YOSEMITE Vegetational Changes in Yosemite Valley NPS Occasional Paper No. 5

Previous sections describe the soils and vegetation samples from Yosemite Valley in quantitative terms. The following sections examine vegetational descriptions more broadly, assess vegetational changes, and discuss the implications this intensely used recreational site has for management.

Forest Vegetation

Seven species of trees characterized the forests in Yosemite Valley: Libocedrus decurrens, Quercus kelloggii, Pinus ponderosa, Quercus chrysolepis, Pseudotsuga menziesii, Abies concolor, and Umbellularia californica. These grew on all the landforms from the sandy soil without profile development near the Merced River to the top of the talus slopes. L. decurrens, Q. kelloggii, P. ponderosa, and A. concolor tended to dominate the valley floor in the size of plants and the number of plants per unit of area. Quercus chrysolepis, U. californica, and P. menziesii dominated on the upper talus slopes. L. decurrens and P. ponderosa dominated on the recent alluvium. Populus trichocarpa, Alnus rhombifolia, and Salix (willow) were restricted to the soils of the valley floor, usually near water or on the banks of former rivers. Pinus lambertiana, Acer glabrum, Arctostaphylos viscida (whiteleaf manzanita), Ceanothus intergerrimus (deer brush), and Sambucus caerulea (elderberry) occurred only on the fans and talus. Two species, Pinus contorta (lodgepole pine) on recent alluvium near the Merced River and Abies magnifica on moraine, appeared essentially out of their normal range. They may have become established from seed that drifted or washed down from the canyon tops 900 m (3,000 feet) or more above the valley.

Crown cover by woody plants amounted to approximately 55% of the area and varied from 41 to 68% by landform. The densest forest occurred on the talus and the thinnest on recent alluvium near the Merced River. Herbaceous cover under forest canopies was highest (41%) on the fine alluvial sands and least on the moraines, fans, and talus. Although forest litter tended to be least near the river and on the upper talus, in these areas litter covered more than 70% of the soil surface. Except in areas heavily trampled by people, no accelerated erosion was found. Litter and live vegetation covered the soil sufficiently to prevent erosion.

Height classes of woody plants indicate the timing of forest regeneration. Because height classes only approximate age, no claim is made for definite years of seedling establishment. However, it is reasonable to assume that a tree species is reproducing itself if individuals are present in a height range from less than 30 cm (12 inches) to tall, mature trees. Immature conifers have pointed tops. Rounded tops indicate maturity. Evaluation of maturity in broad leaved species was most subjective, as indicated by size and vigor. The objectives of national park administrators at the time of field sampling prevented extensive increment borings and cutting of woody plants to determine age.

The earliest sources used in this study indicated that the forest has expanded and thickened in Yosemite Valley during the last 100 years. Paired photographs taken 107 years apart show this vividly (Fig. 14). A study of annual rings of Pinus ponderosa, Libocedrus decurrens, and Abies concolor growing on former meadow areas indicates that invasion of the meadows and thickening of the forest began between 1860 and 1870 (Gibbens and Heady 1964). Those researchers also found that the invasion was continuous because trees of many ages less than 100 years were found. Ernst (1943, 1949) recorded the invasion of trees into meadows, and the commissioners' report (1882) also recognized tree invasion. There appeared to be scattered trees of two earlier age groups, one established about 1800 and the other about 1700 (Reynolds 1959). Decimation of Indian populations by disease coincided roughly with the two oldest age classes of trees. Frequent burning by Indians probably prevented tree regeneration during intermediate times.

Fig. 14. (Upper) Yosemite Valley from Glacier Point in 1866 (photographed by C. E. Watkins) shows a wet meadow at A near photograph center and open forest on alluvial fan materials at B. (Lower) A 1973 photo of the same area (by J. Bartolome) shows less meadow, greatly increased forest cover, and buildings in the location of B.

Individual plants of Libocedrus decurrens, Quercus kelloggii, and Pinus ponderosa occurred in all height classes on all landforms in Yosemite Valley. These species reproduced in all areas and continue to dominate the forest. Mature individuals of Quercus chrysolepis, Pseudotsuga menziesii, and Abies concolor were limited to the fans and talus. Shorter and presumably younger trees occurred in abundance on the alluvial soils of the valley floor, which suggests that they are increasing rapidly along with L. decurrens in former meadows or open forests. Much of the increase in forest area and density was attributed to the last four species mentioned above. Although short and shrubby on the alluvial soils, Umbellularia californica and Quercus chrysolepis may become dominant trees on the valley floor.

The number of plants of each tree species declined in succeedingly taller height classes. Exceptions indicated periods of establishment and changes in the relative dominance of the species. Pinus ponderosa under 15 m (50 ft) in height were less numerous than those over 15 m and immature trees over 15 m were more numerous than mature ones. P. ponderosa appeared to be diminishing in importance in the valley forests. Abundance of seedlings and mature Quercus kelloggii with fewer trees of intermediate sizes suggests that the rate of establishment has been erratic; however, it too could be decreasing relative to other trees, but enough survive to maintain the species. The scarcity of mature Abies concolor and abundance of immature sizes indicated that this species is rapidly becoming more important, especially on the alluvial soils. Pseudotsuga, Umbellularia, and Quercus chrysolepis also were becoming more important on the valley floor but their frequency was generally low. Pinus contorta and Abies magnifica, rarely present in the 1.8-to 6-m (6-20 ft) height class, probably became established only through a peculiar chain of events several decades ago.

Forest vegetation in the valley at the time of discovery consisted of widespread trees of which Quercus kelloggii and Pinus ponderosa were the most numerous individuals. The openness of forest and dominance by the two species probably resulted from periodic fires and the efforts of Indians to maintain orchards of Quercus for acorns. Both these factors have been greatly reduced for over 100 years. Although man's activities have influenced the valley's vegetation in many ways, the reduction of fire has permitted tree species to invade the meadows and alter species relationships in terms of dominance and density. Stand structures show that Libocedrus decurrens, Abies concolor, Pseudotsuga menziesii, and Quercus chrysolepis are present in all size classes. If no catastrophe or managerial impact causes their demise, the forest stands in the valley will become dominated increasingly by these species in the coming decades. The data collected in this study suggest that these four species will not replace Q. kelloggii and P. ponderosa but will combine with them to form a mixed, dense conifer-oak forest.

Forest Understory Vegetation

Understory herbaceous vegetation in Yosemite Valley exhibited relationships to landforms (Table 2). Some 81 herbaceous species were recorded in the understory vegetation. The most widespread and common species were Pteridium aquilinum, Elymus glaucus, Potentilla glandulosa, and Galium bolanderi. Widespread species with low frequency in cluded Stipa columbiana (needlegrass), Bromus marginatus (mountain brome), Bromus tectorum, Apocynum cannabinum (Indian hemp), Artemisia douglasiana, and Asarum hartwegii.

Understory herbaceous cover on the flood plain soils was 20-40%, which is somewhat higher than expected under coniferous forest. This was caused by occasional deposition of silt which had a fertilizer effect, tree crown cover which tended to be less than in the forests on the valley sides, and small, irregular forest stands which permitted side lighting. The herbaceous vegetation in these conditions resembled the meadows, whose dominants were Agrostis alba, Holcus lanatus, Poa pratensis, Carex spp., Mentha arvensis, and Rumex acetosella. On the fans and talus slopes, Poa scabrella, Melica imperfecta, Festuca occidentalis (western fescue), Juncus balticus, Dryopteris arguta, Asarum hartwegii, and Penstemon newberryi characterized the thin understory cover.

Deer Browsing

Although wild animals, especially deer, are a great asset to the recreational resources in Yosemite Valley, overbrowsing by deer has been an intermittent problem for several decades. At times subtle live trapping and removal programs have kept their numbers in check. The meadows show little influence from deer grazing although the animals are commonly seen in the grasslands. Browsing is their principal impact on the valley vegetation because it influences the rate of increase and spread of woody species into the grasslands as well as regeneration of woody species within the forest (Fig. 15). Severely hedged young trees and shrubs occur at the meadow edges. The presence or absence of browsing on the different species was recorded on plots 15 m in diameter.

Fig. 15. Heavily browsed Abies concolor with young unbrowsed Libocedrus decurrens and Pinus ponderosa in the background. Browsed young trees were numerous along the edges of the meadow.

The kinds and relative availability of plants are two of the factors that influenced selectivity of foods by grazing animals. In Table 3 columns labeled A under each landform give the percentage of plots that contained the species in the height classes 0-0.3 m and 0.3-1.8 m. This is a measure of availability. Columns labeled B give the percentage of those plots where the species showed evidence of browsing. The order of the species (Table 3) gives their decreasing importance on the basis of frequency. Two of the dominant species (Libocedrus decurrens and Pinus ponderosa) were seldom browsed. Quercus kelloggii and Quercus chrysolepis were browsed on all landforms, but probably at greater intensities than the frequency data indicate because of their high palatability. Browsing of oaks kills seedlings, leads to the development of browse lines, and reduces the volume of acorns. Deer selected Abies concolor and Pseudotsuga menziesii more than the other conifers. These two species, the two Quercus species, and Umbellularia californica were the major browse plants in Yosemite Valley because of their wide spread distribution. Other palatable species that deer selected but that did not occur widely included Populus trichocarpa, Alnus rhombifolia, Salix, and Ceanothus intergerrimus.

Table 3. Availability of browse (A) expressed as a percentage of plots with plants less than 1.8 m tall on each landform and percentage of those plots which showed browsing (B).

Of the tree species only Abies concolor showed severe and widespread evidence of browsing by deer; however the full range of size classes on all landforms indicated increasing importance of Abies. While deer undoubtedly affected regeneration and spread of the tree species differentially, little conspicuous evidence existed that the recent level of deer populations had affected unduly the woody vegetation. With the possible exception of Ceanothus integerrimus, the range of size classes from seedlings to plants beyond the reach of deer was unbroken for all woody species. Young Quercus kelloggii were present throughout the forest types and Quercus chrysolepis had spread from the fans and talus onto alluvial sites.

Meadow Vegetation

Most meadows in Yosemite Valley border the Merced River. The principal soils developed from alluvial sands which originated from parent materials with the same mineralogical composition. Similar climate occurred throughout the valley without any apparent relationship to vegetational type. Slopes of the meadows were less than 3%. None of the meadow soils approached maturity. They varied in texture from loam to sand <2 mm (0.08 inches) in diameter, in structure from single-grain to weak medium-blocky, and in consistency from loose to slightly hard when dry. Acidity averaged about pH 6 with variation from pH 5 to 7 except for the litter layers which often were more acid than pH 5. Soils with developed profiles of color and organic matter did not have a textural B horizon and recent alluvium existed without profile development. None of these variations in soil characteristics seems sufficient to cause the observed differences in botanical com position of the vegetation.

Drainage or water table levels appeared to be the main cause of different vegetational types within the meadows (Table 4). Where the water table was closer than 75 cm (30 inches) to the soil surface, or perhaps as deep as 1.5 m (5 ft) in the dry season, the top soil horizons were high in organic matter, dark gray to dark grayish-brown in color, and some were slightly gleyed. These soils were usually in closed depressions or seeps and their shape often indicated abandoned river channels. Some contained shallow water throughout the year.

Table 4. Average percentage species composition of meadow vegetation arranged from wet sites on the left to dry sites on the right.

Vegetation on the imperfectly drained soils, poorly drained soils, and wet depressions was dominated by Carex barbarae, C. vesicaria, C. feta, and C. kelloggii. Of 21 locations sampled, percentage composition on a foliage cover basis of all the Carex spp. varied from 40.3 to 86.3%. Other dominants were Poa pratensis, Elymus triticoides (beardless wild-rye), and Agrostis alba (Fig. 16). Occasional patches of Scirpus acutus (common tule), Calamagrostis canadensis (bluejoint redgrass), Typha latifolia (cattail), and Salix spp. were scattered on poorly drained areas. The grasses mentioned above occupied the edges of the wettest soils with Alnus rhombifolia, Libocedrus decurrens, and seedlings of Pinus ponderosa and Quercus kelloggii nearby.

Fig. 16. Perennial grasses invading annual grasses is shown within the marked area.

The moderately well- and well-drained soils, most extensive in the meadows, generally had a smooth, gently sloping surface with few channels. Some were flooded every year but others only rarely or not at all. These soils were grayish-brown, gray, or pale brown in color, loamy very fine sand to fine sandy loam in texture, moderately subangular, blocky in consistency, and slightly hard when dry. Poa pratensis, Elymus triticoides, Agrostis alba, and Carex spp. dominated the vegetation. Fourteen of the 16 sites samples had between 14 and 24% Carex spp.

Poa pratensis, clearly as abundant in the meadows of Yosemite Valley as the group of Carex species, tended to be scarce in the wet sites, especially those with standing water, and abundant on the moderate and well-drained soils. It appeared quickly on newly deposited alluvium, in the mixture of many annual species.

Elymus triticoides occurred in large patches frequently adjacent to areas of poor drainage and on recent, but more than year-old, alluvium that had good drainage. Poa pratensis and Carex spp. grew with it. Sites that were moderately well-drained and seldom flooded usually had Elymus glaucus (Fig. 17). The two Elymus spp. were difficult to distinguish by vegetative means in the field.

Fig. 17. Pinus ponderosa invading perennial grassland.

Annual species such as Bromus tectorum, Lessingia leptoclada (sierra lessingia), Lotus oblongifolius (narrow-leaved lotus), and others dominated on excessively well-drained new alluvium, especially following disturbance (Fig. 18). In most such areas perennials invaded quickly. Abundance of annual grasses suggested a low stage in plant succession rather than permanent vegetation..

Fig. 18. Perennial grasses invading annual grass in a former wet oxbow.

Vegetation on several soil types did not follow the Carex—Poa—Elymus gradient according to decreasing soil moisture. One soil with high organic matter classed as "bottom of depression" (Table 5, Area 1) contained vegetation dominated by Apocynum cannabinum with only 9% Carex spp. This area dried excessively in late summer. A ditch and a road recently had altered drainage in areas 2 and 3. Trees shaded areas 4 and 5. Areas 6-9 showed evidence of recent soil disturbance. Botanical composition in the meadows responded to outside influences as well as to the drainage pattern.

Table 5. Percentage species composition of meadow areas in the soil survey that has uncommon vegetational combinations.

In 1962 the garbage dump was leveled, compacted by bulldozer, and covered with 10 cm of manure and 5-15 cm (2-6 inches) of alluvium. Cynodon dactylon (Bermuda grass) appeared in 1963 as did typical first-year herbs such as Chenopodium album (lamb's quarters), Amaranthus graecizans (amaranth), Lotus sp. (lotus), Brassica spp. (mustard), Lepidium densiflorum (pepper grass), and Rumex crispus (curly dock). Seeds of these species probably came from hay used in the stables. Cynodon did not persist into the second year and other perennial grasses had not invaded by 1968. Seedlings of Quercus kelloggii did not persist. In 1968, Bromus tectorum was 28% of the vegetative composition. In 1973 the five most common species were L. densiflorum, Chenopodium botrys (Jerusalem-oak), B. tectorum, Sisymbrium altissimum (tumble-mustard), and Polygonum aviculare (knotweed), a vegetation much as it was 10 years earlier in 1963. Use of the area for parking and as a staging location for a rock climb slowed plant succession.

In the second year after removal of the Old Village Store and restoration of the site, Sisymbrium altissimum, annual grasses, and many of the same annual herb species mentioned above were present in the vegetation (Gibbens and Heady 1964). Perennial grasses (Elymus spp., Poa pratensis, and Agrostis alba) composed 35% of the cover and Festuca rubra (red fescue), which was seeded, made up 31%. Five years later in 1968 the three resident perennial grasses had increased to 77% of the cover and F. rubra had decreased to 5.5%. In 1973, 67% was Elymus spp., 13% Carex spp., 5% P. pratensis, and 1% F. rubra. Thus, except for the wet sites, plant succession following disturbance (cut and fill sites, new alluvial sand deposits, etc.) began with annual herbs, had an annual grass and herb stage, and proceeded to perennial grasses and Carex spp. Seeded grasses did not persist in abundance.

Botanical composition of the meadows has changed since the valley became a recreational site. Whitney (1868) described the meadows as chiefly sedges and coarse grasses. Sedges remain important and presumably are the same species. The three principal grasses listed by Whitney, Calamagrostis canadensis, Phragmites communis (reed), and Glyceria nervata (manna grass), have been replaced. We found the first to be widely distributed but scattered, the second absent, and the third rare. Poa pratensis, Elymus spp., and Agrostis alba, which probably came to the valley as seed in hay, now cover the meadows. These introduced species are better adapted to the meadow conditions in Yosemite Valley and have permanently replaced the original meadow climax species, just as in other wet meadows throughout the Sierra Nevada.

Numerous other hay and pasture species, introduced both intentionally and unintentionally, have not become dominant elements in the vegetation, although they are present in Lamon Meadow near the stable. Most disturbed areas, such as abandoned roads, about 100 ha (250 acres) of formerly cultivated land, and the former racetrack in Leidig Meadow, are not detectable to the casual observer. Vegetation in Yosemite Valley has a remarkably strong recuperative ability. The meadows are healthy and pleasing to the eye, although quite different from those of 100 years ago. Perhaps they are even better suited to public use than the original meadows.