Effects of light on sediment nutrient flux and water column nutrient stoichiometry in a shallow lake

ARTICLE IN PRESS WAT E R R E S E A R C H

42 (2008) 977 – 986

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Effects of light on sediment nutrient flux and water column nutrient stoichiometry in a shallow lake Bryan M. Spearsa,c,, Laurence Carvalhoa, Rupert Perkinsb, David M. Patersonc a

Centre for Ecology and Hydrology Edinburgh, Penicuik, Midlothian EH26 0QB, Scotland, UK School of Earth, Ocean and Planetary Sciences, Cardiff University, Main Building, Park Place, Cardiff CF10 3YE, Wales, UK c Sediment Ecology Research Group, Gatty Marine Laboratory, University of St. Andrews, Fife KY16 8LB, Scotland, UK b

art i cle info

ab st rac t

Article history:

The effects of light and temperature on nutrient cycling (silica (Si), nitrogen (N) and

Received 2 July 2007

phosphorus (P)) between sediments and water in a shallow eutrophic lake (Loch Leven,

Received in revised form

Scotland), and consequent effects on water column nutrient stoichiometry, were assessed

14 September 2007

using a series of intact sediment core incubation experiments. Estimates of actual seasonal

Accepted 16 September 2007

dark and light P-fluxes were assessed using 24-h incubations. Sediment-P uptake was

Available online 21 September 2007

observed in spring (7 1C) and release in autumn (12 1C) and summer (17 1C), with the highest

Keywords: Sediment Phosphorus Nitrogen Silica

release rates (17 mg PO4-P m2 sediment surface area d1) occurring in summer. In a longer (21-day) experiment in which the effects of light (light (n ¼ 6) and dark (n ¼ 6)) and temperature (five 4-day cycles to represent: 7 1C ) 13 1C ) 23 1C ) 13 1C ) 7 1C) on water column nutrient concentrations were assessed, PO4--P, total P (TP), SiO2 and total silica (TSi) concentrations in the water column were all significantly higher under dark conditions (ANOVA, a ¼ 0.05). NH4-N (ammonium N) water column concentrations were observed to be

Flux

higher under dark conditions at low temperatures and higher under light conditions

Lake

following a high-temperature (23 1C) treatment. No significant light effects were observed

Light Water column

for water column total N (TN) concentration. Flux estimates for all nutrients measured are given. In terms of water column nutrient stoichiometry, TN:TP ratio was significantly higher under light conditions, TSi:TN was significantly lower under light conditions, and TSi:TP did not vary significantly between the dark and light treatments. The main processes acting to regulate diffusive nutrient release appeared to be photosynthetic elevation of bottom water pH and dissolved oxygen concentration (both significantly higher under light conditions) and direct microalgal sequestration. Thus, a feedback mechanism exists in recovering shallow lakes where benthic microalgae can affect the stoichiometry (to favour P/Si limitation) of the plankton, and also of the main source of nutrients back to the sediments via the disproportionate regulation of sediment P, Si and N release. & 2007 Elsevier Ltd. All rights reserved.

1.

Introduction

Benthic microalgae exist at the interface between the sediment and the water column and, as such, are well placed to regulate benthic/pelagic nutrient cycling. Their ability to

perform this ecosystem function may be affected by variations in light and temperature. This includes a range of processes linked to benthic microalgae such as the replenishment of organic nutrient compounds to the sediment (Spears et al., 2007b), direct uptake of nutrients (nitrogen

Corresponding author. Centre for Ecology and Hydrology Edinburgh, Penicuik, Midlothian EH26 0QB, Scotland, UK. Tel.: +44 131 4458536.

E-mail address: [email protected] (B.M. Spears). 0043-1354/$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2007.09.012

ARTICLE IN PRESS 978

WA T E R R E S E A R C H

(N)-uptake: Axler and Reuter, 1996; Webster et al., 2003; phosphorus (P)-uptake: Barlow-Bush et al., 2006; DeNicola et al., 2006; silica (Si)-uptake: Del Amo and Brzezinski, 1999; Penna et al., 2003), oxygenation of the colonisation zone and, therefore, reduction in the release of those nutrient pools that are redox-sensitive (Woodruff et al., 1999), and biostabilisation of the sediment surface via either alteration of the surface structure through mat formation (Dodds, 2003) or by the enhanced cohesion and adhesion of sediment particles caused by the extrusion of extra-cellular polymers (Paterson, 1989; Spears et al., 2007c). Although benthic microalgae have been observed to reduce P, N and Si release from sediments (Jansson, 1980; Kelderman et al., 1988; Jarvie et al., 2002), it is unclear whether this regulation is constant over a diurnal cycle and over a temperature range and whether nutrient-specific regulation occurs resulting in alterations in seasonality of water column stoichiometry. Additionally, no study has considered the effects of these processes on the cycling of these three plant nutrients together with respect to shifts in water column stoichiometry in relation to the Redfield ratio (i.e. 16 Si:16 N:1P; Redfield et al., 1963). This may be particularly important in shaping phytoplankton communities in eutrophic shallow lakes where nutritional sources have been altered (i.e. from the catchment to the sediment) following a reduction in external nutrient load (Sas, 1989). Recent long-term studies have highlighted the seasonality of benthic/pelagic nutrient cycling and water column stoichiometry in recovering shallow lakes (Søndergaard et al., 2005; Jeppesen et al., 2005). The characteristic seasonal dynamics in water column nutrient concentrations observed in Loch Leven are shown (Fig. 1). The contrasting seasonality of important plant nutrients can drive phytoplankton community succession where short-term Si- and N-limitation events are observed in spring and late summer, respectively, and P-limitation is common during all other periods. The occurrence of N-limitation in late summer is driven by highmagnitude sediment-P and Si release (main source to recovering lakes) coupled with denitrification of NO3-N (Søndergaard et al., 2005). Although sediment disturbance

14 12

2.0

10

1.5 SiO2 1.0

8 6

0.5 PO4-P Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

0.0

4

6 5 4 3 2 1 0

SiO2 concentration (mg L-1)

NO3-N

PO4-P concentration (µg L-1)

NO3-N concentration (mg L-1)

2.5

2

Fig. 1 – Average monthly water column concentrations of PO4-P, NO3-N, and SiO2 during the recovery period in Loch Leven. Years included in the average values are 1997, 1998, 1999 and 2001.

42 (2008) 977– 986

(e.g. bioturbation and wind mixing; Welch and Cooke, 1995) can play an important role in the cycling of nutrients between sediments and the water column, the processes described above are predominantly diffusive in nature and are regulated by temperature, sediment-P release occurring on the onset of anoxia (high temperature) resulting from a switch in sediment energy-production from autotrophy to heterotrophy (Bostro¨m et al., 1988), and denitrification from an increase (with temperature) in denitrifying bacterial production (Risgaard-Petersen et al., 1994). Si is released from the sediment in late summer following the dissolution of planktonic diatom frustules (also increases with temperature) that accumulate during spring and summer deposition (Bailey-Watts, 1976a, b; Gibson et al., 2000), a process that has also been linked with bacterial mediation (Bidle et al., 2003). N is mainly released from sediment as NH4-N, which is efficiently converted to NO3-N via nitrification under oxygenated conditions (Van Luijn et al., 1995; Beutel, 2006). All of these processes are expected to be altered under conditions that favour benthic microalgal photosynthesis (i.e. sufficient light conditions); P release should be reduced via the photosynthetic elevation of dissolved oxygen (DO) concentration and moderately elevated pH conditions (ion exchange processes can result in sediment-P release under high pH conditions; Bostro¨m et al., 1988) in the surface sediment and direct algal uptake; Si release should be reduced via direct algal uptake; and NH4-N release should be reduced via algal uptake and increased nitrification via the photosynthetic elevation of surface sediment DO concentrations. The effects of light and temperature on benthic/pelagic nutrient cycling were tested using a series of laboratory-based intact sediment-core incubation experiments. Two approaches were taken: (1) seasonal estimates of ambient daily light and dark P-flux rates were taken on three separate dates using 24-h incubations conducted at the temperature recorded at the time of sampling and (2) estimates of daily water column nutrient (PO4-P, NH4-N and SiO2) concentrations and stoichiometry (total silica (TSi):total N (TN):total P (TP)) were tracked in intact sediment cores over a 21-day incubation period for both light and dark treatments in which temperatures were manipulated to mimic seasonal variations.

2.

Methods

2.1.

Study site

Loch Leven (Fig. 2) is a shallow lake situated in the southeast of Scotland (latitude 561100 N, longitude 31300 W) with a surface area of 13.3 km2 and a mean depth of 3.9 m. The lake has a well-documented history of eutrophication and, in an attempt to improve water quality, external TP load was reduced from about 4.05 mg TP m2 lake surface area d1 in 1985 to 1.62 mg TP m2 lake surface area d1 in 1995 (BaileyWatts and Kirika, 1999). Although improvements in water quality conditions have recently (i.e. post 2002) been observed, significant release-sensitive sediment P pools, and associated summer internal loading events, are still common in the lake (Spears et al., 2006, 2007b). At the time of this

ARTICLE IN PRESS WAT E R R E S E A R C H

979

42 (2008) 977 – 986

Fig. 2 – Bathymetric map of Loch Leven including the location of the sample point (cross). Table 1 – Summary of field observations taken on each sampling date Date

23 August 2005 2 November 2005 18 April 2006

Inc. temp. (1C)

Temp. (1C)

Secchi (m)

Cond. (mS cm)

pH

DO (mg L1)

17.0 12.0 7.0

17.0 10.2 7.6

1.0 1.0 1.1

224 220 232

8.5 8.5 8.3

9.9 10.5 11.1

Inc. temp. ¼ incubation temperature, Temp. ¼ water column temperature at the time of sampling, Secchi ¼ Secchi depth, cond. ¼ water column conductivity, DO ¼ water column dissolved oxygen concentration.

study, therefore, Loch Leven represented a shallow lake under recovery from eutrophication.

2.2.

Sample collection

All sediment cores were collected from the same sampling point (Fig. 2) corresponding to the mean depth (3.9 m) of the lake. Cores were collected from a boat using a Jenkin surfacesediment sampler that facilitated the collection of an undisturbed sediment–water interface (core internal diameter 6.7 cm and height 50 cm) consisting of about 20 cm sediment and 30 cm overlying water column. Cores were transported to the laboratory within 3 h of collection. All incubations were conducted within an environmental chamber in which light and temperature were regulated.

2.3.

Ambient light/dark P-flux incubations

Incubations were designed to estimate the daily dark/light Pflux rates under ambient lake conditions. Twelve cores were collected on three dates (23 August 2005, 2 November 2005 and 18 April 2006). Surface water column temperature, DO concentration, conductivity and pH were measured on each date using an underwater sensor (Hydrolab Water Quality

Monitoring System, Hydrolab Corporation, Colorado, USA) and are summarised in Table 1. Cores were transported to the laboratory and allowed to acclimatise for 1 h in an incubator at temperatures similar to those observed in the lake at the time of sampling (7 1C on 18 April 2004, 12 1C on 2 November 2005 and 17 1C on 23 August 2005). On each date, six cores were left in the light (15 mmol s1 m2 between 8 a.m. and 4 p.m.) and six were darkened by covering with aluminium foil and black polyethylene bags. The light levels were chosen to mimic those observed at about 3.5 m in Loch Leven. Cores were continually bubbled (commenced at the beginning of the incubation period) with air using a pump (Tetratec Whisper AP 200, TetraWerke, Melle, Germany) producing approximately 60 bubbles/min1 with a tubing height of 23 cm above the sediment surface. Water (15 ml) overlying the sediment surface (sampled from 1 cm above the sediment surface) was collected at 3:30 p.m. on the day of sampling using a syringe immediately after the acclimatisation period and at the end of a 24-h incubation. The differences in the concentrations were corrected for sediment surface area and dark and light treatment fluxes expressed as mg PO4-P m2 sediment surface area d1. Only PO4-P flux was considered in the ambient P-flux experiments and P analysis was conducted according to Wetzel and Likens (2000).

ARTICLE IN PRESS

2.4.

WA T E R R E S E A R C H

Light manipulation experiment

Cores were collected on 18 April 2006. Again six cores were left uncovered and six were incubated in the dark. The light levels and cycle period and the bubbling rate were identical to the conditions described for the ambient P-flux experiments. However, temperatures were manipulated at 4 p.m. on day 1 (initial 7 1C), day 5 (initial 13 1C), day 9 (23 1C), day 13 (final 13 1C) and day 17 (final 7 1C) in this experiment. The temperatures were maintained for 4 days following each change. Daily water samples (20 ml) were collected (3:30 p.m.) from 1 cm above the sediment surface (and replaced to surface water with 20 ml distilled water) using a syringe. TP and PO4-P analyses were conducted, using persulphate digestion for the former, following the acid–molybdenumblue colorimetric method (Wetzel and Likens, 2000). NH4-N was analysed using the phenol–hypochlorite method (Wetzel and Likens, 2000). TN was analysed using ion chromatography, following a persulphate digestion. TSi (following a Na2CO3 digestion) and silicate (SiO2) concentrations were analysed using the acid–ammonium–molybdate method (Golterman et al., 1978). Nutrient flux estimates were made using the difference in water column concentrations over each 24-h period and were standardised to sediment surface area and overlying water column volume. pH was measured at the end of the incubation experiment (using an HI 99121 pH meter equipped with HI 1292D soil electrode; Hanna Instruments Limited, Leighton Buzzard, Bedfordshire, UK) at 1 cm above the sediment surface and 0.5, 1.5 and 8 cm below the sediment surface. DO concentrations were measured at about 0.5 cm above the sediment surface at the end of the incubation experiment using a Hach Luminescent Dissolved Oxygen probe (Model HQ10; Hach+Lang Europe, Dusseldorf, Germany). As in all experimental models, intact core incubation systems have inherent problems that complicate direct comparisons with true lake processes. These include sufficient mixing of the overlying water column and potential for disturbance of the sediments. Visual inspection of the cores indicated no disturbance of the sediment surface at the stated bubbling rate and chemical analysis of the water column in a pilot study showed efficient mixing of the overlying water under these conditions. Closed system enclosure effects can occur in the absence of a natural flux of organic matter to the sediment and removal of nutrients from the overlying water column. However, this will have larger implications for longterm incubation studies (e.g. 21-day experiment) and is less important in short-term studies (e.g. 24-h flux estimates). As such, the long-term experiments reported here may be used to assess specific uptake/release mechanisms but may not accurately represent the variations that occurred in the lake during the same period.

2.5.

Statistical analyses

The overall effects of temperature and light in the seasonal ambient P-flux incubations were assessed using two-way analysis of variance (ANOVA; total degrees of freedom (TDF) ¼ 35, a ¼ 0.05).

42 (2008) 977– 986

In the light manipulation experiment, repeated measures one-way ANOVAs (TDF ¼ 251, a ¼ 0.05) were used to assess the overall effects of light on the water column concentrations of NH4-N, PO4-P, SiO2, TP, TSi and TN as well as ratios of TN:TP, TSi:TP and TSi:TN. It was impossible to statistically differentiate between the effects of temperature and incubation time in the 21-day incubation experiments. As a result, visual inspection of the data was used to identify gradual ‘‘incubation effects’’ from more distinct ‘‘threshold’’ temperature effects following temperature shifts. The effects of light on variations in pH and water column DO concentrations at the end of the incubation experiment were assessed using one-way ANOVA.

3.

Results

3.1.

Ambient light/dark P flux incubations

PO4-P flux increased significantly with temperature (po0.01) but not with light (Fig. 3). Sediment uptake of PO4-P was observed at 7 1C with release occurring at 12 and 17 1C.

3.2. Light manipulation experiment: water column nutrient concentrations The results of the repeated measures one-way ANOVA analysis used to determine the significance of the light and dark treatments are summarised (Table 2). The ranges of nutrient flux estimates over the 21-day incubation period are reported (Table 3). Light and dark water column PO4-P (Fig. 4a) and TP (Fig. 4b) concentrations showed similar trends throughout the incubation period remaining low until day 11, following which a sharp increase was observed which coincided with an increase from 13 to 23 1C to suggest more rapid temperature effects. The trends differed from day 17, with a decrease in TP concentration in contrast to the continued increase in PO4-P concentration, during the same period. Concentrations of both were significantly higher under dark conditions than light (Table 2). NH4-N concentrations also differed significantly

Sediment nutrient flux (mg PO4 - P m-2 d-1)

980

30 25 20 15 10 5 0 -5

Light

Dark 7°C

Light

Dark Light Dark 12°C 17°C Temperature and light treatment

Fig. 3 – Light/dark PO4-P flux under ambient lake conditions. Error bars represent standard error of the mean (n ¼ 6).

ARTICLE IN PRESS WAT E R R E S E A R C H

between the light treatments, being higher under dark conditions prior to day 10. Concentrations under light conditions gradually increased from day 7, with a sharper increase on days 12 and 13. However, NH4-N concentrations decreased from day 12 until day 16, and appeared to be triggered by the temperature change. TN decreased between days 1 and 8 and did not vary significantly between light and dark treatments with no sharp temperature effects observed. SiO2 and TSi concentrations increased gradually throughout the incubation. Concentrations were significantly higher under dark compared with light conditions over the 21 days.

3.3. Light manipulation experiment: water column nutrient stoichiometry TN:TP ratio was significantly higher under light conditions (Table 2). TSi:TP did not vary significantly under different light treatments. TSi:TN was significantly higher under dark conditions.

3.4. Light manipulation experiment: pH and dissolved oxygen profiles The DO concentration was significantly higher under light conditions at the end of the incubation period (dark

Table 2 – Results of the one-way repeated measures ANOVA (a ¼ 0.05) analysis used to assess the significance of the difference between water column nutrient concentrations (NH4-N (lg L1), TN (mg L1), PO4-P (lg L1), TP (lg L1), TSi (mg L1), SiO2 (mg L1)) and stoichiometry (TN:TP, TSi:TP and TSi:TN) under dark and light treatments Mean light PO4-P TP NH4-N TN SiO2 TSi TN:TP TSi:TP TSi:TN

24 47 92 0.60 5.6 6.2 42 78 3.0

(24) (25) (63) (0.17) (3.8) (3.3) (25) (32) (1.8)

Mean dark 41 67 62 0.57 7.9 8.4 30 81 4.2

(40) (42) (36) (0.18) (5.1) (4.5) (19) (43) (2.7)

981

42 (2008) 977 – 986

p-Value 0.000 0.000 0.021 0.230 0.000 0.000 0.000 0.613 0.000

Standard deviation of the mean (n ¼ 21) is given in parenthesis.

DO ¼ 9.25 mg L1; light DO ¼ 9.85 mg L1; ANOVA, a ¼ 0.05, n ¼ 6). pH conditions (Fig. 5) were significantly higher under light conditions at 1 cm above and 0.5 cm below the sediment surface but no difference was observed at 1.5 and 8 cm below the sediment surface (ANOVA; a ¼ 0.05).

4.

Discussion

4.1.

Ambient P-flux estimates

Estimates of P-flux rates were in agreement with a number of recent studies in which sediment-P release has been observed in the (warmer) summer, and sediment-P uptake observed in the (cooler) winter months (e.g. Søndergaard et al., 2005; Spears et al., 2007a). Other laboratory studies have found that P release across an oxygenated sediment–water interface becomes independent of DO concentrations at temperatures between 17 (Kamp-Nielson, 1975) and 21 1C (Holdren and Armstrong, 1980). This would suggest that sediment oxygen demand exceeds sediment oxygen production at these temperatures, creating zones of anoxia under which redoxsensitive P complexes can be reduced and PO4-P released from the sediment. However, no significant difference was observed between light and dark ambient P-fluxes. This can be explained by (1) low initial benthic algal biomass or (2) the fact that 24 h is insufficient time for DO shifts to occur. The low secchi depth (1 m on each date) observed on each of the three sampling dates indicated low light conditions at the sediment surface (3.5 m below the water surface) and, therefore, it is likely there was a low initial biomass of benthic microalgae (not measured in this study). In a previous study, the highest epipelic biomass was observed to occur in December/January (i.e. during clearer water periods) in Loch Leven. The measured release rates (Table 3) are comparatively low for eutrophic sediments with similar estimates of ambient P release ranging from 10 to 278 mg P m2 sediment surface area d1 (Penn et al., 2000; Phillips et al., 1994, respectively) compared with the light P-flux at 17 1C in Loch Leven, using the same incubation techniques, estimated at 24 mg P m2 sediment surface area d1 (Spears et al., 2007b). Sediment-P uptake has been observed in other studies and is generally higher under aerated conditions (e.g. 0.22 mg m2 d1 in Lake Cataouatche; Miao et al., 2006). The observed P uptake into the sediments and the comparatively low release rates, even under elevated temperatures, suggest that internal nutrient cycling within Loch Leven has been reduced in

Table 3 – The range of sediment flux estimates measured during the 21-day light manipulation incubation experiment for total phosphorous (TP), PO4-P, total nitrogen (TN), ammonium nitrogen (NH4-N), total silica (TSi) and silicate (SiO2)

Light Dark

TP (mg m2 d1)

PO4-P (mg m2 d1)

TN (mg m2 d1)

NH4-N (mg m2 d1)

TSi (mg m2 d1)

SiO2 (mg m2 d1)

17 to +19 8 to +8

2 to +11 1 to +4

59 to +42 34 to +52

8 to +7 25 to +22

189 to +887 260 to +842

120 to +1126 101 to +777

Negative values indicate sediment uptake of nutrients from the water column and positive values represent sediment release to the water column. All flux estimates are expressed as mg m2 sediment surface area day1.

ARTICLE IN PRESS 982

WA T E R R E S E A R C H

PO4-P 7°C

PO4-P concentration (µg L-1)

120

42 (2008) 977– 986

13°C

100

23°C

13°C

7°C

Dark Light

80 60 40 20 0 1

TP concentration (µg L-1)

250

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21

TP 7°C

200

13°C

23°C

13°C

7°C

150 100 50 0

NH4-N concentration (µg L-1)

1 350 300 250 200 150 100 50 0 1

3

4

5

6

2

3

7

8

13°C

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 23°C

13°C

7°C

9 10 11 12 13 14 15 16 17 18 19 20 21

TN

1.2

TN concentration (mg L-1)

2

NH4-N 7°C

7°C

13°C

23°C

13°C

7°C

1.0 0.8 0.6 0.4 0.2 0.0

SiO2 concentration (mg L-1)

1

3

4

5

6

SiO2 7°C

16 14 12 10 8 6 4 2 0

1

TSi concentration (mg L-1)

2

2 TSi

16 14 12 10 8 6 4 2 0

1

2

3

8

13°C

4

5

6

7°C

3

7

7

8

13°C

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 Incubation day 23°C

13°C

7°C

9 10 11 12 13 14 15 16 17 18 19 20 21 23°C

13°C

7°C

9 10 11 12 13 14 15 16 17 18 19 20 21 Incubation day

Fig. 4 – Variation in water column nutrient concentrations (PO4-P (a), TP (b), NH4-N (c), TN (d), SiO2 (e) and TSi (f)) during the 21-day incubation experiment.

ARTICLE IN PRESS WAT E R R E S E A R C H

Distance from sediment surface (cm)

2 1

pH

0 -1

7

7.2

7.4

7.6

7.8

8

8.2

-2 -3 -4

Dark Light

-5 -6 -7 -8 -9

Fig. 5 – Variation in pH across the sediment–water interface following the long-term incubation experiment.

recent years as a result of water quality management practices.

4.2. Light manipulation experiment: effects of light on benthic/pelagic nutrient cycling Light and dark NH4-N water column concentrations showed conflicting trends throughout the incubation experiment where concentrations under light conditions were lowest (relative to dark conditions) between days 1 and 8, and concentrations under dark conditions were lowest following day 12. The former observation is in agreement with our original hypothesis where increased nitrification (i.e. lower NH4-N concentrations) was expected under light conditions. The higher concentrations under dark conditions are in agreement with a number of other studies in which NH4-N release has been reduced by benthic microalgal production either by direct uptake or increased nitrification through the photosynthetic elevation of DO levels (Jansson, 1980; Risgaard-Petersen et al., 1994; Axler and Reuter, 1996). The shift in apparent impacts of light on NH4-N fluxes occurs from day 9 when the temperature was increased from 13 to 23 1C, suggesting that at this point NH4-N release increased to the extent of masking any ‘‘dampening’’ effects of benthic algae. Apart from the sharp change in NH4-N concentrations under light conditions during days 12–14, all other concentration changes were gradual and it is, therefore, impossible to separate temperature effects from incubation effects. TN and NH4-N variation was not synchronous throughout the incubation period presumably as a result of other nitrification/ denitrification processes involving NO3-N and N2 gas. The observed NH4-N release rates are in agreement with a number of other studies (reviewed by Beutel (2006)) in which flux rates are routinely observed to be lower (or negative; Zhang et al., 2004) under well-aerated conditions. The maximum release values were higher under dark (comparable to eutrophic release conditions) compared with light conditions (comparable to mesotrophic release conditions) (Beutel, 2006) and were generally lower than other highly productive lakes (e.g.

42 (2008) 977 – 986

983

52–71 mg NH4-N m2 d1 in Lake Vechten, Netherlands (Sweerts et al., 1991)). SiO2 concentrations increased steadily up to day 9. The rate of release increased markedly during days 10–13, particularly under dark conditions and coincided with a temperature change from 13 to 23 1C. The lack of change following day 13 may represent the onset of equilibrium between the sediment and the water column in combination with limited sequestration rates of Si by benthic algae. TSi and SiO2 variation was very similar, indicating strong dependence of TSi on sediment-derived SiO2. Flux estimates were in agreement with estimates reported in other studies (e.g. 0.1–0.8 g SiO2 m2 d1 in Loch Leven (Bailey-Watts, 1976b) and 0.1 g SiO2 m2 d1 in Lake Grevelingen, Netherlands (Kelderman et al., 1988)). Water column concentrations were consistently higher under dark conditions, indicating strong regulation of Si release by benthic microalgae. This is likely due to direct uptake. The regulation of sediment Si release by benthic microalgae has also been observed by Kelderman et al. (1988) and, although not observed in this study, sediment-Si release has been reported to increase under high temperatures (i.e. highest release in summer; Bailey-Watts, 1976a, b; Rippey, 1983; Gibson et al., 2000). The pH and DO profiles measured at the end of the 21-day incubation experiment substantiated the hypothesis that actively photosynthesising benthic algae alter the biogeochemical dynamics of sediment N and P cycling through the maintenance of elevated DO and/or pH conditions. Few studies have directly assessed this phenomenon in relation to nutrient fluxes, although the profiles observed in this study are in agreement with those observed in others. Carlton and Wetzel (1988) observed elevated pH conditions in the sediment surface under light conditions with significant decreases in pH (about 1 pH unit) observed down through the sediment. The removal of P from the water column in Ca–P complexes is enhanced under moderate pH conditions (i.e. above pH 6.5–7.0: Wetzel, 2001; Otsuki and Wetzel, 1972). Therefore, under the conditions observed at the end of the incubation experiments, one may expect release of P from Ca–P complexes in the deeper sediment layers and precipitation of Ca–P in the water column and surface sediments (Ca–P makes up 9.5–20.9% sediment TP in Loch Leven; Spears et al., 2007b). However, the observed difference of 0.1 pH unit in this study, between dark and light conditions, is lower than the 1 pH unit difference observed in other studies (Dodds, 2003). DO concentrations were also higher under light (9.85 mg L1) when compared with dark (9.25 mg L1) conditions. Again, this difference is low in comparison to other studies, for example supersaturated DO concentrations (4150% for Phormidium/Cymbella and Ulothrixdominated assemblages) have been observed in the surface layers (upper 2 mm) of dense algal mats (Dodds, 2003), probably as a result of the low (but nevertheless realistic) light levels used in the light manipulation experiments in comparison to these other studies. The elevated DO concentrations under light conditions may indicate a decrease in the release of P from sediment Fe–P and Mn–P complexes (reductant-soluble P makes up 23.2–39.5% sediment TP in Loch Leven; Spears et al., 2007b) to the overlying water column.

ARTICLE IN PRESS 984

WA T E R R E S E A R C H

Although these experiments were aerated, the significantly higher DO concentrations observed under light conditions reflect the elevated DO production by benthic primary producers under illuminated conditions. These observations strengthen the hypothesis that benthic autotrophic production can limit P and N release through the maintenance of elevated DO and pH conditions. Under such conditions the likelihood of Ca–P and Fe–P release is reduced (Bostro¨m et al., 1988) and the reduction of NH4-N by microbial nitrification is facilitated (Risgaard-Petersen et al., 1994). However, direct uptake of nutrients by benthic microalgae must also be considered and is likely to be a key process affecting SiO2 release from the sediment. Additionally, in situ processes may not always be so homogeneous, and so the actual ‘‘in-lake’’ effects may be greater than those observed in these experiments.

4.3. Light manipulation experiment: effects of light on water column nutrient stoichiometry The TN:TP ratio was highest, and TSi:TN ratio lowest, under light conditions. Thus, dependent on absolute concentrations, water column N-limitation may become less likely under light conditions. The main factors regulating this variation are likely to be driven by benthic photosynthesis and include elevated DO concentrations and pH conditions at the sediment surface under light conditions regulating P release more than N release, as discussed above. However, benthic algal production of DO has also been linked with loss of N from the system via elevated rates of coupled nitrification and denitrification (Van Luijn et al., 1995). Few other studies have assessed the effects of sediment nutrient cycling on water column nutrient stoichiometry with the notable exception of Levine and Schindler (1992) in which the TN:TP ratio was observed to decrease (i.e. alleviation of N-limitation) as a result of higher N than P release from sediments of characteristically oligotrophic lakes. This discrepancy highlights the importance of past nutrient loading (and, therefore, trophic status) to a lake with respect to the internal cycling of nutrients. Loch Leven has been impacted by elevated external nutrient loading (both N and P) for over 125 years (estimated using palaeolimnological techniques; unpublished data) and experiences large sediment-P release events as a result of high sediment-bound P stock (Spears et al., 2007b). In comparison, sediment-P uptake was observed in the lakes included in the Levine and Schindler (1992) study throughout the year, probably as a result of low external nutrient loading. Thus, the regulation of sediment-P release by benthic algae in oligotrophic lakes (i.e. low/no sediment nutrient efflux) may be dominated by direct uptake, whereas in situations where algal nutrient requirements are saturated (i.e. eutrophic lake sediments; high sediment nutrient efflux) indirect regulatory mechanisms (e.g. photosynthetic production of DO) may become more important. Additionally, indirect effects will be more important in sediments that experience conditions of anoxia.

4.4.

Wider implications of the study

The sediment nutrient release scenario presented in this paper is representative of a shallow lake recovering from eutrophication. Under other circumstances (i.e. different

42 (2008) 977– 986

nutrient loading history) the effects of temperature and light may be altered, as indicated by the comparison of this study with that of Levine and Schindler (1992). However, the observation of disproportionate regulation of sediment nutrient cycling highlights an interesting feedback mechanism by which benthic microalgae may alter the flux of nutrients to the sediment, therefore altering the stoichiometry of the organic matter reaching the benthos (i.e. bacterial substrate) that, in shallow lake systems such as Loch Leven, ultimately fuels the ecology of the plankton. Eutrophication is a very widespread problem and the need for understanding the processes that could enhance or speed up restoration has never been greater, particularly given European legislation (European Commission, Water Framework Directive) that requires waters to achieve ‘‘good’’ ecological status by 2015. This study provides valuable information on the role of the sediment community in influencing the recovery period. Improved water clarity (and resultant enhancement of sediment light regime) will be a positive feedback on recovery and, in general, N limitation should be reduced as water quality improves.

5.

Conclusions

Ambient P-flux estimates showed sediment P uptake in spring and release in winter and summer (greatest release flux), which is in agreement with the characteristic seasonal P signal in Loch Leven. No significant differences were observed between dark and light ambient P fluxes, probably due to a combination of low initial benthic algal biomass and high light attenuation on the days of sampling. However, a significant decrease in Si and P release was observed under light conditions in the longer incubation experiments most probably as a result of photosynthetic maintenance of DO and pH conditions at the sediment surface (P) and direct uptake by algae (Si and P), altering the transport of nutrients across the sediment–water interface. Light did not significantly alter TN cycling. The threshold effects of light and temperature with respect to the balance between autotrophic (benthic microalgal) and heterotrophic (benthic bacterial) dominance and related alterations to nutrient cycling require further investigation.

Acknowledgements We would like to acknowledge Jamie Montgomery and the staff at the Loch Leven Estates for continued and uninterrupted support and site access and Dr. Ron Smith (CEH Edinburgh) for statistical support. We would also like to thank the reviewers for their helpful and constructive comments. Bryan Spears was funded by a NERC Ph.D. studentship (NERC/ S/A/2003/11324). R E F E R E N C E S

Axler, R.P., Reuter, J.E., 1996. Nitrate uptake by phytoplankton and periphyton: whole-lake enrichments and mesocosm-15N

ARTICLE IN PRESS WAT E R R E S E A R C H

experiments in an oligotrophic lake. Limnol. Oceanogr. 41, 659–671. Bailey-Watts, A.E., 1976a. Planktonic diatoms and some diatom– silica relations in a shallow eutrophic Scottish loch. Freshwater Biol. 6, 69–80. Bailey-Watts, A.E., 1976b. Planktonic diatoms and silica in Loch Leven, Kinross, Scotland: a one month silica budget. Freshwater Biol. 6, 203–213. Bailey-Watts, A.E., Kirika, A., 1999. Poor water quality in Loch Leven (Scotland) in 1995, in spite of reduced P loading since 1985: the influences of catchment management and interannual weather variation. Hydrobiologia 403, 135–151. Barlow-Bush, L., Baulch, H.M., Taylor, W.D., 2006. Phosphate uptake by seston in the Grand River, southern Ontario. Aquat. Sci. 68, 1615–1621. Bidle, K.D., Brzezinski, M.A., Long, R.A., Jones, J.L., Azam, F., 2003. Diminished efficiency in the oceanic silica pump caused by bacteria-mediated silica dissolution. Limnol. Oceanogr. 45, 1855–1868. Bostro¨m, B., Anderson, J.M., Fleischer, S., Jansson, M., 1988. Exchange of phosphorus across the sediment–water interface. Hydrobiologia 170, 229–244. Beutel, M.C., 2006. Inhibition of ammonia release from anoxic profundal sediments in lakes using hypolymnetic oxygenation. Ecol. Eng. 28, 271–279. Carlton, R.G., Wetzel, R.G., 1988. Phosphorus flux from lake sediments: effects of epipelic algal oxygen production. Limnol. Oceanogr. 33, 562–570. Del Amo, Y., Brzezinski, M.A., 1999. The chemical form of dissolved Si taken up by marine diatoms. J. Phycol. 35, 1162–1170. DeNicola, D.M., de Eyto, E., Wemaere, A., Irvine, K., 2006. Periphyton response to nutrient addition in 3 lakes of different benthic productivity. J. North Am. Benthol. Soc. 25, 616–631. Dodds, W.K., 2003. The role of periphyton in phosphorus retention in shallow freshwater aquatic systems. J. Phycol. 39, 840–849. Gibson, C.E., Wang, G., Foy, R.H., 2000. Silica and diatom growth in Lough Neagh: the importance of internal recycling. Freshwater Biol. 45, 285–293. Golterman, H.S., Clymo, R.S., Olmstad, M.A.M., 1978. Methods for Physical and Chemical Analysis of Freshwaters. Blackwell, Oxford, UK. Holdren, G.C., Armstrong, D.E., 1980. Factors affecting phosphorus release from intact lake sediment cores. Environ. Sci. Technol. 14, 79–87. Jansson, M., 1980. Role of benthic algae in transport of nitrogen from sediment to lake water in a shallow clearwater lake. Arch. Hydrobiol. 89, 101–109. Jarvie, H.P., Neal, C., Warwick, A., White, J., Neal, M., Wickham, H.D., Hill, L.K., Andrews, M.C., 2002. Phosphorus uptake into algal biofilms in a lowland chalk river. Sci. Total Environ. 282, 353–373. Jeppesen, E., Søndergaard, M., Jensen, J.P., Havens, K., Anneville, O., Carvalho, L., Coveney, M.F., Deneke, R., Dokulil, M.T., Foy, B., Gerdeaux, D., Hampton, S.E., Hilt, S., Kangur, K., Kohler, J., Lammens, E.H.H.R., Lauridsen, T.L., Manca, M., Miracle, M.R., Moss, B., Noges, P., Persson, G., Phillips, G., Portielje, R., Romo, S., Schelske, C.L., Straile, D., Tatrai, I., Willen, E., Winder, M., 2005. Lake responses to reduced nutrient loading—an analysis of contemporary long-term data from 35 case studies. Freshwater Biol. 50, 1747–1771. Kamp-Nielson, L., 1975. A kinetic approach to the aerobic sediment–water exchange of phosphorus in Lake Esrom. Ecol. Model. 1, 153–160. Kelderman, P., Lindeboom, H.J., Klien, J., 1988. Light dependent sediment–water exchange of dissolved reactive phosphorus and silicon in a producing microflora mat. Hydrobiologia 159, 137–147.

42 (2008) 977 – 986

985

Levine, S.N., Schindler, D.W., 1992. Modification of the N:P ratio in lakes by in situ processes. Limnol. Oceanogr. 37, 917–935. Miao, S.Y., Delaune, R.D., Jugsujinda, A., 2006. Sediment nutrient flux in a coastal lake impacted by diverted Mississippi River water. Chem. Ecol. 22, 437–449. Otsuki, A., Wetzel, R.G., 1972. Coprecipitation of phosphate with carbonates in a marl lake. Limnol. Oceanogr. 17, 763–766. Paterson, D.M., 1989. Short-term changes in the erodibility of intertidal cohesive sediments related to the migratory behaviour of epipelic diatoms. Limnol. Oceanogr. 34, 223–234. Penn, M.R., Auer, M.T., Doerr, S.M., Driscoll, C.T., Brookes, C.M., Effler, S.W., 2000. Seasonality in phosphorus release rates from the sediments of a hypereutrophic lake under a matrix of pH and redox conditions. Can. J. Fish. Aquat. Sci. 57, 1033–1041. Penna, A., Magnani, M., Fenoglio, I., Fubini, B., Cerrano, C., Giovine, M., Bavestrello, G., 2003. Marine diatom growth on different forms of particulate silica: evidence of cell/particle interaction. Aquat. Microb. Ecol. 32, 299–306. Phillips, G., Jackson, R., Bennett, C., Chilvers, A., 1994. The importance of sediment phosphorus release in the restoration of very shallow lakes (The Norfolk Broads, England) and the implications for biomanipulation. Hydrobiologia 275/276, 445–456. Redfield, A.C., Ketchum, B.H., Richards, F.A., 1963. The influence of organisms on the composition of seawater. In: Hill, M.N. (Ed.), The Sea 2: The Composition of Seawater. Wiley, New York, pp. 26–77. Rippey, B., 1983. A laboratory study of the silicon release process from a lake sediment (Lough Neagh, Northern Ireland). Arch. Hydrobiol. 96, 417–433. Risgaard-Petersen, N., Rysgaard, S., Nielsen, L.P., Revsbech, N.P., 1994. Diurnal variation of denitrification and nitrification in sediments colonized by benthic macrophytes. Limnol. Oceanogr. 39, 573–579. Sas, H., 1989. Lake Restoration by Reduction of Nutrient Loading. Academic Verlag Richarz. Søndergaard, M., Jensen, J.P., Jeppesen, E., 2005. Seasonal response of nutrients to reduced phosphorus loading in 12 Danish lakes. Freshwater Biol. 50, 1605–1615. Spears, B.M., Carvalho, L., Perkins, R., Kirika, A., Paterson, D.M., 2006. Spatial and historical variation in sediment phosphorus fractions and mobility in a large shallow lake. Water Res. 40, 383–391. Spears, B.M., Carvalho, L., Paterson, D.M., 2007a. Phosphorus partitioning in a shallow lake: implications for water quality management. Water Environ. J. 21, 47–53. Spears, B.M., Carvalho, L., Perkins, R., Kirika, A., Paterson, D.M., 2007b. Sediment phosphorus cycling in a large shallow lake: spatio-temporal variation in phosphorus pools and release. Hydrobiologia 584, 37–48. Spears, B.M., Funnell, J., Saunders, J., Paterson, D.M., 2007c. On the boundaries: sediment stability measurements across aquatic ecosystems. In: Westrich, B., Fo¨stner, U. (Eds.), Sediment Dynamics and Pollutant Mobility in Rivers: An Interdisciplinary Approach. Springer, Germany. Sweerts, J.-P.R.A., Ba¨r-Gilissen, M.-J., Cornelese, A.A., Cappenberg, T.E., 1991. Oxygen-consumption processes at the profundal and littoral sediment–water interface of a small mesoeutrophic lake (Lake Vechten, The Netherlands). Limnol. Oceanogr. 36, 1124–1133. Van Luijn, F., Van der Molen, D.T., Luttmer, W.J., Boers, P.C.M., 1995. Influence of benthic diatoms on the nutrient release from sediments of shallow lakes recovering from eutrophication. Water Sci. Technol. 32, 89–97. Webster, J.R., Mulholland, P.J., Tank, J.L., Valett, H.M., Dodds, W.K., Peterson, B.J., Bowden, W.B., Dahm, C.N., Findlay, S., Gregory, S.V., Grimm, N.B., Hamilton, S.K., Johnson, S.L., Martı´, E., Mcdowell, W.H., Meyer, J.L., Morrall, D.D., Thomas, S.A.,

ARTICLE IN PRESS 986

WA T E R R E S E A R C H

Wollheim, W.M., 2003. Factors affecting ammonium uptake in streams—an inter-biome perspective. Freshwater Biol. 48, 1329–1352. Welch, E.B., Cooke, GD., 1995. Internal phosphorus loading in shallow lakes: importance and control. Lake Reservoir Manage. 11, 273–281. Wetzel, R.G., 2001. Limnology: Lake and River Ecosystems, third ed. Academic Press, San Diego, USA.

42 (2008) 977– 986

Wetzel, R.G., Likens, G.E., 2000. Limnological Analyses, third ed. Springer, New York, USA. Woodruff, S.L., House, W.A., Callow, M.E., Leadbeater, B.S.C., 1999. The effects of biofilms on chemical processes in surficial sediments. Freshwater Biol. 41, 73–89. Zhang, X.H., Yuan, W.Y., Zhang, G.M., Sai, N., 2004. Control of nutrients after discharge to lakes through wastewater. Water Sci. Technol. 50, 173–178.