The genus Erythronium is widely, though disjunctly, distributed in temperate areas across Eurasia and North America. It occurs in eastern and western North America, eastern and central Asia, and Europe. In western North America about 18 species occur from British Columbia to New Mexico (Allen 1993, Allen and Robertson 2003). Erythronium reaches its greatest diversity in North America, specifically in Oregon and northern California (Allen, Soltis and Soltis 2003).
Globally, E. elegans is known from six site areas in the northern Oregon Coast Range. The range extends approximately 72 kilometers (45 miles) north and south, within a zone that is 16 – 32 kilometers (10 - 20 miles) from the Pacific coast. All sites occur on prominent Coast Range peaks and ridges, separated from one another by up to 48 kilometers (30 miles). This distribution results in a pattern of high-elevation islands of habitat separated by lower elevation coniferous forests.
Table 1. E. elegans known sites
Siuslaw NF, USFS
Lost Prairie Area of Critical Environmental Concern (ACEC)
Salem District, BLM
Willamette Industries, TNC shares management.
Saddlebag Mountain ACEC/Research Natural Area.
Salem District, BLM
Tillamook State Forest, Oregon Dept of Forestry
The cumulative area of all known sites is estimated to be approximately 95 hectares (235 acres) with an estimate of 3,000 total stems (Natureserve 2009). Because of the limited availability of habitat and the disjunct distribution of sites, opportunities for this species to expand beyond currently occupied sites may be limited to non-existent.
C. Population Trends and Demography
A six year study was initiated in 1991 to compare the demographics of a population perceived to be healthy (Mt Hebo) with one thought to be in decline (Lost Prairie) (Guerrant 1999). At each site, two sub-populations were selected for study. Analysis showed that one sub-population was growing vigorously, one was in serious decline, and the remaining two were stable (Guerrant 1999). The sub-population with declining numbers was in a closed canopy coniferous forest with heavy shade and a well developed herbaceous plant layer, while the sub-population with increasing numbers and the two that were stable were in more open habitat where they received more light. The site with the greatest population growth had, on average, the lowest levels of phosphorous, potassium, calcium, magnesium, ammonium, and nitrate in the soil (6 of the 8 nutrients tested for). One hypothesis offered by Guerrant is that E. elegans is generally at a competitive disadvantage relative to its neighboring species and it may do better in poor soil conditions, obtaining sufficient nutrients where other taxa might struggle (Guerrant 1999).
The sub-population with the greatest reproductive success occurred under or directly adjacent to conifers with branches that extend to ground-level, possibly protecting plants from herbivory by elk. Guerrant (1999) suggests that some of the disadvantage of growing in a more shaded environment, in this case, may be offset by the protection the plants appear to receive from elk herbivory. The protective advantage from low-growing branches would tend to disappear in forest habitat with a high closed canopy and low light levels, where herbivory could occur unimpeded.
Early in the demographic study it was recognized that determining mortality would be difficult because plants may remain dormant below the soil surface for some period. In addition, part of the study coincided with a period of high incidence of a fungal pathogen on E. elegans and other Liliaceae species. The pathogen was tentatively identified as Phoma sp. (Guerrant 1999). No evidence of fungal infection was seen during a 2006 site visit to Mt Hebo (M. Stein and R. Exeter, Pers. Obs.2006).
Specific observations from work done by Raven (1994, 1995, 1996) on herbivory and vegetative competition include:
There were clear differences between caged and un-caged plants in a herbivore exclusion experiment: in un-caged plots more flowers and fruits, and bigger leaf portions were removed and presumably eaten by elk or deer.
More mature fruits were produced in caged plots.
Caging is an effective method for excluding herbivory and subsequently improving fruit success.
Light salal cover appears to have both beneficial and harmful effects on E. elegans. Benefits include some protection from deer and elk herbivory and from weather extremes, however, salal cover shelters rodents which feed on the lily, and the cover limits light availability. Heavy salal cover may prevent the establishment of new plants and contribute to the death of established plants.
A characterization of Erythronium elegans habitat presents challenges. Although the species is confined to a small geographic area of the northern Oregon Coast Range, and is restricted to peaks above 790 meters (2,600 feet) in elevation, E. elegans is found in a wide array of habitats within this limited landscape. To capture this habitat variation and diversity, population areas are discussed individually:
1. Mt. Hebo Site
E. elegans is found on Mt. Hebo from 790 meters (2,600 feet) in elevation to the summit at 945 meters (3,100 feet). Plants occur in (a) forest habitats, both open and closed canopy, (b) forb and graminoid dominated meadows, and (c) rocky cliff escarpments within both forest and non-forest habitat. Each of these habitats has its own species assemblage and therefore the list of associated species for E. elegans is large. Essentially, it includes most plant species known from the upper elevations of Mt. Hebo. E. elegans plants appear to have a strong preference for northwest to northeast aspects, particularly in non-forested habitat. The limited numbers of plants that do occur on more southerly aspects, particularly in non-forested habitat, are typically found in sheltered micro-sites next to a clump of shrubs, in a concave depression, or other locations that provide relatively cool, moist conditions. In forested habitats, aspect does not appear to be as important. Associated species at the Mt. Hebo site are listed in Appendix Tables 2 through 4.
(a) Forest Habitats
E. elegans is found in a wide range of forest habitats at Mt. Hebo including forest edges adjacent to meadows and rock outcrops, small gaps in older (80-100 year old) stands, closed canopy stands with up to 80 percent canopy cover, as well as openings created by road corridors, where plants occur on cut slopes and in ditch lines. Plants can occur as widely scattered individuals or in dense patches. Slopes range from 0 to 80 percent with northwest to northeast aspects being most common, although plants can be found in all aspects. Forest canopy cover appears to be a factor as plants are generally more numerous where cover is less than 70 percent and often occur only as scattered individuals where canopy cover is greater than 70 percent. Plants in forest habitat generally do not tolerate a shrub component of more than 20 to 30 percent cover. Vegetative cover for herbaceous plants and bryophytes is variable, ranging from 5 to 90 percent.
Figure 4. E. elegans under a closed-canopy conifer stand on Mt. Hebo. Arrows indicate non-reproductive plants.
Photo by K. Cushman
(b) Meadow Habitat at Mt Hebo
The grass- and forb-dominated meadows on the summit of Mt. Hebo have a complex history. Hammond et al. (1980) recounted information from Scott T. Wells, a Forest Service employee stationed at Hebo Ranger District, that most of the mountain was covered with a mature forest of western red cedar (Thuja plicata) as indicated by the presence of old stumps. Fires that occurred around 1845 and again at the turn of the century replaced much of the forest with open grassland. Sheep grazing by Euro-American settlers also occurred on the mountain around 1900 (Hammond et al. 1980). From 1909 to 1923 a large reforestation effort planted portions of the burn area with off-site Douglas-fir (Pseudotsuga menziesii var. menziesii), Norway spruce (Picea alba) and other experimental non-native tree plantings. In the 1950’s, the U.S. Air Force established a radar base on the summit of Mt. Hebo, complete with barracks and 27 houses for full-time service personnel and their families. This facility occupied a portion of the meadow area and was in operation until 1984. The base facilities were then demolished and the site “reclaimed”. More recently, the meadows have been managed for Oregon silverspot butterfly (Speyeria zerene var. hippolyta), a federal listed species. One of the few extant sites remaining for the silverspot, portions of the meadow are mowed annually to encourage the growth of early purple violet (Viola adunca), the larval food plant for the butterfly.
The meadow complex is located on the broad summit plateau where slopes are generally less than 10 percent and southwest to southeast aspects dominate.
Vegetation is a mix of native and non-native forb and graminoid species, depending on the degree to which soil disturbance has occurred. Prior to the 1845 fire, it is thought that native meadow species were largely restricted to small non-forested areas on rocky outcrops and cliffs, as evidenced by the lack of Roemer’s fescue (Festuca idahoensis Elmer ssp. roemeri (Pavlick) S. Aiken) (= F. rubra) (Hammond at al. 1980). Remnants of this native grass would indicate the presence of stable meadow habitat. Total vegetative cover is usually in excess of 80 percent. The distribution of E. elegans tends to be in small patches of fewer than 10 stems or scattered individuals. Often, plants are in small microsites on the north side of Alaska huckleberry (Vaccinium alaskaense) and serviceberry (Amelanchier alnifolia) shrub patches or in concave depressions. These locations likely have more consistent soil moisture early in the growing season from a lingering snowpack or reduced evaporation.
Figure 5. E. elegans in meadow habitat, Mt. Hebo.
Photo by K. Cushman
(c) Rocky Cliff Escarpment
The largest colonies of E. elegans on Mt. Hebo are found where the relatively flat plateau breaks along a steep escarpment on the north side of the mountain. There is often a sharp contrast between the scattered plants that occur on the gentle slope of the south side of the mostly east-west break, and dense colonies found only a few feet away down slope on the north side, particularly in steep rock or cliff habitat. Aspects are predominately northwest to northeast with slopes of 0 to 60 percent. Where the escarpment has a more westerly or easterly aspect, E. elegans is absent. Within this habitat, shrub cover, mostly salal, also appears to be a limiting factor if taller than approximately 45 centimeters (18 inches). Where E. elegans is present, the vegetation is dominated by herbaceous species and bryophytes with a minor component of trees and shrubs.
Figure 6. Escarpment on the summit of Mt. Hebo.
Photo by M. Stein
2. Lost Prairie Site
This site has a predominant northeastern aspect and is approximately 840 meters (2750 feet) in elevation. The site was clearcut in 1972. The original timber stand prior to cutting was probably dominated by western hemlock (Tsuga heterophylla), Douglas-fir and noble fir (Abies procera). The site is presently dominated by Douglas fir and western hemlock. Vine maple (Acer circinatum), rhododendron (Rhododendron macrophyllum) and salal are the dominant shrubs. The plant series is within the western hemlock/rhododendron/salal plant association. The soils tend to be thin with gravelly soil. Although the majority of known sites of E. elegans occur mostly in the conifer stand, a few individuals are known to occur along the outer edges of a Sphagnum bog. In this respect, the Lost Prairie site is similar to the Fanno Meadows wetland site. There are an approximately 50-100 individual plants on this site. A few of the bryophyte species associated with E. elegans from within the conifer stand include: Racomitrium elongatum, Polytrichum juniperinum, and Eurhynchium oreganum.
3. Rocky Point Site
This site has a predominant northeastern aspect and is approximately 880 meters (2880 feet) in elevation. It is dominated by western hemlock, Douglas-fir, western red cedar and a few noble firs. The stand probably originated in the 1880’s after stand replacing fires. There are approximately 100 individual E. elegans on this site. It is characterized by an approximately 0.1 hectare (0.25 acre) basalt outcrop surrounded by a closed canopy cover. E. elegans occurs on the basalt where the canopy is open and is not known from beneath the dense conifer overstory. Racomitrium elongatum is the dominant moss species associated with the basalt. Oceanspray (Holodiscus bicolor) and salal also occur on the basalt outcrop.
Other associated species at the site include; Polypodium sp., Rosa sp., Vaccinium sp., Selaginella sp. and the following bryophyte species; Eurhynchium oreganum, Hyalocomnium splendens, Rhytidiadelphus loreus, Rhytidiadelphus triquetrus, Racomitrium varium, Isothecium stoloniferum, Polytrichum juniperinum, Polytrichastrium alpinum, and Scapania sp.
4. Fanno Meadows Site
This site occurs on the outer edges of a Sphagnum-sedge wetland. This site is approximately 870 meters (2850 feet) in elevation and has a gentle eastern slope. The conifer stand surrounding the wetland is similar to the BLM lands which surround it, and is estimated to be 65 years old. Dominant conifers include western hemlock, Douglas-fir and noble fir. The shrub layer is mostly thickets of Rhododendron. Few E. elegans individuals were observed (< 20), and were mostly located within or near plots established by Ed Guerrant of Berry Botanical Garden, at the outer edge of the wetland. These sites are becoming overgrown by competing vegetation similar to those at Lost Prairie.
5. Saddlebag Mountain Site
This site occurs on a north-south ridge and has a gentle eastern aspect. The site is approximately 975 meters (3200 feet) in elevation. Western hemlock, noble fir and Pacific silver fir (Abies amabilis) dominate the western slopes and create a canopy cover greater than approximate 90 percent. There is little understory or ground vegetation in this stand. The eastern portion of the ridge was clearcut in approximately 1983 and is now dominated by true firs (Abies spp.) and western hemlock. E. elegans occurs in two distinct habitats at the Saddlebag site. One area is a small 0.08 hectare (0.2 acre) basalt outcrop with an estimated population of less than 30 individual plants. There is a canopy of mostly western hemlock to the west of the basalt outcrop and thickets of hemlock and true fir reproduction to the east. The other site occurs on a flat broad ridge area of approximately 0.2 hectare (0.5 acre) with an estimated population of 150-200 individual plants. This area has thin gravelly soils and scattered exposed basalt boulders. Non-native graminoids are dominant in this area with little existing competing vegetation. This site is also flanked by western hemlock to the west and conifer reproduction to the east. Racomitrium elongatum is the dominant moss of both of these sites.
6. Triangulation Point Site
Discovered in 2000 by Oregon Department of Forestry personnel, this site occupies an area of approximately 80 hectares (200 acres) on the ridge and upper slopes of Triangulation Point in Tillamook State Forest. Elevation ranges from 855 - 1005 meters (2800 - 3300 feet). Fires have repeatedly burned the site during the last century. The 1933 Tillamook Fire killed all the overstory trees and was followed by fires in 1939 and 1945. Salvage logging then took place in 1954 and 1955. A 1954 aerial photo shows a few scattered remnant noble firs growing on the north side of the ridge, but most of the area remained open and non-forested as the result of the salvage harvest and fires. At present, forest stands are dominated by 35 - 40 year old Douglas-fir with scattered noble fir and red alder on the north side of the ridge. There is one stand of young, dense western hemlock on the west ridge of the site. The Douglas-fir was pre-commercially thinned in the late 1980s (D.Clough 2006).
E. elegans is most abundant and widespread on open, wet, brushy slopes with a north aspect. Plants become more scattered and patchy on forested north and west aspects and road cut banks. In forested habitats, plants occur where the shrub component is either lacking or only occurs in minor amounts, and there are small canopy gaps where sunlight can infiltrate. There is also a tendency for plants in forested habitat to occur above the headwalls of streams. E. elegans can also be found on dry brushy slopes with a south aspect where microsite conditions provide some protection. Overall, it appears that E. elegans does best where site conditions are moist to wet, with either partial or full sun.
Other associated species at the site includes copperbush (Cladothamnus pyroliflorus), devil’s club (Oplopanax horridus), huckleberry (Vaccinium spp.), and oxalis (Oxalis oregana).
Aspects of the abundance and distribution of E. elegans suggest a relatively high overall vulnerability for this species. With relatively few sites, a disjunct distribution, and small geographic range, the species seems to be susceptible to range-wide events such as climate change, inbreeding depression, disease and insect outbreaks, invasive plant invasion, herbivory, vegetative succession, illegal collection, and management activities that could alter site conditions.
Large-scale stochastic threats (threats that affect the species range-wide)
1) Climate change.
Dramatic range shifts that have occurred as a response to climate change are well documented. Since the last glacial maximum, (25,000 years before present) the entire range of some tree species have been displaced north to new latitudes, e.g., spruce (Picea spp.), while other species continued to grow at the same latitude, but have shifted higher in elevation in response to climate (Davis and Shaw 2001). The Inter-governmental Panel on Climate Change (IPCC 2007) concluded that there is a very high confidence (>90%) that recent warming is strongly affecting biological systems including pole-ward and upward shifts in plant and animal ranges. Over the past 40 years, biological trends have been found to match climate predictions with a species range shift averaging 6.1 kilometers (10 miles) per decade toward the poles (Parmesan and Yohe 2003). An upward elevation shift of 8-10 meters (26-33 feet) per decade is predicted for nine plant species in the European Alps based on 70 to 90 years of occurrence records (Grabherr et al. 1994).
During the 20th century average temperature increase was greater in the Pacific Northwest (0.8º C, 1.8º F) than globally (0.6º C, 1.1º F), with the maritime portion of the region having the greatest increase (0.9ºC, 1.7º F), occurring primarily during the winter months (Mote 2003). Individual models for predicting global climate change do not work well on a regional scale (Mote et al. 2005). Using ten global climate models to perform simulations of future climate change for the Pacific Northwest, the average warming rate during the next century is expected to be in the range of 0.1º-0.6º C (0.2º -1.1º F.) per decade (Mote et al. 2005). Predictions for changes in total precipitation and seasonality are less certain, although warmer winter and spring temperatures are expected to reduce winter snowpack accumulations, shift the winter snowline to higher elevations and result in an earlier snow melt in the spring (Stewart et al. 2004, Mote et al. 2003).
Climate at E. elegans sites is typically maritime Pacific Northwest, characterized by moderate temperatures and high precipitation with a marked winter maximum. Weather data from the National Weather Service station at Laurel Mountain, located 7 kilometers (4 miles) northeast of the Fanno Meadows site at an elevation 1095 meters (3,590 feet), likely indicate general conditions at E. elegans sites. Average annual precipitation from 1971-2000 was 3.07 meters (121 inches), 90 percent of which occurred from October through May. For the period 1977 – 2008, average annual snowfall was 3.04 meters (120 inches), but varied from 0.58 meters (23 inches) to 8.20 meter (323 in). The coldest month was January with a mean temperature of 1.4º C (34.5º F) and the warmest was August with a mean temperature of 14.3º C (57.7º F) (Western Regional Climate Center 2009).
Using a scenario with increasing temperature, particularly during the winter months, and little change in total precipitation or seasonality, a reasonable outcome would be a reduction in the amount of precipitation falling as snow, and any snowpack that does accumulate would melt earlier in the spring. The near-freezing mean winter temperature at Laurel Mountain suggests that a small winter temperature increase could result in a drastic decrease in the amount of precipitation falling as snow. From 1980-2008, the snowpack at 950 meter (3,110 feet) Saddle Mountain, approximately 32 kilometers (20 miles) north of the Triangulation Point E. elegans site, had melted by early- to mid-April in a typical year. Actual dates ranged from May 31 to no snowpack at all. The mean water content (snow water equivalent) during April was 13 centimeters (5 inches), ranging from 0 (no snowpack) to 79 centimeters (31 inches) (USDA NRCS 2009a).
A possible effect of a reduced snowpack and earlier melt off on E. elegans is a reduction in soil moisture during the early part of the growing season in April and May. The contribution made by the melting snowpack to soil moisture at E. elegans sites is not known, however, precipitation in the form of rain or snow is still substantial at this time of year and it would seem that soil moisture would not be a limiting factor. At Mt. Hebo, E. elegans is most limited in meadow habitat with no tree cover and no slope. The scattered plants that do occur typically occupy small swales, the north side of shrub patches, or other microsites where soil moisture may be retained for a longer period. In contrast, dense colonies can be found at the edge of the meadow near the drip line of trees and where the meadow breaks into a north aspect, microsites where a lingering snowpack and reduced evaporation may lead to increased soil moisture. It is possible that a reduced snowpack in combination with a longer summer drought period could result in soil moisture deficits that affect E. elegans habitat suitability.
The response of E. elegans to changing conditions could be a shift to more suitable microsites that retain soil moisture for a longer period. It is almost certain that E. elegans will not be able to respond to a warming climate by expanding its range northward or shifting higher in elevation, given its current distribution. Based on the Grabherr et al. (1994) estimate for the upward elevation shift needed to compensate for predicted warming, all of the 215 meter (700 feet) vertical band of suitable habitat that E. elegans currently occupies would become unsuitable in 200 – 260 years. Noss (2001) suggested that for isolated habitats that harbor large numbers of endemic species, such as the cloud forests of tropical mountains, climate change is already the dominant threat because species have nowhere to shift in response to changes in both temperature and moisture. In these situations, ex situ preservation of species in zoos and botanical gardens may be the only way to avoid extinction. Ex situ collections should include sufficient genetic diversity to allow adaptation to uncertain conditions in reintroduction sites. Maintaining the full genetic diversity of E. elegans in-situ is a logical preparation for the possibility that ex-situ restoration becomes necessary in the future.