Northern Prairie Wildlife Research Center
Sago grows on bedrock, as well as on mineral bottoms whose particle sizes range from rubble (Wong and Clark 1976; Pip 1987) to fine clay. It also grows on organic bottoms ranging from peats to mucks. Sago has also been found in limnic (marl) bottoms (Rich 1966; Nichols and Mori 1971; Rich et al. 1971; Haag 1983; Bailey and Titman 1984) and in streams with bottoms of bog iron ore that had disintegrated into sands and gravels (Olsen 1950). Sago frequently occurs in a variety of substrates in a single wetland (Moore 1915; Wilson 1941; Anderson 1978).
Many investigators have mentioned that certain bottom substrates produce higher sago biomass. However, in most of these studies, many other factors could have influenced plant growth. The occurrence of sago in the most common bottom textural classes is shown in Appendix C. Therefore, in this section I will only refer to studies where the production or frequency of occurrence of sago or sago propagules was compared among substrates or related to substrate and other factors.
Schmid (1965) found that sago frequency positively correlated with increased coarseness of bottom substrates. After losses during a tropical storm, sago recovered fastest on a site where the proportion of sand in the top 2.4 cm of bottom was greater than before the storm (Oglesby et al. 1976). Sago was more abundant on both silt and gravel than on sand and more abundant on sandy substrates with gravel than on pure sand in a Wisconsin stream (Madsen 1986). The presence of gravel likely provided a firm environment for roots and decreased the erosional nature of the substrate. In another Wisconsin stream, sago cover was directly related to the amount of sand in the substrate because the plant was better adapted to rooting in sand than the other species found there (Madsen and Adams 1985). Mace et al. (1984) found that sago was common in Wisconsin stream stretches with both Type 1 (sand, gravel, or rubble) and Type 2 (silt) bottoms and suspected that macrophytes in Type 1 stretches obtained nutrients mostly from the water column, whereas those in Type 2 obtained nutrients from the sediments. Sandy, gravelly substrates studied by Kautsky (1987) contained only 40% the density of sago drupelets found in sheltered muds, even though plants in the former substrates allocated 4 times the biomass to reproductive parts. Haag (1983) also observed unfavorable drupelet production and seedling growth on sandy substrates characteristic of turbulent, nutrient-poor conditions. The ability of sago to exploit sandy substrates low in nutrients may relate to benefits provided by mycorrhizas on the root hairs (Clayton and Bagyaraj 1984).
Butcher (1933) and Jensen (1940) indicated that best sago growth occurred where organic-rich silt was deposited on other substrates. Howard-Williams and Allanson (1981) showed that thin (1-2 cm) organic layers are a major source of phosphorus for sago and that long-term transfer of particulate nutrients into deep sands below can be a slow process as long as the sago beds remain intact. Laboratory studies of Peltier and Welch (1969), where sago roots were sealed off from the water column, showed that stem lengths increased faster in natural river sediment than in masonry sand regardless of nutrient content of the water column. In the river from which these plants were collected, sago likely met much of its nutrient demand from silts accumulated in the interstices within the gravel bottom. Ravanko (1972) observed that sago was dominant in sand but was replaced by Potamogeton perfoliatus as bottoms became more fine textured.
Silt bottoms are capable of high sago production, and have several times been singled out as especially productive of turions (Wetmore 1921; Craner 1964; Sincock 1965). Ho (1979), nonetheless, rated silt poor for sago production. Silt loams were termed best for sago production by Low and Bellrose (1944). Substrates supporting sago contained more silt and clay than nearby bottoms supporting Potamogeton crispus (Rich 1966).
Sago commonly occurs on clays, and such bottoms can be especially productive of turions (Wetmore 1921; Jensen 1940; Craner 1964). Van Wijk et al. (1988) showed that a mixture of 75% washed clay and 25% washed sand was an excellent medium for sago culture. In an African lake, sago grew in monodominant stands where bottoms had the highest (up to 50.9%) clay content (Denny 1972). Sincock (1965), though, found that sago was infrequent on clays and peats compared with sands and silts and suggested that fine clays could prevent adequate rooting. Ordination models of Paullin (1973) placed sago among a group of plants associated with soils containing a low clay fraction and high organic matter content. Anderson and Jones (1976) and Anderson (1978) also rated clays rather poor for sago production compared with sands, silts, and loams, but in the lakes they studied the heaviest soils were subject to the most wave action. Wetmore (1921) found fewest turions in areas of calcareous hardpan. The probable key to abundant sago production in clays is protection from wave action (Jupp and Spence 1977b). Loams (roughly equal mixtures of sands, silts and clays) have been identified as especially productive of sago (Wetmore 1921; Low and Bellrose 1944; Craner 1964; Anderson 1978; Olson 1979).
Bottom deposits consisting mostly of organic matter in various stages of decomposition are termed fibric (peat), hemic (muck), and sapric (sapropel or gyttja). These often support sago, and McDonald (1951) found that sago was about twice as frequent in bottoms high in organic matter. Nevertheless, Sincock (1965) seldom encountered sago on peats, and Haag (1983) found that the plant was absent for at least 5 years from a site where sediments were highest (27% dry weight) in organic matter, but present for 1 or more years where organic matter ranged from 12% to 26% of dry weight. Table 7 shows the range of organic matter and organic C content in sediments supporting sago.
Several studies indicate that bottom substrate is of little consequence to sago production (Smeins 1967; Van der Valk and Bliss 1971; Olson 1979) or that the plant occurs in a variety of substrates (Pip 1987; Hammer and Heseltine 1988). Sago plants that start from sprouted turions can grow at least 21 days in nutrient solution only (Devlin and Karczmarczyk 1975; Spencer 1986a).
Substrate likely can affect sago's ability to dominate certain sites. For example, in an Indian irrigation area Vallisneria spiralis generally replaced sago where bottoms were silty but not in nutrient poor sandy bottoms (Reeders et al. 1986).
Anderson (1978) found that substrate texture, combined with turbidity and water depth, affected light availability in a large prairie wetland, and he concluded that light intensity limited sago colonization. In his study, firm clays and wave-washed sands found at exposed sites limited high sago production to sheltered shallow bays with peat and sandy-loam soils. Similarly, sago was found throughout the lake studied by Haag (1983), but maximum abundance tended to occur in fine sediments in shallow protected bays.
On a large, shallow (< 1.2 m) clay-bottomed lake, sago was also absent from the central area of greatest fetch, but that bottom was not noticeably different from elsewhere in the lake except the narrow swash zone. Instead, the absence of sago in the central zone was attributed to constant turbidity caused by colloidal or suspended clay particles (Kantrud 1986c, unpublished).
I conclude that sago is not substrate-dependent and that its distribution and abundance often depend greatly on wave action and fetch as they affect turbidity and the distribution of soft, easily colonizable sediments. Aversions noted for certain bottom textural classes likely are caused by other factors. Sago from the same locality may have genotypes adapted to different conditions (Kautsky 1987). Recent work by Van Wijk (1988, 1989) and Van Wijk et al. (1988) suggests that these genotypes differentiate freshwater and brackish-water populations, so it would not be surprising if genetic adaptations to substrate types, or the environmental conditions associated with them, were discovered in the future.
Sedimentation and Disturbance
Sediments are moved by wave action to central deeper water areas or trapped by vegetation in peripheral sheltered areas. Beds of submersed aquatic macrophytes in shallow areas of great fetch are likely to be damaged by unstable bottom sediments (Vicars 1976). Plants with feathery surfaces that are easily coated with or weighted down by sediment are at a disadvantage to the linear leaved sago (Schiemer and Prosser 1976; Vander Zouwen 1982). In openwater areas sago growths can form ring patterns or atolls that formed, according to Varga (1931, cited in Schiemer and Prosser 1976), because of silt deposition in the interior and silt erosion on the periphery.
Although Paullin (1973) thought that sago was ineffective in stabilizing bottom substrates, others (Butcher 1933; Haslam 1978, 1987; Reese and Lubinski 1983) remarked on the species' silt tolerance or ability to consolidate unstable substrates. Madsen and Adams (1985) found that sago was most successful in portions of a Wisconsin stream where growth of other species was limited because of siltation, turbidity, and pollution. Peltier and Welch (1969) saw sago survive 10-30 cm silt deposition in a single season. Sago reached high biomass in silted wetlands (Logan 1975), and Singhal and Singh (1978) considered the species an indicator of siltation. In five water bodies studied by Doherty and La Roi (1973), sago was restricted to the only lake with measurable sediments (3 cm). Sago rhizomes filled an 8-10-cm-thick layer of freshly deposited silt in a single growing season as older rhizomes below died out (Gladyshev and Kogan 1977).
The effects of parent substrate, sedimentation, and decreases in water transparency caused by sediments in the water column are, of course, difficult to separate. Thus Bourn (1932) noted that the great loss of submersed macrophytes in Currituck Sound, North Carolina, occurred almost simultaneously with the sedimentation that resulted from opening and enlarging the Albemarle and Chesapeake Canal. He attributed this loss to the excessive turbidity caused by wind and wave action that kept the fine particulate sediments in almost constant suspension. Bourn believed that only the reestablishment of aquatic plants would return the water to its former clarity.
In heavily silted sites prone to wind-induced turbidity, sago abundance varied greatly with water depth (Jackson and Starrett 1959). Bellrose et al. (1979) documented large losses of sago and other submersed hydrophytes from lakes in the Illinois River valley and concluded that, although altered water levels were locally important, increased turbidity from sedimentation was far more significant.
Otto and Enger (1960) showed that 50-100 ppm suspended natural sediments caused a 50% reduction in weight of cultured sago plants; effects of higher loadings included stem elongation, chlorotic leaves and stems, and apical dominance. Sago was absent from a New Zealand lake where suspended solids measured 100-300 mg/L (Gerbeaux and Ward 1986). Ongoing experiments of Butler and Hanson (1985, 1986, 1988, unpublished) indicate that virtually no sago growth in a lake where concentrations of suspended solids were 40-100 ppm during the growing season. They also found that, even in the absence of fish, resuspension of fine (< 5-micron) sediments was sufficient to maintain extreme turbidity.
Sago is a common inhabitant of heavily traveled boat canals, where bottoms are frequently disturbed or dredged (Van Donselarr et al. 1961; Murphy and Eaton 1983; Haslam 1987). Davis and Brinson (1980) placed sago into a group of five species of submersed hydrophytes in North America able to maintain dominance in disturbed ecosystems and considered this consistent with sago's widespread abundance. Sago occurred in wetlands subject to both weed removal and heavy recreational boating (Kaul and Zutshi 1967). Stuckey (1971) also remarked on sago's tolerance to physical destruction by boat traffic. However, sago disappeared where heavy recreational boating disturbed bottom sediments and greatly increased turbidity (Cragg et al. 1980).
Sago drupelets, deeply buried in dark, reducing environments of lake sediments, might be released from dormancy by disturbance (Haag 1983). Haslam (1978) stated that the overwintering propagules of sago are harmed little by soil disturbances common to rivers. Sago was among a group of macrophytes that increased after a river was dredged (Hannan and Dorris 1970).
Logan (1975) attributed good sago growth in silted impoundments to the species' rhizomatous growth form. Potamogeton turions can be so firmly embedded in clay bottom substrates that removal is difficult (Jupp and Spence 1977b). However, certain loose mucks may increase sago's susceptibility to disease (Bourn and Jenkins 1928).
In conclusion, the linear growth form of sago leaves is an obvious advantage in sites where broad leaved species can be weighted down by sediment deposition. Sago's rhizomatous growth form can consolidate silted bottoms, and sago propagules are resistant to disturbance of bottom substrates. Nevertheless, sedimentation and associated turbidity caused by suspended solids can decimate sago populations. Plants can survive up to 30 cm of silt deposition in a single growing season. Suspended sediments > 100 ppm during the growing season will lower sago biomass greatly at most sites where the plant grows. Sago can reach high biomass in silted wetlands, but not if fetches are large and silts are often and easily suspended by wave action.
Bue (1956) observed little difference in sago frequency in heavily or lightly grazed South Dakota livestock ponds. Yeo (1965) found a dense stand of sago in an irrigation canal heavily grazed by cattle. On Stump Lake, North Dakota, I saw only slightly less sago along heavily grazed compared with ungrazed shorelines, although plants prospered in much shallower water under the latter regime because they were protected by semi-open stands of emergent hydrophytes that grazing had nearly eliminated elsewhere. Barnett (1964) thought the rarity of sago and other submersed macrophytes in a Utah lake was partly attributable to a combination of livestock grazing and carp activity. Here the carp is suspect because Potamogetons are generally favored by the effects of domestic livestock on emergent hydrophytes (Duncan and D'Herbes 1982). In the French Camargue, sago is a dominant submersed macrophyte in many grazed marshes, and it also commonly occupies open areas artificially produced by mowing of emergents (Britton and Podlejski 1981). Sago showed a general decrease in frequency in brackish drainage ditches in England's Norfolk Broads when the area was converted from grazing to the production of cereal grains. However, frequency was maintained in an adjacent area where grazing was continued (Driscoll 1986). I conclude that grazing by domestic livestock encourages sago production, except in sites exposed to wave action or those where bottom-feeding fish are present.
Little information is available on growth of sago on sloping substrates. Nonetheless, the plant seems very tolerant. Boltt et al. (1969) reported it on sandy underwater slopes that angled as much as 31° from the horizontal; they postulated that the absence of several other species from these sites was due to the delicate broad-leaved growth form that increased susceptibility to damage from wave action.
Sago occurs in bottom sediments with salinities of extracted waters up to 24 g/L TDS or conductivities up to 27 mS/cm in inland, likely SO4-dominated, areas (Smeins 1967; Walker and Coupland 1970; Anderson and Jones 1976; Kollman and Wali 1976; Hammer and Heseltine 1988). These published figures are probably conservative because water column salinities of up to 104 g/L TDS have been measured in wetlands supporting sago (Ungar 1970). Ungar (1974) cited studies by Flowers (1934) and Flowers and Evans (1966) that showed that sago grows in Cl-dominated bottoms containing > 2% salts. Craner (1964) recorded > 19 g/L salinity in Cl-dominated sediments 0.3 m deep in wetlands supporting sago.
The salinity of sediments supporting sago is usually higher in the thin oxidized layer at the substrate-water interface than in the reduced layer below it (Kollman and Wali 1976). In a North Dakota wetland, substrate salinity rose slowly during the growing season from 8 g/L to 9 g/L and then rose abruptly to 14 g/L in October (Kollman and Wali 1976). The salinity of the water column also rose during the growing season but fell in October. In Utah wetlands supporting sago, salinities of bottom substrates increased during summer at depths up to 0.15 m but did not change noticeably in deeper strata. No correlations were found between salinities of bottom substrates and production of sago vegetation, turions, or drupelet heads in these wetlands. Instead, production of vegetation and drupelet heads was correlated with salinity of the water column (Craner 1964). Spencer (1987) found that sago turion weight directly related to sediment pH and inversely related to redox potential in California irrigation canals, but turion density was not strongly related to either variable.
Although sago has a poorly developed vascular system (Arber 1920, cited in Stevenson and Confer 1978; Sculthorpe 1967) and absorbs much of its nutrient supply through the leaves (Ho 1979), roots can mobilize all major nutrients (N,P,K) from bottom sediments (Kollman and Wali 1976; Huebert and Gorham 1983). Table 7 lists some of the chemical constituents found in sediments supporting sago.
Jensen (1940) found that there was NH3 depletion in bottoms supporting dense sago stands. Haslam (1978) recorded poor correlations between concentrations of this ion and sago distribution; rather, the plant tended to be found in sediments intermediate in NO3-N. Schiemer and Prosser (1976) found that there was greater sago biomass in soft nearshore sediments higher in N than in shallower offshore sediments. Purohit (1981) also found that the best sago stands were in sediments high in N. Sago was almost always present in Canadian lakes where extracted waters from bottom sediments contained 2.0 mg/L NO3-N (Hammer and Heseltine 1988). Concentrations of N in sago tissue can be 10-18 times that found in the sediments, but 1,000-5,000 times that of the water column (Gopal and Kulshreshtha 1980).
Sediment P seems especially important for good sago production (Schiemer and Prosser 1976; Jupp and Spence 1977b; Anderson 1978; Haslam 1978). Howard-Williams and Allanson (1981) showed that the upper 5 cm of sediment was a major source of P for an extremely dense (> 1,000 shoots per square meter) bed of sago in the littoral of a large wetland where water-column P was low (usually < 5 microg/L). In this wetland, competition for P by macroalgae, epiphytic algae, and invertebrates was strong, and little P was transferred to the pure sands below a surface organic layer < 2 cm thick. An overabundance of sediment P may increase potential competition from angiosperms, for Peverly (1985) found that P values in sediment and in interstitial water at two sites dominated by sago were lower than at a site dominated by Elodea canadensis, Potamogeton crispus, and Fontinalis antipyretica. Jones and Cullimore (1973) indicated that Myriophyllum spicatum is adapted better than sago to wetlands where sediment and water column P both are high. Kollman and Wali (1976) found that P and K content of sediments that supported sago increased constantly during the latter part of the growing season, which they attributed to equilibrium shifts from water to sediment as lake waters became concentrated through evaporation or when organic materials broke down in the sediment. Fetter et al. (1978) found that sago was present in sediments containing up to 9.1 mg/g dry weight P, but submersed plants were absent upstream near a wastewater treatment plant where levels reached 20.6 mg/g. Hammer and Heseltine (1988) found that sago was present in sediments whose extracted waters contained 12 mg/L P. Sago tissue can contain 13-125 times the P concentrations found in the sediments but more than 3,000 times that found in the water column (Gopal and Kulshreshtha 1980).
Anderson (1978) found that sago production, and the proportion of sand in sediments, was negatively correlated with available K in bottom sediments. In his study, sago grew better in substrates with a high sand fraction; the results could have been coincidental or reflected a true aversion of sago to high K concentrations. Haslam (1978) assigned sago to a group of plants found mostly in sediments intermediate in K concentration of the interstitial water. Peverly (1985) found that there were higher K concentrations in both sediment and interstitial water at two stream sites dominated by sago, compared with a site dominated by other submersed macrophytes. Sago occurred in monotypic stands in sediments whose extracted waters contained up to 450 mg/L K (Hammer and Heseltine 1988). Phosphorus in sediments that supported sago increased almost linearly late in the growing season when evaporation occurred, probably because of equilibrium shifts within the water column, which increased many ion concentrations (Kollman and Wali 1976).
Paullin (1973) showed with ordination models that sago tended to occur in wetlands with bottom sediments high in Ca and organic matter, although the plant was among the most widely distributed of all submersed hydrophytes in the area studied. Purohit (1981) also found that best sago stands were in sediments high in Ca and organic matter. Sago showed poorer competetive ability in streams with sandstone (SiO23) bottoms, especially in upper sections where plants with lesser nutrient requirements were found (Madsen and Adams 1985).
Haslam (1978) related sago distribution to ionic concentrations of water extracted from riverine sediments; sago tended to be found in sediments low in Mg and intermediate in Ca and SO4. Correlations between sago distribution and Na concentrations were poor. Gopal and Kulshreshtha (1980) found that sago tissue concentrated Ca and Mg at respective levels of 400 times and 5,000 times that found in the sediments.
In summary, sago grows where salinity (TDS) of waters extracted from bottom substrates is at least 24 g/L. Sago biomass and propagule production seem better related to water-column salinity than to bottom substrate salinity. Sago can tolerate high levels of various elements and compounds found in bottom substrates (Table 7); most of these have not been proven to limit growth or distribution of the plant. Although sago can mobilize all major nutrients (N,P,K) from sediments, water-column nutrients likely are more important in most wetlands, due to sago's ability to absorb needed nutrients through its leaves. Sago distribution sometimes, however, can be related to concentrations of sediment Ca or Mg.