Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Overview INTRODUCTION Realizing that the protection and enhancement of the quality of the nation's surface water and ground water resources had become a priority concern, and that the effective management of these resources requires information on current water quality conditions and trends in their condition, the USGS began to develop a national water quality assessment in 1984. In FY 1986, Congress appropriated funds to initiate the National Water Quality Assessment (NAWQA) pilot program to test and refine the concept and approaches for such an undertaking, and to evaluate the potential use and cost of a fully implemented program. The overall goals of the NAWQA program are to: 1. provide a nationally consistent description of current water quality conditions for a large part of the nation's water resources; 2. define long-term trends (or lack of trends) in water quality; and 3. identify, describe, and explain, to the extent possible, the major factors that affect observed water quality conditions and trends. The program is to be executed through a large set of separate investigations of river basins and aquifer systems, referred to as study units. The USGS postulated that by performing NAWQA as an aggregation of many individual study units, the assessment would provide results that would be useful in understanding and managing the water resources of the study unit, and in an- swering national-scale questions about current conditions, trends, and factors that affect water quality. Further, the program is to focus on conditions that are large scale and persistent in time. 16
Qver~fe~ 77 Emphasis ~10 be placed on reglonsl dcgradst10n of watcr qualky sucb as might occur from both nonpoint and polot sources of contsminan1~ In addltlon 10 collectlog ~ster quality dsts, 1he N^WQ^ program ~ deslgocd 101ske advantage of watcr quslRy lnforma- don compllcd by 01her sgcnclcs for various purpose~ lhus, ono of thc first sctlvlilos ~libln csch study unl1 ls to collsic sod lDtccpre1 1hc svallsblc dsts 10 (1) provldc an lul11~1 dcscclptloD of ~s1cr quslRy condldons, (2) dcvclop hypothcscs about major factors lnfluenclng ~ster quslRy, and (~) define dsts needs. Bccausc of thc cmpbash on 1rc~ds ln ~ster qual~y, 1hc progrsm ls 10 bc pcrenalst rccogolzlag 1ha1 thc cmcrgcacc of nc~ hYdroloelc knowlcd~c, 1mprovcd mc1hods of mcasuromcn1, , ,~ ~ ^, . ~ . , . ^ , ., snd changes 1~1hO lypcs OI cODl~mlDBnlS 01 concern mlgn~ rcqulrc 1ba1 1hc progrsm bc appropclatcly modlflcd. Thc pro- gram ls 10 plBCC 8 high cmphasls on repcthlon of mcssurcmcnts over dmc snd on documcnts110n of 1hc methods of dsis coUcc- don sud snslysk sod of 1bc locs110ns snd characicrk11cs of data-collcc110n sliest Accordlng to 1hC USGS, ~ fulLscalc NAWQA progrsm w1H provldc useful lnformsilon 10 dcclslonmakcrs who set policy, promul~stc rcgulstlons' cs1abllsh pclorlilcs, or manage watcr rcsourccs. As s1~1ed by 1hc uSGS' ~nformailon on 1hc status' 1hc 1rcuds, and 1hc causes of watcr quslRy condldons across 1bc country should bc psrtlcularly usoful to 01hcr sgcnclcs who src involved ln (1) ldentlfylng key substances for posslblc rc~uls110n for which rcscarch ls nccdcd on 1oxlclty, human c~posurc, sud drlnklug-wstcr 1rcatablU1y; (~) sllocs11ng budgotsry rcsourccs among compctlng types of ~stcr qualRy problems; (~) dctcr- mlulng whcthcr dcslrcd goals for ~stcr qualky lmprovcmcn1 arc bclng met; (4) dcslgnlng monltorlng programs ln dlffcrcnt parts of thc country gn 1crms of 1hc consbtucnts snslyzcd, ssmpllng locs110ns, ssmpllns fccquency, snd 1lming of ssmplingt (5) tsrgctlng rc~ula110ns for sclcctcd ~stcr qusllty constituents to psrtlcular gcogrsphlc rcglons or hydrologic scttlngs; (6) dctcf- mlnlng 1hc rcla11vc cffccts on watcr qualky of various types of point snd nonpolnt sources; (7) ldcntlfylng squlfcrs rcqulrlng dlffcrcnt 1ypcs snd dc~rces of ~stcr qusllty proicctlon; sod, (~) cvalustlng managemcnt prsctlccs ln 1crms of thclr largc-scalc cffccts on thc v~stcr qusllty of rivcr baslns snd squlfcr systcms" (Illrsch, ct sl~ 1988\ Four surfacc ~'stcr snd tbrcc ground watcr pHot proicct~ rcprcscutlng a dlvcrshy of hydrologlc cnvlron mcnts and ~sicr quallty condlilons' ~crc sclcctcd by thc USGS ln 1986 to tcst and rcflnc thc ssscssmcnt conccpts of N ~ W ~ A. lhc surfacc watcr pllot proiccts scloctcd 1ncludcd thc uppcr IDlnois Rivcr
18 NAWQA Pilot Program basin in Illinois, Wisconsin, and Indiana; the Kentucky River basin in Kentucky; the lower Kansas River basin in Kansas and Nebraska; and the Yakima River basin in Washington. For the three ground water pilot projects, the USGS selected the Carson basin in western Nevada and eastern California, the Central Oklahoma aquifer in Oklahoma, and the Delmarva Peninsula in Delaware, Maryland, and Virginia. A local liaison committee was established for each pilot project (study unit) consisting of representatives from federal, state, and local agencies and pri- vate organizations involved in water and land management within the area of the project. The charge to each liaison com- mittee was to assist the USGS by ensuring that the scientific information collected by the pilot project was relevant to local and regional interests. To advise the USGS on the overall pilot project program, a National Coordinating Work Group (see Appendix D) also was created with members representing various federal agencies and nonfederal organizations having an involve- ment or interest in water quality information. The committee's assignment to evaluate NAWQA began with a meeting in October 1988. At this meeting, USGS personnel reported that over the prior four years, NAWQA had undergone considerable development, and as a result, a number of refine- ments and modifications of the basic plan had been incor- porated. In fact, the USGS expressed the view that NAWQA would continue to evolve, with certain aspects being further refined and modified, over the next several years. The com- mittee was invited by the USGS to become a part of this evolu- tionary process by making suggestions for improvement or simply by challenging any of the various elements of NAWQA. To assist the committee in its assignment, USGS personnel presented a series of briefings covering the details of all the various elements of NAWQA. During the course of its review, the committee also examined many publications and documents provided by the USGS (see Appendix B) and evaluated the potential usefulness of NAWQA in meeting national, state, and local needs for water quality information by interviewing repre- sentatives of various government agencies (see Appendix E) and the private sector. Additionally, the committee visited, in small teams, five of the seven sites selected as pilot projects: the Carson basin aquifer, Upper Illinois River basin, Yakima River basin, Kentucky River basin, and Central Oklahoma basin aquifer. Meetings were held with USGS project personnel and the local liaison committee. A committee representative also attended several meetings of the National Coordinating Work Group.
Overview 19 The committee, in its deliberations, reviewed all the elements of NAWQA and as a result, identified areas of concern and made suggestions for change. These concerns and suggestions, along with other comments, both positive and negative, were communicated to USGS by the committee through an interim report dated September 25, 1989 (see Appendix A). A major conclusion of the committee, as expressed in the interim report, was "that a national-scale, long-term water quality assessment is in the best interest of the country." This final report addresses those elements of NAWQA dis- cussed in the interim report, but in greater detail. It also eval- uates other considerations deemed important by the committee in designing and implementing a long-term assessment of the qual- ity of the nation's surface and ground waters which, in turn, will produce useful information for those involved in making decisions regarding the management of the nation's water resources. Unfortunately, because the scheduled 4-year study period for the seven pilot projects had not elapsed at the time of the preparation of this report, the committee did not have access to any final products to review, with the exception of five retrospective reports. This has limited the committee's ability to evaluate the anticipated results and usefulness of NAWQA. Therefore, the committee's findings and recommendations are based on the review of many draft documents, briefings by USGS personnel, and the committee's own experience and knowl- edge of surface and ground water quality monitoring and assess- ments. NEED FOR A NATIONAL ASSESSMENT OF WATER QUALITY The committee is convinced that there is a genuine need for a long-term, large spatial scale national assessment of water quality in the United States. Human health and environmental health are inextricably linked to our nation's water quality. As our population continues to grow, our water resources are be- coming more intensively developed, and more potential con- taminants are being produced. Water quality has become an increasingly important component of our political, economic, social, and environmental decisionmaking. Because such deci- sionmaking affects the quality of each individual's life, as well as public and private expenditures of billions of dollars, it cannot proceed without adequate information and understanding.
20 NAPTHA Pilot Program Many significant past and future decisions involving water quality are of national or regional scope. This broad scope arises for several reasons. First, hydrologic boundaries do not follow political boundaries. Therefore, water quality issues are often interjurisdictional. For example, many hydrologic systems, e.g., river basins, lakes, or aquifers, are large and fall within or are adjacent to more than one political unit. These systems are dynamic, flowing systems through which changes propagate over space and time, so that upstream decisions affect downstream users. Second, a number of important water quality problems are widespread throughout the nation. Examples include storm- water runoff quality control and municipal and industrial wastewater treatment. These problems are so widespread that it is often more efficient to make some decisions about them at the national or regional level. Finally, some water quality problems are characterized by long time scales, so that decisions made at one point in time carry impacts far into the future. To the extent that higher levels of government provide continuity over time, these issues may require a national or regional approach. There are many examples of water quality issues requiring or benefiting from national or regional attention and decision- making for one or more of the reasons just discussed. These include evaluating past and guiding future investments in waste- water treatment works; determining the relative contribution of point and nonpoint sources to the loading of contaminants to surface and ground waters; identifying and controlling the water quality impacts of acid deposition, agricultural chemical use (especially pesticides), and tonics; evaluating the effectiveness of federal, state, regional, and local environmental regulations; and controlling eutrophication of inland and coastal water bodies. The future is likely to bring even more issues requiring a large- scale focus, such as determining the value of instream water uses relative to water resources development. This issue has implica- tions for general environmental policy, Indian and non-Indian water rights, and the preservation of threatened and endangered species, among others, and will require attention at many levels of government. Sound decisionmaking about these and many other water quality issues requires that we identify problem areas before they reach crisis proportions, understand the causes of such problems, and are able to predict adequately the effects of changes in water quality and the impacts of attempts to improve or protect water quality. In other words, we need (1) data quantifying hydrologic, chemical, biological, and other
Overview 21 relevant parameters in space and time; (2) information about the past and present states of the system obtained by collating, organizing. and interpreting the available data: and. {3) knowl- _ _ _ . ~ ~ . . . . · . , · , , _ _ edge about the cause and ettect relatlonsnlps Between varlaoles and their evolution over space and time capability. All three are important and build on each other, but ultimately knowledge and understanding, which are essential for predictive capability, must be the goal of any program that supports water quality decision making. As we define it, then, a water quality "assessment" must do much more than "monitor." In our usage, monitoring is a data- collection activity typically directed toward assuring compliance with a regulation or statute, detecting the presence of known contaminants, or operating control facilities and systems. Assess- ment, on the other hand, goes well beyond monitoring and data collection to include the analysis, interpretation, and synthesis of data and theory to enhance our understanding of the environ- ment. While data collection activities are necessary, and indeed are one important component of an assessment program, we are convinced that the strongest current need is for a true national assessment, focusing on enhancing knowledge and understanding. There are several timely examples of the value of such an assessment. One is the issue of pesticides in surface and ground waters. The distribution, mobility, and fate of pesticides in the aquatic environment are controlled by a complex set of physical, geochemical, and biological processes. Mere detection provides no information about sources, pathways, or fate. In addition, detection of a pesticide in one environment typically provides little information about the presence of the same pesticide in another hydrogeochemical environment. For example, aldicarb is often associated with high ground water tables and sandy, mineral soils, where its fate and transport are controlled and its mobility limited by sorption on mineral surfaces and microbio- logical degradation under fully saturated conditions. However, in the presence of a large unsaturated zone and/or more organic soils, aldicarb may behave quite differently because of the potential for partitioning into the soil gas and organic solids. Designing management practices to control contamination by this pesticide and then evaluating those practices cannot be ac- complished without understanding the mechanisms responsible for its fate and transport. A second example is the presence of selenium in agricultural drainage waters in such places as the San Joaquin Valley in California. Effective control and management will be possible
22 NAPTHA Pilot Program only after the complex hydrologic interactions between surface and ground waters are understood and the geochemistry of selenium sorption and oxidation-reduction reactions is delineated. An extensive, in-depth study has been required in order to interpret the initial detections of selenium and to develop poten- tial control and management options (Gilliom, et al., 1989~. Two different (and often competing philosophical ap- proaches can be used to address a complex problem such as a national assessment of water quality. In a purely statistical approach, the collection and analysis of data are based on statis- tical theory. In other words, variable behavior and the relation- ships between variables are assumed to be dominated by random uncertainty. In a process-oriented approach, sampling and data analysis are largely driven by deterministic models of the rele- vant physical processes. It must be stressed that these descrip- tions represent the extremes of a spectrum of approaches. In practice, it is rare to find an approach that is purely statistical or purely process oriented. Statistical approaches are most effective when they exploit an understanding of relevant physi- cal processes, and statistical methods are essential tools for process modeling in the face of data uncertainty and model simplifications. Nonetheless, it is useful to distinguish between these two basic approaches, since the challenge in any given situation is to find the appropriate mix. Because the committee is convinced of the need to develop a much greater understanding of our nation's water quality, it has reached the conclusion that a national assessment must take a strong process-oriented approach. While data uncertainty and conceptual simplifications must be properly addressed, the assess- ment must maintain a strong focus on elucidating cause and effect relationships and developing models that articulate those relationships. In order to meet the goals of a national water quality assess- ment, the assessment must also be long term. First, because of the extraordinary complexity of the physical, chemical, and biological processes controlling water quality, any assessment of the state of water quality in the U. S. must evolve over time, probably iteratively, as our understanding and data bases in- crease. Second, the processes controlling water quality take place over a wide range of time scales. For example, ground water flow rates are very small, and a "snapshot in time," or even several closely spaced snapshots, would provide relatively little information about change. Similarly, the impacts of global climate change on water quality are likely to occur on time scales of decades. On the other hand, mixing processes in moun-
Overview 23 lain streams are very rapid, so that a single sample, or even a few samples, could easily miss important events. In either case, a static, one-time assessment would have no lasting value, but a long-term assessment would have a better chance of detecting the true water quality and its changes. Both the complexity of water quality processes and the wide range of relevant time scales imply that adaptability is an ex- tremely important characteristic of a successful water quality assessment. As new knowledge is gained, new methods devel- oped, or new contaminants discovered or introduced in the environment, or as an existing condition evolves over time, an assessment program must respond and change. A flexible long- term assessment would make that responsiveness possible. A national water quality assessment necessarily warrants a large-scale undertaking. Because of the complexity and spatial diversity of water quality issues, a national-level aggregation and integration would be invaluable in maximizing information gained from local experience. Such integration would enhance the ability to generalize from local experience and to adapt knowledge learned from one location to another. Because of complexity and diversity, multiple lines of evidence are often required to develop necessary understanding. A large-scale assessment makes it possible to develop such lines of evidence. Finally, while there is much completed and ongoing research focusing on cause and effect water quality relationships, this research tends to be directed toward smaller-scale (often labora- tory-scale) understanding. Relatively less is known about the behavior of large systems, such as entire river basins or aquifer systems. For this reason, a national assessment would! be a particularly timely undertaking for scientific reasons alone. Because of the many advantages of a large-scale, long-term water quality assessment, there is tremendous value in devel- oping consistent, compatible, reliable, and accessible water qual- ity data bases. Unfortunately, there is often relatively little consistency between data sets gathered for local or regional purposes. Consequently, generalization and inference at the national level or across state boundaries or from year to year is very difficult. The USGS study of the effects of changes in municipal wastewater treatment on water quality in the Upper Illinois River Basin provides an excellent example of the dif- ficulties caused by inconsistent data bases (see Appendix B. #57~. To summarize, implementing a national water quality assess- ment using consistent data collection, analysis, and reporting procedures is essential if we as a nation are to effectively and efficiently maintain, manage, and control our water resources.
24 NAUSEA Pilot Program Such an assessment must go well beyond mere monitoring and data collection to focus on developing understanding of cause and effect relationships. It should be process oriented, long term, highly adaptable, and of large spatial scale. While the committee is convinced of the need for a national assessment, this is an enormously difficult challenge because of the immense scale of our nationts waters, the diversity of both the natural hydrologic systems and the human activities that affect those systems, and the complexity of the physical, chemical, and biological processes that govern water quality. There is a vast scope to the types of water bodies of importance, encompassing rivers and streams, estuaries, lakes and reservoirs, and ground water aquifers. These water bodies are combined into hydrologic systems with complex interactions between components. They range in size from small streams to extensive aquifers. Important processes occur on scales ranging from microscopic to global and encompass a broad array of scientific disciplines, including hydrology, geology, chemistry, micro- biology, ecology, engineering, and more. Water quality problems range from naturally occurring radon in ground water to the impacts of wastewater discharge on downstream water users. There are many implications of such a vast scope. First of all, except in a few special cases, uniform national assessments are precluded. Understanding must almost always be developed on regional or smaller scales and a national picture must be assembled as a composite of these smaller-scale assessments. Rigorous probabilistic generalizations at the national scale are possible only for a small subset of relatively simple problems that do not require cause and effect analysis, e.g., number of stream miles with low average dissolved oxygen concentrations. Second, a national assessment must be a multidisciplinary under- taking and a work environment and management structure must be established that fosters interactions between different dis- ciplines. For example, the traditional separateness of surface and ground water hydrologists, as well as of physical and life scientists, must be overcome. Third, there is an existing struc- ture for collecting and interpreting a large amount of water quality data. This effort is highly dispersed across many dif- ferent public and private organizations and involves data col- lected for a wide variety of different purposes. Careful coor- dination is essential to avoid duplication of effort and maximum utilization of resources and existing knowledge, and to ensure consistency. Finally, federal agencies have very little experience implementing a truly national assessment of any particular water quality issue, let alone a national assessment of water quality as a whole.
Overview 25 Therefore, such an assessment cannot proceed quickly and will require adequate resources. Conceptual approaches, models, and other technology will need to be developed as an assessment proceeds. At no point can a national assessment become a rou- tine task. Consequently, the success of a national assessment will be highly dependent on the quality of the people directing and implementing it. Staff must be very capable, well educated, broadly experienced, creative, and motivated. To conclude, then, the committee is convinced of the need for a national assessment of water quality. However, such an assessment will face a number of difficulties. It will be of vast scope, it will be highly multidisciplinary, it will need to be well coordinated with the activities of many different organizations, and it will require many experienced, high-quality personnel provided with adequate time and resources. Uniform national analyses or rigorous probabilistic generalizations cannot be expected when a national scope is achieved by assembling a composite of regional or smaller-scale analyses.
