Andrew Ellicott Douglass and the Big Trees

Donald J McGraw. American Scientist. Volume 88, Issue 5. Sep/Oct 2000.

Historians and sociologists of science have long been interested in how scientists choose, and subsequently use, their tools, techniques and test organisms. As Adele Clark and Joan Fujimura put it, “[s]cientific work is constrained or enabled” by the tools available to the practicing scientist. Thus, a first order of business for study is to find and then use what Muriel Lederman and Richard Burian termed “the right organism for the job.” There is probably no better example of “the right organism for the job” than that of Drosophila, the fruit fly, in the development of genetics. Yet when Drosophila became more domesticated, more “constructed,” it took over the people who tended it: Its peculiar biology drove many of the great problems, and solutions, in genetics. When it failed to answer all questions for that nascent field, the need for-and switch to-Neurospora, the red bread mold, was born.

Neurospora became the second extremely significant organism used for the job of unraveling genetics. What the red bread mold did for that field-at a crucial juncture answer questions unanswerable by the prime tool for the earliest aspects of the work-the great red-barked sequoia would do for dendrochronology.

Dendrochronology, or tree-ring dating, was single-handedly created by Andrew Ellicott Douglass in the early 1900s. Born in 1867, Douglass was a brilliant student at Trinity College, where he took his degree in physics, geology and astronomy, with honors in all three. Douglass had no training in the biological sciences nor formal degrees beyond a bachelor’s, but neither of these would constrain his achievements.

After spending five years as an astronomer with the Harvard University Observatory, Douglass caught the attention of the amateur astronomer Percival Lowell, who sent him in the late 1890s to establish what is now known as Lowell Observatory near Flagstaff, Arizona. There he devoted himself to astronomical work and defending, if reluctantly, his patron’s views on Martian civilization. Eventually, however, the two had a falling-out over the subject, and not long thereafter Douglass settled as a faculty member and administrator at the University of Arizona in Tucson, where he would spend the next 56 years. In the early 1900s, Douglass had begun to create a distinct and very powerful new science. The climatological records in the rings of trees would, he hoped, be his entrepot to proving the possible relationship of the 11-year sunspot cycle to weather patterns on earth– his primary research interest.

Since his earliest introduction to astronomy, Douglass had sought evidence for the sunspot cycle. Known to the Chinese of two millennia before, sunspots had been suspected for some time of playing a role in climate control on earth. If so, could Douglass find a record of such control somewhere in the physical resources of this planet? Would such a record tie high sunspot activity with cooler weather on earth? Seeking to answer these questions, he discovered that as early as 1729 the French biologist Comte de Buffon had suggested that tree rings record local weather history. Douglass further explored this idea and eventually founded the new discipline of dendrochronology.

To begin formal research on tree rings, in 1906 Douglass sought financial support from the Carnegie Institution of Washington. Although his $500 request to make a trip to Australia to collect trees in an environment similar to that of Arizona was rejected, Douglass was not discouraged. He created a “working hypothesis” for this project that included the notion that tree rings reflect a tree’s food supply that the food supply reflects available moisture, especially if limited by drought (the arid nature of the Southwest was very much on his mind), and that, finally the rings must reflect this water supply. His first paper, published in 1909 in the journal Monthly Weather Review, noted this theory and described his early fieldwork in Flagstaff with the ponderosa pine, Pinus ponderosa.

By the time of this publication, Douglass had long since realized that wide tree rings indicated years of sufficient moisture for tree growth in the arid northern Arizona region and narrow ones drought years; this pattern thus formed a record of local weather history. He studied a great many stumps and cross-cut logs, committing to his prodigious memory the distinctive patterns of local weather fluctuations for many decades in and near Flagstaff.

This activity led to his establishing the central theoretical construct of the field, cross-dating, that would make dendrochronology the powerful tool that it has become. Simply put, a series of wood specimens can be arranged so that an overlap in any two specimens of a number of years will show correlatable patterns. These patterns allow one to push backward in time a growing local chronology of tree-ring widths. Cross-dating thus allows for moving from the present year (using a freshcut tree or core of a living specimen) backward as far as local wood sources (including those long dead) will permit.

Using his cross-dating technique, Douglass could piece together long chronologies. But he knew that if his discipline were to have broad relevance geographically, he would have to extend his efforts beyond Flagstaff. Australia, of course, had already been foreclosed for lack of funds. It was not until he was able to take a trip to comparatively distant Prescott, Arizona (67 road miles from Flagstaff) in 1911, that Douglass discovered tree-ring patterns in Prescott’s ponderosa pines identical to those of the Flagstaff region. This was crucial in demonstrating that the same weather pattern over many hundreds of square miles would be faithfully recorded in trees. (Of course, he would not have expected such correspondence with Australian weather patterns.) Douglass had, in his 1909 paper, discussed the rainfall data available at that time from a weather station in Prescott, but he had not seen the value of making a sampling trip to Prescott. When he did and when he compared the ring patterns, he was astonished at their fundamental similarity, writing, “The chief feature of the Prescott series [3,500 ring measurements] which places its results on a firmer basis than any previous work is the cross identification of rings between trees. The extent and accuracy of this identification came as a surprise to me.”

Although surprised, he was still only in an early period-that mist-filled dawning when the rings of ponderosa spoke so eloquently to him but did not reveal all the fundamental features of what was yet needed for a firm basis for his science. Nevertheless, the “Prescott paper” was his second on the subject and made the first firm step in stating his own realization that the discipline was much more than Buffon or the flamboyant character about to enter Douglass’s life, Ellsworth Huntington, would have foreseen.

Collaboration in Sequoia Country

Ellsworth Huntington, born in 1876, had by the first decade of the 20th century attained a certain amount of fame-later, many would say notoriety-for his theories on the role of climate on the history of human culture, indeed upon humankind’s full history across all the inhabited continents. A professor of geography at Yale University, Huntington became interested in investigating the Giant Sequoia of the Sierra Nevada, Sequoiadendron giganteum, for evidence to support his contentions. Huntington fell on sequoia as a possible long-lived source of weather history-not because of any astronomical interest, as would later be Douglass’s concern. Huntington was looking for similarities in weather patterns over vast areas, particularly in arid climes, and he wondered whether the Big Tree might aid him in further establishing his central thesis. Having sought out the sequoia during the latter part of a 1911 western expedition, Huntington was clearly aware of their great age. Douglass, apparently was not. He would discover their value only after Huntington had conveyed their wonders to him.

The astronomer and the geographer, after realizing they had similar interests, shared their discoveries and progress via frequent correspondence. In particular, Huntington was aware of Douglass’s efforts to study earth’s climate using tree rings and alerted Douglass that “the connecting link between the past and the present … is found in the Big Trees of California.” Douglass, who had begun to understand the value of long chronologies, would use the rings of the sequoia instead of ponderosa pine for all his further studies. Huntington further advised Douglass on these endeavors, writing, “… it looks as though the relation to rainfall might not be quite so simple as in Arizona.” “Arizona” referred, of course, to the ponderosa that Douglass had used exclusively to that time in developing dendrochronology. The sequoia’s use would “not be quite so simple” in Douglass’s work because Huntington had made a fundamental discovery about the nature of the annual rings in the Big Tree: They are not easily read.

Already Douglass had worked out the basic nature of “sensitive” rings, those easily read for their recording of the weather, versus “complacent” rings, those more challenging to read. Sensitive rings show nicely wide in moist years and narrow in drought years. Complacent rings, on the other hand, are often seen in trees that have sufficient water year round (virtually all those in the wet tropics, for instance) and in ponderosa specimens in arid America that are situated by a stream, for example. What others would later call “semi-complacent” rings were what Huntington had discovered in sequoia.

Douglass himself, however, had yet to see Big Tree rings firsthand. Although he realized that the semi-complacent nature of sequoia rings could be problematic in studying 11-year solar cycles where rings needed to be clearly interpreted year by year, Douglass had to put the sequoias on hold while he traveled to Germany and Scandinavia in 1912. On sabbatical to study the teaching methods used in physics by his German counterparts, Douglass also took advantage of the opportunity to gather ring specimens from a number of Old World tree species. He planned to compare the Northern European tree rings to ponderosa rings. Douglass reported his discoveries in Hunting– ton’s s great tome, The Climatic Factor, writing, “The relation between tree growth and sunspots here shown [by ponderosas in Flagstaff and Prescott) … does not stand alone. A series of measures on 13 tree sections from the forest of Eberswalde, near Berlin, Germany, the first of a number of series to be made on North European pine trees, discloses a striking time relation of the same character.” Douglass also presented, in mathematics he developed for the chapter presentation, the conclusion that “ring thicknesses are proportional to the rainfall with an accuracy of 70 to 82 percent.”

Meanwhile, Huntington continued to struggle with the semi-complacency issue presented by sequoia. He attempted to develop mathematical formulae based on retained moisture that would explain semi-complacency and prevailed on Douglass to aid him in supporting this contention. Douglass did comment in The Climatic Factor, that “only by taking a period of 3 or more years can we form an accurate judgment as to the actual amount of growth which corresponds to a given rainfall.” Yet no mathematical explanation concerning water holding capacity in sequoia root zones has ever been proved.

Establishing Theoretical Constructs

In a 1914 letter to Huntington, Douglass mentioned having developed the “periodograph,” an optical instrument that allowed him to visualize cyclical patterns in his ponderosa pine tree-ring data. Using this device, he commented, “has not been free from difficulties”; in fact, one of the rather arcane diagrams in a Douglass article was later recognized to have been printed upside down! Douglass spent much of his lifetime refining the device, which later matured into the “cyclograph.” The use of these machines became something of an obsession for Douglass; the value he perceived in the periodograms (usually produced as black-and-white photographs) was very great to his interest in solar cycles and (he believed) consequent climatic cycles. The true value of the instruments and the data they produced remains obscure.

With his new device, Douglass sought ever more specimens on which to do more ring studies. He planned to build a greater database, beginning with Huntington’s sequoias. Communicating with the geographer, Douglass solicited advice on how to reach the Big Trees and where to look for examples that might be most useful to study Huntington advised Douglass: “In getting specimens I think you will find the best results are obtained with trees growing in fairly moist places.”

Here lay a fundamental flaw in Huntington’s use of sequoia and one reason why he had said that reading rainfall data in the rings “may not be quite so simple as in Arizona”: Huntington had chosen trees that would bear rings that were hard to interpret because they were nearly completely complacent-for ecological reasons, primarily Thus, Huntington’s preference fox choosing specimen trees in “moist places” made his work approximate at best. It was some time before Douglass realized the crucial point that water-stressed trees would be inherently better for readable rings than well-watered specimens. Yet he was able to understand the deficiency in Huntington’s methods without criticizing heavily: “He was searching for general effects, and accuracy to a year or two was less essential … my chief aim is to get relative and periodic values … for him, the determination of the general curve, with an allowance for larger growth near the center [typical of sequoia], was most important.”

Douglass understood the importance of the specimen trees’ growing sites-irrespective of the inherent additional challenge of semi-complacency-and wrote to Huntington of his work on the sequoia samples, which he eventually collected personally. In a letter to the geographer dated June 17,1916, Douglass said, “I have cross-identified them all [and] … find two very distinct varieties. Those which grow in the uplands where the ground is steep, and the conservation of moisture very limited, or comparatively limited, show a beautiful and very consistent record of variation; whereas those which grow in the flats, where water can accumulate and remain, show what I call a very complacent or phlegmatic growth. The growth of these trees is larger than the others, and very nearly the same year after year. I find well individualized years, two or three in every decade. In the latter class, only two or three in a century, perhaps.”

These findings were of great importance to the establishment of one of dendrochronology’s central theoretical constructs: Seek out specimen trees that have been water-stressed, for they provide easy-to-read sensitive rings. But Douglass, who still had very little experience with a large number of tree species, would not realize just how problematic sequoia could be as a recorder of climatic history. Not until later was sequoia labeled as semi-complacent. In Douglass’s words, however, the semi-complacent nature of sequoia is obvious; his finding “only two or three [good rings] in a century” suggested this semi-complacency challenge. Eventually, single data sets encompassing decades, centuries or even more would be used in research. Because year-byyear rings in sequoia can vary so little, assessing past tree-growth dynamics using multiple years’ worth of rings is often more effective. Even so, Douglass, unlike Huntington, carefully accounted for individual annual rings.

The publication of the first books by the Carnegie Institution represented Douglass’s first statement to the scientific community in general concerning the nature of his new science. Unlike the 1909 article in the journal Monthly Weather Review, the 1919 work described a well-formulated, theoretically sound new science, worthy of serious employment by others. Douglass showed he had a firm grasp of the two main values of Sequoiadendron giganteum to his science: long chronologies and evidence of the importance of ecological settings necessary to avoid complacency. The central theoretical foundations of dendrochronology were thus laid by the time of Douglass’s publication of his initial use of the Giant Sequoia.

Dating Anasazi Ruins

The first real application of dendrochronology was archaeological. Employing sequoia data, in 1915-1916 Douglass attempted to date the ruins of the ancient Anasazi Indians in the Four Comers (Arizona, New Mexico, Utah and Colorado) area. Douglass turned to the distant California sequoias because only they could provide both long chronologies and dates from 4,000 to 2,000 years of age, which he hoped (in vain, as it turned out) would provide correlative climatic data suitable to date the ruins.

Of necessity, Douglass would seek progressively older wood samples from the Four Corners area to build a local long chronology, thus making it possible to date the ancient dwellings. He was optimistic about the correlation to sequoia, though, as he wrote to Huntington, “I hope, however, that with the measurements which I am now making on the sequoias from California I can get a comparison that will have a good deal of reliability. I believe I shall be able to tie up the modern trees to the sequoias by some relation of growth and then perhaps find a similar connection between the old pueblo trees [that is, roof beams] and the sequoias.”

Douglass “hope[d],” “believe[d]” and felt that “perhaps” he might find a usable correlation. Clearly, he did not feel particularly sanguine about the prospects for success. Over a number of years he continued to hope, but it became clear that the only truly reasonable way to gain an insight into the actual dates when the Anasazi were building in the Southwest was to do so via a locally generated chronology, rather than a long chronology using distant California sequoias.

Ultimately, the dilemma of when the Anasazi occupied the mesa country of the Four Corners area was solved employing the “bridging-the-gap” method. This strategy involved building local long chronologies for the Southwest until at some point the various “floating chronologies”—those that represented real roof-beam data from ruins but which had unknown beginning and ending calendrical dates-could be fixed in measured time. The discovery of the famous beam HH-39 on June 22, 1929 in Showlow, Arizona established the dating and changed forever American archaeology

The 1580 Problem

Once the dating of the Anasazi ruins was completed, Douglass returned to his sequoia samples only to encounter an uncertainty. Sure that the first set of sequoia samples from his first (1915) trip would “give a splendid record from about three hundred years before Christ,” Douglass instead found “in this period about six uncertainties. Only one of them seems to me to be serious, and that is for a ring which occurred in [A.D.] 1580 or 1581”—what came to be known as “the 1580 problem.”

Odd-appearing and missing rings around a.D. 1580 or 1581 suggested that a truly extreme drought year may have caused such poor growth that little or no new tissue was added at that juncture. Yet none of the sequoia samples then in Douglass’s possession could clarify whether he was off one year in his chronology or not, an impermissible situation for his 11-year cycle interests.

After securing more funds for research, Douglass returned to gather additional Big Tree specimens. By the end of his 1918 trip, the number of sequoias studied had increased to only 23, but the work on those was done with such accuracy with regard to cross-dating that he was able to extend the chronology by over a thousand years.

By his third trip, the 1580 problem was solved: His sequoia chronology was indeed off one full year, as he had intuited. The final renumbering was made after the 1918 trip, the purpose of which was settling the identity of the doubtful ring occasionally found around 1580 or 1581. “The existence of this ring was established and the necessary corrections … have been made. All subsequent comparisons have verified this identification.”

The solution is difficult to parse out of his sometimes challenging laboratory notes but can be seen by reading each careful reassessment. In the one or two sentences of note text for each tree (of the crucial rings for trees he labeled as D-1 through D-35) the solution appears. The term “1580A” was used commonly at that point to indicate that another ring (“A”) might be present. It was, of course. By adding it to the chronology, the entire dating was increased by one year. This problem is diagrammed in Figure 7, where the rings and corresponding calendar years are shown for Douglass’s specimen D-2.

While Douglass stated that the purpose of the 1918 trip was to “settle the identity of a doubtful ring,” he also went on to say that the “visit to the Big Trees in 1918 was for the purpose of procuring material so that the tree– record from the 2,200 years already secured [in 1915] could be extended to 3,000 years.” After testing about 50 trees, Douglass found a tree that he would name D-21. That, he said, “gives the oldest record … The oldest complete ring in good condition was identified as 1305 s.c.

Thus, the Big Tree was proving its value to the foundational needs of his new science; sequoia’s place had been assured. Douglass had established the importance of choosing specimen trees that are in water-stressed, not water-rich, environments in order to obtain good ring records. He had also found a species of tree, Sequoiadendron giganteum, that provided an ever-lengthening annual record of climatic history, precisely what he needed as he sought proof in nature of earth’s weather response to solar spot cycles. Extending that chronology would remain for decades to come an obsession with the astronomer.

Lord of the Rings

By the early 1920s, Douglass had attained more distinction in archaeology than in astronomy; according to George Ernest Webb, “his archaeological work during the 1920s provided Douglass’s most concrete contributions to American science.” Yet Douglass was, above all, an astronomer on a quest for the potential role of sunspot cycles in climate control; dendrochronology was created to investigate this primary problem. Although he succeeded in solving the single greatest conundrum in the history of Southwestern archaeology, a very distant science from his central concern, he was never exclusively committed to that or any other aspect of what dendrochronology might do. His interests were always cycles, and Douglass continued his quest until his death in 1962 at age 95.

And what of Douglass’s main pursuit? The effects of sunspots on earth’s weather as recorded in tree rings continue to fascinate scientists. An absolute answer with regard to connecting tree rings and possible sunspot-induced changes in earth’s weather, however, eluded Douglass, as it does current investigators. Nonetheless, in the spring of 1999, Shindell et al. argued that the change in solar radiance (sunspot cycle) is very slight-about 0.1 percent-but that the effects on the upper stratosphere could be transmitted to the lower troposphere via heating of ozone. The ozone may then transmit the solar-radiance variations to earth’s surface. The model has been both challenged and supported. Whether tree rings record this activity was not considered in the Shindell et al. study. However, just months ago Tammy Rittenour and Julie Brigham-Grette of the University of Massachusetts at Amherst and Michael Mann of the University of Virginia reported evidence of the double sunspot cycle (22 years) in the sediments of a now-vanished New England lake. The quest continues.

Although these recent studies-two examples among several lately-lid not involve tree rings, Douglass foresaw that many tree-ring studies would take place in his wake. He realized that his young discipline needed institutionalization, and so he established a laboratory at the University of Arizona expressly dedicated to dendrochronology Founded in 1937, the Laboratory of Tree-Ring Research (LTRR) was used heavily in his pursuit of cycles.

By 1945, Douglass could not have better stated the case for the use of the Big Tree as the “right organism for the job”: “The Giant Sequoia (Sequoia gigantea) [he used an earlier binomial] is an important climate-recording tree because of its immense life span, its ability to Live in scattered formation free from the influences of close grouping, its persistence in surviving deficiency of moisture and attacks by pests, and finally, because its ring-growth in sites of special character [well-drained areas] does show obvious relation to precipitation.” Most importantly, however, in the Giant Sequoia of the Sierra Nevada lay the answers to the early developmental history of tree-ring science.