Results and Discussion
Displacement of Birds
When the study period began on 7 May, 1980, Station 1, the lowest in elevation, was under 7.5 cm. (0.25 ft.) of reservoir water. The reservoir steadily rose, inundating five ecotonal stations by 30 May, 1980. In response to reservoir inundation, bird populations and species richness showed significant declines. Figures 5 and 6 depict the decline in these parameters in relation to water depth for the five stations subjected to inundation during the study period.
Figure 5.
Bird populations and Isabella Reservoir
inundation levels, 1980, at five stations.
Figure 6.
Bird species richness and Isabella Reservoir
inundation levels, 1980, at five stations.
Because bird species have different habitat requirements, utilizing the riparian environment differently, their responses to inundation varied. Bird species were grouped into the following four guilds based on their preferred nesting sites: 1) species which nest on the ground; 2) species which nest in shrubs; 3) species which nest among the foliage or branchwork of trees; and 4) species which nest in tree cavities.
The declines in bird populations and species richness are depicted for each guild in figure 7. The figure illustrates proportionately greater decline in ground and shrub nesters. The percent decline represents the average for the five stations once the reservoir level reached each station and assumes a linear decline. Data show, however, the decline in bird species abundance was roughly logarithmic, showing the greatest de-
Figure 7.
Average decline of bird populations and
species richness for four nesting guilds. The
percent decline per foot of inundation once
Isabella Reservoir water level reached each
station was averaged for five ecotone stations.
cline during the first several feet of inundation, then leveling off.
As the water rose, birds utilizing the ground and shrubs for nesting and food sources were displaced to higher ground. Nests were destroyed and unfledged young were drowned. In general, tree nesting species stayed in flooded territory for a short time, but were eventually forced to leave because of reduction in available food and the threat to nests as the reservoir filled. The emigrating forest birds were temporarily replaced by shorebirds foraging along the shoreline of the advancing reservoir. Once inundation was complete, however, this habitat niche was also eliminated.
The influx of emigrating birds to unflooded regions, illustrated by increases in population densities and species richness just prior to inundation of a station, upset the balance of the entire avian community. There was a marked increase in agonistic activity, especially among the wrens and orioles, as competition for scarce unclaimed territories was heightened. The more aggressive species such as Starlings and blackbirds were observed to dominate perch sites previously shared by many species. Interspecific conflicts, indicative of lack of suitable nesting sites, were frequently observed. All of these factors cut into the parents' time and energy budgets. Smaller, more crowded territories, with a less abundant food supply, reduced nesting success and eliminated the opportunity for second and third broods.
Bird Census, 1981
Average aggregate bird densities, species diversity, and species richness for each count in each subhabitat-type are shown in table 1. Densities and diversities include only breeding populations. The highest average aggregate density was 3959 birds per 40 ha. in the closedcanopy forest. The low species diversity in the new-growth subhabitat-type is attributed to the predominance of Redwinged Blackbirds, which comprised over 60% of the total density; when this species is excluded, the new-growth subhabitattype had the lowest density as well.
The exceptionally high bird densities of the South Fork forest well exceed the high density limits established by Peterson (1941), and are comparable even to worldwide maxima (Johnson et al . 1977). The intensive search for and location of 66 nests at four of the survey stations supports the high estimates obtained. Twenty-two active nests were found within a radius of 30 m. from Station 8 alone; the resulting breeding bird density would be over 6,000 birds per 40 ha. Bird densities for individual species within the SFWA are given in Appendix A.
When the rate of decline of bird populations observed in 1980 is applied to the high density estimates of 1981, the effects of reservoir inundation are clearly shown to be devastating.
Vegetation
Caution must be used in drawing conclusions about the effects of reservoir inundation on the condition of vegetation in the SFWA, as it has been affected by land use practices such as grazing, farming, and woodcutting as well (fig. 8). However, it is clear that prolonged inundation has significant immediate and long-term effects on vegetation. While some effects are readily observable after one season, others are not until some years later, when trees may die abruptly after a period of continued growth (Ewel 1978).
The immediate result of flooding, described by Teskey and Hinckley (1977), is the creation of an anaerobic environment around root systems, producing a variety of physiological plant stresses. The duration, depth, and rate of movement of floodwater are important factors controlling oxygen availability to the roots and hence the degree of stress produced. The time of year is also important; plants are much more susceptible to stress during the growing season (Walters etal . 1980).
Table 2 summarizes the data from four vegetative transects along the 5-, 10-, 25-, and 300-year reservoir stage recurrence intervals. Data show that total tree density, tree species diversity, average percent canopy cover, and average tree height all decrease with greater frequencies of reservoir inundation. The number of trees that died during the 1980 growing season (dead trees in 1981, with the previous year's leaves remaining), number of fallen trees, and degree of mistletoe (Phoradendron flavescens ) infestation decreases with increasing elevation.
The decline in total tree density from 550 trees per ha. at transect 4 to 1 tree per ha. at transect 1 is dramatic. Although transect 2 appears to be more dense than transect 3, the filling in of a former open field with young black willow (Salix Gooddingii ), of which 51% are more than 50% decadent, accounts for the high density figure.
Tree condition related to three age-classes for each tree species is shown in figure 9. The histogram illustrates a greater degree of decadence at lower elevations where inundation is most frequent.
Ewel (1978) concluded that the dependence of species composition on hydroperiod and depth of flooding is clearly the most sensitive characteristic of riparian ecosystems. Long periods of flooding eliminate the more sensitive species and generally favor the early successional stage species (Walters etal . 1980). The degree of decadence and mistletoe infestation were especially high among the cottonwoods along the lower elevation transects. In contrast, along the same transects, there was a high proportion of healthy black willow. These results indicate that Fremont cottonwood is least tolerant and black willow most tolerant of prolonged inundation in
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
the SFWA. Teskey and Hinckley (1977) classified Fremont cottonwood as "intermediately tolerant," that is, able to withstand one to three months of partial inundation during the growing season, and black willow as "tolerant," or able to withstand flooding for most of one growing season.
Figure 8.
Scattered willows and cottonwoods of the
SFWA where the forest enters Isabella Reservoir.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Figure 9.
Tree species composition, degree of decadence, and age-class for four vegetative transects within the SFWA, 1981.
The combined effects of grazing and inundation have retarded the regeneration of new trees, especially at lower elevations where large open fields have shown high potential for riparian reforestation. On transect 1, there were no seedling, sapling, or young cottonwoods. All seedling willows were concentrated in one narrow slough and had sprouted in 1981. Transect 2 had a high percentage of young willows concentrated in the interior forest, where aerial photographs show a former open field. This area was probably not easily accessible to cattle. Transect 4, an actively grazed area above gross pool, also had few seedlings or saplings of any species and was the only transect which showed a high percentage of mulefat—a shrub which is less palatable to cattle. The majority of the saplings, preferred food for cattle, were able to survive by growing through the protective center of a mulefat shrub.
The increase in percentage of fallen trees from transect 4 to transect 1 indicates that increased inundation also reduces soil stability, resulting in less support for root systems.
Another effect within the study area was the change in herbaceous vegetation in open fields. Prior to inundation, dominant plants were alfalfa (Medicagosativa ), mustards (Rorrippa and Descurainia spp.), and various grasses (Graminae). When the water receded, the fields were recolonized by cocklebur (Xanthium ) of little wildlife value and a nuisance to visitors.
The plant species composition also changed in a cattail (Typha ) marsh of 4 ha. The cattails did not regenerate after inundation and were replaced with curly dock (Rumex spp.) and monkey flower (Mimulusguttatus ).
Vegetation-Bird Relationships
The longer term effects of reservoir inundation on bird populations will be the result of an overall decrease of habitat value of the forest. Bird densities along ecotone stations are correlated with tree densities, both increasing with elevation. This correlation is also supported by the findings of Carothers and Johnson (1975) in a homogenous cottonwood forest. They found a linear relationship between bird density and tree density with a more rapid decline in bird populations if tree density fell below 25 trees/ha. Johnson (1970) also found that breeding bird populations in a cottonwood forest were reduced when tree density declined.
Tree species composition and age-class diversity also increased with elevation, showing an association between less diverse forest and lower bird densities at lower elevations.
The reduction in density of mature cottonwoods, in particular, has negative implications for bird populations. Nest site locations in the SFWA show that although mature Fremont cottonwood composed only 12% of the forest tree composition in the proximity of stations 8, 10, and 13 (transect 3), they supported 89% of the 45 nests found at these stations. Willow species, constituting 83% of the trees, supported only 9% of the nests. Three percent of the nests were found in herbaceous understory. In a study of nesting birds of a cottonwood/willow community in California, Ingles (1950) also found that hole-nesting birds preferred cottonwood over other tree species.
Although new growth is essential to the continued viability of the forest and contributes to structural diversity, study results show that uniform stands of young trees are less favored habitat for birds than mature groves. Anderson etal . (1978) concluded that along the lower Colorado River, as trees mature they approach their full value to wildlife.
The elimination of the 4 ha. cattail marsh displaced a blackbird colony of several thousand breeding birds in 1980. Because the cattails failed to regenerate, the colony was reduced to less than 50 pairs in 1981.
As Best etal . (1978) has pointed out: "Birds with restricted habitat usage patterns are more vulnerable to changes in land use practices than those occupying a wider variety of environments." The high dependency of birds on riparian vegetation makes them especially susceptible to the effects of habitat degradation.
Additional Effects
Inundation also had important secondary impacts on wildlife and habitat improvements of the SFWA. Fishing activities, popular once the forest was under water, caused considerable disturbance to the Great Blue Heron rookery and to nesting Osprey. Because of the interference, no Osprey were hatched in 1980. Perch poles, necessary for successful utilization of nesting platforms, were washed away.
Reductions in mammals and other vertebrates were not quantified, but animals floating on debris or drowned were encountered daily during the study period in 1980. While mature birds were able to fly ahead of the advancing water level in the reservoir, many other terestrial animals were trapped. A prominent local naturalist has noted the absence of numbers and species of animals throughout the winter after inundation.[15] The recolonization rates for all species are unknown.
Inundation caused substantial damage to CE-constructed improvements in the SFWA. Artificial nesting boxes were inundated, many with occupants. All three footbridges were washed out and required reconstruction. Water control structures, including Waterman gates, levees, and ponds were destroyed, and cattle exclusion fences severely damaged. Repeated inundation will result in the destruction of the water system rebuilt in 1981, as well as other repairs and improvements; and damage or death to newly planted cottonwoods and willows.
Inundation also precluded visitor use of the SFWA for at least three spring and summer months during the season when visitation at Isabella Reservoir is highest.