OVERFLOW OF RED RIVER OF THE NORTH. 81 
safe limit. In any case, when the flow at Fargo would amount to 
9,652 second-feet without control, enough could be diverted through 
the floodway to reduce it below 7,500 second-feet without overtaxing 
the diversion channel. With the stage in the Red high enough to 
cause a flow of 7,500 second-feet at Fargo, the stage in the Wild Rice 
at the entrance to the floodway would be high enough, either from the 
flow of the Wild Rice itself or by backwater from the Red, to allow 
the withdrawal of the 2,250 second- feet at an elevation considerably 
above the bed of the Wild Rice. 
The probable maximum spring flood flow at Fargo, amounting to 
11,580 second-feet, is next considered. To produce such a flow would 
require that all the tributaries above this point should reach very 
nearly the maximum spring flow as noted in Table 5, and that each 
should crest at such particular time as would produce the maximum 
combined effect at Fargo. Although it is not believed that such a 
condition will ever exist in the case of summer floods, it is possible 
that it will occur in the spring. In this event the flow in the Wild 
Rice River would be 4,420 second-feet. By diverting 2,250 second- 
feet through the flood way, the flow of the Red at Fargo would be 
reduced by practically that amount. Assuming that the effect of the 
2,670 second- feet flowing from the Lake Traverse watershed is re- 
duced to 30 per cent before reaching Fargo, which under conditions 
of flood flow is considered liberal, it is seen that by closing the sluice- 
way at the Lake Traverse detention reservoir and completely shut- 
ting off this flow during the spring months the flow at Fargo from 
this source would be reduced by 1,870 second-feet. This, combined 
with a reduction of 2,250 second-feet caused by diversion of this 
amount from the Wild Rice River, would give a total decrease of 
flow at Fargo of 4,120 second-feet, which would bring the flow at 
this point well within 7,500 second-feet. With such control of the 
flow in operation and the recommended channel improvements 
described hereafter accomplished, the elevation of the flow line dur- 
ing such floods as that which occurred in 1897 would be lowered 
about 15 feet. 
The surveys and investigation make it apparent that the overflow, 
during flood periods, of the land along the west side of the Red River 
where the Wild Rice and Sheyenne Rivers flow almost parallel with 
it, are not caused entirely by overflow from the Red. Both of these 
tributaries have channels of insufficient capacity to carry their flood 
flow and are partly responsible for the floods in this region. Conse- 
quently it will be necessary, in addition to providing a diversion 
channel for the Wild Rice, to improve the channels of the Wild Rice 
and Sheyenne Rivers above the points where the floodway joins 
them. While this probably can be more readily accomplished under 
70352°— 22 6 
