Climatic Cycles and Tree-Growth/Chapter 1

by Andrew Ellicott Douglass

CLIMATIC CYCLES AND TREE-GROWTH.

---

I. INTRODUCTION.

The investigation described in the subsequent pages bears close relation to three sciences. It was approached by the author from the standpoint of astronomy and a desire to understand the variations of the sun. It was hoped that these variations could be more accurately studied by correlation with climatic phenomena. But the science of meteorology is still comparatively new and supplies us only with a few decades of records on which to base our conclusions. So botanical aid was sought in order to extend our knowledge of weather changes over hundreds and even thousands of years by making use of the dependence of the annual rings of trees in dry climates on the annual rainfall. If the relationship sought proves to be real, the rings in the trunks of trees give us not only a means of studying climatic changes through long periods of years, but perhaps also of tracing changes in solar activity during the same time. Thus astronomy, meteorology, and botany join in a study to which each contributes essential parts and from which, it is hoped, each may gain a small measure of benefit.

It is entirely natural that the yellow pine, Pinus ponderosa, common on the western Rockies, should have been the first tree studied, since it was an intimate and extensive acquaintance with the forest and with the climate of northern Arizona that led the writer to the thought of possible relation between the two. The climate had been sought for astronomical reasons because its limited rainfall of about 22 inches gave many clear nights and superb skies. The forest with its great extent and stately trees proved wonderfully attractive and the absence of undergrowth or of other species of trees was its most noticeable feature to anyone accustomed to moist climates. Evidently the absence of undergrowth was related to the dryness, and the critical problem of the tree was to survive periods of drought rather than to compete successfully with other species in the struggle to obtain food supply. The following argument, therefore, was naturally suggested: (1) the rings of trees measure the growth; (2) growth depends largely upon the amount of moisture, especially in a climate where the quantity of moisture is limited; (3) in such countries, therefore, the rings are likely to form a measure of precipitation. Relationship to temperature and other weather elements may be very important, but precipitation was thought to be the controlling factor in this region and for the sake of simplicity it is the element fundamentally considered throughout the present study.

In the very beginning of the work it was expected that only in large averages would a relationship be found between tree-growth and climate. Accordingly, something like 10,000 measures had been made on the pines of northern Arizona and the results all tabulated, when it occurred to the writer to compare the annual growth of Flagstaff trees directly with the 8 or 10 years of rainfall records taken at the United States Weather Bureau station recently established there. It was immediately seen that the accuracy with which tree-growth as shown in the rings may represent annual rainfall was far greater than anticipated. In a considerable number of cases, but especially in the dry climate groups, this has been found to be in the neighborhood of 70 per cent, which is raised substantially by applying a formula to allow for some degree of moisture conservation. At the present time, therefore, it is possible to lay a foundation for this study directly in the fact that the rings of trees form an approximate measure of the rainfall.

When the studies were carried to northern Europe an equal exactness in following the rainfall was not found, but a direct correlation was discovered between tree-growth and solar activity. Subsequent groups have been obtained from moist regions of the United States, and one is led to believe that this altered reaction is a question of precipitation and that it must be kept well in mind in any application of the methods hereafter described.

Since the beginning of this investigation, in 1901, assistance has been received from several sources which it is a pleasure to acknowledge at this time. Mr. T. A. Riordian, of Flagstaff, had 24 sections of the early Flagstaff group cut from the ends of logs and shipped to me. Mr. Willard P. Steel assisted in the measuring of the first 25 sections and a number of friends helped in the tabulation. Mr. C. H. Hinderer, of the United States Forest Service, at Prescott, Arizona, assisted in procuring the Prescott groups. Mr. H. S. Graves, Chief of the United States Forest Service, gave me several letters of introduction to foresters in Europe, by which I was greatly assisted in procuring the 9 European groups. I am glad to express my obligation to Dr. H. H. Jelstrup of Christiania, Professor Gunnar Schotte of Stockholm, Professor Dr. A. Schwappach of Eberswalde, and Professor A. Cieslar of Vienna, for especial aid in this connection. Assistance in completing the Vermont group was given by Mr. M. H. Douglass and others, and for aid in procuring the Oregon group I am glad to mention the excellent work of Mr. Robert H. Weitknecht, who was for a time connected with the United States Forest Service at Portland, Oregon. I am indebted to Mr. George A. Hume, of the Sanger Lumber Company, for important help in connection with the sequoia groups. In 1914 a grant of $200 was received from the Elizabeth Thompson Science Fund for study upon the correlation between tree-growth and solar variation. In 1918 a fund of $260 was placed at my disposal by the American Association for the Advancement of Science. This was for the purpose of extending the sequoia ring-record from 2,200 years in length (the result of preceding collection) to 3,000 years. This material was collected in the summer of 1918 and the measurements and tabulation finished soon after. I wish gratefully to acknowledge the courtesy of the editors of the Astrophysical Journal and the Bulletin of the American Geographical Society for permission to use illustrations and extracts from articles of mine which they have published. Plate 9 and figure 31 in the text are from the former journal. Thanks are also extended to Professor Ellsworth Huntington for the use of several text-figures which first appeared in my chapter of his work (1914).

TREES SUITABLE FOR CLIMATIC STUDY.

During the course of this investigation the wood and growth of numerous species of trees have been examined with reference to their adaptability to the purposes herein described. The collections visited include several in London, especially one in the South Kensington Museum, fossils in the Jermyn Street Museum, the lumber-yards of Messrs. W. W. Howard Bros. & Company, tree sections and fossils in the geological museum at Berlin, fossils in the lignite beds of Grube Ilsa near Dresden, and fossils chiefly in Munich and Vienna. In

Table 1.—List of trees in the Jessup collection whose rings were counted.

Scientific name.	Common name.	Locality.	Approximate center, A. D.	Possible periods.	Quality of ring sequence.
Pinus torreyana	Torrey pine	San Diego, Cal	1790	11 years
Pinus radiata	Monterey pine	Monterey, Cal	1855	22 years
Pinus monticola	Western white pine	Oregon	1641	11 years; Brückner
Pinus strobiformis	Mexican white pine	Southern Arizona	1706	22 years
Pinus strobus	White pine	Nova Scotia	1740?	22 years; Brückner	Fair
Pinus tæda	Loblolly pine	Florida	1731
Pinus echinata	Short-leaved pine	Missouri	1650
Tsuga heterophylla	Western hemlock	Canada northwest coast	1700	Uncertain
Tsuga caroliniana	Carolina hemlock	Carolina	1697	Poor. 22yr.; Brückner	Good
Tsuga canadensis	Canadian hemlock	Nova Scotia	1525	11 years; Brückner	Good
Pseudotsuga macrocarpa	Bigcone fir	Southern California	1480?	11 years	Good
Pseudotsuga mucronata	Douglas fir	Oregon	1315	11 years	Good
Picea sitchensis	Tideland spruce	Northwest coast	1798		Very good
Picea rubens	Red spruce	Nova Scotia	1610
Sequoia gigantea	Bigtree	California	550	Brückner	Good
Taxodium distichum	Bald cypress	Florida	1670?
Cupressus macnabiana	Macnab cypress	Northern California	1760		Very good
Toxylon pomiferum	Osage orange	Southern Arkansas	1765		Fair
Ulmus fulva	Slippery elm	Missouri	1770	20 years

America, collections were examined at the Smithsonian Institution in Washington, the horticultural exhibit at the Panama-Pacific Exposition in 1915, the museum at Chicago, but especially the Jessup collection in the American Museum of Natural History in New York City. Much careful counting of rings was done at the latter.^[1]

Considering all the trees examined, the conclusion was reached that the conifers, by the great regions they cover, the great variety of climates they endure, and especially by the prominence of their rings, seem best adapted to the purpose in hand. The chief trees, used with approximate number of rings measured in each, are: the yellow pine (Pinus ponderosa) about 14,000; Scotch pine (P. silvestris) about 9,000; hemlock (Tsuga canadensis) 2,500; Douglas fir (Pseudotsuga mucronata) 2,500; sequoia (Sequoia gigantea) 47,000.

INTRODUCTION TO SPECIAL STUDIES ON THE YELLOW PINE.

Before taking up the details of collection and measurement it is desirable to describe certain preliminary studies, such as those upon the yearly identity of the rings, time of the year of ring formation, and so forth. These studies were made chiefly upon the yellow pine of northern Arizona, but from the similarity between the pine and the other trees used it seems safe to say that the results apply equally to the Scotch pine, sequoia, hemlock and other species employed.

Location. — The yellow pines upon which the studies were made were obtained near Flagstaff, in the central part of northern Arizona, at an elevation of about 6,800 feet above the sea. The northern part of the State is largely a plateau forming the southern extension of the great Colorado Plateau. This high area is intersected some 65 miles north of Flagstaff by the Grand Canyon of the Colorado River. South of the town the high elevation extends 50 to 75 miles, varying only a few hundred feet from place to place, and then falls away abruptly at the "Rim." Oak Creek Canyon begins some 10 miles south of Flagstaff and flows to the south into the Verde River. The general drainage nearer town is gently to the northeast into the Little Colorado River some 40 miles away. Ten miles north of town the plateau culminates in the San Francisco Peaks, which reach an elevation of 12,700 feet. This mountain is a finely shaped volcanic mass with the old crater breaking away into a canyon toward the northeast. The town is in latitude 35° N. and longitude 113° W., and lies between two ancient lava streams 200 to 400 feet in height. It has a "wash" flowing through it from north to south, but this carries water only in time of severe storm or of rapidly melting snow.

	DOUGLASSPLATE 1
A
B

A. Bottomless pits near Flagstaff, illustrating drainage through limestone.

B. Yellow pine forest of northern Arizona.

The general country rock is Kaibab limestone in horizontal layers forming the plateau, surmounted by lavas over extensive areas near the mountain. The bedrock is covered by a thin soil, largely formed in place. The soil over the limestones is porous, while that over the lavas has much clay and holds water. There is no swampy ground and therefore no conservation of moisture from year to year. Consequently variations in moisture-supply are quickly felt by the trees. The pine forest is remarkable for the absence of other kinds of vegetation. It covers all parts of the plateau from about 5,000 feet in elevation to about 9,000. At the lower edge of the pine forest a belt of cedars, smaller than the pines and round in shape and with dark-green, thick foliage, makes an attractive landscape.

Climate and seasonal conditions.—The climate follows naturally from the latitude and altitude and the distance from the ocean. In the winters there may be from 1 to 6 feet of snow on the ground at one time. The storms are of the characteristic temperate-zone cyclonic types, but on account of the altitude the preliminary south or east winds are rarely observed. Storms come from the Pacific coast and rain occurs about a day later than in southern California. Spring and autumn are the dry seasons, and the warmest time of year is usually in June, just before the summer rains begin. The summer rains occur in July and August and often come in "spells" that last a week or two, with thunderstorms in the afternoons or at night, followed by clear mornings. Unlike the winter storms, the summer rains are local and apt to be torrential in character, with heavy run-off.

Meteorological records in northern Arizona are necessarily meager, yet not so deficient as might be expected. The country was first settled in the "fifties," when gold was discovered in Arizona as well as in California, and lines of travel were established from Santa Fe westward across the plateau. The "blazings" on the pine trees marking the earlier roads are still to be distinguished. Soon after the opening of the country the government located military camps at various places, and from that time records of rainfall and temperature were kept. The record at Whipple Barracks, near Prescott, begun in 1867, has been continued at Prescott to the present time. It is the longest consecutive record in the pine forest and is therefore used below.

The extreme range in temperature observed in Flagstaff is from about 20° F. below zero to about 100° F. above. But the town is in a peculiarly sheltered position and exhibits much lower night extremes than the "mesas" 200 to 400 feet above it. I have observed a difference of 26° F. between the top and bottom of the hill west of town at sunrise on a winter morning. During the early years of the Lowell Observatory, which is located on the mesa 350 feet above the town, the lower minima were about 5° F. These figures show the conditions to which the trees are subjected. The unobstructed topography of the plateau where the trees were collected is without doubt a very favorable feature. This leads to very similar conditions for the trees over many miles of country and doubtless greatly assisted in producing concordant tree-records. On the other hand, the San Francisco Peaks, 10 miles north of town, illustrate how meteorological data may vary in rugged localities. The west slopes of these mountains are exposed to the winter westerly storms; and have an immense snowfall. Springs abound and all favorable localities are taken up as ranches. East of the mountain, however, the land is dry and barren, and long distances intervene between watering-places.

In a very rugged country like that about Prescott similar differences between east and west mountain slopes must constantly occur. This is the reason of an early difficulty with the Prescott groups. Nearly 60 trees from various localities were measured before a group was found close enough to Prescott to be compared minutely with records of precipitation at that place.

1. The 17-foot section of sequoia was reviewed with some care and the dates on it checked. The dating is well done, as the errors are mostly under 15 years. The rings are large an do not show marked variations in width. Much repair work has been done on it, and the pieces of wood filling the drying cracks near the year 800 A. D. almost completely interrupt the continuity of the rings.