Water resources in the twenty-first century; a study of the implications of climate uncertainty

By Marshall E. Moss and Harry F. Lins

FOREWORD

The uncertainty associated with global-climate change and attendant implications for water and land resources represents one of the major scientific challenges of the 20th century. The geophysical and geochemical complexities inherent to the study of climate change are enormous. Disciplines from within the atmospheric, hydrospheric, terrestrial, and biological sciences are compelled now, as never before, to cooperate in studying the climate system, especially the intricate interactions of the processes comprising it.

For more than a decade, the U.S. Geological Survey has been an active participant in Federal programs designed to enhance scientific understanding of climate variability and change as well as the social, economic, and political implications of such change. Recent revisions to the U.S. National Climate Program Plan call for a significant increase in the role and responsibilities of the Geological Survey. To meet the challenge of enhanced participation in the study of climate, the Survey is currently engaged in the development of new programs that build upon and expand our previous efforts.

One of the most fundamental aspects of this expansion involves the development of new insights into the potential water-resources implications of climate change. A better understanding of the global-hydrologic cycle is frequently cited as a critical climate research need. "Water Resources in the Twenty-First Century-A Study of the Implications of Climate Uncertainty'' presents the elements of the Geological Survey's program for studying climate and hydrology. It incorporates the Survey's strong and traditional approach that couples research, data collection, and interpretive studies. We believe that it demonstrates the strength of the U.S. Geological Survey's commitment to the national and international attempts to understand the nature and effects of climate change.

Dallas L. Peck Director Foreword III Abstract 1 Introduction 1 Perspectives on climate, hydrology, and the U.S. Geological Survey 2 Climate variability, change, and uncertainty 2 The climate-hydrology continuum 4 The water-related mission of the U.S. Geological Survey 6 Elements of a program on the hydrologic implications of climate uncertainty 7 Research 7

Water-resources interpretive studies 19 Strategy for achieving the program goal 21 References 24

1. Graphs showing annual mean concentrations of atmospheric carbon dioxide measured at the Mauna Loa Observatory and annual changes in mean concentrations 3
2. Graph showing relation of annual runoff to precipitation and temperature 4
3. Map showing regions that exhibit a consistent precipitation response signal to El Nino-Southern Oscillation episodes 5
4. Graphs showing population growth in California; land under irrigation; water storage capacity of reservoirs in the Sacramento and San Joaquin river basin; and historic and projected export of water from the delta by the Federal Central Valley and State water projects 10
5. Maps showing mean sea-level pressure (in millibars minus 1,000) and monthly streamflow patterns for winters of 1976-77 and 1982-83 11
6. Characterization of how the spatial variability of a surface condition such as soil moisture, which varies over a scale of meters, is generalized to a single value through parameterization 13
7. Map showing geographic variation of the climate factor for the 100-yearrecurrence-interval flood 14
8. Map showing locations of U.S. Geological Survey benchmark stations 16
9. Histogram depicting the percentage of continuously operated stream gages in 1987, by drainage area 17
10. Sketch depicting range of scales addressed by the First International Satellite Land Surface Climatology Project field experiment 18

Process-oriented research 7 Model-oriented research 12 Process models 12 Stochastic models 14

Land-based data 15 Interface with remote sensing 17

v

11. Map showing physiography of the Delaware River basin 20
12. Diagram showing primary components of the Delaware River basin study 21
13. Diagrams showing resource balance among the three components of the climate-uncertainty program 22
14. Themes and issues in process-oriented research 7
15. Components of model-oriented research 12
16. Research status and priorities 23

INTRODUCTION

activities in light of the potential for climate change in a decadal timeframe.

This document is the synthesis of ideas derived from discussions with many of the authors' colleagues both from within and from outside the U.S. Geological Survey. The authors particularly would like to acknowledge the contributions of Mark Ayers, Julio Betancourt, Fred Nichols, Dave Peterson, and Tom Winter.

PERSPECTIVES ON CLIMATE, HYDROLOGY, AND THE U.S. GEOLOGICAL SURVEY

The concept of climate is an evolving one, and today there are usages within the scientific community for this term that are at variance with each other. Therefore, a working definition for the purposes of this document is provided here:

Climate-the unconditioned, statistical (probabilistic)

description of the sequence of measures of a suite of properties that describe the atmospheric components of the environment of a locality or region.

In other words, climate comprises the likelihoods, independent of weather-forecasting skill, of such properties as sunshine or percentage of cloud cover, amount and intensity of precipitation, or the magnitude and direction of wind. Weather is a realization or happening at any particular instant from within the realm of possibilities described by the appropriate climate descriptions for that instant.

Both climate and weather impact the study of the Earth's water resources, which is known as hydrology. Hydrology, through its linkage with climate, entails the statistical description of water-related environmental conditions; but it also, through its relation with weather, involves an accounting of the individual realizations in time of such conditions. Thus, conceptually, the term ''hydrology'' refers to a broader field of study than either climatology or meteorology. In this regard, hydrology is more analogous to the term ''atmospheric sciences,'' which encompasses both climatology and meteorology. With these distinctions drawn, it is possible to define a role for the Geological Survey in the study of the hydrologic implications of climate uncertainty.

Climate Variability, Change, and Uncertainty

The classical defmition of climate was a quasi-static one; climate was described by the normal or most common weather conditions. It was recognized that climate could change on a time scale of centuries to millennia, but the averages of limited historical data were the best key to what might happen weatherwise in the future. Such an approach suffered from several shortcomings. First, the averages or norms of weather properties do not comprise a sufficient set of statistics to describe the atmospheric environment. The variability and particularly the extremes of weather are often more useful for climate applications than are norms. Second, examination of longer historical records demonstrated that periods of deviation of weather from its norms could last significantly longer than could be explained on a statistical basis. The Little Ice Age is such an example (Eddy, 1976). Third, for many areas of the world, the periods of data collection are too short and the numbers of meteorological stations are too few to accurately describe the local climate, even on a quasi-static basis. Thus, study of any climate-related phenomenon such as water resources must contend with significant uncertainty even before the consideration of potential anthropogenic change of the Earth's atmosphere.

Climate varies periodically on at least three temporal scales: (1) daily, (2) annually, and (3) over geological ages. Over the daily cycle, changes in insolation and temperature are as distinct as night and day; over the annual cycle, clouds, precipitation, and winds may vary seasonally as well; and at the geologic time scale, ice sheets wax and wane causing drastic swings in all climate parameters. The changes associated with each of these periods can be attributed primarily to the distribution of incident solar radiation across the Earth's surface. However, another factor that causes variation in the Earth's climate is the makeup of its atmosphere, which controls the amount of energy that passes from the Earth back into space. Radiative gases in the atmosphere, like carbon dioxide, methane, ozone, and water itself, are called greenhouse gases. Analogous to the glass panes in the gardener's greenhouse, these gases permit the passage of the Sun's short-wave energy in to the Earth but deter the longer wave energy reradiated by the Earth from dissipating into space. Thus, the greenhouse gases temper the Earth's climate by retaining energy within the atmosphere.

Since the Industrial Revolution, the use of energy has increased immensely. For the last century much of that energy has come from the consumption of fossil fuels like coal, petroleum, and natural gas. A byproduct of the consumption of fossil fuels is carbon dioxide, which is released to the atmosphere. The concentration of carbon dioxide in the atmosphere has been measured at the top of Mauna Loa in Hawaii since 1958; figure 1A illustrates the significant increase in this gas, which frequently has been cited as a likely cause of climatic change. Increasing the carbon dioxide in the atmosphere is analogous to increasing the thickness of the glass panes in a greenhouse-each results in a more efficient deterent to the loss of long-wave energy. Retention of more energy within the atmosphere eventually will result in a global increase in temperature. Perhaps even more disturbing than the increase in atmospheric carbon dioxide is the sporadic but increasing rate at which it is increasing, as is shown in figure 1B.

The climatic impacts of increasing concentrations of carbon dioxide will be compounded by increases in other greenhouse gases in the atmosphere (Clark, 1982). In 1985, a group of eminent scientists convened in Villach, Austria, to discuss this problem. They reached a consensus that the global-mean temperature would increase by 1.5 to 4.5° Celsius during the first half of the 21st century (World Climate Programme, 1986). However, because of the dynamic nature of the Earth's oceans and atmosphere, such a change will not occur uniformly over the surface of the 350.----------------------------------------------------,

c

z

B.

1970 1960 1965 T

s ~. 0 ~

m

i 0

Earth. Some areas will warm in excess of the global mean, others less, and probably some will cool.

An important aspect of the greenhouse effect is the potential for positive or negative feedbacks among the phenomena that cause either exacerbation or modulation of the initial effect. For example, increased energy within the atmosphere will cause added evaporation of water from the Earth's surface. If cloud cover is not increased significantly 1980.

1985 by the added atmospheric water, the greenhouse effect will compound because of the radiative properties of the additional water vapor. On the other hand, if cloud cover increases significantly as a result of the increased evaporation, the clouds will reflect part of the incoming solar radiation and thus will modulate the greenhouse effect.

Concomitant with changes in temperature, other climatic factors also can be expected to vary as a result of the modified atmosphere. Thus, after several decades in which climatic uncertainty seemed to be decreasing as knowledge and information increased, it seems that climatic uncertainty suddenly has increased dramatically.

The Climate-Hydrology Continuum

Unlike climate, the boundaries and definition of hydrology have been rather broadly accepted for several decades. A typical definition is:

Hydrology-the science that deals with the waters of the Earth, their occurrence, circulation, and distribution, their chemical and physical properties, and their reactions with their environment, including their relation to living things (Federal Council for Science and Technology, 1962).

This level of acceptance was not always so. Meinzer (1942) traces the evolution of the meanings of hydrology from the beginning of the 20th century, when the U.S. Geological Survey used "hydrology" to denote only the study of waters beneath the surface of the Earth. At that time,

enw

z0

uw

z

z 25 the study of surface water was referred to as ''hydrography'' and "hydrometry."

Although many hydrologic studies are conducted from a perspective that is more akin to weather than climate, there is a field of endeavor within hydrology that relates directly to climate, and that is the study of the normal behavior of aspects of the land phase of the hydrologic cycle. Some of the earliest reported hydrologic studies were of this type. Meinzer (1942, p. 14-15) attributes the first studies relating hydrology to climate to Pierre Perrault and Edme Marriotte, who were French contemporaries of the 17th century. Each labored independently to demonstrate that rainfall on the drainage basin of the Seine River was sufficient to be the ultimate source of runoff in the river. Their studies were attempts to add evidence to the theory of the hydrologic cycle that had been espoused a century and a half earlier by Leonardo da Vinci.

A more recent, but also classic, study is that of Langbein and others (1949), which graphically relates the normal runoff in rivers of the United States to the climate measures of precipitation and temperature (fig. 2). Some of the more enduring aspects of the Langbein study were its illustration of (1) how runoff, for a given annual precipitation, decreases as temperature increases; (2) how runoff, for a given temperature, increases with precipitation; and (3) how the numerical difference between precipitation and runoff, for a given temperature, increases with precipitation and

D ory

addressed. For example, Barnett and others (1988) have demonstrated the effect that added snow cover and subsequent increased soil moisture in Eurasia have on deterring the subsequent monsoon in South Eastern Asia. This same study shows strong correlations between Eurasian snow cover and the atmospheric conditions over all of North America.

A common characteristic of each of the above examples is that significant levels of uncertainty exist with respect to the exact forms of the interdependence of climate and hydrology and also with respect to the parameters that quantify the interdependence at any given time and location. Therefore, current uncertainties pertaining to the availability and quality of water resources in the 21st century must be viewed as a manifestation of the climatic uncertainty described in the previous section.

Because of the potential for major impacts on social, economic, and environmental conditions in the United States, and because of the uncertainties concerning those impacts, major scientific studies are being planned and some already have begun. These studies are being carried out both at the national and global scales.

The Water-Related Mission of the

The water-related activities of the U.S. Geological Survey are executed by its Water Resources Division. The stated mission of Water Resources Division is to provide the hydrologic information needed for managing the Nation's water resources (Chase and others, 1983). This information is developed by conducting hydrologic research, by collecting basic hydrologic and water-use data, and by performing interpretive studies of the Nation's water resources at various scales from local to national. These three program components are interfaced with the Nation's information users through the release of publications and hydrologic data and through data-coordination, cataloging, and retrieval services.

For almost 100 years, the Geological Survey has been providing hydrologic information to the Nation that is essential for coping with water-supply uncertainty and for protecting the citizenry from the devastations of floods and droughts. For a shorter period of time, the Nation also has depended on the Geological Survey's information to set and implement policies for dealing with the degradation of the quality of its water resources. Thus, the Geological Survey is well positioned by tradition, as well as by the scientific momentum of its current activities, to play a key role in providing information to address the recently recognized increment of 'hydrologic uncertainty associated with the potential for global climate change. In fact, it will be impossible for the Geological Survey to continue to meet its mission responsibilities in the 21st century if the potential impacts of climate change are not taken properly into account.

Although the Geological Survey has not been one of the major Federal participants in the climate-change activities of the last decade, it has not been isolated from them either. In 1976, the Geological Survey convened a workshop to stimulate discussion of the effects of climate variation on the land and water resources of the Nation (Smith, 1978a,b). As a result of the workshop, a Geological Survey climate plan was developed (Howard and Smith, 1978) that to date has been implemented only to a modest degree. The plan is consistent with the responsibilities of the Geological Survey as specified in the first U.S. National Climate Program Plan, a document called for in the National Climate Program Act (Public Law 95-367) and prepared by the National Climate Program Office of the National Oceanic and Atmospheric Administration in conjunction with other Federal agencies. However, as more knowledge and information have accumulated and as Federal agency activities have shifted, the National Climate Program Plan has been revised (National Climate Program Office, 1988), and the responsibilities of the Geological Survey have increased to that of a lead agency for the study of the interactions between climate and hydrology.

For the Geological Survey to become a focal point for hydrology in the climate-change arena, the significant, but diffuse, climate-related activities that currently are being conducted within the Water Resources Division need to be brought into perspective; their roles relevant to a common goal need to be evaluated, and weaknesses and critical gaps need to be identified. By carrying out these steps and rectifying weaknesses and gaps as resources can be made available, the Geological Survey indeed will have a program Much is currently known about many of the hydrologic processes, particularly as they function in the laboratory or on small plots of land. The knowledge developed in the laboratory and on small plots inherently is limited to the lower end of both the spatial and temporal scales of interest in hydrology and frequently is isolated from many of the factors that may be relevant in the natural environments in which the processes function.

Many of the questions concerning the hydrologic implications of climate uncertainty cannot be addressed adequately by improving process understanding at modest scales. Therefore, the majority of the elements contained in the fundamental research component of the program will be oriented toward development of understanding the aggregate of processes at work over reasonably large and div~rse areas of land-the scale problem (Dooge, 1982).

An example of a process that is reasonably well understood at the small scale is the concept of infiltration, which describes the passage of precipitated, ponded, or flowing water from the surface of the Earth into its interior. Much is known about infiltration from the laboratory or at small field-plot scales (Dooge, 1982). Rates of infiltration depend on the nature and water content of the soil or rock that receives the infiltrated water, the chemical, physical, and biological character of the source water, and in the case of precipitation, the rate at which water is supplied to the surface. At many larger scales of interest-that is, agricultural fields or larger areas-most of the factors that control infiltration will vary significantly. Because of the complexity of the relations among the controlling factors, the knowledge of infiltration derived at small scales is difficult to apply at large scales. This quandary is compounded by the current inability to collect and interpret sufficient quantities of precise and accurate data to describe adequately the spatial variability of the controlling factors. Thus, gaining more insight on infiltration over medium to large areas is an essential. element of the program.

Another hydrologic process whose understanding is impacted by scale is the reverse of infiltration, that is evapotranspiration, which takes water out of the Earth and the biosphere and returns it to the atmosphere. In most waterbalance studies, evapotranspiration is computed as the residual after all other relevant flows and reservoirs have been taken into account. There are two primary reasons that it is treated so: (1) evapotranspiration is a highly nonlinear function of the energy available to transform liquid water to water vapor at the surface of the Earth or on a plant surface, and (2) the energy balance on the land surface or on a plant is highly variable both in space and in time. As climate changes, the statistical characteristics of the energy balance will change, as will the availability of water at land surface. In response to significant shifts in water availability, the local plant community also can be expected to change (Eagleson and Segaria, 1985). Therefore, evapotranspiration probably will be very sensitive to climate change. Of the major water balance components, evapotranspiration is the least documented. Unlike precipitation and runoff, evapotranspiration has not been measured systematically across the Nation. Its character at various spatial and temporal scales must be determined.

In many regions of the United States, ground water is an abundant resource that has accumulated over long periods of time. It is a resource that can be drawn upon at rates that vary as functions of the volume of ground water in storage and the rates at which it is replenished. The rate at which ground water is replenished is known as aquifer recharge and, to a first approximation, it is equal to the difference between infiltration and evapotranspiration. Because these processes are controlled by many interrelated factors sensitive to scale, it follows that aquifer recharge must be as well. For ground-water development to proceed in a wise and measured fashion under climate uncertainty, the process of aquifer recharge must be better understood.

As a first approximation, the difference between precipitation and infiltration is surface runoff. Runoff is the primary source of water for most floods producing serious property damage and loss of life in the United States. As a result of its direct linkage to infiltration, the processes controlling the generation of surface runoff are poorly understood at spatial scales larger than field plots. Thus, even if climate change were not a significant issue, research on the process of runoff from catchments and basins is an area that warrants renewed support. The added uncertainty attendant to climate change only serves to augment an existing need for research on this phenomenon.

Another category of the hydrologic continuum requiring process-oriented research is extreme hydrologic conditions. The extremes of the hydrologic continuum, floods and droughts, have been studied largely in statistical contexts. Much of water-resource planning is based on the normal or average conditions over some standard period of instrumented observations. Such an approach ignores the question of how representative the available data may be of the true expected conditions. This question is particularly relevant for decisions that entail planning horizons that extend 20 to 100 years into the future.

Historically, water-resource systems have been designed on the assumption that future climate conditions, hence floods and droughts, would be similar to those observed in the systematic record. This assumption, known as stationarity, has long been controversial in hydrology. If significant climate change occurs in the coming decades, it will invalidate the stationarity assumption and force a change in the procedures for estimating the occurrence of hydrologic extremes; the 100-year flood of today will not be the same as that of the mid-21st century. Importantly, the change will not be attributable simply to a longer record of floods, as is frequently the case today but, rather, to a shift in the

z

u.

za:

a:

a:

tua: ow _.J>

~ ffi 4

oa..

Data Collection

However, items 2 and 3 will be ameliorated as part of the data-collection element of the program.

Climate data for the benchmark drainage basins are sporadic in time and in space; this facet of the network limits its utility for studying climate-hydrology interactions. The first priority for increased data collection under this program will be the installation of permanent climate stations on each benchmark drainage basin. Because of the climate variability over areas within the range of the benchmark drainage basins (approx 5-2,500 km), several climate stations may be desirable in many of the basins. However, the network-design methodology for specifying the ideal number and locations of climate stations for each basin currently does not exist. Therefore, a single climate station will be installed on each, and deployment of other stations will await the methodology that is to be developed in the research element of the program.

Even though new sites can be added to the benchmark network as funds are made. available, any new site will never be as powerful in tracking the impacts of potential climate changes as would the original sites. The information contained in the charter sites never can be recovered fully for the new sites. Nevertheless, if significant climate change is not too immediate, time remains for the establishment of significant data bases at new locations. Criteria for the new sites, both the number of sites and the characteristics of the individual sites, also will be developed within the research element of the program.

A histogram of the percentage of continuous stream gages in the United States by area of their drainage basin, as determined from the files of the National Water Data Exchange (Williams and Knecht, 1981), appears in figure

9. The drainage basins of the benchmark network fall on the

smaller end of the spectrum of gaged watersheds. Because of the incongruity between the spatial resolution of general circulation models and that of the Benchmark Network, the benchmark sites will contribute only a minor amount of information for the calibration and testing of general circulation models. Another set of stations at a more appropriate areal scale will be needed for this purpose. Currently, the Geological Survey is making a detailed search of the existing data base to identify stations that either are active now or could be reactivated to provide the requisite data for model testing and for evaluation of the hydrologic information content of the outputs of the GCM's. It is anticipated that both reactivation of some discontinued stations and the establishment of some new ones will be required to develop an adequate data base to achieve these ends.

One area in which hydrologic data generally are not available for either statistical or modeling purposes is that of evapotranspiration. The reason for this is that a costeffective methodology does not exist to generate the data of the desired temporal and spatial scales. Early in the research element of the program, high priority will be given to the development of the missing methodology. Once the methodology has been demonstrated to be successful, z

Canopy, leaf physiology 1 centimeter - 10 meters

Another example is the development of an in-situ instrument package for measuring the dielectric properties of snow and ice. The Snow and Ice Dielectrics System, developed collaboratively by the Environmental Research Institute of Michigan and the U.S. Geological Survey, is specifically designed to facilitate the spatial extrapolation of snow water-equivalent estimates derived from airborne passive microwave measurements of snow. Activities such as these undoubtedly will expand as large-scale integrated studies of the Earth focus to a greater degree on increasing the detail and versatility of remotely sensed data.

The success of any program designed to study the hydrologic implications of climate uncertainty must be measured in terms of information utility. It is essential that the information generated by a program be carefully matched to the level and scale at which actual policies and decisions affecting the Nation's ability to cope with the uncertainty are made. In terms of water resources, an appropriate areal scale for information synthesis is the major river basin or regional aquifer. Accordingly, the U.S. Geological Survey has initiated a pilot study of the Delaware River basin (Ayers and Leavesley, 1988) (1) to test its capabilities to develop information on the implications of climate uncertainty regarding a number of interrelated aspects of the basin's hydrology; (2) to determine the utility of such information for basinwide water-resources planning in the 21st century; (3) to assess methodological performance and needs for enhancing the utility of such studies; and (4) to evaluate future data requirements for tracking potential impacts of climate change.

Because the Delaware River basin does not encompass the full range of hydrologic phenomena susceptible to climate change, this study is envisioned as the first of several geographically distinct pilot assessments that logically should precede a comprehensive attack on the problem of evaluating the Nation's water resources in the 21st century. The following summary of the Delaware River Basin study provides an example of the scientific complexity associated with such comprehensive, but meaningful, interpretive analyses.

The Delaware River basin encompasses 12,765 miand crosses four distinct physiographic provinces (fig. 11). Its runoff processes are quite diverse, and man's influence on the movement and storage of water adds considerable complexity to the basin system. The Delaware River serves as a major source of water for an estimated 20 million people both in and outside the basin (Delaware River Basin Commission, 1986).

The availability of water to users is based on many complex systems of reservoirs for storage and pipes, tunnels, and canals for diversion and delivery. The two largest diversions out of the basin are through the New York City aquaduct system to New York City and the Delaware and Raritan Canal to northeastern New Jersey. The freshwater portion of the tidal river below Trenton, the Delaware estuary, serves as a source of ground-water recharge for aquifer systems supplying water to southern New Jersey as well as a water-supply source for the city of Philadelphia and many industries by direct diversion. It is critical for these supplies that the Delaware Estuary along these reaches remain potable, even during periods of prolonged low flow (droughts).

Salinity problems in the estuary have historically been associated with New York City diversion rights. The U.S. Supreme Court (1931; 1954) required the city system to maintain certain instream flows at Montague, N. J., to provide both an adequate water supply and salinity controls in the estuary, an issue of considerable concern to downstream interests. Further, ''Good Faith'' agreements have included instream flow requirements at Trenton, N. J. During the drought of record in 1961-65, and twice since 1980, emergency water-use restrictions have been placed on basin water users to meet basic water-supply demands and instream-flow requirements.

Such experiences illustrate the susceptibility of the basin's water-supply to climate variability and emphasize how longer term climate uncertainty poses potentially serious problems for the Delaware River basin. Moreover, they provide insight on how poorly understood the potential effects are of climate change on the water supplies of the basin, including the ability to maintain instream-flow requirements.

The Delaware River basin study is designed to investigate the hydrologic response of the basin (fig. 12), under the existing management infrastructure, to possible climatic conditions in the mid to late 21st century. Specific objectives include defining the spatial and temporal variability of current and projected climate conditions on four important aspects of water availability within the basin. These are

York City and other basin reservoir systems, (2) the ability to maintain instream flow requirements, (3) the upstream movement of saline water in the Delaware estuary associated with changes in sea level and freshwater inflows, and (4) the resulting potential intrusion of saline water into aquifers adjacent to the estuary.

Although the Delaware River basin is a unique hydrologic system within the United States, the interpretive assessment of this basin is intended to serve as a prototype for similar studies across the Nation. Clearly, other regions will have a different mix of physical system, water-supply, legal, and economic problems attendant with a climate change, but the general approach to evaluating those problems should be applicable. That approach, involving the coupling of advances in process research and data collection with existing hydrologic capabilities, is a robust yet flexible design. It should provide a sound framework for analysis regardless of the hydrologic region or type of climate change.

STRATEGY FOR ACHIEVING THE PROGRAM GOAL

The primary facet of a strategy for dealing with the hydrologic implications of cliinate uncertainty is the evolution of the resource balance among the three components of the program: (1) data collection, (2) research, and (3) interpretive studies. The strategy must be an evolutionary one because of the high degree of near-term uncertainty in the climate variables that, along with gravity, drive the hydrologic cycle. Information that reduces this uncertainty must be monitored continually by the program administrators, and, as the new information dictates, the program balance should be modified to address more effectively the program goal. In the early phases of the program, the necessity of feedback of information to develop the short-term plan dominates the planning process to the extent that long-term plans can be viewed only in a statistical or probabilistic context.

The planning process is based on the hypothesis that a mature program will consist of a modest balance between

research and data collection that will continue to generate new knowledge and information but that will have as its major focus the synthesis and interpretation of the existing knowledge and information across broad ranges of both hydroclimatic settings and water-resources problems and opportunities. This hypothesis dictates that a mature program tends to evolve toward the lower-right quarter of the diagram given in figure 13A. As is shown in figure 13B, the water-resources activities of the U.S. Geological Survey historically have gravitated from a data-collection focus to a balance in the 1980's that has in excess of 50 percent of its funds expended for interpretive studies. This traditional evolution can be attributed to the prevailing assumption that natural hydrologic processes are in statistical equilibrium at the time scales of a century or so-stationarity. Under such an assumption, it is efficient to expend relatively large sums collecting and analyzing data that will serve as the basis for projections of future water resources. As the data bases expand, the marginal utility of the next increment of data decreases (Moss, 1970), while the hydrologists' abilities to

B.

perform interpretive studies using the historical data bases increase. Thus, under the assumption of stationarity, there is a natural tendency to evolve from data collection to interpretation. During this evolution, research has been viewed primarily as a modest investment to address future contingencies.

The potential for climate change provides an impetus to the proposed program to start from a different sector than have the traditional water-resources programs. As was described previously, the lack of sufficient understanding of the major interactions of the hydrosphere with the atmosphere contributes significantly to the level of uncertainty regarding the future characteristics of both. Development of this understanding is requisite for useful projections of future hydrologic conditions as well as for understanding the apparent changes that may be detected as a result of existing data-collection activities. Thus, research will be the dominant component of the climate uncertainty program during its early stages, as shown in figure 13B. As research contributes additional understanding, both interpretive studies and data collection will have a strong basis for increased levels of effort.

As shown in table 3, the research needs for this program are partitioned into three categories. The first category, which is called critical, consists of endeavors that currently are underrepresented to a debilitating degree if the program goal is to be addressed in a timely manner. It is noticable that, with one exception, each of these endeavors deals with links between the atmospheric and hydrologic sciences. Because each will be carried out at the disciplinary frontier of hydrology, each increased activity should be planned to take advantage of and interact with complementary activities of other Federal agencies and interdisciplinary, international programs.

The exception to the hydrology-atmosphere interface in this research category is network design. This endeavor is critical because the data-collection component cannot be fully effective until better knowledge and methodologies are available to guide investments in data.

The second category of research endeavors includes facets of existing activities that only may need some redirection and (or) added support to assume their proper role in the program. With the exception of perhaps paleohydrology and biogeochemical cycles, entries in this category are traditional strengths of U.S. Geological Survey hydrologic research.

The final category contains lakes and estuaries, which have been significant components of U.S. Geological Survey research in the past but that will require added emphasis in the decade of the 1990's and beyond. Studies currently are being conducted within the Survey to define more specifically the research needs and opportunities in each of these areas.

As was discussed above, the data-collection component of the program initially will not be of the same magnitude as the research component. However, there are two data-collection activities that merit particular concern. The first of these is augmentation of the benchmark network to provide baseline information for tracking hydrologic change. Modest initial expansion likely would be followed by additional increases as knowledge, methodology, and technology are developed.

The second critical endeavor for data collection is a network for estimating areal evapotranspiration. Although such a network is an immediate need, methodology and technology limit our abilities to deploy the requisite costeffective measurement systems. Interagency collaboration, such as is currently being carried out in the Konza Prairie of Kansas (Brutsaert and others, 1988), will be required both to redress the technological shortcomings as well as to implement the ultimate solution. The Geological Survey has a key role to play in both aspects.

The program component with the greatest potential for growth is interpretive studies. As information and understanding from the other two components of the program accumulate, the demand for interpretive studies can begin to be met. Indications are that the demand for this type of product already exists as a result of the publicity that climate change has received recently. However, current ability to provide substantive interpretations of the implications of climate change is extremely limited. Thus, a cautious approach to performing interpretive studies is in order. Studies like that in the Delaware River basin, described

Where this page came from

This page was imported from U.S. Geological Survey. Published by the U.S. Geological Survey and, as a work of the United States government, in the public domain.

Nobody has written it yet — it is the source material at a new address, which is why search engines are asked to skip it and why no one earns from it. It is up for grabs: take it on, and it is yours to rewrite and to earn from.