Ground Water at Yucca Mountain: How High Can It Rise? (1992)

The charge by the National Research Council to the Panel on Coupled Hydrologic/Tectonic/Hydrothermal Systems at Yucca Mountain was to evaluate an earth science issue, the potential effects on the water table of various natural processes in the Yucca Mountain area. The questions addressed by the panel dealt with the potential for ground water to rise more than 200 meters to the level of the proposed repository below Yucca Mountain: “Has it happened? and Can it happen?” The panel was not asked to, and did not, evaluate the suitability of Yucca Mountain as the location for a high level radioactive waste repository. Nevertheless, although the panel does not take a position regarding Yucca Mountain as a repository site, it recognizes the importance of the responsibility given to the Department of Energy by Congress. The decision concerning the long-term viability of any site being evaluated for a mined geological disposal system (MGDS) must be based on exacting, thorough, and well-coordinated and integrated scientific investigations, if the results are to lead to an understanding of the complex natural systems. It is this understanding that is necessary for prediction of the future behavior of those systems with some acceptable level of confidence. Moreover, predicting for a ten-thousand-year time period, with an exactitude unprecedented in human endeavors, will require a large number of data and substantial understanding of the interactions of those systems. Even with extensive data, predictions will depend heavily on expert judgment. The more that is known and understood about those systems and their

interactions, the greater will be the confidence with which those judgments will be made.

This chapter summarizes what the panel considers its most important conclusions and recommendations that appear throughout the text. It does not cover every finding and recommendation in the report. On the other hand, some of the considerations and recommendations presented here are not directly related to the charge and are therefore not discussed in the body of the report, but bear on the scientific needs of the site characterization program identified by the panel. They may not be included in the present site characterization plan or be given what, in the panel's judgement, is the appropriate degree of attention.

It should be noted that the charge to the panel included an evaluation of the particular concepts described in the report by Szymanski (1989). Those concepts involved seismic pumping as the primary mechanism for driving the deep ground water to the surface in a cyclic progression of crustal stress changes. The panel evaluated the geologic evidence presented for this process and found both the evidence and the seismic pumping model inadequate to support the consequences attributed to them. As the panel was concluding its studies, the “minority” members of the 5-member external review panel selected by DOE and Szymanski to review his report informed the NAS panel that both the interpretation of some of the evidence and the model itself had changed: that Szymanski no longer believed that seismic pumping alone could drive the water up as high as he had stated in his report, and that he now had a new concept involving a thermally driven hydrotectonic cycle. This information was presented at the NAS panel's last meeting. Although there was no time left for the NAS panel to give consideration to a new thesis, nor was there a written document that could be evaluated, the cyclical concept as presented to the NAS panel appeared to have little validity, given that the panel is convinced that the geologic evidence refutes the assertion that ground water has risen repeatedly 100 meters or more in the recent geologic past. Because an essential part of the “cycle” has not yet happened, there is no basis for postulating a cyclical process whatever the proposed mechanisms involved.

The panel took two approaches in its examination of the potential for flooding of the proposed repository at Yucca Mountain by a rise in the ground-water level for a prolonged period. These were (1) to see if there was geological evidence for any large water-table excursions throughout the late Quaternary (last ca. 120 ka), and (2) to examine the results from model calculations that coupled hydrologic responses to tectonic, volcanic, and climatic changes. Given the known geologic and tectonic history of the area, these appeared to

the panel to be the processes most likely to be able to affect the water table. Inasmuch as the only deposits associated with hydrothermal processes in close proximity to Yucca Mountain were formed more than 10 Ma during formation of the tuffs, and the only Quaternary evidence for warm springs observed by the panel was more than 55 km from Yucca Mountain, at Travertine Point (from the earliest Quaternary (2 Ma-700 Ka)), the panel discounted hydrothermal systems as a potential mechanism for raising the water table level in the Yucca Mountain area.

The panel spent several days in the field on separate occasions with proponents of the hypothesis that upwelling ground water caused the surface-parallel soil carbonates and the calcite-silica vein deposits in Trench 14 and elsewhere in the Yucca Mountain area, and with others claiming a surface soil-development, or pedogenic, origin for those deposits. On the basis of the panel members' knowledge, experience, and judgment, which were brought to bear on their observations of field geologic features, none of the evidence cited as proof of ground-water upwelling in and around Yucca Mountain could be reasonably attributed to that process. A few occurrences were equivocal, and some indeterminate, on the basis of field observation alone, but the preponderance of features (1) were clearly related to the much older (13-10 Ma) volcanic eruptive process that produced the tuffs in which the features appear, (2) contained contradictions or inconsistencies that made an upwelling ground-water origin geologically impossible or unreasonable, or (3) were classic pedogenic features recognized worldwide. Some examples follow.

The stages of morphological development of the carbonate accumulation within soils correlated well with the relative age of the various geomorphic surfaces (alluvial fan, pediment, sand ramp, etc.) throughout the area examined, as would be expected if the carbonate developed through progressive pedogenic processes. No such correlation would be expected if the carbonates developed from spring or ground-water upwelling.

On Busted Butte, the slope-parallel surficial carbonate deposits were attributed by proponents of the ascending water thesis to upwelling

of ground water along a vertical, vein-filled fault or fracture zone. At the surface, however, the thickness of the surficial deposits decreases everywhere downslope. The downslope gradient of thickness of the carbonate deposit is uninterrupted where a pronounced vertical fault, one presumed source of the carbonate-bearing fluids, intersects the surface. Had this carbonate-filled vertical fracture zone been the source, the deposits downslope would have been thicker than the deposits immediately upslope. Moreover, the panel could find no evidence upslope for other vertical faults that could have served as sources for ground water upwelling that might have led to the continuous, uninterrupted thinning of the surficial carbonates downslope. The exposures on Busted Butte, where steep gullies cut clearly through the single vertical, vein-filled fault, would have clearly exposed such faults that might have provided conduits to permit such a continuous downslope thickness. Moreover, the presence of abundant root casts in the carbonate horizon is a clear indication that the carbonate formed in the soil zone rather than on the surface. The panel concludes that meteoric water and progressive pedogenic processes produced the surface-parallel carbonate deposits. The hypothesized cyclic upwelling of ground water along faults demonstrably does not account for their presence at Busted Butte.

The abundant carbonate veins in the Yucca Mountain area, if they are the conduits of ascending subsurface waters during specific events of the past, should be capped by abundant tufa, or travertine, mounds. The absence of these surface deposits above purported extinct spring openings argues against ascending water. The excellent mound and feeder veins exposed at Travertine Point near the entrance to Death Valley have several features that confirm their hydrothermal origin, features that differ in all aspects from those of the Yucca Mountain area, as described in Chapter 2 of this report and briefly reiterated below. The one mound present in the Yucca Mountain area at Site 199, about 14 km SW of Yucca Mountain, appears to be the site of a still active seep, possibly of a perched water table, rather than the result of some tectonically driven upwelling in the past.

The modern springs present in the Yucca Mountain area and to the north of it all occur at the base of fractured ridges. The discharge of these springs is very small, in the range of 0.01 to 0.1 liters/sec. At least one of these springs, Cane Spring, has been observed on more than one occasion to become more active after a rainfall, with the water emerging at surface water temperatures.

Finally, the isotopic content of the water is similar to the local rainwater, differing only in having somewhat higher concentrations because of evaporation, and is unlike that of the local ground water. These observations lead the panel to conclude that the spring derives primarily in part from rainwater infiltrating through fractures and emerging at the base through the adjacent fractured mountain mass, not from warm water ascending from great depth.

Several features at Travertine Point are characteristic of deposits from ascending thermal waters: a tufa mound, carbonate veins with coarse-grained calcite (greater than 6 mm up to 1 cm), mirror-image symmetrical banding, and the absence of interlayered bands of amorphous silica, chalcedony or quartz. In contrast, the veins in Trench 14 and other exposures in and around Yucca Mountain show no mounds, are composed of extremely fine-grained calcite (less than 6 microns, or one thousandth the size of the Travertine Point calcites), have no symmetry, and contain thin interlayered bands of low-temperature, amorphous silica, and other materials, such as clay and volcanic ash, commonly found in desert soils. The panel concludes from these features that the trench veins formed under low temperature, descending and evaporating meteoric water conditions, which implies an origin by surficial processes.

The widespread occurrences of breccias cemented by carbonate or carbonate-silica, which are considered evidence of explosive release of highly pressurized ascending fluids by proponents of this thesis, formed by a variety of processes at different times during the geologic history of the region. The origins of the breccias range from formation during the Tertiary high temperature volcanic processes 13-10 Ma to the low temperature mechanical erosion processes that produced the talus at the base of weathering slopes, most likely through the late Quaternary and probably into the Holocene. These breccias, therefore, could not, in the panel's view, constitute evidence for any one process at any one time. However, none of the breccias require formation as a result of upwelling of highly pressurized fluids from great depth. Indeed, the panel concludes that all of the breccias appear to be best explained by processes that do not involve ascension of deep seated fluids.

Those that were not clearly part of the volcanic process, e.g. talus

breccias and fault breccias, had characteristics consistent with a pedogenic origin for the carbonate cements that bind the rock fragments. Carbonate root and twig casts throughout the thickness of the talus breccias suggest progressive development of the carbonate cement simultaneously with accumulation of the talus deposit. Such features are possible only if they resulted from surface processes involving the evaporation of local rain water intermittently flowing downslope carrying dissolved carbonate from the upper slopes. Thin carbonate deposits on the underside of still-uncemented talus blocks of tuff or basalt indicate the process is still continuing.

The fault breccia cements observed by the panel were extremely fine-grained, lacked concentric banding around the fragments, in some cases incorporated detrital minerals of anomalously old ages, clay or volcanic ash derived from the soil, and had isotopic compositions similar to soil carbonates. These characteristics are typical of carbonates formed by pedogenic processes.

The panel concludes that the cements of these fault breccia were deposited by precipitation from evaporating meteoric water as part of the soil-forming processes.

The panel's independent examination of stable and radiogenic isotope data shows that the Trench 14 and Busted Butte vein carbonates formed from the same water and processes as the soils, and that they could not have formed from the present-day ground water of the area. Their carbon (¹³C/¹²C) and oxygen (¹⁸O/¹⁶O) isotopic content overlap the range of values for modern soil carbonates, but differ significantly from the calculated isotopic content of carbonates in isotopic equilibrium with (or derived from) ground water of the isotopic composition measured for the Tertiary/Quaternary aquifer beneath Yucca Mountain. Available data suggest that the isotopic content of ground waters has not changed greatly over the past 300 ka. Additional study is needed, however, of the stable isotope content and isotopic ages of calcite as well as of their fluid inclusions in order to reach reliable conclusions regarding the isotopic content of paleo-ground waters. Moreover, data obtained by the DOE project investigators showed a correlation between depth and the carbon and oxygen isotopic content; δ¹⁸O decreases and δ¹³C increases with greater depth.

Similar results were obtained from the radiogenic isotopes: calcites above the water table showed strontium isotopic ratios that fell within the range of soil calcite and veins, while samples from below

the water table had values within the range for waters from the tuff aquifer. Although some data obtained by the Yucca Mountain project scientists after the panel's analysis showed some inconsistencies in the isotopic composition of some carbonates above the water table in a core from one well, the fact that no other such inconsistencies were found in other wells in the vicinity suggest that it does not indicate a rise in the water table. Further studies are needed to understand the origins and processes involved.

Nonetheless, the isotopic evidence now available indicates that no prolonged excursion of the water table above its present level has occurred in the last ca. 100 ka.

Field evidence from fossils obtained by a panel member and associates several years ago in a study unrelated to that of the panel indicated an absence of perennial spring discharge, or wetlands, in the last 50 ka in the vicinity of Yucca Mountain area, with one significant exception. Wet-ground plant species dated at about 50 ka were found at one site in the middle reaches of now dry Fortymile Canyon, the major drainage east of Yucca Mountain, 60 m above the canyon floor. This need not be interpreted as the result of a significant rise in the water table, because no other nearby fossil sites contain such evidence. It may be related to increased discharge in response to increased recharge in the area north of the steep hydrologic gradient, as described in Chapter 3, or it may result from locally perched water for which some evidence has been observed.

Further efforts should refocus away from the descending/ascending water controversy. Studies should concentrate on improving the knowledge of the ground water history of the Yucca Mountain area, to ascertain the validity of the widely held view that the isotopic composition of ground water of an area does not change much with time and that, therefore, differences in isotopic composition of present day ground water and carbonates at the surface are not the result of differences in the isotopic compositions of present and past ground waters. To that end, the following studies are recommended:

1. Characterization of isotopic age and composition of calcites above and below the water table from cores should be supplemented with information on grain size, chemical variation (especially Ca-Mg-Mn-Sr) and D content of fluid inclusions.

2. Trenching and drilling of the Site 199 tufa mound should be carried out to describe and document the geology, hydrology, and geochemistry of this spring deposit in order to determine if it is the result of a perched water table and if carbonate is present in veins below the surface.

3. Mineralogical, chemical, and isotopic analyses of windblown dust should be conducted in order to determine the magnitude of the contribution of such dust to carbonate deposits.

In considering the various mechanisms that could conceivably cause a significant rise in the water table in the Yucca Mountain area, the panel identified a change from an arid to a pluvial (wet) climate and the consequent increased recharge of the saturated zone as a possible scenario. Analysis of the paleoecological and paleoclimatic information of the area suggests that even at the last glacial maximum during the Pleistocene Wisconsin 18 ka the Yucca Mountain area experienced no more than a 40 percent increase in rainfall over the present. Although there exist some uncertainties as to the climatic conditions during the terminal Wisconsin and early Holocene, most of the available data point to semi-arid to arid conditions for most of the last 50 ka. Thus, any models seeking to calculate water table changes due to increased recharge must take such facts into account.

The only existing model that has examined water table changes due to a change in climate assumed a 100 percent increase in precipitation and a corresponding 15-fold increase in recharge to the ground-water system. The approximately 100 m rise of the water table calculated, using these assumptions, is of concern. The panel recognizes that large uncertainties are associated with methods of calculating recharge, especially in arid and semi-arid regions. The assumed increase in recharge was obtained using an empirical procedure that has been widely applied in Nevada with good results; however, its use for calculating recharge under climatic conditions much different from the present must be viewed with caution. Moreover, strong evidence for an increase in precipitation of no more than 40% of present precipitation during the pluvial climatic episode suggests that the model assumption of a 100% increase is overly conservative.

Nevertheless, according to the only modeling to date, the panel must consider climate change to be a mechanism that has the potential to cause a large rise in the water table.

The panel recommends studies aimed at an improved understanding of the in situ properties and characteristics of the three-dimensional hydrologic system, the paleohydrological setting of Yucca Mountain, and the modern processes that control recharge of aquifers underlying the site to constrain better the models of the effects of increased precipitation. The hydrologic model that is used as a starting point and the results of three-dimensional modeling must be internally consistent with chemical and isotopic variations found in waters in the region. In the body of this report the panel recommends several detailed studies aimed at obtaining important information to achieve these ends.

In assessing what processes are likely to cause a perturbation of the water table, the panel considered the long and complex Tertiary volcanic history of the region. A possible recurrence of the earlier highly explosive silicic volcanism that produced the ash flow tuffs, which are the predominant bedrock of the Yucca Mountain area, was dismissed because the subduction zone that caused it is now extinct in the Great Basin region. Concurrent and subsequent basaltic volcanism, related to the change to extensional tectonics that produced the Basin and Range structure, has experienced a progressive decline in volume, as expressed in the low-volume volcanic eruptions of Crater Flat which bounds Yucca Mountain on the west, and the latest and lowest in volume, Lathrop Wells cone, a short distance to the south.

Thus the geologic record of waning basaltic volcanism indicates that the only likely style of intrusion into the Yucca Mountain area during the lifetime of the repository is a low-volume basaltic dike. Dike occurrences in the area and theoretical considerations of elastic extensional strains discussed in Chapter 4 of this report suggest a near-vertical dike 2-4 m wide. Models of water table responses to a dike intrusion with a top approximately 1 km below the repository horizon, from the points of view of poro-elastic and thermal effects, modeled separately, resulted in rises of the water table of less than 10 m for the former and 12-14 m for the latter. Thus, dike intrusion appears to be inadequate to cause a rise of the water table level of more than 10-20 m.

The calculated probability of occurrence of a dike intrusion that would affect the proposed repository is a very small number, on the order of 10⁻⁸ per year. Although there is considerable uncertainty

in the calculation, this low probability combined with the small effect a dike intrusion would have on the water table means that volcanic effects can be discounted as a primary disruptive event.

To provide a broader basis for predicting water table behavior related to volcanic intrusions, and for establishing probabilities for renewed volcanic activity during the life of the proposed repository, the panel recommends the following additional studies.

1. A study should be undertaken to model the coupling of the poro-elastic and thermal effects of an intruding dike on the water table. This may provide a more realistic basis for predicting the maximum potential effect on the water table in the vicinity of the intrusion. The likelihood of such an intrusion within a significant distance based on such analysis can then be refined accordingly.

2. Earthquake wave studies have identified a columnar zone of low velocity crustal material under Crater Flat extending from the Moho, about 30 km beneath the surface, to about 12 km below. This suggests the possibility of partially molten rock in the form of intrusions at lower crustal depths. To determine the presence or absence of molten rock, the panel recommends more detailed, higher resolution seismic measurements and analysis be undertaken, including the analysis of shear (S) waves, which are more sensitive to the presence of fluids, and the use of fluid-sensing seismic reflection profiling techniques.

Intrusions related to the known Pleistocene basaltic events would not be expected to produce the large volume of low velocity material imaged by the current teleseismic studies. If the material responsible for the velocity anomaly is shown to be partially molten (which could be resolved with the proposed study) it would suggest a different, or at least a more vigorous, style of basaltic intrusion into the crust than is indicated by the recent geologic record.

To evaluate the effects of earthquake-induced changes in crustal stresses on the water table, the panel looked at some of the better known, more recent historic earthquake records and at some model calculations, its own and those of others. The panel was unsuccessful in its attempts to obtain information on water level responses to measured earth strains due to underground nuclear explosions on

the Nevada Test Site that were monitored and about which information exists. In the panel's view, this information may be quite valuable in estimating likely ground-water responses to a nearby earthquake.

While the information the panel was able to consider within the time limits of the study was hardly exhaustive, it provided some idea of the range of what can reasonably be assumed on the basis of contemporary knowledge. The information upon which the panel based its evaluation therefore is suggestive rather than definitive. Many more data and analyses will be required to reduce the uncertainties inherent in judgments made on the basis of the limited site specific information available. However, on the basis of the recent historical record and the results of modeling, the panel concludes that a water table response to seismically induced changes in crustal stresses is at least an order of magnitude less than the amount needed to affect the unsaturated status of the proposed repository.

Given these considerations, along with the apparently low strain rates of the region, the low seismicity (both in magnitude and frequency of occurrence), and the low probability of occurrence of a large earthquake close to Yucca Mountain, the panel finds no reason that site characterization of the area should not proceed as planned.

The panel supports continued site characterization efforts to obtain the critical information necessary for more definitive assessments of the future behavior of the natural systems in the Yucca Mountain region.

Water table changes due to earthquakes of moderate to large magnitudes (M=6-7) do not show a consistent pattern of response. In some cases the water table dropped while stream discharge increased close to the epicenters, as in the case of the Loma Prieta and Anchorage events. However, in the case of Borah Peak, increase in water levels were observed in the epicentral area, including unconfined jets of ground water fountaining to a height of approximately 5 m within 2 km of the fault trace on the downdropped block, and an increase in hydraulic head in the Clayton Silver Mine 50 km from the epicenter. The causes of the hydrologic changes may not be the same for the three earthquakes, because the type of faulting differed in each case, indicating different stress conditions and crustal changes. However, the magnitude of the water table changes for these earthquakes is consistent with modeling results and, therefore, provides further evidence that earthquake strain release mechanisms (or “seismic

pumping”) are inadequate to pose a threat to the unsaturated zone location of the proposed repository. It should be noted that estimates of the probability of occurrence of moderate earthquakes (M=5-6.5) within 0.25 of the fault length from the proposed repository site is less than about 15 percent. If larger areas are considered, the probabilities increase.

Various earthquake models have been used to analyze the response of the water table to stress changes in the crust resulting from earthquakes. It appears that water table rises of more than a few tens of meters are unlikely.

An analysis of the water table changes based on variations in relative compressibilities of the rock aquifer and its mineral constituents yielded a relationship suggesting that detailed knowledge of the elastic properties of the carbonate aquifer at depth is essential to prediction of the earthquake/water table interaction.

The panel recommends that studies be undertaken to determine the properties of the Paleozoic carbonate aquifer at depth, its extent in the Yucca Mountain area, its elastic and hydrologic properties, and its interconnection with the tuff aquifer, because the modeling was based on assumptions that must be verified if the results are to be credible. This information is necessary to increase confidence in predicting the future behavior of the hydrologic system in the event of a large nearby earthquake. In addition, knowledge of the Paleozoic carbonate properties is essential for the general characterization of the flow regime needed to assess effects on increased recharge as described earlier.

Few data are available to constrain the complex hydrologic system acting in the vicinity of the proposed repository in the unsaturated zone. There are a number of specific problems that must be addressed through a comprehensive program of drilling, scientific testing and logging, and core and fluid analysis.

The panel considers an understanding of the local hydrologic system and particularly the nature and source of the steep E-W

trending hydraulic gradient located approximately 1.5 km north of the proposed repository site foremost among the problems that must be addressed early in the site characterization process. Existing hydrologic models of ground-water flow and the hydraulic gradient are somewhat simplistic because of lack of reliable information. The models consider horizontal flow through a layer of uniform thickness and characterize the modern head distribution using blocks with lateral contrasts in transmissivity. These models are then used to predict the impact of increased precipitation related to future climatic changes. However, simply characterizing the lateral blocks does not explain the source of the required enormous (three orders of magnitude) lateral transmissivity contrasts, nor do the models consider the possibility of vertical flow or any interaction between the shallow Tertiary aquifer and the deeper Paleozoic aquifer that is thought to carry the bulk of the flow in this region. Not until the source of the gradient is known can the potential hazard the repository may face due to future climate changes and/or tectonic events be evaluated with a high level of confidence. Specific predictions regarding hydraulic head, head gradients, pore pressure, permeability, thermal gradients and in-situ stress need to be made to distinguish among competing models for the source of this gradient. The panel recommends that data relevant to these parameters be measured and collected in-situ in boreholes.

A focused effort on understanding the source of the steep hydraulic gradient is not enough, however. More data are needed on the regional ground-water flow system. Improved estimates of the hydraulic parameters of the volcanic and carbonate sections, as well as better definition of heads and chemical and isotopic parameters, are essential for improving the calibration of flow models that have been used to evaluate the effects of increased precipitation. Vertical head gradients must be measured so that the three-dimensional nature of the flow field can be assessed. Careful monitoring of fluctuations in head (such as those related to tidal cycles) are also necessary to better define the relationship between rock strain and water table excursions.

The panel regards the general approach to acquiring the data needed for characterization of the Yucca Mountain regional flow system as given in the Study Plan 8.3.1.2.1.3 to be sound. Continued review of available data, coupled with the judicious use of preliminary modeling results, provides a useful framework for guiding

and prioritizing future data collection. However, the panel cautions that the sole justification for data collection cannot be the reduction in uncertainties in existing models of the system. Adequate site characterization for Yucca Mountain will demand an understanding of vertical, as well as lateral, fluxes of ground water and so will require new modeling delineating the flow system in three dimensions, considering the carbonate aquifer, the volcanic aquifers and the unsaturated zone.

To address the hydrologic information needs, the panel recommends new and additional drill hole data. Planning the depth of such drill holes must be done with the above objectives of testing the flow systems in mind and not simply with the goal of better defining the water table.

Direct measurements of hydraulic head, head gradients, hydraulic parameters, and chemical and isotopic compositions of ground waters of both the Tertiary volcanic and the Paleozoic carbonate aquifers are essential. They require a series of thoughtfully placed deep (about 2000 m or greater) wells extending well into the carbonate aquifer. These wells should be located both in the vicinity of the hydraulic gradient and elsewhere for regional characterization. Current plans for “deep” holes described in Study Plan 8.3.1.2.1.3 are, in the panel's view, inadequate: in the crucial area of the unexplained high hydraulic gradient just north of Yucca Mountain only relatively shallow water table wells are planned.

During the course of the panel's study, it became increasingly clear that there was a significant lack of communication among project scientists in different disciplines, especially between those of the hydrologic and solid earth sciences, and among the different scientific organizations involved in the study, such as governmental agencies and national laboratories. Moreover, even among the geologists and geophysicists there seemed to be little integration of their individual spheres of knowledge and data. Because this important site characterization program is large and complex, strong scientific leadership must be provided to the participants and adequate attention must be paid to the continuing coordination and syntheses of scientific results. It is to Szymanski's credit that he brought to everyone's attention the fact that the various scientific disciplines cannot

work in isolation, that the information from one study may be essential to another, and that an integrated approach is the only way in which a true understanding of the complex interactive systems will emerge.

With the foregoing considerations in mind, the panel strongly recommends that DOE appoint a scientist as site characterization project coordinator. Such a person should not be currently associated with any of the participating organizations. That scientist should have a reputation for independence and excellence, as well as the experience in managing and integrating interdisciplinary programs. Such a scientific leader would lend further credibility to these investigations and their results. No large scale multidisciplinary study of this type known to the panel has been undertaken without a strong scientific leader guiding the coordination and integration of the ongoing efforts of the various parts of the project. It is the panel's opinion that had there been such a leader at the inception of this program, the controversy that brought this panel into existence would most likely not have developed, as the various working hypotheses would have been considered and addressed in an earnest and well-coordinated approach early in the program. Moreover, an integrated program guided by a strong scientific coordinator would probably have identified the steep hydrologic gradient early on as a major project-wide concern and would have approached it from a multidisciplinary point of view.

In the panel's view, the anticipated higher-order systems integration efforts would be more effective if the complex solid earth and hydrological sciences studies for site characterization were coordinated and integrated first.

This recommendation should not be misunderstood as being critical of the dedicated scientists of DOE, the United States Geological Survey, the state of Nevada, the relevant national laboratories, and the private sector, whose integrity, professionalism, qualifications and knowledge are respected and admired by the panel. The concern is with enhancing the excellent individual efforts by integrating the results of the multidisciplinary studies so that all segments of the program are aware of the availability of needed information in a timely manner.

The panel recognizes the regulatory requirements that result in the need for carefully detailed study plans that can provide a public record of the methods, data, and results of site investigations. However, plans requiring adherence to a minutely scheduled sequence of observations and a rigid constraint as to analytical methods risk the loss of the use of one of scientists' most valuable tools, their intuition. Moreover, common in scientific investigations is the element of surprise, the unanticipated findings, that may be critical in developing new insights or understanding. The detailed study plans apparently leave little room for possible changes in direction of a study. Such an inflexible approach inhibits scientific progress in achieving the objectives of the studies. The panel, therefore, wishes to register a plea for greater flexibility in allowing the scientists room to exercise their disciplines as they have been trained and as they know their expertise will be most effective. Within the framework of a well-coordinated program guided by a strong scientific leader, the panel believes such an approach can provide quality assurance and other regulatory needs while allowing scientific latitude to flourish. It can only result in greater enhancement of the scientific achievements of the program.