U.S. Geological Survey Open-File Report 96-260

Scott: The account of your Sputnik interest and satellite observations fits right into the period you were discussing, the mid-1950s. That would have been sparked shortly after you had returned to California and embarked on the Nevada geologic mapping work.

Wallace: In 1956, Charles Anderson, who at the time was in charge of the mineral resources program, decided that I should work in Nevada next. I was unhappy about that because I had a really significant project staked out as an extension of the effort in the area of Coeur d'Alene, Idaho. But at the time, Anderson was in charge of the mineral resources program of the Survey, and as such was my boss. He had received his Ph.D. at the University of California at Berkeley, and taught there for several years. While at USGS he become noted for his geologic research, was elected to the National Academy of Sciences, and served as president of several national professional societies, including the Society of Economic Geologists. He served as Chief Geologist of the Survey, and was a strong research leader, as well as practical and familiar with industry.

My Idaho project had been given preliminary approval, but in the end never came to be. I planned to investigate what we call the Lewis and Clark Line (named by the famous economic geologist Paul Billingsley), across Washington, Idaho and Montana. The Osburn fault is a major element of the "line." The proposed project had important implications for both continental tectonics and mineral resources. But instead of pursuing that, I went to work on a geologic map for Nevada.

My Nevada mapping started in 1956. The program Anderson assigned me to in Nevada was being conducted cooperatively with the Nevada Bureau of Mines and Geology. The bureau's director was Vernon Scheid, a long-time friend from Pullman, Washington, days. He had been at the University of Idaho as professor and head of the Geology Department, and director of the Idaho Bureau of Mines and Geology.

Scott: Would you discuss the purposes of the Nevada project?

Wallace: The goal of the program was to prepare a geologic map of Nevada, for which there was only a very general, grossly incomplete and inadequate map at the time. Such state geologic maps assist the exploration for minerals and fuel resources, development of highways, and many industrial projects by private industry, and they form a data base and interpretation of the geologic structure of the region.

Scott: That sounds like a major undertaking. How did you go about it?

Wallace: The plan was to start from scratch by mapping geology at a scale of one-mile-to-one-inch (1:62,500 actually) of selected quadrangles. Unfortunately, there were very few existing topographic maps at that scale available for use as base maps on which to plot and analyze geologic findings. Topographic and geologic programs within the USGS had to be coordinated in order to proceed with the geologic work. Thanks to the USGS Topographic Division (now the National Mapping Division), we did get new topographic maps on schedule. Now, in 1995, it seems hard to remember the time when only a few topographic maps were available in Nevada.

Scott: Regarding the scale of 1:62,500, when you multiply 12 BY 5,280, you get 63,360. So there is a slight discrepancy.

Wallace: Yes, the so-called mile:inch map is actually 1:62,500 rather than 1:63,360. For decades the 1:63,360 scale was used and was stated on the maps, but then it was changed. I don't know what year the change to 1:62,500 occurred, but believe it was after I joined the USGS. The reason was that 1:62,500 is an even fraction of 1:125,000, 1:250,000 and 1:500,000, all scales that were becoming more and mored used by map makers.

In the summer of 1956 I started field studies and geologic mapping of the Humboldt Range in north central Nevada, with colleagues Porter Irwin and Norm Silberling. We started in the Buffalo Mountain quadrangle, which straddles the southern end of the Humboldt Range near Lovelock. (A quadrangle is a map 15 minutes of latitude and longitude on a side.)

Mines in the range had produced silver, gold, quicksilver, iron and some rather rare nonmetallic minerals such as dumortierite, which had been used in the manufacture of spark plugs. Classic localities for a hundred or more species of fossil ammonites were found in the range, and were of special interest to Silberling and to paleontologists around the world. The detailed geologic studies of the Humboldt Range were to serve as an anchor for a reconnaissance study of Pershing County, and preparation of a geologic map for the county. Our geologic studies eventually were to lead to geologic maps of each county in Nevada each published separately. These would then be merged--and, indeed, were merged--into a generalized geologic map of the state.

Wallace: In 1956 the idea had been that I would be responsible for the synthesis of the state geologic map, but other assignments came along to prevent this. Years later the state geologic map responsibility fell to Jack Stewart, who produced a superb geologic map of Nevada. Jack developed many other important syntheses and scientific findings based on information gathered in producing the map.

Scott: Explain a little about your reassignment.

Wallace: In 1960 I was reassigned to serve as Chief of the Branch of Regional Geology in the Southwest States, which included responsibilities for Nevada, Utah, and Arizona. Following common practice, I rotated out of that administrative assignment in 1964. With the work in the Humboldt Range and the wider ranging responsibilities with the Branch, I became deeply involved with and excited about the geology of the Basin and Range province as a whole. I still am. I noted earlier how one usually served as a branch chief for four or five years. I followed the pattern, and after that assignment, fate took me back to studying the San Andreas fault again.

Scott: Were you headquartered here in Menlo Park when working on the Nevada geologic mapping project in Nevada, and after you were made branch chief?

Wallace: Yes, we all were headquartered here in Menlo Park, California. The USGS was just opening up its headquarters here, and in 1956 I moved into a brand-new building. Following the pattern I have already described, the field work was done in Nevada, where we would spend three or four months in the field, and then bring all our rock samples back to Menlo Park to do microscopic work and chemical analyses, and do our report writing. As you recall, when we worked in Alaska, we would go to Washington, D.C., to do the analyses and writing. That had been the operational pattern since the USGS was founded in 1879.

Wallace: There was a great deal to do. It took a lot of original field examination, analysis and plotting of geologic relations on a map. That is why I have come to resist using the term "geologic mapping"--most people interpret "mapping" as a very routine operation. In fact, if geologic mapping is done properly, the process is a fundamental scientific effort and a demanding intellectual exercise.

Scott: What term do you prefer to use instead of "geologic mapping"?

Wallace: I like to say, "geologic analysis of a region." A geologic map can be described as presenting in graphic form a four-dimensional (including time, of course) interpretation of a part of the earth's crust, and a history of how it was created. It is far more complicated than a topographic map, on which rather straight-forward things such as roads, buildings and elevations are depicted.

I maintain that almost every rock outcrop looked at is, in a sense, a research project, and the synthesis of all the outcrop data is a major research undertaking. Multiple working hypotheses are employed at a given time, and trying to weave them together is a demanding intellectual exercise. Since the mid-1960s, for example, interpreting the meaning of an outcrop or a mapped area in the context of plate tectonics demands the best intellectual capabilities one has to offer. Any major new geologic concept that becomes generally accepted then tends to make older geologic maps obsolete.

For example, after the plate tectonic revolution of the late 1960s, one might look at a rock outcrop and say, "This rock type cannot have been created anywhere near that rock type across the valley, so a great conveyor belt of the plate tectonics system must have dragged it here to Nevada." Indeed, the geologic investigations--"mapping"--themselves account for major changes in ideas. The hundreds of kilometers of displacement along the San Andreas fault came from geologic mapping, as did the concept of microplates and terranes.

Scott: I can see that would be very demanding work, but also very rewarding. What were some of the geologic findings in Nevada that stand out in your mind?

Wallace: In Nevada, slabs of rocks many kilometers thick and hundreds of kilometers on a side have been moved eastward 100 km or more, to be telescoped and shoved together along great thrust faults. In 1956, when I first went to Nevada, the concept of such large-scale regional thrust faults was just coming into being, and plate tectonics had not yet been invented, or at least had not matured to a point where the majority of geologists accepted the idea.

In actuality, the concept of continental drift had been strongly argued for decades, and in 1958 S.W. Carey of Tasmania put together an impressive symposium volume on continental drift. (Carey, S.W. (Convenor), Continental Drift: A Symposium, University of Tasmania Symposium Series Proceedings, 1958.) Many of the concepts presented there are now accepted and merged under the name "plate tectonics". But the missing element in 1958 was proof of seafloor spreading, which gradually received major acceptability in the mid-to-late 1960s.

Some geologists doubted that the large thrusts were connected, or that these were relatively large regional phenomena. They continued to argue that thrust faults were rather local cases of smaller slabs or rock having slid off local mountain blocks. We were trying to verify or disprove the reality of such huge thrusts. The great Roberts Mountains thrust and the Sevier orogenic belt were of Paleozoic and Mesozoic origin respectively. Structures like that had characterized the crust of the earth tens of millions of years before the great block faults started to break up the crust to produce what we now call the Basin and Range structure.

Block-fault movement is such that where one mountain-sized block moved downward several kilometers, and the adjacent block moved relatively upward. Upward movement of a block led to formation of a mountain range, and downward motion produced a basin. All of this vertical action in Tertiary time, starting some 40 million or so years ago, was superimposed on the much earlier great thrust structures. The combination produced an extremely complicated geometric and geologic puzzle.

As the Nevada work proceeded, people began to realize that not all of the slab movements were due to thrusts from the rear or from depth. Some slabs 10 or 20 kilometers thick and a hundred km on a side seem to have slid eastward. That is still a puzzle--why did they slide toward the interior of the continent. Perhaps similar blocks also slide toward the ocean basin. That question is one I would like to answer.

Such were the problems. The possibilities, or working hypotheses, had to be kept in mind when considering every outcrop. The synthesis of the data, in part expressed and illustrated on the geologic map, produced the rationale for one interpretation or another. These were exciting times, prompting discussions and arguments whenever two or more geologists got together.

Scott: This background sets the stage very well for describing the kind of investigations you actually did in the field.

Wallace: I mentioned above my personal field activity in the Humboldt Range. In addition, being in charge of the branch, I also took time and made trips through Nevada, Utah and Arizona, and visited our staff that was working in each of these areas.

Scott: They were doing similar things to what you were doing in the Humboldt Range?

Wallace: Yes. I would visit them and we would have field discussions. We all knew about this big picture that we were trying to evolve, and we had wonderful arguments and debates.

In summer months, some of the people, such as Tom Nolan, who was then director of the USGS, worked out of Eureka, Nevada. He was marvelous in that by his own activities he demonstrated the Survey's interest in major geologic problems. In the old days, back to John Wesley Powell, Clarence King and others, they all got out in the field. Nolan was the last director really to do that while he was director. Each summer he ran the Survey by telephone from Eureka, Nevada.

Another who was out there was James Gilluly, who, without doubt, is seen as one of the two or three greatest geologists to work with the USGS. He was attracted to Nevada because of intriguing and difficult problems to be worked out, and a lot of his fame came from his interpretations there. Art Baker, associate director under Tom Nolan, was working in Utah on the big Charleston thrust. And, strangely enough, Nolan, Baker and Gilluly were listed as members of the branch I managed.

Scott: So despite their senior administrative responsibilities and titles, they were also working within a USGS branch doing geologic mapping?

Wallace: Yes. Of course, I really had no administrative control over them. I was really a servant to them. I like to think that the best operation of the Survey is in having the branch chiefs and office chiefs work as the servant of the research staff. Over the years that relationship was promoted by rotation: your field assistant today may be your boss tomorrow. Branch chiefing seldom lasted more than four years.

Scott: Yes, you mentioned rotation before, in connection with your tour of duty in Washington and your return to California. I see how that was very much a pattern in the USGS.

Wallace: Our daily activity in the late 1950s started by driving our government jeeps from Lovelock--later we stayed at Unionville--out to the work site. We would drive to the base of a mountain someplace and start climbing up a ridge. Each day, I would go up one ridge and my partner would go up another, climbing from an elevation of about 4,000 feet up to about 8,000 or 9,000 feet. We worked out a strategy so that we could effectively sample all the rock types and distribution of structures in the Humboldt Range.

We mapped about one-half a square mile a day, or maybe a full square mile. Typically a quadrangle like that would take from three to four years. During that time we would hike it, record the rocks, gather the samples, analyze them chemically, check the fossils as to age, species and so forth, draw the structural diagrams, and then write it up. We really put together a four-dimensional history of the block covered.

Scott: It was a pretty thorough investigation of the territory.

Wallace: It is thorough for its purposes, but, of course, there are various levels of detail in mapping. For "mine mapping," what we did would be considered only the broadest-brush investigation and wholly unacceptable.

Scott: It must have taken a lot of effort to cover that kind of area with that kind of care daily, especially considering the terrain. That was a lot of hiking!

Wallace: Yes, it was day-to-day hiking, looking, and collecting. When we say "fieldwork", we mean it. You are out in the field and in the mountains, hiking. The crest of the Humboldt Range is about 9,000 or 10,000 feet in elevation, and the base of the valleys is usually about 4,000 to 5,000 feet. So it was very rigorous work. We all loved it. You got to know the plants, soils, rocks, and geography intimately--better than almost anybody else, and all features related one to the other. I found it to be very satisfying. USGS is not, however, doing as much of that kind of geologic investigation anymore.

Funding for geologic mapping has decreased dramatically, for several reasons. For one thing, most areas in the U.S. now have some sort of a geologic map, inadequate as they may be. As a result, the original charge to the USGS in 1879 to make a geologic map of the country has almost been forgotten. Most of our earth science colleagues in geophysics and geochemistry, furthermore, have little concept of what a geologic map is or how it can be employed in interpreting the earth.

At present, mineral and energy resources are not perceived as being as critical as they were, for example, during World War II. At least the federal government is not seen as having a major responsibility. But, mark my word, with a limited planet and a burgeoning population, this can only be a temporary situation.

Scott: I heartily agree. You don't have to be an earth scientist to look at trend lines and realize that some crunches lie ahead.

Wallace: I noted that our pattern of operation was usually four months in the field in Nevada, and the rest of the year here in Menlo Park, writing, and getting chemical analyses and paleontological analyses and the like. The end result was a four-dimensional picture--the three dimensions of space, plus the geologic time dimension. In the 1950s Nevada was geologically unknown in many ways and in many of its regions. The idea of great thrust faults was being developed, and then the plate tectonic revolution was changing the way we all looked at the rocks and structures we saw. It was an exciting time and place to be doing that kind of investigation.

The earth's crust is not shaped quickly. I mentioned thrust-faulting periods hundreds of millions of years ago. The crust was fractured early, and those fractures are still there, forming the framework for the more recent faults that control the earthquake story. I have given papers on the "pre-fractured" nature of western North America and its relation to modern earthquake activity. The crust was fractured from early-on. From 200 million years ago on there were great sutures or ruptures that break again and again. Those ruptures set the stage for the later activity, although the stresses and strains change dramatically.

The thrust faulting is usually thought of as a compressional phase--although there is some question about that. For the last 30 to 40 million years, the Basin and Range province activity has been extensional--maybe as much as 100 percent extension. This is particularly true of about the last 14 or 15 million years up to the present--the period of block faulting that characterizes the Basin and Range.

I have long maintained that the Basin and Range story and that of the San Andreas fault are closely linked. Others are beginning to come around to that. In a recent seminar by Tom Brocher at the Survey, and Ben Page at Stanford, they spoke about the structures along the San Andreas fault being related to thrusting from the Basin and Range region. I like their ideas, but I believe that there is more to the story than simple "thrusting".

Scott: The acceptance of plate tectonics must have helped a lot with the new interpretations you have been talking about. As I understand it, before plate tectonics was understood, interpretations assumed that rather static conditions prevailed over very long periods of time.

Wallace: Yes, one school of thought saw things as basically static. But in the 1940s, being weaned on the San Andreas fault, I was indoctrinated by personally finding movements of tens of kilometers. In fact, my thesis mentioned possibly 75 miles (121 km) of displacement. That was extreme for the time, however, and I put it in the thesis on the basis of very fragmentary evidence.

A few years later in 1953 Mason Hill and Tom Dibblee came out with their paper projecting even greater displacement--hundreds of kilometers. Such ideas were considered pretty drastic, of course, and a great many people were not accepting it. Then came plate tectonics. In my mind plate tectonics was easy to accept after I had concluded that great faults could have hundreds of kilometers of slip. (Hill, M.L., and Dibblee, T.W., Jr., "San Andreas, Garlock, and Big Pine Faults, California: A Study of Character, History, and Tectonic Significance of Their Displacements," Geological Society of America Bulletin, v.64, n.4, 1953, pp.443-458.)

In the Soviet Union, some of the great geologists there were still thinking vertical tectonics. In their view, blocks could come up, shed sediment and fill up a basin, but they did not accept the idea of these great lateral movements. Moreover, when I was back at Northwestern, I had been taught that great thrust faults were impossible because rocks were not strong enough. But at Caltech my life in graduate school geology related to great earth movements.

There has been a extremely exciting transition in fundamental physical ideas about how the earth works. For example, when I was at Caltech up to 1942, the term "granitization" had never been mentioned in any of my classes. People like G. E. Goodspeed at the University of Washington, however, had seen that granite-like rocks were produced by metamorphic changes in sediment. No original molten mass was involved. What a controversy that caused!

Scott: The idea of long-distance movement along faults, and granitization, were new ideas broached at about that time?

Wallace: Yes. Those ideas were new in about the early 1940s and 1950s. There were tremendous transitions in thinking, and they are still going on, right now. It is one of the fascinating things about the field. I do get irritated when I see some very simplistic statements about how the earth works, and many of our earthquake hazard-reduction methods still rest on antiquated concepts of that. Nevertheless, we try to push ahead so that what is known is assimilated in the practical or applied end.

Scott: I understand that in recent years studies of faults have focussed on identifying prehistoric earthquakes and estimating slip rates.

Wallace: That is so. For me that meant that I turned to looking at faults in a different light than I had been doing in mineral programs and even differently than when looking at regional tectonics, although the new views do contribute in major ways to concepts of regional tectonics. Field techniques for these studies were very different than for the traditional geologic mapping that I had spent my life doing. I had to invent new ways to do the job, as did others.

Scott: Would you explain that a bit more?

Wallace: Identification of prehistoric earthquakes followed two general lines, one geomorphic, for example, analysis of ancient fault scarps generated during earthquakes, the other by the analysis of offset of sedimentary layers revealed in trenches dug for the purpose. Combinations of the two methods permitted long-term slip rates on faults to be determined. Short-term slip rates are determined by geodetic methods which can be meshed with the long-term rates. In any event, these kinds of studies not only have great theoretical value, but also can have some practical application in estimating the likelihood and size of earthquakes for engineering and planning purposes.

Scott: That is an interesting point. Would you say a little more about the practical applications?

Wallace: The dating of prehistoric earthquakes has shown, for example, that in the Basin and Range province major earthquakes along any one fault occur only at intervals of thousands of years. In contrast, along the San Andreas fault system the interval is measured in hundreds of years. Such differences can not be even recognized by instrumental records only a few decades long. The instrumental records, or even the longer historical records, do not bracket even a single cycle of the great earthquakes in the U.S. How can we possibly study long-term patterns with such a short sample?

As for slip rates, the idea is that elastic or brittle crust must be bent and warped (strained) before it breaks to produce an earthquake. Just as a rubber band must be stretched before it breaks. Slip rates measure that stretching, warping or bending. For a fault to break and create displacements of 5 meters, which could produce a magnitude 8 earthquake, the ground (crust) must first be bent 5 m. Geologic mapping of offset rock units can document the rate of offset over millions of years, and that slip rate can then be translated into how long it would take for a 5 m. bend to be produced, thus how long between great earthquakes.

Scott: Please describe how this has been used practically, in terms of public policy.

Wallace: I'll give two examples. Data on recurrence and slip rates were the fundamental data base that made possible the 1990 estimate of probability of large earthquakes in the San Francisco Bay Region, one of the first detailed analyses of a forecast (synonymous with long-term prediction). This analysis of probability underlies many plans, such as justifying the retrofitting of the Golden Gate Bridge, and adopting new values in building codes.

(Working Group on California Earthquake Probabilities, Probabilities of Large Earthquakes in the San Francisco Bay Region, U.S. Geological Survey Circular 1053, 1990.)

Second, the recognition of fault slip and the demonstrated ability to identify and map seismically active faults led to the enactment of California's Alquist-Priolo Act of 1973 which pertains to controlling building across active faults. That Act was actually the first legislative act based on fault slip that I know of.

Scott: Historic and instrumental records alone do not give a good idea of how often and where earthquakes will happen do they? Say a word or two about their inadequacies.

Wallace: While such data is of course extremely important, it is only a partial and incomplete record. The periods of time covered by instrumental records are not sufficiently long. Also, while historic information in some parts of China and a few other places in the world goes back pretty far, the incompleteness of such records emphasizes the inadequacy of such historical records and the need for paleoseismology.

Scott: Is that especially true for California?

Wallace: In California, our instrumental seismic records do not bracket even one cycle of great earthquakes, which are not frequent events along a given portion of most major faults. In contrast, the stratigraphic record along the San Andreas fault is fairly clear over a few thousand years. In Nevada we can clearly recognize fault-scarp evidence of earthquakes going back to 100,000 years. Dating of scarps that can be used to document earthquakes over the past 10,000 years is getting more and more precise. In Nevada we found evidence of dozens of individual big earthquakes having occurred in prehistoric time.

Through our cooperative programs with the Japanese, Chinese and Russians, interest in paleoseismology has spread around the world. The late 1970s saw the first trenches dug for paleoseismologic purposes in Japan and China.

Scott: Where did the term "paleoseismology" originate?

Wallace: Having worked in Nevada, I knew about the big faults where earthquakes had occurred in historic times--1915, l932, and l954. I decided that if I was going to work on the earthquake program as a geologist, I needed to get geology into the program more effectively. I knew I had something to contribute, but about 1965 I was not sure what would be a good course of action.

I picture these thoughts as leading to what we now call "paleoseismology." Against the strongly expressed advice of some of my colleagues, I started pushing the use of that term. Others did not favor introducing new terms, although that term had been used once or twice before. John Adams of Canada was one of the first to use it.

Scott: Did your plan work out?

Wallace: For one thing, I think use of the term contributed to one long-term goal, that is, seeing that geologists and geophysicists got together more effectively. Seismology was devoted mostly to instrumentation, yet, as I said earlier, the seismic record was far too short to determine the long-term pattern of big earthquakes along the San Andreas fault. Paleoseismic techniques could, and, indeed, have filled that need. Furthermore, the term "paleoseismology" legitimized the use of geologic techniques in the broad field of seismology.

Scott: You have mentioned Nevada several times in discussing paleoseismic work yourself. Is that where you got started?

Prehistoric Fault Scarps in Nevada

Wallace: Yes, I gravitated to Nevada because of my previous work there, and found that I could do paleoseismic analyses there rather easily. I could see not only the 1915 earthquake scarp, but also others along the same line that were more subdued. These were older prehistoric scarps for which there was no instrumental record. The 1915 scarp was superimposed on earlier scarps representing at least two earlier earthquakes. In an area of several 2-degree sheets of the topographic series, I could find evidence of a dozen or more large prehistoric earthquakes for which, of course, there were neither historical nor instrumental records.

Observations of the amount of degradation of the scarps permitted ballpark guesses of how old they were. It worked out that, even working alone, I could quickly measure the profile of a scarp. And from a comparison to the shoreline of glacial Lake Lahontan I could get an idea of how old the scarp might be. The last high stand of Lake Lahontan is known to be about 12,000 years old, and the wave-cut cliff produced at that time has since eroded to a certain slope angle and form. Using that slope and form as a calibration base then permits an estimation of the age of a fault scarp.

I tried several techniques of measuring scarp profiles and finally found that simply laying a fiberglass stadia rod on the scarp in a succession of end-to-end positions, measuring the slope of the rod at each position with an Abney level, gave excellent, reproducible results. It took me a couple of years to latch on to this method, but early-on, no fiberglass stadia rods were on the market. At first I was also not sure just which measurement would be the most valuable.

Scott: What did you find?

Wallace: Very soon I concluded that thousands of years separated the big earthquakes on individual fault segments. Also, the long-standing puzzle about repetition of displacement on narrow strands of a fault was immediately answered. There had been repeated movement on narrow fault strands in Nevada, just as I had found along the San Andreas fault at Wallace Creek.

One regional question I tried to answer was whether in the older record earthquakes had been migrating toward or away from the historic 1915 break, which resulted in a 7.6 earthquake. I wanted to see whether or not I could determine any long-term geographic pattern of strain release. Perhaps that would be a forecasting tool.

Scarp Degradation

Scott: You mentioned using evidence of scarp degradation to estimate the age of prehistoric earthquake. Say a few words about the factors causing degradation.

Wallace: The main agents of degradation are water, wind, and the action of biologic agents such as lichens and mosses. In Nevada I found that the most important agent seemed to be the freeze-thaw cycle--the freezing and thawing of the moisture in the soil. In Nevada approximately 130 times a year the ground goes through a freeze and thaw cycle. Those cycles make the soil swell and collapse and move downhill. Farther north the ground stays frozen longer, and farther south it does not freeze at all.

Scott: I take it that at first the scarp profiling gave only a rather crude approximation of a scarp's age? Did the method progress on from that?

Wallace: Yes. I teamed up with Tom Hanks and we wrote a paper employing a diffusion model to quantitatively describe profiles, and to get a more precise age calculation. Diffusion is a basic process in chemistry and physics, and determines the rate at which particles or concentrations move. Tom found that this could all be put into a formula of down-slope movement of material. My joint study with Tom Hanks started one day as we sat on a scarp along the San Andreas fault. Tom said: "Is that really a fault scarp? It looks like an error function to me." That floored me, but indeed the shape of an "error function curve" does nicely approach the profile of many fault scarps.

Scott: Would you briefly explain "error function curve" so non-geology readers can understand its relation to fault scarp profiles?

Wallace: A simple description of an "error function" for the nontechnical person is difficult. In graphic form, an error function appears as a curve that has a somewhat flatttened and subdued "S" shape which relates directly to the diffusion process. Thus, the error function curve is a convenient means of quantifying the shape of a fault scarp that has been subdued over many centuries by downslope movement (diffusion) of sand and rock particles. From that we can quantitatively deduce the age of the scarp and its related earthquakes.

We could apply some of the same formula to areas in China that had about the same climate as Nevada. I began comparing scarps in Nevada with those in Montana, China, Japan, and elsewhere to assess some of the variables related to climate.

Inasmuch as we have been talking primarily about the Nevada story here, I would prefer to develop the more general story of paleoseismology later, when we talk about the development of the USGS earthquake program. Many others played more important roles than I in the evolution of paleoseismology, and I want to mention them at an appropriate time.

Back to contents -- On to next section

U.S. Department of the Interior | U.S. Geological Survey
URL: https://pubs.usgs.gov/of/1996/of96-260/wallace-V.html
Page Contact Information: Michael Diggles
Page Last Modified: March 9, 2007