Higher Order Streams Research Articles

Conservation of otter (Lutra lutra) in a Mediterranean area: the importance of habitat quality and temporal variation in water availability

Abstract1. This study analyses quantitatively the association between habitat characteristics and annual variation in water availability on otter (Lutra lutra L.) distribution in a large Mediterranean area (13 717 km2) in southern Spain.2. There was a strong positive correlation between habitat quality, estimated after the two first components of a principal component analysis of a matrix of habitat variables×sites, and an index of otter presence/absence, otter presence being most commonly associated with unpolluted and undisturbed sites surrounded by woodland.3. After univariate analysis, the main freshwater habitat features were statistically different between sites with otters and sites lacking them. In general, otters occurred in medium‐sized fluvial habitats, with high bankside vegetation cover, unpolluted, with low or very low human disturbance and surrounded by forests or dehesas (Mediterranean‐like savannah).4. Water availability, inferred from the water balance, strongly influenced otter distribution and the type of water body used the most. As water availability increased so did the otter distribution range and during these periods they tended to colonize low‐order streams. During periods of water shortage, otters were usually found in high‐order streams and reservoirs, the only sites that kept water.5. This work shows that otter distribution in Mediterranean areas is strongly influenced by two factors — man‐induced habitat degradation and natural variability in water availability. The addition of the second factor to the first one introduces additional risks to the survival of otter populations in Mediterranean areas compared with more northerly and humid areas. Copyright © 2001 John Wiley & Sons, Ltd.

Environmental influences on seedling growth of cottonwood species following a major flood

A major flood in June 1995 along most streams in southern Alberta (AB) and southeastern British Columbia (BC), Canada, permitted the comparison of natural seedling establishment of different cottonwood species across different environments. Nine study sites were established along a 340-km corridor from BC, over the Rocky Mountain Continental Divide, and onto the foothills and then prairies of AB. Four native cottonwood species occurred from BC to AB: the black cottonwood, Populus trichocarpa Torr. and Gray, balsam poplar, P. balsamifera L., and narrow-leaf cottonwood, P. angustifolia James, of section Tacamahaca, and the prairie cottonwood, Populus deltoides Bartr. ex Marsh, of section Aigeiros. Cottonwood seedlings of the 1995 cohort were monitored from 1995 through 1998 in quadrats along riparian transects. The study confirmed that a major flood enabled extensive cottonwood recruitment along mountain, foothills, and prairie river reaches and revealed that both, environment and species influenced cottonwood seedling growth in situ. Across the sites, seedling heights varied ten-fold and were closely negatively correlated with site elevation (1995: n = 9, r 2 = 0.93; 1997: n = 4, r 2 = 0.92). The increased growth at lower elevations was probably associated with warmer temperatures and a longer growing season; seedling growth was positively correlated with the accumulation of growing degree days of nearby weather stations (1995: n = 6, r 2 = 0.88). Growth rate was also influenced by species as height varied up to three-fold across species at sites where species co-occurred. P. deltoides seedlings grew fastest followed by P. trichocarpa/P. balsamifera and intersectional hybrids and, finally, P. angustifolia and intrasectional hybrids. The superior seedling growth of P. deltoides is consistent with its life history and distribution. P. deltoides occurs along prairie river reaches with warmer and drier climates and higher-order streams with finer substrate textures; these physical conditions would favor seedling recruitment. The alternate environmental conditions of foothillls and mountain regions probably encourage clonal (asexual) recruitment that may supplement seedling recruitment, particularly for the Tacamahaca species.

Estimation of exposure criteria values for biliary polycyclic aromatic hydrocarbon metabolite concentrations in white suckers (Catostomus commersoni)

A methodology was developed for deriving quantitative exposure criteria useful for comparing a site or a watershed to a reference condition and for defining the occurrence of extreme exposures. The prototype method used indicators of exposures to oil contamination and combustion by-products, naphthalene (NAPH)-type and benzo[a]pyrene (BaP)-type biliary metabolites from white suckers (Catostomus commersoni). The fish were collected from first- to third-order streams in the Regional Environmental Monitoring Assessment Program (REMAP) study of the Eastern Corn Belt Plains ecoregion at randomly selected sites and from third- or higher-order streams in conjunction with the Ohio Environmental Protection Agency (OEPA) Biomonitoring Program, which sampled sites of concern and reference sites selected by ecologists. The polycyclic aromatic hydrocarbon metabolites were measured using fixed-wavelength fluorescence with excitation/emission pairs at 290/335 nm for NAPH-type and 380/430 nm for BaP-type metabolites. Exposure criteria values were selected for each type of metabolite for both the REMAP study and the OEPA reference sites following the recommendations of the International Federation of Clinical Chemistry. Exposure criteria derived for each metabolite were not significantly different for REMAP and OEPA reference sites. More than one half of the OEPA nonreference sites were found to be contaminated, exceeding criteria values for both types of metabolites. This method for finding meaningful exposure criteria can be used to develop criteria of exposure to other contaminants for other wildlife and other ecosystems.

Effects of forest management practices on mid-Atlantic streams.

Agricultural and urban land use activities have affected stream ecosystems throughout the mid-Atlantic region. However, over 60% of the mid-Atlantic region is forested. A study was conducted to investigate the effects of management practices on forested stream ecosystems throughout the mid-Atlantic region. The study consisted of two phases: Phase 1 was a literature synthesis of information available on the effects of forest management practices on stream hydrology, erosion and sedimentation, riparian habitat alteration, chemical addition, and change in biotic diversity in the mid-Atlantic region. In Phase 2, data from mid-Atlantic streams were analyzed to assess the effects of forest land use on stream quality at the regional scale. Typically, it is the larger order streams in which monitoring and assessment occurs—3rd order or higher streams. The impacts of forest management practices, particularly hydrologic modifications and riparian buffer zone alteration, occur predominantly in first and second order streams with cumulative impacts translating to higher order streams. Based on the literature review and mid-Atlantic Highland streams analysis, there are short-term (e.g., 2 to 5 years) effects of forest management practices on stream quality at local scales. However, signatures of cumulative effects from forest management practices are not apparent at regional scales in the Highlands. In general, forested land use is associated with good stream quality in the region compared with other land use practices.

The Effect of Storage on Runoff from a Headwater Subarctic Shield Basin

Canadian Shield runoff production processes have been investigated, but research is needed beyond the hillslope scale to determine the influences on the volume and timing of runoff from headwater basins to higher-order streams. Such research will permit an understanding of the magnitude of climate change impacts at the watershed scale, where changes in the hydrologic regime are felt most by water resource users. To this end, water budget terms were measured over different portions of a headwater lake basin north of Yellowknife, Northwest Territories, during the spring snowmelt of 1998 to determine the relative importance of each component as water moved through the basin. Evaporation made up 16% and runoff 70% of the snowmelt from upland areas. Upland ponding of water during the melt increased the evaporative loss. Headwater lake storage deficits are very important in the timing and volume of runoff that is transmitted downstream: 53% of basin meltwater went to meeting the storage deficit in the lake so that only 7% of the meltwater was routed out of the basin. These results imply that topology of shield headwater basins, notably the location of lakes and antecedent water levels in lakes, are important in determining the runoff response. As storage deficits in surface waters will cause changes over space in runoff response and timing, it may be inappropriate to consider the snowpack or rainfall volume in isolation when forecasting runoff volume. These results are important for water management and hydrological modeling applications of northern shield basins.

Comparative Demography of Clemmys marmorata Populations in the Trinity River of California in the Context of Dam-Induced Alterations

Western pond turtle (Clemmys marmorata) populations were examined on two forks of the Trinity River in northern Califoria, one of which is subject to the impacts of damming and one of which is free-flowing. Mark-recapture was conducted over a three-year period for each population, and the Jolly- Seber model was used to estimate population size. Size/age structures were compared with reference to potential impacts of damming on this species. The density of populations (turtles/ha) on the dammed fork (mainstem Trinity) did not differ significantly from that on the undammed fork (south fork Trinity). How- ever, both populations were much sparser than western pond turtle populations elsewhere. Higher-order streams, in general, do not appear to support high densities of western pond turtles, possibly because of the sparse distribution of resources. The mainstem population had an age structure dominated by adults relative to a population with a known stable age distribution, suggesting that damming of the mainstem Trinity may have negatively impacted juvenile turtles. Populations of western pond turtles (Clemmys marmorata) have experienced declines through-

Dynamics of Wood Transport in Streams: A Flume Experiment

The influence of woody debris on channel morphology and aquatic habitat has been recognized for many years. Unlike sediment, however, little is known about how wood moves through river systems. We examined some dynamics of wood transport in streams through a series of flume experiments and observed three distinct wood transport regimes: uncongested, congested and semi-congested. During uncongested transport, logs move without piece-to-piece interactions and generally occupy less than 10 per cent of the channel area. In congested transport, the logs move together as a single mass and occupy more than 33 per cent of the channel area. Semi-congested transport is intermediate between these two transport regimes. The type of transport regime was most sensitive to changes in a dimensionless input rate, defined as the ratio of log volume delivered to the channel per second (Qlog) to discharge (QW); this ratio varied between 0·015 for uncongested transport and 0·20 for congested transport. Depositional fabrics within stable log jams varied by transport type, with deposits derived from uncongested and semi-congested transport regimes having a higher proportion of pieces orientated normal to flow than those from congested transport. Because wood input rates are higher and channel dimensions decrease relative to piece size in low-order channels, we expect congested transport will be more common in low-order streams while uncongested transport will dominate higher-order streams. Single flotation models can be used to model the stability of individual pieces, especially in higher-order channels, but are insufficient for modelling the more complex intractions that occur in lower-order streams. © 1997 John Wiley & Sons, Ltd.

Dynamics of Wood Transport in Streams: A Flume Experiment

The influence of woody debris on channel morphology and aquatic habitat has been recognized for many years. Unlike sediment, however, little is known about how wood moves through river systems. We examined some dynamics of wood transport in streams through a series of flume experiments and observed three distinct wood transport regimes: uncongested, congested and semi-congested. During uncongested transport, logs move without piece-to-piece interactions and generally occupy less than 10 per cent of the channel area. In congested transport, the logs move together as a single mass and occupy more than 33 per cent of the channel area. Semi-congested transport is intermediate between these two transport regimes. The type of transport regime was most sensitive to changes in a dimensionless input rate, defined as the ratio of log volume delivered to the channel per second (Qlog) to discharge (QW); this ratio varied between 0·015 for uncongested transport and 0·20 for congested transport. Depositional fabrics within stable log jams varied by transport type, with deposits derived from uncongested and semi-congested transport regimes having a higher proportion of pieces orientated normal to flow than those from congested transport. Because wood input rates are higher and channel dimensions decrease relative to piece size in low-order channels, we expect congested transport will be more common in low-order streams while uncongested transport will dominate higher-order streams. Single flotation models can be used to model the stability of individual pieces, especially in higher-order channels, but are insufficient for modelling the more complex intractions that occur in lower-order streams. © 1997 John Wiley & Sons, Ltd.

Sedimentary responses to sub-aerial felsic volcanism from the Late Devonian-Early Carboniferous northern Macalister Synclinorium, southeastern Australia

Active sub-aerial volcanism has the capability to rapidly alter both the topographic and drainage characteristics of a landscape, and thus fundamentally influence resulting sedimentary facies. Relationships between sedimentation and volcanism are explored in this paper, via an investigation of the stratigraphy and early depositional history of the Upper Devonian volcano-sedimentary units of the northern Macalister Synclinorium, east-central Victoria. Complex interfingering relationships exist between sub-aerial felsic volcanic successions (Rose River Volcanics) and alluvial, fluvial and lacustrine sedimentary units (the Bindaree and Howitt Spur Formations). A depositional model is presented for these units, in which the Rose River Volcanics, an outflow (ignimbritic plateau) facies of the Tolmie caldera complex to the north, co-existed with a series of volcaniclastic alluvial fans and freshwater lakes. A dacitic volcanic centre (Refrigerator Gap Dacite) appears to have developed somewhat separately to the south, in the Jamieson River area. These lower successions of the northern Macalister Synclinorium record an episode of sedimentation in close proximity to an active felsic volcanic terrain, and in a landscape of significant topographic relief. The conglomerates of the Bindaree Formation, in particular, record the supply of abundant felsic volcanic detritus via high-gradient proximal alluvial stream systems. Comparisons can be made with sedimentological processes operating during deposition of overlying sedimentary successions, where influences by primary volcanic activity were less important. Well defined channel geometry successions and a diversification in clast and grain types within these younger units indicate ongoing headward erosion into metasedimentary basement rocks, and deposition within lower-gradient, higher-order stream systems.

Incidences de facteurs géomorphologiques dans le fonctionnement d'un bassin en situation de crue : l'exemple de l'Agly (Pyrénées-Orientales)/Relationships between géomorphologic features and hydrological processes in a catchment area during storm events: the case of the Agly river catchment (Pyrénées-Orientales)

Abstract Relationships between geomorphic structures and hydrological processes during storms are studied in the Agly catchment (1 045 km2). This basin is located in the eastern part of the French Pyrenees. The channel network is influenced by structural landforms. High order streams are oriented according to the main east-west trend of faulting and folding. These tectonic features are inherited from the pyrenean uplift partly eroded, and then rejuvenated during the mio-pliocene period. This context, combined with the local mediterranean climate, leads to heavy flood occurrences especially during storm events. Two approaches are followed : simulation of stream flow on gauged basins and analysis of the evolution of floods hydro- graphs along the main river, during a heavy storm. Morphometric characteristics of several sub-catchments are used for the study of their effects on runoff and river flows. Simulations of river flows from Digital Elevation Models (D.EM.) are compared with observed storm flows. Results of the simulation are accurated for time to peak flow. The general shape of the simulated hydrograph reflects the geomorphometric characteristics of the catchment basin. The flood associated with a storm lasting about 3 hours on September 26 , 1992 is depicted at five gauged stations along the main stream of the Agly river. The speed of the flood wave for this storm was quite fast (3 meters/sec.) and this justifies the flood warning equipment of this catchment. More comprehensive research must be conducted to emphasize hydrological processes understanding via geomorphological approach. Global understanding of how catchment basins work is required to face such natural hazard as floods. Geomorphologists have to be involved in this challenge.

Riparian Water Quality Buffers: Estimates of Effectiveness and Minimum Width in an Agricultural Landscape, Western Tennessee

The adverse effects of agricultural non-point source pollution (e.g., sediment, nitrogen, phosphorous) on water quality are widely recognized and documented. Riparian vegetation, however, can help to filter out these pollutants by detaining or processing the pollutants. In this study, we estimate the effectiveness and adequacy of riparian buffers in an agricultural landscape in Fayette County, Tennessee, through the use of a mathematical model—the Riparian Buffer Delineation Equation (RBDE)—and a geographic information system (GIS). Based on edaphic characteristics, slope, and surface roughness, the minimum buffer width needed to filter 90% of nitrates from surface flow was estimated using the RBDE for 18 soil types along five reaches of the Wolf River and in a tributary basin, Hurricane Creek. The minimum buffer widths suggested by the model ranged from 2.3 m to 64.3 m. We then digitized existing land cover, soil type, and hydrography into a GIS and compared the minimum estimated buffer widths with the widths of existing buffers in the study area. Results from the GIS buffer analysis indicated areas where actual buffer widths were less than the proposed minimum buffer widths. Lower-order streams were generally less adequately buffered than higher-order streams, yet previous studies have suggested that more pollutants may enter streams through low-order intermittent and perennial streams. This finding suggests that the Wolf River may be more susceptible to agriculturally derived inputs of sediment and nutrients from tributary streams than from runoff directly into the main stem.

Fish community variation along physical gradients in short streams of northern Spain and the disruptive effect of dams

Relationships between the fish community and selected habitat features were examined in a set of short temperate streams located at the northern end of the Iberian Peninsula. The fish fauna in these streams consists mostly of diadromous or estuarine species. Species richness and diversity increased with stream order, depth and width and decreased with elevation and distance from the sea. Stream order (positively) and elevation (negatively) were the two variables most highly correlated with species richness and diversity; higher order streams (order 3–4) showed greater values of species diversity than lower order ones (order 1) even when the elevation effect was removed. Addition of species in the downstream direction, but no replacement or loss was evidenced. We also compared the observed values of species diversity with those predicted from habitat features for a set of locations above unpassable dams. A great majority of the sites showed lower than predicted diversity values, which is an expected outcome for this mainly migratory fish fauna.

Fine-scale biogeographic differences in the crustacean fauna of a cave system in West Virginia, USA

We examined the distribution patterns of four amphipod and one isopod crustaceans in 23 cave stream segments within a subterranean drainage basin. The patterns indicate that invasion history is a strong determinant of the species' distributions. One set of species likely has invaded the cave system from the zone of interstitial water, and thus is distributed mainly in headwater streams throughout the drainage. One species probably invaded through upstream migration from the resurgence of the cave system, and thus is strongly associated with the larger, higher-order streams. The effect of interspecific interactions on the patterns of distribution is not apparent at our scale of analysis.

Changes in the functioning of wetlands along environmental gradients

One of the prevalent gradients in wetlands is the continuum of depth and frequency of flooding. While much emphasis has been placed on the importance of hydrology as a driving force for wetlands, few other perspectives have emerged to demonstrate unifying patterns and principles. In contrast to the wetness continuum, the functioning of wetlands can be separated into two broad categories: (1) landscape-based transitions that occur within a wetland or group of similar wetland types and (2) resource-based transitions that allow comparisons of the flow of water and processing of nutrients among very different wetland types. Landscape-based continua include the transition from upstream to downstream in riverine wetlands and between aquatic and terrestrial ecosystems within a wetland. Along the upstream-downstream continuum, sources of flood-water delivery change dominance from ground-water discharge and overland runoff, as in low order streams, to dominance by overbank flooding, as in high order streams. With increasing size, properties related to the aquatic-to-terrestrial transition are replaced by properties related to wetland-atmospheric exchanges and by landscape maintenance, the latter not normally acknowledged as a wetland function. Resource-based continua include the extremes of (1) sources of water to wetlands (precipitation, overland flow, and ground water) and (2) the variation in inflows and outflows of nutrients and sediments. Emphasis on water source forces consideration of controls beyond the wetland’s boundaries. A broader view of biogeochemical functioning is gained by categorizing wetlands into groups based on the exchange of nutrients and sediments among landscape units rather than on serving as a sink or source for a particular element. Based on this analysis, the less frequently flooded or saturated portions of wetlands are no less functionally active than wetter portions; the functions are simply different. Efforts to classify wetlands according to their hydroperiod do little to reveal their fundamental properties.

Modelling operating system structures by timed stream processing functions

AbstractSome extensions of the basic formalism of stream processing functions are useful to specify complex structures such as operating systems. In this paper we give the foundations of higher order stream processing functions. These are functions which send and accept not only messages representing atomic data, but also complex elements such as functions. Some special notations are introduced for the specification and manipulation of such functions. A representation of time is outlined, which enables us to model time dependent behaviour. Finally, we demonstrate how characteristic operating system structures can be modelled by timed higher order stream processing functions.

Use of stream order and biological indices to assess water quality in the Osage and Black river basins of Missouri

Significant variations in four biological measures of water quality with stream order and river basin were demonstrated for streams of the Black and Osage river basins of Missouri. Water quality criteria specific for each order and basin were then developed. Benthic macroinvertebrates from springs and stream orders 3–8 in the two river basins were sampled quarterly for one year with riffle nets and artificial substrate samplers. A total of 548 samples were taken at 137 stations. The average annual macroinvertebrate density, index of diversity, number of taxa, and number of mayfly and stonefly taxa were determined for each station. These measures showed significant differences (p < 0.05) across stream order within and between the two river basins. Total taxa, total mayfly and stonefly taxa, and diversity were highest in orders 4 and 5 with decreased values in lower and higher stream orders. Maximum organism densities occurred in intermediate order streams. These differences were attributed to the succession of physical changes from headwaters to mouth within each river and to the unique geomorphology of each catchment basin. Water quality criteria based on three of the four measures described above (with 95% confidence limits) were established for each stream order in each river basin. Criteria for the Osage River basin were then used to identify three streams in the basin affected by environmental disturbances (stream impoundment, channelization and sewage discharge). The use of order- and basin-specific criteria assures that the biological differences between streams caused by environmental disturbance can be distinguished from the natural biological differences between streams of different orders and drainages.

Variation in Species Composition of Diatom Assemblages in an Unpolluted River.

Spatial variation of attached diatom assemblages was surveyed in an unpolluted river, the River Saguri in Niigata Prefecture, Japan, and the causes of the observed variation were analyzed. The result of cluster analysis indicated that the spatial variation of diatom assemblages was correlated to stream order and current velocity. Relative abundance of Fragilaria vaucheriae (Kutz.) Peters, Gomphonema angustum Ag. and Rhoicosphenia abbreviata (Ag.) Lange-B. in the assemblages collected from the first-order streams was less than that in the assemblages collected from the higher-order streams. In the first-order streams, a group of diatoms such as Gomphonema gracile Ehr., G. hebridense Greg., and Navicula angusta Grun. occurred when the current velocity was quite slow. There was little variation in the species composition between different substrata.

Monitoring the condition of the Canadian forest environment: The relevance of the concept of ?ecological indicators?

The Canadian forest environment is characterized by high spatial and temporal variability, especially in the west. Our forests vary according to climate, landform, and surficial geology, and according to the type, intensity, extent of, and the time since the last disturbance. Most Canadian forests have had a history of repeated acute, episodic disturbance from fire, insects, wind, diseases and/or logging, with a frequency of disturbance varying from a few decades to many centuries. These sources of variability have resulted in a complex and continually changing mosaic of forest conditions and stages of successional development.Monitoring the 'quality' of this dynamic forested landscape mosaic is extremely difficult, and in most cases the concept of a relatively simple index of forest ecosystem quality or condition (i.e. an 'ecological indicator') is probably inappropriate. Such ecological indicators are better suited for monitoring chronic anthropogenically induced disturbances that are continuous in their effect (e.g. 'acid rain', heavy metal pollution, air pollution, and the 'greenhouse effect') in ecosystems that, in the absence of such chronic disturbance, exhibit very slow directional change (e.g. lakes, higher order streams and rivers). Monitoring the effects of a chronic anthropogenic disturbance to forest ecosystems to determine if it is resulting in a sustained, directional alteration of environmental 'quality' will require a definition of the expected pattern of episodic disturbance and recovery therefrom (i.e. patterns of secondary succession in the absence of the chronic disturbance). Only when we have such a 'temporal fingerprint' of forest ecosystem condition for 'normal' patterns of disturbance and recovery can we determine if the ecosystem condition is being degraded by chronic human-induced alteration of the environment. Thus, degradation is assessed in terms of deviations from the expected temporal pattern of conditions rather than in terms of an instantaneous assessment of any particular condition. The concept of 'ecological rotation' (the time for a given ecosystem to recover from a given disturbance back to some defined successional condition) is useful in the definition of these 'temporal fingerprints'. This requires information on the intensity of disturbance, the frequency of disturbance, and the rate of successional recovery. Only when all three of these are known or estimated can statements be made as to whether the ecosystem is in a longterm sustainable condition or not.The somewhat overwhelming complexity of this task has led forest ecologists to use ecosystem-level computer simulation models. Appropriately structured and calibrated models of this type can provide predictions of the overall temporal patterns of ecosystem structure and functions that can be expected to accompany a given frequency and character of episodic disturbance. Such models can also be used to examine the long-term consequences of chronic disturbances such as acid rain and climatic change. Predictive ecosystem-level models should be used in conjunction with some method of stratifying the inherent spatial biophysical variability of the forest environment, such as the biogeoclimatic classification system of British Columbia.

Ecosystem Metabolism and Turnover of Organic Carbon along a Blackwater River Continuum

Variation in ecosystem metabolism and organic carbon turnover length was examined in the Ogeechee River (Georgia, USA) basin, a sixth—order blackwater river, to determine if the patterns observed in other rivers are also seen along a river continuum characterized by warm water temperatures, low stream gradient, extensive riparian swamps, and tea—colored water. The study was in two parts: an intensive study of seasonal variation in metabolism and organic carbon turnover length at a fourth—order site, which can be compared with an earlier study at a sixth—order site, and a study of the variation in system metabolism at five second— through sixth—order sites along the continuum. Metabolism varied seasonally at the fourth—order site that was studied most intensively; mean gross primary production (GPP, measured as O2) was 0.8 g°m—2°d—1 and mean community respiration (CR, also measured as O2) was 4.1 g°m—2°d—1. The community production/respiration ratio (P/R) ranged from 0.02 to 0.4 with an annual mean of 0.2. This is clearly a heterotrophic stream reach, where high rates of community respiration are supported by inputs of organic matter from extensive riparian swamps. Mean annual organ carbon turnover length at the fourth—order site was 570 km, although on most dates the length was much less, with a median of 114 km. Turnover lengths are significantly less than those at the sixth—order site, but much greater than observed in other rivers of similar size, probably because most of the organic carbon export in this river is dissolved organic carbon (DOC), which is less efficiently retained. Seasonal variation in turnover length reflected variation in transport. At five sites along the Ogeechee River continuum ranging from second to sixth order, GPP, CR, and P/R increased with stream order, as observed in other rivers; yet P/R was always &lt;1, and net daily metabolism was always negative and became more negative with increasing stream order. In contrast to other river continua studied, metabolism in this low—gradient blackwater river becomes more dependent on allochthonous organic carbon with increasing stream size despite downstream increases in GPP; floodplains and riparian swamps appear to be the source of allochthonous carbon. Riparian influences increase rather than decrease along the continuum in this river. Although first—through third—order streams represent 32% of the channel area in the Sixth—order basin, only 4% of basin—wide GPP and 9% of basin—wide CR occurs in them. Higher order streams are less efficient in organic carbon processing (longer turnover lengths), but are nonetheless responsible for most of the metabolic activity in this river basin.

Fish species richness and stream order in Washington State streams

We sampled fish at pairs of sites of the same stream order on opposite sides of drainage divides in the Cascade Mountains and in the southwest portion of Washington state. Elevation, gradient, drainage area, and stream order were significantly correlated with number of fish species collected at a site. Elevation accounted for the greatest portion of the variation in number of species and stream order for the least, but in low gradient, low elevation streams, stream order was significantly related to number of species. Species richness was greatest in low elevation, low gradient, high order streams. Species richness of a site reflected species richness of the drainage: in paired comparisons, sites in a drainage with a richer ichthyofauna had more fish species than sites in a drainage with fewer species. Addition of species with increasing stream order occurred in most streams, but replacement was more frequent than in other studies relating fish to stream order. The apparently higher frequency of replacement in this study appeared to be a result of headwater introductions of brook charr, Salvelinus fontinalis, and a tendency for cutthroat trout, Salmo clarki, to occupy headwaters when in freshwater.