318 A NATIONAL PLAN FOR AMERICAN FORESTRY 



Fish Creek and during the 6-year period after the fire an average 

 annual increase of 29 percent. Blaney (op. cit.), however, has attrib- 

 uted this increase to the destruction not of the chaparral, which 

 constitutes 97 percent of the watershed, but of the canyon-bottom 

 forest. Hoyt and TroxelFs implied conclusion that in semiarid regions 

 the land should be denuded of forest to increase stream flow cannot be 

 accepted without first considering the certainty of greatly increased 

 erosion and the usableness of the increased run-off. 



Hoyt and Troxell themselves point out that under normal conditions 

 erosion in the watersheds of Fish Creek and adjacent creeks was 

 negligible, but that samples of water collected from these streams 

 during 4 months immediately after the fire showed a total sand and 

 ash content of 20 to 67 percent by volume and 6 to 40 percent by 

 weight. They state also that in the first year after the fire the large 

 deposit of silt from the burned-over area caused considerable damage 

 to orchards, railroads, and highways adjacent to the mountains. 



Cecil, 23 in discussing the usability of water from southern Cali- 

 fornia watersheds, states : 



The prime requisite in water production is that the water must be usable. 

 This factor is of greater importance than the quantity produced and is vastly 

 more important than a minor increase in the sustained summer flow. Probably 

 95 percent of the water used for domestic and industrial purposes, outside the 

 city of Los Angeles itself, is pumped (from underground reservoirs), as is also 

 upward of 80 percent of that used for irrigation. * * * The replenishment 

 of these underground reservoirs * is of paramount importance. In 



order that the water finding its way from the mountain areas onto the coastal 

 plain may perform its maximum of use, as much of it as possible should percolate 

 into the underground strata near the months of the canyons from which it issues. 

 For years past, several communities, represented by the water companies supply- 

 ing them, have spread the flood waters over the detrital cones by means of lateral 

 ditches, increasing the wetted area and materially increasing percolation over that 

 obtaining under natural conditions. The experience of these companies has proved 

 beyond a doubt that, in order that water may be spread successfully and the maxi- 

 mum of percolation secured, it must be free of suspended matter. It is often 

 necessary, during the first run-off of the season, to by-pass to the ocean a varying 

 part of the flood flow. In the case of a watershed that has been run over by fire, 

 the quantity that must be by-passed because of the silt load is many times as great 

 as that under normal conditions. 



Reports of the Forest Service indicate that before the 1924 fire on 

 Fish, Sawpit, and Rogers Creeks practically all the run-off of these 

 streams was either used for direct irrigation or went to replenish 

 underground reservoirs as described by Cecil. After the fire, much 

 of the run-off in 1925 was unusable because of erosion debris. 



Under the semiarid conditions of southern California it ordinarily 

 takes not less than 5 years for enough vegetation to be reestablished 

 on burned watersheds to serve effectively in handling semitorrential 

 rains. In instances where much of the productive top soil is washed 

 off from the slopes as a result of hard rains in the first year, it will take 

 considerably longer than 5 years to reestablish a closed canopy for 

 the soil. 



Farther north in California the foothills of the Sierra Nevada have 

 suffered disastrously from fire. The effect of destruction of cover by 

 fire in the transition between woodland and forest, in Madera County, 

 is shown by experimental plots of the California Forest Experiment 

 Station. In 1929, with 18 inches of the season's precipitation, 747 

 cubic feet of water per acre ran off the surface of burned plots and 



? 3 Cecil, Q. H. " Discussion of ' Forests and Streamflow.' " Proc. Amer. Soc. Civil Eng., December 1932. 



