An Evaluation of the hydrological impacts of Wimbleball Reservoir using the IHA approach
A river acts both as a source and carrier of water for supporting and sustaining the biological diversity and integrity of the aquatic, wetland and riparian species and natural ecosystems. To accomplish these functions, it is necessary that river water meets some essential qualitative and quantitative parameters and the stream-flow exhibits the dynamics and hydrological attributes comparable to that of natural or unaltered river flows (hydrologic regime).
This hydrologic regime is the lifeline of freshwater ecosystem and all diverse variety of aquatic & riparian species are for long accustomed and adapted to the characteristic temporal, spatial and hydrologic variation of water flow cycles attributable to the natural or unaltered water flow. Unfortunately, this regime and its naturally configured variation patterns get disturbed failing to absorb the stresses induced by our ever-increasing demands and environmentally irresponsive use of water.
To evaluate the shifts in the pre and post-reservoir hydrologic parameters, the effect of Wimbleball Reservoir have been analysed based on the long-term flow-patterns of the downstream discharge of the reservoir. The analysis was conducted by a very robust statistical model called the IHA model. Both long term differences and RVA analysis show substantial impacts of manmade reservoir control on the biota of the Exe-catchment.
Water bodies like rivers, streams, channels, etc. serve a dual function being essential source points for our day-to-day water requirements as well being its transporters or carriers by flowing in and channelling water downstream to the river beds, catchments and agricultural fields in the process supporting and sustaining the biological diversity and integrity of the aquatic, wetland and riparian species and natural ecosystems. Our earth is also called the ‘water planet’ as water forms approximately 70% of its total surface (The Ground Water Foundation,2003) but only a part of it is available for our use. This realization has long back prompted us to take up some water management practices.
In the beginning, water management practices were very much focused on issues like water quality and flood control measures and the overall strategy was never so broad to include other aspects like water quantity, stream flow management and restoration (BD, Richter, etal,1997)2. However, issues pertaining to water quantity, flow, restoration, etc. gradually started to get prominence in our policy framework following a landmark order passed by the US Supreme Court identifying the separation of water quality from water quantity and flow as an artificial distinction and recommending incorporation of both water quality and quantity objectives in a broader and comprehensive water management policy framework (US-EPA, 2002)3.
Water quality, quantity & flow conditions are in way inseparable features considering the fact that the amount of flow in a river effects many issues of water quality and water quantity at the same time. Therefore, the assessment on the wholesomeness of water in any system is essentially dictated by the above conditions of quality, quantity and flow characteristics. Going by this approach broadens the overall water policy framework making this a comprehensive management initiative. This shift in water management approach necessitated re-configuration of the erstwhile single or limited objective driven practice of flood& storm water control thereby embracing a comprehensive initiative of total ecosystem management & restoration having multi-utility potentials.
This system is very important and effective because this takes into account the sustainable use of water resources or ‘water takings’ and their possible restoration (Dept. of Fisheries &Oceans, Canada, 2002)4. Under the ambit of this, it is necessary that river water meets some essential qualitative and quantitative parameters and the stream-flow exhibits the dynamics and hydrological attributes (hydrologic regime) comparable to that of natural or unaltered river flow (Richter D. Brian & etal) 5. This hydrologic regime or ‘natural flow regime’ is the lifeline of freshwater ecosystem and all diverse variety of aquatic & riparian species are for long accustomed and adapted to the characteristic temporal, spatial and hydrologic variations of water flow cycles attributable to the natural or unaltered water flow.
Unfortunately, this regime and its naturally configured variation patterns get disturbed (Allan David & HinzLeon, SNRE, 2004)6 failing to absorb the stresses induced by our ever-increasing water takings demands and environmentally irresponsive use of water. In fact, this is the point where human intervention or controls and water integrity issues found themselves in a highly confronting and conflicting platform.
Increased water demands compelling human actions like construction of water reservoirs, dams, impoundments, etc. for storing and using water for domestic, energy and hydropower, artificial parks and various other uses have started taking their toll on river waters and water bodies substantially degrading the quality, quantity and importantly squeezing the downstream water flows(Benue, A. C. 1990). This flow reduction in rivers consequential to manmade flood and irrigation control practices like reservoirs and dams are found to alter the natural hydrologic regime bringing in a series of impairments to overall ecosystem and also opening up a new front in the field of river and hydrology studies.
This paper aims to assess the variations in the hydrological parameters of a river system specifically attributable to impacts of man-made interventions or controls like reservoirs. Primarily, the research ambition is to identify and evaluate the degree of alterations in the hydrologic profile by analysing the long-term historical as well recent water flow records representative of the pre-impact and post-impact period of construction and commissioning of a typical reservoir. An emerging computer tool called the ‘IHA’ (Indicators of Hydrologic Alterations) has been applied to generate scenarios and analyze the data. The records and data needs for this study have been sourced from an existing gauging station in the Exe river of South-West England strategically selected to represent the influence of the ‘Wimbleball Reservoir’.
Natural Flow Regime & Hydrologic Alterations – Ecological Significance
The concept of natural flow regime is based on the understanding that aquatic and riparian organisms depend upon, or can tolerate arrange of flow conditions specific to each species (Puff etal, 1997)7.For example, certain fish species moves into safer floodplain areas during floods to feed and escape from attacks of other species occupying the main water body thereby adapting a mechanism to survive and carry on all by itself. This in a way indicates that if flooding occurs at the right time of the year, and lasts for the right amount of time, these fish populations will benefit from the flood event finally. Again as a contrast to this case, other species may be adversely affected by the same flood.
With the development of the science of hydrology, it has been confirmed with a good degree of confidence that hydrologic regime with all its natural and temporal variations (both intra-annual and inter-annual) are needed to maintain and restore the natural form and function of aquatic ecosystems. However, this prerequisite is not in line with the traditional water management practice which is functionally attuned to influence and dampen natural fluctuations with the objective to provide steady and undisturbed supply of water for different in-stream and out-of-stream activities(Richter et al., 2003) .
Moreover, for intervening and containing extreme drought and flood events, the traditional water management initiatives rather relied on moderating and limiting flow fluctuations. Many studies indicate ‘natural flow regime’ as a determinant toing-stream flow needs of a water body. For example, (Richter et al,1996) and (Puff et al. 1997) generalized that natural flow conditions may indicate and determine in-stream flow requirements. There exists a correlation between stream-flow and other physicochemical characteristics critical to ecological integrity of streams and rivers(Puff etal., 1997).
Precisely, flow can be associated to some direct as well indirect or secondary impacts and as such flow characteristics can be used as surrogates for other in-stream indicators and ecosystem conditions and importantly the components of a flow regime as shown infigure-1, are very much accessible to scientific inquiry (IFC, 2002,Poff et al. 1997, Richter et al., 1996) .
Any disruption, fragmentation and dilution of this natural regime of water-flow leads to ‘Hydrological alteration’ and in general, this can be defined as any anthropogenic disruption in the magnitude or timing of natural river flows (Biosciences, 50-9, 2000). The natural flow regime of a river is dependent on various factors including rainfall, temperature and evaporation when considered in a broader geographic scale or macro-scale and is also influenced by the physical characteristics of a catchment at the catchment level or micro-scale(Rash et al, 1988) .
As mentioned earlier, river flow regimes are also affected directly and indirectly by human activities. Such human interventions disrupting natural flow of a river through construction and operation of reservoirs and dams have the potential of triggering a series of undesirable consequences like extensive ecological degradation, loss of biological diversity, water quality deterioration, groundwater depletion, and also more frequent and intense flooding(Puff et al, 1997).
Reservoir are built to store water to compensate for fluctuations in river flow, thereby providing a measure of human control of water resources, or to raise the level of water upstream to either increase hydraulic head or enable diversion of water into canal. The creation of storage and head allows reservoirs to generate electricity, to supply water for agriculture, industries, and municipalities, to mitigate flooding and to assist river navigation(Rash et al. 1988).
The biological effects of hydrologic alterations are often difficult to disentangle from those of other environmental perturbations in heavily developed catchments as identified by Rosenberg et al. (Environmental Reviews 5: 27–54, 1997) .
The impacts of large-scale hydrological alteration include habitat fragmentation within rivers (Dynes us and Nilsson 1994) , downstream habitat changes, such as loss of floodplains, riparian zones, and adjacent wetlands and deterioration and loss of river deltas and ocean estuaries (Rosenberg et al. 1997)36,deterioration of irrigated terrestrial environments and associated surface waters (McCall 1996) . Hydrological alterations also bring another indirect or secondary impacts on the genetic, ecosystem and global levels.
They can cause genetic isolation through habitat fragmentation (Pringle 1997) , changes in processes such as nutrient cycling and primary productivity (Pringle 1997, Rosenberg et al. 1997),etc.
With the realization of the importance of natural flow regime and the possible dangers posed by human alterations, there emerged a relatively new and promising water and ecology management paradigm. Many researchers started seeing this as a very comprehensive and sound management option and on many occasions stressed regarding the urgency of protecting or restoring “natural” hydrologic regimes (Sparks 1992;National Research Council, Doppler et al. 1993; and Dynes us &Nilsson 1994) . Effective ecosystem management of aquatic, riparian, and wetland system requires that existing hydrologic regimes be characterized using biologically-relevant hydrologic parameters, and that the degree to which human-altered regimes differ from natural or preferred conditions be related to the status and trends of the biota(BD, Richter, etal, 1997).
Ecosystem management efforts should be considered experiments, testing the need to maintain or restore natural hydrologic regime characteristics in order to sustain ecosystem integrity. Only some limited studies have closely examined hydrologic influences on ecosystem integrity and this is mainly because most of the commonly used statistical tools are poorly suited for characterizing hydrologic data into biologically relevant attributes(BD, Richter, etal, 1997). Without such knowledge, ecosystem managers will not be compelled to protect or restore natural hydrologic regime characteristics.
However, recently, there have been some significant developments in the field of hydrological studies and importantly few robust computer statistical tools and models like IHA & Range of Variability Approach (RVA) using the (Indicators of Hydrologic Alterations, BD, Richter, etal, 1997), Wetted Physical Habitat Simulation System (PHABSIM Model, Jowett, 1997)35, Flow Incremental Methodology (FIM), other Hydrologic Modelling Software like GAWSER, Ontario Flow Assessment Techniques (OFAT), etc. are now known to exist(Jowett, 1997).
The following sections attempt to evaluate and assess the possible effects of hydrological alteration specifically induced by human interventions or activities. A very useful computer model called the model (available at Freshwaters.com) has been used for generating and evaluating the effects of flow variations. The ecological zone considered for analysis in this paper is the ‘Exe river Estuary’ region and the gauging station selected is 45001 – Exe at Thorverton.
The Indicators of Hydrologic Alteration (IHA) Method – Approaches & Application
The evaluation and assessment of the flow regime of the Exe-river system and the variations it witnessed after the construction of the ‘Wimbleball Reservoir’ have been accomplished by the application of Avery detailed computer-modelling tool known as the IHA or ‘Indicators of Hydrologic Assessment’ model. The software basically takes birth from the concept of integrity and wholesomeness of the ‘natural flow regime ‘and is configured and capable of determining the relative transformations and variations in this natural flow regime subject to any natural or artificial modifications or alterations (BD, Richter, etal, 1997).
At first, it requires defining and identifying a series of biologically-relevant hydrologic attributes that characterize intra anointer-annual variations in water conditions which are further processed for a robust statistical variation analysis after isolating the data-sets to represent two different periods resembling the pre-impact and post-impact scenarios (Rosenberg, et al, 2002). The Nature Conservancy is now the custodian of this statistical tool, which is very useful for assessing the degree to which human activities have changed flow regimes (US-EPA, 2002). Brian D. Richter and et al. from the Nature Conservancy (Richter D. Brian, etal, 1996-97) have identified four basic for this analysis and they are:
(I) Define the data series (e.g., stream-gauge or well records) for pre- and post-impact periods in the ecosystem of interest.
(ii) Calculate values of hydrologic attributes – Values for each of 32ecologically-relevant hydrologic attributes are calculated for each year in each data series, i.e., one set of values for the pre-impact data series and one for the post-impact data series.
(iii) Compute inter-annual statistics – Compute measures of central tendency and dispersion for the 32 attributes in each data series, based on the values calculated in step 2. This produces a total of 64 inter-annual statistics for each data series (32 measures of central tendency and 32 measures of dispersion).
(iv) Calculate values of the Indicators of Hydrologic Alteration -Compare the 64 inter-annual statistics between the pre- and post-impact data series, and present each result as a percentage deviation of onetime period (the post-impact condition) relative to the other (there-impact condition). The method equally can be used to compare the state of one system to itself over time (e.g., pre- versus post-impacts just described); or it can be used to compare the state of one system to another (e.g., an altered system to a reference system), or to compare current conditions to simulated results based on models of future modification to a system.
The same computational strategies will work with any regular-interval hydrologic data, such as monthly means; however, the sensitivity of the IHA method for detecting hydrologic alteration is increasingly compromised with time intervals longer than a day (Richter. Brian, etal, 1996-97). Detection of certain types of hydrologic impacts, such as the rapid flow fluctuations associated with hydropower generation at dams, may require even shorter (hourly) interval. They have also suggested that ‘the basic data for estimating all attribute values may preferably be daily mean water conditions (levels, heads, flow rates).
Hydrologic conditions in general can vary in four dimensions within an ecosystem (three spatial dimensions and time).However, the three spatial domains can be scaled down to one with the assumption that only one spatial domain exists at any strategic location over time in a river system. Restricting the domain to one specific point within a hydrologic system (like any measuring point in river) makes it simple for us to identify specific water conditions with one spatial and one temporal domain.
These events may be specific water conditions like heads, levels, rate of change, etc. (Richter Brian, etal, 1996) whose temporal variations can be recorded and assessed from that particular spatial point or from a single position. Such temporal changes in water conditions are commonly portrayed as plots of water condition against time, or hydrographs.
Here, we seek to study and analyse the variations in hydrologic conditions using indicators and attributes, which should essentially be biologically relevant as well as responsive to human influences or modifications like reservoir and dam operations, ground water pumping, agricultural activities, etc. at the same time (Richter D. Brian, etal,1996,). Importantly, a variety of features or parameters of hydrologic regime can be used and functionally superimposed (Sense South wood 1977, 1988; Puff & Ward 1990}40 to virtually represent and finally characterize the “physical habitat templates” (Townsend& Hilde, 1994)43 or “environmental filters” (Sense Eddy 1992)42that shape the biotic composition of aquatic, wetland, and riparian ecosystems.
The IHA method is based on 32 biologically relevant hydrologic attributes, which are divided into five major groups to statistically characterize intra-annual hydrologic variation as showman Table-1. These 32 attributes are based upon the following five fundamental characteristics of hydrologic regimes:
1. the magnitude of the water condition at any given time is measure of the availability or suitability of habitat, and defines such habitat attributes as wetted area or habitat volume, or the position of water table relative to wetland or riparian plant rooting zones;
2. the timing of occurrence of particular water conditions can determine whether certain life cycle requirements are met, or influence the degree of stress or mortality associated with extreme water conditions such as floods or droughts;
3. the frequency of occurrence of specific water conditions such as droughts or floods may be tied to reproduction or mortality events for various species, thereby influencing population dynamics;
4. the duration of time over which a specific water condition exists may determine whether a particular life cycle phase can be completed, or the degree to which stressful effects such as inundation or desiccation can accumulate;
5. the rate of change in water conditions may be tied to the stranding of certain organisms along the water’s edge or in pounded depressions, or the ability of plant roots to maintain contact with phreatic water supplies.
A detailed representation of the hydrologic regime can be obtained from these 32 parameters for the purpose of assessing hydrologic alteration. Importantly, all the parameters having good ecological relevance do not call for any parameter specific statistical analysis and all of them can be processed by single and unique approach like they (Kozlowski 1984; Bustard 1984; Puff & Ward 1989)46. Also, because certain stream-flow levels shape physical habitat conditions within river channels, it is needed to identify some hydrologic characteristics that might aid in detection of physical habitat alterations. (Richter D. Brian, etal, 1997).
Sixteen of the hydrologic parameters focus on the magnitude, duration, timing, and frequency of extreme events, because of the pervasive influence of extreme forces in ecosystems (Gaines & Denny 1994)48 and geomorphology (Leopold1994)49 and other 16 parameters measure the central tendency of either the magnitude or rate of change of water conditions (Table-2). The rationale underlying the five major groupings and the specific parameters included within each are described below.
Table-2: Summary of various Hydrological Groups
Groups Descriptions Number of total Hydrologic Parameters
1 Magnitude of monthly water conditions 12
2 Magnitude & duration of annual extremes 10
3 Timing of annual extremes 02
4 Frequency & duration of high & low pulses 04
5 Rate & frequency of change in conditions 04
Group-1: Magnitude of Monthly Water Conditions
This group includes 12 parameters, each of which measures the central tendency (mean) of the daily water conditions for a given month. The monthly mean of the daily water conditions describes “normal” daily conditions for the month, and thus provides a general measure of habitat availability or suitability. The similarity of monthly means within a year reflects conditions of relative hydrologic constancy, whereas inter-annual variation (e.g., coefficient of variation) in the mean water condition of a given Month provides an expression of environmental contingency (Colwell 1974; Puff & Ward1989). The terms “constancy” and “contingency” as used here refer tithe degree to which monthly means vary from month to month (constancy),and the extent to which flows vary within any given month(contingency).
Group-2: Magnitude and Duration of Annual Extreme Water Conditions
The 10 parameters in this group measure the magnitude of extreme(minimum and maximum) annual water conditions of various duration, ranging from daily to seasonal. The durations that we use follow natural or human-imposed cycles, and include the 1-day, 3-day, 7-day(weekly), 30-day (monthly), and 90-day (seasonal) extremes. For any given year, the 1-day maximum (or minimum) is represented by the highest (or lowest) single daily value occurring during the year; thematic-day maximum (or minimum) is represented by the highest (or lowest) multi-day average value occurring during the year. The mean magnitudes of high and low water extremes of various duration provide measures of environmental stress and disturbance during the year; conversely, such extremes may be necessary precursors or triggers for reproduction of certain species. The inter-annual variation (e.g. Coefficient of variation) in the magnitudes of these extremes provides another expression of contingency.
Group-3: Timing of Annual Extreme Water Conditions
This group includes 02 parameters one measuring the Julian date of the 1-day annual minimum water condition, and the other measuring the Julian date of the 1-day maximum water condition. The timing of the highest and lowest water conditions within annual cycles provides another measure of environmental disturbance or stress by describing the seasonal nature of these stresses. Key life cycle phases (e.g. Reproduction) may be intimately linked to the timing of annual extremes, and thus human induced changes in timing may cause reproductive failure, stress, or mortality. The inter-annual variation in timing of extreme events reflects environmental contingency.
Group-4: Frequency and Duration of High and Low Pulses
This group has 04 parameters include two, which measure the number of annual occurrences during which the magnitude of the water condition exceeds an upper threshold or remains below a lower threshold, respectively, and two, which measure the mean duration of such high and low pulses. These measures of frequency and duration of high- and low-water conditions together portray the pulsing behaviour of environmental variation within a year, and provide measures of the shape of these environmental pulses. Hydrologic pulses are defined here as those periods within a year in which the daily mean water condition either rises above the 75th percentile (high pulse) or drops below the25th percentile (low pulse) of all daily values for the pre-impact time period.
Group-5: Rate and Frequency of Change in Water Conditions
The four parameters included in this group measure the number and mean rate of both positive and negative changes in water conditions from one day to the next. The
Rates and frequency of change in water conditions can be described in terms of the abruptness and number of intra-annual cycles of environmental variation, and provide a measure of the rate and frequency of intra-annual environmental change.
Assessing Hydrologic Alteration
In assessing the impact of a perturbation on the hydrologic regime, we want to determine whether the state of the perturbed system differs significantly from what it would have been in the absence of the perturbation. In particular, we want to test whether the central tendency or degree of inter-annual variation of an attribute of interest has been altered by the perturbation (Stewart-Oaten et al.1986)55. The assessment of impacts to natural systems often poses difficult statistical problems, however, because the perturbation of interest cannot be replicated or randomly assigned to experimental units (Carpenter 1989; Carpenter et al. 1989; Hulbert 1984;Stewart-Oaten et al. 1986)66.
The lack of replication does not hinder estimation of the magnitude of an effect, but limits inferences regarding its causes. However, the IHA method is robust and can be easily adapted to more sophisticated experimental designs.
A standard statistical comparison of the 32 IHA parameters between two data series would include tests of the null hypothesis that the central tendency or dispersion of each has not changed. However, this null hypothesis is generally far less interesting in impact assessments than questions about the sizes of detectable changes and their potential biological importance. A standardized process for assessing hydrologic impacts is included within the IHA software.
The Range of Variability Method (RVA) is another analysis frame in which to assess change in structured manner. This method of determining hydrologic alteration is based on the theory that there is natural variability in stream-flow. The RVA software would plot and determine whether an activity, such as water taking, would alter the stream-low outside this normal variability. Significant alteration would occur if the stream-low regime were altered more than one standard deviation from the natural variability, which may have ecological consequences.
Development of Pre- and Post-Impact scenarios
When adequate hydrologic records are available for both there-impact and post-impact time periods, application of the IHA method will be relatively straightforward using the statistical procedures described above. When pre- or post-impact records are nonexistent,include data gaps, or are inadequate in length, however, various datareconstruction or estimation procedures will need to be employed. Examples of such procedures include the hydrologic record extension techniques described by Searcy (1960) and Alley & Burns (1983).Hydrologic simulation modelling or water budgeting techniques can also be used to synthesize hydrologic records for comparison using the IHAmethod (Linsley et al. 1982)73.
Accounting for Climatic Differences
Climatic differences between the pre- and post-impact time periods obviously have the potential to substantially influence the outcome of the IHA analysis. Various statistical techniques can be used to test for climatic differences in the hydrologic data to be compared. When the IHA analysis is to be based upon actual hydrologic measurements rather than estimates produced from models, a reference site or set of sites uninfluenced by the human alterations being examined can be used as climatic controls (Alley & Burns 1983).
For example, stream-gauge may exist upstream of a reservoir thought to have impacted study site. Analyses can establish a statistical relationship between stream-lows at the study site and at the upstream reference site using synchronous pre-dam data sets for the two sites. This relationship can then be used to estimate the stream-low conditions that would have occurred at the study site during the post-impact time period in the absence of the reservoir.
IHA Application- Description of Study Site
As mentioned earlier, the principal motive of this study is tantalize and evaluate the impacts, if any, of human interventions like reservoir operations on the overall sanctity and natural integrity, i.e. the natural hydrologic regime of water bodies like rivers. Here the operation of a well know reservoir in the south-west coast of Britain called the ‘Wimbleball reservoir’ has been identified as the human intervention point which is sufficiently used to store and supply water to cater to human needs like hydropower, drinking water supply, etc. (SW-Environment Agency, 2003)81 and eventually it ends up regulating a river system in the process.
The down-stream water body and habitat, which is expected to come under the influence of the alterations resulting from the Wimbleball reservoir operations, considered here is the Exe-river estuary system. The main motivation for selection of the above reservoir and the river system happens to bathe strategically located river monitoring system (gauge-station),which falls in the influence zone. This station is designated as‘No.45001-Exe at Thorverton’ having a grid reference of ‘21 (SS) 936016’ (NRFA & Data Holdings, 2005)66. Figure-2 (enclosed) shows diagrammatic representation of the Exe-river catchments area along with the positions of the river and reservoir. The national authority NRFA, describes the monitoring station as “Velocity-area station with cableway and flat V-Crump profile weir constructed in 1973 due to unstable bed condition” (NRFA, 2005)66.
There also exists minor culvert flow through mill u/s of station included in rating. Notably, Low flows are affected significantly by the operations of the Wimbleballreservoir post-1979 and by exports to the Taw catchment. Station iscontrol point for operational releases from Wimbleball (NRFA & DataHoldings, 2005)66. The headwaters drain Exmoor and the geology is predominantly Devonian sandstones and Carboniferous Culm Measures, with subordinate Permian sandstones in the east, Moorland, forestry and arrange of agriculture (NRFA & Data Holdings, 2005)66.
The Exe Estuary is partially an enclosed tidal area composed of both aquatic (marine, brackish and freshwater) and terrestrial habitats.
The Estuary makes an important contribution to the diversity of British estuaries by virtue of its unspoilt nature, international conservation importance, recreational opportunities and high landscape value(SW-Environment Agency, 2003) . This Estuary flows through an open landscape with gently rolling hills on either side. It is shallower than many estuaries in the south west of England, so the tide plays significant role, wit