QUANTIFICATION OF HISTORIC FLOOD DAMAGE IN THE MAPLE AND WILD RICE WATERSHEDS OF THE RED RIVER OF THE NORTH BASIN

Transcription

1 QUANTIFICATION OF HISTORIC FLOOD DAMAGE IN THE MAPLE AND WILD RICE WATERSHEDS OF THE RED RIVER OF THE NORTH BASIN A Thesis Submitted to the Graduate Faculty of the North Dakota State University of Agriculture and Applied Science By Michael Edward Kjelland In Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE Major Program: Natural Resources Management Major Department: Agribusiness and Applied Economics March, 2001 Fargo, North Dakota

2 Acid Free Committee Sheet ii

3 ABSTRACT Kjelland, Michael Edward, M.S., Department of Agribusiness and Applied Economics, College of Agriculture, North Dakota State University, March Quantification of Historic Flood Damage for the Maple and Wild Rice Watersheds of the Red River of the North Basin. Major Professor: Dr. Steven D. Shultz. Between 1989 and 1998, flood damage in the Maple Watershed (MW), North Dakota, was an estimated $29.4 million and $102 million in the Wild Rice Watershed (WRW), Minnesota. Non-agricultural damage records were reviewed and attributed to a watershed, or extrapolated to a watershed based on a county s percentage area or population within the watershed. Such adjustments reduced non-agricultural damage estimates by 80 percent (MW) and 27 percent (WRW). Agricultural damage, 55 percent of total damage (MW) and 31 percent (WRW), was estimated by extrapolating previous estimates based on relationships between peak flood stages and cropland flooding. Without extrapolation, agricultural damage was underestimated by 77 percent (MW) and ranged from 69 percent underestimation to 72 percent overestimation (WRW). Without extrapolation and adjustments, total flood damage was underestimated by 59 percent (MW) and between 4 to 32 percent (WRW), indicating the need for improved flood damage data collection, organization, and reporting methods. iii

4 ACKNOWLEDGMENTS I would like to extend a special thank you to my advisor Dr. Steven D. Shultz whose patience, insight, and reviews of my thesis were greatly appreciated. I would also like to express a sincere thank you to the other members of my committee: Dr. Jay A. Leitch, Dr. Lowell A. Disrud, Dr. G. Padmanabhan, and Dr. Carolyn Grygiel, for their insights and suggestions. I would like to thank my family and friends for coping with my absence from events at times due to my research and graduate studies. I need to thank Melanie Bengtson, Dean Bangsund, Dr. F. Larry Leistritz, Dr. (M.D.) Steve Baggin, Aaron Meyer, Pat Fridgen, and Kevin Kermes for their valuable help. Also, I wish to thank the individuals that helped me from the agencies listed within this thesis. Lastly, I would like to thank everyone else at the North Dakota State University Agribusiness and Applied Economics Department that aided me in my graduate career for their friendly can do attitudes. iv

5 DEDICATION This thesis is dedicated to my true friend Karlton York Hoff (July 6, 1976 June 17, 1994). In our youth, and accompanied with imagination, we forged empires; vanquished evil; and explored newly discovered worlds. M.E.K. "Imagination is more important than knowledge. Knowledge is limited. Imagination circles the world" ~Albert Einstein v

6 TABLE OF CONTENTS ABSTRACT...iii ACKNOWLEDGMENTS... iv DEDICATION... v LIST OF TABLES... vi LIST OF FIGURES... vii CHAPTER 1. INTRODUCTION... 1 CHAPTER 2. LITERATURE REVIEW Introduction The Study Sites Red River Basin Maple Watershed Wild Rice Watershed Flood Damage in the RRB, Maple, and Wild Rice Watersheds Red River Basin flood damage assessments Maple Watershed flood damage assessments Wild Rice Watershed flood damage assessment Agencies Disbursing Flood Payments Small Business Administration (SBA) Federal Emergency Management Agency (FEMA) United States Department of Agriculture (USDA) CHAPTER 3. STUDY PROCEDURES Introduction Non-agricultural Flood Damage Estimates Types of Non-agricultural flood damage Extrapolating county-level data to the watershed-level of analysis Administrative Charges with DSR s Agricultural Flood Damage Estimating agricultural flood damage in the Maple Watershed Acres of flooded cropland Flood damage per acre Incidental Flood Damage Estimating agricultural flood damage in the Wild Rice Watershed Acres of flooded cropland Flood damage per acre Incidental flood damage Comparisons of flood damage within each watershed CHAPTER 4. MAPLE WATERSHED FLOOD DAMAGE ESTIMATES Introduction Historic Flood Damage by Category Observations FEMA non-agricultural flood damage payment data Agricultural flood damage vi

7 TABLE OF CONTENTS (Continued) CHAPTER 5. WILD RICE WATERSHED FLOOD DAMAGE ESTIMATES Introduction Historic Flood Damage by Category Observations FEMA non-agricultural flood damage payment data Agricultural flood damage CHAPTER 6. SUMMARY AND CONCLUSIONS Introduction Agricultural Flood Damage Estimates The Effect of Population and Area Adjustments on Flood Damage Estimates Relationships Between Flood Characteristics and Damage Conclusions and Implications LITERATURE CITED LIST OF ACRONYMS APPENDIX A. NON-AGRICULTURAL FLOOD DAMAGE BY YEAR APPENDIX B. FEMA PUBLIC ASSISTANCE BY WATERSHED AND YEAR APPENDIX C. SECONDARY FLOOD EVENT AGRICULTURAL DAMAGE APPENDIX D. HISTORIC FLOOD DAMAGE (ADJUSTED & EXTRAPOLATED) APPENDIX E. MODEL FLOOD DAMAGE REIMBURSEMENT SPREADSHEET vii

8 LIST OF TABLES Table Page 2.1 Flood damage estimates in the RRB Flood damage estimates in the Maple Watershed Flood damage estimates in the Wild Rice Watershed The proportion of county population and acreage within each watershed Procedures in extrapolating county flood damage to the watershed-level Maple Watershed acres flooded Maple Watershed estimated crop flood damage, 1975 and Maple Watershed seasonal crop flood damage per acre by reach Maple Watershed historic flood damage summary table, 1975 and Wild Rice Watershed acres flooded Maple Watershed crop damage per acre (1979 dollars) Maple Watershed agricultural flood damage Wild Rice Watershed crop damage per acre (1979 dollars) Wild Rice Watershed agricultural flood damage The impact of extrapolating and adjusting on flood damage estimates A-1. Maple Watershed non-agricultural flood damage by year A-2. Wild Rice Watershed non-agricultural flood damage by year C-1. C-2. Maple Watershed secondary flood event agricultural damage Wild Rice Watershed secondary flood event agricultural damage D-1. Watershed-level flood damage (adjusted & extrapolated) E-1. Optimal organization of flood damage reimbursement information viii

9 LIST OF FIGURES Figure Page 2.1. Study site location: Maple and Wild Rice Watersheds of the Red River Basin North Dakota counties within the Maple Watershed Minnesota counties within the Wild Rice Watershed Classifications of flood damage Procedures for estimating agricultural flood damage Increasing rate of crop damage over crop production cycle Maple Watershed historic flood damage ( ) Maple Watershed historic flood damage (including infrastructure) Maple Watershed historic flood damage by year (including infrastructure) Maple Watershed FEMA flood damage payments ( ) Maple Watershed historical FEMA flood damage payment programs Population density in townships of the Maple Watershed Maple Watershed FEMA Public Assistance payments for flood damage ( ) Maple Watershed FEMA National Flood Insurance Program payments for flood damage ( ) Maple Watershed flood damage by category for individual flood events Maple Watershed percentage change in agricultural flood damage with respect to season/crop cycle and river peak Wild Rice Watershed historic flood damage ( ) ix

10 Figure LIST OF FIGURES (continued) Page 5.2. Wild Rice Watershed historic flood damage (including infrastructure) Wild Rice Watershed historic flood damage by year (including infrastructure) Wild Rice Watershed FEMA flood damage payments ( ) Wild Rice Watershed historical FEMA flood damage payment programs Population density in townships of the Wild Rice Watershed Wild Rice Watershed FEMA Public Assistance payments for flood damage ( ) Wild Rice Watershed FEMA National Flood Insurance Program payments for flood damage ( ) Wild Rice Watershed flood damage by category for individual flood events Wild Rice Watershed percentage change in agricultural flood damage with respect to season/crop cycle and river peak Alternative annual agricultural flood damage estimates, Maple Watershed Alternative agricultural flood damage ( ), Maple Watershed Alternative annual agricultural flood damage estimates, Wild Rice Watershed Alternative agricultural flood damage ( ), Wild Rice Watershed Alternative estimates of total flood damage (agricultural and non-agricultural), Wild Rice Watershed ( ) Historic flood damage in the Maple Watershed, with and without population and area adjustment strategies Historic flood damage in the Wild Rice Watershed, with and without population and area adjustment strategies Maple Watershed flood damage & hydrologic characteristics x

11 Figure LIST OF FIGURES (continued) Page 6.9. Wild Rice Watershed flood damage & hydrologic characteristics B-1. B-2. B-3. B-4. B-5. B-6. B-7. B-8. B Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Maple Watershed FEMA Public Assistance payments for flood damage Wild Rice Watershed FEMA Public Assistance payments for flood damage Wild Rice Watershed FEMA Public Assistance payments for flood damage B Wild Rice Watershed FEMA Public Assistance payments for flood damage xi

12 CHAPTER 1 INTRODUCTION Extensive flooding in the Red River Basin (RRB) in 1997 increased the demand for more detailed and accurate flood damage estimates in order to evaluate flood mitigation efforts. Specifically, flood damage data are required to insure that appropriate resources are allocated to the areas where they are needed and to justify investments in various types of flood mitigation projects (HDR Engineering, Inc. & Houston Engineering, Inc., 1998). Previous flood damage estimates within the RRB do exist; however, there are some common problems with them. First, the estimates are too general with respect to the classification of damage types. In other words, flood damage estimates are usually not segregated into classifications such as residential, agricultural, industrial, transportation, or commercial. Second, the majority of previous flood damage estimates consist of general averages that do not distinguish between floods of high and low frequencies or springtime and summer time flood events. Third, flood damage assessments do not include very recent flood events, cover only a specific flood event/flood year, or do not provide detailed procedures used to estimate the flood damage. Finally, the flood damage estimates are either too site specific, covering only cities such as Fargo or Grand Forks, or too general in that they attempt to report overall aggregate damage. This makes it difficult to evaluate the economic feasibility of various flood mitigation efforts, at the watershed, community, or township-level of analysis.

13 The primary objective of this study was to evaluate and develop new strategies to estimate detailed (watershed specific) flood damage from 1989 to 1998, in two watersheds of the Red River Basin: the Maple Watershed in North Dakota, and the Wild Rice Watershed in Minnesota. A related objective was to examine the ability of local, state, and federal agencies in providing historical flood damage data. The Maple and Wild Rice Watersheds were chosen as study sites because they are typical watersheds of the Red River Basin that have been the focus of frequent flooding and flood mitigation projects over the last decade. The time frame of study, from 1989 to 1998, was chosen based upon the availability of both hydrologic and flood damage data and the wide range, or different types, of floods that have taken place within the specified time frame. The historical flood damage for each watershed was sub-divided into ten distinct damage categories: agricultural, debris clearance, protective measures, transportation, water control facilities, commercial/industrial, public utility systems, public facilities, residential, and miscellaneous. Historical flood damage in the watersheds was estimated using a variety of procedures. First, a literature review was conducted to solicit flood damage payment data from public and private agencies. Second, available data were categorized by the type of damage and time period in which it occurred. Finally, the existing data were extrapolated to time periods without data and to alternative levels of geographical analysis. Flood damage estimates for the years without data were based, in part, on identifying similar flooding conditions such as: peak flows and flood stages, dates of flooding, and the number of days in which flooding took place. Identifying watershedlevel flood damage, in cases when area based data were available only at the county-level 2

14 of analysis, was accomplished through dividing the total flood damage reported for individual counties by the percentage of the counties land areas within the watershed. Similarly, population based flood damage data at the county-level of analysis were divided by a county s percentage population within the watershed to determine watershed specific flood damage. The results of this study are presented in separate chapters for each watershed and sub-divided by agricultural, residential, and other flood damage types. Chapter 2 summarizes the characteristics of each of the watersheds with a focus on flooding issues, followed by a description of previous studies that have estimated flood damage in the Red River Basin, and in particular, the Maple and Wild Rice Watersheds. Chapter 3 presents the detailed procedures, assumptions, and data sources. Chapters 4 and 5 contain the results of the assessment and estimates of flood damage for the Maple and Wild Rice Watersheds respectively. Chapter 6 provides comparisons of previous flood damage estimates with the present study s estimates, and discusses the impact of utilizing various assumptions and specific methods, namely population and area adjustments. Chapter 7 summarizes and discusses the economic and social implications of the study. In addition to discussing the potential of such procedures in other locations and time frames, specific recommendations are made regarding ways for public and private agencies to better collect, organize, and disseminate flood damage data. 3

15 CHAPTER 2 LITERATURE REVIEW 2.1 Introduction This chapter contains information on the biophysical and socioeconomic conditions, and historical flooding, in the Red River Basin (RRB), Maple Watershed, and the Wild Rice Watershed. The intention is to familiarize the reader with the characteristics of the study areas, including agencies and programs involved with flood-related issues. 2.2 The Study Sites It is important to mention that the Maple and Wild Rice Watersheds of the Red River Basin may also be referred to as the Maple River and Wild Rice River Watersheds. The Maple Watershed is representative of other watersheds in the Red River Basin with very little variation concerning terrain elevation and soil characteristics. The Wild Rice Watershed is representative of watersheds in the Red River Basin having both Glacial Lake Plain and Uplands and thus more variation regarding soil characteristics Red River Basin The RRB covers 17,000 square miles across parts of the states of North Dakota, Minnesota, South Dakota, and the province of Manitoba, Canada (Figure 2.1). The Red River flows through the bed of glacial Lake Agassiz, which occupied the area 10,000 to 15,000 years ago. The former shorelines of glacial Lake Agassiz are marked by low beach ridges. Glacial drift in the uplands is composed mainly of an unsorted and un-stratified mixture of clay, silt, sand, and gravel known as till. The uplands contain the majority of 4

16 Figure 2.1. Study site location: Maple and Wild Rice Watersheds of the Red River Basin. the scattered lakes, ponds, and existing wetlands in the RRB (Flood Response, The Red River, ND MN, 1998). Flooding is a perennial problem in the RRB due to four major reasons. First, soils in the RRB have very low percolation rates and high rates of runoff (Flood Response, The Red River, ND-MN, 1998). Second, the slope of the Red River averages less than 6 inches per mile, slowing the rate at which water can run through the RRB (Red River Flooding, 1998). Third, the Red River flows northward resulting in occasional ice jam problems; and when thawing takes place in the Southern end of the RRB, the Northern end may still be frozen causing the Red River to back up and overspill its banks. Lastly, the Red River channel is small, varying from 200 to 500 feet wide and 10 to 30 feet deep, thereby increasing the ease at which it can overspill its banks (A River Runs North Managing an 5

17 International River, 1993). Five conditions conducive to spring flooding in the RRB include an unusually wet fall, an unusually cold winter, an unusually heavy winter snow accumulation, an unusually late and cool spring followed by a sudden warming trend, or widespread, heavy, warm rainfall during the thawing period (The Raging Red, 1996). The increase in urbanization, population, and the wet cycle of the past ten years, in the RRB, has led to increasingly higher economic losses associated with Red River flooding. The 1997 flood was the largest and most expensive recorded flood in the history of North Dakota and Minnesota (The Floods of 1997, 1997, & MN Recovers Disaster Task Force, 1998). Flood records in Manitoba, Canada, reveal that the most severe Red River flood of modern times occurred in The unregulated discharge (cfs) in 1997 at Winnipeg was 162,000 (cfs) compared to 225,000 (cfs) estimated for Several other major (low frequency) floods in the Red River Basin have occurred. Minor (high frequency) floods occurring this past century in significant portions of the Red River Basin from 1950 onward include the years: 1965, 1969, 1975, 1978, 1989, 1993, and 1998 (A River Runs North Managing an International River, 1993). The floodplain of the Red River constitutes less than 10 percent of the basin s total area, and extends from 10 to 15 miles on either side of the river. Typically, the brunt of flood damage in the RRB occurs near the Red River main stem or in the lower reaches of its tributaries. Several tributaries of the Red River that are more inclined to flooding include the Pembina, Maple, Sheyenne, Wild Rice, Marsh, and Red Lake Rivers. (A River Runs North Managing an International River, 1993). 6

18 2.2.2 Maple Watershed The Maple Watershed in eastern North Dakota lies within five counties including Barnes, Cass, Ransom, Richland, and Steele (Figure 2.2) and 63 townships. A township is an approximately 36-square mile area formed by township and range lines in the rectangular Figure 2.2. North Dakota counties within the Maple Watershed. 7

19 survey system (Floyd & Allen, 1999). The total area of the watershed is approximately 1,024,000 acres, which equates to 1,600 square miles and contains 48 sub-watersheds (USACE, 1990). The watershed consists primarily of Glacial Lake Plain dominated by clay soils. The Maple River flows from north to south through a well-defined valley until the town of Enderlin, from then on it flows northeast onto the lakebed. At Enderlin, the river s velocity slows as it enters the lakebed, causing siltation and lowering of the stream capacity causing increased flooding. About five miles to the north of West Fargo, the Maple River joins the Sheyenne River (USACE, 1990). Agriculture is the predominate land use in the Maple Watershed. Principal agricultural crops include wheat, barley, oats, sugar beets, potatoes, corn, beans, forage grasses, and sunflowers (Red River Basin Information Document, 1997). Typical spring planting begins in late April or early May, after fields have dried out from the snow thaw Wild Rice Watershed The Wild Rice Watershed, a sub-watershed of the RRB in northwestern Minnesota, lies within 86 townships in six counties: Becker, Clay, Clearwater, Mahnomen, Norman, and Polk (Figure 2.3). The total area of the watershed is 1,293,468 acres (2021 square miles) and contains 80 sub-watersheds, excluding the Marsh River portion (Houston Engineering, 1999). There are three distinctive soil groupings in the watershed. Clay soils primarily dominate the Glacial Lake Plain. These soils are very productive agricultural lands, have low permeability, and have poor internal drainage. The soils in the Beach Ridge Area are 8

20 more diversified, ranging from clay loams and sandy loams, to sands and gravels. Clays and silts make up the Glacial Moraine. Drainage within the soils of this area tends to vary from poor to good (Houston Engineering, 1999). Most flood damage typically occurs in the western portion of the watershed, in a 75 square mile area around the city of Ada. The majority of the watershed flood flows are confined to narrow and sharply defined stream valleys (Reconnaissance Report, 1988). Approximately 74 percent of the watershed is dedicated to crop and pasture, especially in Figure 2.3. Minnesota counties within the Wild Rice Watershed. 9

21 the western portion. Principal agricultural crops include wheat, barley, oats, sugar beets, potatoes, corn, beans, forage grasses, and sunflowers (Red River Basin Information Document, 1997). 2.3 Flood Damage in the RRB, Maple, and Wild Rice Watersheds Several flood damage studies have focused on the RRB, and specifically the Maple and Wild Rice Watersheds. However, many of these studies are somewhat limited because they do not specify types of damage in sufficient detail, have limited geographic scopes and time frames, or often report damage amounts that are inconsistent with those reported by other studies. The lack of specificity in past RRB flood damage studies is associated with the reporting of only agricultural or urban/residential flood damage values. Moreover, the flood damage assessments are often for a specific year, like 1997, rather than for extended periods of time, and do not distinguish between high and low frequency, or springtime and summertime, floods. For many flood mitigation planning efforts, it would be useful if flood damage was assessed annually over extended periods of time with additional flood damage classifications for transportation, commercial/industrial, protective measures, and infrastructure. Flood damage estimates have been either too detailed, covering only cities such as Fargo, Ada, and Grand Forks, or too general, covering a whole state or the entire RRB. Flood damage assessments at the watershed-level are important because drainage patterns in that area differ from surrounding watersheds. A watershed board, in charge of planning 10

22 flood damage mitigation in a watershed district, should have watershed specific flood damage information. In some cases, there is a high degree of variance among flood damage estimates from individual studies focusing upon the same area. For example, two independent flood damage estimates in North Dakota as a result of the 1997 flooding were $361 million (LeFever, Bluemle, & Waldkirch, 1999) and $309 million (Carlson, 1998). The city of Grand Forks (1999) estimated 1997 flood damage for Grand Forks, ND at $1 to 2 billion ( The study, Service Assessment & Hydraulic Analysis, Red River of the North 1997 Floods (United States Department of Commerce, National Oceanic & Atmospheric, National Weather Service, 1998) estimated $3.6 billion in flood damage for Grand Forks/East Grand Forks. These large differences between flood damage estimates create a particularly serious problem in comparing mitigation costs and benefits. The qualitative assessment of the past studies of flood damage is organized into three sections: Red River Basin, Maple Watershed, and the Wild Rice Watershed. Data and procedures from several of these studies were used directly or extrapolated for use in the present study Red River Basin flood damage assessments It was estimated in the study, 1993 Red River Valley Final Flood Damage Assessment, that $72 million in payments by federal, state, and local agencies were made for reimbursement of 1993 flood damage in ten Minnesota watersheds (including the Wild Rice Watershed) of the Red River Basin (HDR Engineering, Inc. and Houston Engineering, 11

23 Inc., 1998). Flood damage payments were categorized under residential, commercial/industrial, public facilities, transportation, utilities, emergency costs, and agricultural categories. However, the distinction between springtime and summer time flood events and damage was not made. The primary method for estimating flood damage was contacting public and private agencies responsible for reimbursing both agricultural and non-agricultural flood damage. A second method for determining agricultural flood damage involved averaging crop production data (obtained from the National Agricultural Statistics Service) two years before and one year after the 1993 flood, and subtracting that figure from 1993 crop production levels (HDR Engineering, Inc. and Houston Engineering, Inc., 1998). The Farm Service Agency s Flood Survey provided estimates of agricultural damage associated with the 1997 flood for all counties in Minnesota. Unfortunately, reported flood damage was not separated from a large wind/ice-storm that preceded the flood. An Assessment of Recovery Assistance Provided in Canada and the United States After the 1997 Floods in the Red River Basin (Natural Hazards Center - University of Colorado & Disaster Research Institute - University of Manitoba, 1999) focused primarily on the Red River Basin as a whole, and several specific cities (Winnipeg, Manitoba, Grand Forks, North Dakota (ND), and East Grand Forks, Minnesota (MN)). The study was conducted for the International Red River Basin Task Force International Joint Commission. Several agencies were contacted and provided flood payment amounts listed with the corresponding city receiving the aid. 12

24 The United States Army Corps of Engineers estimated the flood-related damage in Grand Forks, ND and East Grand Forks, MN was between $1 and $1.5 billion (An Assessment of Recovery Assistance Provided in Canada and the United States After the 1997 Floods in the Red River Basin, 1999). In another study, Service Assessment & Hydraulic Analysis, Red River of the North 1997 Floods the Corps of Engineers and the U.S. Department of the Interior estimated that flood damage totaled $4 billion for the entire RRB in 1997 (United States Department of Commerce, National Oceanic & Atmospheric, National Weather Service, 1998). The method used to quantify flood damage was not identified. The Minnesota Recovers Disaster Task Force Anniversary Report 1998 provided descriptive information concerning the Flood of 1997 as well as dollar amounts of assistance and other financial commitments made by county, state, and federal agencies which totaled more than $1.5 billion (State of Minnesota, 1998). The method used to arrive at the estimate was not given. Estimates of flood damage assistance, in the form of mitigation and economic development, were reported for numerous counties, watershed districts, townships, and cities in MN. The Floods of 1997 estimated that combined damage from flooding and the ice storm of 1997 was approximately $2 billion (North Dakota State Water Commission, 1997). The city of Grand Forks provided its own estimate of flood damage as being $1 to $2 billion for 1997 ( The specific method for estimating flood damage amounts was missing. The study, Economic Effects of the 1997 Flood on the Red River Valley (Carlson, 1998), reported flood-related damage for Grand Forks-East Grand Forks as being at least 13

25 $253 million. Another study, Flooding in the Grand Forks-East Grand Forks North Dakota and Minnesota Area (LeFever, Bluemle, & Waldkirch, 1999), listed flood damage for 1997 North Dakota flooding, an estimated $361 million. Several of the studies listed in this section did not include a description of the method to estimate flood damage. Presumably, the method used in these studies consisted of surveying flood damage payment amounts accounted for by various flood-mitigating agencies. The RRB studies and corresponding flood damage estimates are listed in Table 2.1. Table 2.1. Flood damage estimates in the RRB Source Location Flood Damage Note Year Estimate HDR Engineering, Inc. & RRB of 1993 $75.3 million $189 million + Houston Engineering, Inc., 1998 Minnesota (paid) (surveyed) University of Colorado & Grand Forks/ 1997 $1.5 billion Cited by United States University of Manitoba, 1999 EGF Army Corps of Engineers Minnesota Recovers Disaster State of 1997 $1.5 billion + Task Force, 1998 Minnesota North Dakota State Water State of 1997 Almost Includes both flooding Commission, 1997 North Dakota $2 billion & ice storm damage City of Grand Forks, 1999 Grand Forks 1997 Between Listed on website $1-2 billion ( Recovery/finance.htm) U.S. Dept. of Commerce, Entire RRB 1997 $4 billion Cited by United States National Oceanic & Atmospheric, U.S. & Army Corps of Engineers National Weather Service, 1998 Canada Carlson, 1998 Grand Forks 1997 $253 million State of MN 1997 $250 million State of ND 1997 $309 million Manitoba 1997 $70 million (Canadian Dollars) LeFever, Bluemle, & State of ND 1997 $361 million Waldkirch,

26 2.3.2 Maple Watershed flood damage assessments The Corps of Engineers estimated agricultural flood damage in the Maple Watershed for 1989 and 1993 (USACE After Action Reports, 1993 and 1996). The estimates were based on stage damage curves, historic flood discharge levels measured in cubic feet per second (cfs) associated with crop, and soil damage per acre of cropland flooded. No distinction was made between spring and summer flooding. The estimated crop damage per acre of flooding in the Maple Watershed was $15 in both 1989 and Other agricultural damage (non-crop losses such as farm machinery and infrastructure) associated with flooding was assumed to be 20 percent of crop damage ($3/flooded acre). The Corps of Engineers made these estimates using a Computerized Agricultural Crop Flood Damage Assessment System (CACFDAS) model for the watershed. The CACFDAS model computes crop flood damage based on initial crop distributions (acres of crop type), computerized crop budget data (planting and harvest costs), alternative crops, damage duration data (inundation time), and daily historic hydrologic information on river stages and flows (Killcreas, 1981). The CACFDAS estimated flooded acres and crop damage in the Maple Watershed for 32 years, as recent as the flooding in The CACFDAS results included the flood year, the date flooding occurred, assigned river reaches with the number of agricultural acres flooded, and the amount of crop damage in dollars. Crop damage, as defined by the CACFDAS analysis, consisted of damage from foregone net revenue. Data were missing for some years that should have shown a large amount of flood damage, such as However, a flood of similar conditions associated 15

27 with the flood of 1993, namely the flood of 1975, was listed as having 48,621 acres flooded and $9,071,901 (1998 Dollars) of crop damage. The Maple River Dam Cass County, North Dakota (USACE, 1990) study estimated annual average agricultural, urban, non-urban, and transportation damage for the Maple Watershed from 1969 to The number of acres flooded along with the magnitudes of the flood events (average by expected recurrence intervals, i.e. 5-year, 10- year, etc.); were listed along with average annual damage amounts. The average annual agricultural damage amount was based on acres flooded in 1969, 1975, 1978, and 1979 as determined from aerial photographs. Three reaches for the Maple River were defined and annual average damage estimated for each reach (Table 2.2). The CACFDAS was used to compute crop damage per acre per reach for the Maple Watershed. Average annual acres flooded were reported as being 3,440 acres. Table 2.2. Flood damage estimates in the Maple Watershed Source Location Flood Estimate Note Year USACE, Maple 1989 $15/acre crop damage Total number of acres flooded not listed 1993,1996 ND 1993 $3/acre other agricultural USACE, 1990 Maple 1969, Average annual total damage Reach 1 = $31.35/acre crop damage ND 1975, $334,600 - $355,300 Reach 2 = $32.21/acre crop damage 1978, Reach 3 = $26.10/acre crop damage Wild Rice Watershed flood damage assessment The study, 1993 Red River Valley Final Flood Damage Assessment reported that in 1993 total flood damage in the Wild Rice Watershed was $9.4 million (HDR 16

28 Engineering, Inc. and Houston Engineering, Inc., 1998). The method used in this study was previously described in the Red River Basin section (p. 12). The Corps of Engineers estimated agricultural damage using the CACFDAS economic model for the Wild Rice Watershed for the period 1977 to The CACFDAS analysis is based on crop production cost and foregone net revenue associated with crop inundation. One problem with the CACFDAS study is that stream reach data and hence flood damage estimates were missing for various years. The years without data were either not analyzed, or represented years with no apparent flood damage. Also, damage estimates were only reported for Norman County and not the entire Wild Rice Watershed (Norman County represents approximately 40 percent of cropland in the Wild Rice Watershed). The estimated average crop damage from 1977 to 1995 in the Wild Rice Watershed was $15 per acre flooded. Other agricultural damage associated with flooding was estimated as being 20 percent of crop damage ($3/acre). In a separate study, Flood Damage Reduction Study Wild Rice and Marsh Rivers, Norman County, MN, (Watts & Associates, Inc., 1997) flood damage by stream reach was reported. A dollar amount for average annual flood damage, based on a 67-year period, for the city of Ada was estimated to be $10,500. The years 1997 and 1998 were not included in the study. An attempt was made to link urban damage reduction estimates for Ada based upon historical discharges. The Corps of Engineers estimated average annual total flood damage from 1909 to 1995 for the Wild Rice Watershed as being $2.3 million (1995 Dollars). Average annual crop damage for the Wild Rice Watershed was estimated to be $541,800 ($62/acre). This figure is much higher than the Corps of Engineers estimate of 17

29 $15/acre (1996 Dollars) mentioned previously. The discrepancy is probably due to the different number of years being analyzed in the studies. The Red River of the North Reconnaissance Report: Wild Rice - Marsh Rivers Subbasin (USACE, 1980) estimated a dollar amount for crop, other agricultural, and transportation flood damage for Average annual crop damage was estimated to be $1.3 million for the Wild Rice Watershed while average annual other agricultural damage was estimated as being $383,100 (26 percent of crop damage). The methods used to arrive at the flood damage estimates were not discussed in detail. Crop damage was estimated using the CACFDAS. Other agricultural damage was estimated through field surveys of farmsteads. An updated 1988 version provided similar flood damage estimates (Table 2.3). Table 2.3. Flood damage estimates in the Wild Rice Watershed Source Location Flood Year Estimate Note HDR Engineering & WRW of 1993 $9.5 million $6 million of damage Houston Engineering, 1998 MN (damage paid) (agricultural) USACE, 1975 WRW of 1980 $2.9 million Annual agricultural MN average annual damage damage = $2 million 2030 $5.8 million Annual agricultural average annual damage damage = $4.1 million USACE, 1988 WRW of N/A Average annual crop Other agricultural MN damage = $1.3 million damage is 26% of crop Average annual other damage agricultural damage = $338,000 USACE, 1996 WRW of 1996 $15/acre crop damage Total number of acres MN $3/acre other agricultural flooded not listed Watts & Associates, WRW of $62/acre crop damage Average annual crop Inc., 1997 MN (67 years) $2.3 million annual damage = $541,800 watershed damage (USACE, 1995) CACFDAS results 1993 $5.5 million crop damage 18

30 The document called Flood Control Twin Valley Lake Wild Rice River, Minnesota Design Memorandum No. 2, Phase I General Plan Formulation (USACE, 1975) reports an estimate of average annual acres flooded in the Wild Rice Watershed (10,557 acres), along with estimated average annual flood damage for 1980 and The method for estimating future flood damage consisted of a population increase/damage increase correlation and not on flood duration and timing per se. The data for future estimates are highly speculative, especially because the estimates were made in Total flood damage (1998 Dollars) in 1980 was estimated to be $2.9 million ($2 million agricultural damage). Total flood damage (1998 Dollars) predicted for 2030 was estimated to be $6 million ($4 million agricultural damage). 2.4 Agencies Disbursing Flood Payments The following section describes the various public and private agencies that provide a major role in disbursing flood damage payments in the RRB. The intent is to provide the reader with some background information in order to attain a better understanding of the procedures/methods used to quantify historical flood damage payments in Chapter Small Business Administration (SBA) The purpose of the SBA s Disaster Loan Program is to offer financial assistance to those who are trying to rebuild their homes and businesses after a disaster. Disaster loans at below market interest rates can be for homes, to repair physical damage to businesses, and economic injuries. The SBA reports flood damage at the county-level. More 19

31 information about the SBA can be found at the web site address aboutsba/ Federal Emergency Management Agency (FEMA) The Federal Emergency Management Agency s (FEMA s) mission is to reduce the loss of life and property and protect our nation s critical infrastructure from all types of hazards through a comprehensive, risk-based, emergency management program of preparedness, mitigation, response, and recovery ( There are several programs under FEMA that provide disaster assistance including: National Flood Insurance Program, Flood Mitigation Assistance Program, Hazard Mitigation Grant Program, Infrastructure Support Program, Human Services Program, Mission Assignments Program, Flood Risk Reduction Program, Disaster Unemployment Assistance Program, and the Crisis Counseling Program. The programs providing data for this study are described below. The National Flood Insurance Program (NFIP) makes federally-backed flood insurance available in communities that agree to adopt and enforce floodplain management ordinances as a means to reduce future flood damage. The NFIP generates money through the premiums from flood insurance policies ( The NFIP reports flood damage at the county, township, and city-levels. The Hazard Mitigation Grant Program (HMGP) provides grants to states and local governments to implement long-term hazard mitigation measures after a major disaster declaration ( The HMGP reports flood damage at the county, water resource district, and city-levels. 20

32 The Public Assistance Program (Infrastructure Support Program) provides public assistance, which it breaks down into several categories. Damage Survey Reports (DSRs) are used to classify flood damage claims. A DSR is categorized as either of the following: Debris Clearance: clean up of public roads and streets, other public property, private property (when undertaken by local government forces), and structure demolition Protective Measures: measures taken to protect life and safety, property, health, and stream/drainage channels. Road System: damage to roads, bridges, streets, culverts, traffic controls and related miscellaneous damage. Water Control Facilities: damage and construction of dikes, drainage channels, levees, dams, irrigation works, and related miscellaneous damage. Buildings and Equipment: damage to public buildings and equipment, supplies and inventory, vehicles or other equipment, transportation systems, and related miscellaneous damage. Public Utility Systems: damage under the classification of water, storm drainage, sanitary sewage, light/power, and related miscellaneous damage. Other: damage to park facilities, recreational facilities, and related miscellaneous damage. Source: Requirements for Federal Damage Surveys, FEMA Form 90-49, May 94 The Human Services Program provides individual assistance in the form of Individual Family Grants (IFGs) and Disaster Housing Assistance (DH). Individual and 21

33 Family Grants provide funds for the necessary expenses and serious needs of disaster victims that cannot be met through insurance or other forms of disaster assistance. IFG assistance is allocated for real property damage, personal property damage, medical, dental, funeral, and transportation costs ( Disaster Housing Assistance provides grants for emergency residence repairs or temporary housing to both homeowners and renters ( The Human Services Program reports flood damage at the county-level United States Department of Agriculture (USDA) There are several agencies under the Department of Agriculture that provide disaster assistance to farmers and the rural community including the following: Consolidated Farm Service Agency (CFSA), Rural Development Administration (RDA), Natural Resources Conservation Service (NRCS), Farmers Home Administration (FmHA), Food and Nutrition Service (FNS), and the Federal Crop Insurance Corporation (FCIC) ( The agencies and programs providing county-level data for this study are listed below. The Consolidated Farm Service Agency (CFSA), and the individual Farm Service Agencies (FSAs) in particular, can provide emergency loans for farming, ranching, or aquaculture operations when economic injury or property loss occur as a result of a government declared natural disaster. The low-interest loans are disbursed through several different programs including the following: Emergency Conservation Program (ECP), Noninsured Crop Disaster Assistance Program (NAP), Emergency Loan Assistance Program (EM), Emergency Haying and Grazing Assistance Program, and the Crop Loss 22

34 Disaster Assistance Program ( Data from both the Emergency Conservation Program and the Emergency Loan Assistance Program were used in this study. The Rural Development Administration (RDA) provides assistance for flood damage in the form of Water and Waste Disposal Grants and Loans ( rurdev.usda.gov/nd/wtrwaste.htm). The RDA reports flood damage at both the county and city-levels. The Natural Resources Conservation Service (NRCS) has two flood-related programs, the Emergency Watershed Protection Program and the Emergency Wetland Reserve Program ( The purpose of the Emergency Watershed Protection Program (EWP) is to respond to emergencies created by natural disasters. EWP work includes removing debris from stream channels, roads, culverts, and bridges; reshaping and protecting eroded banks; reseeding damaged areas; correcting damaged drainage facilities; repairing levees and structures; and purchasing flood plain easements ( p1566/ewp/ewp.htm). The Emergency Wetland Reserve Program (EWRP) consists of purchasing land, a one-time perpetual easement, to avoid more expensive costs of restoring it or repairing structures such as levees ( ftw.nrcs.usda.gov/p1566/ewp/ewp.htm). The mission of the Food and Nutrition Service (FNS) is to ensure access to nutritious, healthful diets for all Americans ( fns/ menu/ about/ mission/state.htm). One flood-related program under the FNS, the Emergency Food Stamp Program, was initiated as a result of the 1997 flood. ( menu/response/faq.htm). 23

35 CHAPTER 3 STUDY PROCEDURES 3.1 Introduction Flood damage in the Maple and Wild Rice Watersheds was estimated by surveying and classifying flood damage payments made by agencies and organizations in North Dakota and Minnesota between 1989 to 1998, and by modifying and/or extrapolating flood damage over space and time. Non-agricultural flood damage was in most cases represented by reported damage payments from various federal, state, and local agencies. Often these data needed to be extrapolated from the county to the watershed level of analysis. Agricultural-based flood damage was estimated by extrapolating previously calculated damage based on a computerized agricultural crop flood damage assessment system (CACFDAS) and hydrologic data. Figure 3.1 illustrates the types of agricultural and non-agricultural flood damage data quantified by the study while Figure 3.2 illustrates the general procedures, incorporated in this study, to estimate agricultural flood damage. This is followed by a detailed discussion of the methods and procedures used to classify and/or extrapolate both non-agricultural and agricultural based flood damage data over space and time. 3.2 Non-agricultural Flood Damage Estimates The type and amount of flood damage payments and/or estimates of flood damage within the Maple and Wild Rice Watersheds between 1989 and 1998 were requested through telephone, , and traditional mail correspondence from over 30 different federal, state, and local agencies. As well, individual damage statement reports (DSR s) 24

36 TOTAL FLOOD DAMAGE ( ) Agricultural Non-agricultural Existing CACFDAS Data (Extrapolated Data) USDA Data (Insufficient Data) Residential Debris Clearance Commercial/Industrial Protective Measures Transportation Water Control Facilities Public Utility Systems Public Facilities Miscellaneous Infrastructure Figure 3.1. Classifications of flood damage. AGRICULTURAL FLOOD DAMAGE ( ) Extrapolate from Existing CACFDAS Data Identify Similar Flood Events (CACFDAS Flood Events Similar to Flood Events) Estimate the Number of Acres Flooded for Flood Events ( ) Estimate Crop Damage for Flood Events ( ) Estimate Other Agricultural Damage for Flood Events ( ) Figure 3.2. Procedures for estimating agricultural flood damage. 25

37 associated with flood damage payments made by the Public Assistance Program of the Federal Emergency Management Agency (FEMA) were manually reviewed in local FEMA offices in order to classify the type and location of flood damage payments Types of Non-agricultural flood damage Non-agricultural flood damage was categorized as: residential, commercial/industrial, debris clearance, protective measures, transportation, water control facilities, public utility systems, public facilities, infrastructure, and miscellaneous. A description, as incorporated by this study, of each category follows: Residential flood damage included damage to the contents and structure of residential buildings. Commercial/industrial flood damage included damage to commercial and industrial buildings and equipment, supplies and inventory, and vehicles. Debris clearance damage was associated with demolition activities, and floodrelated debris removal from public roads, streets and right of ways, and both public and private property (when undertaken by local government forces). Protective measure flood damage included damage related to the protection of life and safety, property, health, and stream/drainage channels. There were some cases when it was difficult to determine if flood damage payments should be listed under protective measures or water control facilities. It was assumed that temporary dikes should be placed under protective measures and permanent dikes under water control facilities. The flood damage payments pertaining to clearing/ dredging of stream/drainage channels were listed under protective 26