No abstract

As a part of numerous hydrologic studies in this country and elsewhere, intensive experiments have been undertaken to determine the influence of timber‐cutting or other changes in plant‐cover or land‐use upon water‐yields, flood‐flows, erosion, and related factors. In these, studies ordinarily two or more watersheds are subjected to control by means of stream‐gaging stations rain‐gages, snow‐courses, and similar equipment. Records of stream‐flow and precipitation are obtained from each watershed over a period of years, and then he desired treatments are applied to some of the areas, while one or more is kept untreated its a check. Records taken after treatment for another period of years are analyzed to determine the influence of the treatment or treatments, upon water‐yield and other factors.

A statistical method for estimating the necessary duration of a watershed experiment is presented that applies especially to the type of study in which two or more watersheds are placed under control by means of stream‐gaging stations and other equipment. After a suitable period of calibration, all but one may be subjected to some form of treatment such as a change in land use or the removal of forest cover. With minor modifications the method may be applied to other kinds of experiment in which regression analyses are used to obtain statistical control of the variables under scrutiny. Equations are derived with which the investigator can estimate the number of observations which is likely to be required before and after treatment, in order to demonstrate that a difference or treatment effect of any chosen order of magnitude is larger than might have been expected by chance.

As a part of research on the management of forested watersheds, the Rocky Mountain Forest and Range Experiment Station has conducted intensive experiments in Colorado to find out how timber cutting in a lodgepole‐pine forest influences the amount of water available for stream‐flow [see 2 of “References” at end of paper]. Among the various factors involved in this influence, the storage and melting of snow obviously assumes a dominant role. Accordingly, an intensive series of measurements has been devoted to this factor; and now, with completion of the first phase of the investigations, definite information is available on how timber cutting in this forest‐type influences the storage and melting of snow.

No abstract