Printer Friendly

Colonization of a Volcanic Mudflow by an Upper Montane Coniferous Forest at Lassen Volcanic National Park, California.

JOSEPH D. WHITE [1]

ABSTRACT.--Colonization of a relatively undisturbed 1915 debris flow by an upper montane coniferous forest was examined in Lassen Volcanic National Park in northern California in 1987. Seventy 100-[m.sup.2] circular plots were arranged in four transects across the flow and sampled to measure tree densities, heights, basal areas and ages. The composition of the forest changed from a mixture of Abies magnifica, Pinus monticola and P. contorta on steep slopes to a forest dominated by P. contorta on shallow slopes. This pattern is typical for these terrains at the 2000-m elevations of the flow. Age data and historical photographs indicated little successful colonization before the late 1930s and peak colonization rates about 1955. Height growth has generally been slow with most trees being [greater than]20-y-old but [less than]2-m tall; however, some individuals, including some recent colonizers, have shown rapid growth. This variation among individuals suggests (1) that the earliest colonizers are not necessari ly those which will eventually dominate the forest and (2) that opportunities to invade and occupy the canopy may extend for 30 y after the first successful colonization.

INTRODUCTION

Much of the forested land of the northwestern United States is of volcanic origin and continues to be subject to the effects of volcanic eruptions (Kiver, 1982; Harris, 1988). Eruptions may range from catastrophic events, such as the 1980 eruption of Mount St. Helens in Washington that destroyed forests over a 600-[km.sup.2] area (Tilling et al., 1990), to localized events, such as the minor eruptions of Cinder Cone in Lassen Volcanic National Park in northern California that only affected nearby trees (Russell, 1897; Finch, 1937). Forests may be completely destroyed and buried by (1) lava flows, pyroclastic flows and ash falls (geological nomenclature follows Bates and Jackson, 1984 and Scarth, 1994) from active volcanoes or (2) lahars, mudflows, debris flows, earth flows and landslides from either active volcanoes or unstable terrain on the slopes of inactive volcanoes. The resulting deposits of lava, ash, mud or debris may require primary succession to replace the forest (Beardsley and Cannon, 1930; Eggle r, 1941; Heath, 1967; Frenzen et al., 1987). Alternatively, blast effects may destroy standing trees but allow recolonization from propagules in relatively undisturbed soils (Viers, 1987; Halpern et al., 1990).

Although research at Mount St. Helens is providing information on the invasion and recovery of herbs and shrubs on pyroclastic flow, lahar and pumice surfaces (e.g., Means et al., 1982; Chapin and Bliss, 1989; Halvorson et al., 1991; del Moral and Bliss, 1993; del Moral and Wood, 1993; Zobel and Antos, 1997; Titus and del Moral, 1998), less information is available on the slower rates of tree invasion and recovery (Halpern and Harmon, 1983; Frenzen and Franklin, 1985; Antos and Zobel, 1986; Adams and Dale, 1987; del Moral, 1998; Larsen and Bliss, 1998). Moreover, many of the tree species at Mount St. Helens differ from those at more northern and southern sites of volcanic activity (Sudworth, 1927).

The purpose of this study is to describe forest colonization of a debris flow surface deposited during the 1915 eruption in Lassen Volcanic National Park (hereafter, LVNP). The Lassen Peak eruption is particularly useful in measuring forest colonization because of (1) well-documented dates and patterns of eruption (Day and Allen, 1925) and (2) the history of aerial (e.g. Heath, 1967) and oblique photography (e.g., Stillman and Turnage, 1992) of the site. The principal tree species in LVNP (Barbour, 1988; Parker, 1991) are conifers and include red fir (Abies magnifica Andr. Murray), white fir (A. concolor (Gordon & Glend.) Lindley), Jeffrey pine (Pinus jeffreyi Grev. & Balf.), lodgepole pine (P. contorta murrayanna (Grev. & Balf) Critchf.), western white pine (P. monticola Douglas) and mountain hemlock (Tsuga mertensiana (Bong) Carriere) (botanical nomenclature follows Hickman, 1993). The specific objectives of the study were to: (1) document the status of forest development 72 y following the eruption; (2) c ompare the forest composition to that expected for similar elevations and terrains and (3) use tree age data and historical photographs to infer the past history of colonization.

SITE DESCRIPTION

The study was conducted on the northeast slopes of Lassen Peak, a 3100-in volcano (Fig. 1) (Harris, 1988; Scarth, 1994), where the climate is an upland, dry-summer type with winter snows and frequent summer droughts (Griffin, 1967; Parker, 1991). The most recent period of volcanic activity at Lassen Peak began in May 1914 with small eruptions of steam and ash and culminated with two large eruptions in May 1915 (Day and Allen, 1925; Loomis, 1926; Eppler, 1984; Scarth, 1994). On 19 May 1915 magma emerged at the summit, melted the surrounding snowpack and caused a fast-moving lahar (i.e., a meltwater-driven flow of mud and rock) that flowed northeasterly for more than 20 km down Lost Creek and Hat Greek and left deposits averaging 2-rn deep (Eppler, 1984). Thicker lahar deposits occurred in the headwaters of Lost Creek within 5 km of the peak where flows [greater than] 10-m deep destroyed the existing forests and left deposits containing little apparent organic debris (Loomis, 1926; Eppler, 1984). A second larg e eruption on 22 May 1915 involved three main components. The first component was a lateral blast that uprooted or broke off trees in a northeasterly direction for [greater than]5 km (Day and Allen, 1925; Eppler, 1984; Scarth, 1994). Contemporary photographs (Day and Allen, 1925; Loomis, 1926) show that this blast destroyed the forests surrounding the deep lahar deposits in Lost Creek. The second component was another fast-moving lahar that deposited generally [less than]1 m of sediments in Lost Creek (Eppler, 1984). The third component was a relatively slow-moving debris flow that was distinct from the previous lahars because it (1) contained a greater proportion of coarse grain materials, (2) came to rest on sloping terrain within 5 km of the peak and (3) was characterized by a distinct lobate structure whose leading margins were marked by steep-sided edges of [geq]0.3-m thick deposits (Eppler, 1984).

Most of this debris flow was emplaced over the deep deposits of the two previous lahars in the headwaters of Lost Creek (Fig. 1); however, part of its southeastern edge was deposited in an area where the forests and preexisting soils had not been removed by the lahars, but where trees were killed by the lateral blast. This is indicated by the decaying remnants of blown-down trunks in this area. Most lahar deposits have been dissected by stream erosion (Eppler, 1984) or overwashed by subsequent flooding events (e.g., Heath, 1967), but the debris flow surface has remained free of these disturbances. As such, the area of the debris flow that overlays lahar deposits represents an unusual opportunity to document primary forest succession on a known-age volcanic surface that (1) contains little organic debris and (2) has had little subsequent disturbance.

Several studies of herb and tree revegetation have been performed in the area impacted by the Lassen eruptions (Bailey, 1963; Heath 1967; Fessenden, 1984), but relatively little information is available for the debris flow area over the lahar deposits (Fig. 1). Bailey (1963) provided vegetation descriptions and quantitative data on tree abundances for numerous sites in the blast and lahar areas, but few of these sites were located on the debris flow. Heath (1967) measured tree densities and heights on the debris flow, but his transect (Fig. 1) was located where the flow overlies soil and not lahar deposits. Fessenden (1984) measured the species composition and biomass of herbaceous, shrub and tree species in areas she termed the Dense Forest Margin, which corresponds to the area underlain by soil (Fig. 1) and the Sparse Central Forest, which corresponds to the area overlying lahar deposits. Fessenden (1984) showed that herbaceous and shrub species were now only a small component of the community accounting fo r [greater than]1% of the total forest biomass.

METHODS

To measure forest composition and structure in the area underlain by lahar deposits four parallel transects, H1 through H4 (Fig. 1), were established in 1987 oriented across the direction of the flow at 200-m intervals. Transect H1 was located at the top of the debris flow, the other transects were at progressively lower elevations. Circular [100.m.sup.2] sample plots were located at 20-m intervals along each transect. Because the width of the debris flow varied, the numbers of plots per transect varied with 14, 21, 17 and 18 plots on transects H1, H2, H3 and H4, respectively.

The debris flow terrain was characterized by measuring elevation, slope and aspect at each plot. Elevation was measured to a resolution of 1[lem] 1-m using barometric surveying (Davis et al., 1981). Slope was measured to the nearest degree using a clinometer. Aspect was measured to the nearest degree using a compass corrected for magnetic declination.

All individuals of tree species in the plots were identified to species and measured for height and age in 1987. Heights of trees taller than 2-m were measured with a Haga altimeter. For trees [greater than] 1.37-m tall diameters at breast height (dbh) were also measured. Importance values (i.e., the sum of relative density, relative frequency and relative dominance expressed as basal area; Greig-Smith, 1964) were calculated for each transect. Herbaceous and shrub species, which comprise only a small component of the community (Fessenden, 1984), were not sampled.

Ages of trees with dbh [greater than] 0.08-m were determined by counting branch whorls, and the ages of trees with dbh [geq]0.08 m were determined by counting tree rings in increment cores. Increment cores were: (1) extracted as near to the ground surface as possible using a 5.15mm diameter borer; (2) air-dried; (3) glued onto wooden blocks; (4) sanded to expose ring structure and (5) measured to determine the number and widths of rings to the nearest 0.01 mm (Phipps, 1985). Because of problems with increment cores and whorl counts, the ages of approximately 16% of the trees had to be estimated from height and basal area using multiple regression procedures (Draper and Smith, 1981). To obtain a better estimate of the maximum ages of trees on the debris flow, cores were also collected from 36 relatively large (i.e., dbh [geq] 0.2 m) and, therefore, potentially old trees outside the sample plots.

Aerial and oblique photography from the LVNP archives and commercial sources (Still-man and Turnage, 1992) were examined under magnification for evidence of tree invasion. Although the ability to determine the presence of trees is dependent upon photographic quality and scale, trees [greater than]1-m tall should be discernible in oblique photography and trees with canopy diameters [greater than] 1-m should be discernible in aerial photography.

Statistical analyses.--Mean tree densities, heights, basal areas and slopes were compared among transects using Analysis of Variance (hereafter, ANOVA; Milliken and Johnson, 1984) performed using the General Linear Models procedure (hereafter, GLM) of the Statistical Analysis System (SAS Institute, 1989). Transect means were compared using the Studentized Maximum Modules procedure (SAS Institute, 1989). Where the data departed from the assumptions required for ANOVA, analyses were also performed on variance stabilizing transformations (Box et al, 1978), but these analyses are only reported when they resulted in different interpretations than untransformed data. The multiple regression procedures to estimate ages from heights and basal areas were also performed using the GLM procedure.

RESULTS

Terrain characteristics.--Mean ([pm]SD) elevations declined from 2033 [pm] 2-m for transect Hi to 1977 [pm] 4-m for transect H4. The maximum plot elevation of 2036-m occurred on Hi, and the minimum of 1970-rn occurred on H4. Mean slopes for plots differed significantly (F = 4.36; df 3, 66; P [leq] 0.01) from 8 [pm] 5[degrees] on H1 to 4 [pm] 3[degrees] on H4, and the proportion of plots with slopes [geq]4[degrees] decreased from 79% at Hi to 28% at H4. About 29% of the plots on H1 and H2 had slopes [geq]10[degrees]. Aspects changed from primarily northeast at H1 (median aspect = 55[degrees]) to primarily north at H4 (median = 350[degrees]. There was also considerable within-plot heterogeneity in slope and aspect due to the irregular lobate structure of the flow surface and the presence of [leq]1-m diameter dacite blocks that protrude above the flow surface (Eppler, 1984).

Tree densities. -- Mean density ([pm]SD) of trees of all species was 3507 [pm] 1812 trees [ha.sup.-1]. The majority of the trees were Pinus contorta (mean density was 2391 [pm] 1720 trees [ha.sup.-1]. The next two most commonly observed species, Abies magnifica and P. monticola had mean densities of 501 [pm] 413 and 400 [pm] 299 trees per [ha.sup.-1], respectively. These three species each occurred in [greater than]85% of the sample plots. Abies concolor, Pinus jeffteyi and Tsuga mertensiana each occurred in [less than]50% of the plots and had mean densities [less than]100 trees [ha.sup.-1].

There was a statistically significant (Fig. 2; F = 10.3; df = 3, 66; P [leq] 0.01) increase in the density of Pinus contorta from H1 to H4 (Fig. 2). Pinus contorta accounted for 40% of the trees on H1 and 78% of the trees on H4. Pinus monti cola densities declined significantly from H1 through H4 (F = 5.24; df = 3,66; P [leq] 0.01). Abies magnifica densities did not differ significantly (F = 3.05; df = 3, 66; P [greater than] 0.05) among transects.

Age distributions. -- Age distributions (Fig. 3) were computed separately for (1) those individuals that could he aged using increment cores and whorl counts and (2) those individuals whose ages had to be estimated from multiple regression equations. These regressions had [r.sup.2] of 0.72, 0.64 and 0.78 for Pinus contorta, Abies magnifica and Pinus monticola, respectively. The [r.sup.2] for the other species were [greater than]0.76. For trees with measured ages, comparisons of ages estimated from the regressions with the measured age indicated that 94% of the estimated ages were within 10 y of the measured age. Most of the trees with estimated ages were too tall for accurate whorl counts with stems too small for coring.

Some trees of all species were between 40 and 50-y old, but few trees were [greater than]50-y old (Fig. 3). Eight trees had measured ages [greater than]50-y and 5 trees had estimated ages [greater than]50y. The increment cores obtained from large trees outside the sample plots also indicated few trees [greater than]5O-y old. Of 36 large trees, 7 were [greater than]50-y old, and none were [greater than]60-y old.

Age distributions varied among species and indicated different patterns of colonization. For Abies magnifica there were more individuals in the [greater than]20-y old to [leq]40-y old categories than in the younger age categories. This suggests (based on these ages being measured in 1987) that the most successful period of colonization for this species occurred around 1955. The age distribution for A. concolor is similar to that for A. magnifica and indicates a peak colonization around 1960. The age distribution for Pinus contorta has a greater proportion of individuals [leq]30-y old than [greter than]30-y old, which indicates that colonization rates for these species increased beginning about 1960 and have remained relatively high. The patterns for the age distributions of P. jeffreyi P monticola and Tsuga mertensiana are less clear but indicate continuing colonization from at least 1940 to the present. Although the proportion of trees with estimated ages are large for some species and age classes (e.g. the 3l-y to 40-y old A. magnifica) the large proportion of estimated ages that were within 10-y of the measured ages suggests that potential errors in estimated ages do not greatly affect the interpretation of age distributions and colonization patterns.

Tree heights.--Despite the large numbers of trees [greater than]20-y old, most trees were relatively short (Table 1). Thirty-two percent were [less than]2-in tall, 98% were [leq]10-m tall and all the sampled trees were [less than]16-m tall. For Pinus contorta, Abies magnifica and Pinus monticola mean tree heights were [less than]3-m and median tree heights were [less than]2-m. Large numbers of small trees might be expected for P. contorta because of the large proportion of young trees (Fig. 3), but the preponderance of small trees for A. magnifica is surprising given that most individuals of this species were [greater than]2O-y old. The relationship between height and age for A. magnifica with measured ages (Fig. 4) indicated many stunted older trees. Older trees with stunted growth were also noted for P contorta and P monticola.

Basal area.--Most trees also had relatively small diameters. The percent of individuals with dbh [geq]0.08-m for Pinus contorta, Abies magnifica and P monticola were 8, 16 and 23%, respectively. At least 50% of the individuals for all species were either (1) [less than]1.37-m tall or (2) had a dbh [less than]0.025 m. The mean ([pm]sD) total basal area per plot was 8.8 [pm] 7.5 [m.sub.2] [ha.sup.-1], and P contorta accounted for one-half of this total. Basal area per plot (Table 2) increased significantly (F = 5.02; df = 3, 66; P [less than] 0.01; ANOVA using logarithmically transformed basal areas) from transect H1 to transects H2, H3 and H4. Mean basal areas per plot were similar for P contorta, A. magnifica and P. monticola on transects H1 and H2. On transects H3 and H4 the mean basal area for P. contorta was more than 3X that for the other species.

Importance values.--Importance values for the three most common species (Fig. 5) ranged from [greater than]170 for Pinus contorta on H4 to [less than]50 for Abies magnifica on H3 and H4. The values for the remaining species were always [less than]30. Pinus contorta had the greatest importance value on all transects except Hi where the three species had similar values. The importance values for P. contorta increased from H1 to H4, whereas the values for A. magnifica and P. monticola declined.

Photographic evidence of tree invasion.--Trees were not discernible on oblique photography taken in 1930 (photograph number LAVO 406 in the LVNP archives) and 1934 (unnumbered photograph in the LVNP archives) but were discernible in a 1941 aerial photograph (number C. X. W. 11-63 in the LVNP archives) and a 1949 oblique photograph taken by Ansel Adams (Stillman and Turnage, 1992). Small trees were also evident on the debris flow in an undated oblique photograph (number LAVO 400 in the LVNP archives) that appears to have been taken about 1940 based on the correspondence of details between this and other photographs dated from 1940. Aerial photography from 1952 (Heath, 1967), 1966 (CFL 208.2743 in the LVNP archives), 1973 (LAO 2 09089 in the LVNP archives) and 1988 (LAVO 3-4-7794 in the LVNP archives) show continual increases in the numbers and sizes of trees on the debris flow. The photographic sequence also shows the development of trees in the surrounding areas that were denuded by blast effects. Large tree s were absent in the 1915 photography but were evident in the 1941 and 1949 photographs.

DISCUSSION

The species composition of the debris flow forest is typical of the upper montane coniferous forest (Barbour, 1988; Rundel et al., 1990). This forest type is described as a mixture of Pinus contorta, Abies maginifica and Pinus monticola occurring at elevations from 1800-m to 2400-m in the northern Sierra Nevada Mountains and the southern Cascade Range (Barbour, 1988). Species that are characteristic of the mid-montane conifer forest (i.e., Abies concolor) and the subalpine forest (i.e., Tsuga mertensiana) are only minor components of the debris flow forest. The forest composition on transects H2, H3 and H4 is similar to that given by Barbour (1988) for the lodgepole pine phase of the upper montane coniferous forest where P contorta Importance Values may exceed 200 (Barbour, 1988).

The forest composition and distribution on the debris flow are also similar to those described for these elevations and terrains in LVNP by Parker (1991). Parker's (1991) studies indicate that Pinus contorta dominates on areas with slopes [less than]4[degrees] between 1900-m and 2300-m, whereas mixtures of Abies magnifica and P. monticola dominate on steeper slopes. The change in relative species compositions on the debris flow from the steeper to the more level transects parallels the pattern observed by Parker (1991).

Although the species composition of the debris flow is typical for these elevations and terralns, the forest basal areas are smaller than the 50 to 60-[m.sup.2] [ha.sup.-1] reported for Pinus contorta forests (Barbour, 1988; Parker, 1991) and the 70-[m.sup.2] [ha.sup.-1] reported for Abies magnifica forests (Parker, 1991) on similar elevations and terrains. The smaller basal area results, in part, from the large number of young trees but is also due to the large numbers of old, small and apparently stunted trees. Height growth for P. contorta, A. magnifica and Pinus monticola on the debris flow are among the slowest reported for these species. Twenty-year old P. contorta and P. monticola may be expected to reach heights [greater than]7-m on good soils and [geq]2-m on poor soils (Graham, 1981; Lotan and Critchfield, 1981; Koch, 1996). Fifty-year old A. magnifica can reach 18-m tall on good sites and 3-m tall on poor sites (Laacke, 1981; Rundel et al., 1990).

Tree growth has been greater where the debris flow is underlain by soil than on the area underlain by lahar deposits. Heath (1967) reported mean tree heights of 3.8 m for Pinus monticola and 1.4-m for P. contorta in 1964 for the area underlain by soils. These heights are similar to or greater than those observed in 1987 on areas underlain by lahar deposits. Fessenden (1984) reports forest biomass estimates of 1.9 kg [m.sup.-2] for the area overlying soils but only 0.6 kg [m.sup.-2] for the area over lahar deposits. Whether the greater growth in this area is due to (1) the facilitating effect of the remnant, decaying tree stumps and trunks on conifer establishment (Frenzen et al.,

1988), (2) the effects of the underlying soil (Frenzen and Franklin, 1985; Larsen and Bliss, 1998), or (3) some other cause remains to be determined.

Older stunted trees occurred in the same plots as younger more vigorous trees which suggests considerable within-plot variation in microsite quality. This is consistent with (1) the observations of within-plot heterogeneity in slopes and aspect and (2) the presence of protruding dacite blocks that can modify microsite environments. The presence of large, relatively young trees also indicates that the earliest individuals to colonize the surface are not necessarily those that will eventually dominate the forest canopy or produce seeds for succeeding generations of colonizers. The opportunity for successfully invading and reaching a place in the canopy appears to last for [geq]30-y after the first successful colonization.

The composition of the debris flow forest is similar to that described by Bailey (1963) for belt transects sampled in 1962 and by Fessenden (1984) for plots sampled in 1982. Where Bailey's (1963) transects cross the debris flow, he describes (1) increasing Pinus contorta abundance with declining elevation, (2) maximum tree heights [leq]6.1-m and (3) tree densities approximately one-half those observed in 1987. The increasing P. contorta abundance is consistent with the increasing density observed from H1 through H4 (Fig. 2). The [leq]6.1-m tree heights indicate that trees in 1962 were approximately one-half as tall as in 1987 (Table 1). The lower densities are consistent with the recruitment of new trees between 1962 and 1987 (Fig. 3). Fessenden (1984) reports densities of 1946, 602 and 426 trees [ha.sup.-1], respectively, for P. contorta, Abies magnifica and P. monticola in her Sparse Central Forest, which corresponds to the areas of transects H2, H3 and H4. These densities are similar to those observed on these transects in 1987.

The age distributions (Fig. 3) are similar to the pattern of tree invasion apparent in the historical photographs. Both indicate little successful colonization of the debris flow before 1935. Small seedling trees could have established on the debris flow before 1935 and not have been discernible in the photography, but there is no evidence from the age distributions that large numbers of these trees survived until 1987. The age distributions and photographic records are also consistent with Bailey's (1963) qualitative observation of tree ages being [leq]20-y old on the debris flow area in 1961. Thus, there was little successful colonization by tree species for the first 2O-y after the eruption. If tree seedlings did establish during this period, they did not survive in large numbers.

The absence of successful tree colonization during the first 20-y may have several explanations. First, the lack of invasion may simply reflect the lack of seed sources. Photographic evidence from July 1915 (Day and Allen, 1925; Loomis, 1926) shows that trees surrounding the debris flow were destroyed by the blast. Trees do not begin to occur on the debris flow until large trees appear in photographs of the areas surrounding the debris flow. Moreover, the increased colonization rates in the 1950s are consistent with Bailey's (1963) observation of large seed-producing individuals of all six tree species growing at several locations surrounding the debris flow in 1961. These large individuals either grew from small survivors of the blast or from seeds in relatively undisturbed soils. The importance of proximity to seed sources for conifer invasion of lahar surfaces has been documented by del Moral (1998) who observed more rapid invasion of conifers on lahar surfaces nearer to existing forests than on those loc ated farther away. Frenzen et al. (1988), Gecy and Wilson (1990) and Larsen and Bliss (1998) have observed conifer invasion within the first lO-y where lahar and mudflow surfaces were near seed sources.

Second, the lack of successful colonization may be due to the presence of ash (or tephra) deposits on the surface of the debris flow. Contemporary photographs indicate the presence of ash deposits in 1915 (Fig. 42; Loomis, 1926), and further ash falls continued into 1917 (Day and Allen, 1925). Until these ash deposits were weathered, dispersed or eroded by wind and rain, the surface may not have been appropriate for seedling germination or survival. Although tree seedlings can successfully establish in ash deposits under existing tree canopies (Antos and Zobel, 1986), they may have limited success on ash surfaces fully exposed to sun and wind unless their roots have access to underlying soils (Frenzen and Franklin, 1985; Stevens et al., 1987).

Third, the physical surface may have been appropriate, but conifer invasion was reduced by physiological stress or nutrient limitation. Simple physical weathering increases the suitability of lahar and pumice deposits to support vegetation (del Moral and Clampitt, 1985), and this suitability is further enhanced by invading herbaceous vegetation. Nitrogen-fixing lupines and other herbaceous species have been shown to contribute to soil development at Mount St. Helens (Halvorson et al., 1991; Titus and del Moral, 1998) and facilitate, at least in part, the establishment of other species (Morris and Wood, 1989; Titus and del Moral, 1998). Balley (1963) reports the presence in 1961 and 1962 of several lupine species, including Lupinus andersonii S. Watson, and other herbs and shrubs throughout the blast, lahar and debris areas, and Fessenden (1984) observed standing crops of approximately 3 g [m.sup.-2] on the debris flow for L. andersonii and [less than] 2 g [m.sup.-2] for all other herb and shrub species. Thes e plants may have played a role in the initial establishment of trees on the debris flow through their effects on soil fertility and microclimates. Even the moderate shade provided by sparse herbaceous vegetation may have reduced the stressful effects of irradiance and soil temperature on Abies magnifica seedlings (Ustin et al., 1984; Selter et al., 1986).

Unfortunately, no studies were performed in the 1930s to identify the processes and events that permitted initial successful colonization or in the 1950s to identify the processes resulting in increased rates of colonization. Hopefully continuing studies at Mount St. Helens will identify the factors affecting initial tree invasion and colonization.

Most analyses of forest revegetation of lahar and debris flow surfaces have occurred where these surfaces are relatively narrow corridors (e.g., Halpern and Harmon, 1983; Frenzen et al., 1988) surrounded by intact forests whose seed production can dominate the composition of the developing forest (e.g., Gecy and Wilson, 1990). Because blast effects removed surrounding forests (especially those at higher elevations), the debris flow is more isolated from forest seed sources, and this isolation suggests more similarity to the patterns of herbaceous revegetation observed at higher elevations at Mount St. Helens. In a review of this herbaceous revegetation, del Moral and Bliss (1993) comment on the importance of (1) chance events such as the role of isolated survivors and (2) the greater relative invasion by ruderal (sensu Grime, 1977) species due to their more effective seed dispersal. At Lassen Peak chance survival of blast effects by isolated conifers is suggested by the later photographs and by the descripti ons of Bailey (1963), and Pinus contorta with its readily-dispersed, extremely small seeds (Krugman and Jenkinson, 1974; Lotan and Critchfield, 1981; Turner, 1985) has been the most successful invader.

Despite the chance survival of isolated trees to serve as potential seed sources and despite the greater success of the more ruderal Pinus contorta, the pattern of forest development on the debris flow appears to be more predictable than the multiple pathways observed in some other systems composed of a similar number of invading species (McCune and Allen, 1985; Fastie, 1995). The species compositions and distributions across terrain on the debris flow are similar to those for older forests at LVNP and suggest the eventual development of the expected and predicted forests. This similarity and predictability may simply reflect the slow growth rates and, consequently, longer period for invasion and establishment. The debris flow forest in 1987 represented the summation of 40+ y of varying seed inputs and seedling survivals. If growth rates had been faster and the colonization period shorter there may have been a greater chance that founder effects due to random variations may have established less predictable forests (e.g., Fastie, 1995).

Acknowledgments. -- The study was conducted in Lassen Volcanic National Park with the permission and assistance of Gilbert Blinn, Alan Denniston and Steven Zachary. Financial support was provided by the A. W. Mellon Foundation, the U.S. National Park Service, the Texas Christian University Research Foundation and the U.S. Department of Energy's Education and Outreach Program through Financial Assistance Award DE-FC09-96SR18546 with the University of Georgia. We thank S. Basham, L. Beatty, A. Cook, J. Ebe, S. Kroh, H. Edwards Lytle, A. M. Basham May and T. Bertram Rea for their assitance in the field.

(1.) Present address: Department of Biology, P.O. Box 97388, Baylor University, Waco, Texas 76798-7388

LITERATURE CITED

ADAMS, A. B. AND V. H. DALE. 1987. Vegetative succession following glacial and volcanic disturbances in the Cascade mountain range of Washington, U.S.A., p. 70-147. In: D. E. Bilderbeck (ed.). Mount St. Helens 1980: botanical consequences of the explosive eruptions. University of California Press, Berekely, Calif. 360 p.

ANTOS, J. A. AND D. B. ZOBEL. 1986. Seedling establishment in forests affected by tephra from Mount St. Helens. Am. J. Bot., 73:495-499.

BAILEY, W. H. 1963. Revegetation in the 1914-15 Devastated Area of Lassen Volcanic National Park. Ph.D. Thesis. Oregon State University, Corvallis. 195 p.

BARBOUR, M. G. 1988. California upland forests and woodland, p. 131-164. In: M.G. Barbour and W. D. Billings, (eds.). North American terrestrial vegetation. Cambridge University Press, Cambridge.

BATES, R. L. AND J. A. JACKSON (EDs.). 1984. Dictionary of geological terms, 3rd edition. Doubleday, New York, N.Y 571 p.

BEARDSLEY, C. F. AND W. A. CANNON. 1930. Note on the effects of a mud-flow at Mt. Shasta on the vegetation. Ecology, 11:326-336.

BOX, G. E. P., W. G. HUNTER AND J. S. HUNTER. 1978. Statistics for experimenters. John Wiley and Sons, New York, N.Y 653 p.

CHAPIN, D. M. AND L. C. BLISS. 1989. Seedling growth, physiology, and survivorship in a subalpine, volcanic environment. Ecology, 70:1325-1334.

DALE, V. H. 1989. Wind dispersed seeds and plant recovery on the Mount St. Helens debris avalanche. Can. J. Bot., 67:1434-1441.

DAVIS, R. R, F. S. FOOTE, J. M. ANDERSON AND E. M. MIKHAIL. 1981. Surveying theory and practice, sixth edition. McGraw Hill Book Co., New York, N.Y 992 p.

DAY, A. L. AND E. T. ALLEN. 1925. The volcanic activity and hot springs of Lassen Peak. Carnegie Institution of Washington, Washington, DC. 190 p.

DEL MORAL, R. 1998. Early succession on lahars spawned by Mount St. Helens. Am. J. Bot., 85:820-828.

----- AND L. C. BLISS. 1993. Mechanisms of primary succession: Insights resulting from the eruption of Mount St. Helens. Adv. Ecol. Res., 24:1-66.

----- AND D. M. WOOD. 1993. Early primary succession on the volcano Mount St Helens. J Veg. Sci., 4:223-234.

----- AND C. A. CLAMPITT. 1985. Growth of native plant species on recent volcanic substrates from Mount St. Helens. Am. Midl. Nat., 114:374-383.

DRAPER, N. R. AND H. SMITH. 1981. Applied regression analysis. John Wiley and Sons, Inc., New York, N.Y. 709 p.

EGGLER, W. A. 1941. Primary succession on volcanic deposits in southern Idaho. Ecol. Monogr., 11:277-298.

EPPLER, D. B. 1984. Characteristics of volcanic blasts, mudflows and rock-fall avalanches in Lassen Volcanic National Park, California. M. S. Thesis. Arizona State University, Tempe. 261 p.

FASTIE, C. L. 1995. Causes and ecosystem consequences of multiple pathways of primary succession at Glacier Bay, Alaska. Ecology, 76:1899-1916.

FESSENDEN, J. E. 1984. Forest biomass and production estimates for the Devastated Area, Lassen Volcanic National Park, California. M .S. Thesis. Humbolt State University, Arcata. 49 p.

FINCH, R. H. 1937. A tree ring calendar for dating volcanic events at Cinder Cone, Lassen National Park, California. Am. J. Sci., 233:140-146.

FRENZEN, P. M. AND J. E FRANKLIN. 1985. Establishment of conifers from seed on tephra deposited by the 1980 eruptions of Mount St. Helens, Washington. Am. Midl. Nat., 114:84-97.

-----, M. E. KRASNY AND L. P. RIGNEY. 1988. Thirty-three years of plant succession on the Kautz Creek mudflow, Mount Rainier National Park, Washington. Can. J. Bot., 66:130-137.

GECY, J. L. AND M. V. WILSON. 1990. Initial establishment of riparian vegetation after disturbance by debris flows in Oregon. Am. Midi. Nat., 123:282-291.

GRAHAM, R. T. 1981. Western white pine, p. 385-394. In: R. M. Burns and B. H. Honkala (eds.). Silvics of North America 1. Conifers. U.S. Department of Agriculture, Washington, DC. Agricultural Handbook 654. 675 p.

GREIG-SMITH, P. 1964. Quantitative plant ecology. Butterworths, London. 256 p.

GRIFFIN, J. R. 1967. Soil moisture and vegetation patterns in northern California forests. Pacific Southwest Forest and Range Experiment Station, Berkeley, Calif. U.S. Forest Service Research Paper PSW-46. 22 p.

GRIME, J. P. 1997. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. Am. Nat., 111:1169-1194.

HALPERN, C. B., P. M. FRENZEN, J. E. MEANS AND J. F. FRANKLIN. 1990. Plant succession in areas of scorched and blown-down forest after the 1980 eruption of Mount St. Helens, Washington. J. Veg. Sri., 1:181-194.

----- AND M. E. HARMON. 1983. Early plant succession on the Muddy River mudflow, Mount St. Helens, Washington. Am. Midi. Nat., 110:97-106.

HALVORSON, J. J., J. L. SMITH AND E. H. FRANZ. 1991. Lupine influence on soil carbon, nitrogen and microbial activity in developing ecosystems at Mount SL Helens. Oerologia, 87:162-170.

HARRIS, S. L. 1988. Fire mountains of the west: The Cascade and Mono Lake volcanoes. Mountain Press Publishing, Missoula, Mont. 379 p.

HEATH, J. P. 1967. Primary conifer succession, Lassen Volcanic National Park. Ecology, 48:270-275.

HICKMAN, J. C. (ED.). 1993. The Jepson manual. University of California Press, Berkeley, Calif. 1400 p.

KIVER, E. P. 1982. The Cascade volcanos; Comparison of geological and historical records, p. 3-12. In: S. A. C. Keller (ed.). Mount SL Helens: one year later. Eastern Washington University Press, Cheney, Wash.

KRUGMAN, S. L. AND J. L. JENRINSON. 1974. Pinus L., p. 598-638. In: C. S. Schopmeyer (ed.). Seeds of woody plants in the United States. Forest Service, U.S. Department of Agriculture, Washington, DC. 883 p.

KOCH, P. 1996. Lodgepole pine in North America, Vol 1. Forest Products Society, Madison, Wis. 343 p.

LAACKE, R. J. 1981. California red fir, p. 71-79. In: R. M. Burns and B. H. Honkala (eds.). Silvics of North America 1. Conifers. U.S. Department of Agriculture, Washington, DC. Agricultural Handbook 654. 675 p.

LARSEN, D. R. AND L. C. BLISS. 1998. An analysis of structure of tree seedling populations on a lahar. Landscape Ecol., 13:307-322.

LOOMIS, B. F. 1926. Pictorial history of the Lassen volcano. Loomis Museum Association, Mineral, Calif. 96 p.

LOTAN, J. E. AND W. B. CRITCHFIELD. 1981. Lodgepole pine, p. 302-315. In: R. M. Burns and B. H. Honkala (eds.). Silvics of North America 1. Conifers. U.S. Department of Agriculture, Washington, DC. Agricultural Handbook 654. 675 p.

MEANS, J. E., W. A. MCKEE, W. H MOIR AND J. F. FRANKLIN. 1982. Natural revegetation of the northeast portion of the devastated area, p. 93-103. In: S. A. C. Keller (ed.). Mount St. Helens: One year later. Eastern Washington University Press, Cheney, Wash.

MCCUNE, B. AND T. F. H. ALLEN. 1985. Will similar forests develop on similar sites? Can. J. Bot., 63:367-376.

MILLIKEN, C. A. AND D. E. JOHNSON. 1984. Analysis of messy data, volume 1: Designed experiments. Van Nostrand Reinhold Co., New York, N.Y 473 p.

MORISS, W. F. AND D. M. WOOD. 1989. The role of lupine in succession on Mount St. Helens: Facilitation or inhibition? Ecology, 70:697-703.

PARKER, A. J. 1991. Forest/environment relationships in Lassen Volcanic National Park, California, U.S.A. J. Biogeography, 18:543-552.

PHIPPS, R. L. 1985. Collecting, preparing, crossdating, and measuring tree increment cores. U.S. Geological Survey, Washington, DC. Water Resources Investigator Report 85-148. 48 p.

RUNDEL, P. W., D. J. PARSONS AND D. T. GORDON. 1990. Montane and subalpine vegetation of the Sierra Nevada and Cascade Ranges, p. 559-599. In: M. G. Barbour and J. Major. (eds.). Terrestrial vegetation of California. California Native Plant Society, Sacramento, Calif.

RUSSELL, I. C. 1897. Volcanoes of North America. The Macmillan Company, London, 346 p.

SAS INSTITUTE. 1989. SAS/STAT user's guide, version 6, 4th ed., Volume 2. SAS Institute, Cary, N.C. 846 p.

SCARTH, A. 1994. Volcanoes. Texas A & M University Press, College Station, Tex. 273 p.

SELTER, C. M. AND W. D. PITTS. 1986. Site microenvironment and seedling survival of Shasta red fir. Am. Midi. Nat., 115:288-300.

STEVENS, R. C., J. K. WINJUM, R. R. GILCHIUST AND D. A. LESLIE. 1987. Revegetation in the western portion of the Mount St. Helens blast zone during 1980 and 1981, p. 210-245. In: D. E. Bilderbeck (ed.). Mount St. Helens 1980: botanical consequences of the explosive eruptions. University of California Press, Berkeley, Calif. 360 p.

STILLMAN, A. G. AND W. A. TURNAGE (EdS.). 1992. Ansel Adams: Our National Parks. Little Brown and Co., Boston, Mass. 127 p.

SUDWORTH, G. B. 1927. Forest trees of the Pacific slope. Dover, N.Y 455 p.

TILLING, R. I., L. TOPINKA AND D. A. SWANSON. 1990. Eruptions of Mount SL Helens: Past present, and future, revised edition. U.S. Government Printing Office, Washington, DC.

TITUS, J. H. AND R. DEL MORAL. 1998. Seedling establishment in different microsites on Mount St. Helens, Washington, USA. Plant Ecology, 134:13-26.

TURNER, D. P. 1985. Successional relationships and a comparison of biological characteristics among six northwestern conifers. Bull. Torrey Bot. Club, 112:421-428.

USTIN, S. L., R. A. WOODWARD, M. G. BARBOUR AND J. L. HARFIELD. 1984. Relationships between sun-fleck dynamics and red fir seedling distribution. Ecology, 65:1420-1428.

VIERS, S. D. JR. 1987. Response of vegetation within the blast zone, Mount St. Helens, p. 228-245. In: D. E. Bilderbeck (ed.). Mount St. Helens 1980: botanical consequences of the explosive eruptions, University of California Press, Berkeley, Calif. 360 p.

ZOBEL, D. B. AND J. A. ANTOS. 1997. A decade of recovery of understory vegetation buried by volcanic tephra from Mount St. Helens. Ecol. Monogr, 67:317-344.
              Heights (m) for tree species on the debris flow
                                                              Percent
Species             n  Mean SD  Minimum Median Maximum [greater than]5-m tall
Pinus contorta    1674 1.4  1.8  0.03    0.8    13.2             5.8
Abies magnifica    351 1.8  1.8  0.04    1.2    14.7             7.1
Pinus monticola    280 2.5  2.8  0.02    1.7    15.1            15.7
Abies concolor      57 1.6  1.9  0.20    1.1    11.3             7.0
Tsuga mertensiana   47 1.1  2.3  0.10    1.2     5.7             2.1
Pinus jeffreyi      46 3.2  3.4  0.40    2.0    14.9            21.7


Means [pm] SD basal area ([m.sup.2] [ha.sup.-1]) for the major species and the total for all species combined on transects H1 through H4. Means for the basal area of Pinus contoria and the total basal area that are denoted with the same letters were not significantly different at P [leq] 0.05 in an analysis of logarithmically transformed data. There were no significant differences among means for Abies magnifica or Pinus monticola. n = number of plots per transect
                        Transects
                           H1            H2             H3
Species                  n = 14        n = 21         n = 17
Pinus conlorta        0.7 [pm] 1.5a 2.6 [pm] 2.0b  7.5 [pm] 5.8b
Abies magnifica       1.2 [pm] 1.6  2.3 [pm] 2.5   1.0 [pm] 3.1
Pinus monticola       1.4 [pm] 1.0  2.5 [pm] 2.1   1.4 [pm] 1.7
Total for all species 3.4 [pm] 2.0a 8.4 [pm] 4.9b 12.0 [pm] 9.0b
                             H4
Species                    n = 18
Pinus conlorta         6.9 [pm] 6.0b
Abies magnifica        0.8 [pm] 0.9
Pinus monticola        2.0 [pm] 4.4
Total for all species 10.0 [pm] 9.0b
COPYRIGHT 2000 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:KROH, GLENN C.; WHITE, JOSEPH D.; HEATH, SHELLY K.; PINDER III, JOHN E.
Publication:The American Midland Naturalist
Geographic Code:1U9CA
Date:Jan 1, 2000
Words:7081
Previous Article:Overstory Composition and Stand Structure Influence Herbaceous Plant Diversity in the Mixed Aspen Forest of Northern Minnesota.
Next Article:Germination Ecology of a Federally Threatened Endemic Thistle, Cirsium pitcheri, of the Great Lakes.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters