The lysimeter and weather stations are operated continuously through the accretion 

 (winter) and ablation (spring) periods. This paper will deal only with the winter 

 period; November-March, inclusive, on the north aspect and November-January , inclusive, 

 on the south aspect. In 1970, the spring season on the south aspect began in February. 



Data analysis . — Computation of the total volume of outflow for the lysimeter 

 represents a straightforward summation of instantaneous hourly flow rates converted to 

 area-inches. 



A more complex analysis involves the partitioning of outflow into its components of 

 origin--rain percolate (in the small opening) , through fall -drip percolate (in the forest) 

 and snowmelt percolate from the snowpack proper. This analysis is based primarily upon 

 a number of generalized assumptions dealing with the disposition of rainfall, through- 

 fall, and drip that occur during and after storms. The entire scheme is undergoing 

 refinement; the approach presented in this paper is an earlier version and has certain 

 limitations that will be discussed subsequently. 



The partitioning of outflow in the small opening depends upon the amount of observed 

 rainfall. If outflow is equal to or less than observed rainfall, it is defined as rain 

 percolate. Excess outflow, or that which exceeds the observed rainfall, is called snow- 

 melt percolate. In a sheltered opening, such as found on the north aspect, the gage 

 rainfall catch represents a realistic estimate of the amount of rain that actually 

 infiltrates the snowpack. In a windblown opening, on the south aspect, the gage rainfall 

 catch measured less than that which actually reaches the snow surface. Thus, the snow- 

 melt percolate portion of the outflow is very likely an overestimation. 



The separation of outflow in the forest hinges first on two basic assumptions : 

 (a) That the amount of rainfall observed in the opening is the same amount as the rain- 

 fall incident on the canopy; and (b) that a small portion of rainfall (about 20 percent) 

 is intercepted by the canopy and lost for the duration of the storm. The larger portion 

 of rainfall that is not intercepted is defined as throughfall. 



Throughfall-drip percolate includes the following: (a) Throughfall from a canopy 

 bare of snow which in turn percolates through the snowpack into the lysimeter; (b) drip 

 from melting snow held in the canopy during a storm-free period which in turn percolates 

 through the snowpack; or (c) combined throughfall and drip during rain on canopy-held 

 snow. Drip in this experiment refers to drops of melt water falling from canopy-held 

 snow; it does not include small releases of snow. Throughfall-drip percolate may in- 

 clude some melt from the snowpack during hours when the percolate is passing through the 

 snowpack. During warm, storm-free periods and in the absence of snow in the canopy, all 

 outflow in the forest originates as snowmelt percolate. 



Recession flow, which drains slowly from the snowpack, is assigned to the previous 

 outflow event on a prorated basis either as rain, throughfall-drip, or snowmelt percolate 

 Recession flow near the end of depletion may be a few thousandths of an inch per day; 

 therefore, it may require days to accumulate to 0.01 inch in the catchment tank. 



An example of the daily separation of water released from the snowpack at stations 

 42 and 45 for February 6-12, 1970, is presented in table 2. 



6 



