OCR Text |
Show erosion may conveniently be classified according to the size of the area from which the data were obtained. A large amount of data has been collected on the relative infiltration capacity values of different soils under various vegetal covers from infiltrometers, which are essen- tially devices for simulating rainfall on isolated plots or applying a constant head of water to samples of soil. The results, despite many tests, apply to only a portion of the many possible combinations of the principal variables in- volved. Infiltrometer data, nevertheless, do furnish in- dices of relative infiltration capacities, detention, and erosion characteristics of typical soils under certain fixed conditions and are useful in delimiting areas of soil types with respect to such characteristics. Lysimeters are essentially tanks of soil for measuring accurately the amounts of water transpired from plants and evaporated from the soil, and waters percolated through soils. Because of the high cost of lysimeters, only a few types of soil and vegetation have been investigated. Weighing lysimeters also provide information on the con- densation of moisture within the soil. The sprinkling devices used in the infiltrometer tests have also been used to investigate the hydraulics of flow of water across the land surface before channelization takes place (overland flow). These investigations have confirmed the theoretical overland flow equation and established detention coefficients for artificial surfaces, such as pavements and turf, for short lengths of flow. A change in type of flow has been indicated for longer lengths, and an investigation is in progress on paved strips which will provide such basic data. There are virtually no data to establish detention coefficients for turbulent flow on grassed areas. A large number of small plots with uniform soil, slope, and vegetal cover have been operated under natural rain- fall conditions. Unfortunately much of the resulting data have not been analyzed and published in such form as to be readily available. Usually, only the surface elements of runoff and sediment resulting from natural rainfall are measured. Plots afford a high degree of control and have been particularly valuable in erosion studies to establish relative soil losses. Plot data do not ordinarily agree quantitatively with runoff from larger watersheds, al- though relative agreement exists between soil cover com- binations and runoff. Data on effects of land management on rates and total amounts of runoff and erosion are available from numer- ous watersheds ranging in size from one acre to about 10,000 acres. Such data automatically integrate the effects of the many variables involved and become in- creasingly valuable with the length of record. Small watersheds, like plots, are not true indications of runoff and erosion rates on larger watersheds, but they may be used in explaining or interpreting the runoff and sediment production of larger areas. Variability in land management and flow pattern on the larger watersheds makes it difficult to evaluate the sepa- rate effects of different types of land management on total runoff and erosion. In cases where there has been a radi- cal change in land management over a large part of the watershed, good records of precipitation, stream flow, and sediment load before and after the change might have provided data on the apparent effects of the change in land management. See figures 40, 41 and 42. B. Major Deficiencies The effect of land management upon runoff and erosion from plots and small watersheds is well-established for a number of combinations of soil, cover, condition, and cli- mate. But little has been done to bridge the gap between these small areas and large watersheds. Special attention should now be given to the development of methods for the accurate estimation of the effects of land management for larger watersheds. Erosion has received considerable attention from the standpoint of the effect of different types and densities of cover on rates of soil loss, but relatively little attention has been paid to the erosion process itself. Insufficient knowledge is available on the natural laws involved in the detachment and transportation of soil particles, and facts are lacking on the effects of plant cover, litter and biological population in the soil, and on the stabilization of soil particles. Infiltration rates have been investigated with various types of infiltrometer equipment for a relatively large number of plant-soil complexes, but variations in rates under what appear to be similar conditions have been difficult to explain. Improved methods of determin- ing infiltration rates and better utilization of available data will require greater knowledge of the mechanism of the infiltration process, and thereby extend its appli- cation to plant soil complexes not yet investigated. There is a lack of sufficient information to determine surface storage characteristics of different kinds of soil and land use for various rates of rainfall and runoff. More data are needed as to the influence of land man- agement on subsurface storage and on the rate at which storage opportunity is affected by evapo-transpiration. Although considerable data have been accumulated on the rate of transpiration for different individual types of vegetation, much of them are not applicable in water balance computations for combinations of types of cover because they have been obtained largely under laboratory conditions. Data are needed on rates of evapo-trans- piration for various field conditions, particularly in relation to available supplies of water, and the interrelation of evaporation and transpiration under different site con- ditions. The mechanism of evapo-transpiration requires further study. Further studies are needed to determine (1) the dis- position of precipitation after it enters the soil, in order to arrive at the amount which goes to ground water reservoirs, and (2) the influence of watershed management practices in water yields. C. Recommendations 1. Continued study with greater emphasis should be made of the effects of land management practices upon watershed hydrology, with particular attention to the solution of such problems as bridging the gap between conclusions drawn with regard to small experimental plots and watersheds and their application to larger areas, such as river basins. 379 |