Skip Navigation


Bioscience Horizons Advance Access originally published online on February 25, 2009
Bioscience Horizons 2009 2(1):44-54; doi:10.1093/biohorizons/hzp009
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
2/1/44    most recent
hzp009v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Haley, M. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Haley, M. A.
Related Collections
Right arrow Ecology
Right arrow Environmental Science
Right arrow Physiology
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2009 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

The impact of stream support on the hydrology and macrophytes of the upper Bristol Avon

Mark A. Haley*

School of Science and the Environment, Bath Spa University, Bath, UK

* Corresponding author: School of Science and the Environment, Bath Spa University, Bath, UK. Tel: +44 0779 2347073. Email: ancientpathways{at}hotmail.co.uk

Supervisor: Dr D. Watson, Bath Spa University, Bath, UK


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
 References
 
Low river flow due to water abstraction has impacted the macrophyte communities on several rivers in England. Stream Support is a method of alleviating low flows by augmenting water supply to the headwater channels of rivers when flow falls below a trigger level. This project has studied that the effect Stream Support has had on the hydrology of the upper Bristol Avon, and the consequent impact these changes have had on the macrophyte communities of the river. The Winterbourne classification methodology was used to investigate hydrological effects on macrophyte communities since the early 1990s. Stream Support was found to have successfully targeted low flow conditions with the annual Q95 measure of low flow closely associated with augmentation (Tetbury: R2 = 71%, p < 0.001; Sherston: R2 = 78%, p < 0.001). Decade long Q95 (discharge exceeded 95% of the time) measures of low flow on the river were shown to have risen in 1998–2007 by 60% above measures for 1978–1987 and 1988–1997. Aquatic species such as Ranunculus penicillatus pseudofluitans had increased coverage since the 1990s by as much as 40%, but wetland species such as Myosotis scorpioides and Mentha aquatica had marginally lower coverage. These changes may be partly attributable to a reduction in low flow conditions. Stream Support may not be sustainable if global warming leads to annually occurring prolonged periods of summer drought.

Key words: Stream Support, hydrology, macrophytes, aquatic plants, low flows, water abstraction, Bristol Avon


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Conclusion
 References
 
Background
Aquatic plant communities are important to riverine ecosystems as they provide services such as oxygenation and nutrient cycling, as well as reproduction sites and predation refugia for river fauna. Macrophytes have a vital role in stabilizing the substrate and channelling flow leading to scoured gravels, and changes in plant communities can have implications for a river's amenity value and for fisheries.1–3

Flow regime has a strong influence on riverine macrophyte communities by determining the physical structure of the aquatic environment and impacting on the life-history strategies of plants.4 Groundwater abstraction for public water supply from aquifers near the heads of perennial rivers can lead to a reduction in upstream flow, particularly during the summer months, as recharge water is exported to urban areas from which it enters the river via sewage treatment works further downstream or is transferred to a different catchment altogether.5, 6 In the UK, a combination of increased water abstraction for an expanding population and periods of drought during the 1980s and 1990s led to periods of extended low flow on rivers in the south and southeast of England, a region dominated by groundwater fed rivers running over permeable geology made up of various forms of limestone.7

Low flow leads to the reduction of the wetted area and increases in sedimentation, pollutant concentration and water temperature.8, 9 These factors have contributed to the loss of aquatic macrophytes on at least three rivers in eastern England and have led to changes in macrophyte species composition and cover in many rivers on the chalklands of southern England as communities adapt to more ephemeral summer flows.2, 10 Low flows are also implicated in the success of invasive species such as Elodia canadensis.4

Extensive survey in southern England by Holmes11 showed zoning of aquatic macrophyte species according to flow level. Reaches with perennial flow were dominated by the true aquatic species, such as Ranunculus penicillatus pseudofluitans, Sparganium erectum and Callitriche spp., and emergent species, such as Apium nodiflorum and Nasturtium officinale, and wetland species such as Myosotis scorpioides and Mentha aquatica were frequently at the margins, non-aquatic grasses and herbs tended to be absent altogether. Reaches that dried out for 2–4 months during the summer generally had a mixture of emergent species and wetland species such as Alopecurus geniculatus, but non-aquatic species such as Agrostis stolonifera were also present. Reaches that dried for >4 months were frequently dominated by non-aquatic species of grasses and herbs with true aquatic species absent.2

It is important when considering aquatic macrophytes to take into account water velocity as well as water coverage. Separating water flow from water coverage is important when considering species such as Lemna minor, which is a free floating aquatic macrophyte but requires still water to remain in position, or R. penicillatus pseudofluitans, which requires a good flow rate to enable nutrient uptake and respiration.12

Stream ecosystems in England and Wales are given some legal protection from low flows by the minimum acceptable flow concept embodied in the Water Resources Act 1991. This operational mechanism normally gives a minimum river discharge value below which abstraction in an area must cease. However, this approach is only effective in surface water dominated catchments. In areas with permeable geology, the buffering effect of the aquifer means that a reduction in abstraction based on river discharge will not mean an increase in flow without subsequent recharge from rainfall. A method of defining an ecologically acceptable flow for groundwater dominated catchments based on annual flow volume has been proposed by Petts et al., 13 but has no current legal status other than its inclusion in the requirements for specific special areas of conservation.14

More wholesale legal protection for river flows may come into force through the Water Framework Directive (WFD), which is due to be implemented throughout the European Union by 2015. It requires the classification of surface waters based on environmental conditions. The WFD approach integrates water quality, quantity and physical habitat with ecological indicators to produce classifications based on both ecological status and chemical status. Water bodies of good ecological status should only deviate slightly from the biological, structural and chemical characteristics that would be expected under natural conditions.15 The inclusion of water quantity within the classification criteria is important as a river will not achieve ‘good’ ecological status unless flow is close to that which would be expected if abstraction were not taking place.16

Global warming may have an impact on flow regime in the future as winters become wetter and summers drier.2 This change may have a negative effect on aquifer recharge as although winter rainfall is likely to be higher, the recharge period will be shorter and this may lead to a drop in recharge by as much as 30%.17 Greater divergence between winter high flows and summer low flows is therefore expected in rivers in England and Wales as the climate warms, irrespective of water abstraction.18

In a response to reports detailing the ecological impacts of low flows on rivers in England and Wales such as the ‘High and Dry’ report by Drury Hunt and MacGuire,19 and the report on the impact of abstraction on wetland SSSI's issued by English Nature,20 the Environment Agency developed and published the National Environment Programme in 1999. This is a list of improvement schemes to be funded though water company charges, including 118 river sections to be targeted for flow improvements.21 Included in this list are three target rivers in the Wessex Water region, the Chitterne Brook in Wiltshire (a tributary of the Hampshire Avon), the Piddle in Dorset and the upper Bristol Avon in the south Gloucestershire.

Wessex Water,22 working in partnership with the Environment Agency, English Nature and Ofwat, published their ‘Statement of Intent’ document in 2002. This report detailed schemes to augment low flows on the target rivers by the use of ‘Stream Support’. This is the process by which additional water is pumped from a deep aquifer into the river channel when discharge in the river reaches a predetermined minimum. This ensures the river keeps flowing even during extended dry periods.

Project aims
This investigation aims to assess the impact of Stream Support on low flows on the upper Bristol Avon and to identify the consequent effects on the macrophyte communities of the river. This will involve the use of hydrological and climatic data, and both archive and new macrophyte survey data.

Low flows on the upper Bristol Avon are a long recorded phenomenon during the late summer.23 They are believed to be exacerbated by extensive public drinking water abstraction from the underlying Greater Oolite aquifer, and the absence of abstraction would provide higher levels of base flow to the river.24 To address the low summer flow issue, Stream Support has been provided on the upper Bristol Avon since 1981, but the gauged river flow trigger level for providing support was increased in 2002 under the Statement of Intent.22


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Conclusion
 References
 
Survey area
The Sherston and Tetbury branches of the Bristol Avon rise in the Cotswolds, Gloucestershire. Underlying the area is a 60 m thick layer of Greater Oolite limestone that dips slightly from West to East, springs from which feed the headwaters of both branches.25 The Greater Oolite aquifer is a major source of public drinking water to the region.24 The underlying Inferior Oolite is the source aquifer for Stream Support in the area. Between the two Oolites is a layer of Fullers Earth that is only partially permeable.26 As the Tetbury and Sherston branches travel south and eastwards, respectively, their underlying geology changes from limestone to clays and the two branches meet at Malmesbury in Wiltshire (Fig. 1).


Figure 1
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Schematic map of the Tetbury and Sherston branches of the Bristol Avon showing Stream Support, gauging station and WCM Survey Reaches.11, 31

 
Hydrological methods
The data sources used to analyse the effects of Stream Support on low flows on the upper Bristol Avon are shown in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Data sets used to analyse effects of Stream Support on low flows on the upper Bristol Avon

 
Daily discharge values are given in the Tetbury Avon Daily Gauged Flow and Sherston Avon Daily Gauged Flow data sets.27 There is a single Environment Agency gauging station on the Tetbury Avon at Brokenborough and one on the Sherston Avon at Fosseway. Stream Support inputs are given in the Daily Stream Support Inputs at Luckington, Crow Down Springs and Tetbury data sets.28 There are three Stream Support points in the area. Gauging station and Stream Support Point details are presented in Table 2 and Fig. 1.


View this table:
[in this window]
[in a new window]

  		Table 2. Wessex Water Stream Support Points on the upper Bristol Avon and Environment Agency Gauging Stations27, 28

 
Q95 was used (the discharge exceeded 95% of the time), a common method of defining low flow conditions.29 The Tetbury and Sherston Avon Daily Gauged Flow data sets were used to calculate Q95 values for both branches of the river for each year.27

In order to determine the effect of Stream Support on flows, the daily Stream Support values from the Stream Support data sets were subtracted from the daily gauged values to give natural flow values (estimated gauged discharge in the absence of Stream Support).28 The total Stream Support was summed by year, mean values calculated by year for gauged and natural flow and maximum (peak) discharges calculated for gauged discharge.

To investigate relationships between rainfall and flow levels, the Yeovilton monthly rainfall data set was used alongside the mean monthly flows calculated from the Tetbury and Sherston Avon Daily Gauged Flow data sets.27, 30 Monthly rainfall values were summed by year to give an annual total and summed in various combinations for summer rainfall (June–September, May–October and July–September).

Macrophyte survey methods
The Winterbourne classification methodology (WCM) was used to investigate the effect of flow rates on the macrophyte communities of the upper Bristol Avon.11, 31 This methodology was originally devised to help identify the effects of the summer droughts of the early 1990s on riverine flora (Holmes11, 31). On the basis of surveys undertaken between 1992 and 1995 across southern England, including eight sites on the upper Bristol Avon (Table 3, Fig. 1), the method involves estimating floristic cover by species over the whole area of given river reaches. As it does not involve random sampling within reaches, options for statistical analysis are limited. Acknowledging its limitations, the WCM was used for the 2007 survey in order to obtain data for comparison with the original survey.


View this table:
[in this window]
[in a new window]

  		Table 3. Sites surveyed using WCM11, 31 on the upper Bristol Avon between 1992 and 1995

 
The original survey data from the sites on the upper Bristol Avon were made available by N.T.H. Holmes in the form of floristic tables with percentage and Domin cover values (N.T.H. Holmes, personal communication). The eight sites were surveyed again as part of this investigation between 10 and 20 September 2007, and the results of this survey used along with original data to classify each site for each year surveyed using the WCM predictive key,31 which classifies each site according to its predicted flow regime. Classes range along a hydrological gradient from A (perennial flow) to D (winterbourne reaches). These major groups are subdivided into five subgroups (e.g. A1–A5) to show the gradient within the major group (e.g. A1 having a stronger and more reliable flow than A5).

Given that the survey method did not make use of replicate sampling techniques, the floristic table percentage cover values were used to indicate variation in species cover over time by tabulating the change in cover for each species by survey site. This was done by subtracting the mean percentage cover for surveys between 1992 and 1995 from the percentage cover in 2007. The results from this should be treated with caution as is not intended to be a statistically robust method of analysing change, but is rather a method of summarizing the survey data.

Principal components analysis (PCA) was used to display the floristic survey data to show how the data for each year class and site are distributed.32 PCA is a multivariate analysis based on generating eigenvectors and eigenvalues from the survey data. Unlike detrended correspondence analysis, PCA assumes that species are linearly related to each other, and to environmental gradients.33 In this case, as insufficient environmental data were available, only floristic cover data were used.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Conclusion
 References
 
Hydrological results
The annual series of low flows (determined by Q95) and mean flows can be seen for the Tetbury and Sherston branches of the Avon in Fig. 2. The year with the highest Q95 was 1985 which had particularly elevated low flow levels, even though mean flow for the year was not unusually high. This was due to an even pattern of rainfall from January to August 1985 (Fig. 3).


Figure 2
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Tetbury and Sherston Avon low flows and mean flows; Sources: Daily gauged flow at Brokenborough;27 Fosseway;27 Stream Support;28 monthly rainfall at Yeovilton.30

 


Figure 3
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Monthly Rainfall at Yeovilton for 1985–1987.30

 
The year with the lowest natural flow is 1995 when the river had a negative natural flow on the Tetbury branch, showing that flow on this branch was only being maintained by Stream Support, and a very low natural flow on the Sherston branch. In fact, the river stopped flowing at Malmesbury for short periods in 1990 and 1995, which demonstrates that water was being lost from the river between the gauging stations at Fosseway and Brokenborough, where flow was still evident, and Malmesbury where flow ceased, possibly through the perching of water through the river bed into the depressed water table.34

There is a correlation between the total annual rainfall and mean river flow on both branches of the river (Tetbury: R2 = 20%, p = 0.013; Sherston: R2 = 22%, p = 0.007). Stream Support has little impact on the mean flows on either river (Fig. 2).

The total amount of Stream Support used can be seen to generally rise during the period 1978–2007 (Fig. 4). Stream support was not used prior to 1981, or in 1985 when rainfall was spread evenly across the summer, 1986 when annual rainfall was high and very little in 2007 which had a particularly wet early summer. Heavy pumping of Stream Support to test the ability of the Inferior Oolite aquifer to recover over winter causes the anomalous peak in Stream Support use in 2003 (Fig. 4).


Figure 4
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. The effect of Stream Support on the Q95 of the Tetbury and Sherston Avon as determined by difference in Gauged Q95 at less Natural Q95 (Gauged Q95—Stream Support).27, 28

 
The effectiveness of the pumping regime to target low flows can be seen in Fig. 4 by the correlation between total Stream Support values and the difference it has made to the Q95 values (correlations: Tetbury: R2 = 71%, p = 0.012; Sherston: R2 = 78%, p < 0.001). Although there are year-on-year variations, the effect of Stream Support on low flows on a more long-term basis can be seen in Fig. 5. The Q95 values are similar between 1978–1987 and 1988–1997, but increase by around 60% in the decade 1998–2007 for both rivers.


Figure 5
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Changes in low flow Q95 discharge amounts by decade.

 
The bankfull capacity is slightly lower on the Tetbury branch than the Sherston branch, but is around 8 m3/s in the reaches near the gauging stations on both rivers. Fig. 6 shows that high flows (i.e. flows approaching bankfull) do not occur every year, e.g. there were no high flows between 2004 and 2006, and in some years, flow does not exceed 3 m3/s. Stream Support has no direct impact on high discharges as it is not in operation when high flows occur; however, sedimentation during low flows can change the channel capacity, as can the type and density of vegetation in the channel.1 Sustained frontal rainfall saw two major high flows in June and July 2007, and the highest ever gauged flow for the Tetbury branch was recorded on 20 July 2007. The aquifer remained sufficiently charged through the summer of 2007 to sustain flows, and Stream Support was not required.


Figure 6
View larger version (23K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Tetbury and Sherston Avon maximum annual discharge measured at Brokenborough and Fosseway.27, 28

 
Macrophyte survey results
The site classifications based on the WCM are given in Table 4.11, 31 Most sites show stability or improvement between 1992 and 2007, although BAV5 (Easton Grey) appears to have deteriorated and BAV6 (Cowage Farm) improved. BAV7 (Tetbury), above the Stream Support point, remains dry in the summer, as it has been since at least 1927, and therefore has the lowest classification.23


View this table:
[in this window]
[in a new window]

  		Table 4. Classification of sites on the upper Bristol Avon using WCM11, 31

 
Table 5 shows the relative changes in cover by species between the 1992–1995 period and 2007 for each of the eight sites. Across all sites, filamentous algae (mainly Cladophora glomerata and Vaucheria spp.) and the moss Fontinalis antipyretica show the largest increase. The true aquatic macrophytes R. penicillatus pseudofluitans and Callitriche spp. also show good increases at many sites. Nasturtium officinale shows the largest decrease in cover, and many non-aquatic species have also declined markedly.


View this table:
[in this window]
[in a new window]

  		Table 5. Change in macrophyte cover at sites on the upper Bristol Avon using WCM11, 31

 
Fig. 7 displays the WCM Domin cover values (algae excluded) using PCA.32 The 1992–1995 site results group separately from the 2007 site results. Fig. 8, a PCA diagram displaying species as vectors, shows that the true aquatic angiosperm species R. penicillatus pseudofluitans, Callitriche, Veronica anagallis-aquatica (Veronica spp.) and emergent species such as A. nodiflorum, N. officinale and S. erectum act to draw sites in one direction on axis 2, whereas riparian non-aquatic species of grasses and herbs and aquatic mosses draw sites in the opposite direction on axis 2. On axis 1 aquatic mosses, and to a lesser extent aquatic angiosperm species, draw sites in the opposite direction to emergent and wetland species. The resulting pattern is that aquatic species of angiosperms and mosses draw the 2007 sites in one direction and the wetland and riparian species draw the 1992–1995 sites the other way.


Figure 7
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 7. PCA32 using Kovach45 for species on the upper Bristol Avon using WCM Domin values;11, 31 algae not included. Axis 1: 26%; Axis 2: 19%.

 


Figure 8
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 8. PCA32 using Kovach45 showing species as vectors for species on the upper Bristol Avon using WCM Domin values;11, 31 algae species not included. Axis 1: 26%; Axis 2: 19%.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Conclusion
 References
 
The impact of Stream Support on low flows
The use of Stream Support on the upper Bristol Avon has increased since its introduction in 1981 (Fig. 4), and does appear to have been successful in increasing the Q95 measure of low flows on the river (Fig. 5). The drought periods 1989–1992 and 1996–1998 both led to the river drying up at Malmesbury, but during the 2003–2005 drought even the river's lowest flow remained relatively high (Fig. 2). The operational mechanism in place to target Stream Support specifically at low flows, avoiding its use when flow is adequate, also seems to be successful (Fig. 4), and consequently Stream Support has had little effect on mean annual discharge (Fig. 2).

The use of Stream Support inevitably leads to an unnatural temporal pattern of flow within the river. Flow would naturally be expected to drop steadily during a dry period and recover as a response to rainfall. Under the Stream Support regime, flow drops steadily to a point at which augmentation is triggered, after which the flow remains fairly even until subsequent rainfall has raised the water table back above the trigger level. This pattern may lead to a reduced wetland zone on either side of the wetted channel as the gradual fall of the river during the summer is arrested at the trigger point, so areas above the river height at the trigger point stay dry until natural flow is restored, but areas below the trigger level remain permanently inundated.

There is some evidence of perching of Stream Support water due to the water table being lower than the river bed during the periods of drought (e.g. Fig. 2, natural minimum is negative in 1995), effectively leading to a transfer of water from the Inferior to the Greater Oolite aquifer, the rate of perching would be expected to increase as the water table falls and the implications of this need to be considered when looking at the long-term use of Stream Support if river flows decline due to global warming.2, 17

The impact of reduced incidence of low flow episodes on macrophyte communities
On the basis of the WCM, the macrophyte community at the head of the Crow Down Spring tributary of the Sherston Avon (BAV1), just below the Stream Support point, appears to have changed little between the original survey years 1992–1995 and the survey done in 2007.11, 31 The WCM classification varies between A1 in 1992 and 1995 and A3 in 1993, 1994 and 2007 (Table 4).

Table 5 shows that even though the WCM classification has changed little at BAV1, the community assemblage has changed markedly in this period. There has been a reduction in cover of the aquatic angiosperms R. penicillatus pseudofluitans and Callitriche spp., possibly due to grazing pressure at the site from cattle that appeared intense during the 2007 visit.35

There has been an almost complete loss of considerable Glyceria fluitans stand (recorded under WCM as Glyceria spp.), which may be due to the loss of sediment shoulders from this reach, the preferred substrate for G. fluitans, due to flushing during the high flows in early summer 2007.

The 2007 survey saw the first recorded appearance of Schoenoplectus lacustris, an in-channel rush capable of aerating its rhizome system from emergent shoots through well-developed aerenchyma tissue,36, 37 and an increase in the coverage of the emergent species A. nodiflorum, but several wetland species such as M. scorpioides and M. aquatica, which persisted during the period 1992–1995, were no longer present. These changes may indicate a more reliable flow regime and that the margins are no longer drying out during the summer months.11

At the head of the main branch of the Sherston Avon (BAV3), the sampled reach is immediately downstream of a bridge shortly below the Stream Support point. Its WCM classification ranges between B2(a) in the early 1990s, A1 in 1995 and A4 in 2007. This site is described by Holmes11 as being channelled for flood control in 1992, after which a high flow event during the winter of 1994 led to deposition downstream of the bridge at the site and a radical change in the morphology of the reach from a deep channel to a shallow riffle, and results from this site need to be interpreted with this in mind. Table 5 shows a fall in some wetland species in a similar pattern to that seen at BAV1. Large populations of the riparian species Epilobium hirsutum and Solanum dulcamara seem to have been lost altogether; however, results for riparian species should be treated with caution as it is sometimes a difficult judgement on the part of the surveyor as to whether these species are usually within or outside the wetted channel. The true aquatic Callitriche spp. became well established at the site by 2007 from a small population in the 1990s, as had the emergent species A. nodiflorum and Veronica beccabunga. The success of these species may be due to increased flow, but the change in morphology from deep run to riffle is also likely to have been a major influence.

The head of the Tetbury Avon (BAV7) was dry when sampled both in the 1990s and in 2007, and therefore remains a true winterbourne section with mosses dominating the macroflora. Table 5 suggests a shift from the moss Amblystegium riparium to F. antipyretica; however, this may be due to surveyor error as moss species are easily misidentified.

Around 2 miles downstream of the headwater sites on the Sherston Avon are BAV2 and BAV4. These are very similar sections that join together at the end of each reach at the village of Sherston. Both sites show improvements based on their WCM classifications since the early 1990s. The aquatic species R. penicillatus pseudofluitans has increased substantially and the emergent V. beccabunga moderately since the 1990s on both reaches, whereas wetland species have declined. The emergent species N. officinale showed a major reduction at BAV2, which may be due to it being stripped out by the 2007 high flows and replaced by R. penicillatus spp., the growth patterns of these species can alternate cyclically around such high flow events.38 The high abundance of the free floating species L. minor in 1992–1995 at BAV4 suggests water being ponded in littoral zone vegetation during periods of low flow, its absence in 2007 is probably a result of high summer flows, but may also reflect reduced littoral vegetation.39

A further 3 miles downstream on the Sherston Avon is Easton Grey (BAV5). Like BAV3, this site is situated immediately below a bridge and seems to be suffering from the same problem of deposition and widening leading to shallow riffle formation with several islands of sediment in the reach. Deposition of sediment in riffles and behind bridges is a particular issue after high flow events such as those seen in 2007, subsequent lower flows move sediment from riffles back into pools.1, 40 Sediment deposition at the site may account for its apparent deterioration according to the WCM classification which has dropped from A and B classes during the 1990s to C1 status in 2007. Table 5 shows that the wetland species M. scorpioides has increased since the early 1990s, as has moss cover, but emergent species appear to be in decline.

At the sites furthest downstream, Cowage Farm (BAV6) on the Sherston Avon and Back Bridge (BAV8) on the Tetbury Avon, the results contrast markedly. BAV6 shows a big improvement from C1 status in the 1990s to A1 in 2007; however, at BAV8, a deterioration is seen from A1 in 1994 to B2(a) in 2007. The site position at BAV8 may account for its poor score as it is immediately downstream of a bridge and shows signs of sediment build up. The true aquatic R. penicillatus pseudofluitans shows an increase at the site, and wetland species are making some progress at the margins. The big improvement at BAV6 may be illusory as the species accounts recorded by Holmes11 seem to suggest that the site surveyed in the 1990s was in fact not the site at the given grid reference on the Sherston Avon but at a site on a tributary at the entrance to Cowage Farm.

The PCA diagrams (Figs 7 and 8) show that a general trend across sites is for aquatic angiosperm species and aquatic mosses to have benefited since the 1990s at the expense of wetland species.32 This trend is consistent with Stream Support increasing flow rates and creating a larger, more reliable wetted area during the summer, reducing the ability of wetland species to encroach from margins that periodically dry out. The strength of this trend seems to increase with distance downstream as sites in 2007 line up on axis 1 of Figs 7 and 8 in the same order as their position along the river, which could indicate that physical factors such as channel morphology are acting to buffer the effect of reduced incidence of low flow episodes on macrophyte communities at the heads of the rivers.

Algae species were seen to increase dramatically at all sites in 2007, particularly the filamentous algae C. glomerata and Vaucheria spp. This increase is more likely to be associated with water quality in 2007 rather than any long-term trend as populations of these algae can radically change annually depending on climatic and nutrient conditions during a particular growing season.41

The PCA results seem to contradict the WCM classification results, which seem to show a decline in status with distance downstream. This may be due to the WCM method including algae within its key and its sensitivity to increases in wetland species such as M. scorpioides and riparian species such as Phalaris arundinacea at small densities on sediment banks created behind bridges after high flows such as those in June and July 2007.

Long term issues with the use of Stream Support
Climate change is predicted to lead to wetter winters and drier summers.2 Although in the future full winter recharge of aquifers could be expected every year, low flows due to prolonged summer drought may become a regular occurrence.42 There is some evidence of perching of Stream Support water (Fig. 2), through the permeable river bed when the water table drops below the level of the river when it is kept artificially high by Stream Support. Long summer droughts could lead to large quantities of water being required to keep the river flowing as water is lost through perching to an increasingly deep water table. A side effect of perching will be the transfer of water from the Inferior to the Greater Oolite aquifer.

The overall combined effect of recharge due to removing water from the lower aquifer and perching of Stream Support water through the river bed will be mixing of water between the aquifers that have to date remained independent.


   Conclusion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
References
 
Stream Support has successfully reduced the incidence of low flow episodes on the upper Bristol Avon, but has introduced an unnatural pattern in how the river rises and falls over time. Aquatic plant species have benefited from increased Stream Support since the early 1990s, but wetland species have generally declined. High flow events during the early summer of 2007 removed sediment shoulders from upstream sites and deposited material behind bridges downstream. This has led to changes in the macrophyte community structure that partly obfuscate the impact of Stream Support. Longer periods of summer drought due to global warming may lead to an increased need for Stream Support, and perching of water through the river bed to a depressed water table may lead to mixing of water between the Greater and Inferior Oolite aquifers.

The revised classification of British rivers, published by the Joint Nature and Conservation Council (JNCC), describes the Bristol Avon as a type IV low-land clay river, although the upper reaches more closely resemble a chalk river and therefore have a high potential amenity value.43 The types and quantities of plant cover can influence the amenity value placed on a river, and by increasing aquatic species Stream Support is having a positive impact on this value, and this may consequently affect the amount of conservation effort it receives.44 However, should climate change lead to regular prolonged summer droughts, it is uncertain whether Stream Support will be a sustainable solution to low flows.


   Acknowledgements
 
The author would like to express his thanks to Andy House (Wessex Water) for his help in scoping this project, supplying data and ongoing support; Hazel Grace (Environment Agency) for providing hydrological and water quality data; Lindsey Beech (Wessex Water) for providing training in macrophyte surveying on the Avon; Nigel Holmes (Alconbury Environmental Consultants) for providing historical survey data; Dr David Watson for his invaluable help and advice and to Rebecca Haley-Evans for her patience.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
 References
 

  1. Knighton D. Fluvial Forms & Processes – A New Perspective (1998) London: Hodder Arnold.
  2. Westwood GC, Teeuw RM, Wade PM, et al. Prediction of macrophyte communities in drought-affected groundwater-fed headwater streams. Hydrol Process (2006) 20:127–145.[CrossRef]
  3. Elliott JM. Quantitative Ecology and the Brown Trout (1994) Oxford: Oxford University Press.
  4. Bunn SE, Arthington AA. Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environ Manag (2002) 30:492–507.[Medline]
  5. Owen M. Groundwater abstraction and river flows. J Inst Water Environ Manag (1991) 5:697–702.[CrossRef]
  6. Severn Trent. Network Access Code: Treated Water Services (2002) Coventry: Severn Trent Water.
  7. EA. The State of England's Chalk Rivers (2004) Bristol: Environment Agency.
  8. Heicher DW. Instream flow needs: biological literature review. (1993) 149:37. Susquehanna River Basin Commission.
  9. Lake PS. Ecological effects of perturbation by drought in flowing waters. Freshwater Biol (2003) 48:1161–1172.[CrossRef]
  10. Bickerton MA, Petts GE, Armitage PD, et al. Assessing the ecological effects of groundwater abstraction on the ecology of chalk streams. Regul River (1993) 8:121–134.[CrossRef]
  11. Holmes NTH. Classification of Winterbournes. (1996) Huntingdon: Alconbury Environmental Consultants.
  12. Sculthorpe CD. The Biology of Aquatic Vascular Plants (1967) London: Edward Arnold.
  13. Petts GE, Bickerton MA, Crawford C, et al. Flow management to sustain groundwater-dominated stream ecosystems. Hydrol Process (1999) 13:497–513.[CrossRef]
  14. English Nature. River Avon cSAC Conservation Strategy (2007) Sheffield: English Nature.
  15. Logan P. Ecological quality assessment of rivers and integrated catchment management in England & Wales. J Limnol (2001) 60:25–32.
  16. Logan P, Furse M. Preparing for the European Water Framework Directive – making the likes between habitat and aquatic biotia. Aquat Conserv (2002) 12:425–437.[CrossRef]
  17. Arnell NW. Climate change and water resources in Britain. Climatic Change (1998) 39:83–110.[CrossRef][Web of Science]
  18. Cooper DM, Wilkinson WB, Arnell NW. The effects of climate changes on aquifer storage and baseflow. Hydrol Sci (1995) 40:615–631.
  19. Drury Hunt I, MacGuire F. High and Dry: the Impacts of Over-Abstraction of Water on Wildlife. (1996) RSPB: Biodiversity Challenge.
  20. English Nature. Impact of Water Abstraction on Wetland SSSIs (1996) Peterborough: English Nature.
  21. EA. National Environment Programme (1999) Ipswich: Environment Agency.
  22. Wessex Water. Low Flows Statement of Intent (2002) Nailsea: Wessex Water.
  23. Walls E. The Bristol Avon (1927) Bristol: Arrowsmith.
  24. EA. The Bristol Avon Catchment Abstraction Management Strategy (2005) Ipswich: Environment Agency.
  25. Gloucestershire County Council. Cotswolds Area of Outstanding Natural Beauty Landscape Character Assessment (2007) Gloucester: Gloucestershire County Council.
  26. Jones R. Down the Bristol Avon (1983) Jersey: Ex-Libris Press.
  27. EA. Bristol Avon Mean Daily Gauged Flow at Brokenborough and Fosseway (2007) Bridgewater: Environment Agency.
  28. Wessex Water. Stream Support Daily Augmentation to Bristol Avon at Luckington, Crow Down Springs and Tetbury (2007) Saltford: Wessex Water Science Laboratory.
  29. Smakhtin VU. Low flow hydrology: a review. J Hydrol (2001) 240:147–186.[CrossRef]
  30. Met Office. Yeovilton Monthly Rainfall Dataset (2007) Exeter: National Meteorological Library and Archive. (2007).
  31. Holmes NTH. An Investigation of Chalk Stream Headwater Macrophyte Communities (2005) Bridgewater: Environment Agency Internal Report.
  32. Orloci L. Geometric models in ecology. I. The theory and application of some ordination methods. J Ecol (1966) 54:193–215.[CrossRef]
  33. Jolliffe I. Principal Component Analysis (2005) New York: Wiley.
  34. Wessex Water. Statement of Intent, Review of Phase I (2006) Nailsea: Wessex Water.
  35. Preston J, Muise E, Portig A, et al. An investigation of the factors affecting the occurrence and abundance of Ranunculus penicillatus in selected rivers in Northern Ireland. Environment and Heritage Service Research and Development Series No. 01/01, Belfast.
  36. Steinmann F, Brändle R. Carbohydrate and protein-metabolism in the rhizomes of the bulrush (Schoenoplectus lacustris (L) Palla) in relation to natural development of the whole plant. Aquat Bot (1984) 19:53–63.[CrossRef][Web of Science]
  37. Bunce RGH, Barr CJ, Gillespie MK, et al. Vegetation of the British Countryside - the Countryside Vegetation System: ECOFACT (1999) 1. London: Department of the Environment, Transport and the Regions.
  38. Dawson FH, Castellano E, Ladle M. Concept of species succession in relation to river vegetation and management. (1978) Proceedings of 20th Congress, Internationale Vereinigung fur Theoretische und Angewandte Limnologie: Copenhagen, Denmark. 1429–1434.
  39. Grime JP, Hodgeson JG, Hunt R. Comparative Plant Ecology: a Functional Approach to Common British Species (1988) London: Unwin-Hyman.
  40. Maklin MG, Lewin J. Sediment transfer and transformation of an alluvial valley floor: the River South Tyne, Northumbria, UK. Earth Surf Proc Land (1989) 14:233–246.[CrossRef]
  41. Wong SL, Clark B. Field determination of the critical nutrient concentrations for Cladophora in streams. J Fish Res Board Can (1976) 33:85–92.
  42. Brouyere S, Carabin G, Dassargues A. Climate change impacts on groundwater resources. Hydrogeology (2004) 12:1431–2174.
  43. Holmes NTH, Boon P, Rowell T. Vegetation Communities of British Rivers: a Revised Classification (1999) Peterborough: JNCC.
  44. Edwards PJ, Abivardi C. The value of biodiversity: where ecology and economy meet. Biol Conserv (1998) 83:239–246.[CrossRef]
  45. Kovach. Multivariate Statistical Package (MVSP) (2003) Anglesey: Kovach Computing Services.
Submitted on 30 September 2008; accepted on 18 December 2008


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
2/1/44    most recent
hzp009v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Haley, M. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Haley, M. A.
Related Collections
Right arrow Ecology
Right arrow Environmental Science
Right arrow Physiology
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?