USGS Home
Contact USGS
Search USGS

Reduced sedimentation rates of phosphorus (as percent of water column totals) have been associated with presence of fish in previous research (Mazumder et al. 1989, Vanni et al. 1997), and lower sediment accumulation in Sagebraten may reflect domination of the phytoplankton by small-celled algae with low sinking rates (Mazumder et al. 1989). Predominance of small-bodied phytoplankters would be expected in Sagebraten due to the paucity of large-bodied cladocerans. Scarcity of submerged and emergent macrophytes in Sagebraten may also contribute to lower sedimentation rates, as lack of macrophytes promotes water-column turbulence, hindering deposition of particulate mater (Graneli and Solander 1988). Lower sedimentation rates result in phosphorus being retained in the water column of Sagebraten, and is manifested by phytoplankton and seston comprising nearly half the total phosphorus measured in this wetland.

High phytoplankton biomass in Sagebraten likely results from both decreased zooplankton herbivory (an indirect result of fish predation) and high nutrient availability in the water column (Vanni and Layne 1997). Higher algal abundance in Sagebraten undoubtedly contributed to higher turbidity. However, turbidity gradually increased throughout the summer while phytoplankton biomass remained relatively constant, implicating additional factors. Resuspended sediments also may have increased turbidity, and low macrophyte biomass in Sagebraten obviously facilitated mixing and disturbance of sediments.

Reduced abundance of macrophytes and periphyton in Sagebraten is likely due to lower water clarity (Graneli and Solander 1988), with periphyton biomass also limited by lack of macrophyte substrate. Lower invertebrate biomass likely results from minnow predation and limited foraging and refuge areas due to reduced abundance of macrophytes and associated periphyton. Aquatic invertebrates are important foods of fathead minnows (Held and Peterka 1974, Price et al. 1990), and these fish have been associated with reduced invertebrate abundance in prairie wetlands (Hanson and Riggs 1995). Invertebrate biomass and diversity are also higher in macrophyte beds, mainly due to low disturbance (Beckett et al. 1992), decreased risk of predation (Crowder and Cooper 1982), and increased food resources (Carpenter and Lodge 1986). Lower abundance of macrophyton and periphyton in Sagebraten may lead to low diversity of macroinvertebrates and a reduction in functional feeding groups like grazers, scrapers, and shredders.

Higher interstitial concentrations of phosphorus in Rollag may be due to higher abundance of macrophytes or physical characteristics of the wetland site. Dense beds of macrophytes reduce water column mixing and favor stagnation near the sediment surface, supporting reducing conditions and higher interstitial phosphorus levels (Graneli and Solander 1988). Limited turbulence in concert with intense photosynthesis by submerged macrophytes and their associated periphyton may also contribute to high pH at the sediment surface, further enhancing interstitial phosphorus levels (Graneli and Solander 1988).

Higher interstitial phosphorus concentrations in Rollag indicate greater mobilization of sediment phosphorus, which is "available" for incorporation into biotic pools. Possible fates of this available phosphorus include uptake through roots of macrophytes or mobilization to the water column. Yet water-column phosphorus levels in Rollag remained low, as did phytoplankton biomass. This indicates that most of the interstitial phosphorus was taken up by macrophytes directly from sediments, or perhaps was released to the water column but rapidly utilized by periphyton. Chara may play a particularly important role in maintaining low phosphorus and phytoplankton concentrations in Rollag, as these plants have a high capacity to absorb phosphorus (Kufel and Ozimek 1994). In contrast, interstitial phosphorus moving into the water column in Sagebraten was probably incorporated into phytoplankton and seston pools, as these pools increased while the macrophyte and periphyton pools remained largely unchanged.

Higher water-column phosphorus in Sagebraten may also result from fathead minnow activity. Fish increase water-column phosphorus levels in numerous ways, but the magnitude of influence is largely dependant on the species, size structure of the population, and diet (Carpenter et al. 1992a). Fathead minnow populations exhibit several characteristics that increase their potential for phosphorus cycling. Egestion and excretion of nutrients by fish can increase water-column phosphorus concentrations (Schindler et al. 1993, Vanni et al. 1997), and impacts are greater when fish forage on benthic prey due to stoichiometric ratios of predators and prey (Schindler and Eby 1997). Additionally, diets high in benthic or littoral prey facilitate higher water-column phosphorus concentrations by translocating nutrients from inshore and benthic areas (Braband et al. 1990, Carpenter et al. 1992b). A large portion of the diet of fathead minnows in Sagebraten was composed of benthic invertebrates such as chironomids (Zimmer et al. unpublished data), perhaps enhancing concentration of phosphorus in the water column.

Additionally, young-of-the-year (YOY) fish are metabolically very active and have large impacts on phosphorus recycling (Kraft 1992). Fathead minnow populations often appear to be dominated by YOY fish; Payer and Scalet (1978) reported that 99% of annual fathead production in a prairie wetland was by YOY fish. In our study, YOY minnows averaged 53% of total fish biomass in Sagebraten, and on the last sampling date they represented 98% of total fish biomass. Additionally, most adult fathead minnows die after spawning (Peterka 1989) and we also observed high mortality rates of YOY fish. Death and decomposition of fish can result in a substantial nutrient flux to the water column (Threlkeld 1988), and we estimated 259 kg of fathead minnows•ha^-1 died during the sampling period. Assuming 50% of phosphorus in these fish is quickly released upon death and decomposition (Smith et al. 1977), 0.65 kg of phosphorus•ha^-1 was released from this pool. This is a substantial amount, representing 16% of the total amount of phosphorus measured in all pools on 4 August (4.07•ha^-1). Thus, decomposition of post-spawn adults and YOY fish may result in a consistent flux of phosphorus from fish to the water column pool.

Obviously fathead minnows influenced the ecosystem structure of Sagebraten and this site was functionally different than Rollag. Sagebraten demonstrated characteristics typical of the turbid-water state, whereas Rollag emulated the clear-water state. Additionally, Sagebraten was characterized by a "pelagic" type food web, with open-water nutrient pools of phytoplankton, seston, and fathead minnows being most important. In contrast, Rollag displayed a more "littoral" food web, with macrophyton, periphyton, and aquatic invertebrates comprising major nutrient pools. Water quality also differed, with Sagebraten characterized by lower water clarity and higher phytoplankton biomass and water-column phosphorus levels. Most primary production in Sagebraten probably occurs in the phytoplankton, while macrophyton and periphyton contribute most primary production in Rollag. The sum of all phosphorus pools being 5-6 times larger in Rollag was a surprising result. Perhaps phosphorus in Sagebraten remains "tied up" in the sediments through the mechanisms described above, resulting in smaller but active water-column nutrient pools. In contrast, macrophytes in Rollag tap the sediment as a source of phosphorus, resulting in higher phosphorus•ha^-1 in the pools we examined. We feel the differences observed between these two wetlands is due to the presence and absence of fathead minnows and aquatic macrophytes.

We stress that results presented here are not meant to be extrapolated to all prairie wetlands, as this was an unreplicated study. Loss of aquatic macrophytes is a critical feature as wetlands or shallow lakes switch from a clear to a turbid water state. Results of a larger study indicate a strong pattern of higher turbidity in wetlands with fathead minnows, but not of reduced macrophyte abundance (Zimmer et al. unpublished data). This indicates that the turbid-water state is not universally associated with fathead minnows. However, loss of macrophytes is often caused by insufficient light for growth, and light availability for submerged plants is influenced by both turbidity and water depth. A wetland may have relatively high turbidity, yet be shallow enough to permit submerged plant growth. Though fathead minnows probably impact wetlands of all depths, we hypothesize that deeper wetlands with fathead minnows are most vulnerable to shifting toward a turbid-water state.