Age-Class Structure of Old 

 Growth Ponderosa Pine/ 

 Douglas-Fir Stands and Its 

 Relationship to Fire History 



Stephen F. Arno 

 Joe H. Scott 

 Michael G. Hartwell 



Introduction 



A century ago the ponderosa pine {Pinus ponderosa) 

 type covered about 40 million acres in the Western 

 United States (Van Hooser and Keegan 1988). This 

 species continues to dominate (in denser, younger 

 stands) the driest regions, where it represents the 

 potential climax type (Pfister and others 1977; 

 Wellner 1989). In the less-droughty half of its distri- 

 bution, however, ponderosa pine formed a serai, fire- 

 maintained cover type, and is often being replaced 

 successionally by other species (Arno 1988). Prior 

 to 1900 in both the climax and serai pine types fre- 

 quent surface fires, termed underburns, kept most 

 stands in an open park-like condition dominated by 

 large old trees (Cooper 1960; Leiberg 1899; Wickman 

 1992). In the areas where ponderosa pine is serai, un- 

 derbiUTis prevented more shade-tolerant competitors— 

 interior Douglas-fir (Pseudotsuga menziesii var. glauca), 

 grand fir (Abies grandis), and white fir (A. concolor)— 

 fi-om developing an understory and eventually replac- 

 ing the shade-intolerant, but fire-resistant pine (Arno 

 1988; Weaver 1967). 



Underburns have been excluded for 60 to 90 years 

 in most of the serai ponderosa pine forests. As a re- 

 sult, acciimulations of svuface fuels and conifer thickets 

 have developed that allow stand-replacing wildfires 

 to become common (Arno and Brown 1991; Barrett 

 1988; Mutch and others 1993). Past selective cutting 

 has removed much of the pine and encouraged devel- 

 opment of dense fir thickets. The loss of the serai old 

 growth ponderosa pine type is of great concern for wild- 

 life habitat and maintenance of biodiversity on Na- 

 tional Forest lands (Everett 1994). Present-day stands 

 are experiencing extensive mortality from insect and 

 disease epidemics (Wickman 1992). These threats, 

 coupled with a growing appreciation of the importance 

 of natural ecological processes, have spurred interest 

 in strategies to perpetuate serai ponderosa pine forests 

 (Fiedler and others 1992; Mutch and others 1993). 

 Developing such strategies will require knowledge of 



how ponderosa pine and its companion species regen- 

 erated in association with past fires. 



Detailed fire history can be determined from 

 analysis of fire scars on trees (Arno and Sneck 1977; 

 McBride 1983), but accurate age-class data from old 

 growth stands have been difficult to obtain. We re- 

 cently adapted a power increment borer to efficiently 

 sample large trees (Scott and Arno 1992) and deter- 

 mine age structures in old growth stands. Our objec- 

 tives in this study were to investigate the influence 

 of past fires on stand age structure on both dry and 

 moist site-types and identify changes in stand struc- 

 ture that are associated with fire exclusion. 



Study Sites and Fire History 



Old growth ponderosa pine was abundant in the 

 accessible lower-elevation valleys and mountain slopes 

 in western Montana and has been logged heavily for 

 more than 100 years. Because of this, it was not pos- 

 sible to locate large unlogged stands and select sample 

 areas using criteria that would ensure representative- 

 ness. Less than 1 percent of the old growth serai pon- 

 derosa pine type has no history of logging. Much of 

 this remnant unlogged area is confined to precipitous, 

 broken terrain that is difficult to sample. Neverthe- 

 less, it seemed that useful information and interpre- 

 tations on the relationship of fire history to stand 

 structure could be derived fi^om these remnsmt stands. 

 Therefore, we consulted National Forest silvicultur- 

 ists and inspected stand maps to identify unlogged 

 stands on relatively uniform topography. Eventually 

 we were able to locate and sample remnant stands 

 on both dry and moist sites that were large enough 

 to contain a 100-m square (~2.5 acre) sample plot. 



Fire histories were determined for each plot from 

 analysis of partial cross sections from the two to four 

 trees in £md immediately adjacent to the plot that had 

 the most complete and least damaged sequences of 

 multiple fire scars (Arno and Sneck 1977). A master 



1 



