Seasonal Changes in Fungal Production and Biomass on Standing Dead Scirpus lacustris Litter in a Northern Prairie Wetland

Abstract

Decaying macrophytes are an important source of carbon and nutrients in fungal and bacterial communities of northern prairie wetlands. Dead macrophytes do not collapse into the water column immediately after death, and decomposition by fungi and bacteria begins while the plants are standing. The seasonal variations in fungal biomass and production on Scirpus lacustris stems, both above and below water, were measured to assess which environmental factors were dominant in affecting these variations in a typical prairie wetland. Fungal biomass and production were measured from early May to November, just prior to freeze-up. Fungal decomposition began and was greatest in the spring despite low water temperatures. The fungal production, as measured by the incorporation of [1-¹⁴C]acetate into ergosterol, ranged from 1.8 to 376 μg of C g of ash-free dry mass (AFDM)⁻¹ day⁻¹, and the biomass, as estimated by using ergosterol, ranged from nondetectable to 5.8 mg of C g of AFDM⁻¹. There was no significant difference in biomass or production between aerial and submerged portions of Scirpus stems. The water temperature was correlated with fungal production (r = 0.7, P < 0.005) for aerial stem pieces but not for submerged pieces. However, in laboratory experiments water temperature had a measurable effect on both biomass and production in submerged stem pieces. Changes in fungal biomass and productivity on freshly cut green Scirpus stems decaying in the water either exposed to natural solar radiation or protected from UV radiation were monitored over the summer. There was no significant difference in either fungal biomass (P = 0.76) or production (P = 0.96) between the two light treatments. The fungal biomass and rates of production were within the lower range of the values reported elsewhere, probably as a result of the colder climate and perhaps the lower lability of Scirpus stems compared to the labilities of the leaves and different macrophytes examined in other studies performed at lower latitudes.

Emergent macrophytes are an important source of carbon in wetlands of the northern prairies (42). There is limited knowledge concerning the significance and regulating factors of fungal processes in the cycling of carbon and nutrients that generates the high rates of biological productivity of prairie wetlands, which support 50 to 80% of North America's waterfowl populations (45). In previous measurements of macrophyte decomposition in these wetlands workers have had to rely on nutrient and litter mass loss, and there has been little differentiation between the fungal and bacterial components responsible for decay (41, 46, 47). The use of ergosterol as an index molecule to estimate eumycotic fungal biomass (25) and the technique involving incorporation of [1-¹⁴C]acetate into ergosterol (29, 31) to measure in situ fungal production rates have resulted in considerable advances in our understanding of litter decomposition and microbial interactions in a variety of environments. Fungi have been found to be the predominant decomposers in some aquatic environments, accounting for up to 99% of the total microbial biomass and production (21, 22, 24, 32, 33, 42).

Researchers have measured fungal colonization on aquatic macrophytes in the standing position (15, 30, 33, 34) and in plants submerged in water (22, 24). Using the ergosterol labeling technique, these workers have obtained evidence that a significant portion of carbon from decaying plant material is converted into microbial mass while the plants are in the standing position. Work on standing dead macrophytes that have collapsed into the water column has shown that there is an initial lag phase, due perhaps to the death of terrestrial fungi or to the conditioning of the plant material, before colonization by aquatic fungi can occur. As they decompose, standing dead macrophytes in wetlands are often partially under water or under sediment, and fungal decomposition does not occur entirely in the standing or submerged position.

Microbial organisms responsible for decomposition in macrophytes are subject to extremes of temperature and UV solar radiation throughout the year, especially in shallow wetlands. Surprisingly, temperature as a factor that regulates fungal production and biomass in decaying organic material has received little attention (8).

UV radiation is known to have damaging effects on bacterial DNA (20), and UV radiation may also be debilitating to fungi and retard their decomposition activities in aquatic systems. Despite the negative effects of UV radiation, the overall effect may actually be beneficial to bacteria due to the production of more labile forms of dissolved organic carbon (DOC) (19). Likewise, UV radiation may speed up the crumbling process in plant litter (2, 11), thereby making it easier for fungi to extract carbon from plant material. Two studies of the solar radiation effects on fungi as they decompose litter in an aquatic environment have produced different results, with one study showing negative effects (10) and the other showing no effects (9).

There have been no studies that have monitored, with regular sampling from spring to fall, seasonal changes in fungal biomass and production on naturally standing dead litter of emerging macrophytes in an inland aquatic system. In this study we assessed seasonal changes in fungal biomass and production on standing dead litter of the hardstem bulrush Scirpus lacustris in a northern prairie wetland. We also experimentally examined the effects of solar radiation and water temperature on fungal colonization and decomposition of this bulrush.

MATERIALS AND METHODS

Study site.

Field studies were conducted at a wetland that typifies the prairie ecozone at the St. Denis National Wildlife Area 40 km east of Saskatoon, Saskatchewan, Canada (106°06^′W, 52°02^′N). Pond 50 is a permanent, slightly saline (mean concentration of total dissolved salts, 11.1 ± 2.69 g liter⁻¹), mesotrophic wetland with a normal maximum mean depth in July of 1.4 m. During the present study the pond depth declined from a maximum of 58 cm in May to <20 cm by late October. The major water loss is via evaporation; hence, the DOC concentration in the pond on 5 May was 40.8 mg liter⁻¹, and by September 28 it was 120 mg liter⁻¹ (mean, 74 ± 24 mg liter⁻¹) (44). The plant flora contributing the largest fraction of organic matter production to Pond 50 is Scirpus lacustris L. p.p., a hardstem bulrush.

Fungal biomass and production.

Standing dead Scirpus was sampled biweekly for fungal biomass and production from May, after the ice cover left the pond, until just before the ice returned in November. Seven stems were chosen at random on each sampling date, and the air-water interface was marked. Two-centimeter pieces of stem from below the interface were used for the in-water measurements; aerial measurements were made with 2-cm stem pieces from 5 cm above the interface. As the water receded during the summer to the sediment surface, the in-water incubation experiments done on and after July 7 (Julian day 194) involved stem pieces taken from the upper part of the sediments, which were loosely consolidated.

Fungal biomass was measured by determining the ergosterol concentration. Variations of the optimum extraction procedure for ergosterol were examined (43), and the method described by Eash et al. (12) was found to be the most suitable method for plant matrices. The contents of the incubation tubes used to measure fungal productivity (see below) were emptied into 35-ml pressure reaction tubes (Alltech, Deerfield, Ill.). Methanol (15 ml) and 5 ml of a 40-g liter⁻¹ solution of KOH in 95% ethanol were added to the tubes, and the contents were vortexed for 1 min and sonicated (Branson [Danbury, Conn.] model 5210; nonadjustable setting) for 1 min. The tubes were then placed in an 85°C water bath for 30 min and hand mixed after 15 min. The tubes were allowed to cool, and the contents were filtered through Whatman no. 41 paper into separatory funnels. In addition, the filters were rinsed with 5 ml of methanol. The mixture was extracted with pentane three times (10 ml each), and the solvent was evaporated under N₂ gas. The dried extract was redissolved in 5 ml of methanol and filtered through a 0.2-μm-pore-size syringe filter. Syringe filters were tested for absorbency or reactivity with ergosterol by filtering known standards and examining the filtrate for loss of ergosterol. No detectable ergosterol losses were observed. The extract volume was reduced to 1 ml by evaporation with N₂ gas. A 500-μl aliquot of each extract was injected into a high-pressure liquid chromatography (HPLC) system to determine the ergosterol concentration, as well as to separate the ergosterol from other constituents in the extract. The HPLC system was comprised of a Bio-Rad model 1350 pump with a reverse-phase Supelcosil LC-18 5 μm HPLC column (15 cm by 4.6 mm; Supelco, Bellefonte, Pa.). The mobile phase was HPLC grade methanol at a flow rate of 1.81 ml min⁻¹. The ergosterol retention time was approximately 4.5 min, and ergosterol was detected with a Dionex (Sunnyvale, Calif.) variable-wavelength detector set at 282 nm. Ergosterol in the plant samples was identified and quantified by comparison with ergosterol standards (Sigma Chemical Company). The ergosterol standards gave linear calibration (r² = 0.99). The detection limit under these conditions was approximately 0.33 ng on a column. Suspect ergosterol samples were confirmed by mass spectrometric analysis performed as described by Headley et al. (18). In addition to the samples, one glassware blank (all nondetectable) and one spiked blank for the percentage of recovery or loss (average recovery, 85%) were included for quality control purposes on every extraction date. The conversion factor used was 10 μg of ergosterol/mg of living fungal carbon and is based on the assumption that the carbon content of fungi is 50% (16, 24).

Fungal production was measured from the rate of sodium [1-¹⁴C]acetate uptake. Stem pieces were incubated in triplicate in 10-ml polycarbonate tubes (VWR Canada) containing 5 ml of filtered (pore size, 0.2 μm) pond water and 1 mM sodium [1-¹⁴C]acetate with a specific activity of 58 to 59 mCi mmol⁻¹ (Amersham Pharmacia Biotech, Montreal, Canada) for 3 to 4 h. Both the submerged and aerial stem pieces were incubated under natural light conditions in the standing Scirpus canopy. The production values for aerial stem portions are only estimates of the potential production since the incubation method involved wetting the stems. (Kuehn et al. [23] have observed significant increases in microbial activity, as measured by CO₂ [fungal plus bacterial] evolution, within 5 min of wetting of dry Juncus effusus litter.) The tubes containing submerged stem pieces were placed in the pond water during incubation, while the tubes containing aerial stem pieces were placed on the ground. One killed control per sampling date (with organisms killed by using formalin [final concentration 2%]) was included on every incubation date (27). All incubations were stopped by adding formalin (final concentration, 2%). Labeled ergosterol fractions, after they were run through the HPLC system for ergosterol quantification, were collected in scintillation vials, dried under N₂ gas, and redissolved with 5 ml of Insta-Gel XF (Canberra Packard). The numbers of disintegrations per minute were determined with a Packard scintillation counter by using an external standard to determine quench.

In preliminary experiments we measured radioisotope dilution that occurred when Scirpus stems were incubated in Pond 50 water. The concentration of cold (unlabeled) sodium acetate was varied, while the concentration of labeled acetate was kept constant. The concentrations of combined cold and hot sodium acetate used were 1, 2, 4, 6, and 12 mM, and these concentrations were added to duplicate samples. The isotope dilution was established by using a nonlinear equation to analyze the rate of uptake versus the total added acetate concentration (36). Instantaneous rates of fungal production were calculated from the rates at which [1-¹⁴C]acetate was incorporated into ergosterol; the conversion factor (19.3 μg of new fungal biomass per nmol of acetate incorporated) (24, 40) was multiplied by the mean isotope dilution value.

The remaining three stems were cut as indicated above, and the ash-free dry mass (AFDM) was determined after combustion for 4 h at 450°C (24, 27). Both the incubated stem pieces and the stems used to determine AFDM were transported to the laboratory on ice and weighed immediately upon arrival (about 30 min later). Incubated stem samples were stored at 4°C and extracted the following day.

Solar radiation experiment.

A floating, two-compartment chamber constructed of Plexiglas with a mesh bottom to allow water exchange while keeping out macrograzers had two different Plexiglas covers (thickness, ∼3 mm); one cover allowed penetration of full sunlight, including UV radiation (UV⁺), while the other excluded UV radiation (UV⁻) but allowed penetration of all other wavelengths. Scanning measurements for the UV⁻ Plexiglas cover, obtained with an Optronics scanning spectroradiometer (model OL-754), showed that it allowed no UV-B (wavelengths, 280 to 320 nm) through, 0.1% of the UV-A (wavelengths, 320 to 400 nm) through, and 89% of the photosynthetically active radiation (PAR) (wavelengths, 400 to 700 nm) through. The UV⁺ Plexiglas cover allowed 76% of the UV-B through, 85% of the UV-A through, and 88% of the PAR through. The chamber was allowed to acclimate in Pond 50 for 10 days before introduction of plant material. Living Scirpus plants were harvested, and stems having similar diameters were selected and cut into 2-cm pieces in the lab. Twenty-four hours after harvesting, equal amounts of cut stems were placed in the two sides of the chamber. In addition, stem samples were taken at the start and at the end of the experiments to measure particulate organic carbon (POC) and particulate organic nitrogen (PON) contents. POC and PON were analyzed with a CHN analyzer.

The water level in the chambers was maintained at a depth of 4 cm throughout the experiment. Although the data varied over the course of a season, the average values for the depths of penetration of 1% of the UV-A and UV-B were 13 and 4 cm, respectively (44). For the first 2 weeks the majority of the stems floated on the surface of the water and therefore were exposed to the radiation penetrating the Plexiglas. Subsequent to this incubation samples were always taken from stems lying on the screen, which meant that in the UV⁺ treatment the stems were exposed to about 88% of the PAR, from 13 to 27% of the UV-A, and <1% of the UV-B irradiance from June to late August. Due to increasing DOC concentrations, which rose sharply in late August (44), at the end of September the level of exposure to the UV-A had dropped to about 6%. While no underwater radiation data are available for October because the water levels had dropped so low that the spectrometer could not be used, it can be assumed that the stem pieces were exposed to less than 6% of the PAR and UV-A.

On sampling days six stem pieces of Scirpus, plus one stem piece for a killed control, were randomly selected from each chamber. Fungal biomass and production, as well as AFDM parameters, were determined as described above. The incubation tubes were placed in the pond water for incubation. A Campbell Scientific 107B thermistor and a CR10X data logger were used to measure and record water temperature. The transport, AFDM preparation, and ergosterol extraction procedures were the same as those employed for the naturally decaying stems.

Temperature experiments.

Two temperature experiments were conducted in the laboratory. For the first experiment, Pond 50 water and stems from the UV⁻ portion of the incubation chamber were brought to the laboratory. Five-milliliter portions of filtered (pore size, 0.2 μm) pond water, along with the stems, were placed in polycarbonate incubation vials to equilibrate for 24 h in Percival incubators set at 4, 10, 15, 21, and 30°C. After acclimatization the stems were incubated with 1 mM sodium [1-¹⁴C]acetate to measure fungal production. All incubations were done in triplicate and included one killed control.

For a longer-term temperature experiment two 20.8-liter glass aquaria (A-3002; Hagen) were used. Standing, dead Scirpus stems were harvested in November, cut into 2-cm pieces, and placed in the aquaria along with 10 liters of a fresh supply of filtered (pore size, 20 μm) Pond 50 water. The filter was used to remove large grazers but allowed the majority of the microbial population to pass through. One aquarium was maintained at 4°C, and the other was maintained at room temperature (21 ± 3°C). Air was constantly supplied to the aquaria with an air pump (A-799 Elite; Hagen). Both aquaria were agitated periodically to mix the contents, and additional filtered (pore size, 20 μm) pond water was added as required to replenish the system throughout the experiment. Both aquaria were kept in the dark to prevent algal growth. Fungal production and biomass for both treatments were measured as described above, except that preparations used to determine production were incubated in quadruplicate along with one killed control. The same quality control standards, except a glassware blank (as all previous blanks had been nondetectable), of one ergosterol spike (average recovery, 98.3% ± 2.4%) at every extraction date were included.

SigmaStat 2.03 (SPSS Inc.) was used to perform all statistical analyses, including analysis of variance. Parameter means ± standard errors are given below.

RESULTS

Isotope dilution.

The isotope dilution was 283-fold in the first experiment and 284-fold in the second experiment done 6 months later. To calculate instantaneous fungal production, we multiplied the 1-¹⁴C-labeled uptake by 284.

Fungal biomass and production.

The AFDM values for 2-cm stem pieces were similar (P > 0.06) over the course of the sampling season, with means of 0.11 ± 0.006 g for the submerged stem pieces and 0.098 ± 0.01 g for the aerial stem pieces.

On 10 May we conducted an initial survey of Pond 50 to assess the variations in fungal biomass on standing dead Scirpus litter. An average fungal biomass of 4.9 ± 0.8 mg of C g of AFDM⁻¹ was obtained for the tips of standing dead Scirpus with a moisture content of 13.7% ± 4.4%, and a biomass of 3.79 ± 0.60 mg C g of AFDM⁻¹ was obtained for stem pieces just above the water-air interface with a moisture content of 28% ± 5%. The biomass on Scirpus that had dislodged and fallen into the water was 2.41 ± 0.15 mg of C g of AFDM⁻¹. No acetate uptake experiments were performed at this time.

During the ice-free season the biomass on aerial sections of dead litter exhibited two peaks; the first peak was 1.5 mg of C g of AFDM⁻¹ and occurred in July (Julian day 194), and the second peak occurred in late September (Julian day 264) and was 1.3 mg of C g of AFDM⁻¹ (Fig. 1A). The fungal biomass values for the submerged stem portions were much higher and more sustained (Fig. 1A). For submerged litter a biomass of 2.2 mg of C g of AFDM⁻¹ was measured at the end of March (Julian day 144), and this was followed by a general decrease to 0.11 mg of C g of AFDM⁻¹ by 12 July. Then the biomass increased to a peak of 5.8 mg of C g of AFDM⁻¹ in August (Julian day 236) before decreasing to nondetectable levels in October. Surprisingly, the biomass increased again to 4.91 mg of C g of AFDM⁻¹ in November. The differences in fungal biomass in submerged and aerial Scirpus stems on the different sampling dates, however, were not significant (P > 0.05). The mean fungal biomass over the sampling time for the aerial portions was 0.69 ± 0.2 mg of C g of AFDM⁻¹, while for submerged portions it was about three times higher (1.95 ± 0.56 mg of C g of AFDM⁻¹). The mean biomass values for the aerial and submerged stem portions were statistically different (P < 0.04).

FIG. 1.

Seasonal changes in fungal biomass (A) and production (B) on Scirpus in aerial (▵) and submerged (•) parts of stems. The line with no symbols is the average air temperature. The error bars indicate standard errors.

Fungal production also varied over the season. The highest production value for aerial stem portions was measured in July (Julian day 194) and was 376 μg of C g of AFDM⁻¹ day⁻¹; later in the year the production rates were 253 μg of C g of AFDM⁻¹ day⁻¹ in late August (Julian day 236) and 180 μg of C g of AFDM⁻¹ day⁻¹ in October (Fig. 1B). The lowest production rate in aerial stems was 12.7 μg of C g of AFDM⁻¹ day⁻¹ and occurred in early October (Julian day 278). There was no significant difference between the peaks. The second and third peaks were significantly higher (P < 0.01) than the surrounding values, whereas the first peak was not significantly higher (P > 0.05) due to high replicate variation. The highest production values for submerged Scirpus stem portions were 372 μg of C g of AFDM⁻¹ day⁻¹ on 23 May (Julian day 144) and 275 μg of C g of AFDM⁻¹ day⁻¹ on 5 September (Julian day 249) (Fig. 1B), while the lowest production value was 1.8 μg of C g of AFDM⁻¹ day⁻¹ in July (Julian day 194). The differences in fungal production in submerged Scirpus stems on different sampling dates were not significant (P > 0.05). The potential mean production rates were 121.4 ± 41.1 μg of C g of AFDM day⁻¹ for the aerial stem portions and 91.8 ± 32.7 μg of C g of AFDM⁻¹ day⁻¹ for the submerged portions, and these values were not statistically different (P > 0.35). Water temperature was correlated with fungal production (r = 0.7, P < 0.005) for aerial stem pieces but not for submerged pieces. On an annual basis, from ice melt in the spring to freeze-up in the fall, the production rate on submerged stem portions was 28 mg of C g of AFDM⁻¹ year⁻¹, while the potential production rate on aerial stem portions was 36 mg of C g of AFDM⁻¹ year⁻¹.

Solar radiation experiment.

Fungal biomass and production rates were measured 24 h after live Scirpus was harvested. We found a fungal biomass of 0.16 mg of C g of AFDM⁻¹ and a surprisingly high production rate, 384 ± 202 μg of C g of AFDM⁻¹ day⁻¹. Twelve days later the biomass in the stems subjected to the UV⁺ treatment had increased to the seasonal high, 4.2 mg of C g of AFDM⁻¹, and this was followed by a slow decline to 0.36 mg of C g of AFDM⁻¹ by 1 November (Julian day 306) (Fig. 2A). Stems subjected to the UV⁻ treatment exhibited a similar trend but reached a peak biomass of 3.7 mg of C g of AFDM⁻¹ 28 days after the start of the experiment. Thereafter, there was a considerable drop in biomass to 0.55 mg of C g of AFDM⁻¹ in August (Julian day 222), followed by a slow increase to 2.7 mg of C g of AFDM⁻¹ in September before the final decline to 0.46 mg of C g of AFDM⁻¹ by 1 November. The mean fungal biomass for the UV⁺ treatment was 1.5 ± 0.4 mg of C g of AFDM⁻¹, while for the UV⁻ treatment it was 1.6 ± 0.4 mg of C g of AFDM⁻¹. There were no significant differences in biomass (P > 0.76) between the two treatments or within the treatments (P > 0.05) over the course of the experiment.

FIG. 2.

Seasonal changes in fungal biomass (A) and production (B) on Scirpus stems exposed to (•) or protected from (♦) UV radiation. The error bars indicate standard errors.

The production values for the UV⁺ stems ranged from a high of 413 ± 66.7 μg of C g of AFDM⁻¹ day⁻¹ on 23 August (Julian day 236) to a low of 15.1 ± 1.4 μg of C g of AFDM⁻¹ day⁻¹ on 20 September (Julian day 264) (Fig. 2B). The production data for the UV⁻ stems followed the same general rise and fall pattern as the data for the UV⁺ stems. The highest production value for the UV⁻ stems was 464 ± 75.5 μg of C g of AFDM⁻¹ day⁻¹ on 18 October (Julian day 292), and the lowest production value, which was obtained on 20 September (Julian day 264), was 12.4 ± 5.6 μg of C g of AFDM⁻¹ day⁻¹. Over the course of the experiment the average daily fungal production for the UV⁺ stems was 197 ± 49 μg of C g of AFDM⁻¹ day⁻¹, while for the UV⁻ stems it was 195 ± 53 μg of C g of AFDM⁻¹ day⁻¹. There was no statistical difference between the treatments (P > 0.96) or among sampling dates within the treatments (P > 0.05). There was a positive correlation between temperature and production in both the UV⁺ treatments (r² = 0.65, P < 0.05) and the UV⁻ treatments (r² = 0.67, P < 0.004).

At the beginning of the experiment the POC content was 40 mg g⁻¹ and the PON content was 0.78 mg g⁻¹. The values at the end of the experiment for the UV⁻ stems were 24.3 mg of POC g⁻¹ and 1.98 mg of PON g⁻¹; for the UV⁺ stems the POC content was 23.0 mg g⁻¹ and the PON content was 0.81 mg g⁻¹. The C/N ratio at the start of the experiment was 51:1. At the end of the experiment the C/N ratio for the UV⁻ stems was 12:1 and the C/N ratio for the UV⁺ stems was 28:1.

Water temperature.

In the short-term experiment there were no differences (P > 0.05) in the production rates at 4 and 10°C (Fig. 3) or at 15, 21, and 30°C. However, the production rates at 4 and 10°C were significantly lower (P > 0.05) than the production rates at 21 and 30°C, but they were not significantly lower than the production rates at 15°C (P > 0.05).

FIG. 3.

Plot of temperature versus fungal production in Scirpus stems. The error bars indicate standard errors.

In the 4°C treatment in the long-term temperature experiment, the starting fungal biomass, 0.22 mg of C g of AFDM⁻¹, initially dropped to 0.056 mg of C g of AFDM⁻¹, but then the biomass rose 10-fold to 0.52 mg of C g of AFDM⁻¹ by day 55 (Fig. 4A). After this there was a gradual decline and then a sudden increase to 1.56 mg of C g of AFDM⁻¹. The fungal production rate at the start of the experiment was 92 μg of C g of AFDM⁻¹ day⁻¹ (Fig. 4B). The production rate declined dramatically at first in the 4°C treatment and then ranged between about 10.0 and 20 μg of C g of AFDM⁻¹ day⁻¹ for rest of the experiment. On day 255, even though the biomass was 0.43 mg of C g of AFDM⁻¹, there was no measurable uptake of radiolabeled sodium acetate. Over the duration of the experiment the mean biomass was 0.43 ± 0.17 mg of C g of AFDM⁻¹, while the mean production rate was 13.3 ± 2.5 μg of C g of AFDM⁻¹ day⁻¹.

FIG. 4.

Fungal biomass (A) and production (B) on Scirpus stems over time at 4°C (♦) and 21°C (▾ and ▿). The error bars indicate standard errors.

In the 21°C treatment the biomass increased to 1.21 mg of C g of AFDM⁻¹ by day 35 before it decreased to 0.056 mg of C g of AFDM⁻¹ by day 55. For the rest of the experiment the biomass was stable near 2.0 mg of C g of AFDM⁻¹ (Fig. 4A). The production rates declined gradually from the start of the experiment to 9.92 μg of C g of AFDM⁻¹ day⁻¹ by day 55 (Fig. 4B) and then rose sharply to 269.4 μg of C g of AFDM⁻¹ day⁻¹ by day 134. The mean biomass was 1.46 ± 0.34 mg of C g of AFDM⁻¹, while the mean production rate was 76.5 ± 32.6 μg of C g of AFDM⁻¹ day⁻¹. The mean biomass (P < 0.008) and the mean production rate (P < 0.05) were significantly higher in the 21°C treatment than in the 4°C treatment.

DISCUSSION

Fungal biomass and production on standing dead Scirpus litter.

On the first sampling date, 10 May (average daily temperature, 8°C), we detected a high ergosterol content on portions of dead stems above and below the water and even at the tips of the shoots, where the moisture content was only 13%. This observation suggests that the process of plant breakdown begins very early in the spring in prairie wetlands despite the relatively cold temperatures. The highest fungal production values for the growing season for the submerged portions of Scirpus stems were obtained in late May. Unfortunately, no data for earlier dates closer to ice-out are available, and it is possible that the production rates were higher at these times. Due to limitations of the radiolabeled acetate method for measuring fungal production, our production rates for aerial portions of the stems can only be considered estimates of potential production rates. However, they do indicate that when these parts of the dead litter are wetted for as little as 3 to 4 h, as in our incubation, the fungal production rates can be similar to the fungal production rates for submerged stem portions. Intermittent snow, rain, fog, and dew would have provided sufficient moisture to sustain fungi (30) on these aerial stem portions. Kuehn et al. (23) observed an increase in fungal and bacterial activity in as little as 5 min after dry dead litter was wetted. The significance of such rapid increases in fungal production after wetting need to be investigated further in order to better understand the decomposition of dead macrophyte litter in prairie wetlands and other systems.

The conversion factors for fungal biomass and production which we used were obtained from the literature and were not specifically determined for the primary fungal species responsible for decomposition of Scirpus at Pond 50. This may have markedly affected our estimates of both production and biomass (see reference 16 for comparative conversion factors used by other workers). Although we do not know whether this led to over- or underestimates of fungal biomass and production in Pond 50, our observations corroborate those of other researchers (24, 34) because they showed that a significant amount of carbon is sequestered on dead Scirpus while it is in the standing position and before abscission and submergence in water.

Although we did not study fungal succession, it may partially account for the variations in fungal biomass and productivity observed on dead Scirpus stems at Pond 50. Suberkropp (39) noted that biomass accumulation and sporulation occur rapidly during the initial colonization of leaves in streams, and this phase is followed by a decrease in sporulation, as well as a decrease in the respiration rate, which is similar to what occurred in Scirpus stems in the spring. Fungi in the spring may degrade the more labile portions of bound plant carbon in Scirpus, and these fungi are later followed by fungi with specialized enzymes for digesting the more refractive stem constituents. Such a scenario would have contributed to the fungal production and biomass changes observed on decaying Scirpus at Pond 50 during the season.

Invertebrate grazing of fungi may also lead to a loss of fungal biomass and productivity and may have contributed to the variability of the results obtained for Pond 50. Previous researchers have shown the importance of preconditioned plant material as a nutrient source for organisms at higher trophic levels (5, 6). Not only would the fungi provide a more enriched source of nutrition than the plant material itself (28), but the fungi would also have partially predigested plant components that organisms at higher trophic levels would have been unable to digest on their own. Newell and Porter (34) have described significant decreases in fungal biomass due to mycophagy by periwinkle and molluscs. Although we observed no organisms that might have been fungivores on any of the Scirpus stems which we incubated, microscopic mycovores, such as amoebae and mites (28, 34), as well as nematodes (37), may have caused a reduction in fungal biomass during the sampling season.

Although other workers have found evidence that inorganic nutrient availability influences fungal production (30) during smooth cordgrass decomposition in salt marshes, this is unlikely to have been the case for Scirpus in Pond 50. Although we did not measure nutrient concentrations in the stems or water, Waiser (45) assessed possible nutrient limitations of bacterial and algal populations in the water and growing on Scirpus in Pond 50. She found no evidence of either P or N limitation. The PON results at the end of the solar radiation experiment also suggest that nitrogen is unlikely to be limiting, at least in the submerged portions of standing Scirpus stems.

Our values for fungal production in decomposing Scirpus in Pond 50 ranged from 1.8 to 376 μg of C g of AFDM⁻¹ day⁻¹, and the biomass ranged from nondetectable to 5.8 mg of C g of AFDM⁻¹. These values are well within the ranges of values that other researchers have reported, although comparisons may not be entirely valid due to large differences in the nature of the decomposing litter. Newell et al. (33) reported fungal biomass of 0.15 mg of C g of AFDM⁻¹ and production rates ranging from 116 to 665 μg of C g of AFDM⁻¹ day⁻¹ on standing dead leaf blades of Carex walteriana in a freshwater system. These values were later increased by 1.5-fold (30, 34). Kominkova et al. (22) measured fungal biomasses of approximately 10 to 80 mg of C g of AFDM⁻¹ and fungal production rates ranging from 72 to 1,224 μg of C g of AFDM⁻¹ day⁻¹ on submerged leaves of an emergent macrophyte in a hardwater Swiss lake. Kuehn et al. (24) measured fungal biomasses of 7 to 42 mg of C g of AFDM⁻¹ and production rates ranging from 73 to 2,836 μg of C g of AFDM⁻¹ day⁻¹ on decaying leaves of an emergent macrophyte in a freshwater wetland in Alabama. The biomass and production values obtained in our study are in the lower to middle range of the values obtained in the three studies noted above. This may be due in part to the fact that the Scirpus stem is a less labile source of carbon (15, 46) than leaves. In addition, the lower temperatures associated with northern prairie conditions probably also account for the production and biomass values that are lower than the values obtained for other systems.

Solar radiation.

Our experimental conditions were an attempt to mirror the conditions experienced by Scirpus litter in a typical prairie wetland and to measure the effects of solar radiation on aquatic fungal colonization. Dead litter from the previous season stands in a new season after the ice has melted before it falls into the water. Besides being shallow, the waters in prairie wetlands have very high concentrations of DOC that rapidly attenuate UV radiation (3). Therefore, fungal exposure to solar radiation on standing dead litter, and to UV radiation in particular, is limited to a very narrow band beneath the water surface. Once the litter falls into the water and sinks to the sediments, much of it is exposed only to PAR because of the rapid attenuation of UV. To mimic these events in our experiment, we used three distinct types of exposure. First, for about 2 weeks the Scirpus pieces were exposed to PAR (88%), UV-A (85%), and UV-B (76%) in the UV⁺ treatment. Second, after the stems sank to the screens at a depth of 4 cm (by day 194), the average depth of penetration of 1% of the UV-B (44), they were primarily exposed to UV-A (48 to 23%) plus PAR from mid-July to mid-September (days 195 to 270). Third, the UV⁻ and UV⁺ treatments were essentially the same, with only PAR exposure, after day 270 to the end of the experiment. Unlike the levels of exposure in other studies, in which fungi were exposed to the UV flux of almost full sunlight (2, 11), our levels of exposure were more similar to the levels of exposure experienced by most of the underwater fungal population except the fungi in the upper few centimeters of the water column.

During the incubation period there was no significant difference in fungal production or biomass between the UV⁺ and UV⁻ treatments. In the 2-week exposure to UV-B at the beginning of the experiment no impact on fungi was detected. During the next 75 days, when the stem pieces were exposed to only PAR plus UV-A, there was still no significant difference in either biomass or production between the two treatments, even though there seemed to be possible stimulation of fungal production in the UV⁺ treatment from days 236 to 249. Although much work has focused on UV-B, as it is the more energetic form of UV radiation, UV-A has been found to be as important as or more important than UV-B for microbial processes. Aas et al. (1) observed equal inhibition of incorporation of labeled thymidine by bacteria under UV-A and UV-B conditions. Sommaruga et al. (38) observed a 70% decrease in bacterial thymidine and leucine uptake in both freshwater and marine systems and cautiously attributed it entirely to UV-A radiation. The influence of UV-A radiation on fungi has been studied less than the influence of UV-A radiation on bacteria, but there are some indications that fungi, for the most part, are inured to this form of radiation. Denward et al. (10) observed no decrease in aquatic fungal biomass on aquatic macrophytes exposed to levels of PAR plus UV-A similar to those in our study. Moody et al. (26) observed that UV-A effects, as measured by mycelial extension in fungi grown on agar, were largely beneficial to terrestrial fungi. Newsham et al. (35), however, reported that a single fungal species disappeared with UV-A treatment and also that UV-A accelerated the rate of decomposition of leaves in terrestrial litter. These types of UV-A effects may have occurred during our experiments with Scirpus and may have helped mask any differences between the two treatments. This is an area for future research.

The lack of differences between the treatments may also have been due to several other factors. It is possible that aquatic fungi may differentially colonize plant material, with some fungi inhabiting the surface and others preferring to tunnel into decomposing material; this is similar to what occurs in terrestrial litter, in which some fungi inhabit the phylloplane of detritus and others burrow more deeply into the plant litter (26, 35). If this occurs with aquatic fungi, the UV radiation may have had little or no effect on fungi penetrating the stems because once inside, these fungi would have been shaded, while fungi on the surface could have been affected. Additionally, it is possible that if there was increased breakdown of the Scirpus stems in the UV⁺ treatment (11), it may have offset any inhibitory effects on the fungi. This seems unlikely, however, as the POC contents at the end of the experiment were the same in the two treatments.

Another possibility is that there was a change in species composition in the UV⁺ treatment. Moody et al. (26) showed that phylloplane species were largely unaffected by UV-B radiation, whereas other species were usually negatively affected. Similarly, Gehrke et al. (14) and Newsham et al. (35) observed differences in terrestrial fungal species composition between preparations exposed to levels of UV radiation that were 30% greater than ambient levels of UV radiation and preparations that were exposed to ambient solar radiation in decomposing terrestrial litter. Unfortunately, our biomass and production measurements for Pond 50 would not have detected a difference in species composition between the two treatments. UV radiation in the UV⁺ treatment may have led to death or debilitation of some fungi, while it may have favored other fungi that were more tolerant of the exposure conditions. Although the work of Newsham et al. (35) and Gehrke et al. (14) was done with terrestrial fungi, the same conclusions may apply to fungi in aquatic systems. Changes in seasonal patterns of fungal conidia and species have been studied (4, 13, 17) in lotic systems but not in shallow prairie wetlands.

The decreases in POC levels in both treatments during the experiment in the present study indicated that organic carbon from the decomposing Scirpus either leached into the water or was utilized by fungi and other microorganisms. The increase in the PON level in the UV⁻ treatment was probably associated with an increase in the number of microorganisms, possibly bacteria and algae, since the fungal biomass declined. Conversely, in the UV⁺ treatment the level of PON decreased, indicating that bacteria and algae may have been inhibited since the fungal biomass did not change. Therefore, the high C/N ratio measured in the UV⁺ treatment at the end of the experiment may have been due more to the harmful effects on microorganisms other than fungi. Newell and Porter (34) and Battle and Golladay (7) obtained similar results with decomposing cordgrass and leaf litter, respectively, and attributed the lower C/N ratios to higher levels of available nitrogen and more robust microbial activity. Denward and Tranvik (11) observed no change in the C/N ratio for decomposing aquatic macrophyte litter placed in buckets of tap water and subjected to treatments that blocked UV-B or both UV-A and UV-B, possibly because their experiment was performed for only 60 days. Additionally, the increased leaching of organic material into water would have increased the DOC concentration, leading to greater UV attenuation and decreasing the amount of radiation reaching the microorganisms over time.

Water temperature.

It is surprising that temperature as a regulating factor in aquatic fungal dynamics has been largely ignored. We found that water temperature was a strong variable associated with the seasonal changes in fungal biomass and production. Suberkropp and Weyers (40) observed an exponential increase in fungal carbon production as the temperature increased from 10 to 25°C on yellow poplar leaves. Chauvet and Suberkropp (8) assessed the influence of temperature (15, 20, and 25°C) on the sporulation of eight aquatic hyphomycetes. The optimal temperature for sporulation varied for five species, two species produced similar amounts of conidia at 20 and 25°C, and one species sporulated equally at all three temperatures.

There was a positive correlation between temperature and fungal production in the solar radiation experiments, as well as with the aerial stem portions in the seasonal study. The results of our short-term experiment indicated only that fungal production was lower at ≤10°C than it was >15°C. However, in the long-term study there was a significant difference in both biomass and production, both of which were higher at 21°C than at 4°C. These results support the hypothesis that there is a correlation between temperature and production as measured in the field and indicate that in northern prairie wetlands fungal colonization and production on standing dead Scirpus litter are sensitive to temperature fluctuations. However, further research is required to demonstrate the relative importance of water temperature compared to other factors, such as grazing and the quality of the substrate.