systems for the production of corn (grain and 

 silage), wheat, alfalfa, and bromegrass. Three 

 sets of trials are analyzed in this report. One, 

 called the "cropland study," compared pro- 

 duction from a wheat and corn rotation on 

 benches with and without contributing areas, 

 and on natural slopes. The land sloped down- 

 ward toward the east, with a gradient ranging 

 from 1 to 5 percent. The soils were Bainville 

 loam and silt loam, Grail silt loam, Grail- 

 Moline silty clay loam, and Regent silty clay 

 loam. There were four benches — two 50 feet 

 wide and two 70 feet wide. Soil water determi- 

 nations at seeding and harvesttimes were 

 made at 1-foot increments to a depth of 6 feet. 

 Data for this experiment are available for 

 4 years, 1962-65 (9). 



A second experiment, called the "grass- 

 land study," compared alfalfa and bromegrass 

 hay production from 50-foot-wide benches with 

 and without contributing areas, and from na- 

 tural slopes. The land sloped downward toward 

 the west, with a 5-percent gradient. The soil 

 was classified as Grail silty clay loam. Soil 

 water determinations were made at 1-foot 

 increments to a depth of 6 feet in early spring, 

 at about monthly intervals during the growing 

 season, and in late fall about the time the soil 



normally freezes. Data are available for 4 

 years, 1963-66(9). 



In both studies, the contributing areas 

 were equal in size to their respective benches. 



A third set of experiments, called the 

 "contributing area study," investigated the 

 effectiveness of contributing areas of various 

 sizes in supplying runoff water to level benches. 

 The land sloped downward towards the north, 

 with a gradient ranging from 5 to 10 percent. 

 The soil was classified as Morton silt loam 

 and Regent silty clay loam. All the benches 

 were 30 feet wide and were used for producing 

 alfalfa hay. One group of benches collected 

 water from a bromegrass-covered contributing 

 area of 10.4 acres, with possible contributing- 

 area-to-bench ratios ranging from 19:1 to 

 85:1. A second group of benches had narrow 

 bromegrass-covered contributing areas with 

 ratios of 3:1, 2:1, 1:1, and 0:1 (no contributing 

 area). Soil water readings were taken at 1-foot 

 increments to a depth of 8 feet just before and 

 just after snowmelt, after each harvest, and 

 in late fall about the time the soil normally 

 freezes. Data on runoff, soil water, and al- 

 falfa production are available for 5 years, 

 1964-68. 



EXPERIMENTAL RESULTS 



The three sets of experiments demon- 

 strated the bench systems' ability to trap snow 

 and snowmelt water for the benefit of growing 

 crops. In the spring, the benches in the crop- 

 land, grassland, and contributing area studies 

 averaged about 3, 1, and 7 inches more water 

 than the unterraced slopes. For snow collec- 

 tion, the contributing area study was located 

 the most favorably and had the best growth of 

 grass on the dikes. Benches with a contributing 

 area generally contained slightly more water 

 in the spring than others. 



Runoff from the bromegrass-covered 

 watershed in the contributing area study was 

 measured during snowmelt and during each 

 rain. Total runoff from snowmelt and rainfall 

 was as follows: 





Runoff inches 



Year 



Snowmelt 

 1 0.020 



Rainfall 



1964 



0.071 



1965 



1 .050 



.168 



1966 



.147 



.038 



1967 



.462 



.001 



1968 



.023 



.002 



1 Estimated values. 



Snow Control 



The above data show that water collection 

 from runoff alone would not be important in 

 most years if the contributing area was cov- 

 ered with grass. Observation of the benches in 

 the wintertime indicates that two of their chief 



