As the world drifted toward global war from 1938 through 1941, no one really knew the true nature of the fault. It was possible, some postulated, that it was a very old surface feature that for some reason had started to slip laterally just in the last two million years. Of course, everyone in the field knew that it had last slipped catastrophically in 1906, displacing the countryside (as well as fences and roads) in Marin County to the north of San Francisco by more than 5 meters (16.4 feet) at some points, and causing the great San Francisco earthquake in the process.
But how far had it slipped in historic times? And how deep did the rift in the earth go?
The fault trace (the visual surface evidence of it) ran from above Point Reyes north of San Francisco south through Daly City, down the peninsula, through the small farming community of San Juan Bautista, south through the tiny community of Parkfield (and twenty miles to the west of Coalinga), south-southeast through the barren Carrizo Plain, through the middle of the Tejon Pass north of San Fernando and Los Angeles, south of Palmdale, and on to the south-southeast, breaking into several subsidiary faults as it entered Mexico.
It was there, but what it could do, and when it might be expected to do it, were unknowns. Those were the questions which confronted Bob Wallace during the sultry summer of 1938, when he arrived, hot and sweaty, piloting an aging car along the margin of the Mojave Desert (where the fault trace of the San Andreas stands out with amazing clarity as a series of straight ridges running northwest to southeast), intent on finding answers to what seemed to be a purely geologic question.
Levi Noble, a geologist with the survey whose house literally sat on the fault, had produced some papers in the twenties that Wallace had read—papers which concluded that there could have been as much as twenty-four to twenty-six miles of total slippage along the San Andreas since the tear in California’s landscape first formed. The scientists at Berkeley’s School of Earth Sciences, however, steadfastly disagreed. Their position was that the fault had slipped no more than one mile at best.
Standing on the fault itself in nose-to-nose confrontation, Bob Wallace could see the evidence quite clearly: The offset stream and riverbeds and the lateral features and hills on the eastern side were now scores of miles to the southeast of where their counterparts resided. It seemed obvious, geologically, that the fault had slipped far more than even Noble had suggested, at least a hundred miles over millions of years.
It never occurred to Wallace to phone Berkeley and talk to the people who held the opposing view. Long-distance calls cost money he didn’t have (research money was nonexistent, and even the National Science Foundation was yet to be created), and that sort of thing just wasn’t done anyway. He didn’t know the people at Berkeley. To write or call might be unacceptably forward, and he certainly couldn’t afford to travel to Berkeley to discuss the matter. The geologists there could have just as easily been located in Mongolia. Communication about conflicting ideas and theories simply didn’t exist from 1938 to 1942—especially not at the behest of a shy, studious doctoral candidate who had no desire to breach protocol.
With the nation at war by 1942, Bob Wallace ended his four years of research before feeling he had truly finished the project. Four years of studying the San Andreas had not provided the hard evidence he needed to solve all the problems (and he had filled in some of the questions with answers of his own construct); but he had produced an important paper, and in the process he had managed to boldly suggest with elegant understatement that the total slippage on the San Andreas Fault might be as much as seventy-five miles (many years before others discovered not only that it had slipped seventy-five miles, but that the total distance is measured in the hundreds of miles).1
It had been twenty-three years before time, circumstance, and money had permitted Bob Wallace to get back to the San Andreas, but thanks to the effect of the Good Friday quake on the USGS’s research budget, 1965 found him once again on the trail (and in the Carrizo “desert”).
Experienced where he had been green in 1938, tempered now by decades of research, and carrying a fresh mandate to help answer the multitude of questions about earthquake risk in California, Wallace focused on documenting streambed offsets and trying to find ways of pinpointing slip rates at various sites along the fault.
It was geology contributing directly to seismology, providing basic information which might help seismologists construct a basic framework from which to answer the increasing drumbeat of questions coming from government leaders, insurance companies, businessmen, and citizens.
The problem, of course, was the simple fact that no one knew what had happened (if anything) along the San Andreas before the middle of the nineteenth century.
There had been only one great Fort Tejon quake (1857) and only one great San Francisco quake (1906) in the recorded human history of California, but that “history” was only a few hundred years old—a mere heartbeat in geologic time. Without a longer base line of historical knowledge with which to “bracket” major quakes at different locations along the fault, some other method was desperately needed to know when—and if—they had occurred. And though it still seemed rather bizarre in 1965 to the ordered analytical methods of seismological analysis, Bob Wallace, George Plafker, and numerous other geologists were beginning to understand that there might well be ways to look at the sedimentary deposits of the recent geologic past—the past thousand years or so—and find evidence of when and how badly the earth had shaken from fault movement. Perhaps, thought Wallace, the process could be developed into a science in itself, providing seismology with the much needed seismic history undreamed of by previous seismologists.
Of course, there was the bedrock recognition that while geologists were not trained to read seismograms with the sophistication of a Charles Richter, seismologists were not experienced in reading the surficial record—yesterday’s dirt in geologic time scales—and deriving vital information on where significant quakes had happened before, how they had affected the surrounding terrain, and how often they might recur.
The question Bob Wallace had worked on during 1965 and 1966 was very direct and not a little troubling: How dangerous is the San Andreas Fault?
And there was another question, always in the background, flowing inexorably from the realization that anyone who could emerge from a study of the fault with the answers to “when, where, and how much” would be offering the world (and California in particular) a genuine, unabashed earthquake prediction.
Was such a thing possible? The question of whether science would someday be able to predict with accuracy where and when damaging earthquakes would occur had been around for centuries. But finding a method of prediction would be akin to the discovery of dynamite—a breakthrough which would carry with it an entirely new and serious set of problems.
What would major population centers do in response to a prediction? What should they do? What should government leaders do or recommend? How do you prevent either apathy or panic among the general population? And, of course, what happens if you’re wrong, and a predicted earthquake does not occur? Such questions leaped from the neat confines of geophysics and seismology, the observational confines of geology, and the strictures of structural engineering (an art of the possible) and raced with mad abandon through the hitherto unexplored recesses of sociology, psychology, law, public policy, and economics. Such disciplines were not part of the required curriculum for a Ph.D. in seismology, yet the trail from one question to its logical conclusion would require the involvement of all such fields. If the presence of geologists at the doorstep of seismological research was startling, then the image of what stood in the shadows behind the smiling, helpful cadre of geologists (offering their emerging techniques of finding seismic history in the dirt) would be somewhere between upsetting and terrifying to the more insular members of the science. The confines of that neat scientific discipline were about to crumble, and though the best and the brightest of the field (such as Professor Charles Richter) had ne
ver tried to confine the science to the basement lab or insulate its implications and teachings from the real world, research seismology was about to be hauled, blinking and looking somewhat off-balance, into the messy area of real-time human existence to a far greater extent than even Richter had thought possible.2
Prediction was not science fiction. The question of whether it could be done was regarded as a reasonable possibility by the scientific community, given enough research time and sufficient research budgets within both the USGS and the academic communities. Perhaps only long-range warnings of earthquakes would be possible, or maybe a deep understanding of the quake mechanism itself would develop in the years and decades ahead, handing seismologists the key to effective prediction methods on a widespread basis. Possibly (as with so many other scientific breakthroughs in human history) a serendipitous fluke of a discovery would occur somewhere along the line of meticulous research. There was hope that in a few decades (or even years) seismologists would be able to predict earthquakes as regularly and effectively as meteorologists predict the weather. If that were to occur, people in general and Americans in particular could know when their buildings would be shaken to the limit, their freeways imperiled, their schoolchildren threatened by falling ceilings and bricks, and their employers put out of business. Perhaps they could be told with precision exactly how a given piece of real estate would behave in a future quake, as well as how large that quake would be, preventing municipal auditoriums and nuclear power plants from straddling active faults, and multistory apartment buildings from rising on alluvial Jell-O, poised to sink, collapse, or capsize in a moderate tremor.
Since earthquakes can cause such damage, there just must be a way to determine when, where, and to what extent they will occur. That holy grail might be years down the road, they knew, but it was an unspoken, common goal of the broad collection of seismologists and interested earth scientists such as Plafker and Wallace, who were increasingly convinced that perhaps, just perhaps, the answer lay in the recent geologic record, right beneath their feet.
And to an amazing extent, one of Wallace’s heroes had started that very process ninety-five years before in Utah.
Grove Karl Gilbert, one of the most admired geologists in the history of the American West, was best known for producing a classic study in 1907 of the great 1906 San Francisco quake with an attention to detail that had become legendary.3 But Gilbert’s inquisitive mind had married geologic evidence and questions of where earthquakes might occur as far back as 1872. He had participated in the Wheeler surveys of Utah, and recognized that certain scarps (long, relatively straight scars running horizontally, which he named piedmont scarps) along the base of the Wasatch Mountains near Salt Lake City indicated slippage related to earthquakes. For Gilbert’s brilliant mind, it was a short logical distance from that realization to the conclusion that the populace of the Salt Lake area needed to be informed of the risk. In 1883 he published a warning to the citizenry, pointing out that in places where no such scarps exist (such as Salt Lake City), yet where they should exist according to the reading of the geologic evidence in the mountains on either side, there was reason to believe that a damaging earthquake could be expected in the future. His warning—the first serious U.S. earthquake prediction and the very first paper on earthquakes by the fledgling U.S. Geological Survey—was a century before its time (although the predicted quake has not yet occurred). Equally important was the fact that his observations were based entirely on surficial evidence—what he could see in the terrain, evaluated by his “natural talent for deductive reasoning.”4
Gilbert had gone far beyond that propitious start in the decades which followed, but his most important contribution to seismology ended up effectively lost, buried in dusty stacks of his work, never cited in any significant paper until Bob Wallace discovered his amazing blueprint for earthquake prediction and hazard mitigation more than two-thirds of a century later.
In 1909, with San Francisco still rebuilding from the great quake of 1906, Gilbert had given an otherwise unremarkable keynote address to a January 1 meeting of the American Association of Geographers. Though already well known as a geologist and geographer, Gilbert wanted to pour forth a flood of ideas and insights into the problems of coping with, predicting, and preparing populations for damaging earthquakes. With the destruction of 1906 very fresh in his mind, and the power of the San Andreas Fault’s sudden slippage the subject of his insightful study for the previous three years, Gilbert had much to say. His speech, however, fell on deaf ears.
These were geographers. While they were respectful and interested, the words Gilbert lofted into the room held no greater meaning for them than the immediate effect of any intelligent speech from a distinguished scientist. Those words (and the paper on which they were based, “Earthquake Forecasting,” issued the same year), would remain effectively lost for the following seventy-one years. They spoke of the hope that someday the proper seismological tools and understanding would enable the seismologist to serve society the same way the meteorologist can, and they outlined the basic considerations for a major program to minimize—mitigate—the hazards of earthquakes to various populated areas. The areas Gilbert covered included a primer of earthquake prediction, earthquake engineering, land use, risk evaluation, and earthquake insurance. In the distant future, Bob Wallace would refer to the speech as the “most perceptive, encompassing, and balanced analysis” to come out until the 1960’s. But the tragedy was, of course, that those ideas and concepts and the blueprint to action were needed then and there—and they were ignored.
The years that followed would provide ample evidence of what could have been done, how many lives and dollars could have been saved, as quake after quake struck communities, not only in California but throughout the world, that were totally and abysmally unprepared to deal with the threat—or the aftermath. Earthquakes remained natural occurrences to which mankind could only react. The concept of advance preparation or planning was largely nonexistent—though the blueprint had been offered in 1909.
Bob Wallace watched the streaks of late-afternoon sunlight playing like an infinite row of dancing spotlights through the gently moving limbs of the eucalyptus trees that surround Menlo Park, wafting their pleasant and distinctive fragrance through the air and through his window to permeate his office with the scents of nature at her best. It had the feel of a university campus, but without the presence of frenzied students dashing back and forth to classes.
Not that students weren’t a part of the Menlo Park culture. Wallace’s office was just a short distance to the north of Stanford University—a mere bicycle ride away. It had become somewhat axiomatic that Stanford geology and seismology students naturally gravitated to the USGS “campus” for summer research jobs, or just to hang around the working cadre of the survey’s ranks of experts. It was like studying political science in the basement of the White House with full access to the President—an amazing continuous opportunity for intellectual exchange and ferment, the young students learning from the veteran scientists, and (as always) vice versa.
Campus was the right word for it, as well. Unlike any other agency of the United States government, the survey was run like a university, with senior scientists rotating in and out of administrative positions in large measure in accordance with their desires, and with information and planning exchanged in a freewheeling way which led some of the more organizationally oriented (some would call them bureaucratically calcified) members of the U.S. civil service to refer to their organizational structure at USGS-Menlo as “benign anarchy.”
But, it worked. While the loose organizational structure might create coronary instability in the heart of a senior civil service administrator from, say, the Treasury or Commerce Departments, the survey’s way of doing things produced results, creating an ongoing climate of dynamic support for intellectual curiosity that, when gently guided (and occasionally shoved), provided the nation with cutting-edge earth science technology and vital ongo
ing scientific leadership of university programs throughout the nation.
Money, of course, was a constant battle, even after the 1964 Good Friday quake. The mere mention of money—funding priorities—in a major project proposal could raise blood pressure, trigger tempers, and create infighting over scientific research turf by men and women who knew well that at any given time there was only so much of it available from either the taxpayers or various foundations. The monetary pie was very finite, and their task was to divide properly (and to protect their share of) that pie each year. The pie might get larger, as it had in 1964, but everyone’s share was still a matter of percentages.
This reality had already had an impact on the efforts to translate the growing excitement over earthquake potential and possible earthquake hazard mitigation into some sort of coherent program. Just after the catastrophe in Alaska, for instance, Dr. Frank Press had chaired a panel for the President’s Office of Science and Technology which produced a major report and recommendation for an ambitious ten-year program on earthquake prediction, a program which naturally would cost money.5 The report emphasized that no one was adequately studying the threat of earthquakes or what to do to minimize damage and deaths from earthquakes in the United States, a significant national mistake.
The engineering community, however, had been working on ways to plan structures to withstand earthquakes for many decades. It had amassed years of research and stacks of documents and suggested building codes to enable its practitioners to freeze the existing structural technology at any point and build the most resistant structure possible, given what they might know about the earthquake potential of the piece of land they were using.
On Shaky Ground Page 18