Environment, Prehistory & Archaeology of Mount Rainier National Park, Washington
Greg C. Burtchard

Chapter 2:
ENVIRONMENT AND LAND-USE ON MOUNT RAINIER & IN THE SOUTHERN WASHINGTON CASCADES
(continued)

HOLOCENE EVOLUTION OF MOUNT RAINIER'S ENVIRONMENT

The central premise underlying environmental and land-use patterns discussed above holds that critical plant and animal resources are not evenly distributed but rather tend to cluster seasonally in 1) upper-elevation subalpine to alpine habitats, and 2) early fire succession habitats in mid-elevation forests. Assuming that these plant and animal resources are the primary attractors drawing lowland populations to the mountain, it is reasonable to expect human use to have always favored these places over other less productive forest zones. Because of their spatial stability (relative to burn areas) and because they optimize access to the widest range of exploitable resources, subalpine to alpine habitats are expected to have been particularly heavily used during the prehistoric past.

Vegetation boundaries are not perfectly constant over periods. There is little doubt that Holocene climatic changes altered high elevation forest patterns, affected resource productivity, and may have altered human land-use strategies on Mount Rainier. At a general level, there is ample reason to believe most of the northern hemisphere experienced general climatic warming following final retreat of Fraser/Wisconsin glacial epoch about 9,500 years ago. The trend persisted until about 4,500 to 4,000 years ago, then reversed, with generally cooler conditions to the present.

Even though this pattern substantially simplifies actual Holocene climatic oscillations there is reason to expect climatically induced variation in forest and resource structure during the period of time that humans have inhabited the Mount Rainier region. Without active human intervention, periods of sustained environmental warming could be expected to be accompanied by forest advance in upper elevation environments, effectively reducing the area of economically productive subalpine habitat. Alternatively, delayed snowpack melt associated with sustained colder periods would have repressed montane forest cover, ostensibly lowering forest maturity in upper elevation settings, thereby improving ungulate forage and huckleberry habitat. [8]

This section assembles available data to model the late Pleistocene and Holocene climate sequence for the Mount Rainier region. Sources include plant macrofossil and pollen data and interpretive summaries from Mount Rainier, the Cascades and the surrounding region (especially Dunwiddie 1986, Heusser 1977; Whitlock 1992; and Sea and Whitlock 1995); and synchronous North Cascade and Mount Rainier glacial advance-retreat patterns (Crandell and Miller 1974; Porter 1976). Climatic implications of these data are summarized in Table 2.3. The curve is adapted from Heusser's (1977) inferences for the Pacific Slope of Washington as modified by pollen and geological data. The modified Heusser curve is a useful means of modeling climatic change because it displays greater variability and provides a better notion of climatic process than static time-block sequences (see Sea and Witlock 1995:378 for a good summary of the latter). Even though the curve simplifies a more complex natural reality by smoothing shorter-term oscillations, it can be refined as additional data become available and offers a more precise mechanism for comparing human system responses, if any.

The climate curve in Table 2.3 estimates long-term changes in summer temperature based on relative frequency of western hemlock (Tsuga heterophylla) to Douglas fir (Pseudotsuga menziesii) in 300 pollen spectra from the Hoh River Valley on the western Olympic Peninsula and Salmon Springs near Puyallup in the Puget Trough (Heusser 1977), as modified by macrofossil and glacial data noted above. Changes are expressed in degrees Celsius relative to present average July temperature at sea level (zero on the scale). The vertical guides are set arbitrarily at 1° C above and below the modern July mean. This zone approximates historically experienced (i.e., modern) climatic circumstances. Portions of the curve falling to the left (i.e., colder) of the -1° C line are classified by reference to their most commonly applied regional glacial or interglacial nomenclature. Sections falling within the ±1° guidelines are classed as modern interludes. The single significant part of the curve falling to the right of the +1° line (i.e., warmer) is the mid-Holocene xerothermic period labeled here the Hypsithermal Interval.

Table 2.3 Mount Rainier Region Late Pleistocene - Holocene Climatic Sequence

It now appears that Mount Rainier landscapes became sufficiently glacier-free, with adequate time for floral and faunal dispersal processes to have established economically useful plant and animal habitats by about 8,500 years ago. Prior to that time, glacial ice mass would have effectively precluded human use of Mount Rainier proper; although the Puget lowlands, eastern foothills and larger intermontane valleys could have sustained exploitable species as early as the Everson interglacial circa 13,500 years ago (the Pacific coast and southern lowlands earlier still). Assuming, for the moment, that humans did not intervene to alter natural successional processes, basic climate/environmental patterns on Mount Rainier and in the Pacific Northwest approximated the basic pattern shown in Table 2.3. The periods summarized below represent units of time during which the curve extended above or below 1° C of the modern July average. Please keep in mind that these periods are difficult to identify with precision and contain substantial climatic variability not represented in the curve's smoothed contours.

Late Fraser Glaciation: Vashon Stade; circa 17,000 to 13,500 B.P.

The Cordilleran ice sheet reached its maximum development in the Pacific Northwest between about 15,000 to 13,500 years ago (Whitlock 1992:9; Porter 1976:73; cf., Waitt and Thorson 1983). According to these sources, glacial ice covered the central and northern Puget Trough and the Straits of Juan de Fuca. East of the Cascades, Cordilleran ice lobes reached their most southerly extension. Alpine glaciers, however, appear to have been less extensive than earlier Fraser advances. Even so, Mount Rainier for all practical purposes was either ice shrouded or barren. Ice free areas on the mountain and in its immediate vicinity would have appeared essentially as rock or rock-grass islands impoverished of floral and faunal species in the midst of predominantly ice-mantled landscapes.

South of the glacial ice, vegetation took on a more mesic character compared with the preceding several thousand years. Probably responding to increased winter moisture that fed the Vashon stade, low elevation vegetation on the Washington coast, in the southern Puget Trough, and in lowland southwestern Washington shifted from predominantly parkland/tundra to parkland/open forest associations (Whitlock 1992:13-14). East of the Cascades and south of the glacial ice mass, periglacial steppe persisted to an elevation of about 1,500 m (4,900 ft).

Clearly, Mount Rainier, central and northern Puget Trough, Olympic Mountains, and North Cascades could not support human use during Vashon times. Even though plant and animal communities south of the glacial ice could potentially have supported human settlement, it is unlikely that anyone was there to avail themselves of them. At this time, most North American human populations remained locked in the ice-free periglacial Berginian refugium that extended from eastern Siberia to northwestern Alaska. It is unlikely that groups were able to move south of the continental ice in significant numbers (if at all) prior to its rapid wasting at the close of the Fraser/Wisconsin epoch (Burtchard 1987:209-213; Mathews 1982; Young 1982).

Terminal Fraser: Everson Interglacial & Sumas Stade; circa 13,500 to 8,500 B.P.

Floral and faunal patterns changed dramatically with the wasting of cordilleran and continental ice at the end of the Fraser glacial epoch. Despite interruption by at least one major renewed advance about 11,000 to 9,500 years ago (the Sumas Stade, or McNeely drift on Mount Rainier), plant and animal communities dispersed into previously ice mantled areas, changing the character of the Cascades, Puget Trough and Olympic Peninsula dramatically. By virtue of its tolerance for nutrient poor, glacially scoured sediments, Pinus contorta (lodgepole) became the dominant tree species in central and northern Puget Trough early in the period (Whitlock 1992:15). Whitlock notes that through time "P. contorta was joined by Pica sitchensis [Sitka spruce], Pseudotsuga [Douglas fir], and Tsuga heterophylla [western hemlock] to form a more closed forest." Presumably these more nearly closed forest conditions appeared near the end of the period coincident with rapid wasting of the Sumas stade.

As Cascade alpine glaciers retreated, alpine tundra and patchy forest habitat established on lower mountains and intermontane valleys. Similar communities probably were established to a limited extent on Mount Rainier as well. During the McNeely drift, permanent snowline on Mount Rainier is estimated at about 5,900 ft with glacial tongues extending well into the major river valleys (Crandall and Miller 1974:43). At this elevation, many of the Park's larger mid-elevation ridge, cirque basin, and tarn parklands would have been under, or immediately adjacent to, perpetual snowpack. Vegetated habitat would have been limited primarily to low to mid-elevation slopes–probably in dispersed semi-isolated stands around the mountain base. While it is likely that some open forest/tundra species colonized the mountain during the Everson interglacial and persisted through the McNeely Drift (e.g., mountain goats, pika and marmots), their number was probably too low and exploitation cost too high to have been of serious economic interest to the few early Holocene human groups in the region. The probability of human forays into the immediate Mount Rainier area increases sharply, however, with rapid glacial retreat and establishment of more nearly modern plant and animal associations after about 9,000 to 8,500 years ago.

Modern Interlude 1; circa 8,500 to 7,800 B.P.

The demise of the Fraser glacial epoch, and ultimately the onset of the mid-Holocene hypsithermal interval probably was due largely to an increase in summer solar radiation brought about by more pronounced earth axial tilt in conjunction with a near sun orbit (perihelion) during the summer (Kutsback and Guetter 1986; Whitlock 1992:15-16). According to Sea and Whitlock (1995:379), "The direct effects of greater-than-present summer radiation in the Pacific Northwest included higher temperatures and decreased effective moisture. Increased summer radiation also resulted in a strengthening of the east Pacific subtropical high pressure system." In concert, these events induced dramatic environmental changes in the Pacific Northwest and over much of the world.

During this relatively brief interval, environmental conditions on Mount Rainier are likely to have approximated those of the present day–though perhaps with hotter summers. In essence, this is a formative period during which Mount Rainier took on most of its modern vegetative character; even though, geologically, the mountain was rather different than present. The mid-Holocene destruction and rebuilding of Mount Rainier's summit had yet to occur, and many of the newly ice-free landscapes are likely to have been highly unstable. Even though the permanent snowline retreated to approximately its present elevation, the effects of rapid glacial retreat should have lingered. It is likely that water-heavy ice masses stimulated frequent avalanches on upper elevation slopes and generated repeated high energy lahars in glacial river valleys. Newly exposed, poorly consolidated high gradient terrain should have been subject to repeated mass wasting rock avalanche movement.

These geological characteristics probably did not have a marked effect on forest succession above the lahar-prone river valleys and below glacial margins. It is likely that, for the first time, most of Mount Rainier's low to mid-elevation slopes took on a forest mantle roughly comparable to that of the present day, though probably with greater species variability and higher fraction of seral representatives. It is reasonable to expect basic faunal patterns described earlier in this chapter to have become established as suitable habitat became available.

The development of economically useful plant and animal associations and the presence of human populations sets the stage for initial use of Mount Rainier National Park. Even so, because of very low regional population density and probable availability of game in less logistically challenging places elsewhere, it is likely that use of the mountain was limited. Indeed, people may have done little more than observe the massive peak at a distance well beyond present Park boundaries.

Hypsithermal Interval; circa 7,800 to 4,500 B.P.

Because the northern hemisphere had entered into a period of rapid climatic change, the first Modern Interlude was essentially a transient state into a prolonged xerothermic period that dominated the region for over 3,000 years. Evidence for the existence of an early to mid-Holocene drought in the Pacific Northwest is compelling, though timing and local effects remain incompletely understood. Heusser (1977) notes its impact on the Puget Trough and coastal Olympic peninsula by charting an increased fraction of Douglas fir over western hemlock–a change that he believes reflects an increase in about 2° C in mean summer temperatures compared to the present climate. [9] Citing a number of sources, Whitlock (1992:17) reinforces the point, noting further that prairies and open forest conditions expanded throughout most of the Puget Trough. Fire frequency increased, in places reducing forest maturity in a manner that created a "mosaic of forest in various stages of succession." At Battle Ground Lake on the lower western flank of the southern Washington Cascades, Whitlock's pollen data suggest a 40-50% reduction in annual precipitation between 9,500 and 4,500 B.P. Similar to Heusser, she suggests that these data imply an annual temperature 1° to 3° C higher than present.

East of the Cascades, the forest/steppe margin moved higher and further north than present. The Columbia Plateau remained essentially treeless. Higher than present percentages of chenopodium and amaranthus at Carp Lake in southwest Washington indicate warm/dry rather than cold/dry condition were dominant during the period (Whitlock 1992:17; Barnowsky [Whitlock] 1985).

On Mount Rainier, Dunwiddie (1986) examined pollen and macrofossil data from three sites situated between 4,250 to 4,920 ft elevation on the mountain's southern flank atop the 6,000 year old Paradise Lahar. As expected for a very young landscape, Dunwiddie notes a high fraction of early seral species at the base of the profiles (especially noble fir [Abies procera], subalpine fir [A. Lasiocarpa] and lodgepole [Pinus contorta]). However, seral species continue to dominate all levels below Mt. St. Helens series Y_n tephra (circa 4,000 to 3,250 B.P.), suggesting persistence of a warmer/drier than present climate before that date (Dunwiddie 1986:63). Assuming his interpretation is correct, these Mount Rainier data are approximately consistent with the general mid-Holocene xerothermic period as indicated by floral changes elsewhere in the Pacific Northwest. [10]

Glacial drift data from Mount Rainier also indicated a prolonged mid-Holocene period of alpine glacial withdrawal. No major glacial advance can be documented between the final retreat of the Fraser (McNeely Drift) circa 9,500 years ago and the relatively limited advance marked by the Burroughs Mountain Drift about 2,900 years ago (Crandell and Miller 1974). This does not mean that there were not lesser glacial advances and retreats during the period, but that such advances were small and traces obliterated by later, larger advances.

Geological events on Mount Rainier would also have affected environmental patterns temporarily, particularly to the mountain's southern and northeastern slopes. About 6,000 years ago, the Paradise and other lesser lahars reduced the summit and modified landforms on the south face of the mountain. Later, the 5,700 year old Osceola collapse and mudflow lowered the summit approximately 2,000 ft, moving a massive sediment load down the north and east faces of the mountain into the White River drainages and out into the Puget Trough. Undoubtedly, the physical character of the mountain was transformed dramatically. Even so, it is likely that impacts to vegetation patterns were largely temporary. The impact to the archaeological record notwithstanding, effect on land-use itself was probably neither substantial nor long lasting. The primary effect would most likely have been limited to short-term repression of forest cover serving to enhance somewhat the resource productivity and economic utility of affected landforms.

Clearly, regional effects of the Hypsithermal Interval are complex. Since human populations were now established in the region, Hypsithermal climatic changes hold implications for adjustments to regional settlement patterns and for use of Mount Rainier itself. For our purposes, the following general Hypsithermal patterns appear to apply to subregions in the vicinity of Mount Rainier:

East of the Cascades:

1) Increased aridity and degradation of ungulate forage in the Columbia Plateau.

2) Northward and upward shift in forest/steppe margins in the Okanagan and Cascade Mountains respectively.

West of the Cascades (Puget Trough):

1) Increased patchiness and open forest mosaics with increased fraction of mildly drought tolerant species such as oak (Quercus) and Douglas fir (Psuedotsuga).

2) Improved ungulate habitat due to opening of the forest canopy and concomitant expansion of grasses and brushy plant cover.

Southern Washington Cascades (including Mount Rainier):

1) On lower slopes and intermontane valleys, forest cover becomes patchier as drier conditions make forests more susceptible to fire–particularly in leeward settings; ungulate and collectable plant habitat improves.

2) Effect on upper to middle elevation slopes variable with exposure; species composition may have shifted to increased representation of drought tolerant species such as Psuedotsuga; in the absence of massive fire events, closed canopy forest should have dominated in wetter windward settings with increased patchiness in leeward settings and geologically disturbed landforms; effect on ungulate habitat unclear–here considered inconsequential.

3) Forest expansion on high elevation landscapes; forests invade and become more uniform in subalpine and lower alpine zones due to suppressed snowpack and lengthened growing season; upper elevation ungulate habitat is restricted; selective contexts for increased human induced burning to enhance habitat.

As the mid-Holocene drought deepened, it is likely that climatic events combined to reduce forage essential to support ungulate populations and their human predators east of the mountains, while simultaneously increasing them in the Puget Trough and lower elevation Cascade valleys and foothills. Accordingly, the Hypsithermal Interval may have witnessed a punctuated increase in use of landscapes and productive environmental zones in the vicinity of Mount Rainier. It is likely the use of the mountain itself increased during the period, perhaps involving fire modification of upper elevation landscapes.

Modern Interlude 2; circa 4,500 to 2,800 B.P.

Heusser's (1977) data suggest rapid amelioration of mid-Holocene warming starting about 5,000 years ago. By about 4,500 years ago, climatic conditions appear to have cooled to within 1° C of the present summer average (see Table 2.3). Given that climatic conditions were not perfectly stable or as regular as indicated on Table 2.3, the general return to cooler and moister circumstances should nonetheless have induced environmental regimes on Mount Rainier approximating those of the present day. Dunwiddie's (1986) data from Paradise lahar are consistent with a such a change. He suggests that cooler/moister conditions are indicated by increased fractions of mountain hemlock (T. mertensiana) and Alaska yellow cedar (Chamaecyparis nootkatensis) above Mt. St. Helens series Y_n tephra. Because the mountain building Mount Rainier series C eruptions had not yet occurred, Rainier retained its lower, open to the east (but presumably now well vegetated) profile created by the Osceola collapse over a thousand years earlier.

Less xeric climatic conditions following the Hypsithermal Interval should have reversed regional environmental patterns. In the Puget Trough, forest cover returned more completely closed canopy. Forest maturity also increased on lower elevation Cascade valleys and slopes. Higher, forests retreated, reopening subalpine and alpine habitats on Mount Rainier. East of the mountains, forests again moved downslope, and Columbia Plateau rangeland improved.

It is plausible that loss of lowland ungulate habitat in the Puget Trough and lower western slope would have increased resource stress on growing regional populations now well established in the region. Such conditions would have established a selective context favoring intentional burning to maintain forests at a more productive, lower maturity state. On Mount Rainier, seasonal use should have focused on subalpine to alpine habitats and at the margins of naturally and culturally induced burns as discussed in the previous section.

Burroughs Mountain Glacial Advance; circa 2,800 to 2,100 B.P.

The Burroughs Mountain advance is the first stade of the Winthrop Creek glaciation on Mount Rainier (see Crandell and Miller 1974:44-50). It is indicated in the Park by a few remnant moraines, most of which have been crosscut by the later, larger Garda stade. At the peak of the Burroughs advance about 2,600 to 2,400 years ago, permanent snowpack is estimated to have shifted downslope to about 6,400 to 6,200 ft–400 to 600 ft lower than the mid-1960s snowline. In essence, permanent snowpack would have covered what is now the Park's alpine tundra as shown on Figures 2.10 through 2.13. Subalpine associations are likely to have shifted downslope to what are presently Mount Rainier's upper forested slopes.

The Burroughs glacial advance is roughly contemporaneous with the most recent mountain building episode on Mount Rainier (i.e., the Mount Rainier series C eruptions that rebuilt the summit to its present elevation. Series C events not only built the modern summit, but deposited sand to gravel sized tephra over much of the central and northeastern part of the Park (see Figure 2.3). It is possible that the combination of lowered snowline and heightened eruptive activity reduced the value of affected mid to upper elevation landscapes. More likely, however, increased snowpack, shortened growing season and periodic volcanic disruption simply pushed tundra and subalpine associations downslope, and perhaps extended them to the northeast. [11] It is unlikely that these events had marked effect on the productivity of economically useful plants and animals. Since major river valleys are not primary resource areas in any case, expansion of glaciers and glacial outwash events into them should not have significantly influenced human use.

Modern Interlude 3; circa 2,100 to 900 B.P.

Huesser's (1977) lowlands data and Crandell and Miller's (1974) glacial study indicate a circa 1,200 year interval between the two Winthrop Creek glacial advances. Huesser's climatic inferences suggest that temperatures rose overall during the interglacial period, but remained slightly cooler than present (see Table 2.3). During this interlude, regional floral and faunal patterns should once again have returned to near modern conditions as described for the second Modern Interlude above. On Paradise lahar, Dunwiddie's (1986) pollen profiles show an increase in western hemlock (T. heterophylla) about 2,000 years ago at Jay Bath, and Alaska yellow cedar (C. nootkatensis) at Reflection Pond about 1,500 years ago. Both suggest continuing cool and moist conditions, perhaps associated with natural succession processes begun with the amelioration of the Hypsithermal Interval.

Garda Stade Glacial Advance; circa 900 to 500 B.P.

The second major advance of the Winthrop Creek glaciation extended beyond and obliterated most of the moraines of the earlier Burroughs Mountain drift. The Garda stade on Mount Rainier reached its peak about 700 years ago, lowering permanent snowpack to about 6,200 ft–circa 800 ft lower than the mid-1960s (Crandell and Miller 1974:45). There is little floral data from Mount Rainier directly attributable to this period. Dunwiddie (1986:65-66) speculates that predominance mountain hemlock (T. mertensiana), normally represented at higher elevation, and loss of seral species at Jay Bath may be attributed to this glacial advance. 12]

In the absence of more refined floral data, we are limited to inferences based on combined effects of cooling and the lowered snowline. At 6,200 ft, permanent snowpack would cover some of the Park's larger expanses of upper elevation low gradient terrain (e.g., Sunrise Ridge and Spray Park). Even though tundra and subalpine associations would have shifted downslope, the substitution of sideslope habitat for the broader "pasture" of these rolling uplands, implies a loss of ungulate forage overall. However, other areas of flat to rolling terrain–Grand Park, Mist Park, Indian Henry's Hunting Ground–lie below the snowline as inferred and probably retained tundra, if not upper subalpine, qualities. For present purposes, I apply the same assumptions as those for the lesser Burroughs Mountain stade–that is, that net losses of upper elevation subalpine/tundra habitat probably were balanced by habitat increases at lower elevation. From a human standpoint, it is likely that use of the mountain continued, but use intensity may have decreased overall or shifted from higher landforms like Sunrise Ridge, upper Berkeley Park and Spray Park to lower elevation, low gradient landscapes such as Mist Park, Indian Henry's Hunting Ground, or presently forested lower elevation ground. It is important to note, however, that these assumptions are speculative in the absence of paleoecological and archaeological data from a wider selection of Park locations. Such work holds intriguing promise for improving our understanding of paleoclimate, habitat and human interactions and are highly encouraged.

Modern Interlude 4; circa 500 B.P. to Present

The current interlude is marked by renewed climatic warming to within 1° C of the present July average at sea level. The period is associated with a variety of short-term, punctuated climate shifts world-wide (the apparent contrast with earlier periods simply reflecting more complete climatic data for the historic period). The maximum extension of glacial ice associated with the Garda stade retreated sharply at the outset, though renewed minor advances occurring between 500 and 100 years ago (i.e., the Little Ice Age) technically may be considered part of the general Garda event (Crandell and Miller 1974:49-50, also see footnote 12). For the most part, the period from A.D. 1450 to the latter half of the 1800s was cooler than present, but warmer than the Garda stade peak noted above. Twentieth century oscillations also include the sharp, but brief, xerothermic event of the 1930s and a cold rebound in the 1940s.

It is reasonable to assume that climatically sensitive environmental zonation patterns described earlier applied, in general terms, throughout this most recent interlude. Also we assume that resource opportunities described in that section were available to human populations during this time frame and during each of the earlier Holocene modern interludes. The manner and intensity with which they were exploited, however, varied not only with resource availability and abundance, but with population density and social variables affecting human groups in the broader region surrounding Mount Rainier. Consideration of broad-scale, long-term changes in regional land-use patterns and their implications for the archaeology of Mount Rainier National Park is the subject of the following final section of the chapter.