30 
BULLETIN 1017, U. S. DEPARTMENT OF AGRICULTURE. 
A study of the annual and monthly records indicates that no com- 
bination of excessive monthly precipitation is likely to occur which 
will produce a maximum annual precipitation of more than 30 inches 
over that part of the Red River watershed south of the international 
boundary. It appears also that the maximum annual precipitation 
for the extreme western portions of the watershed will probably not 
exceed 25 inches, and that the maximum for areas of 2,000 square 
miles or greater in the eastern sections of the watershed will probably 
not exceed 37 inches. The maximum annual precipitation on the 
watershed above Fargo and on the Red Lake River watershed above 
Crookston will probably not exceed 36 inches. From the maximum 
of 37 inches south of Lake Traverse and 36 inches south of Red Lake 
intensities diminish toward the north by about 2 inches per degree of 
latitude, and toward the west by about 3 inches per degree of longi- 
tude. These maximum conditions are likely to occur but once in a 
long period, perhaps 50 to 75 years. 
The records further show that maximum annual and monthly pre- 
cipitation on the smaller watersheds have not occurred simultaneously 
on the different sub-areas. It is improbable, therefore, that the esti- 
mated maximum precipitation will occur on all the sub-areas in any 
one year. 
STORM PRECIPITATION. 
Records were compiled of all storms within the Red River water- 
shed with precipitation of 2 inches or more in 24 hours. For each 
storm the depths of precipitation for one day, two days, three days, 
and for the duration of the storm were determined for all observa- 
tion stations within or near the watershed. The records cover the 
26-year period 1894 to 1919, and include 130 storms of from two to 
eight days' duration (see Table 8). By plotting isohyetals on a map 
of the Red River watershed the intensity and extent of the storms 
were made evident. The areas covered by storms during the years 
1894 to 1904 are somewhat indefinite, as the number of records is 
small. For the period 1905 to 1919 many more records are available 
and the areas covered can be more accurately determined. During 
many of the storms excessive precipitation occurred in but few locali- 
ties and covered but small areas. The relative importance of the 
storms was determined from the isphyetal maps. 
Table 8. : — Storm periods during iclxicli precipitation of 2 incites or more occurred 
on Red River watershed in tJie United States. 
Year. 
Date. 
Length 
ofstorm 
period. 
Year. 
Date. 
Length 
ofstorm 
period. 
1894 
Apr. 15-20 
*** 1 
3 
4 
5 ! 
1896 
Mar.4-7; 
Days. 
4 
June 26-28 
Am-. 8-13 
6 
1895 
Apr. 25-30 
1 Mav 11-17 
6 
June 6-10 
7 
July 4-8 
June 4-8 
5 
