Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 7
--> 2 Water Management and Information Needs Floods, droughts, and other natural disasters have drawn special attention in both religious and secular literature from the beginning of time, and the modern world seems almost as vulnerable to the whims of nature as ever before. Despite massive investments in dams and drainage and other facilities to reduce the risks associated with these events, economic and environmental damages resulting from them continue to increase. Although the data are admittedly subject to uncertainty, especially prior to 1950, Burton et al. (1993) cite several sources that indicate that worldwide deaths from natural disasters are substantially lower now than in the first half of this century. While the number of deaths has decreased, the number of incidents with at least 100 deaths has sharply increased, possibly resulting from the increased concentration of populations in urban areas. Estimates are that economic losses also are increasing rapidly. Some of those trends may be explained simply by better reporting, but factors such as urbanization and increasing concentration of the population in coastal areas have no doubt contributed to increased risk. Of the reported incidents causing at least 100 deaths, floods are the leading cause (40 percent), and the other primary hydrologic hazard, drought, accounts for an additional 15 percent. Tropical cyclones and earthquakes account for 35 percent. Subsequent sections of this chapter include an overview of the nature and magnitude of damages caused by floods and droughts and an examination of organizational arrangements for addressing floods and droughts in the United States, with particular attention to the roles of various government management agencies. A special section is devoted to the role of the U.S. Geological Survey (USGS). The concluding section identifies key issues and their research and data implications.
OCR for page 8
--> Floods Flooding is the most widespread hydrologic hazard in the United States and throughout the world. A recent compilation of data on events leading to declarations of disasters under the 1988 Stafford Disaster Relief and Emergency Assistance Act indicates that of the 295 declarations during the period December 1988 to May 1996, one-third were due to flooding (Godschalk et al., 1997). Another 11 percent were due to a combination of tornadoes and flooding. Those numbers do not include coastal storms and hurricanes, which cause a combination of wind and water damage. Because damages from a small number of devastating earthquakes and hurricanes account for a disproportionate share of all relief provided under the Stafford Act since 1988, riverine flooding accounted for a lesser but important share of the total, namely 17 percent. Not only is flooding the most frequently occurring disaster, but it is also the most widespread. Flood disaster declarations under the Stafford Act have been made in 42 states. Because the Stafford Act broadened the range of disasters for which relief is made available, the proportion of all types of disasters attributable to flooding decreased after 1988. The Federal Interagency Floodplain Management Task Force (FIFMTF), citing data from the Federal Emergency Management Agency, Figure 2.1 Average annual flood damages over five-year periods. Source: FIFMTF (1992).
OCR for page 9
--> Flood damage to bridge and road. Note person on bridge for scale. Photo courtesy of U.S. Geological Survey. indicated that floods and hurricanes accounted for three-fourths of all presidentially declared disasters from 1965 through 1989 (FIFMTF, 1992). The FIFMTF also reported that over the 70-year period from 1916 to 1985 there were, on average, 101 flood-related deaths annually in the United States and that there was no indication that the number of deaths per capita was changing (FIFMTF, 1992). While no trend was discernible in loss of life, a definite increase in economic losses over that period was noted. Those damages, adjusted for inflation and population increase, are shown in Figure 2.1. Average annual damage in the 1970s was in the neighborhood of $4 billion per year in 1985 dollars. Although that average dropped to just over $2 billion in 1980–1985, the Interagency Floodplain Management Review Committee estimated that the Midwest floods in 1993 caused $12 billion to $16 billion in damages (IFMRC, 1994). Flood Management Strategies Strategies The FIFMTF formulated a well-developed framework for floodplain management, consisting of four complementary strategies: Modify susceptibility to flood damage and disruption through floodplain
OCR for page 10
--> regulations, development and redevelopment policies, disaster preparedness, disaster assistance, floodproofing and flood forecasting, warning systems, and emergency plans. Modify flooding by constructing dams, reservoirs, levees, floodwalls, channel alterations, high-flow diversions, land treatment, and onsite detention. Modify the impact of flooding on individuals and communities through information and education, flood insurance, tax adjustments, and postflood recovery. Restore and preserve natural and cultural floodplain resources. Implementation of these strategies takes place through a complex array of federal, state, and local government and private-sector activities. Federal agencies with prominent roles in managing flood risk are the Federal Emergency Management Agency (FEMA), the U.S. Army Corps of Engineers (USACE), the Natural Resources Conservation Service, the Federal Crop Insurance Corporation, other agencies of the U.S. Department of Agriculture, and the National Weather Service. At times, the web of institutional arrangements and specific policies becomes so complex that it appears to be fragmented, and it is difficult to envision how well the broad policy framework formulated by FIFMTF is being implemented. It is in the details of these institutional arrangements and policies where the strategy's success or failure may rest. Given the bottom line of a rising rate of economic losses related to floods, it appears that there are serious weaknesses in some aspects of these policies. Structural Measures From the mid-1930s until the decade of the 1960s, structural measures to modify flooding, such as the construction of levees and dams, dominated efforts to reduce flood damage in the larger watersheds of the United States. Among the earliest efforts was the construction of levees in the Lower Mississippi Basin, first by individual landowners and later by levee boards. Beginning in the early part of this century, greater reliance was placed on the construction of large reservoirs.1 As of 1910 there were fewer than 300 large reservoirs in the United States with a combined capacity of 14 million acre-feet (MAF). In 1988 there were over 2,700 reservoirs in that size class, and their combined normal storage was 467 MAF (Ruddy and Hitt, 1990), though it is acknowledged that flood control is only one purpose served by most of these multiple-purpose facilities. Major growth in these facilities began in the 1920s and peaked about 1965, as shown in Figure 2.2. 1 Large reservoirs are defined by USGS as those with a nonflood control capacity of at least 5,000 acre-feet or a total capacity of 25,000 acre-feet or more.
OCR for page 11
--> Figure 2.2 Number of large dams completed in successive five-year periods in the United States. Source: Ruddy and Hitt (1990). The sharp decline in construction after 1965 has been attributed to a number of factors. Water resources professionals and environmental groups began to question the benefits of large dam projects. Pasèóge of the Environmental Policy Act of 1970 gave opponents of such projects an entree to the courts while an increasingly urban U.S. population waned in its support for rural construction projects. At the same time, the costs of dams and reservoirs increased as the most attractive sites had already been developed, and debates over how the costs of building these facilities would be shared between federal and nonfederal partners put many projects on hold. Nonstructural Measures In the 1960s, behavioral research by Gilbert White and others began to show that, as the frequency of flood events was reduced by building control structures, protected lands became more attractive for urban development. When the frequent floods did occur, property damage was considerably higher, resulting in even higher average damages. After a series of flood disasters in the early 1960s, White chaired the Bureau of the Budget's Task Force on Federal Flood Control Policy whose report, A Unified National Program for Managing Flood Losses , recommended a broader perspective on flood control that embraced the management of floodplains. Pub-
OCR for page 12
--> lished as House Document 465, that report recommended, among other actions, that: flood-prone areas be delineated by an appropriate federal agency, uniform techniques be established for defining flood frequencies, flood forecasting techniques be improved, and the feasibility of a national flood insurance program be determined. The insurance program was to provide financial incentives for private insurers who would provide protection in flood-prone areas that they would not insure otherwise. It would serve two objectives: first to have property owners bear a share of the risk and respond accordingly and, second, to provide disaster relief through the private sector when damage occurred. When the feasibility study was completed by the U.S. Department of Housing and Urban Development in 1965, its recommendations were translated into policy in 1968 when Congress passed the National Flood Insurance Act (NFIA). Modifications to the law were necessary in 1969 and 1973 to overcome local government hesitancies to adopt sufficiently strict land-use regulations. The Flood Disaster Protection Act of 1973 included a provision that made adoption of those controls a prerequisite for federal financial assistance. FEMA now administers the National Flood Insurance Program and is instrumental in promoting mitigation of the effects of a wide range of hazards. Under the Disaster Relief Act of 1974 as amended and the Stafford Disaster Relief and Emergency Assistance Act of 1988, FEMA is authorized to provide financial assistance to individuals, state and local governments, certain nonprofit organizations, and Indian tribes. The Hazard Mitigation Grant Program, triggered by a presidentially declared emergency, is a federal cost-sharing program authorized under the 1988 act that builds hazard mitigation into postdisaster recovery operations. A major limitation of the National Flood Insurance Program has been its exemption from flood insurance purchase requirements of those structures built prior to enactment of the program. FEMA is also authorized to purchase flood-damaged property and to offer owners an opportunity to relocate. A number of agricultural policies also are directed at hydrologic hazards. Of special importance is the Federal Crop Insurance Program first established under the Agricultural Adjustment Act of 1938, a program that offers farmers flood insurance for at-risk production areas. The U.S. Department of Agriculture offers a variety of other financial and technical assistance programs. Some, like disaster payments under the Agricultural Consumer Protection Act of 1973 and cost sharing under the Emergency Conservation Program in the Agricultural Credit Act of 1978, are targeted specifically to disasters. Others, such as the Wetlands Reserve Program in the 1990 Farm Bill, ''swampbuster'' provisions of the Farm Bill of 1985, and the small watershed program under the Natural Resources Conservation Program, are directed at modification of flood events
OCR for page 13
--> through soil conservation practices, structural measures, and protection of strategically located wetlands. The infrastructure of dams, levees, and floodways built by the USACE is well known, but the Corps has many other flood management activities. Its Flood Emergency Operations and Disaster Assistance programs provide a wide variety of flood-fighting and rescue operations, assistance for repairing flood control works, emergency water supplies, and other services. Its Floodplain Management Services Unit provides nonemergency technical assistance, including flood hazard mapping and planning. The Economic Development Administration and the Small Business Administration are significant in the provision of financial assistance to economic development activities and businesses adversely affected by natural disasters. They have only modest roles in modifying the hazard or the risk of damage. FIFMTF conducted an assessment of progress under NFIA in 1992, noting a long list of achievements—more widespread public perception of the hazard; improved knowledge, standards, and technology; an extensive body of judicial decisions; and well-established development standards. Measuring program effectiveness remained an elusive task, however. The report pointed to the lack of "consistent, reliable data about program activities and their impacts" as a principal complication (FIFMTF, 1992, pp. 60—61). It noted that susceptibility to flooding in the United States is being reduced at individual sites and local communities through a variety of land-use controls and emergency preparedness activities. Evidence reviewed by the task force led its members to conclude, however, that overall vulnerability has either increased or remained the same because of the large amount of preexisting vulnerable development, numerous exceptions in state and local policies, or the inability of governments at all levels to respond quickly to new spurts in development activity. Following the disastrous floods of 1993 in the Mississippi-Missouri River basin, the Interagency Floodplain Management Review Committee (IFMRC) was established in January 1994 to investigate the causes and consequences of that flood, to evaluate the performance of existing floodplain and watershed management programs, and to make recommendations for appropriate changes. Among the committee's numerous findings was that initial estimates of those properties actually covered by flood insurance ranged from below 10 percent up to 20 percent of insurable buildings in identified flood hazard areas in the Midwest. For the nation as a whole, the range is 20 to 30 percent. The committee recommended taking more vigorous steps to market the program and to provide reduced postdisaster support for those who are eligible but do not purchase adequate coverage (IFMRC, 1994, pp. 131—134). Droughts Most extreme events of nature that result in large economic losses or large
OCR for page 14
--> numbers of deaths are relatively sudden events of short duration, for example floods, hurricanes, and earthquakes, but droughts may have equal or greater consequences. Events that signal the beginning of drought disasters are more gradual than more violent disasters, but droughts may persist for several years, as did the most recent one in California, which lasted from 1987 through 1992. For a variety of reasons (discussed later in this section), damages from droughts are also much more difficult to measure than those of other extreme events that occur within short periods of time. A fundamental definition of drought does not exist. There is no drought analysis method that utilizes a probabilistic method that is as well developed as flood frequency analysis. While low-flow analysis is a meaningful measure of drought, it does not provide a sufficient definition of drought. Four different working definitions of drought have been established: meteorological, hydrologic, agricultural, and economic drought. All four are based on the notion of water deficit but differ depending on the perspective of the water user. Meteorological drought can be expressed solely by the degree of dryness and the duration of the dry period (Wilhite, 1993). Normally, the precipitation deficit over a specified length of time determines the degree of dryness. Meteorological drought is very region specific because the atmospheric conditions that result in precipitation deficiencies vary greatly from region to region (Wilhite, 1993). Hydrologic drought is associated with the effects of precipitation deficits on streamflows, reservoir and lake levels, and ground water supplies. That is, the impact of this deficiency is assessed throughout a hydrologic system. Hydrologic droughts are usually out of phase or lag the occurrence of meteorological drought. Hydrologic droughts may impact local water supplies, hydroelectric power production, flood control, irrigation, commercial navigation, and recreation. Land use plays an important role in the determination of hydrologic drought. Changes in land use not only significantly alter the hydrologic characteristics of the immediate watershed but can also greatly impact downstream areas. Increased variability in streamflow owing to urbanization effects will cause an increased likelihood of hydrologic drought downstream. Another factor that influences hydrologic drought is the water supply for human and industrial consumption: periods of droughts cause an increased water need, further depleting water supplies (Wilhite, 1993; Grigg, 1996). Agricultural drought occurs when soil water is inadequate to initiate and sustain normal crop growth over a substantial period of time. Although agricultural drought focuses primarily on soil water deficits, it also incorporates precipitation shortages, differences between actual and potential evapotranspiration, and reduced ground water or reservoir levels. Economic drought relates to the supply and demand of some water-related good or service and concerns the areas of human activity affected by meteorological, hydrologic, or agricultural drought. Therefore, economic drought is strongly based on human needs (Changnon, 1989). The incidence of economic
OCR for page 15
--> drought could increase because of a change in the frequency of physical events, a change in societal vulnerability to water shortages, or both. Because of difficulties of measurement, historical information on the frequency, magnitude, and duration of droughts is far more limited than that of floods and other hazards. Much of the information is anecdotal—the Dust Bowl days of the 1930s, droughts of Southern California in 1927 to 1932 and 1987 to 1992, the drought in the Northeast in the 1960s, and the nationwide drought of 1988. Much of the economic information is also specific to particular effects on agriculture, public water supplies, hydroelectric power, and navigation. Very little information has been compiled to estimate average annual losses across all sectors of the economy. One of the more complete descriptions of drought frequency at the national scale was produced by Riebsame et al. (1991). They calculated a 90-year sequence of area-weighted annual precipitation from 1895 through 1985 (see Figure 2.3) and similar series for selected seasons. For each year of that same period of record, they also calculated the percent of land area in extreme drought as indicated by the Palmer Hydrological Drought Index (see Figure 2.4) and accumulated precipitation deficiencies. All of these measures were calculated for the contiguous United States. The same measures could also be calculated for selected regions of the country. When droughts occur, their impacts are quite diverse. As part of the National Drought Study (USACE, 1994), the Corps of Engineers surveyed all 50 Figure 2.3 Total annual precipitation, 1895 to 1989, area weighted over the contiguous United States. Source: Riebsame et al. (1991).
OCR for page 16
--> Figure 2.4 Percentage of the contiguous United States in severe or extreme drought (Palmer Hydrological Drought Index ≤3), 1895 to 1989. Source: Riebsame et al. (1991). state governors to identify the types of impacts that affected their regions. Shortages of public water supplies and crop losses were among the most frequently cited effects. Environmental damages to wildlife, fisheries populations, and other aquatic ecosystems were also frequently mentioned. Impacts on the Everglades were of special concern. In the Northwest, loss of hydropower was especially important, and in the Midwest, interruptions to navigation on the Ohio and Mississippi rivers were cited. A number of states cited effects on aquifers that were subject to saltwater intrusion. Economic losses for the 1988 drought as estimated by the Interagency Drought Policy Committee totaled $39.2 billion, over 60 percent of which was attributed to losses in farm production and increased food costs (Riebsame et al., 1991). The National Drought Study identified a number of difficulties in measuring the consequences of droughts. Agricultural losses in one region of the country can be offset by increases in other regions. Regional industrial losses can at least be partially offset by increases elsewhere. Because droughts extend over relatively long periods of time, separating their effects from others such as shifts in the general economy and changes in management practices is often difficult. In addition, techniques that focus only on losses during droughts fail to account for capital expenditures made in earlier years to mitigate the effects of droughts,
OCR for page 17
--> including the building of reservoirs for public water supply, hydroelectric power, low-flow augmentation, and navigation. Drought Management Strategies Unlike that for floods, no unified national management strategy has been developed for droughts. By contrast with specific strategies outlined by the FIFMTF, the National Study of Water Management During Drought (USACE, 1994) contains only a broad planning framework for drought management. That is not to say that droughts have been ignored. Substantial capital investments have been made by USACE, the Bureau of Reclamation, the Tennessee Valley Authority, the Natural Resources Conservation Service, and other federal agencies to develop conservation storage for hydroelectric power, navigation, municipal and industrial water supplies, and low-flow augmentation. Furthermore, there has been a substantial history of federally legislated financial assistance to farmers for drought relief both in general and in particular years. Local governments and the private sector have likewise made substantial investments in reservoirs to augment flows for public water supplies and hydro-electric power. About 85 percent of all municipal water supplies in the United States are provided by local governments, and a substantial share of that is taken from surface water reservoirs. In some cases those supplies are provided by federal reservoirs, and in California the state has made a large investment to develop water supply reservoirs and distribution systems. A special challenge facing all operators of surface water and ground water reservoirs is how to allocate available resources during a drought. While these facilities are usually designed to provide a given yield under drought conditions of specified frequency as determined from historical records, operators face considerable uncertainty during any drought as to what its magnitude and duration will be. They are usually confronted with the need to hedge against the possibility that a particular drought may be more severe than that for which the system was designed. USGS Role in Hydrologic Hazards The USGS provides several types of support to water resource managers and emergency management officials for addressing the issues associated with extreme floods and droughts, including (1) determining the probability of occurrence of extreme hydrologic events; (2) improving our understanding of the processes that determine the severity of extreme events, including anthropogenic factors; (3) understanding ancillary impacts of extreme hydrologic events such as bridge scour associated with floods; (4) providing tools for assessing alternative strategies for mitigating the impacts of floods and droughts; (5) monitoring of ground water levels and streamflow conditions; and (6) undertaking assessments
OCR for page 18
--> of the magnitude and extent of floods and droughts in support of emergency management officials. The long-term streamflow and ground water-level monitoring programs of the USGS provide the base information for determining the probability of occurrence of extreme hydrologic events. The current network of 7,000 daily streamflow stations and the more than 27,000 other stations that have previously been operated as daily discharge stations or peak discharge or low-flow stations provide a robust database for assessing flood and drought potential in many parts of the nation. It is not practical or economical to monitor every stream, so the USGS uses statistical techniques to estimate the probability of occurrence of floods and low streamflow of various durations for ungaged streams. These techniques typically employ regression equations for estimating specific flow characteristics based on physical characteristics, such as area of the watershed, average annual precipitation, slope of the stream channels, land use, and amount of water storage (lakes and reservoirs) in the watershed. The USGS's National Flood Frequency Program should soon be available on the World Wide Web for estimating peak flow characteristics throughout the nation. Regression equations for estimating low-flow characteristics are available in reports published by individual USGS district offices, generally in cooperation with the primary water management or natural resource agency of each state. A subelement of the probability of occurrence of floods and droughts is understanding the processes that determine the severity of a flood or drought. The climate conditions prevailing at the time of the event and antecedent conditions are major factors affecting the magnitude or severity of a flood or drought, but sometimes even these factors are obscured by human activities such as land-use practices, long-term water use, and changes in management of river systems. The USGS periodically conducts assessments of how streamflow characteristics change because of anthropogenic or climatic conditions. For example, many of the states that have regression equations for estimating flood characteristics of streams with undisturbed or natural watersheds also have ancillary equations for estimating flow characteristics of streams with watersheds with varying amounts of urban development. The USGS also has conducted studies to determine the effects of deforestation, drainage for improving crop productivity, and other land-use alterations. Impacts of flooding besides inundation include deposition of sediments in river channels, reservoirs, and floodplains and scour of river channels, particularly the foundations of bridge piers and abutments. The USGS monitors streams for sediment transport and has conducted various studies to determine the variation of sediment yield from watersheds with different land-use characteristics. Determination of the potential for large-scale input of sediment to streams from landslides and channel erosion is addressed as part of this activity. This information is used for identifying areas subject to extensive sediment transport and deposition to optimize the location of water storage and treatment facilities. It
OCR for page 19
--> Discharge measurements are especially important during extreme flow events, even though conditions are least favorable. Photo courtesy of U.S. Geological Survey. also provides a basis for designing flood control facilities, including providing adequate storage for water control and sediment accumulation. The USGS also has been conducting research on stream and bridge site characteristics that affect the potential for scour of the foundations of bridge piers and abutments. Physical characteristics such as channel slope and size of streambed materials have been used in conjunction with scour measurements to develop equations for estimating the potential for scour at existing or planned highway bridges. The research also has resulted in guidelines for state highway departments to consider while inspecting and designing bridges.
OCR for page 20
--> The development of tools to help resource managers and emergency officials plan efficient mitigation strategies is another aspect of USGS activities related to hydrologic hazards. The agency has developed various numerical models and other analytical techniques for predicting runoff from watersheds and for routing flood peaks from the headwater streams to major rivers. Watershed models are being used for assessing various alternatives for operating reservoirs for reducing flood hazards and strategies for distributing water to water-deficient areas during droughts. Watershed models also are used for evaluating nonstructural alternatives for reducing the impacts of floods. The WSPRO (Water-Surface Profiles) model developed in the mid-1980s has been used extensively to determine areas subject to inundation for floodplain management and flood insurance rate determinations. WSPRO and multidimensional models such as FESWMS (finite-element surface-water modeling system) also have been used to design culverts and bridges at single and multiple stream crossings of local, state, and interstate highways. Improved understanding of fluvial geomorphology through USGS research has resulted in sediment transport models for assessing the effects of floods and water regime modifications on aquatic and riparian ecosystems. The USGS role in collecting and distributing real-time streamflow and ground water-level data has expanded dramatically in the past two decades. There have been long-term agreements with the National Weather Service, the Army Corps of Engineers, and other agencies needing data for forecasting or operational decisions to provide access to the data through telephone or radio links to USGS monitoring sites. In the 1980s the USGS's capability to distribute hydrologic data directly to resource management and emergency management organizations was enhanced with the availability of instruments and satellites for transmitting data from remote locations. Today, the USGS serves real-time streamflow data from more than 3,000 monitoring sites to the World Wide Web via Geostationary Operational Environmental Satellites (GOES). These data are now available to resource managers, local emergency officials, and individual property owners who need to make decisions on how best to deal with either a flood or a scarcity of water. The other area where the USGS plays a significant role in hazard programs of various organizations is in rapid assessment of the magnitude and extent of floods. USGS personnel respond to extreme hydrologic hazards by collecting critical data during and immediately after the events to characterize the extent of flooding. This information includes the determination of flood stages and discharges to classify the floods for disaster assistance and insurance purposes. It also includes documentation of flood profiles and flood inundation maps for both short-and long-term planning. Emergency management officials use the information to make informed logistical decisions for distributing flood-fighting resources, rescue teams, potable water, food, medical supplies, and temporary housing. The information is also used for land-use planning and for assessing floodplain management strategies.
OCR for page 21
--> Issues and Implications for Information Needs A number of issues arise out of the national strategy to manage floods and droughts, and the need for a continued flow of information to support the strategy is readily apparent. This report addresses issues directly related to the USGS. Many of those issues and information needs are not new but are worthy of restating nonetheless. For the management of floods, there is a need for continued improvements in methods for estimating the frequency and severity of extreme events and their consequences. In addition to a need to continue efforts to improve techniques for long-range forecasting of floods, at least two other avenues are worthy of pursuit. First, on regulated streams and in urban areas and other places where hydrology is being modified by land development, the probabilities of flood peaks and volumes are being altered, often dramatically, as demonstrated by a number of field studies. The USGS periodically assesses those impacts using regression equations, but the generalization of results needs improvement. Hydrologists need enhanced procedures for adjusting the probabilities of extreme events in a timely manner. A second issue is the lack of readily available and up-to-date information about the consequences of extreme events for which probabilities have been increased by development activity. Despite technical information and warnings, communities are often caught off guard by flood events. The reasons for this are unclear but there are several contributing factors. Maintaining "readiness" requires a commitment of financial and other resources. Perhaps also local officials do not understand probabilities very well or the consequences of an event of given probability; perhaps the information is communicated in technical language not readily understood by the affected communities. Unfortunately there are also many instances where risk is well understood but ignored by local officials who may act unwisely (as, for example, local planning boards are often subject to political pressure to develop floodplain lands). Flood management also requires conditional forecasts of flood events given an occurrence of exceptional precursor events. Many flood events, like those in Northern California and the Red River, follow periods of exceptional snowfall. Others follow periods of sustained rainfall that leave soil in near-saturated conditions. Knowledge that events of this kind have occurred can substantially increase the probability of flooding in comparison to long-term probabilities. Such information is of significant value to flood preparedness activities. In the best of times, budget resources for the collection of basic information are limited, and the stream-gaging program is no exception to that rule. In an era of downsized government, every noncritical expenditure has come under closer scrutiny. Although the gaging program is critical, especially during extreme events, justification for continued support of the program has become more difficult in recent years. Such pressures continue to underscore the need for a more economically efficient means for improved data collection. Recent technological
OCR for page 22
--> advances have led to improved methods for gathering, storing, and transmitting data, but there is little evidence to suggest that the cost of data acquisition has decreased. Improved techniques for regionalizing the results of available gages also could be enhanced. The hydrology of droughts is much less well understood than the hydrology of floods. Unlike flood hazards, where flood-prone areas have been delineated to some degree at least by the probability of inundation, there is no systematic process for estimating the probabilities of drought. Droughts are defined differently for different purposes. An agricultural drought is not necessarily the same as a municipal water supply drought. It may be necessary to establish different definitions of droughts and different estimates of probabilities to respond to different needs. Improved techniques for estimating drought probabilities are needed, as are improved methods for communicating those probabilities and related consequences to the public. As with flooding, conditional forecasting of droughts could be helpful. It is one thing to know the long-term probability that a drought will occur in a particular region. It is another to know the probability of the duration and magnitude of a drought once it is recognized that one has started. Such information is crucial to the allocation of resources during drought events, to decisions whether to continue investments in crop production, or to decisions about the movement of goods by navigation or alternative means. Incorporation of knowledge about prevailing large-scale meteorological conditions and probabilities of when those conditions will change could produce more useful statements of probability.
Representative terms from entire chapter: