Table 9 — Difference between assumptions and actual or adjusted data (pounds/acre) 



Range 



RPA 



Environmental statement 



Soil report 



Actual 

 production 



Assumption 



Difference 

 from Actual 



Assumption 



Difference 

 from Actual 



Assumption 



Difference 

 from Actual 



Good condition ranges 



High production 740 



Moderately high production 740 



Average production 740 



Moderately low production 740 



Fair condition ranges 



High production 530 



Moderately high production 530 



Average production 530 



Moderately low production 530 



Data adjusted to represent 500 

 good range condition and 

 average soil and 

 precipitation conditions 



1.200 

 1,000 

 850 

 700 



1.100 

 900 

 750 

 600 



850 



480 

 260 

 110 

 {-)40 



570 

 370 

 220 

 70 



350 



'806 



66 



^1,415 

 ^780 



675 

 40 



'606 



76 



'1.038 



538 



'Estimate was 500-700-lb/aae increase under "favorable" conditions which suggests average or above average precipitation. Pretreatment production was 106 

 lb/acre at Oak Creek. 



n'he 1 ,415-Ib per acre estimate represents a typical estimate of production for the Oak Creek area in favorable years, and the 780 lb/awe represents an estimate 

 of production for below-average soils in the Oak Creek area in favorable years (Stoneman and others 1978). 

 'Average for estimates of favorable and unfavorable years for typical soil conditions. 



In addition, the production data from Clay Springs were 

 further adjusted to account for less productive eolian soils 

 to place these values on a par with the other three study 

 areas (table 7). The adjusted production data were as- 

 sumed to more accurately represent changes over time 

 under conditions of average precipitation and average soil 

 conditions. The data suggest a peaking of production 6 to 

 8 years after seeding at about 550 lb per acre and a grad- 

 ual decline to approximately 450 lb per acre in older seed- 

 ings (fig. 3). Using 500 lb as the best estimate of average 

 long-term production potential, the study areas at Oak 

 Creek still appeared to be at least one productivity class 

 less than assumed in the RPA scenario for an "average" 

 pinyon-juniper site (table 9). The adjusted average value 

 for the swale bottom topographic position (628 lb per acre) 

 also fell short of the "average" expected production 

 (850 lb per acre). 



DISCUSSION 



The reasons plant production did not meet assumed 

 production levels are not known. A major problem occurs 

 in fitting a specific area into a nonspecific array. Without 

 clear identifiers the proper location within an array, such 

 as that in table 3, may involve a bit of guesswork. Some 

 elements of the environment such as precipitation and 

 soils provide clues, but no specific guides are available. 

 When a viable classification of the pinyon-juniper type is 

 developed, the proper positioning of a particular area 

 within such an array will be more likely. The best ap- 

 proach to evaluate or test an array of production coeffi- 

 cients for a wide-ranging vegetation zone would be to draw 



unbiased samples from across the vegetation zone and 

 determine the statistical distribution of measured out- 

 puts. This would alleviate the need to identify the exact 

 "pigeon-hole" into which an individual location fits. 



A second question is whether the production levels were 

 assumed to represent grazed or nongrazed situations. 

 A review of recent literature suggests that, after tree 

 removal, grazed pinyon-juniper areas rarely exceed 800 lb 

 per acre except in localized situations. Conversely, after 

 tree removal ungrazed pinyon-juniper areas oflen exceed 

 1,200 lb per acre (Clary 1987). Presumably, the grazed 

 situation should apply to the RPA assumptions because 

 the key issue was a question of AUM's of grazing capacity. 

 However, this problem was not recognized in the RPA 

 scenario nor in the local assumptions. 



Another possible problem concerns the production 

 trends of seeded stands. Stands of good initial density, 

 which are often established when drilling or other inten- 

 sive revegetation techniques are used, typically peak in 

 production after 2 to 3 years (White 1985). These stands 

 then decline through at least the sixth year to about 50 

 percent of their peak value (fig. 5). If production sampling 

 occurs during the peak production period, the values ob- 

 tained are highly biased estimates of long-term average 

 production. A difTerent pattern occurs when the seeded 

 stand is sparse or when no seeding is done and the native 

 vegetation is allowed to recover naturally. A thickening of 

 the stand and an increase in production may occur for 

 some years, particularly if sodformers are part of the 

 composition. In their Southwestern study, Arnold and 

 others (1964) found approximately 10 years were required 

 to achieve maximum production of herbage after juniper 

 removal (fig. 6). A similar pattern undoubtedly occurs in 



9 



