Managing Surface Runoff

by D. B. KRIMGOLD

RAIN OR SNOW WATER that runs off the surface of farm land instead of sinking into the soil may serve a useful purpose if it is properly managed, but it may cause great damage if it is not controlled.  Many farmers and ranchers in the United States have no springs, no wells, and no streams on their land, and must depend on impounded runoff water for livestock and other purposes at all times. On the other hand, water flowing over unprotected land loosens and carries away topsoil; further damage results when this soil is deposited in stream channels and reservoirs and on valley floors and when fertile bottom land is flooded.

Our studies have shown that more of the rainfall runs off where soils are tight, shallow, and wet, vegetation is poor and thin, and rainfall is heavy and intense; and that the larger the amount of runoff water and the faster it moves, the greater is its cutting power and the greater is the load of sediment and debris it can carry. The speed with which surface water runs off a small agricultural area depends on its amount, the slope along which it flows, whether its course is straight, the size and shape of the drainage area, the number of watercourses, and whether its path is obstructed.

We have learned to reduce the amount of surface runoff by rotating crops and protecting the soil with trees and grasses and with crop residues, and by practicing contour cultivation, strip cropping, terracing, pasture furrowing, and basin listing. To limit the speed of flow and protect the soil against erosion, we use drainage terraces, diversion ditches, spillways, gulley plugs, culverts, and channels lined with vegetation and with masonry. The size and cost of such structures, the space they require, and the extent to which they interfere with farming operations vary according to the quantities of water expected.

To control and make use of surface runoff from agricultural areas we must find out how much water flows off, in terms of gallons, acre-feet, or cubic feet, and how fast it flows, in terms of gallons per minute or cubic feet per second. A great part of this task has been to develop suitable instruments and procedures. Through laboratory experiments, weirs and flumes have now been developed with which we can accurately measure flows ranging from less than one-half gallon a minute to 800,000 gallons a minute. These devices can be used for measuring all flows except those carrying exceptionally heavy loads of debris.

Rates of runoff and total amount of water running off an area within a given time are calculated from the depth of water flowing over a weir or through a flume. Rate of surface runoff from a small area changes rapidly and irregularly. Therefore, to determine amounts of surface runoff from such an area we must have a record of the depth of flow for each 10-, 5-, 2-, or even 1-minute interval. A new type of water level recorder has been devised that gives a continuous record of the depth of flow to the nearest 0.01 foot for every minute. In this recorder a chart is mounted on a cylinder, which is rotated by a fast moving clockwork. A pen moved by a float rests on the chart with the result that a continuous record of depth of flow appears on the chart.

Since the development of adequate devices for measuring runoff from small areas, studies of runoff have been undertaken in some of the major agricultural areas of the United States, on small drainage basins of various sizes that are typical as to soils, vegetation, and other factors.  Altogether, more than 100 experimental watersheds have been used.

The runoff from small areas is extremely variable. A question therefore arises as to what sorts of flow should be provided for in designing control structures. When we build terraces, diversion ditches, and other structures for control of runoff, should we make them large enough to carry the greatest flows expected at any time?—or should we make them only large enough to carry flows that, on an average, are expected once in 10, in 15, in 25, or in 50 years? On a small agricultural area, the farmer must ask himself which would in the long run be more economical—a smaller structure that might overtop or even fail at long intervals, or a larger one that might cost a good deal more, might occupy more land, and might interfere more with farming operations. Soil conservation structures on the farm are usually made large enough to carry flows expected once in 10 or 25 years. Occasionally, one is made large enough for the flow expected once in 50 years.

Early results of the runoff studies already mentioned showed that rainfall of a certain intensity on a certain area does not always result in the same rate of runoff; that, for instance, the most intense rainfall in 15 years on a certain area does not necessarily produce the heaviest runoff in 15 years from that area. Accordingly, in designing measures for control of runoff we use runoff records rather than rainfall records.  Any reliable estimate as to how heavy a runoff should be expected from a certain area once in 10 or 25 years, for example, must be based on records of runoff over a long period. The longer the record, the more reliably can such things be estimated.

Where runoff water is used for livestock and other farm purposes, it is usually stored in ponds or reservoirs. We estimate that in 1943 there were more than a million farm and ranch ponds in the United States.  A great deal of work should be done to improve those existing ponds, and many more are needed. Reports of the postwar planning boards of 12 States called for a total of more than 250,000 new ponds and reservoirs. The need for farm ponds and reservoirs is by no means limited to the arid and semiarid parts of the country; the postwar planning board of Georgia, for example, estimates that 34,000 are needed in that State.

In planning a farm pond, we need to assure ourselves that the water supply will be sufficient for the intended purpose after evaporation and seepage take their toll. We cannot do much to control evaporation, and very seldom can we entirely eliminate seepage. To estimate properly how much usable water we can expect to obtain from the pond, we must set up a sort of bookkeeping, based on records of past rainfall, runoff, and evaporation, and on characteristics of the proposed pond site. Our ledger must show, for any given period, how much water would flow into the proposed pond and how much rain and snow would fall on it, how much water would go over the spillway, how much would evaporate from the pond surface, and how much would be lost by seepage through the bottom and through the dam or dyke.  If the balance is too small for the intended purpose, and we cannot make up the deficit by building a higher dam, we shall need to increase either the drainage area or the quantity of runoff. One way to increase the drainage area is to locate the pond farther downstream; another is to divert water from an adjoining drainage basin by means of terraces or ditches. The amount of runoff can sometimes be increased by putting a larger portion of the drainage area in cultivated crops.  Care must be taken to protect the pond from silting. This is done by following conservation practices on the cultivated land within the drainage area and by providing a protective strip of grass around the pond.

The bookkeeping suggested cannot safely be done on the basis of average conditions. Rainfall, runoff, and evaporation at any place vary greatly from time to time; and when rainfall and runoff are least, evaporation and the quantity of water needed for farm purposes may be greatest. Obviously, the balance may be favorable in 1 year and unfavorable in another. How often the balance must meet farm needs if the pond is to be worth what it cost depends on how badly the water is needed, whether the pond will be the only source of water on the farm, the losses that would result from lack of water, and the cost of providing water from other sources. With these practical considerations in mind, the farmer can decide whether the proposed pond must provide this balance in 4 out of 5, 9 out of 10, 14 out of 15, or 24 out of 25 years.

Records of runoff and of evaporation are still meager and far too short.  However, with the understanding of surface runoff and of related factors gained through research, these limited records are being put to good use.

Technical reports giving information needed in planning farm ponds in the claypan prairies of Missouri, Oklahoma, Iowa, Kansas, Illinois, and Indiana have been issued by the Soil Conservation Service in coopertion with State agricultural experiment stations.

Because the information we now have to guide us in controlling and utilizing runoff water is so limited, we have to play safe and use structures and practices that often turn out to be more expensive than necessary. However, if adequate records of surface runoff and information on evaporation and seepage are obtained, in all parts of the country, we can look forward to managing runoff better and more cheaply as time goes on.

THE AUTHOR
D. B. Krimgold, a soil conservationist, joined the Soil Conservation Service shortly after its establishment. He helped set up the Hydrologic Division and with C. E. Ramser wrote the working plan for the experimental watersheds. He directed the selection of the watersheds in Ohio, Texas, and Nebraska, and the technical phases of the runoff studies in some 22 States.

FOR FURTHER READING
Harrold, L. L., and Krimgold, D. B.: Devices for Measuring Rates and Amounts of Runoff Employed in Soil Conservation Research, Soil Conservation Service Technical Bulletin 51, 1944.
Krimgold, D. B.: Runoff from Small Drainage Basins, Agricultural Engineering, volume 19, pages 439-446, October 1938.
Krimgold, D. B.: What is There to Know About Farm Ponds, Agricultural Engineering, volume 26, pages 283284, July 1945.
Krimgold, D. B., and Minshall, H. E.: Hydrologic Design of Farm Ponds and Rates of Runoff for Design of Conservation Structures in the Claypan Prairies, Soil Conservation Service Technical Bulletin 56, 1945.
Ramser, C. E.: Runoff from Small Agricultural Areas, Journal of Agricultural Research, volume 34, pages 797-823, September 1927.


HERE ARE RAINDROPS driving into unprotected soil at 14 miles an hour. On the rebound the splashes may carry as much as 40 percent soil, making a muddy mixture that more rain can and usually does wash away.  To determine the destructive effect of rain hitting unprotected soil, this relatively simple machine was built (below). It can imitate rain of high or low intensity, large or small drops, and rain at controlled velocity.




The soil sample at left, with coins to simulate protected surface, was photographed after 45 seconds of rainfall. The white splash board, 6 inches away, is already flecked with mud. The other sample was snapped 1¼ hours later and shows what rain splash will do to open soil.


Here, in an open field, is proof that beating raindrops, as well as surface flow, will gradually carve away the good earth. Each pedestal of soil is shaped to the rock or crust on top.


Only 1 inch of driving rain on erodible soil (below) can splash and move as much as 170 tons an acre, mostly downhill. Soon—too soon—the light-colored, unproductive soil shows up.


But not all this splashed soil moves downhill.  The finest and richest part can even be floated out of contour furrows, leaving only the coarser, less fertile soil behind. Good cover crops will help prevent destructive splash erosion.


The farmer who owned this land (below) was surprised to find a rail fence under about 4 feet of soil at the base of a long field slope. Good cover crops would have prevented this; contouring also would have helped.



Other damage frequently caused by raindrop splash is that it seals the surface of the land so that water goes off—not into—the soil.   This puddled soil is practically waterproof.


W. D. Ellison of the Soil Conservation Service shows samples, left to right, of the splash collected from a bare field, a field with some cover, and one that had good protective cover.


A field of young corn (left) is a wide-open target to splash erosion. Even when corn has grown to more than knee height (right) many raindrops can get through to bombard the soil.

From an eye-high level the young oats crop (below) would look like a tight cover—but looking straight down at it we can readily see how the driving rains might raise havoc.



Beating rain has little chance of getting through good vegetative cover like this combination of clover and timothy. The splash process is prevented, the soil remains stable, and the water soaks into it. Of course, all land cannot be kept covered all the time. But by using what we know about protecting the land the rate of soil loss can definitely be retarded.


THE CONTROL OF WATER is vital in the care of the land. On one Iowa farm (above, left) heavy rains washed tons of soil across a cornfield. On another Iowa farm (above, right) contour furrows prevent downgrade washing and promote good conservation farming. Below, left, erosion is ruining a pasture; right, vegetation-lined waterways help protect fields.



Sedimentation between eroding farm fields and the oceans is costing us dearly. An example is deep gully erosion in Wisconsin (above, left) from which a vast amount of sterile sand was washed over nearby bottomland cornfields (right). Silting of reservoirs, like this one in South Carolina (below, left), is an outstanding example of sedimentation. It can be stopped in various ways: By gully-control structures (right), dams, and revetments to halt erosion.