If we assume a climate like that of the present, a period of 500 to 1,000 years may well have been required to fill the lake to its present level. Much of the measured seepage may leave the lake 500 feet or more below the present water level. If so, the rise of the lake above that level must have been slow, but it is not possible to approximate the probable levels of seepage loss nor deduce the time of filling to the present level. The lake may have been generally at its present level, in balance with the present climate, for a similar length of time. Nelson (1961, p. 130) concluded that deposits of colloidal clay derived from volcanic ash, reworked in places by submarine landsliding, have helped to seal the bed of the lake. Such sealing may still be in progress. But continued sealing by any means during a period of stable climate should result in a slowly rising lake level and submergence of mature forest trees, of which we have as yet no confirmed evidence.
There is, however, a strong probability that the climate was not like that of the present time for about 3,000 years after the formation of the caldera. Students of post-Pleistocene climate are generally agreed that the middle third of the last 10,000 years-7,000 to 4,000 years before the present-was warm and dry in western North America. For example, Harding (1965, p. 13T-140) observed stumps of trees that grew on the shore of Lake Tahoe, Calif., in a period when the lake was below the level of the root systems of the trees and below its own outlet. One stump with its base 2 feet below the basin rim had 100 countable rings in 1934. Carbon-14 dating on samples from two such stumps show ages of 4,250±200 years and 4,460+250 years. Obviously, to permit the growth of these trees a protracted warm-dry period must have occurred when inflow to Lake Tahoe did not exceed evaporation from its surface. By contrast, the present average annual inflow is more than three times the average evaporation loss (Harding, 1965, p. 136). Similar evidences of the warm-dry period have been reported in many parts of the world (Matthes, 1942, p. 208-215; Flint and Brandtner, 1961, fig. 1; Gentilli, 1961, p. 491).
Crater Lake was born in that warm-dry period. No doubt a small lake formed soon after volcanic activity in the crater ceased, but the level of the water surface must have been well below the present level until perhaps 1,000 to 4,000 years ago. The period of 2 to 5 millenniums in which the lake was low would have given ample opportunity for forest trees to grow on the crater walls well below the present lake surface. If and when the stumps of such trees can be found and dated, their altitude and age will fill a gap in our knowledge of the development of Crater Lake.
The lower slopes of Mount Mazama are drained by tributaries of the Rogue River to the west ‘and the Klamath River to the south and east. The drainage paths for some parts of the area surrounding the basin of Crater Lake (especially on the north) are uncertain, and some ground water may interchange across topographic divides.
The pumice cover on the higher outer slopes of Mount Mazama, and much of the underlying volcanic materials, are so highly permeable that in places ‘all precipitation infiltrates where it falls. Large areas have no stream channels or other signs of erosion by running water. Downslope from the permeable areas, and within 25 miles of Crater Lake, large springs occur in the basins of the Klamath, Rogue, and Umpqua Rivers. The less extensive areas of glacial deposits and solidified volcanic ash on Mount Mazama are relatively impermeable.
No perennial streams flow into Crater Lake, and there is no overflow. In the summer season, the melting of accumulated snow produces many rivulets that course down the steep caldera walls into the lake. The total measured flow of 63 of these streams was 10.75 cfs (cubic feet per second) in mid-July 1901 (Diller and Patton, 1902, p. 60). Presumably, these streams were not constant in flow, but varied considerably during each day with changes in the rate of snowmelt; also, some unmeasured melt water doubtless was then entering the lake as small seeps at the shore or as underground percolation.
