0 293.4324 0.0198 0 .0202 0.00040 1983.7279 4 .84604 N34 46 36.0162 0.21521 W118 1 3 7.0208 1.483 7 2.04845 N34 32 21.77672 W118 24 26.9933 1412.2018 1642.87108 W118 N34 31 42.9001 1.8605 -0.000C 0.0214 0.0822 2.6644 .0187 .0238 0.5494 -0.38772 N34 38 39.71126 N34 2:3 12.66125 N34 2 1 29.6 4.0175 0.1312 -0.07655 W118 3 3 12.3?92 0.44595 W119 1 2 13.79688 W118 1 2 55.3789 840.328 9 0.323 5 0.1135 1 0.000 0.012 3 1983.0213 0.000 0.0202 9.0000 0.3600 0.589 6 0.1149 0.68421 N34 3'7 51.000 0.4579 2.92070 W118 5 4 16.2861 -0.4371 4.55435 W118 1 1 5.66608 W119 28 30.000 0 1983.0223 0.0196 0.0000 0.000 16.0202 0.000 0.45406 N34 3 1 39.000 6.0241 0 .1216 1334.000 0.880 1451.0236 0.3941 1896.0000 0. N34 55 12.000 2140.4381 4.000 7.51791 N34 9 48.4205 4.7558 1332.1097 1.000 0.60739 W118 N34 12 55.0?3999 2144.0242 0.0.05163 N34 3 7 52.46355 W118 N34 227 35.1161 0.0000 0.000 7.0000 1983.04883 W119 2.000 0.0009 0.8125 -( 0.9525 4.021C 0.31056 W118 2 4.0143 0.9901 0.78244 N34 48 13.021 5 0.0000 1983.6498 0.2672 3.0647 16.1089 0.5188 -0.611:3 0.4718 3.0187 0.000 960.0998 0.01569 N34 5:1 45.0215 0.0000 1983.0290 0 .6211 0.7732 -0.6434 1450.0097 0.0204 0.49925 N35 0 38.1761 0.017:3 0.000 1446.0607 0.03838 N34 1 4 49.0292 -0.0186 0.0330 0.0222 00.4115 3.000 119.9936 896.000 0.0205 0.668 3 0.0000 1983.0000 0.1000 0.000 1! 1523.0133 0.47847 N34 48 46.09952 N34 54 10.4598 9.23434 W117 5 3 54.61772 W119 12 13.CCOO 619.62617 W118 49 21.9932 2655.0201 C0.000 0.6684 16.53673 N34 30 48.4394 -0.6830 -0.1000 2411.000 0.1730 0.00040 1983.

0049 1065.85269 3.9426 4.1961 4.05892 W119 4 N35 10 56.1497 3.0198 166.0118 393.1179 0.4942 4.0247 1070.2218 0.0159 -0.5505 0.000 0 1983.22760 1 32.4178 3.0304 0.9937 0.32136 W118 3 N34 41 35.0187 1291.9937 0.0100 0.0242 0.1354 -0.1396 -0.36093 W120 : N35 33 6.000 0.56551 5 45.4792 -0.1840 0.3105 0.000 0.76649 14.0288 0.0000 1993.72126 W120 N35 55 12.2859 ".1656 0.000 5.78050 W120 N35 36 17.0.5961 0.2700 0.02261 W119 N35 23 8.93662 W119 N35 53 17.73621 W120 N35 23 33.3513 4.1375 e) 1595 0.2504 1.000 0.4669 0.08083 52.0000 1993.1681 0.1960 0.60180 i 5.3325 4.4679 -0.3811 0 8730 0.9504 0.0000 1983.91351 37.82772 I 19.0351 0.000 0.0185 0.000 0.3407 0.0201 708.0000 0.60360 W120 I N35 7 31.0084 1050.9599 -0.8122 --0.000C0 1993.000 0.2199 0.3405 -0.0989 0.2627 0.70575 W120 N35 53 16.73209 W120 N36 0 2.8253 --0.0333 0.0000 4.0674 0.000 0.0000 0.3147 0.0143 1052.0000 1993.0268 0.3333 -0.000 0.65193 W119 4 N35 26 28.01730 5 1.0000 1993.6243 4.1324 0.0242 0.0212 0.2650 0.99012 W120 N35 56 56.000 0.4303 0.0989 0.0226 0.0000 1983.000 0.2357 974.0000 1983.61473 i 27.93962 W120 N35 21 5.80267 W120 N35 49 50.9937 4.0224 0.4393 2.4384 0.0133 471.9937 4.000 0.0264 0.9935 0.2697 -0.2419 --0.0283 0.000 0 1983.2258 4.000 0 1983.47182 1439.3405 -0.0000 1983.3905 0.32600 W120 N35 21 29.0121 513.1307 0.32084 W120 : N35 21 31..37258 29.0270 -0.35894 W120 N35 21 30.62846 W120 N35 41 9.4907 4.0226 1164.94246 15.0231 0.06793 47.0065 1983.2326 0.7059 -0.0195 951.40613 4.0104 12.0244 1173.0000 1993.6411 0.000 0.0144 -0.2 1 9 d 4 -928 4.1385 0.0336 0.3114 -0.0071 1993.0252 -0.0192 789.0000 1983.75507 W120 N35 45 20.0286 0.0332 0.84484 W120 N35 47 15.0244 0.5701 .9929 0.0187 668.000 1993.9573 2.4363 -0.0307 0.58428 W120 N35 58 9.5460 .0049 1169.0000 1993.3690 4.0000 1993.01843 3.00000 1983.0.0000 1993.0000 1983.1467 0.1080 -0.0303 0.0330 0.000 1983.0219 1070.0223 0.000(0 1983.000 0 1983.0000 0.000 0.8520 -0.000 1983.1911 0.1080 -0.3610 0.0000 1983.0000 1983.0236 252.1861 1983.46411 W119 ' N35 21 31.0162 518.9932 4.0000 985.000 0.0228 0.2348 -0.0000 1983.2123 0.000 1983.1352 0.0214 1164.000 0 1993.4994 0.0310 0.000 0 1983.0000 1993.2179 -0.13828 F 33.0000 0.23451 41.0304 0.0331 0.21537 W117 5 N34 37 51.0000 1993.59581 W118 1 N34 41 35.0224 0.0183 958.81628 L 14.000 0 1983.2131 0.0293 0.8587 -0.000 0.4827 0.8242 0.3404 0.0320 0.0000 1993.9005 -0.23361 W120 N35 4 32.0247 762.0228 0.0192 1522.3112 0.17236 W119 ! N35 4 22.6341 6.19469 53.53516 4.63994 W119 N35 53 17.0203 735.2277 0.4826 -0.0000 1983.0239 556.78211 W119 4 N35 8 0.000 0.0380 0.0252 472.0071 1993.42802 W119 N35 6 46.000 0 1983.08083 50.31278 3 18.0250 1287.0099 751.4841 .69809 W120 N35 40 26.000 1.000 0.5056 2.1392 0.9500 9.99- 9.9936 4.5017 0.9573 0.000 0.0100 4.000 0.87968 W120 N35 17 26.10078 W120 N35 23 33.0192 1171.1904 0.3138 0.4921 -0.33520 W118 1 N34 30 0.00000 1993.0319 0.0198 1523.87064 W120 N35 55 29.0356 0.0000 0.0231 0.5289 0.43639 41.0000 1983.0312 0.0467 0.1896 0.0000 1983.0273 1522.24985 3.9904 4.1369 0.96787 W120 N35 2 10.6130 -0. N35 27 20.0355 0.43015 24.0182 0.0138 -0.3167 0.0200 0.76401 W119 N35 26 27.05163 6.4961 4.4476 0.000 0 1983.6937 -0.1432 0.0027 587.1404 0.9938 4.2547 1.0111 -0.7228 0.19710 25.2469 0.000 13.4017 0.000 0.03093 L 51.0233 724.0327 0.148 port_rml sanga_net saw esa sanga_net sawmi sa sanga net siselsle sanga_net sanga_net tank sa sanga net tom carri net allen carri net berros biddle carri net calle ca carri net carrizo carri net carrl net cauvel crock 88 carri net crocker carri net carr _net glo 13 carrinet gould carrl net hath 2 kittncer carri net madre 80 carrl net madreecc carrl net mcktt ca carri net painted_ carri_net saltos carri net stanley2 carr _net sycamore carr _net carr _net turkey welport_ carr _net airway_ sanlunet sanlu net almond sanlu net barren bench sanlu net blackhll sanlu net blhllres sanlu net sanlu net bonnie sanlu net break buck ecc sanlu net buckhill sanlu net caliente sanlu net sanlu net car car rml sanlu net car rm2 sanlu net castle .000 0.1637 0.12835 23.1396 0.000 16.0695 4.5007 --0.000 0.000 19.07093 W119 5 N35 21 28.77911 19.77188 W120 N35 6 45.1477 0.2157 0.36000 W118 3 N34 16 8.1844 0.000 0.0000 0.800 0.1000 0.000 0 1983.37258 2 29.000 0.8944 0.88854 W120 N35 3 37.000 18.2046 516.0143 797.7803 0.2324 0.00000 1993.0000 1983.79463 8 36.0524 -0.3383 0.000 0.000 0.0346 0.65634 50.0000 1983.0000 0.000 18.9622 2.2538 0.2821 -0.6353 0.2175 0.24268 W120 N35 17 26.0182 -0.03502 7.000 1983.0225 561.72277 28.0000 1993.0233 0.5605 0.0200 0.3411 0.74809 I 49.0.3954 -0.0000 0.1034 -0.0000 1983.1578 0.0287 0.000 0.0000 1993.000 0.1435 0.000 0 1983.1544 0.3295 0.22983 7 25.9937 4.000 0.2605 0.45064 20.0212 0.8070 4.2790 3 7440 0.2836 0.000 0.6663 4.4238 -0.67382 W119 4 N35 4 32.0319 0.0191 0.0267 166.000 0.0000 1070.29999 W120 N35 41 9.0000 1993.0000 751.27410 54.1946 0.sanlunet chi2 nol sanlu net chiches2 sanlu net cholame sanlu net cobra sanlu net sanlu net cotton sanlu net davis flattop_ sanlu_net sanlu_net gastro gifford sanlu_net sanlunet gold sanlu net hatch hatchrml sanlu net sanlu net hearat hopprm3 sanlu_net sanlu_net hopper kenger_ sanlu_net sanlu net kerr kitchen sanlunet lyeguas sanlu_net mason sanlu net mcdonrm2 sanlu net mckttrck sanlu net mel ecc sanlu net melville sanlu net mid e sa sanlu net sanlu net mid f mine mt sanlu net mine rm2 sanlu net sanlu net mth__ montosa sanlu net moose sanlu net napoleon sanlu_net sanlu_net park sanlunet pelato sanlu_net pfl sanlunet pltt prospect sanlu net ranchito sanlu net red hill sanlu net sanlu net shade sammler sanlu_net smithecc sanlu net table sanlu net temb rml sanlu net temblor sanlunet tess__ sanlu net N34 23 11.3400 0.0306 0.90664 14.3641 0.0.7580 960.44325 W120 N35 56 58.0000 1993.51765 W120 N35 53 15.57241 W120 N35 30 12.98076 50.2627 0.1204 0.0000 1983.1341 0.01459 3 17.8351 0.0226 -0.0409 0.08153 20.81038 W120 N35 58 9.09124 37.000 18.71435 7.04544 4.1466 0.000 0.0285 758.4292 0.2350 0.9345 -0.7079 0.000 0.6301 -0.85516 i 5.3015 0.1345 0.0065 1983.0144 807.69911 W120 47.2497 0.000C0 1983.0349 0.3153 0.000 0.79608 16.0262 1285.3516 -0.9934 0.0000 0.8605 0.0319 0.0088 1334.0219 0.2105 -0.000 0.0126 854.40070 0.6375 .70466 54.000 0.74334 55.000 0.5369 3.0.1692 0.000 0.0000 0.9937 0.0036 1007.000 18.0162 749.0000 1933.7221 -0.0288 0.0269 479.2969 0.59607 7.0327 0.000 0.1218 3.2529 0.1118 4.0242 -0.70144 55.9022 -0.0231 0.0183 517.0000 1993.1402 0.3886 -0.69290 W119 4 N35 19 30.0162 0..0380 0.4533 0.000 0.21852 W120 N35 40 26.0234 0.81932 53.3395 1.7205 -0.000 19.0000 1983.0403 0.90678 W120 N35 13 58.0242 0.2069 0.1178 0.000 0.000 0.17290 W120 1 N35 17 27.192 0.59321 W120 N35 54 49.0027 838.0216 632.0000 1993.1499 0.8290 0.71263 W120 N34 55 16.0278 0.2123 0.0278 525.000 0.000 0 1983.2034 0.4342 4.0161 732.88046 11.0421 0.9917 0.8614 -0.0195 578.0187 479.7917 2.5957 -0.76853 20.1344 0.36402 53.0230 0.4921 14.000 0.000 0.9723 0.46123 W120 N35 45 57.4990 0.0000 0.16836 29.000 0.000 0.0000 1983.0000 0.36477 W119 3 N35 24 49.3563 -0.000 0 1983.42821 W120 N35 56 10.0200 0.0306 0.0221 0.000 0.0000 1983.80225 17.1257 -0.1462 0.0214 0.4178 0.0797 .79433 W120 N36 4 43.7595 --0.89134 W120 N35 55 29.37832 5 50.4081 0.25871 ) 53.0615 -0.000 0.0000 786.000 0.000 0.000 0.0137 1287.1018 0.0192 93.6750 .0197 534.1355 0.0784 0.93662 W119 4 N35 15 0.91211 W120 N35 8 57.0292 1115.1675 -0.55765 W119 N35 53 43.000 0.5180 .0536 0.0161 371.0319 0.3150 0.1186 -0.98992 W119 5 N35 14 31.000 0.17848 39.2969 3.0222 0.0264 756.0334 0.3736 0.0864 0.2433 0.9924 3.9491 -0.1684 2.41121 W120 N35 50 6.54265 W120 2 N35 9 40.0306 0.6905 -0.0524 .9873 0.68514 14.95060 W120 N35 56 20.3267 6.66672 1.79210 W117 5 N35 11 6.0234 1647.0390 0.1206 7.9913 4.0225 479.9292 0.07629 49.000 0 1983.0234 0.6605 1515.4999 4.02339 12.0359 0.0000 4.06149 W120 N35 55 28.0216 0.94614 47.36975 W120 .52770 W119 N35 52 12.0000 1983.0339 471.1809 0.0000 1983.000 0 1983.90522 W119 N35 49 57.5579 --0.1435 0.0242 0.1808 0.3012 0.0095 562.0000 0.8440 1.00000 1992. N35 53 18.0000 1983.000 0 1983.89316 W119 ' N35 15 40.30419 28.7113 -0.3267 0.0192 1172.1427 0.7372 2.2039 0.45222 W120 2 N35 2 10.83391 W120 4 N35 52 39.5408 -0.000 0.0000 1983.3444 -0.1766 0.9126 .4267 2.15735 W120 N35 48 56.2747 0.0251 0.1347 2.22538 54.83441 L 29.0238 825.1921 0.000 0 17i7.2397 0.3205 -0.1731 0.0183 584.0278 487.9869 -0.73120 i 41.2892 0.83476 44.00000 1983.0194 793.000 0.2045 4.0203 0.000 0.45438 W120 N35 44 44.2069 0.000 0.0000 1983.2598 0.32111 W120 N35 58 16.000 0.0000 1993.9919 0.9418 4.0000 1993.000 0.0262 838.000 14.78806 W120 N35 52 12.99094 * 27.000 0 1983.000 0.000 1983.6105 0.1132 0.21136 58.82784 W120 N36 1 46.1761 0.1708 0.5891 -0.0301 0.395 1702 .0301 0.84919 W119 4 N35 14 31.6408 0.0242 0.0241 0.8680 -0.0267 856.000 0.000 0.000 0.0000 1993.3021 40.1355 0.0308 0.0417 0.0129 732.0493 0.0160 661.1664 0.9938 0.0000 711.11-0.3239 0.0325 0.2315 0.4355 0.0312 0.2771 0.0219 0.37811 534.4500 9.13719 0 56.3530 -0.8919 2.0089 552.0.39129 2 29.9502 0.53796 I 21.1309 0.-0.2814 0.0211 706.0.30440 W120 N35 54 0.97654 W120 N34 55 12.0200 0.4225 0.1534 4.0203 0.2923 -0.49087 W120 N36 4 43.9574 0.000 1983.57132 5 49.0254 617.1806 0.9932 3.5015 -0.85461 41.0000 0.58404 40.1000 3.1299 0.0199 701.000 0.52770 W119 4 N35 9 12.000 0.51299 8.0000 0.6562 6.99956 25.9614 0 8 0.0108 3.000 0.000 1983.1844 0.35476 5 45.93330 5 38.69651 W120 N35 28 20.2878 0.000 0 1983.0181 1069.99317 1.1144 -0.0000 1983.151 0.5343 -0.3064 0.1382 0.5273 -0.0268 0.9615 1646.0000 1983.00000 1983.1345 0.0250 958.2503 0.0251 0.0033 0.24149 W120 .90183 ) 18.3403 4.0201 480.42541 W120 N35 51 51.0312 0.1135 0.0462 -0.3161 0.1618 0.7071 -0.000 0.2590 0.3015 0.

0329 0.14609 15.0243 -21.0186 0.1080 -0.1523 5.000 0.0000 173.0216 0.9937 0.9999 0.0786 530.0231 0.0000 1983.56058 49.95877 N32 23 21.57767 N32 27 14.6425 0.000 0.88267 45.4972 1.0873 0.0000 1983.0000 0.0343 0.000 0.9454 -0.00C3 1.0000 1983.4820 0.0079 -0.43490 1345.0000 1992.000 0.900 1.00690 7 6 1.0250 0.81744 5.2590 0.13506 W115 11 44.000 0.0263 0.4420 8.45588 37.0000 0.882 1.0000 0.70314 31.000 0.0267 0.8787 2.0121 453.13719 N33 41 16.9999 38.0000 1983.20368 47.0000 1983.71?2 1967.9138 -0.0309 0.5526 1.0234 0.1639 0.6109 0.0000 1983.8885 19.8591 -0.8543 1.4792 0.4661 1983.36525 W115 11 28.0155 0.1336 0.2697 -0.36975 W120 27 11.0000 1983.85937 W116 W116 W116 W117 W117 W116 45 45 18 44 44 34 32.7577 0.3582 1.91211 W120 28 39.0050 0.9767 -0.1495 8.5843 910.5605 0.0222 876.6000 19.6116 19.000 0.64183 57.4558 0.0000 1983.000 0.0327 0.0211 0.7927 9.0268 1545.0266 -0.6306 10.7214 44.925 1983.0000 1983.9345 -0.000 0.000 0.4826 -0.7155 0.0000 0.0000 0.0261 -0.06793 W120 25 N35 53 17.0000 1983.0229 0.000 0.000 0.4775 0.0196 -0.000 11.4517 16.0000 1982.1215 0.000 3.2131 0.04538 N32 27 56.4727 0.9133 11.5385 367.0291 -0.6111 0.8333 0.7532 8.9291 1.60180 9.0320 0.1612 0.0010 1983.1118 0.0000 0.002 0.54072 W116 47 57.0281 berdo an black brink an brinkrma coahulla david an davis double elsinore eve eve rml gander_ hayrml_ haystack ida anzanet anza_ net anza__ net anza net anza net anza net anza net anza net anza net anza net anza net anza__net anza_net anza__net anza net N33 N33 N34 N34 N33 N33 N33 N33 N33 N33 N33 N33 N33 N33 N33 51 48 0 0 35 54 56 43 36 38 38 57 40 40 47 40.0000 19.0000 1983.0000 1992.5253 9.97252 N35 33 6.1076 12.000 0.69944 N35 28 N35 45 47.000 0.6069 19.96287 kitching kitchrm2 laqul_an lomancer lomas lookout N33 N33 N33 N33 N33 N33 anza net anza__ net anza__net anza net anza net anza net 58 58 42 45 45 33 31.27121 N35 29 15.7605 81.38534 58.6169 0.6311 10.48114 38.46660 N34 49 23.72047 W120 1 5.94303 W117 27 N35 2 10.12235 45.0000 1983.69983 -36.000 0.0000 0.1518 0.4426 0.0275 0.000 0.7927 923.0000 1983.4324 0.0266 -0.000 0.: 1.2123 0.56498 55.57708 W116 37 0.000 0.0288 0.9573 0.0032 1032.9976 0.000 1.1808 0.0000 0.0000 0.4342 0.000 0.1576 0.0177 0.5000 1.83206 W115 18 21.8679 1.0000 0.4398 0.0250 1488.0325 0.2156 507.0000 1983.7214 1983.04111 44.61770 48.4025 0.8925 6.0612 -0.0235 0.0526 15.77047 17.0000 1982.0000 6.97527 57.0000 1983.0000 1522.000 0.43459 54.002 0.98386 N33 44 56.8940 13.0000 0.23438 23.0000 1983.000 0.50CC 9.2093 0.000 0.4267 0.5633 0.98518 23.4361 11.0989 0.7370 7.1000 0.1591 175.000 0.0000 0.1005 -0.93662 W119 45 37.55835 W115 47 45.9000 17.4274 1.0270 -0.4252 0.3469 361.000 0.5520 4.0284 122.0000 1983.7696 743.0328 0.3514 1983.0270 0.0325 0.000 0.4643 1983.0036 1521.3678 12.0433 ]ason anzanet ]urup_an anzaret N33 33 41.7546 9.2529 0.40246 W116 W117 W117 W117 W116 W116 W116 W117 W117 W116 W116 W117 W116 W116 W117 5 39 8 8 46 59 59 7 20 33 33 6 27 27 19 20.0316 0.0270 0.000 0.0000 1983.5000 1.3614 0.0266 -0.6116 -0.000 0.000 8.0205 0.1904 0.85269 N33 52 13.37897 W118 44 34.50194 51.0050 0.8312 19.0149 -22.3147 0.002 0.1499 0.97659 54.61120 44.1076 12.0000 0.0000 840.0000 0.0403 0.0000 1983.0236 0.12393 8.2006 0.0000 0.05737 W116 11 15.0284 587.0187 762.8555 -21.0130 0.33416 32.03035 1.2747 0.6000 0.44634 3 37969 W121 0 22.09115 N34 1 56.9058 5.9110 15.0000 1983.3935 -0.76364 39.91966 N35 28 3.1003 2116.4819 0.4305 0.06813 W118 33 12.6062 -0.7082 -0.484: 0.22738 W116 23 29.0088 983.0230 0.5152 10.3666 4.5494 -0.000 0.000 0.10829 14.08083 50.0734 0.36615 W116 10 16.1945 1983.24092 N34 49 6.12606 W115 25 38.0::D 1983.7213 44.0268 0.8023 19.00354 N32 18 31.1341 0.4459 0.000 0.0000 1983.0343 -0.0003 0.000 0.6828 0.7029 755.81038 W120 26 N35 58 N35 53 43.0000 1983.600: 1.9931 0.0000 12.7460 -0.0000 1983.11212 55.56240 W120 19 9.000 0.0062 0.1394 -0.16316 N32 19 7.6454 19.0170 -0.0064 19.9615 1983.000 0.0000 1983.90678 W120 13 20.0195 524.3806 6.0349 0.82784 W120 15 38.0006 0.50:: 0.49:I 14.2576 -16.0000 1983.0000 1983.149 sanlu net towers twis rm2 sanlu-net twissel sanlu net two oaks sanlu-net vill2rm2 sanlu net villa 2 sanlu net sanlu net wild almon mo monit net beacon 2 monit net bm603 39 monit net monit net bolt bull mo monit net monitnet ca]on calie mo monit net mo monit net car chola mo monit-net cotto mo monit net monit-net edom 2 monit net gap gastr mo monit net monit net lgo_70 litte mo monit net mid ecc_ monit net mine mo monit net park mo monit net pattiway monit net red h mo monit net monit net resort salisbur monit net monit net sal rm saw e mo monit net sawmi mo monit net teon 41 monit net monit net tram monit net view 2 whita mo monit net mexic net 17 mexic net 24 3 mexic net mexic net 36 N35 47 20.2432 0.38891 29.43825 N34 48 2.0000 1983.2309 530.59382 35.0000 1983.0312 0.0232 -0.9185 -0.0326 0.07728 26.802 0.41405 54.426 11.000 0.5007 4.0000 0.4833 0.0285 -0.0322 0.9723 -0.0406 0.0281 0.58947 48.51980 W116 25 51.73410 W119 45 37.0622 -0.000 0.24961 26.15799 37.0000 0.000 0.36316 45.0000 1983.46646 W117 26 33.12235 32.3584 1674.1402 0.95583 22.8602 9.6449 1983.0742 -0.94809 N33 55 35.0000 1983.44733 13.0000 0.82772 N34 57 35.C-11 1983.000 0.0000 -0.71502 W120 29 53.0275 0.000 44.0000 1983.2469 0.0090 122.10731 17.48837 N34 49 22.24149 N35 58 16.000 0.1802 19.000 0.5289 6.8313 1.4582 0.0275 0.5424 0.1107 2.0142 -0.0208 0.0266 0.3161 0.0183 1646.0640 10.6C65 0.0169 0.1312 -0.08916 1.56683 N34 34 3.3513 0.36478 W119 42 47.52367 22.8871 0.36779 14.000 0.12138 N34 41 35.0235 0.1488 4.3664 4.4499 12.5536 0.0221 0.0002 1215.16606 29.0267 809.0000 0.60497 26.002 0.8605 -0.0170 -0.8923 1659.93141 37.6012 9.7609 663.1533 0.0000 0.0288 0.0000 0.800 0.40960 2.0180 751.8585 8.2544 -0.0000 1983.2459 701.0000 1983.2111 1983.0000 1983.0000 0.2538 0.0000 1983.6437 44.85747 17.67688 5.0192 0.0666 1058.2672 0.0189 749.0000 0.0000 0.3993 -0.4583 0.1349 0.0139 32.9043 662.0199 735.0000 1983.0000 0.0607 0.3225 1126.0330 0.0000 0.68392 W115 6 21.50.1550 0.4103 3.7071 -0.0000 -0.0000 0.000 1.54143 54.000 0.5300 1.0000 1983.0000 1982.33475 46.1565 -0.000 0.76166 11.7599 1.7818 11.9619 1.0000 0.0996 1.0000 1983.3333 -0.33391 W118 33 41.5311 1.0176 0.7101 -0.23844 53.4450 19.6177 0.3275 0.49570 W120 25 13.2547 0.30391 N35 51 31.0000 1983.1288 5.08153 N34 48 13.08161 W118 48 56.4783 22.000 0.0282 -0.6506 -0.0000 1983.0000 1983.6116 19.0491 498.84540 W120 28 8.0000 -25.0291 -0.67968 47.7100 44.0000 1983.31412 N35 29 15.0000 1983.7260 9.5623 1.97654 W120 18 21.000 19.2459 -14.0000 1983.0274 0.5181 0.0934 18.0000 1983.0000 -0.000 0.35180 W115 33 36.6301 -0.0000 0.27523 W115 52 23.000 0.0222 -0.0000 11.6=00 0.5003 3.6555 2.1995 1.0270 -0.4833 11.13025 19.7523 0.30806 N32 N32 N32 N32 N32 N32 N32 N32 26 28 27 12 25 49 37 27 14.3789 0.2441 0.0000 0.0014 1983.0000 0.1965 0.8136 4 8 9 928 bnp10065 carri me centinel cila 6 mexic net mexic net mexic net mexic net mexicnet mexic net mexic net mexic net david diablo- mexic net mexic net N32 13 20.0721 0.0236 0 0201 0 0343 0.0308 0.0285 0.6570 13.0411 1983.7980 2.000 0.3107 0.3794 0.7180 15.1398 719.0000 1164.9169 1211.79608 N35 47 15.0286 0.0000 1983.36000 W118 33 41.0211 188.0000 0.000 0.0000 1983.000 9.2419 -0.0000 -0.000 0.2771 0.2563 0.0230 0.01886 W116 17 15.24037 W120 1 5.7213 8.2683 -26.2218 0.0268 0.000 1.0346 0.2502 0.25343 N34 41 35.95:3 0.000 0.0205 0.1347 0.88115 1376.91418 15.0000 1983.0077 0.4647 0.56743 49.5241 1.0286 0.7541 1545.0327 0.3348 -0.43491 W116 17 17.0199 0.4581 0.03733 W116 33 52.1757 6.0314 0.0115 0.1660 4.0000 0.8638 2114.83638 54.0000 0.87475 33.6410 -0.9869 1127.000 0.0000 1545.0004 1983.0000 1983.000 0.000 0.4484 3.0000 1983.8182 0.0000 0.7093 -0.8805 1572.000 0.0059 -0.06:0 1983.0084 1221.000 0.2139 0.0266 0.0319 0.2145 3.0000 1983.7000 19.0631 188.04883 W119 25 56.5363 0.0003 0.16826 640.90914 W115 W115 W115 W115 W115 W116 W115 W115 16 15 11 39 15 0 42 55 28.4774 0.1637 366.62429 49.91:3 Dacumme mayor off 225m off 2 me prieto_ puerta 7-8-28 arling_ arling80 asbearml asbestos bachelor bee mexicnet mexic net mexic net mexic net mexic net mexicnet anza net anza__net anza_ net anza net anza net anza net anza net N32 N32 N32 N32 N32 N32 N33 N33 N33 N33 N33 N33 N33 41 5 38 37 25 22 32 52 52 37 37 36 43 52.43958 N35 36 17.6169 794.1304 0.000 2.7990 0.1925 0.91896 N34 20 49.0211 487.999 0.41121 W120 34 54.8085 1983.0000 1983.1899 0.0320 0.2037 0.94246 N35 2 10.01569 N33 52 12.1024 19.63656 N35 56 58.1000 0.82809 33.5723 0.39257 W116 W115 W115 W115 W115 W115 W117 W117 W117 W116 W116 W117 W117 9 18 43 55 18 27 43 28 28 27 27 3 41 52.49898 W119 42 48.9001 0.10492 N33 46 56.0C: 0.5338 1.0000 0.4994 0.000 0.81971 48.97021 21.0225 1412.79903 N33 46 53.10829 31.0000 1983.9179 1.0287 -0.0000 1983.9000 17.83939 26.9919 .4205 -0.0000 1983.3144 1661.9988 70.0317 0.9513 0.26770 W118 48 9.0000 0.0000 19.0822 2.70462 55.1708 0.0000 -0.05145 3.0263 0.49878 36.9995 0.4112 0.0000 1983.000 0.6720 1983.0000 -0.0000 1983.1578 0.000 0.000 0.50174 22.17828 N36 0 2.0181 1069.0133 1309.5700 19.3106 1.6062 0.1324 0.9930.0319 0.4709 0.4210 -0.0182 1450.4889 1.9555 1983.1525 -0.0000 1983.5824 11.4409 808.3929 0.0000 0.6000 0.48933 15.0186 -18.0268 1648.21290 30.0216 -0.84410 54.7068 44.000 0.0343 0.8125 -0.29734 0.0206 0.5000 0.4406 7.000 0.78539 N32 20 3.0218 0.4586 8.000 0.000 0.9999 0.7824 1983.5440 -0.000 0.4153 0.18203 56.0231 0.000 0.42444 19.0222 0.67492 39.0736 0.0225 479.0181 0.27089 1577.8092 0.6364 0.0239 494.6569 765.0000 0.0000 1983.4942 0.000 0.0000 0.0352 0.0000 0.0195 877.4241 0.000 17.0000 0.05495 33.0365 0.2263 dixie me mexic net mexic net fierro N32 47 27.1316 0.2576 19.000 0.0248 642.0233 0.5000 0.000 1.0275 0.67516 39.03001 W121 0 22.3190 -0.12835 N33 55 25.0000 1983.0000 1983.0168 -10.0000 1983.0049 786.4337 1567.6605 -0.0278 -0.8104 898.

55692 W117 39 13.0260 0.8569 -0.34968 9.0058 1.4706 6.4972 1.4446 -0.000 0.0163 0.5021 6.0099 1276.62884 W116 26 58.0047 701.9291 6.000 0.0000 1983.0131 1880.000 1.4377 0.0000 1983.28217 4.22046 W:16 46 27.34687 36.0250 0.000 42.25937 3:16 1 46.45132 W116 40 48.40424 W116 10 31.93:2 0.53271 W116 55 35.000 0.6116 0.3785 0.1966 -0.05689 5.8399 -0.0000 0.93814 44.0106 0.4908 14.0000 1983.000 1983.0000 1943.1282 0.6353 -0.09017 W116 54 38.6733 0.0311 1345.2865 0.0000 1983.0270 0.0000 0.36 : 0.0132 609.0000 1983.0000 1983.7470 19.0000 1983.6731 0.000 0.0526 15.8170 7.6814 0.3874 0.82551 W116 25 32.0222 1485.0000 1993.0000 1983.0283 663.0146 1315.0000 1983.54176 44.6834 8.3972 0.0000 1983.0091 0.1797 44.1388 -0.82421 W116 25 32.5855 3.0084 -0.0002 1983.9786 -0.82263 1116 26 58.5573 0.0000 1983.3443* 12.5526 1.0278 0.82897 48.000 0.54550 W117 44 17.0242 2617.000 0.3939 0.0156 453.1502 0.000 1983.000 0.33651 11.91194 115 36 40.0269 1533.56174 53.0000 1983.0000 1983.000 0.0000 1983.000 0.0079 0.4080 0.0065 0.17296 W116 11 27.7211 0.18945 W116 46 27.0079 0.37100 16.24979 W116 36 24.0265 359.0000 1983.07239 56.0270 0.000 0.0000 1983.000 13.6155 -72.0266 1558.42.5166 -0.4911 14.6819 -0.000 0.9271 -0.7214 0.74524 37.000 38.0100 1951.1110 19.1650 -0.0041 0.000 0.3725 -0.0000 1983.7558 0.51980 W11116 25 51.07:1 0.65288 52.0024 0.0062 -0.7781 0.000 4.0035 0.0276 0.000 0.0177 1232.0062 -0.13981 3:15 49 25.000 1.7220 -0.1565 -0.0088 0.5552 0.0000 1983.67666 W117 13 43.46-: 0.8140 0.7350 0.0049 748.0118 361.7927 -0.000 0.32663 11.0052 0.0211 0.22220 49.0304 508.6116 0.0061 0.77798 46.000 1.30997 58.000 16.49878 42.0035 0.77047 W116 0 54.5826 0.5022 6.27907 52.03977 W116 22 35.5095 0.6062 0.3443 0.0000 1983.0000 1983.51247 50.0258 633.2469 -0.4805 8.77336 W116 6 48.0236 0.8=95 0.5899 0.18760 25.27507 W116 13 59.4187 0.22378 16.0199 2617.3447 0.7596 0.65:3 0.8009 0.6677 0.0000 1983.0115 1316.0C: 0.0000 1983.3135 0.93827 W117 44 3.6410 -0.0003 1983.0007 1357.0000 104.000 0.000 0.3603 -0.0103 894.66973 W117 48 43.0256 0.53:5 0.0139 1879.0430 10.0000 0.6364 0.4830 -0.64183 56.32674 W116 6 32.72667 21.0000 1983.9764 8.734: 0.1488 4.3941 2.07997 5.42733 1:15 38 15.0163 0.0319 1000.7213 44.0060 0o.2309 4.28806 28.6957 -0.0000 250.89235 40.3614 0.000 18.77916 W116 41 57.0000 1983.0000 1983.5764 -0.2275 0.2933 19.0106 0.: 0.OOO 1.000 15.2728 0.24583 37.0298 0.000 0.0000 1983.0104 0.2472 -0.0000 1983.6169 -0.0000 1983.1949 0.0289 343.0142 833.8656 44.49 0.55768 32.9864 12.51:7 0.4914 0.5292 13.0143 0.0000 1983.000 44.000 0.7408 19.4236 -0.0297 9.22174 W116 23 23.0139 0.195: 0.29139 1:16 7 34.6211 0.0000 975.5255 0.5529 0.0203 0.0343 498.11784 9.4581 0.6220 6.0000 1983.7214 44.0169 0.95406 52.1899 0.05689 3.93740 52.0142 0.1124 44.7469 8.0000 1577.000 0.7457 9.24174 25.0078 1527.5352 -0.91345 58.96:5 0.0000 1983.7767 -0.1288 5.98888 30.5255 -0.0211 0.0000 0.5660 -0.0000 1983.51112 16.000 0.0480 10.82519 W116 40 58.5332 0.000 0.0053 0.10710 W116 11 43.0270 1691.2226 19.99?9 0.0000 1983.0491 -0.80460 57.0000 1983.4865 0.9897 0.0254 0.000 0.0100 1127.96690 17.60:2 0.0314 0.9495 14.6036 0.9512 1.3302 19.6954 10.94:3 0.05495 41.000 42.05348 27.65146 W115 21 29.0469 -0.77018 W117 4 14.0270 735.0025 0.5824 0.4824 1.47i" 0.7696 -0.0286 1695.000 0.4892 0.5424 6.74562 W115 58 8.0003 1983.9703 44.8544 19.31260 W116 40 50.0143 0.24961 44.98386 8.9997 0.3935 11.2701 0.0003 1983.6282 6.0243 0.0000 1983.0498 0.000 0.8569 -0.8092 0.0000 1983.000 14.61109 3116 27 30.3811 0.0295 -85.1046 -0.13396 16.0260 0.0000 1983.3885 0.0275 894.54143 W116 18 45.0000 1983.0079 0.5711 19.0000 1983.0270 0.0584 1.0366 0.67968 8.000 0.6169 -0.0251 0.0102 0.0079 1028.000 0.17296 W116 11 27.4155 0.41606 W116 18 21.3514 -0.59013 7.75860 W116 4 44.37631 0.4957 0.24881 27.0000 0.0000 1983.12804 52.3846 17.7256 0.000 0.1433 -0.000 0.0206 -71.0049 672.19138 27.6211 0.0037 1577.63459 13.0911 12.0049 2031.95406 18.7100 0.0177 174.1000 14.5577 3.7213 44.6862 14.7343 0.69.0321 -71.000 0.7532 8.25815 W116 24 25.6569 0.000 42.84757 W116 18 22.9928 0.000 1.4277 -0.5891 8. 0.85252 W115 57 5.7130 -0.0190 0.000 1.8856 0.0000 1983.06371 W115 20 41.4332 0.5412 -0.75883 W:16 4 44.31-3 0.5099 14.1370 -0.000 0.3146 0.75636 W117 31 59.0267 0.0000 2.000 12.68389 W116 11 43.13203 W:15 43 31.000 0.5569 0.10583 51.6418 13.6115 0.8602 9.000 1983.7163 4.1118 44.4274 1.19149 3116 55 35.0064 1983.0264 -0.000 2.000 0.000 0.3890 0.000 1.57767 42.000 0.1835 -0.000 0.0191 675.4617 0.67688 W115 55 57.8602 9.0096 -0.9433 19.53044 W117 11 28.9569 -0.0933 0.0268 1700.0088 105.9282 16.32339 W117 43 49.0000 1983.0309 264.0267 0.26759 W:16 6 32.7187 19.0118 0.7017 7.3888 19.8467 18.0115 485.9994 0.5032 0.1046 31.99043 24.0000 893.1945 0.9344 0.8C!5 0.07:9 0.7214 0.0640 10.0000 1983.51632 5.4031 1.0258 750.0628 1.5074 0.0203 0.3845 0.5901 -0.3950 5.4003 5.0068 0.56507 27.0612 0.0242 2040.93738 11.0354 0.7521 44.0000 0.5441 -0.39550 W115 58 50.0050 585.55881 W116 5 20.3074 -0.10OC 0.6C:3 0.8925 6.0064 0.7602 -0.1830 -0.4392 0.70369 W116 18 21.81831 14.6062 0.5096 0.0254 0.0000 0.0000 1983.000 2.0293 9.1996 -0.2714 0.0282 818.6766 3.150 martl an anza net menifee_ anza net micro anza net microrml anza__net moore anza net moss anza net nelson anza net niguel_ anzanet niguel a anza__net pollycgs anza net pollydmg anza net ranger_ anza_net road ansa net roundtop anza net san an ana net san joaq anza net sanjuan anza_net santiago anza net sheep_an anza__net sier anza net stubb an anza net thom_cgs anzanet thom2asu anza net thomasec anza net thomecc2 anzae net toro anza__net toro 80 anza net 29_palms joshu net beaoo_jo joshunet berdojo joshunet creole_ joshu net creolrml joshunet creolrm2 joshunet creolrm3 joshu net dome joehunet dome_ecc joshu net edok jo joshu net insprm2 joshu net inspncer joshu net joshunet keys laquijo joshu net ummj3o joshe net joshu net meeks mesquite joshunet pax ncer joshu_net queen joshu_net joshu net rich_ sandhill joshu net segreset joshu_net segundo_ ]oshu net stubbe_ joshu_net valmtec joshu net joshu net warren alamo salto net beals salto net berdoo salto net blbuvlbi salto net bluff salto net bluffres salto net bluffrml salto_ net butte salto net carrizo salto net coach_ salto net coolidge salto net cottonwo salto net cuyamaca salto net salto net diixe elephant salto net fish salto net frink_ salto net frinkrml salto net granite salto net jacuba_ salto_net kane salto net laquinta saltonet martinez salto net mecca salto net mon rea salto net obsidian salto net oco rmal saltonet ocotillo salto net off 225 salto net off 229 salto net oldbeach salto net orocopia salto net palm____ alto_net pt_101__ altonet salton salto net N33 N33 N33 N33 N33 N33 N33 N33 N33 M33 N33 N33 U33 N33 N33 W33 N33 N33 N33 N33 N33 N33 N33 N33 N33 N33 N33 N34 133 N33 N34 N34 N34 N34 U33 N33 N33 U33 133 N33 133 334 334 334 N334 334 N34 N334 N334 N34 N33 N34 N34 133 N33 133 N33 N33 N33 N33 N33 N32 N33 133 133 N32 132 N33 132 N33 133 N33 N32 133 N33 N33 N33 N32 133 133 332 1132 N33 133 133 N33 133 133 25.1009 -0.6413 -0.000 3.000 0.5157 7.97478 14.000 1983.03120 14.0000 1983.0000 1983.0138 1694.46633 W115 43 11.0198 0.36368 W116 40 50.4028 1.6209 0.308: 0.27763 52.8710 19.5646 13.33033 49.6381 7.25420 16.0034 15.000 4.52576 W:15 54 35.3771 0.0152 0.67968 25.7007 2.0000 -36.55835 W115 47 45.0062 1677.0003 1983.4302 0.4824 1.5552 0.000 1.98167 41.0267 81.0000 1983.000 0.8755 0.5385 -0.8085 0.0148 1028.0000 1983.6418 13.0631 -0.338 12.0062 -0.0060 0. 000 0.0232 0.6766 11.0325 7.0000 1983.57.0000 0.0040 0.4324 8.0498 0.28449 30.15534 41.0001 1983.000 0.000 0.0000 1983.4532 0.000 0.0064 494.0277 0.0000 1983.0000 1983.53:2 193.29714 174.49099 W116 11 26.0330 -15.89474 W115 39 22.52691 W117 13 25.0000 1983.1562 0.0083 0.0002 1983.3077 -0.55881 W116 5 20.0265 1340.000 1.4957 0.0233 0.4649 0.1959 0.0142 0.0000 1983.0278 0.3689 0.0203 0.29490 W117 11 27.33033 44.35194 16.5721 0.0003 1983.00149 WI15 38 49.25-7 0.5385 -0.2977 0.0226 0.6116 2.0065 652.6Cit 0.6069 -0.000 0.0000 1983.9569 -0.4349 0.1046 31.000 0.17818 W116 4 38.3215 -0.0079 0.0000 1983.65568 W15 46 47.0082 0.79903 40.0000 757.5533 -0.4027 1.3903 0.3498 44.3755 0.6927 0.54143 W116 18 45.0152 0.5819 -0.16916 37.0040 1137.000 1.0050 1317.74216 54.8619 12.47342 W115 48 56.0002 1983.3669 11.255! 0.6102 0.75451 W116 28 7.44404 39.5695 0.0282 0.18833 24.0122 1330.57708 W116 37 0.0000 1983.1377 0.000 0.0298 676.0155 1664.1684 11.38318 24.000 0.0003 1983.52460 W116 25 14.4426 0.53534 W:15 30 4.0000 1983.000 0.4869 0.56151 W116 35 21.1469 -0.40945 W116 18 21.12673 8.4028 -0.9907 11.000 0.000 1.000 0.4915 0.17192 44.5187 -0.0003 1983.9684 0.5166 -0.11555 8.0242 2050.0220 0 0245 361.7214 44.7532 8.80000 50.0000 1983.0000 1983.99092 W11116 16 43.5989 0.4556 -0.6920 15.2659 0.0193 1300.0000 0.000 0.0063 0.0242 2048.27198 W116 13 59.0211 0.0002 1983.8085 19.5279 -0.7473 -0.0000 1983.000 4.49104 W116 5 49.00107 0.1796 -0.3242 0 .4861 0.48262 14.8093 .000 44.33672 16.98038 W117 32 3.000 0.1436 0.000 0.0203 0.4372 0.8995 -0.0000 1983.000 0.1523 5.11188 28.81959 W116 49 31.0000 0.000 19.78569 24.8610 -0.0000 1983.0156 583.0265 360.08869 22.0001 0.9837 0.191 479.0256 279.0000 1983.7469 8.0260 0.7916 -0.0259 -0.6569 0.5020 6.0050 1315.2536 23.0000 362.7196 -0.000 0.4547 0.6509 15.0298 9.8375 -0.0086 790.000 0.0000 1983.0000 1983.07728 W116 9 52.0277 0.8154 0.9688 -0.000 0.45105 W116 37 2.0000 1983.7257 35.2931 0.5795 7.464) 0.12410 W115 38 15.8679 1.0324 1983.0266 985.1 0.10011 W115 38 49.04912 W116 28 46.9166 0.000 0.000 0.77680 44.000 1.6364 0.0000 1662.14e5 0.0066 0.000 0.5577 3.17957 W116 13 58.

4096 0.0215 0.0000 0.0030 -0.0316 0.8983 1081.000 1983.000 1983.2583 -0.8293 0.0001 0.96416 54.65786 6.0083 0.7886 .73003 6.5981 0.000 1983.0002 1983.4424 1.0251 0.74694 43.00044 594.0107 0.0001 0.7456 6.0341 0.7782 0.4032 0.0247 -0.99499 43.0317 0.2916 1594.000 1983.26926 17.3875 1363.0000 0.2821 0.8009 8.000 1983.0029 0.4216 0.0253 -0.99481 1.0001 0.11664 30.0317 0.000 1.8180 1594.0259 -0.6041 0.0148 0.6180 4.57596 W115 W115 W116 W115 W116 W116 W116 W116 15 55 17 49 34 34 26 16 41.32205 4.0285 0.15601 42.6815 17.151 salva salto net soda salto net salto net stage sup_ saltonet volcan salto net volcan 2 salto net wilson salto net salto_net yak 1.0298 0.39365 39.6408 10.22493 8.5212 5.8944 191.0476 31.0311 0.0083 -0.0272 -0.5624 6.0004 0.73013 38.0083 1.000 1983.0283 0.9068 824.000 1983.0171 -0.6408 10.38765 8.0001 0.

152

Table 4: VLBI/GPS sites used in this analysis
* frame: WGS84
* u = V(east) v = V(north) w = V(up),
unit: m/year
* uncertainty of coordinates: Sn = north, Se = east, Su = up,
*

site
YUMA7894
BLKB7269
DEAD7267
MONP7274
PINY GPS
GOLDVENU
MOJAGPS
SIO1_GPS
SOLJ_GPS
NIGU_GPS
PEAR7254
JPIM_GPS
OVRO_GPS
PVER_GPS
BRSH_GPS
BLUF_GPS
SAFEGPS
LOVEGPS
WHIT_GPS
CATOGPS
HAPYGPS
HOPPGPS
SNPAGPS
SNP2 GPS
SCLAGPS
MPNSGPS
COTR_GPS
MUNS_GPS
SOLI_GPS
FIBRGPS
TWIN_GPS
YAM2_GPS
LACU GPS
CENTGPS
MADCGPS
GAVI_GPS
ALAM_GPS
POZO_GPS
GRAS GPS
LOSPGPS
VNDNGPS
BLRL GPS
BLAN_GPS

full-name
glbk_sln
glbk_sl1n
glbk sln
glbk sln
glbk_sln
glbk sln
glbk sln
glbk sln
glbk aln
glbk_ln
glbk sln
glbk_sln
glbk sln
glbk_sln
glbk_ ln
glbk sln
glbk_ sln
glbk_sln
glbk_ sln
glbk_sln
glbk aln
glbk sln
glbk sln
glbk sin
glbk_sln
glbk_sln
glbk_siln
glbk_sln
glbk_sln
glbk_sln
glbk sln
glbk_aln
glbk_ ln
glbk_sln
glbk_sln
glbk_sln
glbk sln
glbk sln
glbk sln
glbk_ sln
glbk_sln
glbk aln
glbk_sln

latitude
N32 56 20.89178
N33 39 49.49969
N34 15 17.99416
N32 53 30.35936
N33 36 33.29567
N35 14 51.72318
N35 19 53.60277
N32 52
3.98589
N32 50 23.52199
N33 30 52.32750
N34 30 43.67421
N34 12 17.35069
N37 13 57.22076
N33 44 37.54758
N33 24 25.25418
N32 55 36.40844
N34 19 49.57244
N34 29 46.75128
N34 34 2.77603
N34 - 5 8.92891
N34 21 28.70043
N34 28 39.80814
N34 23 16.33306
N34 26 25.26217
N34 19 32.44823
N34 48 46.17029
N34
7 12.68080
N34 38 8.81283
N34 17 54.02647
N35 23 54.58928
N33 13 54.29920
N34 51
9.16981
N34 29 39.87590
N33 59 41.43538
N35 4 32.23428
N34 30 6.52010
N34 47 54.60789
N35 20 45.59111
N34 43 50.04385
N34 53 37.47457
N34 33 22.55601
N35 21 31.36183
N35 39 52.44917

longitude
W114 12 11.30341
W115 43 11.33474
W116 16 43.97113
W116 25 22.14310
W116 27 31.68935
W116 47 41.60213
W116 53 17.35055
W117 15 8.39604
W117 15 8.99042
W117 43 49.05408
W117 55 20.61791
W118 10 23.59868
W118 17 37.65250
W118 24 12.78565
W118 24 17.47350
W118 31 6.75401
W118 36 4.85026
W118 40 7.17291
W118 44 34.00343
W118 47 8.75839
W118 51 0.52712
W118 51 55.96788
W118 59 55.65153
W119 0 34.72668
W119 2 21.12539
W119 8 43.47866
W:19 9 14.39109
W119 18 1.94763
W119 20 33.67371
W119 23 38.45561
W119 28 44.25935
W119 29 3.89002
W119 42 50.01706
W119 45 10.51788
WI20 4 1.66736
W120 11 55.65845
W:20 15 24.47970
W120 17 55.29107
W120 24 50.64455
W120 36 22.28021
W:20 36 58.31329
W120 49 54.22078
W121 17 4.04932

unit: meter

height
238.4184
489.3921
833.0976
1838.7331
1235.5440
1062.8826
904.5336
7.5681
215.4982
235.7494
890.9723
423.9774
1178.0603
69.4543
448.5902
297.3293
1102.9961
727.0558
1221.3702
825.5900
669.0879
1344.9731
184.7995
1476.6934
655.3611
2662.6006
-34.0748
2107.5871
-11.6259
54.7118
200.1605
806.6027
1164.8576
389.5322
958.0918
713.3918
457.2234
728.6537
331.1783
463.1409
-11.4871
166.4970
-24.8115

u
0.0004
-0.0054
-0.0040
-0.0317
-0.0165
-0.0024
-0.0039
-0.0322
-0.0323
-0.0314
-0.0150
-0.0270
-0.0058
-0.0291
-0.0307
-0.0306
-0.0261
-0.0248
-0.0121
-0.0291
-0.0248
-0.0227
-0.0272
-0.0242
-0.0264
-0.0223
-0.0289
-0.0280
-0.0292
-0.0088
-0.0324
-0.0280
-0.0298
-0.0278
-0.0248
-0.0323
-0.0312
-0.0242
-0.0286
-0.0279
-0.0302
-0.0264
-0.0246

v
0.0002
0.0010
-0.0002
0.0253
0.0212
0.0082
0.0077
0.0299
0.0285
0.0307
0.0205
0.0242
0.0076
0.0301
0.0333
0.0359
0.0255
0.0218
0.0251
0.0291
0.0296
0.0236
0.0270
0.0231
0 0286
0.0198
0.0340
0.0249
0.0281
0.0127
0.0355
0.0284
0.0301
0.0342
0.0306
0.0344
0.0347
0.0316
0 0369
0.0347
0.0363
0.0357
0.0359

w
0.0105
-0.0031
0.0541
-0.0064
-0.0018
-0.0081
-0.0032
0.0153
0.0069
0.0011
-0.0158
0.0126
-0.0010
0.0092
0.0088
0.0070
0.0094
0.0068
0.0009
0.0014
0.0144
0.0123
0.0062
-0.0023
0.0195
0.0233
-0.0271
0.0044
0.0046
0.0040
-0.0026
0.0010
0.0036
0.0065
0.0040
-0.0098
0.0282
0.0144
0.0165
0.0050
0.0030
0.0068
-0.0089

epoch
Sn
1987.353 0.0657
1987.079 0.0658
1987.642 0.0659
1987.544
0.0657
1986.480 0.0636
1989.506 0 0658
1991.161 0.0619
1991.876 0.0635
1988.170 0.0636
1989.787 0.0628
1986.877 0.0659
1991.901 0.0620
1989.908 0.0597
1990.778 0.0622
1989.508 0.0625
1989.313 0.0627
1991.498 0.0617
1991.749 0.0615
1990.550 0.0615
1991.589 0.0618
1991.762 0.0615
1991.881
0.0614
1990.597
0.0614
1992.127 0.0614
1990.215
0.0615
1991.705
0.0610
1989.603 0.0616
1991.396 0.0611
1991.471 0.0613
1989.906 0.0604
1989.613 0.0620
1990.462 0.0608
1991.061 0.0609
1989.853 0.0612
1990.801 0.0603
1989.950 0.0606
1991.478 0.0604
1989.899 0.0600
1991.231 0.0603
1990.706 0.0601
1991.399 0.0603
1989.712 0.0596
1989.960 0.0591

Se
0.0647
0.0649
0.0651
0.0646
0.0332
0.0647
0.0317
0.0327
0.0331
0.0331
0.0651
0.0332
0.0323
0.0337
0.0339
0.0343
0.0337
0.0337
0.0340
0.0340
0.0340
0.0339
0.0341
0.0341
0.0344
0.0341
0.0346
0.0343
0.0345
0.0341
0.0352
0.0345
0.0348
0.0351
0.0349
0.0354
0.0353
0.0352
0.0354
0.0356
0.0357
0.0356
0.0360

Su
0.0653
0.0654
0.0655
0.0653
0.01C2
0.0653
0.0075
0.0069
0.0078
0.0072
0.0656
0.0067
0.0069
0.0069
0.0072
0.00'3
0.0069
0.0070
0.0076
0.0070
0.0071
0.0063
0.00C9
0.0069
0.00'6
0.00':
0.00-B
0.00'
0.0C69
0.0070
0.0073
0.0072
0.0069
0.0070
0.0069
0.0073
0.0069
0.0072
0.0069
0.0069
0.0068
0.0070
0.0071

153

Table 5: USGS trilateration subnetworks
sitea

min. vel. direction

scatters

San Gabriel - Tehachapi
Los Padres - Tehachapi
Carrizo - San Luis
San Luis - Parkfield

52
35
26
67

N25 0 E
N30 0E
N500 E
N50 0E

5.70 mm, 0.17 ppm
4.23 mm, 0.20 ppm
3.07 mm, 0.22 ppm
4.84 mm, 0.25 ppm

Salton - Mexicali
Anza - Joshua

63

N50 0 E
N50 0 E

3.35 mm, 0.22 ppm
3.64 mm; 0.22 ppm

subnetwork

83

a. Some sites are used by two subnetworks. The total number of sites is 323
Table 6: imposed constraints for minimum constrained solutions
function
constraint
section
control section translation
southern V(Mecca) = 0
control section rotation
V(Salva) towards N135 0 E
control section translation
V(Pattiway) = 0
northern
control section rotation
V(Tank) towards N115 0E
V(Red hill) towards N400W remedy connection defect between
San Luis and Carrizo subnetworks
stabilize Carrizo subnetwork
V(Gifford) = V(Allen)

154

Table 7: Velocity constraints

VG site
DEAD7267
NIGU_GPS
PINY_GPS
MONP7274
BLHL_GPS
SAFE_GPS
MPNS_GPS
SNP2_GPS
JPLM_GPS
PEAR7254
WHITGPS
MUNS_GPS

EDM site
Sandhill
Niguel
Asbestos
Monu_res
Blhllres
Piconcer
Mtpinos
Santapau
Jpllrml
Tank
whitaker
Reyes

I

rli]

ti

t

distance
0 lnkm
0.43 km

11

2.07 kman
0.22 km
0.02 km
0.03 km

0 km
0 km
0.21 km
1.34 km
0.02 km
1.83 km

i

i

final status
used
used
used
not used
used
used
used
not used
not used
used
not used
used

01093596 -0.00230387 -0.01315690 0.00881368 -0.173 0.66456696 compatibility criterion (95 dVe dVn 0.048 0.07561848 34.00265779 0.0079 0.07133113 0.079 0.04037959 -0.066 0.082 0.060 0.01159851 -0.00144013 -0.558 1.1215 0.05568800 -0.39865512 241.085 0.06852287 0.00924951 -0.81282355 34.1238 0.052 0.073 1.54119793 243.02804152 -0.56743609 34.059 0.00629703 0.09740505 -0.071 0.037 0.059 0.74767081 242.28018503 243.33043485 34.49440779 33.099 0.01003713 0.1891 0.00769894 -0.28610896 240.02840351 -0.1770 0.125 0.0002 0.3777 9:11:10 9:11:10 site YUMA7894 BLKB7269 DEAD7267 MONP7274 PINY GPS GOLDVENU MOJA GPS SIO1 GPS SOLJ GPS NIGU GPS PEAR7254 JPIM GPS OVRO GPS PVER GPS BRSH GPS BLUF GPS SAFE GPS LOVE GPS WHIT GPS CATO GPS HAPY GPS HOPP GPS SNPA GPS SNP2 GPS SCLA GPS MPNS GPS COTR GPS MUNS GPS SOLI GPS FIBR GPS TWIN GPS YAM2 GPS LACU GPS CENT GPS MADC GPS GAVI GPS ALAM GPS POZO GPS GRAS GPS LOSP GPS VNDN GPS BLHL GPS BLAN GPS .26970952 242.02023866 0.34599556 34.55626281 35.96080051 240.89414712 -0.72111933 243.0174 0.06183294 0.83986590 33.2082 0.00136413 0.51453355 34.2946 0.20511069 243.03334909 -0.112 -0.4- -- 1-- - - _ 155 Table 8a: Compatibility test: formal uncertaities of VG solution scaled by a factor of 2 **t********************************************************************************************* Compatibility test: using VG final combined sites.090 0.66374988 34.57718401 ! 243.21423648 241.52104141 240.053 0.389 0.123 0.065 0.118 -0.04762380 0.050 0.48145932 241.82678075 241.00905166 0.map.3831 0.088 0.07769973 0. 2 sigma scaling * and solvemcomb.068 0.00853319 -0.92677808 34.85254527 34.0442 0.4527 0.85459139 240.02540618 0.0327 0.3535 0.10485400 0.4248 0.02744370 0.38786845 34.35341706 0.24770040 35.22896639 -0.04608349 -0.93287233 239.064 0.083 0.087 0.63577947 34.39055221 -0.5619 0.058 0.00955126 0.93913660 33.11105169 0.25499839 32.010 0.74320302 239.073 0.3541 0.00357143 0.047 0.2492 0.126 0.24708046 239.010 0.99035625 240.01240376 0.29833854 35.087 0.5409 0.032 0.99484105 35.15998870 -0.08124144 -0.29972965 0.078 0.099 0.56813894 0./net mld/solvem scum.1354 0.50180898 34.01247149 0.05806882 0.183 0.00620707 -0.33155577 32.71554378 failed to pass lat 32.075 0.03555383 0.053 0.45676708 .074 0.02264104 -0.0002 0.4458 1.59645089 241.2635 0.33134321 241.089 0.79686014 244.0118 0.03119595 0.01672173 0.154 0.972 1.59514843 241.08581136 34.064 -0.53368052 -1. between: .2slg and solvem comb.088 0.00600201 -0.60924826 35.39381344 239.65731550 240.00185869 -0.2460 0.map2sig ****dif.0012 0.071 0.010 4.23256085 33.40701310 32..44034866 34.010 .map2sig time: 1993/ 6/19 between: .16827399 238.01268432 -0.084 0.115 0.01103823 -0.010 0.2807 0.01671035 -0.23408981 -0.01445015 0.010 0.0035 0.49631810 34.258 0.079 0.027 0.00574471 -0.133 0.35870946 35.69946154 240.6166 0.74750448 242.11668614 -0.092 0.94341074 -0.35719731 -0.0217 0.02452755 0.02482963 -0.039 0.27162688 -0.00768912 0.28616706 0.5056 0.08073921 0.0001 0.73056499 34.51213101 34.0641 0.12018709 34.93363227 0.14.080 0.84600455 240.map./net**mid/solvescum. * dlf.05496714 -0.13445555 241.80120849 239.066 0.145 0.874 0.39849624 33.2sig time: 1993/ 6/19 * * symbol ! means ion 245.010 0.00997472 0.00583750 0.166 0.60598525 240.059 0.03383853 -0.067 0.15725304 0.70130990 239.00700461 0.0114 0.47772261 34.210 -0.158 0.051 0.23174766 34.tt.101 -0.35797001 34.51558838 240.12466139 -0.08541753 0.0000 0.03997701 -0.503 0.3680 0.16629373 0.20481771 37.289 -0.70620810 241.89176545 33.376 0.23830264 0.0.065 0.79849954 35.6352 0.2558 0.0009 0.38380469 239.151 -0.3211 0.32567828 34.89374052 34.329 0.11184743 i 242.25722227 241.07910686 -0.58593468 239.14985593 241.09311457 -0.00121016 240.74376112 33.128 0.34017860 0.865 0.139 0.07760677 241.087 0.075 0.86777149 32.06122698 % confidence level) Cvn corr Cve 0.351 0.1356 1.010 0.01927014 0.1298 0.62119023 -0.

493 7.081 0.083 0.13445555 241.04121216 0.02103266 -0.1224 34.2316 -0.143 33.123 0.534 -0.03898145 0.00298646 0.28610896 240.92677808 -0.1705 -0.187 0.8062 35.05775850 0.03034332 0.0000 35.2363 34.55243931 1.12775462 0.57709513 .064 0.49631810 -0.0090 34.164 34.59677372 1.__~ __ ___ _I_~ ___ 156 Table 8b: Compatibility test: formal uncertaities of VG solution scaled by a factor of 3 **tt***t*t***** t***********************************************t t * Compatibility test: using VG final combined sites.05712972 1.50180898 0.33155577 -0.map and solvem comb.58593468 239.between: solvem****scumap and solvecomb.125 0.2276 site YUMA7894 BLKB7269 DEAD7267 MONP7274 PINY GPS GOLDVENU MOJA GPS SIO1 GPS SOLJ GPS NIGU GPS PEAR7254 JPLM GPS OVRO GPS PVER GPS BRSH GPS BLUF GPS SAFE GPS LOVE GPS WHIT GPS CATO GPS HAPY GPS HOPP GPS SNPA GPS SNP2 GPS SCLA GPS MPNS GPS COTR GPS MUNS GPS SOLI GPS FIBR GPS TWIN GPS YAM2 GPS LACU GPS CENT GPS MADC GPS GAVI GPS AL4 GrPS GRAS LOSP VNDN BLHL BLAN - GFS GPS GPS GPS ___ ~ ___ .288 0.23174766 -0.66374988 0.71554378 failed to pass compatibility criterion (95 % confidence level) lat dVe dVn Cvn Cve corr 32.47673354 -0.44715790 0.010 0.04346573 34.685 0.134 0.00773401 0.125 -0.49641544 0.39865512 241.13980394 0.113 33.507 -0.126 0.3208 -0.142 0.157 -0.47772261 -0.59645089 241.01427152 0.138 0.04306524 0.25722227 241.73056499 0.139 -0.80120849 239.74376112 0.29833854 -0.2492 -0.79686014 244.3805 34.048 -0.463 0.93287233 239.35797001 0.85459139 240.24708046 239.40834621 0.04459018 -0.28018503 243.02387236 0.266 0.65731550 240.07561848 0.99484105 0.69946154 240.81282355 -0.07081243 0.668 0.02890729 0.227 0.51558838 240.15432592 34.137 0.124 0.39212608 -0.04783318 0.99035625 240.60598525 240.07692382 0.295 -0.20396856 0._ii/_~~ I_ X~iii _ __ :.0001 34.0483 0.07261411 0.02951935 0.89176545 -0.55626281 -0.5763 0.2550 34.106 0.38786845 0.31948650 0.21423648 241.02636650 0.83986590 -1.622 35.058 0.20481771 0.103 0.74767081 242.0854 34.56636179 1.3662 0.23256085 0.0898 2.125 -0.101 0.185 -0.9018 0.25499839 2.115 0.37961713 33.96080051 240.247 0.21086890 0.14941471 34.33043485 -0.1042 34.4817 -0.08581136 -0.12018709 -0.228 0.03626075 0.19091242 -0.07760677 241.06308288 0.07398258 0.05067990 0.00925721 0.01169488 0.3795 34.00110082 0.3441 32.03529394 0.3683 0.26970952 242.243 0.026 0.32770323 0.48145932 241.03291356 0.214 -0.70130990 239.54119793 243.068 0.1161 34.128 0.055 -0.25995412 -0.02124843 -0.86777149 0.210 0.34599556 0.63577947 0.49440779 0.010 0.17561412 0.74750448 242.00369599 0.05770642 0.127 0.058 0.44034866 -0.map time: 1993/ 6/18 10:38: 8 * symbol I means lon 245.2375 34.35870946 0.365 0.115 0.82678075 241.159 0.51213101 -0.00284685 0.110 0.16827399 238.154 34.02355854 0.128 34.010 0.00121016 240.32567828 -0.07326673 34.5712 0.655 1.3424 0.00567670 35.3356 0.56743609 -0.01161305 0.4297 35.0492 32.135 0.662 0.02852606 0.79849954 0.93913660 0.186 -0.1054 0.091 0.13929515 0.349 -0. between: solvemscum.725 -2.932 34.92675860 3.2982 0.105 32.130 0.01423932 33.172 -0.38380469 239.02756850 0.51453355 -1.02745191 0.27230285 33.map time: 1993/ 6/18 10:38: 8 * dlf.00086074 0.243 0.14888767 0.44382895 0.00156087 0.1368 0.12751644 0.03382335 0.103 0.14.24770040 -0.89374052 0.12692127 0.59514843 241.52104141 240.06726543 0.20485314 0.0741 -0.0002 -0.127 0.317 0.14985593 241.57718401 243.03755696 0.60924826 0.141 -0.0614 -0.84600455 240.74320302 239.072 0.1914 0.129 0.11184743 242.85254527 -0. 3 sigma scaling ***dif.4656 34.272 34.139 0.72407307 0.2489 -0.046 33.176 0.099 0.236 0.02141486 0.39849624 -0.72111933 243.01022904 0.33134321 241.70620810 241.12629734 33.078 0.150 37.3762 0.40701310 -0.01687425 0.05412186 35.103 32.66456696 0.3309 35.03599830 0.01514391 34.306 0.39381344 239.405 0.20511069 243.6476 34.05655611 0.00716116 0.15316821 0.

115 0.365 0.246 0.89176545 0.2767 0.00062210 0.58593468 239.47772261 0.191 -0.07760677 241.16316473 33.046 0.164 1.01754592 0.01742585 0.298 -0.66374988 7.02148697 -0.7991 0.28610896 240.24770040 -0.07797983 0.3424 32.04595694 0.30436696 -0.3595 34.150 0.49440779 -0.03056316 37.03557993 0.8924 0.1777 0.23174766 0.06861393 -0.0492 0.147 -0.73056499 -0.078 0.736 -1.51213101 -0.04153991 0.00252674 0.93287233 239. /comb/solvem.131 0.97691006 0.57623295 -14.79849954 -0.01186425 0.0097 1.124 0.103 0.2982 0.48145932 241.243 0.73009727 0.00573141 0.09003169 0.01125985 0.92677808 -0.0365 -0.176 0.74750448 242.44034866 3.16591662 0.82678075 241.66456696 site YUMA7894 BLK7269 DEAD7267 MONP7274 PINY GPS GOLDVEN MOJA GPS SIO1 GPS SOLJ GPS NIGO GPS PEAR7254 JPIM GPS OV1d GPS PVER BRSB BLF SAFE LOVE WHIT CATO GPS GPS GPS GPS GPS GPS GPS HAPY GPS HOPP GPS SNPA GPS SNP2 GPS SCL GPS MPUS GPS COTR GPS MONS GPS SOLI GPS FIBR TWIN YAM2 LACd CENT MADC GAVI ALAN POZO GRAS LOSP VNDN BLBL BLAN GPS GPS GPS GPS GPS GPS GPS GPS GPS GPS GPS GPS GPS GPS .6096 34.2489 0.137 34.099 0.4656 35.84600455 240.493 34.137 0.59645089 241.068 0.01944794 0.134 0.058 34.781 0.14985593 241.5712 0.01451327 0.317 0.51453355 1.map and .99035625 240.214 -0.14598127 0.08908248 34.39865512 241.96080051 240.655 0.01342988 0.113 33.02924975 1.60924826 0.243 0.1335 34.11046139 0.083 0.01205101 0.228 0.110 0.71554378 time: 1993/ 6/18 18:56:46 failed to pass compatibility criterion (95 % confidence level) Cvn corr Cve lat dVe dVn 0.03640911 0.04638422 0.60598525 240.1906 0.110 35.139 0.130 0.02630640 0.01679431 0.127 0.35870946 0.85254527 -0.103 -0.40701310 -0.125 0.31178821 -1.01424201 0. 3 sigma scaling *tween: solvemttscum.090 0.21423648 241.74320302 239.141 -0.47665767 0.39738183 33.157 Table 8c: Compatibility test: add constraint on site SNP2 with scaling factor of 3 * t*dif.0002 -0.668 0.3390 33.058 0.30765412 0.185 -0.0000 0.35797001 0.42883385 32.07761282 34.243 0.306 0.93913660 0.39049791 0.38786845 -0.63577947 -0.272 0.148 -0.18070759 0.39849624 0.0897 34.25722227 241.23940957 0.010 0.13445555 241.678 34.4332 35.02188190 0.28018503 243.159 0.74767081 242.055 0.50180898 0.01086821 0.1409 34.12866179 0.495 -0.70620810 241.29150033 -0.03997369 0.20789753 1.01043949 0.56825584 -0.54119793 243.0001 33.128 32.93775455 0*3441 0.89374052 -0.115 0.40689957 -0.101 34.00831644 0.0748 34.52104141 240.4340 34.80120849 239.2433 35.010 0..2276 35.86777149 0.ma np2 time: 1993/ 6/18 18:56:46 * dif.138 0.3662 -1.43123129 -0.map.534 -0.3356 0.349 -0.55626281 -0.103 33.5017 -0.4187 34.9223 34. bet Compatibility test: using VG final combined sites + SNP2.6817 0.02486192 34.07170299 34.02949113 0.74376112 -0.127 0.010 0.54412173 0.23256085 0.2417 0.933 2.04022318 0.091 0.106 -0.39381344 239.33155577 -0.34599556 0.02052562 0.622 0.26970952 242.072 35.32567828 1.177 -0.20511069 243.24790731 -0.03986010 -0.16827399 238.83986590 -1.465 0.29833854 -0.99484105 0.04858753 0.11184743 242.2550 -0.3609 34.143 32.187 35.25139274 34.355 0.70130990 239. between: solvemscum.65731550 240.49282079 -0.266 0.3745 0.59514843241.637 0.07561848 0.026 0.2268 -0.69946154 240.snp2 * * symbol I means l~on 245.09346625 -2.20481771 0.03237065 0.57718401 243.map and .05818341 -0.49631810 0.064 0.17639476 0.05299823 0.79686014 244.06869857 34.308 0.02956983 0.04127723 0.405 3.33134321 241./comb/solvemcomb.160 0.15482160 -0.25499839 2.048 0.521 -0.38380469 239.00967397 0.20397400 0.04980978 0.12018709 0.comb.24708046 239.128 32.51558838 240.00121016 240.218 -0.33043485 0.44445764 0.08581136 0.56743609 0.2363 -0.81282355 0.02660506 34.135 0.68450200 1.3743 0.03297855 0..142 -0.72111933 243.10910137 0.56120081 -0.85459139 240.1188 34.71270209 33.123 0.5889 0.02136957 0.