Leaf litter chemistry, decomposition and assimilation by ... - Springer Link

Hydrobiologia (2012) 680:63–77 DOI 10.1007/s10750-011-0903-1

PRIMARY RESEARCH PAPER

Leaf litter chemistry, decomposition and assimilation by macroinvertebrates in two tropical streams Neil E. Pettit • Tegan Davies • Jason B. Fellman Pauline F. Grierson • Danielle M. Warfe • Peter M. Davies

•

Received: 23 March 2011 / Revised: 15 September 2011 / Accepted: 25 September 2011 / Published online: 18 October 2011 Ó Springer Science+Business Media B.V. 2011

Abstract Riparian vegetation typically provides substantial allochthonous material to aquatic ecosystems where micro-organisms can play an important role in organic matter degradation which can support consumer biomass. We examined the effects of leaf litter quality (e.g., leaf nutrients, lignin and cellulose content), leaf species mixing, and microbial community diversity on in-stream breakdown rates of litter from dominant riparian trees (Melaleuca argentea, M. leucadendra, and Nauclea orientalis) in both a perennial and intermittent river in Australia’s wet-dry tropics. Leaf mass remaining after 82 days of in-stream incubation was negatively correlated (P \ 0.05) with initial leaf N and P content while initial lignin and cellulose content had

Handling editor: David J. Hoeinghaus

no statistically significant effect. Breakdown rates of incubated leaves of both Melaleuca and Nauclea were significantly higher in mixed litter bags compared with single species litter bags. Although it was expected that leaf N content would decrease from initial levels during decomposition, we found either similar or slightly higher N content following in-stream incubation suggesting microbial colonisation increased overall N content. Stable isotopes of d13C and d15N for the major sources and consumers in both rivers provide evidence that leaf litter was an important macroinvertebrate food source in the perennial river where heavy shading may limit algal production. However, in the intermittent river where riparian cover was low, benthic algae were the major organic carbon source for consumers. Our findings suggest that riparian tree species influence rates of in-stream organic matter processing, microbial community composition, and aquatic food web dynamics in tropical wet-dry streams.

N. E. Pettit (&) P. M. Davies Centre of Excellence in Natural Resource Management, The University of Western Australia, Albany, WA, Australia e-mail: [email protected]

Keywords Melaleuca Nauclea Microbial community Mass loss Detrital processing Interactive effects Aquatic food webs

T. Davies J. B. Fellman P. F. Grierson Ecosystems Research Group, School of Plant Biology, The University of Western Australia, Crawley, WA, Australia

Introduction

D. M. Warfe Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, NT, Australia

Leaf litter from riparian plants provides an important source of nutrients, energy and shelter in freshwater ecosystems (Henderson & Walker, 1986; Wallace

123

64

et al., 1997), particularly in temperate small forested streams where dense canopy coverage limits light penetration and subsequent autochthonous production (Vannote et al., 1980; Davies et al., 2008). In contrast, streams and floodplain lakes where light and nutrients do not limit algal growth often have high autochthonous production and algae provide the main basal source of energy for aquatic food webs (Bunn et al., 2003; Douglas et al., 2005; Molina et al., 2011). This is primarily because benthic algae are considered a higher quality food with a much lower carbon to nitrogen (C/N) ratio compared to leaf litter, which commonly contain high levels of recalcitrant compounds such as lignin and cellulose (Hamilton et al., 1992; Bunn et al., 1997; Lewis et al., 2001). In the wet–dry tropics of northern Australia, much of the riparian tree canopy is from the Myrtaceace family. Common genera such as Melaleuca and Eucalyptus have highly sclerophyllous leaves such that most leaf litter that enters streams is refractory and high in polyphenolic compounds (Coley & Barone, 1996), which in turn can affect leaf breakdown (Ardoń & Pringle, 2009). Once leaf material enters a stream, soluble organic material and other nutrients are rapidly leached. Leaves are either transported downstream or remain in the stream for long periods of time, where refractory structural biopolymers are slowly consumed by microorganisms (e.g. mainly bacteria and fungi; Glazebrook & Robertson, 1999; Gonçalves et al., 2006; Encalada et al., 2010). There is a paucity of shredders in tropical river systems, although caddis fly larvae (Phylloicus sp.) and some macro-crustaceans are important shredders in some neotropical streams (Crowl et al., 2001; Encalada et al., 2010). This is in contrast to temperate forested streams where macroinvertebrate shredders play a major role in the degradation of leaf material (Dobson et al., 2002; Boulton et al., 2008; Leigh & Sheldon, 2009). Consequently, microbes likely control leaf litter breakdown and subsequent release of nutrients in tropical streams (Franken et al., 2005; Reid et al., 2008). The breakdown of leaf litter by aquatic microbial communities could therefore be an essential but poorly recognized component of tropical aquatic food webs, providing an ecological link between terrestrial primary production and aquatic secondary production (Biddanda & Cotner, 2002). Litter from a number of species form mixed leaf packs on the stream bed (Kominoski et al., 2009).

123

Hydrobiologia (2012) 680:63–77

Although information about the decomposition rates of individual species is useful, litter mixing studies have shown that decomposition is not always predictable from the dynamics of single species (Kominoski et al., 2007; Pe´rez-Harguindeguy et al., 2008). This is due to the interactive effects of leaf species and initial leaf chemistry that cause different decomposition rates within in situ leaf packs. Interactive effects are likely due to either the translocation of stimulatory or inhibitory nutrients from leaf species with high concentrations to those with lower concentrations (McArthur et al., 1994; Ball et al., 2008), or the creation of a heterogeneous micro-environment for leaf consumers to coexist (Stout, 1989; Leff & McArthur, 1989; Bastian et al., 2008). Different litter types are also likely to have varying effects on microbial community structure and productivity with subsequent effects on litter breakdown. Consequently, the mixing of different leaf species may be more important than individual leaf quality alone, particularly when plant communities are diverse (Kominsoki et al., 2009). While leaf chemistry, leaf identity and microbial communities contribute to litter decomposition, abiotic conditions can also play a role. Water chemistry and stream discharge both influence breakdown rates of leaf litter in streams (Davies et al., 2008). In particular, microbial activity and decomposition rates of litter are strongly influenced by water temperature, turbidity, pH, salinity, dissolved organic carbon (DOC), nutrients and oxygen (Wetzel, 1983; Wantzen et al., 2008). Variability in stream discharge can also affect litter breakdown rates where high stream flows increase physical fragmentation of leaves as well as increasing the downstream fluxes of carbon and nutrients to microbes and other biota, thereby increasing mass loss of leaf litter (Glazebrook & Robertson, 1999; Gonçalves et al., 2006). In northern Australia, the strong climatic seasonal cycle of wet and dry seasons influences river hydrology (Boulton et al., 2008, Warfe et al., 2011), which in turn affect the dynamics of the aquatic food web (Douglas et al., 2005; Jardine et al., 2011). However, an understanding of how riverine physical conditions (e.g. discharge) influence litter breakdown and subsequent incorporation of carbon into tropical river food webs is still lacking. This study examined rates of leaf litter breakdown of key riparian species with contrasting leaf chemistry


in two rivers in the wet–dry tropics of northern Australia: one intermittent and one with permanent flow. Our objective was to determine how leaf litter quality influenced in-stream breakdown rates and the incorporation of leaf litter into the metazoan food web of these two tropical rivers. We hypothesised that initial leaf litter quality (evaluated by leaf percent lignin, N, P, and C content, lignin:N, and C:N) would control mass loss, and that a mixture of leaf species would create interactive effects leading to faster breakdown rates than those observed for each species alone. We further hypothesised that biomass and composition of microbial communities would differ between litter species, reflecting litter quality that would, in turn, affect consumer preferences (Glazebrook & Robertson, 1999; Castillo et al., 2003; Gonçalves et al., 2006).

65

Nauclea orientalis, and Terminalia platyphylla. As the aim of the study was to examine species with contrasting leaf chemistry, we have focused on Melaleuca and Nauclea for our experiment. Melaleuca and Nauclea are also both widespread across rivers of northern Australia.. Both M. argentea and M. leucadendra are included in the M. leucadendra complex and are taxonomically similar (Barlow, 1988) with broad sclerophyllous leaves high in volatile polyphenolics; these species are also difficult to distinguish in the field. Consequently, we made no distinction between M. argentea and M. leucadendra while collecting leaves. Thus, the combined leaves used for experiments and samples will herein be referred to as Melaleuca. In contrast to Melaleuca, N. orientalis (Family Rubiaceae, herein referred to as Nauclea) was selected as a more mesophyllic species with large soft, non-sclerophyllous leaves that we expected to be of higher substrate quality for aquatic biota.

Methods Study sites Leaf litter breakdown experiments were conducted in the Fergusson (14°04.30 S–131°57.20 E) and Edith (14°10.20 S-132°07.10 E) Rivers, both third-order tributaries of the Daly River in northern Australia (Fig. 1). Mean annual rainfall at Katherine (50 km south of study area) is 987 mm, of which less than 2% falls between May and September (Bureau of Meteorology, 2009). The Fergusson River is intermittent and surface water becomes restricted to a series of pools during the dry season (May to October). In contrast, the Edith River is groundwater-fed and surface flow persists throughout most years, so is considered perennial (Tickell, 2009). The study reach for both rivers consisted of a pool-run-pool sequence that extended over 200 m at Edith River, and over 500 m at Fergusson River. The Edith River averaged 7 m wide along the study reach and the streambed substrate was predominately rocky, while the Fergusson River was *20 m wide with a sand and rock substrate. The Edith and Fergusson rivers flow through relatively intact tropical savannah vegetation; the Edith River is heavily shaded (*50% canopy cover), while the Fergusson River had around 15% canopy cover along the study reach. Riparian trees for both rivers included Melaleuca argentea, M. leucandendra, Pandanus aquaticus, Eucalyptus camaldulensis,

Experimental design Field measurements of litter decomposition were made over a 12-week period in the dry season from the 7th May to the 28th July, 2009. In order to assess differences in leaf litter mass loss (representing loss from both physical and biological processes) between species, a fully replicated sampling design was established at both the Fergusson and Edith Rivers. Three evenly spaced transects perpendicular to each stream bank were selected over a 200–500 m reach of each river, specifically targeting pools where we observed litter that had naturally accumulated. One replicate of each litter treatment (Melaleuca, Nauclea and Melaleuca ? Nauclea mixed) was anchored on the stream bed at five positions evenly spaced along each transect ([1 m apart); litter bags remained submerged for 82 days. Consequently, our overall design encompassed three litter treatments (Melaleuca only, Nauclea only and Melaleuca ? Nauclea mixed) 9 fifteen replicates at each of two sites of contrasting flow (ephemeral and perennial). Leaf litterbag preparation, installation and sample processing Fresh leaves were collected from all sides of the canopy from five trees of Melaleuca and Nauclea

123

66


Darwin 250 kms

" Pine Creek

Fergusson R

Edith R

¯

0 3 6

12

18

24 Kilometers

"

Katherine

Fig. 1 Map of site locations on the Fergusson (intermittent) and Edith (perennial) rivers in relation to the nearby towns of Pine Creek and Katherine and the city of Darwin in the Northern Territory, Australia

along each study stream reach. Approximately 5 g of air-dried leaves (air-dried for 24 h) were placed in separate mesh litterbags (20 cm 9 15 cm, mesh size of 1 9 2 mm), and 5 g of a mix of Melaleuca and Nauclea leaves were placed in litterbags in equal mass proportions. Together, these litterbags represented the three litter treatments. To allow access to leaves for larger aquatic macroinvertebrates, 12 round holes (5 mm diameter) were cut in each side of the bags. Five subsamples of leaves of each species was kept for analysis of initial leaf chemistry. After 82 days of incubation, the litterbags were carefully removed and placed into individual plastic bags and transported on ice to the laboratory. One replicate of each litter treatment from each transect (i.e. three replicates per

123

river) was frozen for analysis of phospholipid fatty acids (PLFAs) to determine microbial biomass and community composition within litterbags (see below). All remaining replicates were carefully washed to remove fine debris and sediment. Macroinvertebrates found in the litterbags were identified to order or family and, where possible, assigned to functional feeding groups (FFGs). Leaf content of each litterbag was air-dried and weighed by species to the nearest 0.01 g for mass loss determination. Approximately 0.6 g was then removed from the single species litterbags only and retained for lignin and cellulose analyses. All litterbags were then oven-dried at 50°C for 24 h, reweighed and approximately 0.3 g was removed from each


67

single species litterbag and retained for analysis of % P, C and N, and carbon and nitrogen stable isotopes (d13C and d15N). Approximately 0.1 g of leaf material was also collected from each species within mixed litterbags for % C and N, and for d15N and d13C analysis. Mass loss over the 82 days was then calculated for all litterbags to determine breakdown rates using the exponential decay model:

nitrogen (TDN) were determined from three replicate filtered (Whatman glass fibre filters, nominal pore size 0.7 lm) river water samples from each site and then analysed via high temperature combustion on a DOC/ TDN analyser (Shimadzu, Kyoto, Japan). Discharge of both rivers was estimated on both dates using a handheld FlowTracker (Biolab Australia, Clayton, Victoria, Australia).

Wt ¼ Wo ekt

Stable isotope analysis for trophic structure

where Wt is the remaining mass at 82 days, Wo is the initial mass, t is time (82 days), and k is the breakdown rate (Webster & Benfield, 1986; Glazebrook & Robertson, 1999; Boyero et al., 2006). Litter chemistry Subsamples of air-dried fresh leaves and incubated litter were roughly ground in a coffee grinder. Lignin and cellulose contents of fresh leaves and decomposed leaves from the single species litterbags were determined using the acid-detergent fibre (ADF) method (Van Soest, 1963). Leaf material from litterbags was analysed for total P content on an auto-analyser, after acid digestion in concentrated H2SO4. All samples were analysed for C and N content (%), and d15N and d13C (%) using a Automated Nitrogen Carbon Analyser-Mass Spectrometer consisting of a 20/20 mass spectrometer connected with an ANCA-S1 preparation system (Europa Scientific Ltd., Crewe, UK). Samples were standardised against a secondary reference of radish collegate (3.167% N; d15N 5.71%; 41.51% C; d13C 28.61%) that was subsequently standardised against primary analytical standards (IAEA, Vienna). Accuracy was measured as 0.07% and precision as 0.03%. Stream physical and chemical characteristics Chemical and physical stream conditions were measured at the initiation (May 2009) and end (July 2009) of the 82-day incubation period. Water physico-chemical parameters, including pH, turbidity, temperature, conductivity and dissolved oxygen were measured with a handheld multi-parameter meter (Hydrolab QUANTA, ECO Environmental, Perth, Western Australia, Australia) and a HACH turbidity meter (Biolab Australia, Clayton, Victoria, Australia). DOC and total dissolved

Stable isotopes of carbon (d13C) and nitrogen (d15N) were analysed to assess carbon sources and trophic linkages between leaf litter, benthic algae, biofilm and aquatic macroinvertebrates. Leaf litter biofilm, benthic algae and aquatic macroinvertebrates were collected from each river at the time of litterbag retrieval. Biofilm was collected from leaves by scraping them with a razorblade, benthic algae were collected by scrubbing rocks, and macroinvertebrates were picked from kick- and sweep-net samples. In each case, samples were collected from a range of habitats across the reach so that they were considered representative of the diversity of biota occurring with the reach. Samples were kept on ice during transport from the field to the laboratory and then preserved by freezing. After sorting to family, whole bodies of invertebrates were used for stable isotope analysis, however, the shells were first removed from shrimp. Samples were prepared for isotope analysis by oven-drying at 50°C and were ground to a fine powder before being analysed via combustion and mass spectrometry (as above). The d13C values were corrected for lipid contents using C:N ratios (see Logan et al., 2008). Microbial biomass and community composition Microbial biomass and community (bacteria and fungi) composition of decomposed litter were assessed using PLFA analysis (Pennanen et al., 1999; Pietikinena et al., 2007). The amount of total PLFAs provides a proxy measure of microbial biomass and the different types indicate community composition (Pietikinena et al., 2007). The biofilm layer on the leaves in the litterbags that developed during incubation could not be easily isolated from the conditioned leaf fragments; consequently, PLFA analyses were conducted on combined leaf fragments and attached biofilm. However, PLFA analyses were

123

68

also conducted on biofilm scraped from the surface of Melaleuca and Nauclea leaf litter collected directly from the sites to supplement the data from the litterbags. Total lipids were extracted from thawed litter (1–2 g) using a modified single-phase extraction procedure (Greenwood et al., 2009). The PLFAs were isolated from free and neutral fatty acids by column chromatography to assist analysis of their methyl ester analogues on an Agilent 6890/5973 Gas Chromatograph linked to a Mass Spectrometer (GC-MS). The fatty acids and their derivatives were then analysed on the same GC-MS (operational parameters of the GC-MS are detailed by Greenwood et al. (2009)). Fatty acid methyl-ester (FAME) abundances were determined by the integration of total ion chromatogram peak areas. The concentration of each fatty acid was determined relative to the internal standard and calculated using the following formula: ðP 17:6 ng StdÞ= ðW lg individual fatty acid per gram litter=biofilmÞ: where P is percentage fatty acid compared to C19:0, 17.6 ng Std is the concentration of the internal standard (ng mL-1 solvent), and W is the fresh weight for extraction. A commercial PLFA standard was separately analysed to assist product identification.


indicated. All data were analysed using StatView Version 5.0 (SAS Institute Inc, Cary, USA). Analysis of Similarities (ANOSIM) and non-metric multidimensional scaling (nMDS) ordinations were used for comparisons of microbial community composition among leaf litter treatments and between rivers. Multivariate tests of PLFA data were based on Bray-Curtis dissimilarities calculated among observations. Data were normalised using square-root transformations. A similarity percentages routine (SIMPER) was also used to reveal the contributions of each PLFA to any community difference (Clarke & Gorley, 2006). For stable isotope analysis the relative importance of the different sources sampled (Melaleuca litter, Nauclea litter, biofilm and benthic algae) to consumers was assessed using the IsoSource mixing model program (Phillips & Gregg, 2003). We calculated the range of percentage contributions (1–99 percentiles) of each source to each consumer’s d13C value, with increments of 2% and a tolerance of 0.2%. Sources with a low 99th percentile are not likely to be important for that consumer, while a source with a high first percentile would likely be a significant food source for a consumer (Phillips & Gregg, 2003).

Results

Data analyses

Stream environment

Differences between species in initial total leaf chemistry were analysed using multivariate ordination of the range of parameters measured for leaves, with the Gower metric used as a measure of dissimilarity and the significance of dissimilarities tested using a permutation procedure applied to the dissimilarity matrix (Analysis of Similarity (ANOSIM), Primer v6.0; Clarke & Gorley, 2006). Differences in initial (0 day) and final (82 days) leaf chemistry between leaf species as well as rates of mass loss for Melaleuca and Nauclea were determined using analysis of variance (ANOVA) followed by Fisher’s PLSD tests. Abundance of PLFAs between leaf species and between rivers was also analysed using ANOVA followed by Fisher’s PLSD. Data were either square-root or log10 transformed where necessary to improve homogeneity of variances and meet assumptions of normality; the significance level was P \ 0.05 unless otherwise

The intermittent Fergusson River had ceased flowing and contracted to a series of large disconnected pools by the end of the experimental period (28th July), whereas the Edith River was still flowing, albeit at lower flow rates (Table 1). These changes in river flow were reflected in the water chemistry, as conductivity and pH did not change for the Edith River between sample dates but large increases were observed in the Fergusson River. Concentrations of DOC and TDN were also greater in the Fergusson than in the Edith River (Table 1).

123

Initial leaf chemistry Overall chemistry of the leaves prior to incubation significantly differed between Melaleuca and Nauclea (ANOSIM, R = 0.385, P \ 0.001; Table 2) as well as between rivers (R = 0.171, P = 0.042). For initial


69

Table 1 Stream conditions for the Fergusson and Edith rivers at the initiation (7th May, mid-afternoon) and conclusion (28th July, mid-afternoon), of the field incubation period in 2009, (mean ± SE, n = 5) pH

Discharge (m3/s)

DOC (mg/l)

TDN (mg/l)

Month

River

Temp (°C)

Cond. (lS/cm)

Dissolved O2 (mg/l)

May

Fergusson

26.5 (0.3)a

28 (0.0)b

7.4 (0.05)b

7.0 (0.2)

0.193

–

–

c

7.3 (0.03)b

6.9 (0.1)b

0.400

–

–

No flow

3.60 (0.45)

0.34 (0.12)

0.179

1.53 (0.03)

0.09 (0.01)

ab

May

Edith

25.9 (0.0)

July

Fergusson

25.4 (0.6)b

45 (3.0)a

7.23 (0.05)b

8.2 (0.0)a

c

c

a

b

July

Edith

21.3 (0.1)

14 (0.0) 15 (0.0)

8.09 (0.14)

6.9 (0.3)

Different letters within a column indicate significant differences at P \ 0.05 between rivers and sampling times (– indicates no sample taken)

leaf chemistry, Nauclea had a significantly higher initial % N content (F2,17 = 53.63, P \ 0.001), lower % C content (F2,17 = 65.19, P \ 0.001), and higher total P concentration (in the Fergusson only) than Melaleuca. The differences in initial leaf C and N content were reflected in the C:N ratio, as the mean C:N ratio of Melaleuca was significantly greater than Nauclea (F2,17 = 51.17, P \ 0.001). The lignin:N and cellulose:N ratios were significantly higher in Melaleuca leaves than Nauclea even though species did not differ in the initial percentage of lignin or cellulose in the leaves (Table 2).

Chemistry of mixed species litterbags at the end of the incubation period showed variable results. The N content was significantly higher (*23%) in Nauclea leaves in the mixed litterbags at the Edith River compared with final N content in the single species litterbags (P \ 0.01), whereas there was no significant difference for either species in the Fergusson River (Table 2). Similarly, at the end of the incubation period, Melaleuca leaves in mixed species litterbags showed a significantly higher percent C content than in single species litterbags in the Fergusson River (P \ 0.01), while C of Nauclea was 16% higher in mixed species litterbags in the Edith River (P \ 0.01, Table 2).

Changes in leaf litter chemistry Mass loss of leaf litter Nutrient concentrations and C fractions of leaves in single species litterbags changed significantly (all P \ 0.01) in both rivers following 82-day field incubations (Table 2). Percent N content did not change following incubation in most treatments (single and mixed species). In contrast, percent C decreased by as much as 15% in all Nauclea treatments (all F2,8 [ 18.42, P \ 0.05, Table 2), and P concentration was two to three times lower for Melaleuca and Nauclea after field incubation (all F2,8 [ 103.96, P \ 0.001). Moreover, the C:N ratio of Nauclea did not change following in-stream incubation. However, for Melaleuca the C:N ratios decreased for samples in the Fergusson River (p \ 0.001) but increased in the Edith River (P = 0.038). Evaluating the C fractions of the litter showed that the relative contribution of lignin increased *10–20% and cellulose by 4–36% for all treatments by the end of the incubation (Table 2). The lignin:N ratio did not change for either species incubated in the Fergusson River, but was three times higher for Melaleuca and twice as high for Nauclea in the Edith River (Table 2).

Mixed species litterbags in the Edith River had breakdown rates up to twice as fast as in the Fergusson River (Fig. 2). In the Fergusson River, mass loss of single species Nauclea was similar to single species Melaleuca (F2,8 = 3.47, P = 0.054). However, mixing promoted mass loss in Nauclea leaves, which had a significantly greater mass loss than Melaleuca in single (P = 0.012) and mixed species litterbags (F2,8 = 7.57, P = 0.012; Fig. 2a). In the Edith River, mass loss was greater in Nauclea compared with Melaleuca litterbags, and in mixed species litterbags for both species (F2,8 = 7.59, P = 0.012 for both species, Fig. 2b). Breakdown rates of single Nauclea (mean k = 0.010 day-1) were 1.5 times faster than single Melaleuca (k = 0.007 day-1) in the Edith River (F2,8 = 0.19.8, P \ 0.001; Fig. 2b). Nauclea breakdown was almost twice as fast when mixed with Melaleuca (mean k = 0.020 day-1) than when incubated separately (k = 0.010 day-1, F2,8 = 21.72, P \ 0.001; Fig. 3b), losing 77 ± 5% of the initial mass. Melaleuca in mixed litter bags lost an

123

70


Table 2 Mean chemical composition values (± SE) for Melaleuca and Nauclea leaf litter in single and mixed species bags before and after 82 days in-stream incubations at (a) the Fergusson and (b) the Edith River Melaleuca

Nauclea

Initial

Final single

Final mix

Initial

Final single

Final mix

33.1 (2.1)c

49.5 (0.7)a

–

b

a

32.1 (2.3)c

39.4 (2.5)b

–

c

–

18.2 (1.1)b

–

(a) Fergusson Lignin (%) Cellulose (%) Nitrogen (%) Carbon (%) Phosphorus (%) C:N

17.8 (1.0)

24.3 (0.4)

1.2 (0.1)c

1.6 (0.1)ab

b

56.6 (0.1)

56.2 (0.4)

b

1.6 (0.09)ab a

c

2.2 (0.2)a

1.8 (0.2)ab

d

41.1 (3.0)d

52.4 (0.3)

47.7 (1.9)

–

0.15 (0.01)a

0.07 (0.0)

0.05 (0.0)b

a

35.6 (2.6)

b a

–

a

1.9 (0.1)ab

61.1 (1.4)

0.1 (0.0)c 47.4 (3.2)

14.3 (1.2)

b

38.4 (2.3)

c

27.5 (0.9)

b

23.7 (1.4)c

b

– –

23.1 (1.3)

b

Lignin:N

28.2 (3.1)

31.5 (2.6)

Cellulose:N

14.6 (0.7)a

15.4 (1.1)a

–

7.5 (0.7)b

9.1 (0.9)b

N:P

12.9 (0.8)b

35.2 (2.9)a

–

13.5 (0.5)b

34.1 (5.3)a

C:P Lignin:P

b

a

16.8 (1.2)

595.1 (30.0) 238.0 (30.4)c

1206 (69.7) 1063 (66.3)a

– –

368 (17.7) 202 (9.3)c

23.6 (0.7)d

53.8 (4.6)a

– –

19.4 (2.0)

c

–

c

– b

733 (57.2) 600 (49.1)b

– –

27.6 (1.6)c

45.8 (1.6)b

–

a

11.9 (1.1)b

–

1.9 (0.0)a

1.5 (0.1)b

b

d

(b) Edith Lignin (%) Cellulose (%) Nitrogen (%) Carbon (%)

a

23.4 (2.87)

1.4 (0.1)b

1.2 (0.0)b

22.4 (0.7)

a

58.6 (0.4)

Phosphorus (%)

0.19 (0.01)

C:N

42.7 (3.2)b c

57.2 (0.7) a

a

1.1 (0.0)b

ab

0.07 (0.00)

23.9 (1.1)

b

48.4 (1.3)ab

a

59.8 (1.6)

54.1 (0.8)

0.15 (0.01) 28.4 (1.2)c

27.8 (0.9)c

c

b

–

8.1 (0.8)c

–

a

14.5 (1.0)

–

12.5 (0.3)c

45.6 (3.8)

16.3 (1.3)b

19.7 (2.3)a

c

b

bc

0.04 (0.00)

d

54.6 (2.0)a –

17.1 (1.2)

Cellulose:N

47.5 (3.2)c

41.0 (2.2) b

–

a

Lignin:N

2.0 (0.1)a

32.2 (2.8)

– 24.6 (2.2)c

N:P

7.7 (0.5)

17.0 (1.1)

–

13.1 (1.2)

37.0 (4.0)

–

C:P

317 (7.2)c

816 (45.5)b

–

362 (21.0)c

1011 (81.9)a

–

Lignin:P

127 (2.8)c

753 (51.4)b

–

182 (8.4)c

1146 (94.3)a

–

Different letters within a row indicate significant differences (at P \ 0.05) between litter bag types and initial and final dates of incubations

extra 17 ± 3% mass than single Melaleuca (F2,8 = 30.3, P \ 0.001) and had breakdown rates comparable to single Nauclea (mean k = 0.011 day-1, k = 0.010 day-1, respectively). Importantly, leaf litter mass remaining for both species and in both rivers was negatively correlated with initial percent N and P concentration, and positively correlated with C:N and lignin:N ratios (n = 36, Fig. 3). Macroinvertebrate fauna associated with litter bags Invertebrate abundance, diversity or community composition did not differ between litter types, but there were greater differences between streams than between litter types. Similarly, there was no difference in the

123

relative abundance in functional feeding groups between litter bag types (Table 3). Filter feeders were the dominant functional feeding group (FFG) in terms of relative abundance, accounting for over 50% of total macroinvertebrates in litter bags at the Fergusson R and 70% at the Edith (Table 3). Predators and omnivores (mainly shrimp) were generally much less abundant and no shredders were identified in the litter bags (Table 3). The three most abundant taxa found in the litter bags at both sites were the mayfly families Baetidae and Caenidae and non-biting midge larvae (Chironomidae). Microbial biomass and community composition Microbial biomass was significantly higher in single Nauclea litterbags (17.2 ± 1.5 lg g-1) than Nauclea

Hydrobiologia (2012) 680:63–77 100 80

71

(a) Fergusson a

K (day-1) = 0.005 (0.0003)

a

K (day-1) = 0.005 (0.0004)

ab

Mass remaining (%)

60

b

K (day-1) = 0.008 (0.001)

(Global R = -0.125, P = 0.01), and was used to support, but not directly compare trends from the litterbags.

K (day-1) = 0.009 (0.001)

40

Leaf litter links to food webs

20 0 100

(b) Edith

80

K (day-1) = 0.007 (0.0004)

a

K (day-1) = 0.011 (0.001)

60 b

40

K (day-1) = 0.010 (0.001)

b

K (day-1) = 0.020 (0.002)

c

20 0

Single Mixed

Single Mixed

Melaleuca

Nauclea

Fig. 2 Mean values (±SE) of % leaf mass remaining of Melaleuca and Nauclea in the a Fergusson and b Edith rivers after 82 days in-stream incubation as single and mixed species. Bars (n = 12) capped with the same letter within each figure are not significantly different (P \ 0.05). Breakdown rates, k (day-1)

in mixed species litterbags (10.3 ± 1.6 lg g-1), but did not differ significantly from microbial biomass on Melaleuca in single (13.9 ± 2.3 lg g-1) or mixed species (12.8 ± 1.7 lg g-1) litterbags (F2,8 = 5.06, P \ 0.035). There was also no significant difference in microbial biomass between the Edith and Fergusson rivers (F2,8 = 0.01, P \ 0.939). Microbial type composition (assessed by PLFAs) did not differ between litter species within litterbags (ANOSIM Global R = 0.027, P = 0.319) or in the biofilm collected from standing leaf litter of the two species (Global R = 0.043, P = 0.247). However, microbial community composition did differ between rivers, in both the litterbags (Global R = 0.434, P = 0.001; Fig. 4a) as well as in collected standing litter (Global R = 0.289, P = 0.002; Fig. 4b). The PLFA assessment indicated that the ubiquitous fatty acids 16:0 (common to bacteria and eukaryotes) and 18:1 x9 (which can derive from fungal, alga, higher plants or gramnegative bacteria sources) contributed most towards dissimilarity. These fatty acids were also quantitatively the most significant PLFAs, together making up 60 ± 3% of PLFAs from litterbags in the Fergusson River and 44 ± 2% in the Edith River. Standing in-stream leaf litter had the same microbial community composition as the litterbags in both rivers

Stable isotope analysis showed that the biofilm d13C signal (-28.5%) was similar to the benthic algal d13C signal (-28.3%) in the perennial Edith River (Fig. 5). However, biofilm d13C values in the intermittent Fergusson River were more depleted (-31 to -33%) and similar to values for Melaleuca and Nauclea litter. These d13C values are likely indicative of differences in the major carbon source for biofilms in each river. Similarly, the basal carbon source supporting first- and second-order consumers varied between rivers. Benthic algae were the primary carbon source for consumers in the Fergusson River, as benthic algae d13C values (-26%) were similar to some first-order consumers (-26.5%), most of the second-order consumers (-24 to -27%) and shrimp (-25.4%), whereas Nauclea and Melaleuca leaves and biofilm values were more depleted (Fig. 5a). Benthic algae as the primary carbon source was confirmed by mixing models; the range of potential contribution (1–99th percentile) for benthic algae was 44–88% for first-order consumers and 68–88% for second-order consumers compared with 0–40% for the other sources. In contrast, we observed a greater range of d13C values for first- -26 to -34%) and second-order consumers (-24 to -33%) in the Edith River, which encompassed benthic algal (-28%) and leaf litter (-31 to 34%) signatures, indicating that consumers in the Edith River assimilated a mixture of benthic algae, terrestrial leaves and/or biofilm (Fig. 5b). Mixing model results for the Edith River also indicate that there was a mixture of sources, where the strongest contribution for first-order consumers was from Melaleuca litter (30–88%), with Nauclea litter (0–55%) and benthic algae (0–30%) unlikely to be an important source. For second-order consumers, Nauclea litter (8–86%), benthic algae (0–80%) and biofilm (0–66%) all had relatively high maximum values and therefore could be potential sources, while Melaleuca litter (0–30%) is unlikely to be an important source. Predatory bugs (Nepidae) and beetles (Dytiscidae) had more enriched d13C signatures (-24 to -22%), indicating these two consumers were likely consum-

123

72


Fig. 3 Relationship between percent leaf mass remaining and initial leaf chemistry of combined Melaleuca and Nauclea leaves for both rivers

Table 3 Mean relative abundance (% ± SE) of aquatic macroinvertebrate functional feeding groups for each leaf pack type and at each river Feeding group

Fergusson R

Edith R

Melaleuca

Nauclea

Mixed

Detritivores

20.7 ± 3.1

18.5 ± 4.5

Omnivores

1.4 ± 1.4

4.8 ± 3.7

Predators

18.6 ± 12.9

Grazers Filter feeders

35.9 ± 7.0 23.4 ± 8.6

Shredders

0.0

Melaleuca

Nauclea

Mixed

18.8 ± 2.1

20.6 ± 7.8

24.3 ± 16.9

17.0 ± 17.0

3.5 ± 2.5

14.7 ± 5.9

2.7 ± 2.7

6.4 ± 3.7

8.1 ± 2.1

11.8 ± 8.7

20.6 ± 5.9

16.2 ± 9.4

4.3 ± 4.3

15.3 ± 5.3 53.2 ± 14.6

27.8 ± 0.7 38.2 ± 20

0.0 38.2 ± 11.8

13.5 ± 9.7 43.2 ± 19.5

2.1 ± 2.1 70.2 ± 54.2

0.0

ing carbon from a source not sampled in this study, possibly via the consumption of terrestrial invertebrates. The d15N values were similar at each river and indicated the trophic level of invertebrate consumers; first-order consumers generally had more depleted

123

0.0

5.9 ± 3.9

0.0

0.0

values (1.5–4.6%) than second-order consumers (3–8.5%), and this difference is equivalent to that generally considered to represent one trophic level (Post, 2002). The shrimp (Macrobrachium rosenburgii) had the most enriched d15N values for both


(a)

73

E

E

E

F

E

E

E F

E

E

F

E

F

F

Discussion

F

F

F

E

E F

F

rivers (7 and 8.5%) indicating assimilation of mostly other consumers rather than primary sources such as algae or detritus.

E

This study provides new insight on how leaf litter chemistry and stream microbial communities influence leaf breakdown in two tropical rivers with contrasting hydrology. For instance, initial N and P content were related to in-stream mass loss of leaves. However, ‘‘structural’’ leaf material such as lignin and cellulose had no effect. Furthermore, species mixing of leaves resulted in faster breakdown rates than individual species alone. Colonisation of leaf litter by both aquatic macroinvertebrates and microbial communities did not differ between leaf species suggesting no litter quality preferences. Rather, these communities are likely to reflect the different environmental conditions present in both the perennial and intermittent streams. This of course needs to be tested with further experimentation under a range of river conditions of flow and chemistry. Our finding that macroinvertebrate or microbial biomass and community composition could not explain the difference in mass loss between single and mixed species leaf litter provides evidence that physical and/or chemical characteristics of a river have a greater influence on mass loss than microbial

F

Melaleuca Nauclea Mixed - Melaleuca Mixed - Nauclea

F Similarity 80%

(b) F

F

E E

E

E

F

E

E

F

F

F

Melaleuca Nauclea

Similarity 80%

Fig. 4 NMDS (nonmetric multidimensional scaling) ordination of PLFA (phospholipid fatty acids) abundances associated with Melaleuca and Nauclea leaves in a litter bags, and b biofilm from in-stream standing litter, where (F) is the Fergusson River and (E) the Edith River. The NMDS figure is derived from a Bray-Curtis dissimilarity matrix of square-root transformed samples clustered at 80% similarity. The two dimensional stress for ordinations where a 0.07, and b 0.02

(a) Fergusson

(b) Edith

δ 15N (0/00)

8

8

Shr Hem

6 4

Gyr Dyt

Omnivore 1st consum. 2nd consum.

Dyt

Lep

4

Nauclea Melaleuca

Nep

Tab Zyg Sim Epi

6

Tab

Ger Epi

Shr

Benth. algae Bae Eph

2

Cae

Biofilm

2

0

0 -35

-30

-25

-35

-20

δ 13C (0/

-30

-25

-20

00)

Fig. 5 Carbon and nitrogen stable isotope biplots of each leaf litter species, benthic algae, biofilms, primary consumers, secondary consumers and omnivores in a the Fergusson River, and b the Edith River. Consumers are abbreviated as: Bae Baetidae (mayflies), Cae Caenidae (mayflies), Dyt Dytiscidae (water beetles), Epi Epiproctophora (dragonflies), Eph

Ephemeroptera (mayflies), Ger Gerridae (water striders), Gyr Gyrindae (whirlygig beetles), Hem Hemiptera (water bugs), Lep Leptophlebiidae (mayflies), Nep Nepidae (water scorpions), Shr Shrimp (Palaemonidae), Tab Tabanidae (march flies), Sim Simuliidae (black flies), Zyg Zygoptera (damselflies)

123

74

activity. For instance, the greater flow in the Edith River could effectively wash fine matter out of the litterbags, or physically weaken the structural integrity of leaf membranes and facilitate the leaching of dissolved and particulate organic matter. This finding is consistent with previous research in an Amazon floodplain stream, which found irregular hydrologic pulses had greater impact on leaf breakdown than leaf associated-invertebrates (Rueda-Delgado et al., 2006). Low flow conditions at the Fergusson River could also reduce leaf breakdown rates through the accumulation of soluble polyphenols that can inhibit microbial activity (Bunn, 1988). Although we did not observe a decrease in dissolved oxygen (DO) concentration in the Fergusson River during in-stream incubations, low DO concentrations can restrict microbial processing of leaf litter (Wantzen et al., 2008) and localised areas of low DO may have reduced breakdown rates in this river. Leaf breakdown rates were also controlled by initial N content and to a lesser extent, initial leaf P. While recent studies in tropical streams have shown that leaf toughness (usually lignin and cellulose content) rather than N and P control litter breakdown rates (Ardoń et al., 2009; Li et al., 2009), we observed no influence of lignin on rates of mass loss. This is despite the relatively high lignin content in our study species (Melaleuca and Nauclea) compared to other species from the wet tropics (e.g. Ardoń et al., 2009). The initial N content of our study leaves (1.2–1.9%) is slightly higher than levels reported for litter fall studies in Amazonia (0.6–1.8%, Wantzen et al., 2008). This may promote greater decomposition by invertebrates and microbial communities as well as increase leaching of N (and probably other nutrients), thus stimulating the growth of microorganisms and thereby increasing the rate of leaf breakdown (Mathuriau & Chauvet, 2002). Leaves with higher N content are also more likely to encourage rapid fungal and microbial colonisation with fungal hyphae and bacterial enzymes penetrating and breaking down leaf structures. However, breakdown rates reported here (0.005–0.020 k day-1) are within the range reported for tree litterbag studies in tropical streams elsewhere (0.001–0.0651 k day-1; see Wantzen et al., 2008). High ultraviolet radiation in the tropics and semi-arid areas will accelerate breakdown of lignin in leaf litter (Austin & Vivanco, 2006), so that differences in breakdown rates between species with different leaf

123


chemistry may be reduced and mixing effects are not as pronounced as has been shown in temperate streams (Kominoski et al., 2007, 2009). In-stream mass loss of mixed species leaf litter was significantly greater compared with single species alone at Edith River, although this is again likely to be dependent on river conditions (Lecerf & Richardson, 2010), particularly flow, because only a subtle mixing effect on mass loss was observed in the Fergusson River. Other studies generally show additive effects on leaf litter breakdown of mixed species compared with single species alone (Gartner & Cardon, 2004; Kominoski et al., 2009). However, the diversity of consumer organisms may also have stronger effects on organic matter processing than resource diversity in the form of leaf litter consisting of a number of species (Kominoski et al., 2010). Another possibility is that leaves of poor quality could act as a physical substrate for micro-organisms to feed on the more palatable species (Stout, 1980; Kominoski et al., 2009). A lack of mixing or interactive effects is also not uncommon and has been observed in a tropical stream in Brazil (Moretti et al., 2007) and many temperate streams (Lecerf et al., 2007; Taylor et al., 2007; Kominoski et al., 2009). Thus, the lack of interactive effects are most likely related to local river conditions such as microbial community composition, water chemistry, flow regime and the type and cover of the riparian forest. Our finding that breakdown rates were higher in mixed species litterbags compared to single species litterbags in the Edith but less so in the Fergusson may reflect differences in the microbial community composition between the two rivers. Stream water chemistry is known to directly influence microbial communities in biofilms (Lyautey et al., 2005) and on leaf litter (Fischer et al., 2009). The Fergusson River had higher concentrations of DOC and TDN, conductivity and pH when there was no flow (in July), and combined with the static hydrological conditions, may have led to a divergence in the microbial communities between the rivers. Stable isotopes analysed from the major sources and macroinvertebrate consumers in both rivers suggested that leaf litter was an important macroinvertebrate food source in the Edith River. A possible reason for this finding is the smaller stream size of the Edith R creates edge effects that allow for greater riparian inputs and greater shading, which can minimise algal production (Davies et al., 2008). In contrast, benthic


algae were the primary source of organic C supporting the food web in the more open Fergusson River. Although many studies of tropical rivers have shown that benthic algae are an important food source (Bunn et al., 2003; March & Pringle, 2003; Brito et al., 2006, Molina et al., 2011), especially in Australia (Douglas et al., 2005; Leigh et al., 2010). A seasonal switch between the wet to dry season from algae to plant detritus as the major C source has been observed in rivers of both temperate Australia (Reid et al., 2008) and also Hong Kong (Lau et al., 2009) as well for a tropical floodplain lake (Molina et al., 2011). Given that algae are considered to be the preferred C source over sclerophyllous leaves (e.g. Melaleuca spp.), aquatic consumers may utilize riparian inputs of plant detritus as a C source when environmental conditions restrict algal growth. Taken together, our findings suggest that riparian forest coverage influence the relative importance of different C sources for food webs in tropical streams. Decomposing leaves may become more attractive as a food source to aquatic consumers through changes in leaf chemistry. For example, leaf N content was similar or slightly greater following in-stream incubation in the Fergusson River. This finding suggests either microbial colonisation of leaves or preferential loss of C via microbial respiration could be altering leaf nutrient content. Although we found no evidence of greater leaf decomposition in leaves with elevated N content, other studies have shown that elevated leaf N content associated with microbial biomass can promote greater decomposition rates (Gessner & Chauvet, 1994; Glazebrook & Robertson, 1999). Alternatively, a large fraction of DOC leached from leaf litter (e.g., Eucalyptus) can be metabolized by bacteria (Baldwin, 1999) that are, in turn, grazed by microinvertebrates (e.g. protozoa) that also may be consumed by higher order consumers (Meyer, 1994). This trophic pathway may provide a means by which DOC leached from recalcitrant leaf litter can support metazoan food webs. However, further work is required to understand the fate of leaf litter C in tropical streams and the potential mechanisms that may lead to its incorporation into the aquatic food web, especially the role of stream environmental conditions in leaf litter breakdown. In summary, our findings suggest that clearing riparian forests and/or intensive prescribed burning regimes could alter aquatic food web dynamics by

75

increasing available light and stimulating benthic algal production, while at the same time reduce leaf litter inputs, including nutrients, to streams. This could alter the contribution of allochthonous relative to autochthonous carbon as well as the individual leaf species entering streams. Consequently, there could be a shift in the relative importance of both sources of carbon for food webs in northern Australian tropical rivers. Acknowledgements The authors gratefully acknowledge the Wagiman and Jawoyn communities in the Northern Territory for access to study sites. We thank Erica Garcia, Peter Kyne, and Peter Novak for assistance with fieldwork. We thank Paul Greenwood (Western Australian Biochemistry Centre, UWA) for help with the PLFA analysis, Greg Skyrzypek (Western Australian Biochemistry Centre, UWA) for assistance with isotope analyses and Tim Jardine (Griffith University) for lipid corrections on the stable isotope data. We also thank two anonymous reviewers for constructive comments on a previous version. Funding for this project was provided by the Tropical Rivers and Coastal Knowledge (TRaCK) research programme which receives major funding for its research through the Australian Government’s Commonwealth Environment Research Facilities initiative, the Australian Government’s Raising National Water Standards Programme, the Fisheries Research and Development Corporation, and the Queensland Government’s Smart State Innovation Fund.

References Ardoń, M. & C. M. Pringle, 2008. Do secondary compounds inhibit microbial- and insect-mediated leaf breakdown in a tropical rainforest stream, Costa Rica? Oecologia 155: 311–323. Ardoń, M., C. M. Pringle & S. L. Eggert, 2009. Does leaf chemistry differentially affect breakdown in tropical vs temperate streams? Importance of standardized analytical techniques to measure leaf chemistry. Journal of the North American Benthological Society 28: 440–453. Austin, A. T. & L. Vivanco, 2006. Plant litter decomposition in a semi-arid ecosystem controlled by photodegradation. Nature 442(3): 555–558. Baldwin, D. S., 1999. DOM and phosphorus leached from fresh and ‘terrestrially’ aged river red gum leaves: implications for assessing river-floodplain interactions. Freshwater Biology 41: 675–685. Ball, B. A., M. D. Hunter, J. S. Kominoski, C. M. Swan & M. A. Bradford, 2008. Consequences of non-random species loss for decomposition dynamics: experimental evidence for additive and non-additive effects. Journal of Ecology 96: 303–313. Barlow, B. A., 1988. Patterns of differentiation in tropical species of Melaleuca. Proceedings of the Ecological Society of Australia 15: 239–247. Bastian, M., R. G. Pearson & L. Boyero, 2008. Effects of diversity loss on ecosystem function across trophic levels

123

76 and ecosystems: a test in a detritus-based tropical food web. Austral Ecology 33: 301–306. Biddanda, A. B. & J. B. Cotner, 2002. Love handles in aquatic ecosystems: the role of dissolved organic carbon drawdown, resuspended sediments, and terrigenous inputs in the carbon balance of Lake Michigan. Ecosystems 5: 431–445. Boulton, A. J., L. Boyero, A. P. Covich, M. Dobson, P. S. Lake & R. Pearson, 2008. Are tropical streams ecologically different from temperate streams? Chapter 9. In Dudgeon, D. (ed.), Tropical Stream Ecology. Elsevier, London: 257–284. Boyero, L., R. G. Pearson & R. Camacho, 2006. Leaf breakdown in tropical streams: the role of different species in ecosystem functioning. Archives of Hydrobiologia 166: 453–466. Brito, E. F., T. P. Moulton, M. L. Souza & S. E. Bunn, 2006. Stable isotope analysis in microalgae as the predominant food source of fauna in a coastal forest stream, south-east Brazil. Austral Ecology 31: 623–633. Bunn, S. E., 1988. Processing of leaf litter in two northern jarrah forest streams, Western Australia: II. The role of macroinvertebrates and the influence of soluble polyphenols and inorganic sediment. Hydrobiologia 162: 211–223. Bunn, S. E., P. M. Davies & D. M. Kellaway, 1997. Contributions of sugar cane and invasive pasture grass to the aquatic food web of a tropical lowland stream. Marine and Freshwater Research 48: 173–179. Bunn, S. E., P. M. Davies & M. Winning, 2003. Sources of organic carbon supporting the food web of an arid zone floodplain river. Freshwater Biology 48: 1–17. Bureau of Meteorology, 2009. Commonwealth of Australian Government [available on internet at www.bom.gov.au]. Castillo, M. M., G. W. Kling & J. D. Allan, 2003. Bottom-up controls on bacterial production in tropical lowland rivers. Limnology and Oceanography 48: 1466–1475. Clarke, K. R. & R. N. Gorley, 2006. PRIMER v6: User Manual/ Tutorial. Primer-E, Plymouth. Coley, P. D. & J. A. Barone, 1996. Herbivory and plant defences in tropical forests. Annual Review of Ecology and Systematics 27: 305–335. Crowl, T. A., W. H. McDowell, A. P. Covich & S. L. Johnson, 2001. Freshwater shrimp effects on detrital processing and nutrients in a tropical headwater stream. Ecology 82: 775–783. Davies, P. M., S. E. Bunn & S. K. Hamilton, 2008. Primary production in tropical streams and rivers, Chapter 2. In Dudgeon, D. (ed.), Tropical Stream Ecology. Elsevier, London: 23–42. Dobson, M., A. Magana, J. M. Mathooko & F. K. Ndegwa, 2002. Detritivores in Kenyan highland streams: more evidence for the paucity of shredders in the tropics? Freshwater Biology 47: 909–919. Douglas, M. M., S. E. Bunn & P. M. Davies, 2005. River and wetland food webs in Australia’s wet-dry tropics: general principles and implications for management. Marine Freshwater Research 56: 329–342. Encalada, A. C., J. Calles, V. Ferreira, C. M. Canhoto & A. S. Graça, 2010. Riparian land use and the relationship between the benthos and litter decomposition in tropical montane streams. Freshwater Biology 55: 1719–1733.

123

Hydrobiologia (2012) 680:63–77 Fischer, H., J. Bergfur, W. Goedkoop & L. Tranvik, 2009. Microbial leaf degraders in boreal streams: bringing together stochastic and deterministic regulators of community composition. Freshwater Biology 54: 2276–2289. Franken, R. J. M., B. Waluto, E. T. H. M. Peeters, J. J. P. Gardeniers, J. A. J. Beijer & M. Scheffer, 2005. Growth of shredders on leaf litter biofilms: the effect of light intensity. Freshwater Biology 50: 459–466. Gartner, T. B. & Z. G. Cardon, 2004. Decomposition dynamics in mixed-species leaf litter. Oikos 104: 230–246. Gessner, M. O. & E. Chauvet, 1994. Importance of stream microfungi in controlling breakdown rates of leaf litter. Ecology 75: 1807–1817. Glazebrook, H. S. & A. I. Robertson, 1999. The effect of flooding and flood timing on leaf litter breakdown rates and nutrient dynamics in a river red gum (Eucalyptus camaldulensis) forest. Australian Journal of Ecology 24: 625–635. Gonçalves, J. F., M. A. S. Graça & M. Callisto, 2006. Litter breakdown dynamics at three streams in temperate, mediterranean and tropical Cerrado climates. Journal of the North American Benthological Society 25: 344–355. Greenwood, P. F., S. Wibrow, S. J. George & M. Tibbett, 2009. Hydrocarbon biodegradation and microbial community response to repeated oil exposure. Organic Geochemistry 40: 293–300. Hamilton, S. K., W. M. Lewis & S. J. Sippel, 1992. Energy sources for aquatic animals in the Orinoco River floodplain: evidence from stable isotopes. Oecologia 89: 324–330. Henderson, P. A. & I. Walker, 1986. On the leaf litter community of the Amazonian blackwater stream Tarumazinho. Journal of Tropical Ecology 2: 1–16. Jardine, T. D., B. J. Pusey, S. K. Hamilton, N. E. Pettit, P. M. Davies, M. M. Douglas, V. Sinnamon, I. A. Halliday & S. E. Bunn, 2011. Fish mediate high food web connectivity in the lower reaches of a tropical floodplain river. Oecologia. doi:10.1007/S00442-011-2148-0. Kominoski, J. S., C. M. Pringle, B. A. Ball, M. A. Bradford, D. C. Coleman, D. B. Hall & D. B. Hunter, 2007. Non additive effects of leaf litter species diversity on breakdown dynamics in a detritus-based stream. Ecology 88: 1167–1176. Kominoski, J. S., T. J. Hoellein, J. J. Kelly & C. M. Pringle, 2009. Does mixing litter of different qualities alter stream microbial diversity and functioning on individual litter species? Oikos 118: 457–463. Kominoski, J. S., T. J. Hoellein, C. J. Leroy, C. M. Pringle & C. M. Swan, 2010. Beyond species richness: expanding biodiversity-ecosystem functioning theory in detritus-based streams. River Research and Applications 26: 67–75. Lau, D. C. P., K. M. Y. Leung & D. Dudgeon, 2009. What does stable isotope analysis reveal about trophic relationships and the relative importance of allochthonous and autochthonous resources in tropical streams? A synthetic study from Hong Kong. Freshwater Biology 54: 127–141. Lecerf, A. & J. S. Richardson, 2010. Biodiversity-ecosystem function research: insights gained from streams. River Research and Applications 26: 45–54. Lecerf, A., G. Risnoveanu, C. Popescu, M. O. Gessner & E. Chauvet, 2007. Decomposition of diverse litter mixtures in streams. Ecology 88: 219–227.

Hydrobiologia (2012) 680:63–77 Leff, L. G. & J. V. McArthur, 1989. The effect of leaf pack composition on processing: a comparison of mixed and single species packs. Hydrobiologia 182: 219–224. Leigh, C. & F. Sheldon, 2009. Hydrological connectivity drives patterns of macroinvertebrate biodiversity in floodplain rivers of the Australian wet/dry tropics. Freshwater Biology 54: 549–571. Leigh, C., M. A. Burford, F. Sheldon & S. E. Bunn, 2010. Dynamic stability in dry season food webs within tropical floodplain rivers. Marine and Freshwater Research 61: 357–368. Lewis, W. M., S. K. Hamilton, M. A. Rodriguez, J. F. Saunders & M. A. Lasi, 2001. Food web analysis of the Orinoco floodplain based on production estimates and stable isotope data. Journal of the North American Benthological Society 20: 241–254. Li, O. Y., C. Y. Ng & D. Dudgeon, 2009. Effects of leaf toughness and nitrogen content on litter breakdown and macroinvertebrates in a tropical stream. Aquatic Science 71: 80–93. Logan, J. M., T. D. Jardine, T. J. Miller, S. E. Bunn, R. A. Cunjak & M. E. Lutcavage, 2008. Lipid corrections in carbon and nitrogen stable isotope analyses: comparison of chemical extraction and modelling methods. Journal of Animal Ecology 77: 838–846. Lyautey, E., C. R. Jackson, J. Cayrou, J. L. Rols & F. Garabetian, 2005. Bacterial community succession in natural river biofilm assemblages. Microbial Ecology 50: 589–601. March, J. G. & C. M. Pringle, 2003. Food web structure and basal resource utilization along a tropical island stream continuum, Puerto Rico. Biotropica 35: 84–93. Mathuriau, C. & E. Chauvet, 2002. Breakdown of leaf litter in a neotropical stream. Journal of the North American Benthological Society 21: 384–396. McArthur, J. V., J. M. Aho, R. B. Rader & G. L. Mills, 1994. Interspecific leaf interactions during decomposition in aquatic and floodplain ecosystems. Journal of the North American Benthological Society 13: 57–67. Meyer, J. L., 1994. The microbial loop in flowing waters. Microbial Ecology 28: 195–199. Molina, C. I., F.-M. Gibon, T. Oberdorff, E. Dominquez, J. Pinto, R. Marin & M. Roulet, 2011. Macroinvertebrate food web structure in a floodplain lake of the Bolivian Amazon. Hydrobiologia 663: 135–153. Moretti, M., J. F. Goncalves & M. Callisto, 2007. Leaf breakdown in two tropical streams: Differences between single and mixed species packs. Limnologica 37: 250–258. Pennanen, T., J. Liski, E. Baath, V. Kitunen, J. Uotila, C. J. Westman & H. Fritze, 1999. Structure of the microbial communities in coniferous forest soils in relation to site fertility and stand development stage. Microbial Ecology 38: 168–179. Pe´rez-Harguindeguy, N., C. M. Blundo, D. E. Gurvich, S. Dıáz & E. Cuevas, 2008. More than the sum of its parts? Assessing litter heterogeneity effects on the decomposition of litter mixtures through leaf chemistry. Plant Soil 303: 151–159.

77 Phillips, D. L. & J. W. Gregg, 2003. Source portioning using stable isotopes: coping with too many sources. Oecologia 136: 261–269. Pietikinena, J., P. J. Tikkab, S. Valkonenb, A. Isomakib & H. Fritze, 2007. Is the soil microbial community related to the basal area of trees in a Scots pine stand? Soil Biology and Biochemistry 39: 1832–1834. Post, D. M., 2002. Using stable isotopes to estimate trophic position: models, methods and assumptions. Ecology 83: 703–718. Reid, D. J., G. P. Quinn, P. S. Lake & P. Reich, 2008. Terrestrial detritus supports the food webs in lowland intermittent streams of south-eastern Australia: a stable isotope study. Freshwater Biology 53: 2036–2050. Rueda-Delgado, G., K. M. Wantzen & M. B. Tolosa, 2006. Leaf-litter decomposition in an Amazonian floodplain stream: effects of seasonal hydrological changes. Journal of the North American Benthological Society 25: 233–249. Stout, J., 1980. Leaf decomposition rates in Costa Rican lowland tropical rainforest streams. Biotropica 12: 264–272. Stout, R. J., 1989. Effects of condensed tannins on leaf processing in mid-latitude and tropical streams: a theoretical approach. Canadian Journal of Fisheries and Aquatic Sciences 46: 1097–1106. Taylor, B. R., C. Mallaley & J. F. Cairns, 2007. Limited evidence that mixing leaf litter accelerates decomposition or increases diversity of decomposers in streams of eastern Canada. Hydrobiologia 592: 405–422. Tickell, S., 2009. Groundwater in the Daly Basin. Northern Territory. Technical Report No. 27/2008D. Government Department of Natural Resources, The Environment, The Arts and Sport, Darwin, Australia. Van Soest, P. J., 1963. Use of detergents in the analysis of fibrous feeds II. A rapid method for determination of fibre and lignin. Journal Association of Official Agricultural Chemistry 46: 829–835. Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell & C. E. Cushing, 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37: 130–137. Wallace, J. B., S. L. Eggert, J. L. Meyer & J. R. Webster, 1997. Multiple trophic levels of a forest stream linked to terrestrial litter inputs. Science 277: 102–104. Wantzen, K. M., C. M. Yule, J. M. Mathooko & C. M. Pringle, 2008. Organic matter processing in tropical streams, Chapter 3. In Dudgeon, D. (ed.), Tropical Stream Ecology. Elsevier, London: 44–64. Warfe, D. M., N. E. Pettit, P. M. Davies, B. J. Pusey, S. K. Hamilton, P. Bayliss, D. P. Ward, M. J. Kennard, S. Townsend, M. M. Douglas, M. A. Burford, M. Finn, S. E. Bunn & I. Halliday, 2011. The wet-dry in the wet-dry tropics drives ecosystem structure and function of Australian rivers. Freshwater Biology 56: 2169–2195. Webster, J. R. & E. F. Benfield, 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics 17: 567–594. Wetzel, R. G., 1983. Limnology. Saunders College Publishing, Philadelphia.

123