Studies of the composition of aquatic macrophyte communities in Otsego Lake date back to 1935. Muenscher (1936) iden tified 47 species in the lake and ranked them subjectively into four abundance categories (Table 26). At that time he found emergent plants to be "...very rare...except in the few small protected coves." (Muenscher, 1936), although photographs from the 1920s show sedges and bullrushes along even exposed shorelines. In 1935, Chara vulgaris [includes C. contraria (Wood and Muenscher, 1956)] dominated the substrates from 1 to 2 m in depth. In many cases, this macroscopic alga was accompani ed by Potamogeton gramineus. In 1935, Nitella flexilis was reported in water as deep as 7 m and was considered to have the deepest distribution of any macroscopic plant found in Otsego Lake.
Muencher's procedure was repeated by Harman and Doane (1970) in 1969, and Brady and Lamb (1977) in 1976. In the late 1960s and 1970s, rushes were found in extensive stands along the beaches on the north end of the lake. At the same time, beds of Nuphar variegatum and Nymphaea odora ta were present in practically all embayments, while Typha spp., Sparganium eurycarpum, Scirpus acutus, and Eleocharis spp. were encountered discontinuously along the protected shorelines. Chara vulgaris was still prevalent in these localities, but P. gramineus was no longer a dominant plant.
Many vascular plants that Muenscher had recorded in up to 5 m of water, including Najas flexilis, Heteranthera dubia, Elodea canadensis, Myriophyllum exal bescens, Potamogeton praelongus, and P. richardsonii, were found only in shallower areas in the 1970s. None of these plants formed dense beds, except for H. dubia, which covered extensive local areas. Dense beds of P. crispus were almost always in association with N. flexilis and often appeared to supplant it. Not a native pondweed, this Potamogeton is of European origin and dense stands are often associated with enriched waters (Fassett, 1960). It was not found in Otsego Lake in 1935. Titus (1994a) suggests the spread of P. crispus statewide in the 1930-40 period was associated with fish stocking activities. With the exception of one bed of P. pectinatus at the northern end of the lake, P. crispus was the only Potamogeton that formed extensive beds with plants reaching the water's surface.
In June and July of 1970, an almost continuous strip of the latter followed the 5-7 m contours completely around the lake. By the mid dle of July, as is typical for the species, these plants died back and were no longer visible at the surface (Harman et al., 1980). Beds of P. crispus occurred no deeper than 5 m in the northern end of the lake in the 1970s, while at the southern end beds were commonly found at 7 m. All other species showed somewhat similar but shallower distributions. The transparency at the northern end of the lake is consistently much less than that at the southern end (see section on "Transparency" above). This suggests that the resultant decrease in transmitted light affects the depth at which the compensation points of most species occur and is reflected in the maximum depths at which various plants are able to exist.
Macrophyte phenology in Rat Cove, 1970-71, (Harman, 1974b)
In 1970 and 1971, transects about 500 m long were plotted from the shore to the outer limits of the littoral zone in Rat Cove. Each week during the ice-free seasons, divers measured pla nts growing along these transects to determine the emergence time of the spring cohorts, the greatest maximum heights attained, times of flowering, and times of death and decomposition of each species.
In any one area along the Rat Cove transects between 50 and 550 cm in depth, several species of hydrophytes usually occurred. This often resulted in a complex physiognomy exhibiting an overstory, usually of low density, a moderately dense intermediate stratum, and a low, very dense understory. Sometimes up to seven distinct strata were present. Vertical stratification of the community continually changed during the growing season as the various species of macrophytes emerged from the substrate, attained maximum height and density, and then fell to the substrate and decomposed.
Nuphar variegatum, the yellow water lily, grew in scattered clones in water from 20 to 60 cm in depth in association with Elodea canadensis and Megalodonta beckii. Water from 50 to 500 cm in depth maintained all the remaining macrophytes studied except for Nitella flexilis and P. crispus, which occurred in water from 500 to 700 cm in depth. In some years (e.g. 1975), the lower littoral and upper sublittoral supported dense populations of filamentous green algae (Spirogyra spp.) that completely covered the substrate.
The following are descriptions of the phenology and physiognomy of the plants at 4 sites (0.5, 2.0, 4.0, and 5.5 m in depth) along the transec t (Figure 43a-e) in 1971.
1. Water depth 0.5 m: Nuphar variegatum emerged from the substrate the first week in May; by June 15, 1971, the leaves had reached the surface. They remained there until they decomposed in early September. Elodea canadensis and Megalodonta beckii started their seasonal growth in late April and the middle of May, respectively. They maintained a maximum height of about 30 cm beneath the Nuphar during August. By September 15, M. beckii had collapsed to the bottom and was decomposing. At the same time, E. canadensis started to decompose from the base and the entire plants had become very brittle. By November 15, the winter condition was reached. Megalodonta beckii and E. canadensis were prone on the bottom with winter buds 1-2 cm long developing at the nodes. During July and August, two obvious layers of vegetation were present, N. variegatum at the water's surface and E. canadensis and M. beck ii below. From autumn into the winter, there were no actively growing macrophytes at this site. Figure 43b illustrates these changes at five selected dates during the year.
2. Water depth 2.0 m: This site was always dominated by dense stands of C. vulgaris. Only in June was M. beckii able to emerge above this species. Elodea canadensis occurred only slightly higher than the Chara in June, July, and August before dying in the fall. In late August and early Sept ember Najas flexilis was present, but never grew higher than the Chara. During August and September, there were two definite vegetational strata present. A diffuse overstory of V. americana and M. exalbescens occurred above C. vulgaris. With the decomposition of V. americana and M. exalbescens in November, only Chara remained. The first week in May, dense stands of C. vulgaris were at their lowest height of the entire year. Chara gr ew steadily until the middle of November. As the rate of apical growth decreased in the winter, plant height was reduced because of concurrent basal decomposition.
Megalodonta beckii and E. canadensis started growth at the same time as they did in the shallower water but attained somewhat greater heights. Megalodonta beckii reached approximately 35 cm in height in early September and then began decomposing. Myriophyllum exalbescens appeared in early May, grew stead ily until September, then fell to the bottom, with new shoots appearing at the nodes. Vallisneria americana emerged from the substrate the middle of June, grew more rapidly than its competitors, attaining its maximum height of about 125 cm in September, and then rapidly decomposed. Najas flexilis emerged from the substrate in the middle of July, grew rapidly to about 60 cm high in early September, and disappeared. Figure 43c shows the strata in 2.0 m of water at various times during 1971 .
3. Water depth 4.0 m: as at the 2 m site, C. vulgaris grew in dense beds at a consistent rate throughout the summer. Elodea canadensis was present until August as before. Early in May, Potamogeton pusillus, P. zosteriformis, P. Richardsonii, and P. Illinoensis emerged from the substrate. By July 15, P. Richardsonii and P. Illinoensis attained about 200 cm in height while P. pusillus and P. zosteriformis reached only about 100 cm. All of the species of Potamogeton were decomposing by August, although a second cohort of P. Richardsonii remained until early September. In late May, Heteranthera dubia emerged from the substrate and maintained steady growth until late September when decomposition occured.
In June, the morphology of the plant community was very complex, with seven species of actively growing hydrophytes in the water column, all at different heights. By July 3, obvious strata wer e present: a diffuse overstory composed of P. Richardsonii and P. Illinoensis, a discontinuous stratum of intermediate height composed of P. pusillus and P. zosteriformis, and a dense understory of Chara, Elodea, and Heteranthera. In September, this same area was entirely different in appearance, with 2 major strata present. There was a rather dense overstory of H. dubia underlain by C. vulgaris. By November, the Chara was all that remained. Figure 43d illustrates the physiognomy at this depth at several selected dates.
4. Water depth 5.5 m: In these deeper waters, Nitella flexilis replaced C. vulgaris. Dense beds attaining 95 cm occurred in this area. The only macrophyte associated with this alga was P. crispus, which appeared from the substrate the first week in May. The latter species grew extremely rapidly, attaining 250 cm in height by the middle of July. It then collapsed to the substrate and decomposed. In June and July, two definite layers of vegetation were present at this depth. Potamogeton crispus formed a diffuse overstory while N. flexilis composed a dense understory. From August on into the winter N. flexilis was the only species present at the site (Figure 43e).
The macrophytes studied can be separated into four groups according to their morphology during the winter season. These characterize the appearance of the substrate during this period.
Group 1. Chara vulgaris and Nitella flexilis. Once maturity is reached, the oldest parts of these algae are continually decomposing. During the winter period, entire stands decrease in height since the rate of decomposition greatly exceeds the rate of growth. The apical meristems remain in a healthy condition throughout the year. In early spring, germinating oospores add young plants to the population that cannot be distinguished from the older plants remaining during the same periods .
Group 2. Myriophyllum exalbescens and Elodea canadensis. When these organisms degenerate (late September and late August, respectively), the old growth lies on the substrate and winter buds, 2-3 cm in height, appear from the nodes. These grow extremely slowly over the winter, reaching 4-6 cm by early May. At that time the roots have become established, internodal tissues from the parent plants have rotted away, and the new plants grow rapidly until the fall, when the process i s repeated.
Group 3. Heteranthera dubia. This plant collapses to the bottom in a living condition (in mid-October), but the new shoots do not appear at the nodes until late May or early June of the next year. They grow rapidly until fall and the cycle starts again. The parent stems remain intact throughout much of the growing season.
Group 4. Potamogeton crispus, P. Richardsonii, P. Illinoensis, P. zosterformis, P. pusillus, Vallisneria am ericana, Najas flexilis, Megalondonta beckii, and Nuphar variegatum. This group contains all the remaining plants studied. These macrophytes grow rapidly from buried winter buds, or underground stems or rootstocks, and then decompose, with shoots not appearing above the substrate until the next growing season.
Rat Cove productivity studies, 1977
During the summer of 1977, the productivity of the three most abundant macrophytes in Rat Cove at the ti me (Potamogeton crispus, Heteranthera dubia, and Chara vulgaris) was ascertained.
The transect along which the phenology data were collected (bearing west to east) was bisected by another (bearing north to south), effectively dividing the cove into four quadrants (Figure 44). The biomass collections necessary for the determination of productivity were taken in areas where monospecific stands of each species exhibited their maximum standing crops. Annual net production fro m these areas is presented in Table 27. The productivities of Heteranthera dubia and C. vulgaris were estimated and then transformed to ascertain their productivities throughout that quadrant (Table 27). Those values agree closely with productivity data derived from several oligotrophic-to-mesotrophic lakes cited by Wetzel (1975) .
Historically, Chara vulgaris has formed homogeneous stands completely covering the littoral substrates in Otsego Lake (Harman and Doane, 1970). In 1976, openings were first observed in the Chara beds, exposing the inorganic substrate below (In 1994, extensive areas of barren substrates still remain exposed). Crawford (1977) believed that this phenomenon was a step indicating the successional transition from an oligotrophic to a mesotrophic environment in which Chara is replaced by rooted vascular plants.
To understand better the factors directly responsible for the present distribution of Chara, samples were tak en at several stations along the aforementioned axes. Measured were biomass and the organically bound phosphorus in the substrate. Total phosphorus, soluble reactive phosphorus, soluble unreactive phosphorus, chlorophyll a, and phaeophytin were measured in the overlying water (Table 28, Figures 45-47).
Phosphorus values in both the substrate and the water are inversely correlated with Chara biomass (Figures 45-47). Forsberg (1964) maintained that extensive production of Chara occurs only when habitats possess levels of orthophosphate equal to 0.02 mg/l or less. The distribution of Chara observed could, therefore, be interpreted as the result of inhibition by high phosphorus levels, supporting Crawford's (1977) hypothesis.
Alternatively, the low phosphorus concentrations in association with viable stands of Chara may result from their removal from the environment by the physiological activities of the algae itself. Chara removes seston from the water, including organic chelators (Schelske et al., 1962). This process stimulates the precipitation of CaHPO4 and, subsequently, the marl formations typically encrusting the Characeae (Otsuki and Wetzel, 1972).
Descriptive ecology of Rat Cove macrophytes, 1978
In 1978, an intensive study of the aquatic macrophyte communities in Rat Cove was conducted (Vertucci et al., 1981). The ecology of 20 species of aquatic macrophytes inhabiting 3,300 25 m2 quadrats (Figure 48) in Rat Cove, Otsego Lake was investigated. All quadrats were surveyed for percentage cover of each species. Macrophyte productivity was estimated through biomass determinations. Plant tissue of each species was analyzed for N, P, Mg, Ca, K, and Fe. The environmental characterization of Rat Cove included measurement of water depth, wind and wave exposure and the analyses of sediment and pore water at 38 sampling stations for major and minor nutrients (38 different parameters in all). Indirect and direct gradient analysis were used to analyze plant-environment relationships. Among the methods used were; polar ordination, reciprocal averaging, detrended correspondence analysis and composite clustering. Direct ordinations were derived using a species importance index that incorporated both biomass and percentage cover of the macrophytes in the study area. Detailed methodology is presented in the original publication (Vertucci et al., 1981).
Results are included here that describe or include examples of the floristics, vertical structure, horizontal pattern, phenology, niche relationships, species importance via productivity estimates, and species diversity of Rat Cove macrophyte communities. Selected results of the microhabitat characterization are presented in distribution maps (Figures 49-56). Direct ordinations of species along the major gradients are presented. These include the species distribution along both the depth and sediment gradients.
Biotic characterization
A list of species, their growth forms, methods of reproduction and overwintering is presented in Table 29. To facilitate description, the species are divided into four groups representing different intergrading community types.
Some of the aquatic macrophyte community types in Rat Cove have a distinct vertical structure (Figures 57-60). Very few workers, however, recognize vertical stratification within macrophyte communiti es. This is more likely due to a lack of direct field observations than it is the absence of this characteristic in nature. Forsberg (1960) reported finding macrophytes in monospecific meadows, Swindale and Curtis (1957) found heterogeneous stands while Harman (1974b) and Hendrey and Vertucci (1980) describe the occurrence of both structured and single species stands.
Community type I is typified as in Figure 57. Here, a diffuse understory of Ranunculus aquatilis and Ceratophyllu m demersum is overshadowed by discontinuous clones of the floating leaved species, Nuphar variegatum and Nymphea odorata. Much barren substrate, at times colonized by epipelic filamentous algae, is present beneath the dense cover of floating leaves.
A more complex canopy structure is evident in community type II, (Figure 58). In late summer through fall Heteranthera dubia dominates the overstory with P. zosteriformis occurring sporadically above H. dubia< /i>. There is a subcanopy composed of patches of Myriophyllum exalbescens and Elodea canadensis. Found commonly below these are Vallisneria americana and Meglodonta beckii. The understory is also composed of patches of Chara and some Ceratophyllum.
The most striking vertical zonation of species is found in Type III (Figure 59). By early July, P. crispus grows from 5-6 m deep up to the surface and forms a dense canopy. In patches below thi s overstory, Ceratophyllum and Elodea are found. A diffuse carpet of Nitella flexilis covers much of the bottom below these plants.
During the mapping of the quadrants having this community structure, different fish species occupying the different strata of vegetation were observed. For example, commonly seen were; bullheads (Ictalurus nebulosus) resting in the Nitella, bluegills (Lepomis gibbosus and L. macrochirus), rock bass (Ambloplite s rupestris), and pickerel (Esox niger) in the subcanopy, and numerous minnows (Notemigonus crysoleucas and Notropis hudsonius) schooling near the surface under the protection of the dense overstory of Potamogeton. In early July, pairs of mating carp (Cyprinus carpio) concentrated in these areas to spawn in dense beds of Potamogeton.
In contrast with the aforementioned communities, species composing community type IV tend to grow in monospecific patches with little vertical structure (Figure 60). Meadows of C. vulgaris dominate this community. Dense monospecific stands of P. pectinatus and P. amplifolius also occur in discrete patches. Small clones of P. prealongus and P. Richardsonnii, (6 or 7 plants connected by rhizomes forming rows) sometimes cover a diffuse understory of Chara or Najas flexilis.
Fish species commonly found in other more structured plant communities in Rat Cov e were rarely observed here. These observations imply an apparent relationship between littoral fish communities and vegetation structure that parallels the relationship between terrestrial vegetation structure and avian communities (c.f. Whittaker, 1975). Marine fish species diversity has also been found to correlate with the degree of structural complexity of coral reef habitats (Jones and Thompson, 1978). Vertical structure can be an important characteristic of aquatic macrophyte communities. T his structure, however, is superimposed on a marked horizontal pattern.
The pattern of vegetation cover in Rat Cove can be seen from examples of the distribution maps of selected species (Figures 49-56). From these maps the following general observations can be made. Each species has its own characteristic distribution that may be similar, but not identical, to another species. The distributions are not random or regular, but tend to be somewhat clumped or contiguous. This is expected sin ce vegetative reproduction through rhizomes is so predominant (c.f. Table 29). It can be assumed that the patterns are the result of each individual species' response to the environment, interspecific competition and chance, or some combination of these factors. Chance refers to those stochastic processes related to vegetative and sexual reproduction that fortuitously determine whether a given plant's propagule (seed, oospore, turion, plant fragment, etc.) will be distributed to a given suitable lo cation. Where there are extensive areas of exposed substrate, chance initially may be the dominant determinant of the resultant vegetative pattern. The patterns of individual species distributions will be discussed in more detail later where they are related to the environment. Consideration of spatial pattern alone, however, would neglect the seasonal patterns of growth of these plants as addressed by Harman (1977) above. After recognition of a species spatial and temporal position in the community, a description of its niche may be approximated.
The position of plant species along three major niche axes (height, horizontal pattern and time) can be used to locate species in a simplified niche space (Whittaker, 1975). These axes are taken to delimit some portion of the more complex n-dimensional niche hyperspace [sensu Hutchinson (1958)] that is used to define, in a meaningful abstract way, a species' fundamental niche. While abstract, these concepts provide a framework from whic h an improved understanding of community organization and evolution can develop (Whittaker, 1975).
A summarization of Rat Cove macrophyte species' height, pattern and phenology is presented by positioning them in niche space (Figure 61). This niche space is defined by two axes representing horizontal distribution (pattern) and a third corresponding to the maximum height individuals of a species attain during the growing season. The former two correspond to the sediment gradient and depth. The position of species on the sediment pattern axis is derived from the first axis of a DECORANA species ordination described in the original work (Vertucci et al., 1981). The time of maximum height and the below-to-above-sediment biomass ratio of each species is also expressed in the niche diagram (Figure 61).
It is apparent that the species are dispersed throughout the area, with each one having a unique position, sometimes adjacent to, but never the same, as any other species. T his observation conforms with the principle of Gause (1934). Where species are closely associated on the pattern axis, dissimilar vertical and temporal patterns of growth may allow for species co-existence. The species are more or less cluttered in niche space into discernible groupings that represent the four community types illustrated in Figures 57-60. Note that the community types with the most vertical structure are more closely aggregated on the pattern axis than the unstructured community typ e IV. The clustered appearance of these species suggests that those evolutionary processes that tend to disperse species in niche space, and thereby reduce competitive interactions, may not have been very effective regarding these species. The reason for this may include the instability and ephemeral nature of individual lake environments and the relatively short time these species have been interacting (post Pleistocene). Species with similar niches could also persist in other lake systems where competitive species were absent. Since time and some degree of environmental stability are needed for niche differentiation to occur, these factors could impair selective processes that might show the species to be more equitably distributed in niche space.
There are apparent differences between the below-to-above sediment biomass ratios of species in each community type (Table 30, Figure 61). This may reflect the relative importance of the different functions of below ground tissues in the various locations. In type I, more anchorage may be needed by these shallow water species in order to compensate for the effect of waves. Nicholson and Aroyo (1975) also reported finding highest root/shoot ratios in shallow (less than 1 m) water. Deep-water species (type III) would be under no such constraint. Species found on substrates that are more nutrient poor (type IV), may need greater rooting and rhizome storage volume to accrue enough sediment bound nutrients since macrophytes acquire mos t of their phosphorus from sediments (Carignan and Kalff, 1980).
The consideration of macrophyte species niche does not provide for any understanding of the area of niche space occupied by each species. This area is proportional to the amount of community resources utilized and can be indirectly measured by determining the productivity of each species. Thus the way species divide up the available resources in a community may be measured by determining species importance. Table 30 lists spe cies percentage ash content, average maximum annual biomass and the percentage cover of each species in Rat Cove. These data were used as described in the methods to derive the productivity estimates listed in Table 31. This table reports species maximum annual biomass (GAFDW), maximum annual biomass weighted by coverage in Rat Cove, net and relative primary productivity (g C m-2 yr-1) of species in the Cove, net primary productivity (g C m-2 yr-1) on a l ake-wide basis and the net primary productivity on a per day lake-wide basis in g C m-2 day-1 lake-1.
Species importance is commonly displayed by using a species-importance value or dominance-diversity curve (Whittaker, 1975) (Figure 62). The x axis is a listing of the species from most to least important and the y axis is the logarithm of species importance as determined from the productivity data. The form of the curve is related to one of many hypotheses that suggest how species divide up the available resources of a community (Whittaker, 1975). The data presented here (Figure 62) nearly fit a geometric series and, as such, can be interpreted under the niche pre-emption hypothesis where the size of the niche space and the amount of resources available to each species will depend upon what has not been pre-empted by the more successful species. This results in each successive species occupying or utilizing some fraction of the space or resources rem aining. The ratio of the importance of a species to the next more important species indicates something about the way niche (resource) space is filled by species. The average ratio between species pairs in Rat Cove is 0.67 +/- 0.25 SD. This ratio, or an inspection of Figure 63, suggests that there is an even distribution of species relative productivity. That is, no range of relative productivity has significantly more species than any other range. This suggests that Rat Cove macrophyte communitie s exhibit strong dominance. Species are probably competing intensely for the available resources once established. Those species established first may have a competitive advantage over the species that arrive later. The application of this concept of primacy was first applied to aquatic plants by Vaarama (1938).
For some communities (especially when a collection of communities are grouped together, as here) a plot of the number of species found within octaves of some measure of importance may result in a log-normal distribution with the greatest number of species being found in some mid-range of importance octaves (Preston, 1948). Yet, when this is done for Rat Cove macrophytes following Preston (1948), using relative productivity, considered by Whittaker (1975) to be the most appropriate measure of species importance, the resulting distribution is relatively flat (Figure 63). In a similar plot, using a different importance measure (relative frequency of species occurrence for all q uadrats of Rat Cove) the distribution is, again, not log-normal but negatively skewed (Figure 64).
The apparent discrepancy between these figures can be explained by the following. The establishment of one or more pioneer species promotes the entrapment of other species, thus increasing the frequency of occurrence. However, dominance by the pioneer species does not usually allow subsequent species to become productive. This phenomena has been reported elsewhere (Auclair et al., 1973 ; Nicholson and Aroyo, 1975) and is further evidence of strong dominance within macrophyte stands. The even distribution shown in Figure 63 is expected from the form of the importance-value curve (Figure 64). Since the species that comprise the four community types are not necessarily competitively related, their relative importances vary widely along the logarithmic relative importance axis.
This consideration of species importance has provided some description of how species divide up the ir niche and resource space. The mechanisms responsible for these community characteristics are open to much speculation. It is understood that there are many reasonable alternatives to the hypotheses briefly mentioned here.
A community characteristic closely related to species importance is species diversity. The form of importance-value curves is dependent upon the number of different species in the communities and their relative importances. An index of species diversity is a statistic that sums up the properties of these curves. Diversity indices are used to compare and interpret differences between the macrophyte community types found in Rat Cove.
Community type I as seen from the water surface gives the impression of being dominated by the floating leaved species. Yet those samples assigned to this community type had an average of 6.2 +/- 0.880 ( (SD/n) t. at 95% C.I.) species per quadrat, the dominance concentration (Simpson's index S) was 0.21 +/- 0.066 and H', spec ies equitability, was 2.5 +/- 0.352. From close inspection of this community type it appears that the petioles of Nuphar and Nymphaea effectively trap fragments of other plant species. These then establish themselves under and around the clones of Nuphar and Nymphaea resulting in greater species richness. (It should be pointed out that since these indexes were calculated using percent cover as the importance measure, dominance is underestimated compared to what it would be if estimates of primary productivity were used).
Samples taken from the structurally most complex community type (II) had the highest species richness (7.1 +/- 0.617 species per quadrat), the least dominance (S= 0.16 +/- 0.021) and the highest equitability (H'= 2.9 +/- 0.164). Located in deeper water, community type III had fewest species per quadrat (4.4 +/- 0.618), was dominated by P. crispus (S= 0.27 +/- 0.041) and had the lowest equitability (H'= 2.1 +/- 0.185). The quadrats assigned to the community type IV had 4.8 +/- 0.2 species per quadrat, were dominated by Chara meadows (S= 0.28 +/- 0.04) and also had low species equitability (H'= 2.1 +/- 0.16). Overall, this subsample of Rat Cove quadrats had an average of 5.6 +/- 0.198 species per quadrat, dominance concentration (S) was 0.23 +/- 0.02 and equitability (H') averaged 2.4 +/- 0.06.
The relationship between species diversity and productivity is not simple. Within Rat Cove, however, those community type s exhibiting more dominance (IV and III) were responsible for the greater fraction of total productivity. Table 32 compares macrophyte maximum seasonal biomass and the reciprocal of dominance concentration (Simpson's index) for a variety of lake types. There is no apparent relationship between species diversity, as measured here, and community production of the systems studied. Otsego Lake macrophyte diversity and production was within the range observed for other hard water mesotrophic lakes (Wetz el, 1975).
Environmental characterization
Many attributes of the Rat Cove environment which play a role in the development of macrophyte community characteristics are reported below. They are also of relevance to a more generalized discussion of lake trophic status (see below).
A bathymetric map of the Rat Cove littoral zone depicts the way the bottom gradually deepens to a depth of 9 m (Figure 65). No macrophyte species occurred beyond the deepest contours pictured .
The results of detailed analyses of Rat Cove sediment and pore water samples are reported in Tables 33 and 34. Also included are results of textural analyses of Rat Cove sediments attained by Rigby (1978). After being scaled, these data were drawn on maps of the Cove. Examples are presented in Figures 66-73. These data depict a complex pattern of changing sediment characteristics. Many of these sediment factors are distributed, however, in similar ways, forming a complex gradient. To q uantify these relationships and to identify sediment parameters useful as indicators of other pertinent sediment characteristics, correlation coefficients were calculated between each of the variables listed in Tables 33 and 34 using the statistical analysis system (SAS) of Helwig and Council (1979). Five of these variables and those characteristics significantly correlated with them are those with a p value (p = the probability of having a greater value) of 0.05 or less.
Sediment organic matter content positively correlates with other characteristics that are potentially important to aquatic vegetation. These include factors such as the concentration of Fe, P, Zn, Mn, and K (Table 35). In direct contrast, the percentage of total sediment carbonates (expressed as the percent CaCO3 equivalent) was negatively correlated with many of the parameters indicative of sediment fertility, especially those measuring the P concentration. The pore water concentration of soluble reactive P and ammonium N were correlated with each other as well as other factors indicative of a reduced environment. The amount of extractable Fe [extractable cation concentrations are approximately equal to 80 percent of the concentration of the exchangeable ions (Greweling and Peech 1965)] was significantly correlated with more parameters than any other factor. Otsego Lake littoral sediments have a low clay content. Therefore, most of the ion exchange sites (80 percent) are associated with the sediment orga nic matter (Boyd, 1970; Toth and Ott, 1970).
Sediment inorganic P is commonly associated with hydrous Fe oxides. The P availability is consequently a function of the sediment Fe oxidation state, with P more available in reduced environments. The degree to which sediments are reduced is a function of the availability of electron donors (i.e. organic carbon) and electron acceptors (i.e. oxygen) in the sediment (Holdren et al., 1977). The positive correlations between sediment organic matter content, Fe, P, ammonium and certain cations can possibly be explained in terms of these relationships.
Parameters like those listed in the left hand column of Table 35 may be used as indicators of general sediment characteristics. These environmental scalors, sensu Loucks (1962), serve in subsequent analysis of environment-plant relationships. The sediment and pore water data presented are comparable with data from Cayuga Lake (Vogel, 1973). Rat Cove sediments are higher in organic matter content (8.4-17.2% compared to 0.5-5.0%) and subsequently have higher concentrations of inorganic nutrients. Data indicate that there is an ample supply of nutrients in many of the Rat Cove sediments compared to the overlying water.
The percentage of mid-lake wave height (used as an index of wind exposure under conditions typical of predominant winds) at the 38 sampling stations in Rat Cove is pictured in Figure 74. These data show a gradient of exposure that is, in part, patterned after the bay morphometry. Sediment characteristics are influenced by the amount of exposure. For example, turbulence at the sediment water interface probably aerates the sediment through mixing processes. The degree of exposure also differentially affects the relative amount of sedimentation verses transport of particles of varying sizes and densities (Spence, 1967). Also, much allochthonous material collects and is deposited in the most protected portions of Rat Cove. The exposu re gradient then is related in complex ways to basin morphometry, and therefore sediment perturbation and development.
The relative amount of total daylight received by the quadrats of Rat Cove is reported in Figure 75 as the percentage of total daylength illuminated. As expected, those quadrats near shore are more shaded and received less daily illumination than those quadrats further from shore. The quantity of light reaching any point on the bottom is exponentially related to depth (Gold man and Horne, 1983).
Characterization of the ecocline
The relationships between the plant distributions and the patterns of environmental variation depicted in Figures 49-56 and 66-75 are summarized in terms of four predominant factor gradients (Figures 76-79). The ordinate of each Figure is the index of species importance relativized by the sum of all importance values for all isopleths of each parameter. In order to facilitate comparison of these ordinations, the scale on t he ordinate varies according to the relative breadth of species occurrence along the gradient. For convenience, each Figure was divided into four parts, each corresponding to one of the community types previously described.
These factor gradients (depth, organic matter, complex-gradients affecting intralacustrine plant distributions, CaCO3 and extractable Fe) are reasonable indicators of the major complex-gradients affecting intralacustrine plant distributions, namely depth, degre e of exposure and sediment composition.
These ordinations also appear as Gaussian curves that overlap continuously along each gradient. The community types, however, are distinguishable since species within types have similar distribution curves.
The distribution of species along the depth gradient
The relationship between macrophyte species and the depth-complex gradient is depicted in Figure 76. The shallow water community is dominated by Nuphar variegatum and Nymphaea odorata and is further characterized by the presence of Ranunculus aquatilis. The floating leaved growth form of nymphaied species physically restricts them from inhabiting deeper, less protected areas of the Cove (Sculthorpe, 1967; Hutchinson, 1975). The coiled petioles of Nymphaea odorata may allow for greater tolerance of wave action and as such might explain why this species is found in slightly deeper, less protected water than Nuphar variegatum. Ranunc ulus was also found to be confined to shallow water (<l m) in a study by Rickett (1922).
The species comprising community type II are distributed more broadly along the depth gradient. The modes of their distributions are at 1 and 1.5 m depth. Vallisneria americana occurs more abundantly in shallow water (l m) than the other species in this group. Its depth distribution is similar in Lake George, N.Y. (Secchi depth approximately 6-7 m) (Sheldon and Boylen, 1977). Based upon ob servations of this species' depth distribution, it has been suggested that Vallisneria may require higher light intensities than species commonly associated with it (Vertucci et al., 1981).
The distribution of Heteranthera dubia follows the 1.5 m depth contour (Figures 51 and 65). In Lake George, Heteranthera is found abundantly in 1-3 m of water (Sheldon and Boylen, 1977). The compensation point of this species has been reported to be between 1.8% and 0.68% surface illumination (Meyer et al., 1943). However, Wilkinson (1960) criticized this result and presented evidence from more long term green-house experiments. He found that limiting light levels varied with the season. The critical quantity of light needed for Heteranthera growth was 9.6% (available sunlight) for the period April 1-29, 5.8% for May 6 to June 3, 2.8% for June 10 to July 8, and 5.8% for the remainder of the growing season. Because of their questionable accuracy, pub lished values of limiting light levels are used here for comparative purposes only.
A comparison of the distributions of these two species suggests that Heteranthera dubia may be excluding Vallisneria americana from areas of the depth gradient both shallower and deeper than the region that Heteranthera dubia dominates. In eutrophic Lake Chautauqua, Nicholson and Aroyo (1975) found a similar distribution for these two species. An investigation into the possible competiti ve interaction between these two species, as implied from their distributions in Rat Cove, has been conducted in Otsego Lake by Titus (1994b).
Megalodonta beckii and Myriophyllum exalbescens also have their maximum occurrence at 1.5 m depth. M. beckii has a more narrowly defined depth distribution than the latter species. Sheldon and Boylen (1977) reported M. beckii in Lake George to be prevalent in from 1.5 to 6.5 m depth.
The species of this type with the broadest depth distributions are Potamogeton zosteriformis and Elodea canadensis. The distribution of P. zosteriformis is bimodal with maxima at 1.5 and 4 m depth. In contrast with the other species, the distribution of E. canadensis shows little depth preference. Similar distributions were reported for these species by Sheldon and Boylen (1977).
The species making up community type III are also found over much of the depth gradient with their distribution m axima at 5-6 m depth. The depth distribution of Potamogeton crispus is most broad. Its maximum abundance is at 5 m depth. In Lake George, however, it was predominantly found in shallow water (1-3 m) (Sheldon and Boylen, 1977). Since the 1978 mapping of Rat Cove, this species has rapidly spread into shallower water. Turions of P. crispus abundantly produced in the dense deep water stands settle in areas of exposed substrate in shallow water previously occupied by Chara vulgaris.
Ceratophyllum demersum has a bimodal depth distribution, with maxima at 0.5 and 5.5 m depth. This rootless species is notably tolerant of low light conditions and often exhibits a broad depth distribution (Hutchinson, 1975). The compensation point of this species has been reported to be only 2% of full sunlight (Chapman et al., 1974). When it occurs in shallow water it is often in the understory of a dense canopy (Figures 49, 58). Lind and Cottam (1969) also report finding Ceratophyllum under a canopy of floating-leaved plants.
The maximum occurrence of Nitella flexilis is at 5.5 m depth. Nitella and Ceratophyllum penetrate deeper (9 m) than any other macrophyte species. Meadows of N. flexilis are often found beyond the depth limit of angiosperms, and it is commonly considered a shade tolerant plant (Hutchinson, 1975). Pearsall (1920; 1921) implied that Nitella was also tolerant of much silting. In Rat Cove, Nitella is found only in deeper water, or as an understory associated with Potamogeton crispus, where both light is low and siltation great. An attempt to define the mechanisms controlling the depth distribution of Nitella flexilis in Lake George was reported by Stross (1979). Stross proposed that the density of the Nitella meadows were nutrient limited, while depth limits were set by a photomorphogenetic switch mechanism. This mechanism is supposedly invoked by a phytochrome-like response to the spectral quality and intensity of the ambient light.
The depth distributions of the species in the fourth community type are also broad, with maxima ranging from 2 to 4 m depth. Vertical structure of the individual species can be seen in Figure 60. A comparison of their distribution maps shows (Figures 55, 56) that the distribution of Chara vulgaris is distinct from that of Nitella flexilis. Separate zones of Charaphyte species were first described by Wood (195 0). Chara is found throughout the depth gradient occurring from less than 0.5 to 7 m depth. Similar depth ordinations were published by Denny (1972a) for a species of Chara in an African lake.
Potamogeton pectinatus is most abundant at depths of 2 to 4 m. Others report finding this species restricted to shallow waters (Love and Robinson, 1977; Sheldon and Boylen, 1977). Occurring between 1.5 to 5 m of water, the depth distribution of Potamogeton praelongus in Rat Cove is consistent with its occurrence in other lakes (Campbell, 1971; Love and Robinson, 1977; and Sheldon and Boylen, 1977). Spence and Chrystal (1970) describe this species as shade tolerant.
The distribution of Potamogeton amplifolius is restricted to one cluster located in quadrant 1 of Rat Cove (Figure 54). Love and Robinson (1977) also found this species in a narrow band in 2 to 4 m of water. Other studies found this species in 1 to 1.75 m (Wilson, 1935) and 2 to 7 m depth (Sheld on and Boylen, 1977). Of all the Potamogeton species in Rat Cove, P. pusillus occurs at the greatest depth. Wilson (1935) and Campbell (1971) both found this species to be the deepest occurring macrophyte in the lakes studied. In Lake George, this species is found to a depth of 9 m (Sheldon and Boylen, 1977).
Najas flexilis (an annual plant overwintering by seed only) is mostly found between 1.5 and 4.5 m depth. It may be restricted from shallower water since seed ge rmination is inhibited by light (Wetzel and McGregor, 1968). The maximum abundance of Potamogeton Richardsonii is at 2 m depth. This corresponds to the depth of maximum photosynthesis observed for this species by Love and Robinson (1977). Scattered populations of Potamogeton Illinoensis are found in Rat Cove between 1 and 6 m depth. This depth ordination reflects the patchiness of this species' depth distribution. Love and Robinson (1977) describe a similar distribution for this speci es in West Blue Lake, Manitoba, Canada.
The distribution of species along the sediment gradient
Direct ordinations of macrophyte species distributions along the sediment factor gradients of sediment organic matter content, % CaCO3 equivalent and extractable Fe, are presented in Figures 77, 78, and 79, respectively. Sediments high in organic matter and Fe, but low in CaCO3, are occupied by species of both community types I and III. Species included in type II are less abundant at gradient extremes but otherwise tend to occur throughout the rest of the sediment gradient. Species in community type IV appear to be restricted to sediments of low organic matter and Fe content.
The species of community type I, Nuphar, Nymphaea and Ranunculus, are most abundant on the end of the sediment gradient characterized by high Fe and organic matter content. These species are not generally described as preferring one sediment type over an other. Therefore, the apparent sediment preference may be coincidental (Hutchinson, 1975). The interrelationship between the depth, exposure and sediment gradient is most evident in the area of Rat Cove occupied by these species. Here, the water is shallow, natural wave exposure is minimal and the sediment is highly enriched. These species are most likely distributed in response to the prior two factors rather than sediment composition because of their restrictive growth forms.
Community t ype III species, Potamogeton crispus, Ceratophyllum demersum and Nitella flexilis also occur more on enriched sediments. However, they are found in deeper, more exposed sites than the species of type I (c.f. distribution maps Figures 49-56). In a study by Nicholson et al., (1975), the distribution of Potamogeton crispus was found to correlate weakly with sediment particle size. This species is generally considered to be eurytopic (Stuckey, 1979). Though Ceratophy llum demersum is without roots, its growth, when suspended over enriched mud, was found to be two times that when grown over nutrient poor sand (Denny, 1972b). Rickett (1922) also reported finding this species most often on muddy sediments. Strong correlations were found between the occurrence of Ceratophyllum and sediment textural composition (Nicholson et al., 1975). In Rat Cove, this species did not usually occur on the coarse textured, nutrient poor sediments of quadrant I (F igures 49,71).
In general, the textural composition of the sediment did not correlate well with either species distributions or nutrient concentrations in the sediment (compare Figures 70-73 with 49-56 and 66-69). Nitella growth in Rat Cove is dependent on the greater nutrient supply available from the richer sediments on which it is found. This hypothesis is supported by the fact that Nitella is not found on the poorer sediments of the same depth in quadrant I of the cove (com pare Figure 55 with Figures 66-69).
The broad distributions of the species grouped in community type II overlap with species of each of the other types. Heteranthera dubia and Vallisneria americana occur indiscriminantly along the sediment gradient. Elsewhere they have been found to grow equally well on both sandy and muddy substrates (Rickett, 1922; Vogel, 1973). Nicholson et al., (1975) found no correlation between the occurrence of these species and the textural comp osition of the sediment. It is interesting that these species show little response to sediment composition when they both have been shown to derive essentially all of their P from the sediments (Carignan and Kalff, 1980).
Megalodonta beckii is more abundant on sediments having high CaCO3, low Fe and low organic content. Each of the remaining species, Myriophyllum exalbescens, Potamogeton zosteriformis and Elodea canadensis, as seen here and elsewhere, tend to occur more on rich organic muds than sand (Rickett, 1922; Vogel, 1973; Hutchinson, 1975).
The dominant species of community type IV is Chara vulgaris. This species is most abundant on the nutrient poor, low Fe, high CaCO3 sediments of quadrant I (Figure 56). However, its distribution is not as restricted to these sediments as some of the other species in this group. Other lakes in which Chara dominates also have a low P content (Forsberg, 1965a; b). Sedimen ts can play a major role in the nutrition of Chara since P uptake was demonstrated to occur equally well from shoots and rhizoids. Further, an increase in sediment organic matter content was implicated in causing Chara to be outcompeted by angiosperms (Wohlschlag, 1950). Moreover, Chara itself affects the substrate composition by the deposition of marl [Chara was found to have an ash content of 68 %, with CaCO3 being the major component (Table 30)].
Potamogeton Richardsonii, P. praelongus and P. pusillus are each more abundant on the low organic matter, low Fe, high CaCO3 end of the sediment gradient. They can be found throughout the gradient except on the most enriched sediments. Rickett (1922; 1924) found P. Richardsonii on both sandy and muddy sediments. The sporadic occurrence of P. Illinoensis is primarily associated with high CaCO3, low Fe and organic matter sediments.
Na jas flexilis is not found on the high Fe and high organic matter sediments of Rat Cove. This may be due to the inability of this annual species' seeds to germinate in the more anaerobic sediments (Wetzel and McGregor (1968).
Most of the dense stands of P. pectinatus occur on the low Fe, high CaCO3 sediments. Elsewhere, this species has been found on both sandy and muddy sediments (Rickett, 1924). This species has also been reported to be notably tolerant of silt (Sc ulthorpe, 1967).
The single stand of P. amplifolius is located on the nutrient poor sediments of quadrant I (Figure 54). This species generally is abundant only on rich,muddy sediments (Rickett, 1922). It is likely that this species has, by chance, established itself only in one part of Rat Cove. Any subsequent expansion has probably been limited by the relatively slow vegetative reproductive outgrowth from the parent stand.
Introduction and distribution of Myriophyllum spicatum, 1986-93
In 1986, Myriophyllum spicatum was discovered in Otsego Lake (Dayton and Swift, 1987). In 1987, the species was found in isolated clones in Hyde Bay, the north end of the lake from Goodyear Swamp Sanctuary to the east shore, and in the shallows between Sunken Island and the west shore (Merrifield and Harman, 1988). By the summer of 1988, isolated small beds of M. spicatum occurred in all t he shallow bays from Black Ash Swamp (north of the mouth of Leatherstocking Creek) to the Village waterfront in the south-western portion of the lake (Stalter, 1989). By 1990, all of the littoral substrates were supporting stands of M. spicatum. In 1990, Sanders (1991) utilized a series of quadrat samples in Hyde Bay, the north end of the lake and in Blackbird Bay to show that M. spicatum effectively dominated littoral substrates throughout the lake.
Myriophyllum exalbescens has become rare concurrent with the establishment of M. spicatum. Even so, M. spicatum has not generally formed monospecific stands precluding navigation. Most clones never reach the surface, even in water less than 2 m in depth. This may be, in part, due to a larval pyralid moth, Acentria nivea (Oliver), which has been associated with M. spicatum in Otsego Lake since at least 1987 (Merrifield and Harman, 1988). Apical meristems are often harvested by these organisms resulting in the production of lateral branches without the normal increase in height exhibited by undamaged plants.
In addition to M. spicatum, three other species of rooted macrophytes have been considered pests by the recreational community. Potamogeton crispus, a species introduced in the 1930s, (mentioned above); Potamogeton pectinatus, a native plant that grows actively during the summer; and Heteranthera dubia, a native which h as negative impacts in the late summer and fall.
In order to establish the current distributions of these organisms, Gergel (Unpubl.) made three flights photographing the entire littoral area of Otsego Lake during the 1993 growing season. The distribution of P. crispus was documented on July 1, P. pectinatus and M. spicatum on August 1, and H. dubia on September 19. The distributions of the species under consideration, as well as any other obvious population s, were derived from the photographs. SCUBA and/or grab samples were used to validate species identification from the aerial photographs and to more accurately define the infralittoral boundaries of some distributions.
Final maps were made of selected areas. Figures 80, 81, and 82 illustrate the situation in Hyde Bay on July 1, August 1, and September 19 and are included as examples. Other areas where preliminary maps are being developed include the north end of the Lake and the Rat Cove (B FS) area. The data necessary for maps of the entire littoral are filed at the BFS.
In 1993, 23 species of submergent plants were recognized, compared to 25 in 1935 (Table 26). Potamogeton amplifolius, found in only one locality through the 1970s, is now commonly encountered along the west shore. Emergent plants, with the exception of the introduced purple loosetrife (Lythrum salicaria), occurred only sporadically on protected shorelines. Otherwise, macrophyte distributions are as mentioned above.