RIVER RESEARCH AND APPLICATIONS River Res. Applic. (2011) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/rra.1511 SHORT COMMUNICATION A PRESUMPTIVE STANDARD FOR ENVIRONMENTAL FLOW PROTECTION B. D. RICHTER,a* M. M. DAVIS,b C. APSEc and C. KONRADd a The Nature Conservancy, 490 Westfield Road, Charlottesville, Virginia 22901, USA b The Nature Conservancy, Atlanta, Georgia, USA c The Nature Conservancy, Brunswick, Maine, USA d The Nature Conservancy, Seattle, Washington, USA ABSTRACT The vast majority of the world’s rivers are now being tapped for their water supplies, yet only a tiny fraction of these rivers are protected by any sort of environmental flow standard. While important advances have been made in reducing the cost and time required to determine the environmental flow needs of both individual rivers and types of rivers in specific geographies, it is highly unlikely that such approaches will be applied to all, or even most, rivers within the forseeable future. As a result, the vast majority of the planet’s rivers remain vulnerable to exploitation without limits. Clearly, there is great need for adoption of a “presumptive standard” that can fill this gap. In this paper we present such a presumptive standard, based on the Sustainability Boundary Approach of Richter (2009) which involves restricting hydrologic alterations to within a percentage‐based range around natural or historic flow variability. We also discuss water management implications in applying our standard. Our presumptive standard is intended for application only where detailed scientific assessments of environmental flow needs cannot be undertaken in the near term. Copyright © 2011 John Wiley & Sons, Ltd. key words: environmental flow; sustainability; Sustainable Boundary Approach; river management; corporate water use; water stewardship; water allocation; water scarcity Received 7 December 2010; Revised 16 January 2011; Accepted 7 February 2011 Available freshwater supplies are being increasingly strained by growing human demands for water, particularly for irrigated agriculture and urban uses. The global population is growing by 80 million people each year, and if consumption patterns evolve as expected, two‐thirds of the world’s population will live in water‐stressed areas by 2025 (WWAP, 2009). Whereas differing patterns of population growth, lifestyle changes and climate change will pose different scenarios on each continent, water managers and planners are challenged to meet growing water needs virtually everywhere. At the same time, societies around the world are increasingly demanding that water managers also protect the natural freshwater ecosystems that are being tapped for water supplies. The need to protect ‘environmental flows’—defined as the quantity, timing and quality of water flows required to sustain freshwater and estuarine ecosystems and the human livelihoods and well‐being that depend on these ecosystems (Brisbane Declaration, 2007)—is now being addressed in many governmental water allocation policies, dam development plans and urban water supply plans. The stimuli for protecting environmental flows are varied and many, *Correspondence to: B. D. Richter, The Nature Conservancy, 490 Westfield Road, Charlottesville, Virginia 22901, USA. E-mail: brichter@tnc.org Copyright © 2011 John Wiley & Sons, Ltd. including the desire to protect biodiversity, ecosystem services (especially fisheries production), water‐based tourism or recreation, economic activities such as hydropower generation and other cultural or spiritual values (Postel and Richter, 2003). However, many good intentions to protect environmental flows have stalled upon encountering confusing and conflicting information about which method for environmental flow assessment is appropriate or ‘best’ and perceptions that the more credible and sophisticated methods require considerable investment of time, expertise and money to apply. These real and perceived hurdles have too often resulted in doing nothing to protect environmental flows, leaving the vast majority of rivers on the planet vulnerable to over‐exploitation (Richter, 2009). The environmental flow science community has long been attuned and responsive to the need for more cost‐efficient and time‐efficient approaches to determining environmental flow needs. Beginning in the 1970s with the Tennant (1976) method and continuing with the recent publication of the ‘Ecological Limits of Hydrologic Alteration’ (ELOHA; Poff et al., 2010), a long series of efforts have been put forth by scientists to streamline and expedite environmental flow assessment while maintaining scientific credibility. However, widespread environmental flow protection across the planet’s river networks has yet to be attained. B. D. RICHTER ET AL. Of particular concern and relevance to this paper is the fact that it is proving difficult to implement ELOHA in some jurisdictions even though the approach was explicitly designed to address the issues that have prevented other methods from being applied widely. The four co‐authors of this paper have been actively encouraging government entities to apply the ELOHA framework; the difficulties we have experienced in these efforts have provided strong motivation for writing this paper. As we explain later in this paper, we continue to believe that ELOHA provides the best available balance between scientific rigor and cost of application for setting environmental flow standards for many rivers simultaneously. The ELOHA framework is currently being applied in various jurisdictions around the world. However, we are finding that many government entities are unable (or unwilling) to afford the cost of applying ELOHA (generally ranging from $100k to $2M), especially in situations where existing biological data and hydrologic models have poor spatial coverage. Time constraints are an even more frequent hindrance to the implementation of the ELOHA framework, particularly for jurisdictions embroiled in politically challenging situations such as responding to extreme droughts, legislative mandates or lawsuits. We suggest that until ELOHA or some variation can be applied everywhere, a presumptive, risk‐based environmental flow standard is needed to provide interim protection for all rivers. Another strong motivation for putting forth a presumptive standard at this time is the fact that many large water‐using corporations are now looking for environmental indicators that can help them screen their operations and supply chains for water‐related risks (e.g. SABMiller and WWF‐UK, 2009). These corporations are increasingly coming to understand that, when environmental flows are not adequately protected, freshwater ecosystems will be stressed, jeopardizing ecosystem services valued by many people for their livelihoods and well‐being. This can lead to conflicts that can ultimately endanger a company’s ‘social licence to operate’ (Orr et al., 2009). Presently, many corporations are using estimates for environmental flow requirements put forth by Smakhtin et al. (2004); these estimates range globally from 20% to 50% of the mean annual river flow in each basin. We agree with Arthington et al. (2006) that such a low level of protection as suggested by Smakhtin ‘would almost certainly cause profound ecological degradation, based on current scientific knowledge’. We hope that the presumptive standard we offer in this paper will replace corporate use of the Smakhtin estimates for water risk screening. The presumptive standard for environmental flow protection put forth in this paper is intended for use only in situations where the application of ELOHA or site‐specific environmental flow determinations (e.g. Richter et al., 2006) cannot be applied in the near future; in other words, it is Copyright © 2011 John Wiley & Sons, Ltd. intended for use as a default placeholder. This presumptive standard is derived from the sustainability boundary approach (SBA) described by Richter (2009), which involves maintaining flows within a certain percentage‐based range around natural flows (i.e. flows in the absence of dam regulation or water withdrawals). Before discussing our proposed presumptive standard in greater detail, we provide a short discussion of the advantages of ‘per cent‐of‐flow’ (POF) approaches such as the SBA for expressing environmental flow requirements. We then summarize efforts around the world to apply flow protection standards based on POF expressions. Finally, we propose a specific presumptive standard using risk bands placed around natural flow variability and conclude with management implications in applying this presumptive standard. APPROACHES FOR SETTING FLOW PROTECTION STANDARDS A primary challenge in setting flow protection standards is to employ a practical method that limits water withdrawals and dam operations in such a way as to protect essential flow variability. As described by Richter (2009), a large body of scientific literature supports the ‘natural flow paradigm’ as an important ecological objective to guide river management (Richter et al., 1997; Poff et al., 1997; Bunn and Arthington, 2002; Postel and Richter, 2003; Arthington et al., 2006). Stated simply, the key premises of the natural flow paradigm are that maintaining some semblance of natural flow regimes is essential to sustaining the health of river ecosystems and that health is placed at increasing risk with increasing alteration of natural flows (Richter et al., 2003; Richter, 2009). Three basic approaches have been employed for setting environmental flow standards across broad geographies such as states or nations: minimum flow thresholds, statistically based standards and POF approaches. The most commonly applied approach to date has been to set a minimum flow level that must be maintained. For example, the most widely used minimum flow standard in the USA is the annual 7Q10, which is defined as the lowest flow for seven consecutive days that occurs every 10 years on average. Whereas the original intent of the annual 7Q10 flow standard was to protect water quality under the federal Clean Water Act of 1972, it has become either explicitly in rule or by default the minimum flow threshold in many states (Gillilan and Brown, 1997; IFC, 2001). The growing recognition that this threshold was not sufficiently protective of aquatic habitats led in the 1980s and 1990s to several states setting higher flow thresholds, such as by setting the minimum level at 30% of the mean annual flow (MAF) or by setting thresholds that vary seasonally, such as at the River Res. Applic. (2011) DOI: 10.1002/rra PRESUMPTIVE STANDARD FOR ENVIRONMENTAL FLOW PROTECTION level of 60% of MAF in winter, 30% of MAF in summer and 40% of MAF in spring and fall (Gillilan and Brown, 1997; IFC, 2001). More recently, statistically based standards have been used to maintain certain characteristics of the flow regime. For example, such a standard may call for protecting a high flow of a specified magnitude, with specified duration, to occur with a specified inter‐annual frequency. The application of a statistically based standard in regulating water use generally involves using computerized hydrologic models to simulate the cumulative effects of licenced or proposed water withdrawals and dam operations on the flow regime; hydrologic changes are allowed to accumulate until the statistical standards would be violated by further withdrawals or dam effects. Flow standards set in the USA, the European Union and elsewhere in the past decade have increasingly been based on a POF approach (see case studies later in this paper). This approach explicitly recognizes the importance of natural flow variability and sets protection standards by using allowable departures from natural conditions, expressed as percentage alteration. The POF approach has several strong advantages over other approaches. For instance, the POF approach is considerably more protective of flow variability than the minimum threshold standards. Minimum‐threshold‐based standards can allow flow variability to become ‘flat‐lined’ as water allocation pressure increases and reservoir operations are designed only to meet minimum release requirements. Statistically based standards, although usually more protective of flow regimes than minimum thresholds, can be confusing to non‐technical stakeholders, and complex statistical targets have proven difficult for water managers to implement (Richter, 2009). By comparison, POF approaches are conceptually simple, can provide a very high degree of protection for natural flow variability and can also be relatively simple to implement (i.e. a dam operator simply releases the prescribed percentage of inflow, or cumulative water withdrawals must not reduce flow by more than the prescribed percentage). Sustainability boundary approach Recognizing that human‐induced flow alterations can both deplete and unnaturally augment natural flows to the detriment of ecological health, Richter (2009) expanded upon the POF approach by suggesting that bands of allowable alteration called ‘sustainability boundaries’ could be placed around natural flow conditions as a means of expressing environmental flow needs, as depicted in Figure 1. To apply the SBA, the natural flow conditions for any point of interest along a river are estimated on a daily basis, representing the flows that would have existed in the absence of reservoir regulation, water withdrawals and return flows (Richter, 2009). Limits of flow alteration, referred to as sustainability boundaries, are then set on the basis of allowable perturbations from the natural condition, expressed as percentage‐based deviations from natural flows. Those withdrawing water or operating dams are then required to maintain downstream river flows within sustainability boundaries. Whereas maintaining flows within the targeted range may be infeasible on a real‐time basis in many cases, such management can be facilitated by creating computerized hydrologic models to evaluate what the likely perturbation to natural flows would be under existing or proposed scenarios of water withdrawal and dam operations and by licencing such water uses accordingly. Figure 1. Illustration of the sustainability boundary approach from Richter (2009; reprinted with permission). The sustainability boundaries set limits on the degree to which natural flows can be altered, expressed as a percentage of natural flows. Copyright © 2011 John Wiley & Sons, Ltd. River Res. Applic. (2011) DOI: 10.1002/rra B. D. RICHTER ET AL. The allowable degree of alteration from the natural condition can differ from one point to another along the same river. This determination for any point of interest along a river requires a negotiation or optimization between the following: (i) the desired consumption or dam regulation of water upstream, which might either deplete or unnaturally augment river flows; (ii) the desired uses of water downstream; and (iii) the desired ecological condition and ecosystem services to be maintained. As such, the SBA forces an explicit integration of environmental flow objectives with water withdrawals and dam operations. We recognize and emphasize that this is a socio‐political decision‐making process as much as it is a scientific one. As suggested by Richter (2009), the application of the SBA in setting river flow management goals requires transparent, inclusive and well‐informed stakeholder engagement. The basic challenge confronting environmental flow proponents is the difficulty of determining how much alteration from natural flows can be tolerated without compromising ecological health and ecosystem services to an undesirable degree. In the absence of such an understanding, water managers and governmental regulators have focused solely on water withdrawals and dam operations, providing only minimum flow protection or neglecting ecosystem considerations altogether. This highly undesirable situation calls for the adoption of a precautionary approach to fill the gap, until more detailed and regionally tailored studies of environmental flow needs can be completed and used to set flow protection standards. We believe that sufficient scientific evidence and knowledge now exist to propose an SBA‐based presumptive standard that can serve as initial guidance for regulating water withdrawals and dam operations in rivers. In designing the presumptive standard recommended later in this paper, we reviewed numerous other efforts to set environmental flow standards that apply across broad regions and many different rivers. CASE STUDY REVIEW The following case studies represent environmental flow policies and management guidelines that are being applied in the USA and Europe to limit flow alteration and to achieve relatively high levels of ecological protection, while allowing for carefully managed water development to proceed. Whereas not all of these cases can be characterized as pure POF approaches, we believe that these case studies illustrate useful and progressive water management policies that fulfill the intent of the SBA. They are described here to demonstrate the feasibility of applying standards in a manner consistent with the SBA and to support our recommendations for the presumptive standard described later in this paper. Copyright © 2011 John Wiley & Sons, Ltd. Example #1—Southwest Florida Water Management District Under the Florida state law, the state’s five water management districts must determine ‘minimum flows and levels’ (MFLs) for priority water bodies of the state. Methods to determine MFLs differ among the five districts. The Southwest Florida Water Management District (SWFWMD) uses a POF‐based approach that includes use of multiple environmental flow assessment methods, including the Instream Flow Incremental Methodology and the Wetted Perimeter approach (see IFC, 2001 for descriptions of these methods), to inform the setting of percentage alteration limits. The intent of the resulting MFLs is to limit water withdrawals such that physical habitat losses do not exceed 15% (Flannery et al., 2002, 2008). The allowable flow reduction, which is referenced to as previous‐day flows at a specified river gauge, can vary with season and with magnitude of flow and includes a ‘hands‐off’ low flow threshold, meaning that all withdrawals are curtailed once the flow threshold is reached (see Rules of the Southwest Florida Water Management District, Chapter 40D‐8, Water Levels and Rates of Flow, Section 40D‐8.041 Minimum Flows at www.swfwmd.state.fl.us). These MFLs are used in water management planning and incorporated as water withdrawal permit conditions. The percentage of allowable depletion has been set by SWFWMD for five non‐tidal rivers in the district, ranging from 8% to 15% during high flows and 10% to 19% during low flows. Allowable depletions tend to be larger for freshwater flows into estuaries. For example, the lower Alafia River can be depleted up to 19% as it enters its estuary, based on limiting fish habitat loss caused by changes in salinity and dissolved oxygen to no more than 15%. No withdrawals are allowed when flows fall below 120 ft3/s, based on chlorophyll residence time in the estuary, fish, dissolved oxygen and comb jellyfish. The proposed MFL for the Lower Peace River and its estuary limits withdrawals to flows above 130 ft3/s, with allowable 16% reduction of daily flow up to a flow rate of 625 ft3/s, 29% flow reductions in fall/winter and 38% flow reductions in summer above 625 ft3/s (Flannery et al., 2002, 2008). Example #2—Michigan’s Water Withdrawal Assessment Tool Approach The Great Lakes–St Lawrence River Water Resources Compact and related state law require limits on water withdrawals to prevent ‘adverse resource impact’, defined as the point when ‘a stream’s ability to support characteristic fish populations is functionally impaired’. Zorn et al. (2008) documented the work of the Michigan Department of Natural Resources to develop a predictive model of how River Res. Applic. (2011) DOI: 10.1002/rra PRESUMPTIVE STANDARD FOR ENVIRONMENTAL FLOW PROTECTION fish assemblages in different types of Michigan streams would change in response to decreased summer base flows, using habitat suitability information for over 40 Michigan fish species. The approach involved classification of all river segments in the state based on size and temperature regime and the development of a fish response curve that relates assemblage richness to an index flow (median August streamflow) for each of the 11 river classes. This index flow serves as a surrogate for withdrawals as a POF. Across the majority of river types in Michigan, ‘baseline or existing’ ecological conditions are predicted to be maintained with cumulative withdrawals less than 6–15% of the index flow, depending on the stream type (Seelbach et al., 2009). This is roughly equivalent to maintaining excellent ecological condition for many rivers, but some rivers that have historically been degraded would only be maintained in their current condition (Paul Seelbach, personal communication, University of Michigan, Ann Arbor). Adverse resource impacts are predicted to occur on most types of rivers with withdrawals greater than 17–25% of index flow. Rivers classified as ‘transitional’ between cold and cool rivers are very sensitive to withdrawals and are limited to withdrawals of 2–4% index flows before adverse resource impact is predicted to occur. The Michigan Water Withdrawal Assessment Tool (WWAT) allows estimation of the likely impact of a water withdrawal on nearby streams and rivers using these threshold values. Use of the WWAT is required of anyone proposing to make a (large) new or increased withdrawal from the waters of the state, including all groundwater and surface water sources, prior to beginning the withdrawal. The WWAT is online at http://www.miwwat.org/. Unlike Florida’s POF approach, which references allowable depletions to a percentage of the previous day’s flow, the Michigan approach references its withdrawal limits only to the August median flow. Because August is typically the lowest flow month in Michigan and Michigan flow regimes are fairly predictable, it is unlikely that cumulative withdrawals beyond the adverse resource impact level would frequently exceed the percentage guideline in other months. However, in very dry summers, one would expect the adverse resource impact percentage to be exceeded for a portion of the summer. already be ‘heavily modified’(Acreman et al., 2006). It is assumed that meeting the Good Ecological Status requires protecting or restoring ecologically appropriate hydrological regimes, but the Water Framework Directive itself does not define environmental flow standards for any country in the EU (Acreman and Ferguson, 2010). In the UK, a Technical Advisory Group worked with conservation agencies and academics to begin defining environmental standards for physio‐chemical and hydromorphological conditions necessary to meet different levels of ecological status (Acreman et al., 2006). A key part of this work was defining thresholds of allowable water withdrawal as a percentage of natural flow. To achieve this, a literature review was prepared, and numerous expert workshops were convened. Each river in the UK was assigned to one of 10 classes, based on physical watershed characteristics, to facilitate application of withdrawal thresholds (Acreman and Ferguson, 2010). Withdrawal standards were based on professional knowledge and discussion of the flow needs of various plant and animal communities—primarily macrophytes, macroinvertebrates and fish. Quantitative standards for achieving Good Ecological Status were specified for four groupings of river types, two seasons and four tiers of withdrawal standards based on annual flow characteristics (Table I). The allowable abstraction values in Table I are intended to be restrictions on cumulative withdrawals, applicable to any point on a river of that type. The withdrawal standards in Table I were derived from an expert consensus workshop approach by using the precautionary principle to deal with considerable uncertainty. Different tolerances to flow alteration were recognized across taxa groups, but a 10% flow alteration was generally seen by experts as likely to have negligible impact for most taxa, stream types and hydrologic conditions (Acreman and Ferguson, 2010). The workgroup also generally agreed upon a Q95 (i.e. fifth percentile) flow as being ‘hands‐off’, meaning that at that flow withdrawal would either stop or be significantly reduced. The recommended allowable withdrawal levels increase with magnitude of flow and in cooler months. Thus, permissible alterations range from 7.5% to 20% in warm months at lower flows (below Q70) and from 20% to 35% during cooler months at higher flows (Acreman et al., 2006). Example #3—UK Application of the European Union Water Framework Directive Example #4—Maine sustainable water use rule The European Union (EU) Water Framework Directive, passed in 2000, was designed to protect and restore aquatic ecosystems by setting common ecological objectives across EU member states. The Water Framework Directive requires member states to achieve a ‘Good Ecological Status’ in all surface waters and groundwaters that are not determined to In 2001, the Maine State Legislature passed a law requiring ‘water use standards for maintaining instream flows…lake or pond water levels…protective of aquatic life and other uses…based on the natural variation of flows’. The resulting environmental flow and water level protection rule, finalized in 2007, establishes a set of tiered flow protection criteria Copyright © 2011 John Wiley & Sons, Ltd. River Res. Applic. (2011) DOI: 10.1002/rra B. D. RICHTER ET AL. Table I. Standards for UK river types/subtypes for achieving Good Ecological Status, given as per cent allowable abstraction of natural flow (thresholds are for annual flow statistics) Type or subtype A1 A2 (downstream), B1, B2, C1, D1 A2 (headwaters) C2, D2 Salmonid spawning and nursery areas Season Flow >Q60 Flow >Q70 Flow >Q95 Flow <Q95 Apr–Oct Nov–Mar Apr–Oct Nov–Mar Apr–Oct Nov–Mar Jun–Sep Oct–May 30 35 25 30 20 25 25 20 25 30 20 25 15 20 20 15 20 25 15 20 10 15 15 Flow >Q80 15 20 10 15 7.5 10 10 Flow <Q80 From Acreman and Ferguson (2010). linked to different stream condition classes (Maine DEP, 2010a). The environmental flow standards may be established by one of two methods: a standard allowable alteration of flow or a site‐specific flow assessment. The standard allowable alteration is based on the natural flow regime theory (Poff et al., 1997; Richter et al., 1997) and was informed by considerable scientific research on environmental flow requirements for the eastern USA (e.g. Freeman and Marcinek, 2006). For all streams falling into the state’s best‐condition class (class AA), 90% of the total natural flow must be maintained when the flow exceeds the spring or early winter ‘aquatic base flow’ (Maine DEP, 2010b). This aquatic base flow is defined as the median monthly flow of the central month of each season (Maine DEP, 2006). In other seasons, withdrawals of up to 10% of daily flow can only occur when daily flows exceed 1.1 to 1.5 times the seasonal aquatic baseflow. No flow alteration is allowed in any season when flows are below aquatic base flow levels. In addition, all rivers and streams that flow into class AA waters must meet the POF standard. Although used only for those waters with the highest ecological condition goals, which make up approximately 6% of state waters, the Maine standard provides a good example of use of a hands‐off flow level combined with a POF approach. Summary of case study findings The case studies summarized here have much in common (Table II). In each case, standards were developed with a general intent to avoid ecological degradation of riverine ecosystems. The specifics of management goals vary from case study to case study, but common among them is the desire to maintain ecological conditions that are good to excellent or to avoid ecological harm. Each of these efforts to set standards has utilized the best available science for their region, and each has engaged large numbers of scientists familiar with flow–ecology science, using expert‐ based decision‐making processes. We found the recommendations for flow protection emerging from these expert groups to be quite consistent, typically resulting in a range of allowable cumulative Table II. Summary of per cent‐of‐flow environmental flow standards from case studies Location Florida (SWFWMD) Michigan Maine European Union Ecological goal Avoid significant ecological harm (max. 15% habitat loss) Maintain baseline or existing condition Protect class AA: ‘outstanding natural resources’ Maintain good ecological condition Cumulative allowable depletion 8–19% of daily flow 6–15% of August median flow 10% of daily flow 7.5–20% of daily flow 20–35% of daily flow Copyright © 2011 John Wiley & Sons, Ltd. Considerations Decision process Seasonally variable extraction limit; ‘hands‐off’ flow Single extraction limit for all flow levels Single extraction limit for all flow levels above a ‘hands‐off’ flow level Lower flow; warmer months; ‘hands‐off’ flow Higher flow; cooler months Scientific peer review of site‐specific studies Stakeholders with scientific support Expert derived Expert derived River Res. Applic. (2011) DOI: 10.1002/rra PRESUMPTIVE STANDARD FOR ENVIRONMENTAL FLOW PROTECTION depletion of 6% to 20% of normal to low flows, but with occasional allowance for greater depletion in seasons or flow levels during which aquatic species are thought to be less sensitive (Table II). These results suggest a consensus that modest alteration of water flows can be allowed with minimal to no harm to aquatic ecosystems and species. A PROPOSED PRESUMPTIVE STANDARD Our review of the case studies described above suggests that an appropriate presumptive standard for environmental flow protection can be proposed at this time, subject to some important caveats. We suggest that a high level of ecological protection will be provided when daily flow alterations are no greater than 10%; a high level of protection means that the natural structure and function of the riverine ecosystem will be maintained with minimal changes. A moderate level of protection is provided when flows are altered by 11–20%; a moderate level of protection means that there may be measurable changes in structure and minimal changes in ecosystem functions. Alterations greater than 20% will likely result in moderate to major changes in natural structure and ecosystem functions, with greater risk associated with greater levels of alteration in daily flows. These thresholds are well supported by our case study review, as well as from our experiences in conducting environmental flow assessments for individual rivers (e.g. Richter et al., 2003, 2006; Esselman and Opperman, 2010). This level of protection is also generally consistent with findings from regional analyses such as the ‘benchmarking’ study in Queensland, Australia, by Brizga et al. (2002) and by a national (US) analysis of hydrologic alteration which documented that biological impairment was observed in some sites with hydrologic alteration of 0–25% (the lowest class of alteration assessed) and in an increasing percentage of sites beyond 25% hydrologic alteration (Carlisle et al., 2010). This presumptive standard can be represented graphically as shown in Figure 2, using the convention of the SBA (Richter, 2009), with risk bands bracketing the daily natural flow conditions. When a single threshold value or standard is needed, such as for corporate risk screening or water supply planning purposes, we suggest that protecting 80% of daily flows will maintain ecological integrity in most rivers. A higher percentage of flow (90%) may be needed to protect rivers with at‐risk species and exceptional biodiversity. Whereas we believe that such a presumptive standard of limiting daily flow alterations to 20% or less is conservative and precautionary, we also caution that it may be insufficient to fully protect ecological values in certain types of rivers, particularly smaller or intermittent streams. Seasonal adjustments of the per cent of allowable depletion may be advisable. Several of our case studies utilized ‘hands‐off’ flow thresholds to limit impacts to the frequency and duration of low‐flow events. This may be an additional consideration where fish passage, water quality or other conditions are impaired by low flows. Also, when applying this presumptive standard to rivers affected by hydropower dams, imposing our suggested limits on daily flow averages may be insufficient to protect ecological integrity because of the propensity for peaking power operations to cause river flows to fluctuate considerably within each day. In such cases, our presumptive standard may need to be applied on an hourly, rather than daily, basis. Adjustments to our suggested values Figure 2. Presumptive standards are suggested for providing moderate to high levels of ecological protection. The greater the departure from natural flow conditions, the greater is the ecological risk to be expected. This figure is available in colour online at wileyonlinelibrary.com/journal/rra. Copyright © 2011 John Wiley & Sons, Ltd. River Res. Applic. (2011) DOI: 10.1002/rra B. D. RICHTER ET AL. should be considered when local or regional ecological knowledge indicates that narrower bands of allowable alteration are needed. Most importantly, continued investment in detailed, site‐ specific or regional environmental flow assessment is urgently needed. Such research must continue to inform our understanding of flow–ecology relations and refine our presumptions about the adequacy of protecting different percentages of natural flows. MANAGEMENT IMPLICATIONS To properly apply our presumptive standard, water managers and other water stakeholders, such as corporations concerned about the sustainability of water use in particular river basins, will need to be able to do three basic things: • Develop modelling tool(s) to estimate natural (unregulated and undepleted) flows on a daily basis; this provides the natural or ‘baseline’ flow data illustrated in Figure 1; • Use the modelling tool(s) to evaluate whether proposed withdrawals, dam operations or other changes—when added to already‐existing water uses—would cause the presumptive standard to be violated; • Monitor daily flows at key locations, such as upstream and downstream of major water withdrawals and return flows, and at points of inflow to reservoirs, as a means for verifying and refining the modelling results and for regulatory enforcement purposes. The capability to evaluate proposed hydrologic changes (second bullet in the above list) enables water managers to avoid issuing water use permits that would cause hydrologic variations to deviate outside of the sustainability boundaries set by the presumptive standard (±20%). Obviously, if a particular river’s flow regime has already been altered more than ±20% during part or all of the time, water managers and stakeholders would need to decide whether to restore flows to a level consistent with the presumptive standard or adopt some other standard. Application in over‐allocated basins Ongoing efforts to develop sustainable approaches to water management in the Murray‐Darling river basin in Australia offer a highly relevant and useful example of re‐balancing environmental and economic goals in a previously over‐ allocated basin. In response to considerable ecological degradation, heavy competition among water users, prolonged drought and climate change projections, the Commonwealth Parliament in 2007 passed a national water act calling for the development of a basin plan that would provide for integrated and sustainable management of Copyright © 2011 John Wiley & Sons, Ltd. water resources (MDBA, 2009). The Basin Plan is required to set enforceable limits on the quantities of surface water and groundwater that can be taken from the basin’s water resources. These limits must be set at a level that the Murray‐Darling Basin Authority, using the best available scientific knowledge, determines to be environmentally sustainable. This is defined as the level at which water in the basin can be taken from a water source without compromising the key environmental assets, the key ecosystem functions, the productive base or the key environmental outcomes of the water source. Considerable scientific analysis is being undertaken to determine environmental water requirements that will inform the determination of ‘sustainable diversion limits’. Recent appropriations of federal funding to enable the buyback of historical entitlements can be used to reduce water usage to levels compatible with these diversion limits (Garrick et al., 2009). The scientific assessment and decision‐ making being undertaken in the Murray‐Darling basin exemplifies a situation in which our presumptive standard would have been violated by past water allocations, yet water managers and stakeholders are now striving to restore a level of ecological health and water use sustainability similar to the goals of our presumptive standard. Technology requirements The technology and capacity to manage water in this manner exist in many parts of the world, but we acknowledge that many water management institutions and corporations have not yet developed hydrologic modelling tools with the required level of temporal resolution (i.e. daily) to implement our presumptive standard. Similarly, few countries have been able to install and maintain daily flow monitoring networks with adequate spatial distribution to facilitate data collection and regulation of water uses in the manner we suggest. However, recent and ongoing advances in modelling approaches and technologies, as well as improvements in flow monitoring instrumentation, are driving down the expense of implementing this type of water monitoring and modelling programme. Given growing water scarcity and its economic implications, investment in this level of water management capacity should be given high priority by governments at all levels. We recognize that many water planners continue to use hydrologic models that operate on a monthly time step. We can offer some guidance and caution. Although it is consistent with our presumptive standard to assume for planning purposes that 20% of the natural monthly mean flow can be allocated for consumptive use, this does not mean that a volume of water equivalent to 20% of the monthly mean can be allocated on a fixed basis without violating our presumptive standard. We illustrate this point River Res. Applic. (2011) DOI: 10.1002/rra PRESUMPTIVE STANDARD FOR ENVIRONMENTAL FLOW PROTECTION with a simple hypothetical example. Let us say that the mean monthly flow in July is 100 m3/s. You allocate a sum total of 20 m3/s (20% of mean) for that month. Our presumptive standard will be violated each day in July that natural daily flows (recorded upstream or modeled) drop below100 m3/s, which will be the case during the majority of the time for most river types. Therefore, the only way to be assured that our presumptive standard will not be violated given a monthly allocation will be to subsequently model the system at a daily time step to check for compatibility with the standard under the range of flows typically experienced by the river. Once such compatibility is assured, the water authority can confidently grant water use permits based on fixed amounts (i.e. monthly allocations or continuous rates of use) that provide the water user with desirable certainty. Utility for water planning Although implementation of our presumptive standard will require considerable investment in adequate technology and expertise as outlined previously in this paper, we want to emphasize that our presumptive standard will also be quite useful for initial water planning purposes that require less technological investment. As discussed in our introduction, many large corporations have become quite concerned about their water‐related business risk and are interested in approaches that can help them screen for such risk across many facilities and parts of their supply chains. We suggest that our presumptive standard will be highly appropriate in risk screening, wherein estimates of water availability and use are available for river basins of interest. Our presumptive standard can be used to identify river basins in which water flows appear to have been altered by more than 20%, thereby posing considerable potential risk. In this sense, we are pleased to see the incorporation of a variation of our presumptive standard in the Water Footprint Assessment Manual (Hoekstra et al., 2011), which is already being used by many corporations. Implications for water supply and storage We recognize that in most hydrologic settings, storage will be required to enable full utilization of up to 20% of the available daily flow for consumptive use. Creating such storage can lead to ecological impacts (such as impediments to fish migrations or blocking sediment transport) that can undo the ecological benefits that our presumptive flow standard is trying to protect. Therefore, we strongly urge water managers and engineers to employ innovative options for water storage—such as off‐stream reservoirs or groundwater storage—that do not involve on‐stream reservoirs. Alternatively, in systems in which storage reservoirs already exist, enlarging the capacity of those existing facilities will in most cases be far preferable to building new reservoirs. Copyright © 2011 John Wiley & Sons, Ltd. Some water managers will feel excessively constrained by having to operate within the constraints of the presumptive sustainability boundaries suggested here. However, managing water sustainably necessarily implies living within limits (Richter et al., 2003; Postel and Richter, 2003; Richter, 2009). We suggest that a strong social imperative has emerged that calls for setting those limits at a level that avoids damaging natural systems and the benefits they provide, at least as a default presumption. Where other socio‐economic priorities suggest the need for relaxation of the presumptive sustainability boundaries we suggest here, we strongly encourage governments and local communities to invest in thorough assessments of flow–ecology relationships (Richter et al., 2006; Poff et al., 2010), so that decision‐ making can be informed with scientific assessment of the ecological values that would likely be compromised when lesser degrees of flow protection are adopted. In our experiences in working with water and dam managers, we have found that a remarkable degree of creativity and innovation emerges when engineers and planners are challenged to meet targeted or forecasted water demands with the least disruption to natural flow patterns. Solving the water equation will require new thinking about how and where to store water, conjunctive use of surface water and groundwater, sizing diversion structures or pumps to enable extraction of more water when more is available during high flows, sizing hydropower turbines such that maximum power can be generated across a fuller range of flows, and other innovations. When such creativity is applied as widespread common practice, human impacts on freshwater ecosystems will most certainly be reduced substantially. REFERENCES Acreman MC, Ferguson AJD. 2010. Environmental flows and European Water Framework Directive. Freshwater Biology 55: 32–48. Acreman MC, Dunbar MH, Hannaford J, Bragg OM, Black AR, Rowan JS, King J. 2006. Development of environmental standards (water resources): final report. Project WFD48. Scotland & Northern Ireland Forum for Environmental Research (SNIFFER): Edinburgh. Arthington AH, Bunn SE, Poff NL, Naiman RJ. 2006. The challenge of providing environmental flow rules to sustain river ecosystems. Ecological Applications 16: 1311–1318. Brisbane Declaration. 2007. The Brisbane Declaration: environmental flows are essential for freshwater ecosystem health and human well‐being. 10th International River Symposium, 3–6 September 2007, Brisbane. Brizga SO, Arthington AH, Choy SC, Kennard MJ, Mackay SJ, Pusey BJ, Werren GL. 2002. Benchmarking, a ‘top‐down’ methodology for assessing environmental flows in Australian rivers. In Proceedings of International Conference on Environmental Flows for Rivers. Cape Town: Southern Waters Consulting. Bunn SE, Arthington AH. 2002. Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30: 492–507. Carlisle DM, Wolock DM, Meador MR. 2010. Alteration of streamflow magnitudes and potential ecological consequences: a multi‐regional River Res. Applic. (2011) DOI: 10.1002/rra B. D. RICHTER ET AL. assessment. Frontiers in Ecology and the Environment 2010; doi:10.1890/100053. Esselman PC, Opperman JJ. 2010. Overcoming information limitations for the prescription of an environmental flow regime for a Central American river. Ecology and Society 15: 6. URL http://www.ecologyandsociety. org/vol15/iss1/art6/. Flannery MS, Peebles EB, Montgomery RT. 2002. A percent‐of‐flow approach for managing reductions of freshwater inflows from unimpounded rivers to southwest Florida estuaries. Estuaries 25: 1318–1332. Flannery M, Chen X, Heyl M, Munson A, Dachsteiner M. 2008. The Determination of Minimum Flows for the Lower Alafia River Estuary. Ecologic Evaluation Section, Resource Projects Department, Southwest Florida Water Management District: Brooksville, Florida. URL http://www. swfwmd.state.fl.us/projects/mfl/reports/mfl_alafia_estuary.pdf [accessed 12 January 2011]. Freeman MC, Marcinek PA. 2006. Fish assemblage responses to water withdrawals and water supply reservoirs in piedmont streams. Environmental Management 38: 435–450. Garrick D, Wigington R, Aylward B, Hubert G (eds). 2009. The nuts & bolts of flow reallocation. Proceedings of a Workshop held February 22nd, 2009 in Port Elizabeth as part of the International Conference on Implementing Environmental Water Allocations. The Nature Conservancy: Boulder, CO. URL http://conserveonline.org/workspaces/eloha/documents/ implementing‐policies [accessed 12 January 2011]. Gillilan DM, Brown TC. 1997. Instream Flow Protection: Seeking A Balance in Western Water Use. Island Press: Washington, DC. Hoekstra AY, Chapagain AK, Aldaya MM, Mekonnen MM. 2011. The Water Footprint Assessment Manual: Setting the Global Standard. Earthscan: London. IFC (Instream Flow Council). 2001. Instream flows for riverine resource stewardship. URL http://www.instreamflowcouncil.org [accessed 12 January 2011]. Maine DEP (Department of Environmental Protection). 2010b. Standards for classification of fresh surface waters. URL http://www.mainelegislature. org/legis/statutes/38/title38sec465.html [accessed 24 September 2010]. Maine DEP (Maine Department of Environmental Protection). 2006. Draft basis statement. Chapter 587—Instream flow and water level standards. URL http://www.maine.gov/dep/blwq/topic/flow/about.pdf [accessed 8 November 2010]. Maine DEP (Maine Department of Environmental Protection). 2010a. Chapter 587—In‐stream flow and lake and pond water levels information site. URL http://www.maine.gov/dep/blwq/topic/flow/index.htm [accessed 24 September 2010]. MDBA. 2009. Issues paper: development of sustainable diversion limits for the Murray‐Darling Basin. MDBA publication no. 48/09. Murray‐ Darling Basin Authority: Canberra, Australian Capital Territory. Copyright © 2011 John Wiley & Sons, Ltd. Orr S, Cartwright A, Tickner D. 2009. Understanding Water Risks: A Primer on the Consequences of Water Scarcity for Government and Business. WWF Water Security Series 4. World Wildlife Fund UK: Godalming. Poff NL, Allan JD, Bain MB, Karr JR, Prestegaard KL, Richter BD, Sparks RE, Stromberg JC. 1997. The natural flow regime: a paradigm for river conservation and restoration. Bioscience 47: 769–784. Poff NL, Richter B, Arthington A, Bunn SE, Naiman RJ, Kendy E, Acreman M, Apse C, Bledsoe BP, Freeman M, Henriksen J, Jacobsen RB, Kennen J, Merritt DM, O’Keefe J, Olden J, Rogers K, Tharme RE, Warner A. 2010. The ecological limits of hydrologic alteration (ELOHA): a new framework for developing regional environmental flow standards. Freshwater Biology 55: 147–170. Postel S, Richter B. 2003. Rivers for Life: Managing Water for People and Nature. Island Press: Washington, DC. Richter BD. 2009. Re‐thinking environmental flows: from allocations and reserves to sustainability boundaries. Rivers Research and Applications 25: 1–12. Richter BD, Baumgartner JV, Wigington R, Braun DP. 1997. How much water does a river need? Freshwater Biology 37: 231–249. Richter BD, Mathews R, Harrison DL, Wigington R. 2003. Ecologically sustainable water management: managing river flows for ecological integrity. Ecological Applications 13: 206–224. Richter BD, Warner AT, Meyer JL, Lutz K. 2006. A collaborative and adaptive process for developing environmental flow recommendations. River Research and Applications 22: 297–318. SABMiller, WWF‐UK. 2009. Water futures: working together for a secure water future. URL http://www.sabmiller.com/files/reports/ water_future_report.pdf [accessed 4 December 2010]. Seelbach PW, Zorn TG, Allan JW, Hamilton DA. 2009. A suite of six environmental flow tools used to frame new water management policy in Michigan. Presentation to the North American Benthological Society Annual Meeting, Grand Rapids, Michigan. Smakhtin V, Revenga C, Döll P. 2004. Taking into account environmental water requirements in global‐scale water resources assessments. Comprehensive Assessment Research Report 2. Comprehensive Assessment Secretariat: Colombo. Tennant DL. 1976. Instream flow regimens for fish, wildlife, recreation and related environmental resources. Fisheries 1: 6–10. WWAP (World Water Assessment Programme). 2009. The United Nations World Water Development Report 3: Water in a Changing World. UNESCO/Earthscan: Paris/London. Zorn TG, Seelbach PW, Rutherford ES, Wills TC, Cheng ST, Wiley MJ. 2008. A regional‐scale habitat suitability model to assess the effects of flow reduction on fish assemblages in Michigan streams. Michigan Department of Natural Resources, Fisheries Research Report 2089, Ann Arbor. River Res. Applic. (2011) DOI: 10.1002/rra