What is the ‘‘END’’ (emulation of natural disturbance) in forest ecosystem management? An open question Nicole Klenk, Gary Bull, and Dave Cohen Abstract: The emulation of natural disturbance (END) is said to be the most promising avenue for implementing sustain- able forest management; however, there appears to be no consensus as to the meaning of the END. We have interviewed forest scientists across Canada and, with the use of mental models and network textual analyses, created a shared mental model of the END. Results from both quantitative and qualitative analyses suggest that there are contrasting and contradic- tory views about the meaning of the END, which raise serious implications for the use of the END in policy making. Re ´sume ´: On dit que la reproduction des perturbations naturelles est la voie la plus prometteuse pour instaurer un ame ´na- gement forestier durable; cependant, il ne semble pas exister de consensus au sujet de la signification de cette approche. Nous avons interviewe ´ des scientifiques de la fore ˆt partout au Canada et, a ` l’aide de mode `les mentaux et d’analyses tex- tuelles de re ´seaux, nous avons cre ´e ´ un mode `le mental commun de cette approche. Les re ´sultats d’analyses tant quantita- tives que qualitatives indiquent qu’il y a des points de vue diffe ´rents et contradictoires au sujet de la signification de la reproduction des perturbations naturelles, ce qui entraı ˆne de se ´rieuses implications pour l’utilisation de cette approche dans la formulation de politiques. [Traduit par la Re ´daction] Introduction Forest management in much of Canada is moving towards the idea of ‘‘ecosystem management.’’ The stated goal of this type of management is to maintain the ecological integ- rity and health of the forest (CFS 1998). To achieve this goal, some authors have proposed a ‘‘coarse filter’’ approach to forest ecosystem management, namely the emulation of natural disturbance (END; Hunter 1993). The END is justified by using evolutionary reasoning. Some of the ecological premises of the END are (i) that pe- riodic disturbances are inherent to dynamic forest ecosys- tems (Pickett and White 1985); (ii) that natural disturbances are strong determinants of species composition as well as ecosystems structure and function (Attiwill 1994; White and Walker 1997), and (iii) that forest ecosystems and their spe- cies composition have adapted to the disturbances (Bunnell 1995). In other words, it is thought that by maintaining stand and landscape compositions and structures similar to those resulting from natural disturbances, we can reduce the negative impacts of timber harvest on biodiversity and main- tain essential ecological functions (Angelstam 1998; Attiwill 1994; Bergeron and Harvey 1997; Bunnell 1995). In prac- tice, a common objective of the END is to design forest management practices that fit within a historical range of variability of ‘‘natural’’ disturbance regimes (Landres et al.1999). Despite past efforts to clarify the conceptual meaning of the END (see Ecological Applications Vol. 9, 1999; Forest Ecology and Management Vol. 155, 2002; and Silva Fenn- ica Vol. 36, 2002), there appears to be no convergence to- wards a consensus (Landres et al. 1999; McRae et al. 2001). Descriptions of the END differ in their comprehen- siveness and conceptual precision (Perera et al. 2004). Kimmins (2004, p. 13) provides a particularly theoretically comprehensive characterization of the END: ... management over ecologically significant temporal and spatial scales that attempts to emulate the ecosystem ef- fects of physical (allogenic) or biotic (biogenic) distur- bance events, the frequency and (or) severity of which have been changed by human action but which have his- torically determined the potential pathways, patterns, and rates of autogenic successional development in the eco- system in question. Such emulation aims to maintain the historical range of variation, or a socially acceptable sub- set thereof, in desired ecosystem conditions and functions over defined spatial and temporal scales. Although characterizations such as Kimmins’ strongly an- chor the END to theoretical ecology, other characterizations are broader, but less precise: Natural variability is the ecological conditions, and the spatial and temporal variation in these conditions, that are relatively unaffected by people, within a period of time and geographical area appropriate to an expressed goal (Landres et al. 1999, p. 1181). Received 30 July 2007. Accepted 7 April 2008. Published on the NRC Research Press Web site at cjfr.nrc.ca on 28 June 2008. N. Klenk 1,2 and G. Bull. The University of British Columbia, 2022 – 2424 Main Mall, Vancouver BC V6T 1Z4, Canada. D. Cohen. The University of British Columbia, 2900 – 2424 Main Mall, Vancouver BC V6T 1Z4, Canada. 1 Corresponding author (e-mail: nklenk@interchange.ubc.ca). 2 Present address: P.O. Box 195, Palgrave, ON L7E 3T2, Canada. 2159 Can. J. For. Res. 38: 2159–2168 (2008) doi:10.1139/X08-054 # 2008 NRC Canada