Shredder colonization and decomposition of green and senescent leaves during summer in a headwater stream in northern Japan

We conducted a decomposition experiment using green and senescent maple and alder leaves in a coastal headwater stream in Hokkaido, northern Japan, during June and July 2000. We estimated whether shredder colonization on the leaves and leaf breakdown differed between green and senescent leaves during the experimental period. Late-instar Lepidostoma complicatum (Trichoptera) and Sternomoera rhyaca (Amphipoda) were the predominant shredder taxa among the macroinvertebrates that colonized litterbags. There was no significant difference in shredder colonization between green and senescent leaves although we found a significant difference between maple and alder leaves. The colonization patterns of large individuals of L. complicatum and S. rhyaca differed from those of small individuals. All decomposition coefficients of green and senescent leaves were high. During the experiment, decomposition was significantly faster in maple than in alder leaves, although no significant difference was found between green and senescent leaves. However, the fragmented nitrogen portion was higher in green leaves than in senescent leaves during the experiment. Higher nitrogen release (2–2.5 times more) as particulate organic matter in green than in senescent leaves indicates that green leaves may be a potentially valuable food resource for other macroinvertebrate collector–gatherer species.     

Flow-substrate interactions create and mediate leaf litter resource patches in streams

1. The roles that streambed geometry, channel morphology, and water velocity play in the retention and subsequent breakdown of leaf litter in small streams were examined by conducting a series of field and laboratory experiments. 2. In the first experiment, conditioned red alder (Alnus rubra Bongard) leaves were released individually in three riffles and three pools in a second‐order stream. The transport distance of each leaf was measured. Several channel and streambed variables were measured at each leaf settlement location and compared with a similar number of measurements taken at regular intervals along streambed transects (‘reference locations’). Channel features (such as water depth) and substrate variables (including stone height, stone height‐to‐width ratio, and relative protrusion) were the most important factors in leaf retention. 3. In the second experiment, the role of settlement location and reach type in determining the rate of leaf litter breakdown was examined by placing individual conditioned red alder leaves in exposed and sheltered locations (on the upper and lower edges of the upstream face of streambed stones, respectively) in riffle and pool habitats. After 10 days, percent mass remaining of each leaf was measured. Generally, leaves broke down faster in pools than in riffles. However, the role of exposure in breakdown rate differed between reach types (exposed pool > sheltered pool > sheltered riffle > exposed riffle). 4. In the third experiment, the importance of substrate geometry on leaf litter retention was examined by individually releasing artificial leaves upstream of a series of substrate models of varying shape. Substrates with high‐angle upstream faces (were vertical or close to vertical), and that had high aspect ratios (were tall relative to their width), retained leaves more effectively. 5. These results show that streambed morphology is an important factor in leaf litter retention and breakdown. Interactions between substrate and flow characteristics lead to the creation of detrital resource patchiness, and may partition leaf litter inputs between riffles and pools in streams at baseflow conditions.     

Differences in patches of retention among leaves,woods and small litter particles in a headwater stream: the importance of particle morphology

The location of retention in the channel can influence invertebrate assemblage and breakdown processes associated with a litter particle in heterogeneous streams. We previously identified four types of litter patches that formed on riffles or different locations in pools (middle, edge, alcove), and demonstrated that middle patches had higher litter processing rates than the other patches. In this study, we examined differences in retention on the four patch types among leaves, woody materials and small litter particles, and among leaves of different sizes, by sampling natural and manipulated litter particles that were newly retained. Proportionally more woody materials, leaf pieces (16–50 mm) and particulate organic matter (1–16 mm) than leaves (>50 mm) were retained on middle patches, while proportionally more leaves than the other litter particles were retained on riffle and edge patches. The retention pattern of leaf species with different leaf sizes and a released experiment of leaf particles revealed that proportionally more leaf particles with smaller sizes were retained on middle patches. The flexibility, shape and dryness of litter particles also seem to affect the location of retention in the channel. These results suggest that the morphology of litter particles have the potential to affect the biological use and breakdown of litter particles through determining the location of retention within the channel. The size of leaves and processes that alter leaf size may have important roles on the breakdown and utilisation by invertebrates in these heterogeneous streams.     

Towards a predictive description of forest canopies from litter properties

1. Specific leaf area (SLA), carbon isotope ratio (δ¹³C) and leaf nitrogen content (N) were measured in litter in 1993 and in the litter layer and canopy in 1996 in a pedunculate Oak ( Quercus robur ) wood at Wytham, Oxford.
2. The frequency distribution of SLA of leaves falling through the late summer and autumn of 1993 changed with a greater contribution of leaves of high SLA earlier in the period. For the same SLA leaves falling earlier in the season had less negative δ¹³C and higher N content than leaves falling later.
3. The frequency distribution of SLA of litter leaves was similar in both years but litter leaves had lower N and less negative δ¹³C in 1996.
4. In the canopy, there was a highly significant negative relationship between height in the canopy and SLA in green and senescent leaves. The relationship between SLA and both δ¹³C ratio and leaf N was different between green, senescent and litter leaves.
5. The frequency distribution of foliage in the canopy estimated from the frequency of SLA in the litter and the vertical pattern of SLA in the canopy was similar to the distribution measured by the 'drop-line' method.     

Hot spots of leaf breakdown within a headwater stream reach: comparing breakdown rates among litter patch types with different macroinvertebrate assemblages

1. Detecting hot spots of litter decomposition will promote understanding of litter processing in a heterogeneous system. To identify hot spots of leaf breakdown within a headwater stream reach, we examined the difference in leaf breakdown rate among four types of litter patches, one that formed in riffles and three that formed in pools (middle, alcove, edge), in different seasons. 2. Middle patches showed the highest breakdown rate in some seasons; the rate in middle patches was 1.5–4 times higher than in the other patches. Thus middle patches can be regarded as hot spots of leaf breakdown in the study reach. This result contrasted with other studies showing higher breakdown rate in riffles than in pools. 3. Significant relationships between abundance of caddisfly shredders and breakdown rate were observed in seasons when the rate differed among patch types. Greater abundance of Lepidostoma seems to be responsible for middle patches being hot spots of leaf breakdown. 4. It is expected that when the proportion of leaves retained in middle patches within a reach is higher, the breakdown rate of the entire reach will be increased. Clarifying how the proportion of leaves retained on middle patches within a reach varies temporally and spatially would improve our understanding of leaf breakdown in headwater streams.     

The effect of 6-benzylaminopurine (benzyladenine) on senescence and chocolate spot (Botrytis fabae) of winter beans (Viciafaba)

In laboratory trials on detached bean leaves, benzyladenine delayed both senescence and aggressive attack of chocolate spot disease. In field plots, over 80% of the aggressive spots were on senescent leaves and the distribution of aggressively attacked leaves was predictable from the distribution of senescent leaves. The total frequency of aggressively attacked leaves per shoot was apparently related to the frequency of senescent spotted leaves. Although benzyladenine significantly reduced leaf senescence in field plots sprayed in May, the senescent leaves were more frequently attacked aggressively. Wider plant spacing (14 in instead of 8 in) similarly increased the ratio of aggressive attack to senescence, and the individual fields retained until June the differences in infection earlier established by frost damage.     

长白山源头溪流河床凋落物的分布特征及季节动态

在北方寒冷区,凋落物于秋季大量输入溪流,是水生生物越冬生存的关键.河床凋落物的堆积和组成会直接影响凋落叶分解等关键生态过程,但目前国内关于北方地区溪流河床凋落物分布特征的研究匮乏.在长白山地区一条源头溪流,采用原位取样的方法,探究了溪流河床凋落物的分布特征及季节动态.结果表明:深潭型凋落物斑块的堆积面积和水深显著大于浅...     

Green leaves enhance the growth and development of a stream macroinvertebrate shredder when senescent leaves are available

1. Freshly fallen green leaves and flowers of terrestrial plants enter temperate streams in spring and summer, when senescent leaf litter is often scarce. These resources appear to provide good supplementary food for macroinvertebrate shredders, but have some potential shortcomings as food or case material for caddisflies. 2. To compare suitability of green leaves or flowers and senescent leaves for the growth and development of stream shredders, we reared the caddisfly Lepidostoma complicatum in the laboratory with treatments that provided larvae with senescent (oak) and green (oak or maple) leaves separately, and also together, in case the combined use of both types of leaf may benefit the shredder. 3. Larvae supplied with green leaves alone grew at 65% of the rate of those provided with senescent leaves alone, due to their lower consumption rate. No individuals given green leaves alone developed into adults, whereas 70% of the individuals given senescent leaves alone did. Green leaves may inhibit larval consumption due to their high phenol content, or they may be unsuitable for case material because they are less tough than senescent leaves. 4. Larvae supplied with both senescent and green leaves (or flowers) had a higher growth rate and developed faster, than those given senescent leaves alone, whereas the proportions of successfully emerged individuals did not differ. Lepidostoma probably benefits from the higher nitrogen content of the green leaves when used together with senescent leaves. 5. These results suggest that green leaves (or flowers) cannot serve as an alternative food resource to senescent leaves, but that they can enhance the growth and development of a Lepidostoma stream shredder if senescent leaves are also available.     

Differences in litter characteristics and macroinvertebrate assemblages between litter patches in pools and riffles in a headwater stream

We compared the abundance of litter categories (coarse particulate organic matter 1–16 mm, leaves >16 mm, and small woody detritus 16–100 mm) and macroinvertebrate assemblages between natural litter patches in pools and riffles in a headwater stream. Litter patches in pools were formed under conditions of almost no current (<6 cm s⁻¹), but in riffles they were formed under variable current velocities (13–89 cm s⁻¹). Although the abundance of each litter category exhibited seasonal change, leaves were more abundant in riffles, and coarse particulate organic matter and small woody detritus were more abundant in pools throughout the study period. Macroinvertebrate assemblages in pools and riffles also changed seasonally but distinctly differed from each other. Shredders, collector-gatherers, and predators were the dominant functional groups in abundance in both pools and riffles, but the dominant shredders were caddisflies in pools and stoneflies in riffles. It is considered that the hydraulic conditions affected macroinvertebrate assemblages directly and indirectly through influences on the characteristics of litter retained in the patches. Our results suggest that the relative abundance of litter patches in pools and riffles largely affects the macroinvertebrate community structure of the headwater stream. Received: July 19, 2001 / Accepted: December 19, 2001     

Autumn leaf colours as indicators of decomposition rate in sycamore (Acer pseudoplatanus L.)

We tested the hypothesis that there is a causal connection between autumn colour, nutrient concentration and decomposibility of fresh leaf litter. Samples from patches of different autumn colours within the leaves of the deciduous tree sycamore (Acer pseudoplatanus) were sealed into litter bags and incubated for one winter in an outdoor leaf mould bed. Green leaf patches were decomposed faster than yellow or brown patches and this corresponded with the higher N and P concentrations in the former. Black patches, indicating colonisation by the tar spot fungus Rhytisma acerinum, were particularly high in P, but were decomposed very slowly, owing probably to resource immobilisation by the fungus. The results supported the hypothesis and were consistent with a previous study reporting an interspecific link between autumn coloration and decomposition rate. Autumn leaf colour of deciduous woody plants may serve as a useful predictor of litter decomposibility in ecosystem or biome scale studies where extensive direct measurements of litter chemistry and decomposition are not feasible. This revised version was published online in June 2006 with corrections to the Cover Date.     

Chemical composition and phytotoxicity of volatile essential oil from intact and fallen leaves of Eucalyptus citriodora

A total of 23 volatile constituents was identified and characterized by GC and GC-MS in the volatile essential oil extracted from intact (juvenile and adult) and fallen (senescent and leaf litter) leaves of lemon-scented eucalyptus (Eucalyptus citriodora Hook.). The leaves differed in their pigment, water and protein content, and C/N ratio. The oils were, in general, monoterpenoid in nature with 18 monoterpenes and 5 sesquiterpenes. However, a great variability in the amount of essential oils and their individual constituents was observed in different leaf tissues. The amount was maximum in the senescent leaves collected from the floor of the tree closely followed by that from juvenile leaves. In all, 19 constituents were identified in oil from juvenile and senescent leaves compared to 23 in adult leaves and 20 in leaf litter, respectively. Citronellal, a characteristic monoterpene of the oil reported hitherto was found to be more (77-78%) in the juvenile and senescent leaves compared to 48 and 54%, respectively, in the adult leaves and leaf litter. In the adult leaves, however, the content of citronellol--another important monoterpene-- was very high (21.9%) compared to other leaf types (7.8-12.2%). Essential oil and its two major monoterpenes viz. citronellal and citronellol were tested for their phytotoxicity against two weeds (Amaranthus viridis and Echinochloa crus-galli) and two crops (Triticum aestivum and Oryza sativa) under laboratory conditions. A difference in the phytotoxicity, measured in terms of seedling length and dry weight, of oil from different leaves and major monoterpenes was observed. Oil from adult leaves was found to be most phytotoxic although it occurs in smaller amount (on unit weight basis). The different toxicity of different oil types was due to the relative amount of individual monoterpenes present in the oil, their solubility and interactive action. The study concludes that oil from senescent and juvenile leaves being rich in citronellal could be used as commercial source of citronellal whereas that from adult leaves for weed management programmes as it was the most phytotoxic.     

Red young leaves have less mechanical defence than green young leaves

In many plants, leaves that are young and/or old (senescent) are not green. One adaptive hypothesis proposed that leaf color change could be a warning signal reducing insect attack. If leaf coloration involves less herbivory, it remains unclear why leaves in many species are constantly green. To examine whether green leaves reduce herbivory by physical defense as an alternative to the supposed warning signal of red leaves, we conducted comparative analyses of leaf color and protective tissues of 76 woody species in spring. The protective features (trichomes, enhanced cuticle and multiple epidermis) and the distribution of red pigments within leaves were examined in both young and mature leaves. We observed that redness was more frequent in young leaves than in senescent leaves. Compared to 36 species with red young leaves, 40 species with green young leaves showed a significantly higher incidence of enhanced cuticle and trichomes in both phylogenetic and non‐phylogenetic analyses. The phylogenetic analysis indicated that the multiple origins of mechanical protection were generally associated with loss of red coloration. Our finding of relatively poor mechanical protection in red young leaves provides additional evidence for the adaptive explanation of leaf color change.     

Effects of retention site on breakdown of organic matter in a mountain stream

相似文献   

Effects of burial on leaf litter quality,microbial conditioning and palatability to three shredder taxa

Summary 1. Heterotrophic microorganisms are crucial for mineralising leaf litter and rendering it more palatable to leaf‐shredding invertebrates. A substantial part of leaf litter entering running waters may be buried in the streambed and thus be exposed to the constraining conditions prevailing in the hyporheic zone. The fate of this buried organic matter and particularly the role of microbial conditioning in this habitat remain largely unexplored. 2. The aim of this study was to determine how the location of leaf litter within the streambed (i.e. at the surface or buried), as well as the leaf litter burial history, may affect the leaf‐associated aquatic hyphomycete communities and therefore leaf consumption by invertebrate detritivores. We tested the hypotheses that (i) burial of leaf litter would result in lower decomposition rates associated with changes in microbial assemblages compared with leaf litter at the surface and (ii) altered microbial conditioning of buried leaf litter would lead to decreased quality and palatability to their consumers, translating into lower growth rates of detritivores. 3. These hypotheses were tested experimentally in a second‐order stream where leaf‐associated microbial communities, as well as leaf litter decomposition rates, elemental composition and toughness, were compared across controlled treatments differing by their location within the streambed. We examined the effects of the diverse conditioning treatments on decaying leaf palatability to consumers through feeding trials on three shredder taxa including a freshwater amphipod, of which we also determined the growth rate. 4. Microbial leaf litter decomposition, fungal biomass and sporulation rates were reduced when leaf litter was buried in the hyporheic zone. While the total species richness of fungal assemblages was similar among treatments, the composition of fungal assemblages was affected by leaf litter burial in sediment. 5. Leaf litter burial markedly affected the food quality (especially P content) of leaf material, probably due to the changes in microbial conditioning. Leaf litter palatability to shredders was highest for leaves exposed at the sediment surface and tended to be negatively related to leaf litter toughness and C/P ratio. In addition, burial of leaf litter led to lower amphipod growth rates, which were positively correlated with leaf litter P content. 6. These results emphasise the importance of leaf colonisation by aquatic fungi in the hyporheic zone of headwater streams, where fungal conditioning of leaf litter appears particularly critical for nutrient and energy transfer to higher trophic levels.     

Factors affecting litter processing by Anisocentropus kirramus (Trichoptera: Calamoceratidae) from an Australian tropical rainforest stream

• 1 Larvae of the caddisfly Anisocentropus kirramus are common leaf shredders in rainforest streams in tropical Queensland. Laboratory experiments were undertaken to examine the effects of (a) differences in leaf species, age and degree of conditioning, (b) leaf preferences, (c) temperature and (d) larval stage on processing of leaf litter by A. kirramus.
• 2 Leaf species (five tested), age (fresh v senescent) and condition (conditioned in stream v unconditioned) each had a significant effect on the amount of leaf material that was processed. For most species, conditioned leaves were processed faster but for one species unconditioned leaves were processed faster. Senescent leaves were processed more rapidly than green leaves in three species but not in the other two. Given a choice of leaf types A. kirramus actively selected leaves that were processed faster; no preference was shown between two different ‘fast’ leaves.
• 3 Processing occurred at all temperatures tested (10, 18 and 25°C), with the processing rate increasing with temperature. Younger instar larvae processed leaves at a greater rate per unit body weight (up to 343% day^?1) than older instars. Final instar (5) larvae were capable of processing some tough leaves that younger instars did not process.
• 4 In rainforest streams, processing of leaves by A. kirramus takes place throughout the year. Its ability to process green leaves is important because of the high input of fresh green leaves into tropical streams, and because of the severe depletion of the supply of conditioned leaves and fine detritus after floods.

     

Decomposition of litter from three major plant species of jarrah (Eucalyptus marginata Donn ex Sm.) forest in relation to site fire history and soil type

Rates of weight loss and release of N, P, K, S, Ca, Mg, Na, and Cl from litter of several species in jarrah (E. marginata Donn ex Sm.) forest were measured in relation to site fire history and soil type. Weight loss from leaf litter decreased in the order jarrah > marri (E. calophylla R. Br. ex Lindl.) > Banksia grandis Willd. After 18 months on the forest floor senesced leaves of jarrah, marri and Banksia had lost 45%, 42% and 19%, respectively, of their original weight. Although greatest rates of decomposition occurred on a site burnt 3 y previously by an intense autumn fire and slowest rates on a site which had not been burnt for 8 y, the differences between burn sites were small in comparison with the total weight loss from decomposing litter. The order of release of nutrients from decomposing eucalypt litter was P<N<Ca<S<Mg<Cl<K<Na. There appears to be only slow release of N and P from the litter layer of these forests in the period between successive control burns. Fresh jarrah leaves, which are similar in chemical composition to leaf litter falling after crowns have been scorched by intense fire, decompose rapidly in comparison with senescent leaf tissue. Release of nutrients, particularly N and P, is also more rapid from fresh leaves than from leaf litter. Rates of decomposition of green leaves differed between soil types in the order reddish gravels > dark sandy duplex soil > yellow gravels. These differences may be related to the higher nutrient status of the reddish gravel soils.     

Response of invertebrates to lotic disturbance: a test of the hyporheic refuge hypothesis

Summary Recovery following hydrological disturbances is usually rapid for lotic invertebrates. Stream ecologists have assumed that recovery is facilitated by behavioral migrations during floods down into the hyporheic zone (the interstitial spaces of a streambed) to seek temporary refuge from possible erosion (the hyporheic refuge hypothesis). We provide the first explicit test of this hypothesis by evaluating three predictions of the hypothesis. We coupled field observations of the response of meiofaunal invertebrates to floods with field and flume experiments. The study site was a sandy-bottom stream in northern Virginia. Prediction 1, that loss of fauna from a streambed during floods should be minimal as long as the depth of scour in the streambed is less than the depth of the hyporheic zone, was not supported for any taxon. For two floods which varied considerably in magnitude, 50–90% of the fauna was lost from the bed despite the fact that the depth of scour (10–30 cm) was significantly less than the total depth of the hyporheic zone (50 cm). Prediction 2, that fauna should move deeper into the bed at higher flows, was supported by field observations during only one of two floods and then only for rotifers. In flume experiments that tested for finer scale behavioral movements, significant vertical migrations were found for copepods and chironomids which moved 1.5–3.5 cm downward as mean velocity (3 cm off bottom) was increased from 5–23 cm/s. Movements down by rotifers were not found in the flume experiments. Prediction 3, that the hyporheic zone is the most important source of colonists to defaunated areas, was supported in part by field experiments. The hyporheic route was not the primary route for any taxon but it was as important for the rotifers and copepods as water column or streambed surface routes. We conclude that, even though smallscale (cm's) migrations into the streambed in response to increased flow may be observed for some taxa and the hyporheic zone may serve as a partial source of colonists following disturbances, movements down are not adequate in preventing significant losses of meiofauna during floods.     

Green and senescent leaf phenolics showed spatial autocorrelation in a Quercus robur population in northwestern Spain

Intraspecific variation in polyphenols may be important both for the resistance of plants to herbivory and for nutrient cycling in terrestrial ecosystems. The spatial pattern and scale of polyphenol concentration in natural populations are practically unknown, despite multiple evidence that resources influencing leaf phenolics (such as light or nutrient availability) show spatial dependence at the scale of meters to tens of meters. By using geostatistical methods, we evaluated the spatial variability in polyphenol concentration in green and senescent leaves in a pedunculate oak (Quercus robur L.) population of 125 individuals. The spatial pattern of light and nutrient availability for plants was also evaluated. Leaf polyphenols were more variable than other leaf properties, and green polyphenols were significantly correlated with radiation, leaf mass per unit area and leaf N concentration. Both green and senescent leaf polyphenol concentration showed spatial dependence at distances below 6 to 10 m, with structural (spatial) variance explaining between 50 and 80% of the total variance. The variance not explained by space was assumed to be due to individual genotypic variability. Stochastic maps of leaf phenolics in the area showed that the probability of finding high phenolics level was not random. These findings may be important to understand the spatial heterogeneity of plant-herbivore interactions, leaf litter decomposition and mineralization rate.     

Leaf litter quality affects aquatic insect emergence: contrasting patterns from two foundation trees

Reciprocal subsidies between rivers and terrestrial habitats are common where terrestrial leaf litter provides energy to aquatic invertebrates while emerging aquatic insects provide energy to terrestrial predators (e.g., birds, lizards, spiders). We examined how aquatic insect emergence changed seasonally with litter from two foundation riparian trees, whose litter often dominates riparian streams of the southwestern United States: Fremont (Populus fremontii) and narrowleaf (Populus angustifolia) cottonwood. P. fremontii litter is fast-decomposing and lower in defensive phytochemicals (i.e., condensed tannins, lignin) relative to P. angustifolia. We experimentally manipulated leaf litter from these two species by placing them in leaf enclosures with emergence traps attached in order to determine how leaf type influenced insect emergence. Contrary to our initial predictions, we found that packs with slow-decomposing leaves tended to support more emergent insects relative to packs with fast-decomposing leaves. Three findings emerged. Firstly, abundance (number of emerging insects m^?2 day^?1) was 25 % higher on narrowleaf compared to Fremont leaves for the spring but did not differ in the fall, demonstrating that leaf quality from two dominant trees of the same genus yielded different emergence patterns and that these patterns changed seasonally. Secondly, functional feeding groups of emerging insects differed between treatments and seasons. Specifically, in the spring collector-gatherer abundance and biomass were higher on narrowleaf leaves, whereas collector-filterer abundance and biomass were higher on Fremont leaves. Shredder abundance and biomass were higher on narrowleaf leaves in the fall. Thirdly, diversity (Shannon’s H′) was higher on Fremont leaves in the spring, but no differences were found in the fall, showing that fast-decomposing leaves can support a more diverse, complex emergent insect assemblage during certain times of the year. Collectively, these results challenge the notion that leaf quality is a simple function of decomposition, suggesting instead that aquatic insects benefit differentially from different leaf types, such that some use slow-decomposing litter for habitat and its temporal longevity and others utilize fast-decomposing litter with more immediate nutrient release.     

The effect of flooding and flood timing on leaf litter breakdown rates and nutrient dynamics in a river red gum (Eucalyptus camaldulensis) forest

Comparisons of litter standing-stocks in low-lying and higher areas of the floodplain and the effects of controlled flooding events on leaf litter decomposition and leaf litter nutrients were examined during autumn and winter in a southeastern Australian river red gum (Eucalyptus camaldulensis) floodplain forest. The mean mass of total litter and some litter components was significantly greater in autumn than in winter but there were few differences in litter mass between low-lying flood runners and higher sites (1.5 m) on the floodplain, regardless of season. Leaf decomposition was more rapid in flooded areas than in non-flooded areas and was significantly faster in autumn than in winter. In flooded leaves, concentrations of phosphorus and nitrogen dropped rapidly during the first 3 days of each experiment, increased to near original after 7–10 weeks and then decreased again. After 112 days of decomposition the C:N:P ratios of leaf litter increased, but this effect was most marked for flooded leaves. Simple models of leaf litter dynamics indicated that leaf litter standing-stocks in low-lying flood runners would be reduced by flooding, particularly during autumn. In contrast, models predicted a net gain in standing-stocks of leaf litter to be higher on the floodplain, particularly in autumn. Alteration to the seasonal timing of floods by river regulation has probably decreased litter standing-stocks and nutrients available in low-lying areas of the floodplain to support the production of macrophytes and biofilms during winter and spring floods.