Table 3. --Average percent increase in height gromth following 

 fertilization of ponderosa pine on granitic road fills 



Year } 1968 trees ; 1969 trees 



-------- Percent -------- 



1969 32 



1970 60 71 



1971 70 116 



1972 38 68 



sets, except those for 1969 and 1972 on the 1968 trees. Even in these years, however, 

 fertilization did increase growth over some unfertilized plots. Mulching appears to 

 have increased growth somewhat, especially for the 1969 trees; however, increases were 

 significant only at the 90 percent level, and then in but a few cases. Wider spacing 

 also tended to increase growth in some instances, but again increases were detected 

 only at the 90 percent level. 



Table 3 was developed to present a clearer picture of average fertilizer effects. 

 Growth is expressed as the average percent increase in growth on plots treated with 

 mulch and fertilizer compared to plots treated with mulch alone. There is a tendency 

 for fertilizer effects to decrease 3 to 4 years after application. Although fertilizer 

 was effective in increasing growth on both sets of trees, the effect was greatest on 

 the 1969 trees. The additional grovrth increase was probably caused by the fact that a 

 second fertilizer pellet was used when trees were replanted in 1969. This type of 

 fertilizer releases nutrients slowly; some original tree pellets placed in 1968 were 

 sti.ll recognizable at the time of planting in 1969. Thus, the 1969 trees received 

 more fertilizer than the 1968 trees. 



Erosion Control 



Erosion data were summarized by water years (Oct. 1-Sept. 30) beginning in October 

 1969; consequently, data were provided for 3 complete years, 1970, 1971, and 1972. 

 Additional data were available for the summer of 1969, but were not analyzed because 

 of undue soil disturbance resulting from plot maintenance during the period. 



Previous studies (Megahan and Kidd 1972) indicated that the standard deviations 

 of data from erosion plots similar to those used in the present study tend to be pro- 

 portional to the mean. Consequently, a log (X + 1.0) transformation of erosion data 

 was used to provide a more reliable statistical test of treatment effects. 



Statistical analyses were made for each year of data and included a standard 

 analysis of variance to detect overall treatment and block effects and a sequential 

 multiple range test to compare individual treatment effects. 



Analysis of variance showed highly significant treatment effects (99 percent level), 

 but no block effects for all years of data tested. Results of the multiple range tests 

 are summarized in figure 8. Variations in erosion by years primarily reflect annual 

 variation in erosion energy available from raindrop impact and wind. No significant 

 differences in erosion were observed among the various mulched plots nor the 1.5- by 

 1.5- and 2.5- by 2.5-foot unmulched plots. Compared to control plots, planted trees 

 without mulch reduced average annual erosion rates by 48, 32, and 51 percent in 1970, 1971, 



13 



