PURPOSE OF THE PRELIMINARY REPORT
This preliminary report of the Scientific Assessment and Strategy Team (SAST) is Part V of the Interagency Floodplain Management Review Committee (FMRC) report, Sharing The Challenge: Floodplain Management Into The 21st Century. This preliminary report documents much of the scientific information provided by the SAST for use by the full FMRC in its deliberations and summarizes SAST activities to date. The scientific assessments and analyses presented in this report are preliminary and have not received full critical scientific review. A comprehensive report is planned that provides detailed documentation of the analysis, mapping, and database activities conducted by the SAST. The level of detail in this preliminary report is uneven as a result of different levels of detail that were required for FMRC deliberations on specific topics and time limitations imposed on the FMRC. The preparation of this preliminary report is in partial completion of the goals and objectives of the SAST. A partial list of the other scientific and technical products provided to aid FMRC analysis is included at the end of this document (Appendix A.).

GOAL AND OBJECTIVES OF SAST
The flood of 1993 in the Upper Mississippi River Basin had peak discharges at many locations that exceeded any other peak discharges on record for those locations (Parrett and others, 1993). Although floods of the magnitude and duration of the 1993 event are rare occurrences, they are natural and will recur. The human and economic costs were high; yet, there were ecological benefits, such as improved spawning areas for some fish species and reconnection of some backwater areas to the channel. The negative and positive effects of the flood raised old concerns about flood control measures and habitat restoration. The Scientific Assessment and Strategy Team (SAST) was formed to help decision makers address those concerns by providing scientific advice on flood recovery and future management of the floodplain.

While the SAST was charged with focusing on floods and the structural and nonstructural methods of river basin management, the SAST members acknowledge that any action designed to adapt to or mitigate flooding must also take into account adapting to or mitigating low flows. Further, sustainable development of the river basin is considered in the gathering of scientific data and in scientific analysis.

The goal of the SAST is to provide scientific advice and assistance to Federal officials responsible for making decisions with respect to flood recovery in the Upper Mississippi River Basin and to develop and provide information to support the decision making process regarding both nonstructural and structural approaches to river basin management. The SAST objectives are:

• To develop a database of readily available data to support map production, scientific analysis, and decision making;
  
• to produce maps showing base information and vulnerability to flooding;
  
• and to prepare reports documenting the products of SAST and the methodology and analysis used to produce them, and identifying the ongoing monitoring, research, modeling, data-management and distribution requirements needed to support integrated river basin management.

There are 19 full time SAST members, 6 associate members, numerous ad-hoc members, and 3 project staff members. Initial funding was provided by FEMA with follow-on funding provided through two supplemental appropriations to the U.S. Geological Survey (USGS). Salaries are borne by the home agencies of the scientists and engineers involved. Significant in-kind support is provided by all of the agencies involved. In-depth technical and scientific support is provided by the Earth Resources Observation Systems (EROS) Data Center (EDC). States and a number of nongovernment organizations have been helpful as well by providing data, information, and analysis. Contributors are listed at the end of this report (Appendix C).

The SAST met at EDC near Sioux Falls, South Dakota, for an initial workshop during the week of December 13, 1993, to identify the scope of the problem and to begin to address the many logistical issues that the team would encounter. It began concentrated efforts at EDC on January 3, 1994, and continued until March 11, 1994. Since that date, the SAST continues to function as a distributed team with members working at their home offices or laboratories. The team also conducts workshops to address specific issues.

While at EDC, the SAST built a vast multilayer, multiresolution database covering the Upper Mississippi River Basin. The data densities vary spatially depending on the intensity of study that is required of the SAST. The most concentrated and complete data are along the floodplains of the upper Mississippi and lower Missouri Rivers because these floodplains represent the areas of most immediate interest to policy makers dealing with questions about response to the 1993 flood, the Federal levee system, and habitat restoration. The data are most sparse on the Upper Missouri River Basin upstream of Gavins Point, South Dakota. The primary purpose of the Upper Missouri River Basin data sets is to form a baseline of data and information for future studies since that area did not contribute appreciably to the flood of 1993. Intermediate data densities are in the areas that contributed to the 1993 floods. The database contains advanced very high resolution radiometer (AVHRR), Landsat Thematic Mapper (TM), and other satellite data, elevation data, selected digitized photographs, historical channel geometries, artificial structures, geologic, biologic, hydrologic, hydrographic, hazardous/toxic, and soil survey data, and data on many other topics.

Some of the SAST products include special maps, demonstrations of data applications, decision rules for identifying high priority habitat sites, methods for identifying reasonable alternative levee locations, and new understanding of the influence of variables such as focused flood-flow energy, the relationship between historical and current channel and sedimentation and scour, and land use on the impact of floods on the lower Missouri and upper Mississippi floodplains. The use of these products for management and decision making will be the subject of future scientific and management activities.

Data to populate the database and information for the preliminary report came from many sources including Federal agencies, state governments, universities, and nongovernment organizations (private industry and interest groups). Most data sets were modified from existing available data to make them intercomparable, to format them uniformly, or to otherwise improve their quality. Due to time limitations, many problems encountered in the data were identified in the metadata and not corrected when entered into the database. Quality assurance is an ongoing effort (see Chapter 2). Many data sets that were in the form of maps and tabular listings were digitized and included if they were useful for answering questions raised in the decision making process. Some data sets were built completely from scratch because of the critical need for them and their lack of availability.

APPROACH
To use the scientific method, observations must be made, hypotheses should be made based on those observations, and the hypotheses should be tested by determining how closely responses of the natural system to forcing functions correspond to predictions. Typically, there is an iterative cycle of refinement of the understanding and predictions. The system must be monitored to gather successive data on the observable variables. Those data must be compared to the predicted values of the variables.

The approach consisted of identifying potential questions that must be answered to support integrated river basin management, identifying the data needed to answer those questions, identifying sources for those data, acquiring those data and incorporating them into a compatible database, conducting analysis with the data, identifying additional data needs or developing results from the analysis, and when necessary, acquiring additional data.

It has been clear for many years (Changnon and others, 1983) that incomplete and inconsistent collection of data is a severe handicap to developing policy for floodplain management. In recent years, technology has been developed that aids in data analysis. Geographic information systems (GIS) and image processing systems provide powerful analytical tools that can aid research and assessment of river basin management on local to regional scales. Computerized process models provide a method of estimating process change under different conditions. Time-series analysis and visualization techniques provide methods to understand great amounts of data associated with events. Still, data acquisition and database development are the most difficult and expensive parts of implementing these modern technologies. In this case, for the management of the Upper Mississippi River Basin, a significant amount of data has been acquired already. These data include historical land use and land cover, historical river channels, precipitation, river flows, and many others. However, most of these data are not in digital form and must be digitized. In many cases, where the acquired data are in digital form, they are not in compatible formats; these data are also in the process of being converted. In addition, certain necessary data did not previously exist. These had to be interpreted from aerial photographs or satellite images, collected in the field, or otherwise acquired.

The database was the central focus of initial activities because analysis could not begin until sufficient data were available. The objective of the first analysis was to identify gaps in the original estimation of data needs. These needs were then satisfied either by acquiring additional existing data or by conducting field, remote sensing, process, or modeling analyses.

PRODUCTS
Products from the database include data, maps, illustrations, analysis results, and statistics. These are provided to policy makers as they become available. The generation of these products further refines the understanding of the process and aids in developing decision rules for scientific river basin management.

Ultimately, the results are reported. The SAST is producing a multivolume report that provides scientific information and makes recommendations. The volumes of this report are as follows:

Volume 1. Preliminary Report documents general scientific background and specific narrowly defined analyses provided to the full FMRC for use in deliberations to produce the Report of the Interagency Floodplain Management Review Committee to the Administration Floodplain Management Task Force,

Volume 2. Scientific Report documents in-depth analysis conducted by the SAST using existing and newly developed data and techniques to provide improved understanding of scientific information for river-basin and floodplain management. Techniques, understanding, and management recommendations are presented to aid policy makers and resource managers. This report is currently unfunded.

Volume 3. Database Report provides a detailed description of the data and database for users. This includes metadata, descriptions of the strengths and weaknesses of the data, acquisition methods, data maintenance plans, and data distribution methods,

Volume 4. Scientific Background Report contains the papers and study reports that were commissioned by the SAST to answer specific questions. These papers will be published to ensure that this publicly funded analysis is readily available to the public, and

Volume 5. Proceedings of the SAST Hydraulic, Hydrologic, and Ecologic Modeling Workshop (February 15 & 16, 1994) contains papers presented by workshop speakers and selected discussions of the workshop participants.

Additional reports and scientific papers will be published in the scientific literature as they become available.

Basin Extent

GEOGRAPHIC DESCRIPTION OF THE UPPER MISSISSIPPI RIVER BASIN
The Upper Mississippi River Basin consists of areas drained by the Mississippi River above the confluence with the Ohio River at Cairo, Illinois. It includes the entire Missouri River basin, which joins the Mississippi above St. Louis and extends west to the Continental Divide. The Missouri River and its tributaries drain most of the northern Great Plains. The Upper Mississippi River Basin covers 23 percent of the conterminous United States, an area of 700,000 square miles. The entire Mississippi River basin drains 1,270,000 square miles, or 41 percent of the conterminous United States.

The states that are entirely or partly contained in the SAST detailed study area are Iowa, Missouri, Illinois, Wisconsin, Minnesota, North Dakota, South Dakota, Nebraska, and Kansas. The SAST detailed study area lies almost completely in the Central Lowlands physiographic province, with small areas in the Ozark Plateaus and Superior Uplands provinces. Figure 1.1 illustrates the geographic extent of the study area.

For issues related to floodplain management and the areas most severely affected by flooding in 1993, three smaller study areas were identified. These are 1) the main stem of the Mississippi River above Cairo, Illinois, 2) the main stem of the Missouri River below Gavins Point Dam near Yankton, South Dakota, and, for most of the issues, 3) the main stem of the Illinois River. For the SAST study, the lower Missouri River extends from the Gavins Point Dam near Yankton, South Dakota, where the river enters the Central Lowland physiographic province, to the confluence with the Mississippi River at St. Louis, Missouri. Gavins Point Dam is considered the upstream boundary of the detailed study area because that dam significantly limited the contributions of the upper Missouri River basin to the flood of 1993. Defined more precisely, this area excludes the Missouri River above Fort Randall Dam and the upstream reaches of the western tributaries to the Missouri River -- the Platte and Osage Rivers. It includes the Big Sioux, Vermillion, and James Rivers, and the lower reaches of the Kansas, Platte, and Osage Rivers.

The geographic setting of the Upper Mississippi River Basin is described in many general reference works. For the description here, climate data are from Baldwin (1973) and drainage areas, river miles, and elevations are from the published records in the annual USGS Water Data Reports. Useful summaries of surface water hydrology, floods, and droughts in the Mississippi River basin (and throughout the nation) are given in National Water Summary reports published in the USGS Water-Supply Papers (Moody and others, 1986; Paulson and others, 1991).

The area of the Missouri River drainage basin above Yankton is about 280,000 square miles; the area at the mouth at St. Louis is about 526,000 square miles. (These areas exclude about 3,000 square miles of noncontributing area in the Great Divide closed basin). The lower Missouri River flows about 800 miles from Yankton (river gage datum elevation 1,140 feet above sea level) to the mouth at St. Louis (400 ft), with a nearly uniform slope of about 0.9 feet/mile. The mainstem floodplain ranges from about 2 to 10 miles in width; it is relatively narrower in the last 200 miles of the river's course than upstream. The total area of the lower Missouri floodplain is about 2 million acres (3,100 square miles).

Most of the drainage area of the Missouri River is in the Great Plains. Average annual precipitation depths range from less than 12 inches near the Rocky Mountains to about 24 inches in eastern South Dakota and 32 inches in eastern Kansas; average annual precipitation increases to about 39 inches near St. Louis. Most of the precipitation occurs in the summer, late spring, and early fall months. Mean annual pan evaporation greatly exceeds precipitation throughout most of the basin. Mean annual runoff for the mainstem lower Missouri River is correspondingly low, ranging from 1.35 inches at Yankton and Omaha to 1.99 inches at Hermann, Missouri, the lowest gage on the Missouri River (mile 97). The high runoff season is in the spring, when precipitation and snowmelt are high, soil is bare, and evaporation losses have not yet increased to their high summer levels. The seasonal distribution of flow in the main stem is heavily regulated for navigation, power generation, and flood control purposes by six major reservoirs in the upper basin above Yankton. About 1/3 of the flow at Hermann is derived from the drainage area between Kansas City and Hermann, which amounts to about 1/8 of the total basin area. Mean annual runoff from the smaller tributary basins ranges from about 3 to 5 inches in western Iowa and eastern Nebraska to 10 - 15 inches in central and southern Missouri. The SAST detailed study area covers 270,000 square miles and parts of 9 states, but some data have been collected and analysis has been conducted for the other parts of the Upper Mississippi River Basin because those data and analyses help explain the processes in the detailed study area.

The Missouri River drains 73 percent of the Upper Mississippi River Basin, but contributes an average of 36 percent of the total streamflow from the basin. Long-term average streamflow of the Missouri River at Hermann, Missouri, is 72,000 cubic feet per second (cfs), compared to 198,000 cfs for the Mississippi River at Thebes, Illinois, about 20 miles upstream from Cairo (Moody and others, 1986). Peak flows during the 1993 floods were rated at 750,000 cfs at Hermann and 975,000 cfs at Thebes (Parrett and others, 1993).

The area drained by the upper Mississippi River extends from Lake Itasca (elevation 1467 feet above sea level) in northern Minnesota to the mouth of the Ohio River at Cairo, Illinois. For about 660 miles between Minneapolis-St. Paul and St. Louis, the river flows through a series of navigation pools controlled by a total of 29 locks and dams. The average river slope is about 0.5 feet per mile between St. Paul (river mile 839 above Cairo, gage datum elevation 684 ft) and St. Louis (mile 180, datum 380 ft). Between St. Louis and Thebes, Illinois, the lowest gage on the upper Mississippi (mile 44, datum 300,000), the river is undammed and has a slope of about 0.6 feet per mile.

The drainage area of the upper Mississippi River (excluding the Missouri River basin) lies in the continental humid climate zone. Mean annual precipitation is about 24 to 30 inches in southern Minnesota and increases to about 44 inches near Cairo. Precipitation is concentrated mainly in the summer, late spring, and early fall months. Mean annual runoff of the upper Mississippi River is between 7 and 8 inches upstream from St. Louis and between 3.5 and 3.8 inches in the middle Mississippi River downstream from the confluence of the Missouri River. Although the upper Mississippi River is dammed for navigation between St. Paul and St. Louis, the water storage capacity of the pools is small, and, in contrast to the case on the Mississippi River, the dams have negligible effect on flood flows and seasonal distribution of runoff. Average annual runoff depth from small tributary basins ranges from about 3 inches in southwest Minnesota to 6 inches in central Iowa, to about 8 - 10 inches in Illinois, to 10 - 12 inches in Wisconsin, paralleling the trends in pan evaporation and precipitation and incorporating the effects of varying landforms, soils, and surficial materials.