This webpage cites historical evidence pertaining to the condition of forests and other landscape features in the San Juans as they appeared prior to and at the dawn of Euro-American settlement. Consideration of the forests of the past can help us to understand our forests today and to project potential forest conditions in the future.
Yellow Island Change
(5 screens of text with 11 references)
TODAY'S LANDSCAPE IN San Juan County, as most everywhere in the world, is a product of both natural and human activities. Millennia of aboriginal occupancy in the islands (which was seasonal) must have affected the landcover, but the nature and extent of those influences remains unknown. Later, it was Euro-American year-round settlement with its accompanying ranching, farming and logging that initiated the principal alterations in land use and land cover that are with us today. This process started gradually in the 1870s, accelerated through the peak in homesteading (around 1885-1890), and advanced with varying amounts of serious logging (from 1890 to1930), the expansion of agriculture (1890 to 1950), and residential in-filling (commencing in the late 20th century). Even though tthe islands' underlying topography has changed little, the land-covering vegetation has been altered considerably.
The covering vegetation prior to settlement can be partially deciphered from historical accounts, although these descriptive accounts are limited. The early Spanish explorers and first traders left no descriptions whatsoever; it was the British expeditions of discovery that recorded the first really insightful observations. The very first descritions were made during Capt. George Vancouver's brief visit to the archipelago in 1792.
EARLY IN THE evening of June 8, 1792 Vancouver's naturalist Archibald Menzies joined Lt. Broughton in a small sailboat that probed westward from the anchorage at Cypress Island toward Blakely and Orcas Islands. These men had already formed preconceptions about the landscapes of the broader region because they had recently sailed across the top of the Olympic Peninsula, looped into Puget Sound, and then cruised along Whidbey Island and recorded those shorescapes for the first time. So, when they arrived in the San Juan Islands, however, Menzies immediately recognized that the archipelago was novel in terms of geomorphology and vegetation, relative to what they had recently encountered:
Vancouver seconded these observations a few days later in his own journal:
On June 18, 1892 Menzies re-entered the San Juans, this time to approach Matia Island from the mainland coast near present day Bellingham, and he again noted the conspicuous difference in landscapes:
"Straddling forest," "a little herbage," and "thinly coverd with stinted pines" are the earliest characterizations of the forests of the San Juans. Not exactly superlative remarks about our pre-settlement forests! Nearly two centuries would pass before the distinctive forest growth in the San Juans would begin to be understood as the combined result of thin soil underlain with impervious hardpan or bedrock, of uncommonly diminished rainfall especially in summer (when soil moisture is most needed for plant growth), of exposure to turbulent windstorms, and of centuries of occasional wildfires.
AFTER VANCOUVER AND Menzies visited in 1792, a half century would pass without further Euro-American contact. Then in 1853, soon after moving its trading headquartes to Fort Victoria on Vancouver Island, the British-licensed Hudson's Bay Company introduced commercial flocks of sheep onto San Juan Island. The company's objectives were three-fold: to exploit "prairie" forage that was easily visible from Victoria, to suppress civilian settlement, and also to stake a British land claim in the face of increasing territorial pressure from U.S. pioneers. Wherever they injected their commercial activities, Hudson's Bay Company agents diligently recorded descriptions of the local landscape into their trading journals.
Regarding trading operations on San Juan Island, Hudson's Bay journal narratives stressed their half dozen sheep stations and the daily doings of the handful of employed ranch hands. The sheep stations were all established on pre-existing open forage areas (variously described as meadows, prairies, and grasslands), thereby confirming that the islands were not thoroughly forested even in the mid-18th Century, contrary to common misconceptions that the islands were totally forested prior to settlement. (Other popular misconceptions today are that the early forests were "uniform," "magnificent" and "pristine.")
The largest sheep station on San Juan was Bellevue or Home Prairie overlooking present day Grandma's and Eagle Coves. It was described as "about two miles long by half a mile wide." (11, p. 137). Not far away, in the "basin" of the island now called San Juan Valley, Oak Prairie extended over 1 1/2 square miles (1000 acres); it was characterized by "groves of oak scattered over it" (Garry oaks still occur on the bordering rocky ridges). Oak Prairie was described as drained by a semi-permanent stream (San Juan Creek flowing into False Bay), and in low-lying places it collected standing water in wet winters. (3, pp. 35-38; 5, p. 137). Within a few decades settlers ditched this bottomland in order to reduce flooding, to advance spring planting, and thus to facilitate summer grain crops.
north, an open valley of significant size and "covered
with a luxuriant growth of grass" lay opposite Henry
Island where Mitchell Bay Road is today. (11, p. 137). On the
other side of the island, a few hundred acres of open land near
a protected bay was also used for grazing sheep; that bay would
soon be called Friday's Harbor, after the Hawaiian shepherd and
pioneer, Joe "Friday" Poalie, who was employed to manage
the station and who later settled on his own. (7). The exact
boundaries of Friday's open land are unknown today, but judging
by the pattern of early construction and in-town orchards, the
major part of modern Friday Harbor town, at least from First
Street inland, probably corresponds to his nearly-treeless grassland
(for an extended account of lanscape changes in that area see
"Lost & Found Prarie" in this website).
IN THE San Juans around 1860 was frequently described as damaged
and substandard compared to mainland timber, and these conditions
were customarily attributed to fire, although with little specific
justification. For example, in reference to San Juan Island,
"the lumber of the adjacent shores of Puget Sound is
superior to that of the island, the latter having all more or
less suffered from frequent conflagrations." (11, p.
137). As will be discussed presently, blaming the compromised
condition of the pre-settlement forests solely on fires may not
be justified. But first let us explore how forests of individual
islands were described in those early years, remembering that
the era was marked by a strongly utilitarian viewpoint.
Island was also largely forested, but at least in some places
"the timber is not the best quality as the Indians and
white men too, in search of deer have, from time to time, fired
the forest, thus greatly injuring the growth of trees. Doubtless,
hereafter, when the more desirable lumber of other localities,
especially on the adjacent shores of Puget Sound, has been somewhat
exhausted, mills will be erected on these beautiful shores."
(11, p. 140). The dream of water-powered sawmills never materialized,
with the negligible exception of Newhall's small, late-19th C
operation. (N.B. The remark about white men setting fires requires
some comment, if only because Euro-Americans were so few at the
time. Who were the supposed perpetrators of these fires? The
original hand-written field notes that were later sanitized for
publication refers to "Indians and sometimes barbarians
of lighter hue." (6, p. 52). The well-connected American
author would have known that a deer-hunting party, including
Orcas trapper, fisherman, and patriarch Louis Cayou, had been
sent to Deer Harbor by Hudson's Bay Company a year earlier and
was still active seasonally. (9). The author's florid mode of
expression appears to reflect a specific chauvinistic animosity
toward the British as well as an alarmist cultural bias against
fires in forestland.)
a boosterist newspaper account of that early era (1869), the
landcover on Orcas was again described derisively: "The
timber is of a scrubby growth with little or no underbrush, and
is well adapted for sheep grazing. It is well supplied with water,
there being a lake and small ponds on the highest part."
Island in the 1860s was recognized as having a low-lying western
half covered in forest except for a "small grassy prairie
containing about 100 acres" and a hilly eastern half
that contained "much good grass." (11, p. 138).
The forests of Blakely Island were described as "much
injured by frequent fires, and for this reason there are no inducements
for lumbermen." (11, p. 141); grass flourished on the
mountain slopes. On Decatur Island "there is much good
cedar timber, which growing in the low and moist lands, has escaped
the repeated fires which have swept through the forest."
(11, p. 141).
One hundred forty years ago (and persisting today in many circles) the determining roles of geology and local climate for forest growth were not well understood. However, in a moment of prescient insight, George Davidson, who was later a celebrated coastal surveyor and scientist, and who in 1855 visited the San Juans, observed: "The soil is scarce and poor, and very dry during the summer. The islands generally are covered with a thick growth of Oregon pine [meaning the drought-resistant Douglas-fir], other kinds of wood being exceptional." (4, p. 177). His comments along this vein reveal that he connected reduced forest growth with stressful environmental conditions.
TO CONCLUDE THIS overview, the pre-settlement forests of San Juan County were contemporaneosly described as stunted, sparse, and damaged by fire. A few decades later at the dawn of settlement the forests were described as highly varied and over-crowded (see Section II). These early reports are historically invaluable, yet they contain cultural biases and suppositions that are sometimes hard to reconcile with modern views. For example, in historical accounts fire is always portrayed as a regretable agent of unmitigated forest destruction; the implied assumption was that, without a long history of fires, local forests would have been much more majestic than they actually were. Furthermore, early observers occasionally expressed confidence that, once settlers arrived to prevent fires, bountiful forests comparable to those of the mainland would spring back and that a vibrant timber-cutting industry would soon follow.
optimistic expectations were (and remain) unrealistic, because
forest growth in the San Juans is intrinsically much reduced
and retaded compared to mainland forests. It is more appropriate
to realize that slow forest growth in the San Juans is governed
by inherent and unremitting ecological constraints. Furthermore,
in the absence of enlightened forest management and truly judicious
tree cutting, frequent and low-intensity underburning could have
actually benefited forest health and stocking density.
Sometimes such fires in pre-settlement times were aided by local
native people, but even then the practice was likely very limited
in scope and timing. Over millennia, however, wildfires surely
occurred and influenced our forests.
Although early reporters noticed "deficiencies" in the forests of the San Juans, they probably missed the main contributing factors because they lacked sophisticated ecological knowledge. They could not have known that forest productivity in the archipelago is inherently low, regardless of the fire regime, due to negatively reinforcing limitations of soil fertility, poor winter drainage, severe summer drought, shallow rooting depth, destructive winds, and even aerosol sea-salt. Modern insights into the ecology of forest growth and the consequences of various disturbance regimes allow us to reconsider with somewhat greater clarity the conclusions and perspectives of earlier commentators. Their unedited comments, however, invite the imagination to dip backward into local environmental history and to develop an open-minded quest for factual information before leaping to unsubstantiated presumptions.
1. Agee, James K. (1984). Historic landscapes of San Juan Island National Historical Park. CPSU/UW 84-2. On file in San Juan Island National Historical Park, American Camp Visitor's Center.
2. Agee, James K. and Peter Dunwiddie (1984). Recent forest development on Yellow Island, San Juan County, WA. Can. J. Bot. 62, 2074-2080.
3. Custer, Henry (1859). "Report [to the U.S. Boundary Commission] of Henry Custer, Assistant, of a reconnaissance of San Juan Island, and the Saturna Group. April 11, 1859. On file in San Juan National Historical Park, American Camp Visitor's Center.
4. Davidson, George (1855). Report to the Superintendent, U.S. Coast and Geodetic Survey. Appendix No. 26.
5. Gibbs, George (1859). "Report of George Gibbs, Geologist, of a Geological Reconnaissance of Islands of the Haro Archipelago." Transcription of original in National Archives in San Juan Island National Historical Park, American Camp Visitor's Center.
6. Kennerly, C.B.R. (1860). "Report [to the U.S. Boundary Commission] of Dr. C.B.R. Kennerly, Surgeon and Naturalist, of a reconnaissance of San Juan Island, and the Saturna Group. April 11, 1859." On file in San Juan Island National Historical Park, American Camp Visitor's Center.
7. Koppel, T. (1995) Kanaka: The Untold Story of the Hawaiian Pioneers in British Columbia and the Pacific Northwest.Whitecap Books, Vancouver, B.C.
8. Menzies, Archibald (1792). Menzies' Journal of Vancouver's Voyage. Edited by C.F. Newcombe. Provincial Archives of British Columbia. 5 (1923).
9. Richardson, David (1971). Pig War Islands: The San Juans of Northwest Washington. Orcas Publishing Co., Eastsound, WA
10. Territorial Republican, May 17, 1869, p. 1.
11. U.S. Congress (1867). "The Island of San Juan." Executive Document No. 29. Senate, 40th Congress, 2nd Session. Report of the Secretary of State. Washington, DC.
12. Vancouver, George (1798). Voyage of Discovery to the North Pacific Ocean and Round the World (1790-1796). Reprinted 1967, Da Capo Press, NY. Vol. 1.
OF SETTLEMENT - 1874
(9 screens of text)
The first thorough survey in San Juan County was conducted by the federal General Land Office (today's Bureau of Land Management) in the autumn of 1874, which was a mere two years after the international boundary was resolved and before any significant settlement. Though by no means free from surveying errors, the 1874 GLO survey identified all townships and sections (36 and 1 square miles, respectively). The GLO surveyed section and township lines (1 and 6 miles apart, respectively), monumented section corners, measured and recorded witness trees, and commented upon the character and "useful" potential of the land and its vegetation cover.
T.M. Reed, J.T. Sheets and J.M. Whitworth were the contract surveyors who produced the 1874 GLO survey of the San Juans. After surveying a township, these professional s described its landscape in brief narrative form. Conforming to the utilitarian directives underlying the rectangular survey, the surveyors summarized each township in terms of agricultural and pastoral potential, and exploitable timber and mineral resources. Upland forestland in the San Juans is most often described as range land suitable for pasturing sheep and goats, which indicates a certain openness, as compared to dense forest.
References to the islands' forests are brief and rarely complimentary. The most frequently repeated terms are "young fir thicket" and "thicket of young pine and fir." "Open timber" is rarely mentioned, and hardly ever did the surveyors gush with enthusiasm. The descriptive narratives were necessarily quite superficial but they do reveal an attempt at accuracy and objectivity. The surveyors measured, pinpointed and identified more than 2000 witness trees in the GLO survey, and those data are presented, in part, in Section 3 of this chapter.
The GLO field notes of 1874 confirm the existence of numerous lowland open areas within a generally forested matrix. By that date pre-existing openings were already largely claimed by settling farmers and were fenced; the settlers did not create such clearings by actively cutting away the forest, they were already in existence. Many of the openings were identified as "swamps" or other wetlands but others were cited as prairies, pastures or fields. Some of these areas coincided with low areas and others were on bench lands and hillsides. Between the 1860s and 1874 limited settlement was becoming evident, but time of occupancy was brief and the population was still very small (there were only 93 men in the entire county in 1870). So few Euro-Americans would not have produced profound changes upon the forested landscape by 1874, when the GLO survey wrote their desciptions, especially since the settlers were primarily preoccupied with farming and ranching the pre-existing openings, rather than logging (an arduous activity that had not yet acquired any commercial reward). Although small-scale tree cutting had begun in order to produce such items as locally used fuel wood, fence rails, shakes, and poles, no appreciable logging had yet begun.The GLO survey of 1874 thus predates logging in the San Juans by about 20 years, and therefore the following passages are impressions of the pre-settlement land-cover conditions in the San Juans before much had been changed as a result of settlement.
township may be said to be chiefly the top, sides and base of
Mt. Constitution, which is 2400 ft. in height. There are many
settlers living along the beach and good gardens and nurseries,
the fruit being very fine.
[The summary for this township is missing from the record. However, the GLO field notes contain more localized summaries for each mile-long section boundary. These substitute summaries are reproduced below; "E of 14" signifies the eastern boundary of section 14 in T35NR3W (which is also W of 13, of course). Due to the shoreline, portions of only 10 sections are included in this township.]
Township 35 North, Range 2 West (N. Lopez Island and Pear Pt./Turn Pt., San Juan Island)
Township 35 North, Range 2 West (S. Shaw Island)
Township 36 North, Range 2 West (Shaw Island)
Township 37North, Ranges 2/3 West
Townships 35/36 North, Range 1 West (Blakely Island)
Township 35 North, Range 1 West (Decatur Island)
Township 35 North, Range 1 West (James Island)
Township 35 North, Range 1 West (Center &Trump Island)
Township 35 North, Range 1 West (Frost Island)
Township 36North, Range 4 West (Guss Island)
Township 36 North, Range 3 West (Cliff aka Brush Island)
Township 36 North, Range 2 West (Bell Island)
Township 36 North, Range 2 West (Sheep aka Picnic Island)
Township 36 North, Range 2 West (Double Island)
Township 38 North, Range 2 West (Sucia Islands)
Township 36 North, Range 3 West (Spieden Island)
Township 36 North, Range 3 West (Flat Top Island)
Township 34 North, Range 2 West (Long Island)
Township 35 North, Range 2 West (Turn Island)
Township 37 North, Range 4 West (Cemetery and George Islands)
Township 37 North, Range 4 West (Johns Island)
Township 37 North, Range 4 West (Stuart Island)
Township 37 North, Range 4 West (James aka Satellite Island)
Township 36 North, Range 4 West (Sentinel Island)
Township 36 North, Range 4 West (Small and Cactus Islands
Township 36 North, Range 2 West (Crane Island)
"The shores of the Island are very rocky & broken. Timber fir, cedar, alder, laurel, and underbrush. Some level land in center of island. Land second rate, fir scrubby. Island not enhanced in value by location near to any town or village, Friday Harbor being the nearest village of about 30 inhabitants about 6 miles distant. Walter Cadwell has a log house on the northeastern end of the Island with about 1/4 of an acre fenced in around the house and about 4 acres an 1/8 of a mile Southwest of the house in cultivation. Value of improvements about three hundred dollars." E. Vongohren, March 7, 1884. GLO Field Notes, p. 244.
General Land Office (1874). "Surveyor's Field Notes of San Juan County, Washington." Typed transcript (c. 1970), Dept. of Public Works, San Juan County. Friday Harbor, WA. Manuscript versions with different pagination are also available at Public Works and Bureau of Land Management offices in Portland, OR and Washington, DC.
Section 3. FOREST QUANTITATION - 1874 vs 1990
(3 screens of text with 3 charts)
For a variety of reasons, forests ineluctably change with the passage of time. Some processes of change are natural: growth, crowding, windthrow, disease, succession. Other processes are more clearly caused by people: tree cutting, root excavation, soil compaction, species rearrangement. Still other agents of change, such as grazing of saplings and brush by livestock and burning initiated by people, lie somewhere between the purely natural and the purely anthropogenic because complex disturbances such as fire and grazers can be so unpredictable and the changes so unintended.
A quantitative comparison of tree species, sizes, numbers, and density at two points in time offers a certain insight into forests, but it devalues (by ignoring) what may be more interesting, namely the causal factors of forest change. Nevertheless, superficial comparisons of two obscure data bases of trees in San Juan County allow a kind of "then and now" comparison spanning more than 100 years. The data bases themselves are different in character (and therefore of limited "scientific" merit, but they are the best that are available.
The first data base, from 1874, includes over 2000 trees that were systematically sampled on a scientific grid covering the entire county. The data are in the form of witness trees in the first land survey of the San Juans, as conducted by the General Land Office (GLO, which is today's Bureau of Land Management). Because of the early date - only two years after resolution of the boundary dispute or so-called Pig War - these tree data are useful for characterizing the forests at the dawn of EuroAmerican settlement. They were collected prior to any appreciable logging - the total county population in 1870 was only 93 men and the dominant occupation was farming, not logging - and thus the data significantly reflect the late pre-settlement condition. (The earlier, prehistoric condition of the forests remains uncertain; the fairly heavy seasonal residence by native people must have had profound effects upon the local vegetation, but the details are poorly understood; the presence of these people and their effects upon the forests declined abruptly in the 1830s through 1850s as the societies were devastatingly depopulated and disrupted.)
The second data base derives from a scientific inventory performed by the US Forest Service in 1990. It is based on a small sample of only 600 trees. Although too limited for formal statistics, it allows some objective grounds for comparison with the earlier GLO data.
This brief report compares the 1874 vs 1990 data and addresses the following questions:
The present analysis omits actual numbers of trees and therefore also tree densities at the two dates. Such density analysis is possible but it has not yet been carried out in full detail; suffice it to say, the preliminary conclusion is that at maturity pre-settlement forests were stocked with fewer but larger trees than occur in today's forests. Even so, forest stands were very diverse in character. Broadly speaking, there may be as much wood in today's forests as in 1874, but the average trees today are about half the diameter and much closer together so that today there may be about three times as many trees as before settlement.
The chart above shows tree frequency by species in 1874 and in 1990. Long ago, just as today, Douglas-fir stems were vastly more abundant than stems of other species. Comparison of the data shows that the pre-settlement forests had somewhat higher proportions of red alder and pine than now and somewhat lower proportions of grand fir, redcedar, and Douglas-fir.
The chart below shows that conifers in the pre-settlement forests were somewhat larger in diameter in 1874 than they are now. The difference is most apparent in the commercially valuable species at the far right. Conversely, decidous or broadleaf trees are larger today. Bigleaf maples are notably wider now than they were 116 years ago.
"Relative dominance" in San Juan County's forests is depicted below. This chart demonstrates that the Douglas-fir is - and was also in former times - the overwhelmingly dominant species. Douglas-fir accounts for three-fourths or more of the wood in our forests; this is because it is both quite large and clearly the most abundant tree. From 1874 to 1990 this important species declined in dominance from 86% to 69%, probably as a consequence of being high graded for its commercial value. Grand fir, redcedar and bigleaf maple have increased in dominance.
In conclusion, these brief comparisons begin to illustrate some of the changes that have occurred in the composition of our forests over a period of 116 years. Growth, cutting, natural disturbances, clearing and forest grazing in the past have all impacted our forests and helped to create what they are today.
Return to fOREST iNFO Home