Red River basin precipitation

Introduction

Many investigators are studying possible changes in the frequency and severity of flooding along the Red River of the North. An oft-debated question is whether fluctuations in flooding behaviors are the result of changing land-use practices, climatic factors, or both. In order to facilitate such investigations, the following brief overview will offer a look at precipitation data gathered in the Red River basin over the past century. Graphed and tabular data are offered below as a resource for investigators.

Data Source and Methodology

The National Climatic Data Center (U.S. Department of Commerce) summarizes climate data over multi-county clusters called "climate divisions". These divisions are thought to have homogeneous climate characteristics. All available National Weather Service climate data within the divisions are pooled, and divisional averages are assembled. Data from roughly 10 to 20 locations are used in each division. This data set begins in 1895 and is updated monthly.

All or part of five climate divisions in North Dakota, three in Minnesota, and one in South Dakota, encompass the U. S. portion of the Red River of the North watershed. Most of the basin is found within three North Dakota climate divisions and two Minnesota climate divisions (Figure 1). Precipitation data from these five climate divisions are used to provide the figures and discussions that follow. The data from the five climate divisions are grouped by simple averaging with no efforts made to weight for size of division or number of observing locations within a division. For purposes of this discussion, precipitation is defined as the combination of rainfall and the water content of the falling snow.

Discussion

Four attached figures depict temporal patterns in the U.S. portion of the Red River basin. Figure 2 offers annual precipitation totals for the period 1895 through 1997. Figure 3 shows December through March precipitation totals for 1896 through 1998. Figure 4 presents the October through April precipitation totals for 1896 through 1997. And Figure 5 summarizes May through September totals for 1895 through 1997.

The annual precipitation totals in Figure 2 display a distinct drying pattern from the turn of the century though and including the "Dust Bowl" years of the 1930's. Note the lack of "wet" years over that period. (Although the issue of the day is flooding, it is crucial to note that this extended dry period is a documented component of the climate of the Red River basin. Any long-term planning efforts involving water-dependent activities must account for this scenario.) Beginning in roughly 1940, the precipitation trend takes a ladder-step up to a relatively flat plateau that has stayed in place for nearly 60 years. However, over these sixty years, a great deal of variability can be observed. 1976 was an extraordinarily dry year. Then, in a dramatic turn-about, 1977 is the wettest year on record. Note the lack of years below 20 inches in the early and mid-1980's, then the sequence of years below 20 inches in the late 1980's. The precipitation deficit in the late 1980's led to significant drought. Precipitation totals rise once again from 1991 through 1997. This period of wetness manifests itself today in high lake levels at Devils Lake and other lakes in the region.

Figure 3 shows December through March precipitation totals as a surrogate for potential spring snowmelt runoff. While spring flooding is correlated with winter precipitation, it should be noted that other climate factors; such as the timing and duration of snowmelt, antecedent soil moisture conditions, and spring rainfall are important factors not considered in this graphic. As would be expected in a continental climate, a large amount of interannual variability is apparent. Of particular interest is the relatively large value shown for the winter of 1988-1989. Flooding from spring snowmelt runoff can be a temporally isolated episode. Ironically, Red River flooding which occurred in 1989, did so in the midst of a prolonged and intense dry period as shown by the annual data. Another interesting feature of the December through March data is the lack of winter totals less than two inches since 1986-1987.

Figure 4 depicts October through April precipitation totals. This sequence of months was chosen in an effort to capture autumn soil moisture recharge, winter precipitation, and the April precipitation that often compounds snowmelt runoff flood events. Temporal variability is again quite obvious. Note that the 1996-1997 total is the largest value on record.

Figure 5 summarizes May through September precipitation totals, the bulk of the growing season. On average, two thirds of annual Red River basin precipitation falls during these five months. Interannual variability is again quite obvious. As was the case with annual precipitation patterns, a distinct drying is obvious for the first four decades of the 20th century. Beginning in the 1940's, growing season precipitation totals dramatically increase, and then level off. The 1990's have been notable for the lack of growing seasons with below average precipitation.

Data Resources