Historic Variability for the Upland Vegetation of the Bighorn National Forest, Wyoming Carolyn B. Meyer, Dennis H. Knight, and Gregory K. Dillon Department of Botany, University of Wyoming Laramie, WY 82071-3165 August 22, 2003 Submitted to the Bighorn National Forest, Sheridan, Wyoming and the Division of Renewable Resources, Rocky Mountain Region, U. S. Forest Service, Lakewood, Colorado USFS Agreement No. 1102-0003-98-043 i TABLE OF CONTENTS ACKNOWLEDGEMENTS vi EXECUTIVE SUMMARY viii 1. INTRODUCTION 1 2. METHODS 7 3. THE BIGHORN NATIONAL FOREST 12 3.1. Physical setting 12 3.2. Ecological setting 14 3.3. Cultural setting 17 4. DOMINANT PLANT DISTRIBUTION AND SUCCESSIONAL PATTERNS ALONG ECOLOGICAL GRADIENTS 25 4.1. Dominant plant distribution patterns 25 4.2. Successional patterns 30 4.2.1. High elevation forests 30 4.2.2. Low elevation forest 32 4.2.3. Aspen woodlands 33 4.2.4. Non-forest vegetation 34 5. HISTORIC RANGE OF VARIABILITY FOR DISTURBANCES COMPARED TO EXISTING CONDITIONS 35 5.1. Fire 36 5.1.1. High-elevation forests 36 5.1.2. Low-elevation forests 45 5.1.3. Aspen woodlands 50 5.1.4. Non-forest vegetation 52 5.2. Insects 53 5.2.1. High-elevation forests 53 5.2.2. Low-elevation forests 58 5.2.3. Aspen woodlands 62 5.2.4. Non-forest vegetation 63 ii 5.3. Disease 63 5.3.1. High-elevation forests 63 5.3.2. Low-elevation forests 67 5.3.3. Aspen and non-forest vegetation 69 5.4. Wind 69 5.4.1. High-elevation forests 69 5.4.2. Low-elevation forests (including aspen woodlands) 71 6. HISTORIC RANGE OF VARIABILITY FOR STAND AND LANDSCAPE STRUCTURE COMPARED TO EXISTING CONDITIONS 72 6.1. Stand structure 73 6.1.1. High-elevation forests 74 Tree density 75 Regeneration time and seedling/sapling density 76 Percent canopy cover and rate of gap formation 77 Density of canopy gaps 78 Density and cover of understory plants 78 Plant species diversity 79 Size- and age-class structure of stands 81 Forest floor depth 82 Mineral soil exposed, disrupted, or compacted 83 Snag density 84 Coarse woody debris 85 6.1.2. Low-elevation forests 87 Tree and sapling density 88 Plant species diversity, canopy cover and gaps, and understory plant composition and cover 90 Size- and age-class structure of trees 91 Spatial distribution of trees 92 Forest floor depth 93 Snags and coarse woody debris 93 6.1.3. Aspen woodlands 95 iii 6.1.4. Non-forest vegetation 96 6.2. Landscape structure 99 6.2.1. High-elevation landsapes 100 Number and proportion of land cover types 104 Forest/non-forest ratio 105 Proportion of forests in different successional stages and old-growth 105 Proportion of landscape in low canopy cover 107 Proportion of landscape in high-density classes for snags and coarse woody debris 108 Edge, interior forest habitat, patch shape, and patchiness 108 Disturbance return intervals and rate at which new patches are formed 112 6.2.2. Low-elevation landscapes (including aspen and non-forerst Vegetation) 113 Stand structure at the landscape scale 113 Proportion of landscape in different land cover types 114 Forest patch sizes and configuration 115 Proportion of landscape dominated by old trees 116 Proportion of landscape with high-densities of snags and coarse woody debris 117 7. SUMMARY OF PROBABLE HRV DEVIATIONS 117 8. LITERATURE CITED 125 9. APPENDIX 147 10. TABLES 148 11. FIGURES 159 iv v ACKNOWLEDGMENTS We thank the following individuals for their assistance in the preparation of this report: Claudia Regan, Regional Office, U.S. Forest Service, and Department of Forest Sciences, Colorado State University Bernie Bornong, Bighorn National Forest Rick Laurent, Bighorn National Forest DeeDee Arzy, Bighorn National Forest David Anderson, Bighorn National Forest Pam Ziesler, Bighorn National Forest David Myers, Bighorn National Forest Bob Daniels, Bighorn National Forest R. Scott Gall, Bighorn National Forest Ron Stellingwerf, Bighorn National Forest Dave Beard, Bighorn National Forest Bill Biastoch, Bighorn National Forest Chris Thomas, Bighorn National Forest Bryce Bohn, Bighorn National Forest Harold Golden, Bighorn National Forest Judy von Ahlefeldt, retired, Medicine Bow National Forest Greg Hayward, Regional Office, U.S. Forest Service, and Department of Zoology and Physiology, University of Wyoming Kurt Allen, Regional Office, U.S. Forest Service Burt Jellison, Wyoming Game and Fish Department Thomas T. Veblen, Department of Geography, University of Colorado William H. Romme, Dept. of Forest Sciences, Colorado State University Walt Fertig, Wyoming Natural Diversity Database, University of Wyoming Gary Beauvais, Wyoming Natural Diversity Database, Univ. of Wyoming George Jones, Wyoming Natural Diversity Database, Univ. of Wyoming William A. Reiners, Wyoming Geographical Information Sciences Center and Department of Botany, University of Wyoming William L. Baker, Department of Geography, University of Wyoming Stephen T. Jackson, Department of Botany, University of Wyoming Stephen T. Gray, Department of Botany, University of Wyoming Jodi Norris, Department of Botany, University of Wyoming Christopher L. Fastie, Department of Biology, Middlebury College Five anonymous reviewers vi vii EXECUTIVE SUMMARY The challenges of sustainable land management have led to an increased emphasis on incorporating the results of science in the decision making process. One approach for accomplishing this objective is through the analysis of the historic range of variability (HRV) for key ecosystem variables that are affected by management activities. Such analyses provide a means of synthesizing the scientific literature. The rationale for HRV analyses is that the chances of sustainable forest management are greater if the variation in managed ecosystems includes the range of conditions that are expected at various scales in ecosystems relatively uninfluenced by humans. This report provides an HRV analysis for the Bighorn National Forest (BNF) of northcentral Wyoming. By definition, HRV analyses require the identification of specific variables and an estimation of how those variables fluctuated, at more than one scale, prior to the advent of resource extraction and management by European- Americans. A complete list of the variables that we examined is found in Table 7. Examples include live tree density, dead tree (snag) density, canopy cover, abundance of coarse woody debris, species diversity, fire return intervals, the abundance of various diseases, the proportion of the landscape in different land cover types, and the degree of patchiness in the landscape. We examined variables at two scales, namely, the stand and the landscape. For this report we separated high-elevation landscapes from low-elevation landscapes. Much of the report pertains to forests dominated by lodgepole pine, Engelmann spruce, and subalpine fire at high elevations, and ponderosa pine, Douglas-fir, and aspen viii at lower elevations. Our analysis emphasizes forests rather than grasslands and shrublands because more historical information is available for the forests. Two significant challenges for HRV analyses are 1) selecting a “reference period” for comparison to present day conditions, and 2) making decisions with very little data about the range of variability of a variable during the reference period. We defined the HRV reference period for the BNF as approximately 1600 to 1890, as some data sets on fire history go back several centuries and the influence of European-Americans in the BNF was minimal until about 1890. We also compared BNF ecosystems to comparable ecosystems in natural areas relatively uninfluenced by management activities, such as Yellowstone National Park. When data were lacking, our approach was to first make plausible though qualitative judgements about the conditions that must have existed in the reference period based on recent studies of plant adaptations and current knowledge about ecosystem structure and function. This approach is an example of deductive science. There will always be some uncertainty in such conclusions, but our conclusions were evaluated and supported by a panel of six anonymous peer reviewers. For each of our conclusions about whether a stand or landscape variable is within the HRV or trending away from that norm in managed landscapes, and the Forest as a whole, we indicate whether our confidence level is low, moderate or high. Our report describes how the forests of the BNF have evolved with regular disturbances, and how the kinds and frequency of disturbances have changed in some areas due to modern management practices. Only ~18% of the forested ix area has been harvested for timber (high- and low-elevation landscapes combined). Roughly 12% has been harvested on two occasions. The effects of fire suppression and livestock grazing are more widespread, especially at the high elevations. Section 7 presents a summary of our conclusions. The following variables appear to have exceeded their HRV in some areas affected by management activities: Within stands at high elevations affected by timber harvest (~20% of the high-elevation forests on the BNF): 1. Canopy cover probably is lower and the size and density of canopy gaps probably is higher in harvested stands than the HRV for these variables in unmanaged forests of comparable age and site conditions due to selective and shelterwood cuts, and fewer standing-dead trees (moderate confidence). 2. Where timber harvesting has occurred, snag density and the amount of coarse woody debris is lower than the HRV for unmanaged stands of comparable age and site conditions (high confidence). Whole-tree