Rehabilitation of Stream Ecosystem Functions through the Reintroduction of Coarse Particulate Organic Matter

Kane T. Aldridge,1,2 Justin D. Brookes,3 and George G. Ganf1

Abstract In streams, coarse particulate organic matter (CPOM) acts as a substrate for microbial activity, which promotes nutrient retention. However, in urban areas, increased peak ows within streams lead to decreased retention of CPOM. The aim of this study was to investigate whether the reintroduction of CPOM, in the form of leaf litter, into a degraded urban stream would increase biolm activity and phosphorus retention, two ecosystem functions that reect the integrity of the ecosystem. Stream metabolism and nutrient retention were assessed in treated (T) and control (C) channels of the Torrens River Catchment, South Australia, before and after CPOM addition. Gross primary production and community respiration (CR) were measured as oxygen production and consumption within benthic chambers. Phosphorus retention was measured through a series of short-term lterable reactive phosphorus (FRP) addition experiments. Before CPOM addition,
there were no differences in CR, but C retained 6.8% more FRP than T. After CPOM addition, CR was greater in T than in C (572 and 276 mg O2m22day21, respectively), and T retained 7.7% more FRP than C. The increase in FRP retention in T compared to C was attributed to phosphorus limitation of the CPOM and increased demand for phosphorus of the attached microbial heterotrophic community. The reintroduction of CPOM into degraded streams will be an important step in the restoration of stream metabolism and nutrient retention. Maintenance of CPOM may be achieved through restoration of riparian vegetation, a reduction in the increased peak ows, and rehabilitation of stream morphology.
Key words: biolm, carbon, coarse particulate organic matter, community respiration, ecosystem function, leaf litter, phosphorus retention, stream, urbanization.

Introduction Human populations derive many benets from ecosystem functions (Costanza et al. 1997), and one important function of aquatic ecosystems is their capacity to retain nutrients (McColl 1974; Meyer & Likens 1979; Hart et al. 1991; Mulholland et al. 2001; Webster et al. 2003). The efciency of nutrient retention within streams will alter the amount and availability of nutrients discharged to receiving water bodies. Increased inputs of nutrients from external sources, such as fertilizers, have led to the articial eutrophication of lakes and reservoirs throughout the world. One way of minimizing the external load is to ensure that processes upstream of receiving water bodies that sequester and transform nutrients operate efciently. The capacity of aquatic ecosystems to intercept nutrients is due to their physical and biological diversity (Naeem et al. 1994; Hulot
1 School of Earth and Environmental Sciences, The University of Adelaide and CRC for Freshwater Ecology, Adelaide, SA 5005, Australia 2 Address correspondence to K. T. Aldridge, email kane.aldridge@adelaide. edu.au 3 School of Earth and Environmental Sciences, The University of Adelaide and CRC for Water Quality and Treatment, Adelaide, SA 5005, Australia

2007 Society for Ecological Restoration International doi: 10.1111/j.1526-100X.2007.00338.x

et al. 2000; Tilman 2000; Cardinale et al. 2002; Mulder et al. 2002), which provides multiple pathways for nutrient interception (Brookes et al. 2005). For example, obstacles, such as stones, create a heterogeneous streambed and increase solute transport to the hyporheic zone, thus increasing the likelihood of nutrient interception (Hutchinson & Webster 1998). Similarly, Engelhardt and Ritchie (2001) demonstrated that macrophyte diversity resulted in increased phosphorus uptake by supporting elevated plant (macrophyte and algal) biomass. Coarse particulate organic matter (CPOM) is pivotal for nutrient interception and transformation in streams (Moore et al. 2004). CPOM enters streams via litter fall, longitudinal transport from upstream reaches, or lateral transport from the terrestrial landscape. Upon entering streams, CPOM acts as a surface area for biolm establishment, thus increasing stream productivity and demand for nutrients. It is an energy (carbon) source for heterotrophic microbial organisms (Crenshaw et al. 2002; Wiegner et al. 2005) and so forms a basal resource for stream ecosystems (Hicks 1997; Wallace et al. 1999). Because the molar carbonnutrient ratios of CPOM (Cross et al. 2005) are an order of magnitude greater than its principal decomposers (bacteria and fungi), its breakdown is often nutrient limited (Rosemond et al. 2002; Gulis & Suberkropp

JANUARY 2009

Restoration Ecology Vol. 17, No. 1, pp. 97106

97

Rehabilitation of Stream Ecosystem Functions

2003; Stelzer et al. 2003). Thus, the presence of CPOM is likely to increase the activity of autotrophic and heterotrophic microbial organisms, increase the in-stream demand for nutrients, and therefore reduce nutrient loads to downstream ecosystems. Crenshaw et al. (2002) and Webster et al. (2000) demonstrated that the exclusion of organic matter (OM) inputs to streams reduces their capacity to retain nutrients. Furthermore, Meyer et al. (2005) related the reduced capacity of degraded urban streams to retain nutrients to their reduced standing stocks of particulate OM. Urban streams have lower standing stocks of CPOM due to the removal of riparian vegetation that reduces CPOM deposition and by channelization of stream morphology that reduces streams capacities to retain CPOM (Lepori et al. 2005). Elevated peak ows experienced in urban streams (Booth & Jackson 1997; Walsh et al. 2004) are also likely to remove CPOM directly through erosion. Urban stream reaches in the Torrens River Catchment (Fig. 1), South Australia, are characterized by lower standing stocks of CPOM and a reduced capacity to retain phosphorus in comparison to rural reaches (Aldridge 2005). The aim of this study was to investigate whether the capacity of a degraded urban stream to retain nutrients could be restored through the reintroduction of CPOM. It was hypothesized that if CPOM was added to a degraded urban stream, it would act as an attachment site and energy source for microbial communities and thus increase stream metabolic rates, increase demand for nutrients, and increase the uptake of phosphorus from stream water.

Methods
Study Area

The Torrens River Catchment is highly modied, containing a majority of metropolitan Adelaide, including 500,000 people, and supplying up to 60% of South Australias domestic water (Tonkin Consulting 2002). Early accounts suggest that the area was open woodland timbered with River Red Gum (Eucalyptus camaldulensis) and Blue Gum (E. leucoxylon), with a diverse understory (Kraehenbuehl 1978). Within the streambeds, there probably existed Common Reed (Phragmites australis) and small sedges (Kraehenbuehl 1978). Within Adelaide, the presence of E. camaldulensis and E. leucoxylon is now restricted to small parklands and along the banks of some streams. For this study, an urban reach of Fourth Creek (Fig. 1) was chosen to reect the degraded condition of urban streams within Adelaide. The use of degraded herein refers to both the channelized morphology and the reduced ecological condition of this stream. The studied reach has experienced considerable erosion and bifurcates into two channels of approximately equal dimensions. These two channels were chosen as the experimental site, one acted as a control (C) and to the other, CPOM was

Figure 1. (A) Location of Torrens River Catchment within South Australia and (B) studied stream reach in Fourth Creek. In B, dark lines within boundary represent water bodies. Modied from Tonkin Consulting (2002).

added (treated, T). Although the use of one stream is problematic (pseudoreplication) when scaling up, the variable nature of streams makes it difcult to choose appropriate independent replicates. Consequently, studies of this nature have often focused on individual streams (Bernhardt & Likens 2002; Crenshaw et al. 2002). Fourth Creek follows a northwesterly path from the Mount Lofty Ranges through Adelaide and into the Torrens River (Fig. 1). In 2003, 33% of the Fourth Creek Catchment was within metropolitan Adelaide, 41% was used for agricultural and residential purposes, and 26% remained as native vegetation. Fourth Creek is well mixed owing to the rapid, turbulent ow and shallow water. Flow is highly variable although seasonal. The majority of ow occurs during winter and spring and ceases during summer through autumn reecting the mediterranean climate of the region. Streambeds are typically dominated by rock and cobble substrates, which have attached biolms.

98

Restoration Ecology

JANUARY 2009

Rehabilitation of Stream Ecosystem Functions

Coarse Particulate Organic Matter

Leaves of E. camaldulensis were chosen as the CPOM source for the experiment. After inundation, E. camaldulensis leaves rapidly leach high concentrations of dissolved organic carbon (DOC) that is readily available to microbial organisms (OConnell et al. 2000; Francis & Sheldon 2002). Although the studied reach is lined with E. camaldulensis trees, CPOM is absent from the stream due to the elevated peak ows experienced in urban streams. This has the direct effect of resuspending deposited CPOM and transporting it downstream. In addition, it has the indirect effect of eroding the stream bank and removing obstructions that would otherwise trap CPOM and prevent it from being resuspended and transported downstream. Leaves were collected from trees at the experimental site on 9 November 2004 and stored in the dark at 3C. On 14 November 2004 (late spring), 2,193 g dry weight of leaf litter was separated into 20 groups, packed into 0.5 3 0.5m plastic mesh with 1 3 1cm pores, and distributed along the southwestern channel (treated, T) below the water surface. Mesh was also placed in the control channel, C. Mesh was secured by placing rocks upon the edges of the mesh. Leaf litter dry weight was estimated by determining the relationship between fresh and dry weights of 10 additional samples; the latter was measured after drying to a constant weight at 60C. The leaf litter load equated to 19.4 g/m2 (stream reach 100 m in length and average width of 1.13 m). This was considered representative of natural loads, as Towns (1985) estimated a leaf litter load of 56.9 g/m2 within a pool of a nearby unmodied stream. On 15, 17, 19, and 23 November 2004, leaf samples were collected and analyzed for total phosphorus (TP), total carbon (TC), and total nitrogen (TN) concentrations. TP was determined by the persulfate digestion method of standard method 4500-N C (Eaton et al. 1995). TC and TN were determined using a LECO TruSpec Carbon/ Hydrogen/Nitrogen Determinator (LECO, St. Joseph, MO, U.S.A.).
Stream Metabolism Measurements

In C, before and after CPOM addition, randomly chosen rocks were placed within three water-lled chambers, which were distributed randomly within the 100 m at known depths. Metabolic rates of biolms attached to rocks were calculated as described below and are referred to herein as rock metabolism. This was also done for T before the addition of CPOM. However, after the addition of CPOM, leaf litter from leaf packs was placed in the chambers with rocks to reect the leaf litter standing stocks of 19.4 g/m2. An additional three chambers were also placed in T, which contained leaf litter but no rocks. In this case, rock metabolism was calculated by correcting metabolic rates in chambers containing rocks and leaf litter by rates in chambers containing only leaf litter. Metabolic rates of biolms attached to leaf litter, referred to herein as leaf metabolism, were assumed to be equal with and without the presence of rocks. Community respiration (CR) was calculated as the nocturnal change in dissolved oxygen concentration, extrapolated to 24 hours (assumed to be equal during the day and night). Gross primary production (GPP) over 24 hours was calculated as the sum of the change in dissolved oxygen concentration during the day and the predicted CR during the day. Rates were adjusted for the chamber volume and the active rock surface area and were considered to be indicative of stream metabolism because rocks formed the streambed. The active rock surface area was assumed to equal half of the total surface area (top surface), which was determined by wrapping the rocks in aluminium foil, removing excess aluminium foil, and measuring the surface area of the aluminium foil using a T Area Meter (Delta-T Devices, Cambridge, U.K.).
Phosphorus Addition Experiments

The metabolic activity (measured as oxygen production/ consumption) of biolm communities was measured prior to CPOM addition (8 November) and 1 and 8 days after CPOM addition (15 and 23 November). It was impossible to extend the experiment beyond this date as ow ceased shortly after 23 November. Metabolic activity was measured by placing substrates within sealed recirculating chambers and monitoring dissolved oxygen levels (Bott et al. 1997). Chambers consisted of a Perspex dome with an internal base diameter of 28.5 cm and volume of 10 L. The chambers were sealed, and a 12 V pump recirculated water through the chamber and over the surface of the dissolved oxygen probe, which was inserted at the top of the chamber. Measurements of dissolved oxygen concentration and temperature were recorded using TPS WP-82 dissolved oxygentemperature meters (TPS Pty. Ltd., Brisbane, Australia).

Prior to CPOM addition, three lterable reactive phosphorus (FRP)sodium chloride (NaCl) addition experiments were carried out concurrently in each channel (8, 10, and 12 November). After CPOM addition, four experiments were carried out in each channel (15, 17, 19, and 23 November). Prior to each experiment, 50 L of stream water was removed and enriched with dipotassium hydrogen orthophosphate (K2HPO4) and NaCl. Sodium chloride was used as a conservative tracer because changes in its concentration (conductivity) reected dilution and dispersion. The desired FRP concentration at the point of addition after dilution with stream water was below 40 lg/L, which is lower than the ve to six times background concentrations that is considered to saturate uptake pathways (Hart et al. 1992). The desired peak conductivity at 100 m downstream was double the background conductivity (300500 lS/cm). Four measurements of conductivity of the mixed tracer solution were taken using a TPS WP-84 Conductivity-Salinity-Temperature Meter (TPS Pty. Ltd.), and four samples were taken for determination of FRP concentration (described below). The solution was

JANUARY 2009

Restoration Ecology

99

Rehabilitation of Stream Ecosystem Functions

pumped into the stream over a 2-minute period. Shortterm addition experiments were employed because the aim was to assess phosphorus retention at that particular time and not to assess the response to extended phosphorus addition (Stream Solute Workshop 1990). At 100 m downstream, conductivity and time were logged and 70 mL grab samples taken from the time of addition until conductivity returned to background. Conductivity measurements and water samples were taken at the midpoint of the channel width and water depth at intervals between 1 and 5 minutes to allow suitable coverage of changes in FRP over time. These samples were considered to be representative of the water column because the channels are well mixed and measurements of water temperature revealed no thermal discontinuities. Expected FRP concentrations if there was no FRP uptake at 100 m downstream of the point of injection were calculated by:

ExpFRPt

ECt ECto 1 FRPto ; ECs =FRPs

leach detectable levels of nutrients. Within 24 hours, half the samples were analyzed for FRP following the molybdenum blue technique (Mackereth et al. 1978). This was conducted using a Hitachi U-2000 spectrophotometer (Hitachi Ltd., Tokyo, Japan) with a path length of 10 mm. The remaining half of the samples were analyzed for DOC using an SGE ANATOC II total organic carbon analyzer (SGE, Melbourne, Australia). This was carried out in nonpurgeable organic carbon mode using titanium dioxide as a catalyst in the presence of near-UV light. The pH of each sample was adjusted to 2.8 with 0.1 M perchloric acid prior to analysis. At the time of stream metabolism and phosphorus addition experiments, measurements were also taken for stream discharge at three locations in each channel as the product of average stream velocity and cross-sectional area. Velocity was measured at three positions across the stream cross-section using a BFM002 Open Channel Flow Meter (Valeport Ltd., Dartmouth, U.K.). Cross-sectional area was calculated as the product of stream width and average water depth, measured at 5-cm intervals across the channel.
Statistical Analysis

where Exp[FRP]t is the expected FRP concentration with no uptake, ECt is conductivity at 100 m at time t, ECto is background conductivity, ECs is conductivity of the spike solution, [FRP]s is the FRP concentration of the spike solution, and [FRP]to is the background FRP concentration. Observed and expected FRP concentrations were compared over time, and percent FRP retention was calculated by comparing the area beneath the curves (between the time when expected FRP concentrations began to rise and returned to background). If there was no FRP uptake, observed FRP concentrations would match expected FRP concentrations. However, if observed FRP concentrations diverged from expected FRP concentrations with no uptake, the stream acted as a sink or a source.

Environmental Parameters

Material attached to rocks within chambers was removed with a scrubbing brush. Material attached to leaves within chambers was removed by running ngers along leaf surfaces. Removed material was suspended in reverse osmosis water, and two subsamples were concentrated onto precombusted Whatman International GF-C lters. The rst was used to measure attached OM as volatile solids following standard method 2540E (Eaton et al. 1995). The second subsample was used to measure chlorophyll a following Golterman et al. (1978). From each chamber, initial and nal water samples were taken, and prior to each phosphorus addition experiment, four grab samples were collected from each reach. Immediately following collection, samples were stored in the dark on ice. On return to the laboratory, samples were ltered through a 0.45-lm Millipore membrane lters. Filters were not prerinsed because they were not found to

Statistical analyses were performed using JMP-IN (version 3.2.1; SAS Institute, Inc., Cary, NC, U.S.A.). All treatments were tested for homogeneity of variance (OBrien test) and normality (ShapiroWilk test). Differences in percent FRP retention and environmental conditions between the two channels before and after CPOM addition and metabolic rates prior to CPOM addition were compared through t tests. Metabolic rates after CPOM addition were compared through a two-way analysis of variance with channel and time as xed effects (model 1). Differences in rock metabolic rates and leaf metabolic rates were also compared through two-way analysis of variance with substrate and time as xed effects. Interactions between two effects herein are denoted by 3. Relationships of metabolic rates and FRP retention with environmental conditions were analyzed by regression analysis. Statistically signicant relationships were accepted if p values were less than 0.05, except when interactions were tested because interactions place doubt over the F ratios of the main effects (p < 0.01 for the main effects). Variability between replicates is reported as standard errors.

Results
Stream Metabolism

There were no differences in environmental conditions between channels (Table 1). Although discharge was generally higher in C than in T, it was not statistically different on 8, 15, or 23 November (p 0.0584, 0.2962, and 0.7214, respectively, a 0.05, df 4). Similarly, the

100

Restoration Ecology

JANUARY 2009

Rehabilitation of Stream Ecosystem Functions

Table 1. Environmental parameters during measurements of stream metabolism in treated and control channels before and after CPOM addition. Time Relative to CPOM Addition

Parameter

Date

Treated

Control

Discharge (L/second) Water temperature (C) Chamber depth (m) Initial FRP (lg/L) Final FRP (lg/L) Initial DOC (mg/L) Final DOC (mg/L) Rock attached OM (g/m2) Rock attached chlorophyll a (mg/m2) Leaf attached OM (g/m2) Leaf attached chlorophyll a (mg/m2)
" SE. The values are expressed as X

Before After Before After Before After Before After Before After Before After Before After Before After Before After After After

8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 8 November 15 November 23 November 15 November 23 November 15 November 23 November

10.4 2.87 49.5 17.29 4.1 1.46 14.7 0.42 16.9 0.06 18.6 0.32 22.7 1.67 22.3 0.88 22.0 0.00 11.1 1.11 10.3 0.67 11.6 0.56 5.6 0.56 12.3 2.40 13.8 6.94 6.9 1.14 5.4 0.38 7.5 1.72 7.7 0.54 11.8 3.01 7.4 1.01 3.6 0.50 3.1 0.35 4.1 0.67 10.0 2.49 7.1 0.77 5.4 2.75 1.9 0.44 1.6 0.06 0.2 0.04 0.8 0.23

23.1 3.88 73.0 9.15 5.6 3.56 14.9 0.13 16.6 0.03 16.8 0.15 23.0 1.00 23.3 0.67 23.3 0.67 9.4 0.56 8.9 0.67 13.8 1.92 6.1 1.47 8.3 0.67 8.8 1.92 7.8 1.89 5.7 1.07 7.1 1.48 9.2 0.71 4.7 0.16 5.4 0.51 4.5 0.55 3.6 0.26 4.1 0.59 11.2 0.36 7.9 0.48 6.8 1.17

apparent differences in DOC on 15 November were not signicant due to the variation experienced in T. However, after the addition of CPOM to T, total attached OM increased (p 0.0001, a 0.05, df 9) owing to the OM attached to leaf surfaces (Table 1). Stream metabolic rates were similar within the two channels prior to CPOM addition (Fig. 2). After CPOM addition, there was no change in stream GPP, but stream CR was greater in T than in C on days 1 and 8 (Table 2; Fig. 2). The increase in total attached OM explained 80% of the increase in stream CR (p 0.0001, a 0.05, r2 0.7962, df 16; Fig. 3). The altered stream metabolic rates were initially a result of metabolism upon leaf litter. The day after the addition of CPOM, there were no differences in the GPP upon rocks and leaves, but CR was higher upon leaves than rocks (Fig. 4). However, by day 8, rock CR was greater than that upon leaf litter, explaining the effect of substrate 3 time (Table 2). Although the addition of CPOM appeared to increase rock CR in T in comparison to C on day 8 (Fig. 4), the difference was not statistically different (Table 2).

Phosphorus Retention

Environmental conditions were similar within both channels during the phosphorus addition experiments (Table 3). Discharge appeared higher in C, but this difference was not

Figure 2. Stream metabolic rates before, 1, and 8 days after the addition of leaf litter. CR and GPP in control (shaded) treated (unshaded) channels. Rates are given as per unit active surface area of rock substrate. Error bars represent standard errors.

JANUARY 2009

Restoration Ecology

101

Rehabilitation of Stream Ecosystem Functions

Table 2. p values obtained for the effects of channel and time on stream and rock metabolic rates (control and treated channel) and the effects of substrate and time on leaf and rock metabolic rates (treated channel). Test Parameters Effect GPP CR

Stream metabolic rates Rock metabolic rates Rock and leaf metabolic rates

Channel Time Channel 3 time Channel Time Channel 3 time Substrate Time Substrate 3 time

0.9994 0.5140 0.1396 0.3075 0.6798 0.2002 0.0810 0.2875 0.5164

0.0001 0.1030 0.5987 0.1187 0.6273 0.1185 0.5349 0.3266 0.0119

Figure 4. Rock and leaf CR and GPP 1 and 8 days after the addition of leaf litter. Rock metabolic rates in control (dark shaded) and treated channels (unshaded) and leaf metabolic rates in treated channel (light shaded). Rates are given as per unit surface area of rock (active) and leaf. Error bars represent standard errors.

For all effects df 1, for main effects a 0.01, and for interaction effects a 0.05.

signicant before or after CPOM addition (p 0.8810, a 0.05, df 4 and p 0.5244, a 0.05, df 6, respectively) because both channels experienced discharges between approximately 4 and 95 L/second (Table 3). However, C experienced lower background FRP concentrations (Table 3) and higher molar ratios of DOC compared to FRP than T (p 0.0008, a 0.05, df 12 and p 0.0006, a 0.05, df 12, respectively). Prior to CPOM addition, T retained 5.0% 2.3 of the added FRP, whereas C retained 11.8% 2.30, equating to a difference 6.8% 0.97 (Fig. 5). However, following the addition of CPOM, the opposite pattern was observed, with T retaining 17.4% 4.75, but C retained 9.7% 5.44 of the added FRP. This meant that there was an overall increase in percent FRP retention in T in comparison to C of 14.5%. The difference in percent FRP retention between C and T before and after CPOM addition was significant (p 0.0163, a 0.05, df 6). The increase in percent FRP retention in T (in comparison to C) coincided with increasing number of days since

CPOM addition and decreasing phosphorus availability of the CPOM (Table 3). Percent FRP retention progressively increased with the number of days since the addition of CPOM in T (p 0.0492, a 0.05, r2 0.9040, df 3, with square root transform of the number of days; Fig. 6). No such relationship was obtained in C. Also, following CPOM addition, percent FRP retention had an inverse relationship with leaf TP concentration (Fig. 7; p 0.0061, a 0.05, r2 0.9879, df 3, with natural log transform of percent FRP retention), a positive relationship with leaf TC to TP molar ratio (p 0.0066, a 0.05, r2 0.9869, df 3), and a positive relationship with leaf TN to TP molar ratio (p 0.0446, a 0.05, r2 0.9129, df 3, with square root transform of percent FRP retention).

Figure 3. Inuence of total attached OM (per unit active rock surface area) on stream CR. Measurements taken in treated (shaded triangles) and control (unshaded squares) channels of Fourth Creek following the addition of leaf litter. Full line denotes tted linear regression (p 0.0001, a 0.05, r2 0.7962, df 16).

Discussion The majority of pristine stream ecosystems depend upon allochthonous OM inputs (Vannote et al. 1980; Mulholland et al. 2001). Consequently, decreased inputs and increased exports of particulate OM, associated with anthropogenic changes, will have major implications for the functioning of stream ecosystems. Meyer et al. (2005) attributed much of the altered ecosystem functioning of degraded urban streams to reduced standing stocks of OM. In this study, nutrient retention was used as a measure of the functioning of a degraded urban stream. The amount of phosphorus retained in a degraded stream channel increased relative to a control channel following the reintroduction of CPOM. The highly dynamic nature of stream ecosystems makes the choice of replicates at larger spatial scales difcult. Thus, this study and others (Bernhardt & Likens 2002; Crenshaw et al. 2002; Wilcock et al. 2002) have chosen to focus efforts on individual streams when manipulating a component of the stream to gain a more comprehensive understanding of the factors controlling these ecosystems.

102

Restoration Ecology

JANUARY 2009

Rehabilitation of Stream Ecosystem Functions

Table 3. Environmental parameters during phosphorus addition experiments in treated and control channels before and after leaf addition. Time Relative to Leaf Addition

Parameter

Date

Treated

Control

Discharge (L/second)

Before After

Water temperature (C)

Before After

Background FRP (lg/L)

Before After

FRP at point of addition (lg/L)

Before After

DOC (mg/L)

Before After

Leaf TP (mg/g)

After

Leaf TC (mg/g)

After

Leaf TN (mg/g)

After

8 November 10 November 12 November 15 November 17 November 19 November 23 November 8 November 10 November 12 November 15 November 17 November 19 November 23 November 8 November 10 November 12 November 15 November 17 November 19 November 23 November 8 November 10 November 12 November 15 November 17 November 19 November 23 November 8 November 10 November 12 November 15 November 17 November 19 November 23 November 15 November 17 November 19 November 23 November 15 November 17 November 19 November 23 November 15 November 17 November 19 November 23 November

10.4 2.87 10.2 1.47 93.2 22.82 49.5 17.29 17.4 9.25 14.5 2.57 4.1 1.46 14.0 0.05 16.1 0.04 15.4 0.01 15.5 0.01 17.7 0.08 15.1 0.06 15.6 0.16 6.5 0.68 8.3 0.48 8.4 0.42 8.6 0.58 9.2 0.48 6.4 0.42 9.7 0.48 165.2 322.4 42.1 125.6 232.3 158.2 441.1 4.1 0.08 4.3 0.09 5.2 0.03 4.6 0.06 4.4 0.03 4.2 0.05 4.3 0.07 0.78 0.164 0.65 0.091 0.49 0.54 0.46 0.064 497.3 4.4 478.6 9.5 500.4 5.4 487.6 5.4 11.2 1.0 16.5 0.5 16.2 1.4 16.3 0.7

23.1 3.88 17.8 1.58 90.2 22.64 73.0 9.15 37.5 7.25 17.3 4.60 5.6 3.56 14.4 0.03 16.0 0.02 15.6 0.02 15.5 0.02 17.4 0.03 15.6 0.04 16.7 0.07 13.2 0.96 14.7 0.42 11.8 1.25 11.6 1.98 11.1 0.56 10.2 0.48 10.1 0.42 111.6 167.0 57.6 53.7 93.3 173.0 225.5 4.1 0.07 4.1 0.04 5.1 0.05 4.5 0.10 4.3 0.05 4.3 0.15 4.4 0.22

" SE. The values are expressed as X

The likely mechanism for the increase in FRP retention was the increased rates of CR associated with CPOM addition. This has also been found for stream sediments (Crenshaw et al. 2002). DOC is a key source of carbon for microbes, and Meyer et al. (1998) related DOC concentrations to leaf litter standing stocks. In this study, there were no signicant differences in DOC concentrations following CPOM addition. This may have been because most DOC from Eucalyptus camaldulensis leaves was released

within 24 hours of inundation (Francis & Sheldon 2002), and the DOC was transported out of the reach. Alternatively, leached DOC may have been incorporated into the microbial community and retained as particulate-bound carbon. Indeed, there was an increase in the total attached organic matter that was attached to the leaves during the timescale of this experiment. The nutrient ratios of CPOM in this study suggested that microbial assimilation of CPOM was limited by

JANUARY 2009

Restoration Ecology

103

Rehabilitation of Stream Ecosystem Functions

Figure 5. Percent FRP retention before (shaded) and after (unshaded) the addition of leaf litter. Measurements in the treated (T) and control (C) channels of Fourth Creek and the difference between treated and control channels (difference) before and after the addition of leaf litter. Differences were calculated as the percent FRP retention in the treated channel minus the percent FRP retention in the control channel. Error bars represent standard errors.

Figure 7. Inuence of leaf TP concentration on percent FRP retention. Measurements from the treated channel following the addition of leaf litter. Full line denotes tted regression with natural log transform of percent FRP retention (p 0.0061, a 0.05, r2 0.9879, df 3).

nutrients. In general, CPOM TC/TP molar ratios were 1,6002,700 compared with ratios of microbial organisms of less than 100 reported by Stelzer et al. (2003) and Cross et al. (2005). Consequently, increased CR in this study could only have continued if nutrient sources became available. Indeed, as time since CPOM addition increased, so too did the percent FRP retained. Because phosphorus was preferentially lost from CPOM through time, FRP retention continued to increase, probably because microbial heterotrophs required more water column phosphorus to sustain their increased rates of respiration. Crenshaw et al. (2002) and Bernhardt and Likens (2002) also found that carbon availability has a strong inuence on nutrient retention. The increased microbial activity and FRP retention may have also been due to the CPOM acting as surface for attachment, with an additional 30 m2 (approximate)

made available. This is signicant compared to an aerial cover of approximately 113 m2 for the 100-m reach. However, Boulton (1991) demonstrated that microbial biomass associated with Eucalyptus leaves did not increase until 20 days after submergence. Consequently, the increased CR may have been indirectly caused by an increased capacity of the stream to retain ne particulate organic matter (FPOM), which would contain an active microbial community. Microbial organisms associated with FPOM have been shown to be less important for nutrient uptake than those associated with CPOM (Sanzone et al. 2001), and so the source of the observed changes requires further investigation. Whether the benets of CPOM addition were direct or indirect, it is evident that there is a capacity for recovery of the functioning of degraded stream ecosystems with the reintroduction of CPOM. Although CPOM may initially act as a source of phosphorus, nutrient uptake from the water column by heterotrophic organisms associated with its decomposition is probably more signicant (Webster et al. 2000). The benets of rehabilitation of CPOM standing stocks would also cascade through the stream community because microbial heterotrophs and CPOM will contribute to higher trophic levels (Hall & Meyer 1998; Wallace et al. 1999) and render CPOM more palatable to other invertebrates (Suberkropp & Klug 1980). Consequently, CPOM may provide pathways of energy inputs into food webs, thus increasing biological diversity and ecosystem stability (Moore et al. 2004; Brookes et al. 2005), in addition to retaining more phosphorus to fuel further growth. Indeed, Moss (2000) has suggested that the restoration of ecosystem functioning may be the most appropriate means of achieving optimal biodiversity in freshwater ecosystems. Conclusions Efcient resource interception, an important ecosystem function, is sustained by the physical and biological diversity of natural ecosystems (Brookes et al. 2005). CPOM

Figure 6. Inuence of number of days since leaf addition on percent FRP uptake in treated (shaded triangles) and control (unshaded squares) channels. Full line denotes tted regression for the treated channel with a square root transform of the number of days (p 0.0492, a 0.05, r2 0.9040, df 3). No relationship was obtained for the control channel.

104

Restoration Ecology

JANUARY 2009

Rehabilitation of Stream Ecosystem Functions

provides physical diversity, by increasing the physical complexity of streams; and biological diversity, by providing pathways of energy inputs into food webs (Moore et al. 2004). Degraded stream ecosystems have a reduced capacity to intercept nutrients, which has been attributed to reduced standing stocks of OM (Meyer et al. 2005). In this study, it was demonstrated that it was possible to increase the amount of phosphorus retained by a degraded stream through the reintroduction of CPOM, reecting an improved functioning of the ecosystem. The reintroduction of CPOM will benet not only in-stream communities but also downstream communities through improved water quality.

Implications for Practice The rehabilitation of the heterogeneity in stream morphology and riparian vegetation is an important step in increasing inputs of CPOM (Tabacchi et al. 1998) and reestablishing the connection of streams with their catchment. d Although the rehabilitation of stream structures can successfully increase standing stocks of CPOM, the effect is unlikely to be sustained unless the overriding cause of stream degradation, altered hydrology, is addressed. d In urban streams, a reduction in peak ows may be achieved by capturing stormwater and slowly releasing it into streams, removing direct stormwater inputs into streams, and diverting stormwater over vegetated areas to promote soil inltration (Wong et al. 1999). d Such rehabilitation would provide refuge for a range of organisms (Boulton & Lake 1988; Townsend & Scarsbrook 1997), restore the functioning of ecosystems, and provide humans with resources and aesthetic and recreational benets (Bolund & Hanhammar 1999).
d

Acknowledgments We would like to thank T. Wallace, B. Deegan, and A. Payne for eld and laboratory assistance and the University of Adelaide, Cooperative Research Centre for Freshwater Ecology and Department of Water, Land and Biodiversity Conservation, Government of South Australia for nancial support.

LITERATURE CITED
Aldridge, K. T. 2005. Phosphorus retention and metabolism: indicators of stream deterioration across a rural-urban gradient? Ph.D. dissertation. The University of Adelaide, Adelaide, Australia. Bernhardt, E. S., and G. E. Likens. 2002. Dissolved organic carbon enrichment alters nitrogen dynamics in a forest stream. Ecology 83:16891700.

Bolund, P., and S. Hanhammar. 1999. Ecosystem services in urban areas. Ecological Economics 29:293301. Booth, D. B., and C. R. Jackson. 1997. Urbanization of aquatic systems: degradation thresholds, stormwater detection, and the limits of mitigation. Journal of the American Water Resources Association 33:10771090. Bott, T. L., J. T. Brock, A. Baattrup-Pedersen, P. A. Chambers, W. K. Dodds, K. T. Himbeault, J. R. Lawrence, D. Planas, E. Snyder, and G. M. Wolfaardt. 1997. An evaluation of techniques for measuring periphyton metabolism in chambers. Canadian Journal of Fisheries and Aquatic Sciences 54:715725. Boulton, A. J. 1991. Eucalypt leaf decomposition in an intermittent stream in south-eastern Australia. Hydrobiologia 211:123136. Boulton, A. J., and P. S. Lake. 1988. Australian temporary streamssome ecological characteristics. Internationale Vereinigung fur Theoretische und Angewandte Limnologie 23:13801383. Brookes, J. D., K. Aldridge, T. Wallace, L. Linden, and G. G. Ganf. 2005. Multiple interception pathways for resource utilisation and increased ecosystem resilience. Hydrobiologia 552:135146. Cardinale, B. J., M. A. Palmer, and S. L. Collins. 2002. Species diversity enhances ecosystem functioning through interspecic facilitation. Nature 415:426429. Costanza, R., R. dArge, R. de Groot, S. Farber, M. Grasso, B. Hannon, et al. 1997. The value of the worlds ecosystem services and natural capital. Nature 387:253260. Crenshaw, C. L., H. M. Valett, and J. R. Webster. 2002. Effects of augmentation of coarse particulate organic matter on metabolism and nutrient retention in hyporheic sediments. Freshwater Biology 47:18201831. Cross, W. F., J. P. Benstead, P. C. Frost, and S. A. Thomas. 2005. Ecological stoichiometry in freshwater benthic systems: recent progress and perspectives. Freshwater Biology 50:18951912. Eaton, A. D., L. S. Clesceri, and A. E. Greenberg. 1995. Fixed and volatile solids. Pages 257 in A. D. Eaton, L. S. Clesceri, E. W. Rice and A. E. Greenberg, editors. Standard methods for the examination of water and wastewater. American Public Health Association, American Water Works Association and Water Environment Federation, Washington, D.C. Engelhardt, K. A. M., and M. E. Ritchie. 2001. Effects of macrophyte species richness on wetland ecosystem functioning and services. Nature 411:687689. Francis, C., and F. Sheldon. 2002. River Red Gum (Eucalyptus camaldulensis Dehnh.) organic matter as a carbon source in the lower Darling River, Australia. Hydrobiologia 481:113124. Golterman, H. L., R. S. Clymo, and M. A. M. Ohnstad. 1978. Methods for physical and chemical analysis of fresh waters. Pages 162163 in H. L. Golterman, editor. International Biological Programme Handbook. Blackwell Scientic Publications, Oxford, United Kingdom. Gulis, V., and K. Suberkropp. 2003. Leaf litter decomposition and microbial activity in nutrient-enriched and unaltered reaches of a headwater stream. Freshwater Biology 48:123134. Hall, R. O., and J. L. Meyer. 1998. The trophic signicance of bacteria in a detritus-based stream food web. Ecology 79:19952012. Hart, B. T., P. Freeman, and I. D. McKelvie. 1992. Whole-stream phosphorus release studies: variation in uptake length with initial phosphorus concentration. Hydrobiologia 235/236:573584. Hart, B. T., P. Freeman, I. D. McKelvie, S. Pearse, and D. G. Ross. 1991. Phosphorus spiralling in Myrtle Creek, Victoria, Australia. Internationale Vereinigung fur Theoretische und Angewandte Limnologie 24:20652070. Hicks, B. J. 1997. Food webs in forest and pasture streams in the Waikato region, New Zealand: a study based on analyses of stable isotopes of carbon and nitrogen, and sh gut contents. New Zealand Journal of Marine and Freshwater Research 31:651664.

JANUARY 2009

Restoration Ecology

105

Rehabilitation of Stream Ecosystem Functions

, and M. Loreau. 2000. Hulot, F. D., G. Lacroix, F. Lescher-Moutoue Functional diversity governs ecosystem response to nutrient enrichment. Nature 405:340344. Hutchinson, P. A., and I. T. Webster. 1998. Solute uptake in aquatic sediments due to current-obstacle interactions. Journal of Environmental Engineering 124:419426. Kraehenbuehl, D. N. 1978. Flora of the Adelaide Plains. Pages 1117 in J. W. Warburton, editor. Five creeks of the River Torrens. The Department of Adult Education, The University of Adelaide, Adelaide, Australia. Lepori, F., D. Palm, and B. Malmqvist. 2005. Effects of stream restoration on ecosystem functioning: detritus retentiveness and decomposition. Journal of Applied Ecology 42:228238. Mackereth, F. J. H., J. Heron, and J. F. Talling. 1978. Phosphorus. Pages 8388 in F. J. H. Mackereth, J. Heron and J. F. Talling, editors. Water analysis: some revised methods for limnologists. Titus Wilson and Son Ltd., Kendal, United Kingdom. McColl, R. H. S. 1974. Self-purication of small freshwater streams: phosphate, nitrate, and ammonia removal. New Zealand Journal of Marine and Freshwater Research 8:375388. Meyer, J. L., and G. E. Likens. 1979. Transport and transformation of phosphorus in forest stream ecosystem. Ecology 60:12551269. Meyer, J. L., M. J. Paul, and W. K. Taulbee. 2005. Stream ecosystem function in urbanizing landscapes. Journal of the North American Benthological Society 24:602612. Meyer, J. L., J. B. Wallace, and S. L. Eggert. 1998. Leaf litter as a source of dissolved organic carbon in streams. Ecosystems 1:240249. Moore, J. C., E. L. Berlow, D. C. Coleman, P. C. de Ruiter, Q. Dong, A. Hastings, et al. 2004. Detritus, trophic dynamics and biodiversity. Ecology Letters 7:584600. Moss, B. 2000. Biodiversity in fresh watersan issue of species preservation or system functioning? Environmental Conservation 27:14. gberg, and K. Huss-Danell. 2002. Mulder, C. P. H., A. Jumpponen, P. Ho How plant diversity and legumes affect nitrogen dynamics in experimental grassland communities. Oecologia 133:412421. Mulholland, P. J., C. S. Fellows, J. L. Tank, N. B. Grimm, J. R. Webster, S. K. Hamilton, et al. 2001. Inter-biome comparison of factors controlling stream metabolism. Freshwater Biology 46:15031517. Naeem, S., L. J. Thompson, S. P. Lawler, J. H. Lawton, and R. M. Woodn. 1994. Declining biodiversity can alter the performance of ecosystems. Nature 368:734737. OConnell, M., D. S. Baldwin, A. I. Robertson, and G. Rees. 2000. Release and bioavailability of dissolved organic matter from oodplain litter: inuence of origin and oxygen levels. Freshwater Biology 45:333342. Rosemond, A. D., C. M. Pringle, A. Ramirez, M. J. Paul, and J. L. Meyer. 2002. Landscape variation in phosphorus concentration and effects on detritus-based tropical streams. Limnology and Oceanography 47:278289. Sanzone, D. M., J. L. Tank, J. L. Meyer, P. J. Mulholland, and S. E. G. Findlay. 2001. Microbial incorporation of nitrogen in stream detritus. Hydrobiologia 464:2735. Stelzer, R. S., J. Heffernan, and G. E. Likens. 2003. The inuence of dissolved nutrients and particulate organic matter quality on microbial

respiration and biomass in a forest stream. Freshwater Biology 48:19251937. Stream Solute Workshop. 1990. Concepts and methods for assessing solute dynamics in stream ecosystems. Journal of the North American Benthological Society 9:95119. Suberkropp, K., and M. J. Klug. 1980. The maceration of deciduous leaf litter by aquatic hyphomycetes. Canadian Journal of Botany 58:10251031. Tabacchi, E., D. L. Correll, R. Hauer, G. Pinay, A.-M. Planty-Tabacchi, and R. C. Wissmar. 1998. Development, maintenance and role of riparian vegetation in the river landscape. Freshwater Biology 40:497516. Tilman, D. 2000. Causes, consequences and the ethics of biodiversity. Nature 405:208211. Tonkin Consulting. 2002. Torrens Catchment water management plan. Torrens Catchment Water Management Board, Adelaide, Australia. Towns, D. R. 1985. Limnological characteristics of a South Australian intermittent stream, Brown Hill Creek. Australian Journal of Marine and Freshwater Research 36:821837. Townsend, C. R., and M. R. Scarsbrook. 1997. The intermediate disturbance hypothesis, refugia, and biodiversity. Limnology and Oceanography 42:938949. Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130137. Wallace, J. B., S. L. Eggert, J. L. Meyer, and J. R. Webster. 1999. Effects of resource limitation on a detrital-based ecosystem. Ecological Monographs 69:409442. Walsh, C. J., A. W. Leonard, A. R. Ladson, and T. D. Fletcher. 2004. Urban stormwater and the ecology of streams. Cooperative Research Centre for Freshwater Ecology and Cooperative Research Centre for Catchment Hydrology, Canberra, Australia. Webster, J. R., P. J. Mulholland, J. L. Tank, H. M. Valett, W. K. Dodds, B. J. Peterson, et al. 2003. Factors affecting ammonium uptake in streamsan inter-biome perspective. Freshwater Biology 48: 13291352. Webster, J. R., J. L. Tank, J. B. Wallace, J. L. Meyer, S. L. Eggert, T. P. Ehrman, B. R. Ward, B. L. Bennett, P. F. Wagner, and M. E. McTammany. 2000. Effects of litter exclusion and wood removal on phosphorus and nitrogen retention in a forest stream. Internationale Vereinigung fur Theoretiche und Angewandle Limnologie 27: 13371340. Wiegner, T. N., L. A. Kaplan, and J. D. Newbold. 2005. Contribution of dissolved organic C to stream metabolism: a mesocosm study using 13 C-enriched tree-tissue leachate. Journal of the North American Benthological Society 24:4867. Wilcock, R. J., M. R. Scarsbrook, K. J. Costley, and J. W. Nagels. 2002. Controlled release experiments to determine the effects of shade and plants on nutrient retention in a lowland stream. Hydrobiologia 485:153162. Wong, T. H. F., P. F. Breen, and S. D. Lloyd. 1999. Retrotting urban drainage systems for integrated stormwater management. Pages 271 279 in First South Pacic conference on comprehensive stormwater and aquatic ecosystem management, Auckland, New Zealand.

106

Restoration Ecology

JANUARY 2009