Section IV Case Studies in Integrated Drought and Water Management: The Role of Science, Technology, Management, and Policy

Floods Punctuated by Drought: Developing an Early Warning System for the Missouri River Basin in the Midst of Alternating Extremes

Chad McNutt, Doug Kluck, Dennis Todey,

Brian A. Fuchs, Mark D. Svoboda, and Courtney Black

The Missouri River is an enigma. Despite over a century of human attempts to understand and control it, the longest waterway in the United States remains perplexing and mysterious. The river eludes the efforts of modern technology to tame it fully and continues to provide water of uncertain quantity and quality. It either furnishes too much of it, as it did during the disastrous floods of 1844, 1881, 1943, 1951, and 1993, or too little of it, as was the case during the severe droughts of the 1860's, 1890's, 1920's, 1930's, late 1980's through early 1990's, and early 2000's.

Lawson 2009

Introduction

The Missouri River is the longest river in North America. Its drainage basin incorporates the states of Montana, North and South Dakota, Wyoming, Nebraska, Iowa, Colorado, Kansas, Minnesota, and Missouri, an area of more than 520,000 square miles. The Missouri River starts its course in western Montana and terminates over 2,300 miles downstream near St. Louis, Missouri. Between 1944 and 1964, the US Army Corps of Engineers (Corps) and the Bureau of Reclamation (BoR) built a series of dams and reservoirs under the Pick-Sloan program. The geographic area of the Missouri River Basin along with the major Corps dams is shown in Figure 15.1. The projects were developed to account for the Missouri Basin's tendency to deliver too much water in some years and not enough in others. The climate along the Missouri Basin can also vary greatly between the upper and lower parts of the basin. For example, the climate in the Upper Basin (North Dakota, South Dakota, Wyoming, and Montana) is semiarid with annual precipitation of 254 mm-508 mm (10 in-20 in) and widely varying temperatures due to exposure to Arctic and Pacific air masses. The Lower Basin (Nebraska, Colorado, Kansas, Iowa, and Missouri), however, is more consistent with a humid-continental climate (with the exception of Colorado) and can receive upwards of 1,000 mm (~40 in) of precipitation annually. The historical feature that has so often defined the Missouri Basin is the extreme year-to-year variation in runoff and sensitivity to precipitation.

FIGURE 15.1

Geographic area of the Missouri Basin, with states, cities, and the six major Corps dams on the Missouri mainstem labeled.

A good example of this variation is the record flood in the basin in 2011 and the subsequent severe drought in 2012. While the two extremes coming in quick succession were not necessarily unique for the Missouri Basin, the fact that the record drought emerged so quickly from record flooding the previous year and with such magnitude when many were forecasting another flood makes this particularly salient for understanding and improving how we provide early warning for drought in a basin as variable as the Missouri. Figure 15.2 shows the historical annual runoff for the Upper Basin of the Missouri and helps put into context just how anomalous 2011 and 2012 were. Following these events in 2014, the National integrated drought information system (NIDIS), an interagency program lead by the National Oceanic and Atmospheric Administration (NOAA), began developing a drought early warning information system (DEWS) for the Missouri Basin. The Missouri DEWS is just one of several DEWS that NIDIS has been developing since late 2006, when the US Congress created NIDIS and charged it with developing a series of regional drought early warning systems across the United States.

The objective of this case study is to highlight the challenges of developing a drought early warning system in a basin that exhibits such extreme annual variability in runoff and sensitivity to precipitation variability. This includes how improvements in monitoring and prediction can be leveraged to address both needs to understand slow-onset disasters like drought and rapidly developing extremes like large runoff events. This topic is particularly timely given new evidence that the Missouri Basin is becoming even more variable in terms of runoff. Livneh et al. (2016) assessed meteorological trends

FIGURE 15.2

Missouri Basin (mainstem) annual runoff in million acre-feet above Sioux City, Iowa. Light shaded bars indicate normal or above-normal years while dark shaded years show below-normal runoff years. Double asterisk (**) indicates runoff for 2011 and asterisk (*) shows runoff for 2012.

favorable for high runoff events in the Missouri Basin, finding that annual runoff variability has nearly doubled in the last 20 years. The variability has mostly come from high runoff events and in particular from above-normal precipitation in the Upper Missouri Basin (above Gavins Point Dam) during the October-March timeframe. For example, the authors note that nine of the ten largest runoff events have occurred since 1975 (Livneh et al. 2016). While the basin may be tilting toward more frequent large runoff years, drought will still punctuate these events. The 2012 drought, while not prolonged, was an extremely intense event, and a good example of a climate surprise (Hoerling et al. 2013) that had large impacts across the basin. A chilling prospect is that a more prolonged drought similar to what occurred in the 1930s and 1950s and more recently from 2000 to 2007 in the basin is still a possibility, and it could come while we are preparing for a flood.