PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY Cave Junction, Oregon May 29-31, 2003

Klamath-Siskiyou

Kalmiopsis Ecoregon

Siskiyou Mountains

Marble Mountains

Trinity Alps

California and Oregon

Yolla Bollies

0 10 20 40 Miles

Editors: Kristi L. Mergenthaler, Jack E. Williams, and Erik S. Jules THANKS TO THE FOLLOWING FOR THEIR GENEROUS SUPPORT

CONFERENCE CO-SPONSORS: AuCoin Institute at Southern Oregon University Klamath-Siskiyou Wildlands Center Native Plant Society of Oregon Oregon Caves National Monument Siskiyou Field Institute The Siskiyou Project World Wildlife Fund

MAJOR FUNDING BY: AuCoin Institute at Southern Oregon University The Mountaineers Foundation Native Plant Society of Oregon (State & Siskiyou Chapters) The Nature Conservancy Siskiyou Regional Education Project

ADDITIONAL COPIES AVAILABLE FROM:

Siskiyou Field Institute 9335 Takilma Road Cave Junction, Oregon 97523 Cost: $8.00 (includes postage)

SISKIYOU FIELD INSTITUTE PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY

Cave Junction, Oregon May 29-31, 2003

Editors: Kristi L. Mergenthaler, Jack E. Williams, and Erik S. Jules

Siskiyou Field Institute, Cave Junction, Oregon 2004 ACKNOWLEDGEMENTS

Peer reviews are an integral component of the scientific process. We would like to thank the following individuals who graciously donated their time and expertise in reviewing the papers presented here: Geoff Babb Dr. Michael Parker Frank Betlejewski Daniel Sarr Shannon Clery Dr. John Sawyer Dr. Dominick DellaSala Nathaniel Seavy Dr. Jim Duncan Cecile Shohet Dr. Donald J. Goheen Dr. George Shook Ellen Goheen Dr. Darlene Southworth Dr. Stewart Janes Dr. Karen Stone Dr. Thomas Jimerson Molly Sullivan Dr. Erik Jules Robert Sweeney Dr. Matthew Kauffman Dr. Pepper Trail Kristi Mergenthaler Lee Webb Barbara Mumblo Dr. Charles Welden Dr. Michael Murray Dr. Jack Williams Joel Pagel Anonymous Reviewers

We would like to express our appreciation for use of map and illustrations from the following artists: Bob Cremins (numerous line drawings) David Hicks (Great Blue Heron) Jamie O’Donnell (cover map)

We would like to acknowledge the final design and layout by: Susan DeRosia

Printed and sponsored in part by: Home Run Graphics, LLC Portland, OR 97206 (503) 504-2273 Email [email protected]

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY ii TABLE OF CONTENTS

EDITORS’INTRODUCTION Kristi L. Mergenthaler, Jack E. Williams, and Erik S. Jules 1

PLENARY TALK State of the Klamath Knot: how far have we come and where are we going? Dominick A. DellaSala 2

FULL ARTICLES

PHYTOPHTHORA DISEASE ECOLOGY

Phytophthora in the world’s forests Everett Hansen 10

Assessing the landscape spread of the fatal Port Orford cedar root rot Erik S. Jules, Matthew J. Kauffman, Allyson L. Carroll, and William D. Ritts 13

Exotic pathogens, resistant seed and restoration of forest tree species in western North America Richard A. Sniezko, Diana F. Tomback, Regina M. Rochefort, Ellen Goheen, Rich Hunt, Jerry S. Beatty, Michael Murray, and Frank Betlejewski 21

Genetic resistance in Port-Orford-cedar to the non-native root rot pathogen Phytophthora lateralis: a tool to aid in restoration in infested riparian areas Richard Sniezko, Leslie Elliott, Everett Hansen, and Don Goheen 27

WILDLIFE ECOLOGY

Using a wide-scale landbird monitoring network to determine landbird distribution and productivity in the Klamath-Siskiyou Region John D. Alexander, C. John Ralph, Kimberly Hollinger and Bill Hogoboom 33

Understanding effects of fire suppression, fuels treatment, and wildfire on bird communities in the Klamath-Siskiyou Ecoregion John D. Alexander, C. John Ralph, Bill Hogoboom, Nathaniel E. Seavy, and Stewart Janes 42

Population trends among landbirds of the Klamath-Siskiyou Ecoregion: an analysis of Breeding Bird Survey data Pepper W. Trail 47

Conservation status of American martens and fishers in the Klamath-Siskiyou Bioregion Keith M. Slauson and William J. Zielinski 60

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY iii DISTURBANCE AND CHANGE

Fire and vegetation dynamics in the western Klamath Mountains Dennis C. Odion, Evan J. Frost, Dominick A. DellaSala, James R. Strittholt, Hong Jiang, and Max A. Moritz 71

Losing ground: wilderness meadows and tree invasion over a 55-year period in California’s Klamath Range Michael P. Murray 81

Vascular plant species of Baby Foot Lake Botanical Area before the Biscuit Fre, with comments on post-fire flora Linda Vorobik 86

VEGETATION ECOLOGY

The Oregon Flora Project Linda K. Hardison and Scott Sundburg 98

Analysis of Pacific yew habitat in northwest California Thomas M. Jimerson and Stanley Scher 105

Woody plant distributions in western Oregon riparian forests: insights for restoration and management Daniel A. Sarr and David E. Hibbs 119

Conifers of the Klamath Mountains John O. Sawyer 128

Biodiversity below ground: mycorrhizal fungi and Oregon white oaks (Quercus garryana) Lori Valentine, Carolyn Petersen, Heather Tugaw, Aaron Hart, Mariah Moser, Jonathan Frank, Harold Berninghausen, and DarleneSouthworth 135

CULTURAL ECOLOGY

Restoring indigenous history and culture to the Klamath-Siskiyou Ecoregion: conservation, restoration, and wood fiber production in the forest matrix Dennis Martinez 140

ABSTRACTS

Breeding system and seed production of an endangered plant, Fritillaria getneri, in southern Oregon Kelly Amsbury and Bob Meinke 148

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY iv Effects of fire on rare plants and vegetation in serpentine savanna and fen communities in southern Oregon Darren Borgias, Nathan Rudd and Len Eisenberg 149

Postglacial vegetation and fire history near Bolan Lake in the northern Siskiyou Mountains of Oregon Christy Briles and Cathy Whitlock 150

The return of the wolf to the Klamath-Siskiyou Region: implications for ecosystem conservation Carlos Carroll 150

The impacts of the Quartz Gulch Fire on fish and aquatic amphibian population dynamics in Glade Creek,Oregon Jake Chambers 151

Potential pollinators and insect visitors to threatened and endangered Fritillaria getneri (Liliaceae) and closely related species Fritillaria affinis and Fritillaria recurva Kathleen Donham and Carol Ferguson 151

A rare orchid, Cypripedum fasciculatum (Orchidaceae), breaks the rules for pollination Carol Ferguson and Kathleen Donham 152

Turnover, rate of species accumulation, and persistence of species rich assemblages of butterflies in the Marble Mountain Wilderness, Siskiyou County, California Robert F. Fernau 152

Distribution and environmental/ habitat relations of five endemic plants associated with serpentine fens Evan Frost 153

Phytophthora ramorum: the cause of Sudden Oak Death Ellen Goheen 153

Fish and the Biscuit Fire Tom Halferty 154

Broad-leaved noxious weed abundance and distribution across the Cascade-Siskiyou National Monument, Southern Oregon Paul Hosten 154

Native American management and prescribed burning of riparian zones Frank K. Lake 155

Forgotten birds of the riparian system: monitoring stream-associated birds as a metric for watershed quality Sherri L. Miller and C. John Ralph 155

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY v Pacific lamprey redd densities in the Illinois River of southwestern Oregon Richard K. Nawa and Greg Bennet 156

An ecological study of Preston Peak’s flora: establishing baseline data for climate change research on subalpine vegetation Jamie O’Donnell 156

Analysis of environmental factors limiting Pacific yew occurrence and regeneration in southwestern Oregon Stanley Scher 157

Impacts of the Quartz Gulch Fire on stream channel characteristics within Glade Creek Ted Sedell 157

Matrix habitat effects on the health of patchy plant populations: serpentine plants in changing landscape Priya Shahani 158

Atlas of rare plant species on Lake Peak, Oregon: baseline data for global climate change monitoring Lori Sims 158

The Biscuit Fire: vegetation changes and post-fire rehabilitation Cecile Shohet and Wayne Rolle 159

Uncertain harvest: patterns of acorn production in interior southwest Oregon and far-northern California Donn L. Todt and Nan Hannon 159

Overview of the history of public lands and the Siskiyou National Forest: the early years and the Oregon Land Fraud Greg Walter 159

Basin-scale restoration planning and monitoring in southwest Oregon Debra Whitall 160

Guiding principles of ecologically-sound natural resource management in fire-adapted ecosystems of the West: developing scientific consensus Cindy Deacon Williams, Jerry F. Franklin, Jack E. Williams, and Dominick A. DellaSala 160

Potential impact of the Sudden Oak Death pathogen (Phytophthora ramorum) on the Siskiyou flora James W. “Djibo” Zanzot and Jennifer L. Parke 161

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY vi INTRODUCTION

Introduction to the Second Conference on Klamath-Siskiyou Ecology

any of us are somehow drawn to the diversity, beauty can be powerful forces that allow us to forge common bonds with and wild nature of this portion of northwest California those who we otherwise might feel estranged or separate from. MMand southern Oregon we call the Klamath-Siskiyou We all feel the urge to belong to something larger and more Region. This is a region of great geological and topographical permanent than ourselves. All of us need time to reflect about diversity, including the nexus of the Sierra Nevada, Cascade and such matters, and to find our place in these lands and waters Coast Ranges as well as smaller but nonetheless spectacular around us. Siskiyou, Marble, Trinity, and Yolla Bolly Mountains. David The purpose behind these proceedings is to provide a Rains Wallace (1983) called this land of steep valleys and twisted greater understanding of this place called the Klamath-Siskiyou mountains The Klamath Knot. It is a virtual collision of mountain Region and to assist in stewardship and restoration of its ranges. The Klamath-Siskiyou is a region of wild rivers, dark resources. Like many other parts of our globe, people here are in caverns, tall trees, and serpentine outcrops, whose habitats harbor a continuing struggle to understand how best to live with the land. a diversity of life seldom matched in other temperate regions. The The Klamath-Siskiyou Region is not without its threats: Sudden forests of the Klamath-Siskiyou are recognized as one of the four Oak Death, climate change, wildfire suppression, loss of roadless biologically richest temperate coniferous forests in the world areas, and the spread of exotic weeds. Almost daily, we are faced (DellaSala et al. 1999). As Frank Lang has with questions about how best to manage said, “The Klamath-Siskiyou Region We shall never achieve the land and the people — how is one of those places on our “ to deal with the aftermath of planet that can evoke wonder, harmony with land, any more than we shall the Biscuit Fire of 2002, reverence, and unending achieve absolute justice or liberty for people. how to protect curiosity among all who dwindling salmon runs, venture there.” It is also a In these higher aspirations the important thing is not to how to diversify rural region we call home. achieve, but to strive. economies, or how to The First Conference on ” deal with increasing Klamath-Siskiyou Ecology was held in —Aldo Leopold human demands for 1997 and provided concrete evidence of the long- recreation and resource standing claim that the Klamath-Siskiyou Region is a unique place consumption. These are some of our greatest challenges and and one of the preeminent spots for biodiversity in North America. worthy of our best efforts. This volume provides the proceedings of the Second Conference Aldo Leopold (Leopold 1953) once said: “We shall never on Klamath-Siskiyou Ecology, held on May 29-31, 2003, in Cave achieve harmony with land, any more than we shall achieve Junction, Oregon. All authors of the keynote address, oral absolute justice or liberty for people. In these higher aspirations presentations, poster presentations, and banquet presentation were the important thing is not to achieve, but to strive.” We believe invited to submit either a formal paper or abstract for the that a better understanding of the land Ð of that place in which we proceedings. This resulted in 18 full papers and 26 abstracts that live, raise our families, work and seek spiritual renewal Ð is a are included herein. Each paper was reviewed by two or more prerequisite in our struggle to live a sustainable lifestyle, or as peers as well as one or more of the editors. After the plenary Leopold spoke of it, that harmonious life. We gratefully paper on the State of the Knot, the full papers are divided into the acknowledge all the contributors and sponsors of these following topical areas: Phytophthora Disease Ecology, Wildlife proceedings, those who attended the conference, and others that Ecology, Disturbance and Change, Vegetation Ecology, and seek to improve our knowledge of the Klamath-Siskiyou Region. Cultural Ecology. Abstracts follow, listed alphabetically by author. KRISTI L. MERGENTHALER Philosophers and far-sighted educators increasingly tell us JACK E. WILLIAMS about the importance of “place” (Orr 1994; Sobel 2004). Places ERIK S. JULES

Literature Cited Orr, D.W. 1994. Earth in mind: on education, environment, and the human prospect. Island Press, Washington, D.C. DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt, and D.M. Sobel, D. 2004. Place-based education: connecting class rooms Olson. 1999. A global perspective on biodiversity of the and communities. Orion Society, Great Barrington, Klamath-Siskiyou ecoregion. Natural Areas Journal Massachusetts. 19(4):300-319. Wallace, D.R. 1983. The Klamath Knot: explorations on myth Leopold, L.B., editor. 1953. Round river: from the journals of and evolution. Sierra Club Books, San Francisco, Aldo Leopold. Oxford University Press, New York. California.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 1

PLENARY TALK

STATE OF THE KLAMATH KNOT: HOW FAR HAV E WE COME AND WHERE ARE WE GOING?

Dominick A. DellaSala World Wildlife Fund, 116 Lithia Way, Suite 7, Ashland, OR 97520 Email: [email protected]

Introduction In spite of its many global accolades, the area is under enormous pressure from unsustainable land and water management practices that threaten to unravel the biological he Klamath-Siskiyou Ecoregion of northwest California tapestry of the Klamath Knot. Here, I summarize the “State of and southwest Oregon has been long-regarded for its The Knot” by: (1) reviewing the scientific literature on the extraordinary biodiversity (see DellaSala et al. 1999; Jules TT biology of place; (2) synthesizing research on the status, et al. 1999 for reviews). Many pioneering botanists, the likes of condition, and threats to biodiversity; and (3) outlining challenges Alice Eastwood, Edward Greene, Lois Henderson, and T.J. to a sustainable future for the ecoregion. This paper provides the Howell, combed the region in search of plants unknown to science context for this second conference on the “Ecology of the that today bare their names. But it was not until the 1950s Siskiyous,” the first which took place in 1997, and lays out through the seminal work of Robert Whittaker that the ecoregion conservation priorities that need to be embraced by scientists, the began to garnish its notoriety within scientific circles. Whittaker’s public, land managers, and decision makers if we are going to ecological studies put both the Klamath-Siskiyou and the southern save this magnificent place. Appalachia ecoregions on a national map of botanical importance. In addition, Whittaker (1960) aptly noted that logging in the Siskiyous was beginning to degrade the landscape: The Biology of Place

“With the relative depletion of more available timber farther Delineating boundaries - the specific geographic boundaries north and the consequent shift of lumbering activity toward the of the Klamath-Siskiyou Ecoregion depend on the type of south, much of the forest area of the Siskiyous is being rapidly ecoregional classification system used by researchers. Ricketts et cut; and conservation and sustained-yield programs are little in al. (1999) described the ecoregions of North America using evidence.” terrestrial classifications developed mainly by Omernick (1995). In a related study, Abell et al. (2000) classified North American Whittaker (1960) was witnessing the early stages of forest ecoregions based on freshwater features. Notably, an ecoregion is fragmentation ushered in after World War II with accelerated defined as a relatively large land or water area that contains a logging on public lands. Logging and road building continued geographically distinct assemblage of natural communities that largely unabated for decades until the listing of the northern share a large majority of their species, ecological processes, and spotted owl (Strix occidentalis caurina) pursuant to the environmental conditions (see Ricketts et al. 1999; DellaSala et al. Endangered Species Act and the completion of the Northwest 1999). The Klamath-Siskiyou Ecoregion, in particular, represents Forest Plan in 1993 that set aside significant tracks of late- the grouping of mountains with similar geological origins south of successional reserves throughout the Pacific Northwest while the Klamath River (i.e., “the Klamaths”), and the Siskiyous that reducing the annual cut on federal lands such the Siskiyou are considered a geological subset of the Klamaths to the north. National Forest from around 165 million board feet to an average Previous studies (Ricketts et al. 1999; DellaSala et al. 1999) of 27 million board feet. delineating the ecoregion relied on boundaries derived mainly Since Whittaker’s time, however, the Klamath-Siskiyou has from land-based classifications. This presented some received growing recognition from conservationists concerned disagreements among taxonomic experts in determining which about its unique conservation status. The groundbreaking essays aquatic species to include in the Klamath-Siskiyou because of David Rains Wallace (1983) inspired a folklore-like affinity for aquatic distributions did not line up precisely with terrestrially the region while popularizing its name as the “Klamath Knot.” derived boundaries (DellaSala et al. 1999). Thus, the ecoregional More recently, the ecoregion has been regarded as an area of descriptions provided by Noss et al. (1999) are a hybridization of global botanical significance by the World Conservation Union terrestrial and freshwater classifications more appropriate for the (IUCN – 1992), a proposed “World Heritage Site” and UNESCO Klamath-Siskiyou Ecoregion and are therefore used herein (Figure “Biosphere Reserve” (Vance-Borland et al. 1995), a global “center 1). While the mapping of ecoregions is widely accepted in of plant diversity” (Wagner 1997), and a “Global 200” ecoregion scientific circles, precise boundaries are often in dispute. by the World Wildlife Fund (Ricketts et al. 1999; DellaSala et al. However, Ricketts et al. (1999) noted that different classification 1999). systems often yielded common centroids (central position of the

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 2

PLENARY TALK ecoregion) regardless of the classification system used. Siskiyou’s extraordinary biodiversity can be simplified as follows: World-Class Accolades - perhaps the most striking aspect of the Klamath-Siskiyou’s diverse ecology is that the more scientists Complex Geology + Ancient Landscape + Varied Climate + explore it, the more diverse and unique the ecoregion appears. Ecological and Evolutionary Processes = While the ecoregion lacks the “charismatic megafauna” of other “Galapagos of North America.” nationally regarded areas (such as bison Bison bison and gray wolf Canis lupus in Yellowstone), the biology of place is subtle but The analogy to the Galapagos is warranted in the sense that nevertheless most evident. the Klamath-Siskiyou originated from subduction that plunged The following attributes, also summarized in DellaSala et al. large amounts of buoyant trench sediments deep into the (1999), place the Klamath-Siskiyou among the top ten temperate subduction zone, leading to an uplift of the ecoregion as the first conifer ecoregions on Earth: of coastal mountain ranges (an ancient island system) to spring up from the sea floor some 150 million years ago (Coleman and ¥ Approximately 3,500 plant species, including 220 endemic Kruckeberg 1999). What began as an island is now an ancient and vascular plant taxa (including subspecies and varieties) twisted assortment of mountain ranges whose evolutionary within northwest California and southwest Oregon (J. significance is arguably as important as the Galapagos. The Sawyer, personal communication; element of time (ancient landscape) in the biodiversity formula http://www.Humboldt.edu/%7Eherb/). and the considerable topo-edaphic gradients have contributed to ¥ Nearly two-thirds of the entire California floristic province adaptive radiations in many taxa, placing the ecoregion on similar (Smith and Sawyer 1988). footing with the Galapagos as a “maternity ward” for speciation. ¥ At least 30 conifer species (depending on specific This is most evident in the exceptional levels of plant and boundaries some researchers report as many as 40 species; J. invertebrate endemism (particularly snail endemism and richness Sawyer, personal communication). that are much lower in younger geological landscapes surrounding ¥ Up to 115 species of butterfly (Cascade-Siskiyou National the ecoregion; T. Frest, personal communication). No doubt Monument, E. Runquist, unpublished data). evolutionary processes will continue to remain at work here, ¥ At least 235 mollusk taxa, including 60% of which are generating new species through allopatric or sympatric speciation endemic (Frest and Johannes 1999). events. For instance, the spatio-temporal isolation characteristic of ¥ The most diverse herpetofauna of any similarly sized the ecoregion may be an important factor in allopatric speciations mountain range in the Pacific Northwest, including 79% of (i.e., speciation resulting from geographical isolation of all amphibian and reptile species in the Northwest (Bury populations from the parent species) as evident by the numerous 1999). narrowly restricted endemics. However, rarer sympatric speciation ¥ Among the highest diversity of dwarf mistletoe (11 taxa) in events also may occur for populations that overlap but eventually the United States (Mathiasen and Marshall 1999); develop distinct adaptations resulting in reproductive isolation and ¥ One of the greatest concentrations of ultramafic bedrock population divergence. Examples include several subspecies of geology in western North America (Coleman and Lewisia (endemics) known to hybridize in the western Siskiyous Kruckeberg 1999). and thus this species maybe in the early stages of a sympatric ¥ Exceptional beta-diversity (changes in plant communities speciation event. Notably, Coleman and Kruckeberg (1999) along environmental gradients; Whittaker [1960], Ricketts et indicated that the preponderance of neoendemics (i.e., newly al. [1999]). arisen endemics) in this ecoregion is indicative of rapid speciation ¥ High levels of fish richness (33 taxa) and endemism (42%). events. Further, they state that the ecoregion’s central position ¥ The largest complex of unprotected roadless lands along the provides refugia for certain elements of the western flora to Pacific Coast from the Baja to Canada (e.g., Siskiyou Wild migrate during changing climatic periods (Coleman and Rivers Area; Strittholt and DellaSala 2001). Kruckeberg 1999). In particular, during the dry late Tertiary climatic period, many mesic-associated species migrated into the Because of its geographic position, the Klamath-Siskiyou more mesic portions of the Klamath-Siskiyou accounting for some Ecoregion is of central significance (Whittaker 1960; Smith and of its relict species (e.g., Brewer’s spruce Picea breweriana, Sawyer 1988). The ecoregion is juxtaposed with several nearby foxtail pine Pinus balfouriana, and Kalmiopsis leachiana; ecoregions of global importance, including the Northern California Coleman and Kruckeberg 1999). Notably, the varied climate of Coastal Forests (to the west); Central Pacific Coastal Forests (to the ecoregion may be especially important as climatic refugia for the north); and Sierra Nevada Conifer Forests (to the southeast). species otherwise affected by global warming. In addition, the ecoregion is transitional to the Great Basin, The complex geology of the Klamath-Siskiyou, particularly Oregon Coast Range, Cascades Range, and California’s Central its ultramafic geology is the foundation for a botanical oasis. Valley. Mixing of plant communities from vastly different regions Ultramafic bedrock is an important mountain-forming rock has contributed to the area’s exceptional beta-diversity and this can consisting of serpentinite and peridotite rocks and soils typically be clearly seen in places like the Cascade-Siskiyou National deficient in certain minerals (calcium, nitrogen, phosphorus, and Monument (Odion and Frost 2002). potassium) yet naturally toxic to plants in others (iron, magnesium, Biodiversity Formula - the formula for the Klamath- chromium, cobalt, and nickel; Coleman and Kruckeberg [1999]).

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 3

PLENARY TALK Coleman and Kruckeberg (1999) report at least 40 plant species appreciated by land managers and decision makers as they seek to considered endemic “serpentine taxa” of the Klamath-Siskiyou, balance socio-economic interests with fish and wildlife concerns. however, this number was considered conservative. Moreover, In World Wildlife Fund’s global assessment, the DellaSala et al. (1999) indicate 11% of 154 species with state- or biodiversity of the Klamath-Siskiyou was ranked as endangered federal-listing status are confined to ultramafic areas within the based on a number of indicators of anthropogenically induced ecoregion. In fact, the largest concentration of ultramafic bedrock ecosystem degradation, as summarized by DellaSala et al. (1999): in western North America occurs along the Josephine ophiolite, which spans southwest Oregon’s Siskiyou National Forest and ¥ Low levels (12%) of protection Ð compared to other portions of northern California. Additionally, the varied climate in temperate conifer ecoregions in the United States, the the Klamath-Siskiyou contributes to major differences in Klamath-Siskiyou ecoregion has low levels of protection. environmental gradients that in turn drive patterns in species Although several large Wilderness complexes exist (e.g., distributions and hence the ecoregion’s high levels of beta- Kalmiopsis, Siskiyou, Marble Mountains, Trinity Alps, diversity. This phenomenon originally was described by Yolla Bollys), a representative network of reserves, Whittaker (1960). fundamental to conservation planning, is missing from the Many ecological processes are at work influencing species ecoregion (Noss et al. 1999). Further, the Northwest Forest distribution patterns across the Klamath-Siskiyou. Most notably, Plan is facing rollbacks in protection through weakening of fire and hydrology are major drivers of ecosystem dynamics and the Aquatic Conservation Strategy and Survey and Manage the area’s remarkable diversity. While fire history in the requirements that will further remove the ecoregion from ecoregion has not been thoroughly documented (however see having a representative reserve network. Frost and Sweeney 2000; Odion et al. 2004 for reviews), historic burning practices by Native Americans combined with frequent ¥ Extensive habitat fragmentation Ð over 30,000 miles of lightning-caused fires have been largely responsible for an roads criss-cross the ecoregion and logging has replaced incredibly varied landscape. The Silver Fire in 1987 and Biscuit biologically diverse old-growth forests with sterile-tree Fire in 2002 are but two examples of large fire events that plantations at an average annual rate of about 50,000-acres continue to shape the ecoregion’s complexity and integrity. per year since the 1970s (Staus et al. 2002). Fragmentation Likewise, hydrological processes play a key role in structuring the from road building and logging is a major problem facing composition of aquatic communities with periodic flooding and all ecoregions across the United States as well as many scouring events contributing to diverse riparian areas and healthy across the globe (Heilman et al. 2002). aquatic ecosystems. Notably, beavers (Castor canadensis) are architects of riparian areas, providing habitat such as shallow ¥ Many (154) terrestrial species with state or federal ponds for many aquatic species. This species has been on the conservation status Ðat risk species in this ecoregion are decline in much of the Northwest. largely the result of habitat fragmentation and degradation. Finally, much of the information about the ecoregion Plants by far comprise the majority of at-risk species (75%), reflects new research that has come about in the time between followed by birds (13%), herpetofauna (7%), mammals Siskiyou Ecology conferences (1997-2003). There is no question (4%), and invertebrates (1%); although this list undoubtly that as this area receives additional interest from researchers more reflects listing preferences among researchers and agencies. of its outstanding qualities will be discovered. Further, 11 fish taxa have official conservation status under federal or state programs and most mollusks have experienced extensive (>90%) range contractions (T. Frest, Is the Klamath Knot Beginning to Unravel? personal communication) ¥ Several (5) known extirpations Ð gone from the ecoregion Status, Condition, And Threats - the Klamath-Siskiyou are the grizzly bear (Ursus arctos), gray wolf (Canis lupus), Ecoregion is at a critical juncture. On the one hand, many pronghorn (Antiolocapra americana), California condor impacts largely occurring over the past four to five decades have (Gymnogyps californianus), and bighorn sheep (Ovis had a cumulative effect on the ecoregion’s integrity. But on the canadensis). These species were extirpated in the early part other hand, the ecoregion still has relatively large tracts of of the last century as settlers cleared land and eliminated unroaded lands central to its integrity (Strittholt and DellaSala competing species. 2001). While periodic natural disturbances (largely fire and floods) are the norm here, recent anthropogenic disturbances have ¥ Extensive degradation of key ecosystem processes (fire and fragmented otherwise intact terrestrial and aquatic systems, water) – over half of the ecoregion’s 877 watersheds show disrupted key ecosystem processes like fire and hydrology, extensive degradation and are in rapid decline (Staus and introduced invasive exotics, and eliminated several species. Strittholt 2001; Bredensteiner et al. 2003). Road building, Collectively, the rate and scale of human disturbances are beyond logging in the uplands, numerous barriers to fish passage the capacity of the ecosystem’s ability to self-repair (see below). (e.g., over 1,100 on the Rogue River watershed alone), Such cumulative impacts are often underestimated or under- livestock grazing, and over-allocation of instream flows and

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 4 PLENARY TALK

water quality problems contribute to poor watershed health. mapping analysis of potential “source areas” (i.e., areas of trees Fire suppression and logging contribute primarily to shifts and vegetation that survived the fire) indicate that most potential in fire intensity toward more severe fires in some portions natural seed sources are within 660 feet of severely burned areas; of the ecoregion, mainly low to mid elevation areas (Frost thereby, negating the need for massive re-planting and artificial and Sweeney 2000). seeding (Strittholt and DellaSala, in review). Finally, the Biscuit Fire recovery area has been proposed by the Forest Service as an ¥ Exotic species invasions Ð exotic species have replaced experiment in salvage logging with nearly 70% of the entire native plant communities in many places. Most notably, the volume coming from late-successional reserves and roadless area. endemic Port Orford cedar (Chamaecyparis lawsoniana) Such a grandiose experiment for the Siskiyous is ill-conceived, has been devastated by an exotic root-rot fungus particularly with the preponderance of lands (especially private (Phytophthora lateralis) that is threatening the functional lands) already undergoing salvage operations. A globally role of this cedar as a keystone species of riparian areas. significant region is no place for large-scale salvage logging Reports of Sudden Oak Death Syndrome are starting to turn experiments and instead the area should receive a combination of up in Curry County, Oregon and may pose risks if this truly restorative activities (such as plantation thinnings and destructive Phytophthora spreads. Yellow-star thistle prescribed fire, see Conservation Biology Institute 2003) and new (Centaurea solstitalis) and other exotic thistles and protected areas designations (e.g., Fire Research Natural Areas medusahead (Taeniatherium caput-medusae) are spreading where natural recovery processes can be studied; National rapidly throughout the Shasta Valley into southern Oregon, Conservation Area which is managed for biodiversity and particularly within the Cascade-Siskiyou National ecosystem recovery). Notably, DellaSala et al. (2004) indicate Monument. that post-fire early successional habitat is one of the rarest habitat types in the Pacific Northwest and such areas should be allowed ¥ Decline of endemics and globally imperiled communities - to recover with little or no intervention. More than half (56%) of the at-risk taxa are endemic and most (92%) are plants (DellaSala et al. 1999). Ten plant Conservation Challenges communities listed by the Heritage Program as G1 (globally critically imperiled) and G2 (globally imperiled) occur in the ecoregion. Many of these are on private lands. To promote sound investments in an ecologically sustainable future, the Klamath-Siskiyou Ecoregion is in need of Fire and Water Related Degradation - in addition to the “CPR”– conservation, protection, and restoration – and an above threats, pre- and post-fire management have now risen to ecoregional ethic that builds on Aldo Leopold’s (1949) land ethic. the top of the list as causes of ecosystem degradation. The The following are seven key conservation priorities key to such a proposed Biscuit Fire salvage operation and “restoration” program vision: of the Forest Service represents the largest timber sale on federal lands in nearly a century (over 370 million board feet is ¥ Achieve an ecologically representative network of reserves proposed). While the agency claims salvage logging is a Ð the Roadless Conservation Rule of 2000 was a landmark “restorative” action, and that only about 5% (29,000 acres) of the decision to protect 58.5 million acres of inventoried burn area will be entered, this proposal could alter recovery roadless lands, including over 1 million acres in the processes and the area’s unique plant communities for decades. Klamath-Siskiyou ecoregion. Roadless areas perform many Many scientists (e.g., Conservation Biology Institute 2003; vital ecological functions (DellaSala and Strittholt 2001; ECONorthwest 2003) have questioned the validity of this proposal Strittholt and DellaSala 2001) and fire regimes in these as being ecologically destructive, economically questionable, and areas are most likely to be operating within historic inconsistent with the body of literature on salvage logging impacts parameters (DellaSala and Frost 2001). When combined (e.g., Beschta et al. 1995; Minshall et al. 2003; Beschta et al. with Wilderness, roadless areas are key components to an 2004). At stake is over 12,000 acres of inventoried roadless areas ecologically representative reserve network for the and up to 57,000 acres that could be disqualified from future ecoregion (Noss et al. 1999; Strittholt and DellaSala 2001). wilderness designations. Equally alarming is the agencies’ Unfortunately, the roadless rule has been challenged by the proposal to plant over 50,000 acres with conifers or seed mixtures. Bush administration, which is failing to defend the rule in The conversion of a biologically diverse landscape that has been the courts and is proposing large-scale salvage logging in shaped by fire for centuries, to one that will be more characterized post-fire recovering landscapes (e.g., Biscuit Fire) while by biologically simplified and flammable tree plantations (Odion seeking exemptions for the Tongass and Chugach national et al. 2004) may be irreversible and should not be considered a forests in Alaska. Notably, both the California Wild “restorative” action. However, limited tree planting and seeding Heritage (www.calwild.org) and Oregon Wild may be necessary along bulldozed fire lines and roads using (www.oregonwild.org) campaigns seek to protect unroaded native species mixes characteristic of the site and planted in low, landscapes that qualify for wilderness designation under the variably spaced densities, but only after monitoring indicates lack Wilderness Act. With over 800,000 acres of potential lands of nearby seed sources (DellaSala et al. 2004). Notably, a GIS in the California portion of the ecoregion (e.g., additions to

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 5 PLENARY TALK

the Marble and Trinity Alps Wilderness) and an additional ¥ Protect remaining late-successional and old-growth habitat - 140,000 acres in southern Oregon portion (e.g., Soda old-growth forests have declined precipitously throughout Mountain, Elk River, Zane Grey, Siskiyou Crest and the Northwest and in the Klamath-Siskiyou (Heilman et al. proposed additions to the Kalmiopsis and Rogue-Umpqua 2001; Strittholt et al. in review), and legislation has been Divide wilderness areas), these areas are indeed the “crown proposed (www.nwoldgrowth.org) to protect these areas. jewels” of what remains of relatively intact systems in the These forests harbor species absent from or in lower Klamath-Siskiyou. Finally, many rural communities have numbers in plantations (see FEMAT 1993 for over 1,000 resisted protected area designations due to socio-economic species associated with late-seral forests). Many businesses concerns. However, recent economic analyses indicate that have made commitments to source wood from forest areas protected areas can have a net positive effect on rural not considered endangered (see www.ran.org), including economies by serving as a “magnet” for attracting new Home Depot and Staples, and recently Boise Corporation businesses and residents concerned about relocating to the has agreed in principle to suspend logging operations in old- area for quality of life reasons (Southwick Associates 2000). growth forests. The Forest Stewardship Council (www.fscus.org) is an organization backed by consumers, ¥ Strengthen protections for existing conservation areas such conservationists, and some timber companies because their as the Cascade-Siskiyou National Monument Ð this nearly management standards appropriately balance ecological, 53,000-acre national monument is the only high elevation economic, and social interests in forests while avoiding land bridge joining the globally outstanding Siskiyous to the logging in endangered forests. Such measures show Cascades. It harbors one of the most diverse (115 species Ð promise for shifting logging away from mature and old- recent surveys by E. Runquist) butterfly populations in the growth forests and toward more responsible forestry western U.S. and contains up to 12 endemic mollusks practices. (recent surveys by T. Frest, personal communication). The monument’s land-bridge function, however, is threatened by ¥ Promote watershed conservation initiatives of local widespread logging on private lands (see Odion and Frost communities Ð efforts to restore anadromous fish are 2002) that warrant conservation-incentives (e.g., Forest picking up steam in local communities, many of which have Stewardship Certification of private lands management; fee organized as watershed councils (in Oregon). Watershed title acquisitions and conservation easements from willing councils show promise as “bottom-up” community-led landowners). In addition, livestock grazing has been initiatives that largely address instream conservation. recognized as a potential threat to the area’s unique plant However, for such efforts to maximize recovery of and wildlife species, requiring well-designed field studies to salmonids and other aquatic organisms they must address assess impacts and develop appropriate mitigation strategies upland management practices also influencing watershed (Odion and Frost 2002). integrity. In addition, watershed restoration efforts need to be multi-scaled, including development of basin-wide • Protect “hot spot” areas such as the Siskiyou Wild Rivers recovery and prioritization strategies to achieve maximum Area Ð the 1.1-million acre Siskiyou Wild Rivers Area was recovery benefits (see Staus and Strittholt 2001; proposed as a national monument in 2000. This area Bredensteiner et al. 2003). received acknowledgments for its incredible biodiversity from Pulitzer Prize winner Edward O. Wilson and former ¥ Encourage responsible forestry and fish-friendly farming Interior Secretary Bruce Babbitt. The Siskiyou Wild Rivers practices Ð in many ways, conservation in this ecoregion area is: (1) a “hot spot” of biodiversity and evolutionary needs to embrace the social and economic interests of processes in the ecoregion, (2) contains the highest people, particularly the incredibly socially diverse rural concentration of rare plants of any national forest in the communities, in order to achieve a sustainable future. Two nation (Strittholt et al. 1999), (3) the best wild salmon programs that show much promise in addressing rural fishery along the Pacific Coast, and (4) largest complex of community needs while providing habitat for fish and unprotected roadless areas along the coast from the Baja to wildlife are Salmon Safe (www.salmonsafe.org) and the Canada (www.siskiyouproject.org). In 2000, the Forest Forest Stewardship Council (www.fscus.org) certification Service, under an Executive Order, withdrew new mining programs. Both programs rely on transparent, verifiable claims so that the incoming administration could study it for standards to define stewardship activities while providing future protections. Unfortunately, the Bush administration access to “green markets” where consumers can make a has lifted the mining moratorium and now is proposing conservation difference at the point of purchase. Such large-scale salvage logging in response to the Biscuit fire. programs, however, continue to suffer from an inability to Clearly, an area considered a “cradle of evolution” for the gain market access due to low consumer demand and ecoregion, warrants increased protections especially limited interest on the part of landowners. Considerable following a major fire event like the Biscuit. efforts at national, ecoregional, and local levels will be necessary to create additional exposure for the pioneering

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 6

PLENARY TALK efforts of responsible landowners so that they and others is hope for a sustainable future. A vision of wild rivers packed can eventually enjoy the market place benefits. with spawning wild salmon, restored species, recovered ecosystems, abundant ancient forests, untrammeled solitude, ¥ Restore degraded lands and watersheds and link such the “birthing” of new species, and a truly sustainable society projects to a restoration-based economy Ð restoration has are but a commitment away. Science has a definitive role to the potential to act as an “olive branch” for bringing play in explaining what is at stake, but an ecoregional ethic needs together diverse interests concerned about public lands to emerge and be embraced by society if we are truly going to (Williams et al. 1997). However, for approaches to be create a more sustainable future for both its current occupants and effective ecologically and socially they need to weave social our children’s children. and economic desires within the appropriate ecological framework (see DellaSala et al. 2003). There is much Acknowledgements and Author’s End Note promise in the field of restoration and its ability to proactively “heal” the land, however, these efforts must begin with an emphasis on ecological integrity and an This manuscript benefited from reviews provided by John understanding that economic benefits are a by-product of Sawyer (Humboldt State University) and Jack Williams (Southern restoration projects and jobs in restoration are sound Oregon University). The website addresses reflected in the investments in an ecologically sustainable future. The references were active as of the date of this publication. It is Lomakatsi Project in Williams, Oregon, for example, prides hoped that the concepts and overview provided by this keynote itself on sound principles of ecosystem restoration while address will inspire others to communicate the conservation employing local people in fuels reduction and vegetation importance of this extraordinary ecoregion to decision-makers, management. Other restoration projects include an land managers, and the general public while there is still time for exemplary partnership between Redwood National and State change. Parks and Save-The-Redwoods League in northern California designed to restore ecological integrity to large Literature Cited landscapes (e.g., Mill Creek, California) within the redwood portion of the Klamath-Siskiyou Ecoregion. Abell, R.A., D.M. Olson, E. Dinerstein, P.T. Hurley, J.T. Diggs, W. Eichbaum, S. Walters, W. Wettengel, T. Allnutt, C.J. Conclusions Loucks, and P. Hedao. 2000. Freshwater ecoregions of North America: a conservation assessment. Island Press, Washington, D.C. Whittaker’s concerns about ecosystem degradation echoed Bredensteiner, K., K. Palacios, and J.R. Strittholt. 2003. in the 1960s are even more relevant today. If current land-use Assessment of aquatic habitat monitoring data in the Rogue trends continue, the ecoregion’s integrity will be irrevocably River Basin and Southern Oregon coastal streams. compromised at the expense of both ecology and economy. Only Unpublished Report (www.consbio.org). through a sound investment in conservation and stewardship that Beschta, R.L., C.A. Frissell, R. Gresswell, R. Hauer, J.R. Karr, W. begins with the recognition of global importance of place that is Minshall, D.A. Perry, and J.J. Rhodes. 1995. Wildlife and put into practice at the local, watershed, and ecoregional levels salvage logging: recommendations for ecologically sound can we begin to turn these alarming trends around. In addition, post-fire salvage management and other post-fire treatments we must examine our actions and impacts across generational time on federal lands in the West. Unpublished Report to the lines Ð as measured by the Seven Generations Concept of Native Pacific Rivers Council. Americans. Beschta, R.L., J.J. Rhodes, J.B. Kauffman, R.E. Gresswell, G.W. The Klamath-Siskiyou Ecoregion is one of remarkable Minshall, C.A. Frissell, D.A. Perry, R. Hauer, and J.R. Karr. superlatives originally explored by the 19th century botanists, 2004. Postfire management on forested public lands of the embraced by early ecologists, and now marveled by many. It is western USA. Conservation Biology 18: in press. an ecoregion whose unique biological diversity lies concealed Bury, R.B. 1999. Klamath-Siskiyou herpetofauna: biogeographic behind unusual rocks and soils, and hidden in its contorted patterns and conservation strategies. Natural Areas Journal mountains where the often difficult to find, un-charismatic species 19(4):341-350. reside. While the ecoregion is a “cradle of evolution,” it is on an Coleman, R.G., and A.R. Kruckeberg. 1999. Geology and plant unsustainable trajectory toward continued degradation and species life of the Klamath-Siskiyou mountain region. Natural Areas extirpations. In Aldo Leopold’s (1949) historic treatise of a land Journal 19(4):320-340. ethic, he lamented about how well or poorly we treat the land is Conservation Biology Institute. 2003. Ecological issues evident by the scars we leave behind. Humanity’s scars on this underlying proposals to conduct salvage logging in areas ecoregion are many; they unmistakably show up on landsat burned by the Biscuit Fire. Unpublished Report, images and aerial photos or simply from above through the Conservation Biology Institute (www.consbio.org). window of an airplane. While the clock is ticking on this DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt, and D.M. ecoregion, and the ecological stakes have never been greater, there Olson. 1999. A global perspective on the biodiversity of the

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 7

PLENARY TALK

Klamath-Siskiyou ecoregion. Natural Areas Journal Research Laboratory. Map at 1:7,500,000. 19(4):300-319. Odion, D.C., and E. Frost (eds). 2002. Protecting objects of DellaSala, D.A., and J.R. Strittholt. 2000. Scientific basis for scientific interest in the Cascade-Siskiyou National roadless area conservation Ð review and management Monument: status, threats, and management recommendations. World Wildlife Fund publication recommendations. Unpublished Report, World Wildlife submitted to the Roadless Area Conservation team, U.S. Fund (www.worldwildlife.org/klamathsiskiyou.) Forest Service. Odion, D.C., J.R. Strittholt, H. Jiang, E. Frost, and D.A. DellaSala. DellaSala, D.A., and E. Frost. 2001. An ecologically based 2004. Fire severity patterns and forest management in the strategy for fire and fuels management in National Forest Klamath National Forest, northwest California, USA. roadless areas. Fire Management Today 61(2):12-23. Conservation Biology 18:in press. DellaSala, D.A., A. Martin, R. Spivak, T. Schulke, B. Bird, M. Smith, J.P., and J.O. Sawyer, Jr. 1988. Endemic vascular plants of Criley, C. Vvan Daalen, J. Kreilick, R. Brown, and G. Aplet. northwestern California and southwestern Oregon. Madrono 2003. A citizens’ call for ecological forest restoration: forest 35:54-69. restoration principles and criteria. Ecological Restoration Southwick Associates. 2000. Historic economic performances of 21(1):14-23. Oregon and Western counties associated with roadless and DellaSala, D.A., J.E. Williams, C. Deacon-Williams, and J.R. wilderness areas. Unpublished Report prepared for World Franklin. 2004. Beyond smoke and mirrors: a synthesis of Wildlife Fund and the Oregon Natural Resources Council forest science and policy. Conservation Biology 18:in press. (www.worldwildlife.org/forests/attachments/report_aug15.pdf) ECONorthwest. 2003. Economic issues underlying proposals to Staus, N.L., and J.R. Strittholt. 2001. Conservation planning for conduct salvage logging in areas burned by the Biscuit Fire. aquatic biological integrity in the Klamath-Siskiyou Unpublished Report ECONorthwest, Eugene, Oregon. ecoregion using multiple spatial scales. Unpublished Report FEMAT (Forest Ecosystem Management Assessment Team). to the World Wildlife Fund (www.consbio.org). 1993. Forest ecosystem management: an ecological, Staus, N.L., J.R. Strittholt, D.A. DellaSala, and R. Robinson. economic, and social assessment. 1993-793-071. U.S. 2002. Rate and pattern of forest disturbance in the Klamath- Government Printing Office, Washington, D.C. Siskiyou ecoregion, U.S.A. Landscape Ecology 17:455-470. Frest, T.J., and E. J. Johannes. 1999. Mollusk surveys of Strittholt, J.R., R.F. Noss, P.A. Frost, K. Vance-Borland, C. Carroll, southwestern Oregon, with emphasis on the Rogue and and G. Heilman, Jr. 1999. A conservation assessment and Umpqua River drainages. Unpublished Report. Deixis science-based plan for the Klamath-Siskiyou ecoregion. Consultants, Seattle, Washington. Unpublished Report to the Siskiyou Project Frost, E., and R. Sweeney. 2000. Fire regimes, fire history, and (www.consbio.org). forest conditions in the Klamath-Siskiyou region: an Strittholt, J.R., and D.A. DellaSala. 2001. Importance of roadless overview and synthesis of knowledge. Unpublished Report areas in biodiversity conservation in forested ecosystems: a to World Wildlife Fund case study Ð Klamath-Siskiyou ecoregion, U.S.A. (www.worldwildlife.org/klamathsiskiyou). Conservation Biology 15(6):1742-1754. Heilman, G.E. Jr., J.R. Strittholt, N.C. Slosser, and D.A. DellaSala. Ricketts, T.H., E. Dinerstein, D.M. Olson, C.J. Loucks, W. 2002. Forest fragmentation of the conterminous United Eichbaum, D. DellaSala, K. Kavanagh, P. Hedao, P.T. States: assessing forest intactness through road density and Hurley, K.M. Carney, R. Abell, and S. Walters. 1999. spatial characteristics. BioScience 52(5):411-422. Terrestrial ecoregions of North America: a conservation Jules, E.S., D.A. DellaSala, and J.K. Marsden. 1999. The assessment. Island Press, Washington, D.C. Klamath-Siskiyou region: introduction to theme. Natural Vance-Borland, K.R., R.F. Noss, J. Strittholt, P. Forst, C. Carroll, Areas Journal 19(4):295-297. and R. Nawa. 1995. A biodiversity conservation plan for Leopold, A. 1949. A sand county almanac: and sketches here and the Klamath/Siskiyou region. Wild Earth 5(4):52-59. there. Oxford University Press, New York. Wagner, D.H. 1997. Klamath-Siskiyou region, California and Mathiasen, R., and K. Marshall. 1999. Dwarf mistletoe diversity in Oregon, U.S.A. Pages 74-76 in S.D. Davis, V.H. Heywood, the Siskiyou-Klamath mountain region. Natural Areas O. Herrera-MacBryde, J. Willa-Lobos, and A.C. Hamilton Journal 19(4):379-385. (eds). Centres of plant diversity. Vol. 3: the Americas. World Minshall, G.W. 2003. Response of stream benthic Wide Fund for Nature and IUCN. Information Press, macroinvertebrates to fire. Forest Ecology and Management Oxford, England. 178:155-161. Wallace, D.R. 1983. The Klamath Knot. Sierra Club Books, San Noss, R.F., J.R. Strittholt, K. Vance-Borland, C. Carroll, and P. Francisco, California. Frost. 1999. A conservation plan for the Klamath-Siskiyou Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, ecoregion. Natural Areas Journal 19(4):392-411. Oregon and California. Ecological Monographs 30:279-238. Omernick, J.M. 1995. Level III ecoregions of the continental Williams, J.E., C.A. Wood, and M.P. Dombeck (eds.). 1997. United States. U.S. Department of Agriculture, Forest Watershed restoration: principles and practices. American Protection Agency, National Health and Environment Effects Fisheries Society, Bethesda, Maryland.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 8 PLENARY TALK

Figure 1. Map of the Klamath-Siskiyou Ecoregion of northwestern California and southwestern Oregon.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 9 PHYTOPHTHORA DISEASE ECOLOGY

1 PHYTOPHTHORA IN THE WORLD’S FORESTS

Everett Hansen Department of Botany & Plant Pathology, Oregon State University, Corvallis, OR 97331 Email: [email protected]

IIntroduction and before long tree seedlings were succumbing to some of the same diseases that they had struggled with, including those caused by Phytophthora species. Millions of trees died in the nurseries he genus Phytophthora (Phy-toph-thor-a) is a group of before we figured out what was going on, and that it could be aquatic “fungi,” classified as Oomycetes, that are actually prevented by careful management of water and soil. Of even more closely related to the kelps and brown algae than to TT greater concern, however, were the many seedlings that were mushrooms, bread molds, and other true Fungi. Despite its infected but not killed before being transplanted to forest sites. unlikely relatives, Phytophthora acts like a fungus in many ways, Were we creating a new epidemic in our forests? It took a couple and we commonly call the 70 or so species “water molds.” While of years of research, but we were able to conclude that, no, the Phytophthoras are best known as agricultural pathogens (the name nursery Phytophthora species didn’t survive under forest means “plant destroyer”) some species are also forest denizens, capable of ecological as well as economic destruction. Residents conditions. The weakened trees often died, but the pathogens of southwest Oregon are all too familiar with the Port-Orford- didn’t spread in forest soils. cedar root rot pathogen, Phytophthora lateralis. This is not to say that there aren’t Phytophthora species that In this paper, I want to introduce you to forest are very successful in forests, and lots of them. Many are species Phytophthoras, here and around the world. I hope to give you that we didn’t even suspect existed until quite recently. Of the 15 some sense of why pathologists and ecologists are so concerned species we know to be present in Oregon forests, six are new or about the present and potential impacts of this group of plant still undescribed. Only three are associated with dramatic disease pathogens on forest ecosystems, and open a discussion on ways to in Oregon forests. The ecological roles of the others are very combat the rapid ecological changes that are being triggered by poorly understood. Perhaps they are nibbling on roots or leaves invasive pathogens. without causing dramatic disease. By contrast, the three that do kill large numbers of trees are each exotic, invasive, and changing our forests. The Phytophthora Way of Life Alien Invasion Phytophthoras are versatile pathogens. They do require water, but that might be waterlogged soil, a muddy road, or just a wet leaf. Most infection results from swimming zoospores that Phytophthora lateralis is the best known of these alien sense susceptible plant tissues and home in on them. Most of the invaders to Oregonians, because it has been around the longest. It body of these Oomycetes develops as microscopic threads within first showed up in Coos Bay about 1950, and has since been the plant, exuding toxins to kill cells, and absorbing the released spread throughout the natural range of Port-Orford-cedar (POC) in plant nutrients. When plant tissue is exhausted, or when it gets Oregon and California. It is a very aggressive pathogen on POC, too hot or dry for further growth, some species can survive as killing trees of all ages. Other papers in these Proceedings resting spores. These thick-walled spores in turn may be carried by describe it, the damage it causes, and the disease management people, animals, or vehicles for long distances in soil or plant program in detail, and I won’t repeat. Phytophthora ramorum, the material. Some Phytophthora species do these things in leaves sudden oak death pathogen, has also been described by others. and stems instead of roots, and are transported by rain splash and We don’t know where this one came from, but it arrived in Oregon wind. Both P. infestans, the first species described and the cause less than five years ago, and despite our best efforts, we have not of potato late blight and human famine in 19th century Ireland, been able to eradicate it, although the effort continues and has at and P. ramorum, one of the newest species and cause of sudden least succeeded in greatly slowing the spread. oak death, are aerial pathogens. Phytophthora cambivora is the third species killing trees in My introduction to Phytophthora came in the forest tree Oregon forests. This one is a real puzzle. The pathogen is well nurseries that grow the seedlings that are planted to reforest areas known in European forests where it causes root rot and basal harvested or burned in the Northwest. In the 1970s, forest cankers on chestnut and several other hardwood tree species. It harvests accelerated in Oregon, followed by increased demand for has been in Oregon orchards and agriculture for many years, but seedlings. New nurseries were established, often on farm land, we only recently found it in the forest, and by accident. We isolated it from soil in several places, apparently not causing any 1 Plenary paper for the Phytophthora disease ecology section. disease. But three or four years ago it started killing golden

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 10 PHYTOPHTHORA DISEASE ECOLOGY chinquapins in the southern and central Oregon Cascades. It is bringing together species that had never met in nature? And what very aggressive on chinquapin, causing lethal cankers much like P. more efficient means of distribution of their uniquely pathogenic lateralis on POC. Mortality was first noticed after several offspring than on nursery plants sold around the world. Despite unusually wet winters, but this is the only clue we have so far for the demonstrated susceptibility of North American alder species to its dramatic appearance. this new pathogen, and the key roles they play in riparian Phytophthora cinnamomi is the most notorious forest ecosystems, there is no special effort to block its introduction Phytophthora, in other parts of the world. It is present in at least here. some Oregon forests, but so far, not causing any real damage. Europe also offers the best documented illustration of This pathogen attacks literally thousands of plant species, and is a indigenous Phytophthora species in forests. In contrast to the chronic problem in horticultural nurseries in the Northwest. We havoc sometimes wrought by invading species, those that evolved think that Oregon forest soils are generally too dry in the summer, within an ecosystem generally are much better mannered. Where and too cold in the winter for this beast. In the southeastern host plants and pathogens have evolved together, the interactions United States, by contrast, it killed the American chestnut from are generally characterized by complex genetic checks and the southern part of its range 100 years before the infamous balances between resistance and virulence, acting within the chestnut blight was introduced from the north, and continues to environmental constraints of the particular ecosystem. In cause chronic disease problems, notably littleleaf disease of some European oak forests, at least 13 new Phytophthora species have southern pines, on abused soils with poor drainage. In Australia, it recently been found. No one knew they were there because the has established itself as one of the five worst environmental diseases they cause are very subtle. Roots and leaves are killed, threats on the continent. and sometimes whole trees, but not populations of trees. An Phytophthora cinnamomi is now widespread in Australia, unexplained disease called “oak decline” has occasionally plagued but causes dramatic disease mainly in the south and west of the European forests for centuries. No simple cause has been found, country, usually on seasonally dry sites where much of the but it now appears that some or all of these newly discovered vegetation is susceptible. In the jarrah eucalyptus forests of West forest Phytophthora species may be players. They are for the Australia, it has been carried by vehicles, especially in exploration most part root nibblers, not tree killers, and a healthy tree replaces for bauxite mining, and now spreads away from the roads and its fine feeder roots every year anyway. Perhaps in very wet tracks in a slowly advancing wave, transforming a complex and years, however, the pathogens can take a bigger bite, and if the very diverse forest community into grassy scrub. The heathlands trees are coincidentally stressed by subsequent drought or insect of far southern West Australia have some of the highest levels of defoliation, perhaps the trees cannot recover so quickly. It is an plant diversity recorded anywhere in the world, and most of the uncertain scenario, but supported by increasing evidence and the species, including many rare endemic species, are susceptible to P. lack of satisfying alternative explanations. cinnamomi. The spreading pathogen is being combated in some of the most sensitive areas with aerial applications of fungicides. Change The chemical used is phosphonate, essentially a phosphorous fertilizer, that shows almost no non-target effects. This then is the Jekyll and Hyde picture of Phytophthora in Phytophthora cinnamomi is also introduced and destructive world forests. On the one hand there is a poorly understood in the Ohia forests of Hawaii and cork oak forests of the community of indigenous species that nibble away at roots or Mediterranean. It causes a very different disease in oak forests of leaves, occasionally being so bold as to kill a tree here or there. southern France. Northern red oak, native to the eastern US, has They would be greedier, but are kept in check by the physical been widely planted as a timber tree. The pathogen infects roots, environment of the forest system where they evolved and by and kills strips of bark up to the lower main stem. In cold winters, genetic resistance in the plant community they occupy. The trees however, it is killed from the above ground parts of the oaks, and would be rid of the pathogens entirely, through increased the wounds callus over. There are often repeated infections resistance, but the pathogens change in turn, and the standoff through the years, each contained by the tree and eliminated by continues. On the other hand, we suffer ecosystem changing winter. On good sites, there is no measurable impact on tree epidemics from Phytophthora species, when the rules of détente growth. are altered. In a rapidly changing world, destructive epidemics Another new Phytophthora disease of current concern in seem to be increasingly frequent. The examples mentioned here Europe attacks alder trees along rivers and canals. P. alni spreads illustrate epidemics triggered by: readily down the waterways, and across low-lying forests and plantations in flood waters. Its sudden appearance across the ¥ Invasion of new plant communities that lack resistance. continent suggests some additional means of transport, however. ¥ Global warming and changes in rainfall patterns. It appears that it has been spread widely from forest tree nurseries. ¥ New, human assisted, means of local dispersal. And how did it get into the nurseries? Perhaps it evolved there! Phytophthoras evolved to go downhill with water- people P. alni is a recent hybrid species between P. cambivora, well are carrying them uphill. established in tree nurseries but not pathogenic to alder, and some ¥ New genetic capability from hybridization. as yet undiscovered parent. What better place than a nursery for

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 11 PHYTOPHTHORA DISEASE ECOLOGY

Phytophthora cinnamomi thrives in warm temperate regions well prove futile? After all, change is the pattern of life on earth. with susceptible plants. In cooler areas it is blocked by winter Much of Oregon was tropical swamp forest 15 million years ago, cold or summer drought or both. But as climates change, so does and the continents continue to drift. Some native pathogens, such the distribution of pathogens. In Europe, they think that they are as the laminated root rot fungus, kill trees just as effectively as P. seeing the first northward probes of P. cinnamomi, and are lateralis, and have an even greater impact on forest structure and alarmed. diversity. It isn’t dramatic changes in plant species composition International trade is not new; it spread P. cinnamomi around that are new, or the reality that fungi kill trees and change the the world 250 years ago from its ancestral home in the islands of forest. SE Asia. Movement of people and goods is accelerating, The scale and rate of change is perhaps unprecedented, however, and so is the movement of potential invasive organisms. however, and this time, it is the consequence of human We don’t know where P. lateralis or P. ramorum came from, but thoughtlessness and a devaluing of the natural world, not a natural both appeared first in horticultural nurseries. Long distance process. Clearly people are part of the earth’s ecosystem, not transport of nursery plants is a particularly dangerous pathway for alien to it. People need food, shelter, air, and water from the invasion. ecosystem, as well as solitude and spiritual sustenance. We also Phytophthora alni, the new hybrid species that is killing value the past—but which past? Just what is our “desired future streamside alders in Europe, is a human creation. It wasn’t condition,” in a changing world? deliberate, but when you bring sexual organisms together in Human needs and values from nature need not be mutually unsupervised places, well, things happen. Again, the suspicion exclusive, although it may well be impossible to have all things on goes to the plant nurseries and the trade in ornamental plants. every acre. Consensus on the allocation of nature’s resources is Phytophthora lateralis didn’t move through the forest on not in sight. But perhaps we can agree that alien invasions nursery stock, however. It was carried on truck wheels and grader threaten all of the values. blades, along the new road network that was built along the ridge tops to access the timber of the Siskiyous. We carry it uphill; Suggested Reading gravity takes care of the cedar growing downslope, or downstream. Davidson, J.M., S. Werres, M. Garbelotto, E.M. Hansen, and “Fight Them on the Beaches, or Let the New D.M. Rizzo. 2003. Sudden oak death and associated Order Begin.” diseases caused by Phytophthora ramorum. Plant Health Progress information online at: http://www.plantmanage - Hal Mooney mentnetwork.org/pub/php/diagnosticguide/2003/sod/ Hansen, E.M., and C. Delatour. 1999. Phytophthora species in Professor Hal Moony, an ecologist at Stanford University, is a oak forests of north-east France. Annals of Forest Forest leader in the international effort to stop the transport of invasive Science 56:539-547. organisms around the globe. It isn’t just pathogenic fungi, but Jung, T., E.M. Hansen, L. Winton, W. Osswald, and C. Delatour. weeds, like Scotch broom and Himalayan blackberry, and alien 2002. Three new species of Phytophthora from European animals, including worms and snails and tree-killing insects. Our oak forests. Mycological Research 106:397-411. natural world is being homogenized by invaders. It is no longer a McComb, J.A., G.E. StJ Hardy and I.C. Tommerup (eds). 2001. question of restoring the “natural” condition; too much has Phytophthora in forests and natural ecosystems. 2nd changed already to reclaim the past. But if we can find the will, International Union of Forestry Research Organizations we can stop, or at least slow the invasion and preserve a greater Working Party 7.02.09 Meeting. Murdoch University Print. portion of our native wild diversity. Albany, Western Australia. But why invest the time and expense into an effort that may

Illustration by Bob Cremins

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 12 PHYTOPHTHORA DISEASE ECOLOGY

ASSESSING THE LANDSCAPE SPREAD OF THE FATAL PORT ORFORD CEDAR ROOT DISEASE

Erik S. Jules Department of Biological Sciences, Humboldt State University, Arcata, CA 95521 Email: [email protected]

Matthew J. Kauffman Division of Biological Sciences, The University of Montana, 32 Campus Drive, Missoula, MT 59812 Email: [email protected]

Allyson L. Carroll Department of Biological Sciences, Humboldt State University, Arcata, CA 95521 Email: [email protected]

William D. Ritts Department of Forest Science, Oregon State University, Corvallis, OR 97331 Email: [email protected]

Abstract

Phytophthora lateralis is a fatal non-native pathogen that began infecting Port Orford cedar (Chameacyparis lawsoniana) stands in Oregon in 1952. Since then, the pathogen has spread to most regions within the cedar’s natural range. Spread is accomplished by infected organic material being carried by vehicles traveling along roads and by foot traffic traveling overland. Aerial dispersal of P. lateralis is exceptionally rare and does not occur in our study area. We reconstructed the history of P. lateralis spread across a 37 km2 area in southwestern Oregon and northern California. We used our data to (1) assess the factors that govern the likelihood of infection for a given population, (2) compare the distances that vehicles and foot traffic carry spores, and (3) evaluate past surveys of the same study area by the U.S. Forest Service. Our results show that three factors can increase the risk of infection at a site: greater amounts of water, higher densities of the target host (i.e., cedars), and shorter distances from the road surface to the first potentially infected host. Our results also show that vehicles disperse the pathogen over much longer distances than foot traffic. Finally, our results also show that the U.S. Forest Service survey grossly underestimated the number of P. lateralis infections in our study area. We discuss some of the management implications of our work.

Introduction Coos Bay, Oregon. Spread of the disease occurs via spores that are transported by vehicles along roads, by foot traffic such as the hoofed feet of elk, and by passive dispersal downhill in flowing ne of the most pressing conservation issues facing the water (Hansen et al. 2000). Movement of the disease can also Klamath Region is the continuous influx of invasive, occur between adjacent trees when roots are grafted. Though less OOnon-native species. Non-native species can have both common, infections of Pacific yew (Taxus brevifolia) by P. profound ecological impacts and can result in enormous economic lateralis can be found in areas of infected Port Orford cedar costs (Drake et al. 1989; Pimentel et al. 2000). Perhaps the most (DeNitto, G.A. and J.T. Kliejunas 1991). striking and damaging invasion has been that of the deadly water Port Orford cedar is a long-lived coniferous tree endemic to mold, Phytophthora lateralis, which has been spreading for over southwestern Oregon and northern California (Figure 1). It 50 years across the range of its primary host, Port Orford cedar inhabits a wide range of parent materials within this range, (Chamaecyparis lawsoniana; Sinkiewicz and Jules in press). including quartz diorites, metavolcanics, sedimentary materials, Phytophthora lateralis is a pathogenic organism that infects and ultramafics (including serpentinite, dunite, and peridotite; the roots of Port Orford cedar and causes mortality within weeks Zobel 1985). It’s distribution, however, is limited to areas of low to several years. In 1923, it was first discovered infecting Port water stress, and in some cases, especially dryer inland sites, the Orford cedar nurseries near Seattle, Washington, well outside the cedar can be restricted to areas directly adjacent to creeks (Zobel cedar’s natural range. Two decades later, in 1952, the pathogen and Hawk 1980; Zobel et al. 1985). was first noted within the cedar’s range, near the coastal town of Port Orford cedar is thought to play a critical role in

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 13 PHYTOPHTHORA DISEASE ECOLOGY structuring many riparian areas. Larger cedars form robust root hazards regression (SAS procedure PHREG; Cantor 1997). In this systems that act to stabilize banks, and fallen cedars create pools analysis, we obtained a survival function that describes the important for salmon. The wood of Port Orford cedar is probability of a cedar population escaping infection for at least t impressively rot-resistant (Jules et al. 2002), thus pools formed by years based on field collected population attributes. Attributes Port Orford cedar are likely to be long-lived. On ultramafic sites, used as explanatory variables included elevation, slope, solar the cedar is often the only tree found directly within the riparian radiation, catchment area, host abundance, distance to the nearest area. infection, and road crossing length. The purpose of the study presented here was to assess the We also obtained records of USFS surveys of the Page factors that are controlling the spread of P. lateralis to new, Mountain study area and assessed their accuracy. In this analysis, uninfected sites across a forested landscape. To do this, we we consider our observations of the study area to be a complete reconstructed the historical spread of the pathogen across four census, as we visited all possible infection sites. We then watersheds that contain Port Orford cedar and asked the following compared the number of infections found in our study to those questions: (1) was there a difference between sites that became recorded in the USFS survey. It should be noted that the USFS infected vs. sites that did not?; (2) were most new infections due survey was never intended to be as comprehensive as our census, to the movement of vehicles or foot traffic?; and (3) how but a comparison of the two studies was made to assess the accurately have past U.S. Forest Service (USFS) surveys of P. available datasets found for other areas across the cedar’s range. lateralis reflected the actual extent of the disease?

Methods Results

We evaluated the spread of P. lateralis in a 37-km2 study Of the 63 km of mapped creeks containing Port Orford area that includes four watersheds within the Siskiyou National cedar, P. lateralis infection was found along 29 km (46%; Figure Forest (Figure 2), which surround Page Mountain. The Page 2). Twenty six of 36 (72%) independent infection events began at Mountain study area is dominated by mixed conifer forest where locations where roads cross creeks. Forty three percent of the 86 the most common trees are Douglas fir (Pseudotsuga menziesii), road crossings containing Port Orford cedar (Road infections) incense cedar (Calocedrus decurrens), tanoak (Lithocarpus were infected by 1999. Only 10 of the 65 creeks without road densiflorus), and Pacific madrone (Arbutus menziesii). Most of crossings (Non-Road infections) were infected. Even though the roads present in the study area were built to aid in timber Non-Road infections were not in direct contact with any roads, all harvesting, the most common land management practice employed of these infections had a road somewhere above them, such that in the area. only creeks in headwater positions were uniformly uninfected In the summers of 1998 and 1999 we inventoried and (Figure 2). mapped all creeks within the Page Mountain study area for the The average minimum distance that infections moved via presence of Port Orford cedar and P. lateralis. Infected sites were vehicle traffic along roads since 1977 was 758 +193 m (n = 23 categorized as either initiating from vehicles along roads (i.e., infections), while the average minimum distance that infections infections that clearly start at points where roads cross creeks) or moved by foot traffic was 168 +24 m (n = 9 infections). The initiating from foot traffic (i.e., those starting a points without difference between the distance P. lateralis moved via vehicular roads directly above the infection). For all possible sites of new and foot traffic was significant (t = -3.03, df =23, P = 0.003). infection initiated by vehicles traveling roads, we measured Our dendrochronology study showed that the study area was several key factors that might influence the risk of infection: uninfected in 1976 and that the first infection occurred in 1977 at distance to nearest tree, host abundance, road crossing length (a a road crossing. The infection continued to spread along the road measure of degree of contact with infectable area), water flow system and it was not until 1984 that the first Non-Road infection (measured as “catchment area” drained above road crossing), occurred. The proportional hazards regression indicated that for elevations, slope, and solar radiation (Figure 3). creeks crossed by roads the likelihood of infection increased with In addition, we determined the date each new infection catchment area, host abundance, and proximity of the first cedar to initiated by using cross-dating techniques developed in the field of the road (Tables 1 and 2). Host abundance and distance to nearest dendrochronology (Stokes 1968). At the top of each infected site tree were correlated (Pearsons correlation; rho = -0.317, P = (i.e., the initiation of each infection) we cored six dead trees and 0.006). However, we tested the effect of each variable on a determined their date of infection. Detailed methods are described reduced model which included the effect of the other correlated in Jules et al. (2002). This approach was used to determine the variable, indicating that both measures are significant predictors of shortest possible distance that the pathogen dispersed in each year infection risk. the infection progressed across the Page Mountain study area. We USFS surveys of the Page Mountain study area revealed assessed the differences between disease movement by vehicles 48% (15 of the 31) of the infections known from our study to be and foot traffic using a t test. present at road crossings in 1995 (Figure 5). One of seven (14%) To assess if differences between the attributes of infected Non-Road infections were found in the USFS surveys, and 3 and uninfected sites could be detected, we used proportional uninfected locations were misidentified as infected. In total, 58%

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 14 PHYTOPHTHORA DISEASE ECOLOGY

(22 out of 38) of the P. lateralis infections were not detected in USFS surveys of P. lateralis distributions may underestimate the the USFS survey. true extent of this important disease invasion. In our study region, the USFS surveys revealed less than half of the total infections. Discussion While the USFS surveys were certainly not intended to be as comprehensive as our own, it is critical to acknowledge that Phytophthora lateralis has spread within the Page Mountain management strategies based on USFS surveys may be severely study area since 1977 through a combination of vectors, including underestimating the amount of inoculum across the landscape. vehicles and foot traffic. Vehicles have been responsible for the majority of new infections (72%) and they have spread the Literature Cited infections longer distances than foot traffic. The minimum distance the infection moved via vehicles was 4.5 times greater Cantor, A. 1997. Extending SAS survival analysis techniques for than the minimum distance infection moved by foot traffic. medical research. SAS Institute Inc., Cary, North Carolina. Infections were clearly tied to the road system; even the infections DeNitto, G.A., and J.T. Kliejunas. 1991. First report of that started by foot traffic were in roaded areas. No infections Phytophthora lateralis on Pacific yew. Plant Disease occurred in headwater areas that were above the road system. 75:968. Characteristics of each road crossing appear to control the Drake, J.A., H.A. Mooney, F. Di Castri, R.H. Groves, F.J. risk of infection. Road crossings that (1) experience greater Krueger, M. Rejmanke, and M. Williamson. 1989. amounts of water flow, (2) contain larger numbers of cedars Biological invasions: a global perspective. John Wiley and (potential targets), and (3) have short distances from the road Sons, Chichester, United Kingdom. surface to the first potential host tree were more likely to be Hansen, E.M., D.J. Goheen, E.S. Jules, and B. Ullian. 2000. infected over the 23 year period. These results fit well with what Managing Port Orford cedar and the introduced pathogen, is known about the biology of the pathogen. Phytophthora Phytophthora lateralis. Plant Disease 84:4-10. lateralis requires moist conditions for survival and successful Jules, E.S., M.J. Kaufmann, W. Ritts, and A.L. Carroll. 2002. infection of its host, thus the amount of water flowing at the Spread of an invasive pathogen over a variable landscape: a crossing would seem to strongly enhance infection risk. As well, non-native root rot on Port Orford cedar. Ecology 83:3167- the number of hosts within close proximity to the source of spores 3181. (vehicles) should be linked to risk, because spores are not as likely Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. to survive long distances traveling in moving water. Environmental and economic costs of nonindigenous While at first glance our results might suggest that removal species in the United States. BioScience 50: 53-64. of trees downslope of roads is a useful strategy to decrease Sinkiewicz, C.A. and E.S. Jules. in press. Port Orford cedar and infection risk, we caution a poor interpretation of our work. Our the non-native pathogen, Phytophthora lateralis. Fremontia. work does not support the proposed management strategy of Stokes, M.A., and T.L. Smiley. 1968. An introduction to tree-ring removing Port Orford cedars (sanitation logging) because of the dating. University of Chicago Press, Chicago, Illinois. confounding effects of post-sanitation processes. In areas where Zobel, D.B., and G. M. Hawk. 1980. The environment of cedars are removed, our field experience suggests that cedars will Chamaecyparis lawsoniana. The American Midland quickly re-seed and initiate new populations that maintain Naturalist 103:280-297. infection risk. Only continuous removal of cedars over long creek Zobel, D.B., L.F. Roth, and G.M. Hawk. 1985. Ecology, distances could accomplish significant reductions in infection risk, pathology, and management of Port-Orford-cedar though in our opinion this type of management is entirely (Chamaecyparis lawsoniana). USDA Forest Service, impractical across the large regions that Port Orford cedars General Technical Report PNW-184, Pacific Northwest occupy. Forest and Range Experimental Station, Portland, Oregon. Finally, our study suggests that, at least in some areas, the

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 15 PHYTOPHTHORA DISEASE ECOLOGY

Table 1. Summary statistics for road crossings containing uninfected and infected Port Orford cedar. Infected crossings summarized here are only those crossings that were infected by independent infection events via vehicles, not infection that arrived in flowing water from upstream infection. Factors in bold indicate those with significant effects on infection risk (see Table 2). The text includes details concerning independent infections and explanatory factors. Reprinted with permission from Jules et al. (2002).

Uninfected Infected

N = 48 N = 26

Units Mean SE Mean SE

Distance to Nearest Tree meters 117.7 32.0 10.5 6.3

Host Abundance # of trees 6.3 1.6 18.5 3.2

Slope degrees 14.2 0.9 13.8 1.3

Crossing Size meters 40.5 1.9 43.5 4.3

Elevation meters 1157.3 27.1 1082.0 31.4

2 Catchment Area km 1759.3 368.9 3924.5 910.6

2 7 5 7 5 Solar Radiation joules/m 1.48 x 10 1.40 x 10 1.44 x 10 2.53 x 10

Table 2. Likelihood ratio tests for significance of explanatory covariates in the proportional hazards regression model. For each explanatory variable, the log-likelihood is shown for the reduced model which excludes that variable. The likelihood ratio χ2 value is calculated as -χ2(Lreduced Ð Lfull). The reduced model used to test the full model has no covariates and a log-likelihood value of -107.205. Factors in bold have significant effects on infection risk. Reprinted with permission from Jules et al. (2002).

Explanatory Variable Log Likelihood χ2 d.f. P-value

Distance to Nearest Tree -97.342 3.927 1 0.0475

Host Abundance -98.262 5.767 1 0.0163

Slope -95.652 0.546 1 0.4600

Crossing Size -95.501 0.244 1 0.6213

Elevation -95.379 0.001 1 0.9748

Catchment Area -98.195 5.633 1 0.0176

Solar Radiation -95.782 0.806 1 0.3693

Full Model -95.379 23.4093 6 0.0007

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 16 PHYTOPHTHORA DISEASE ECOLOGY

Figure 1. Range map of Port Orford cedar. Map produced by W.D. Ritts.

Range of Port Orford Cedar Coos Bay Major Road

Pacific

Ocean Grants Pass

Kilometers

OREGON

Area Enlarged CALIFORNIA

Eureka Redding

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 17 PHYTOPHTHORA DISEASE ECOLOGY

Figure 2. Distribution of Port Orford cedar and its associated pathogen, Phytophthora lateralis, in 1999 within the Page Mountain study area, southwestern Oregon and northwestern California, USA. Cedars in the study area are generally restricted to riparian areas, such that their distribution follows the network of creeks. Independent infection events are those that required transport from an infected creek to an uninfected creek. Squares indicate independent infections via vehicle traffic along roads and triangles indicate independent infections via animal and foot traffic. Infection from these events spread downstream in water. Reprinted with permission from Jules et al. (2002).

Creek L Stream ittl er e Eld Infected Not Infected E ld e r Independent Infection Road Creek Non-Road ------Road

0 2

Page Creek

OREGON Study Area

C ALIFORNIA

Creek

Dunn

Fork North

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 18 PHYTOPHTHORA DISEASE ECOLOGY

Figure 3. Example of a road crossing a creek where host cedars may come in contact with spores of Phytophthora lateralis deposited by vehicles. Host target abundance below road, distance to nearest cedar, and road distance in contact with creek are shown. Road distance is the amount of road within 15 m of the creek. Reprinted with permission from Jules et al. (2002).

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 19 PHYTOPHTHORA DISEASE ECOLOGY

Figure 4. The number of new infection events of Phytophthora lateralis on Port Orford cedar occurring between 1977 and 1999. Infections that traveled by vehicles (Road infections; solid bars) and animals/hikers (Non-Road infections; hatched bars) are distinguished. Reprinted with permission from Jules et al. (2002). INFECTIONS NUMBER OF NEW 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 YEAR

Figure 5. Comparison of a U.S. Forest Service survey and the census performed in the Page Mountain study area, southwestern Oregon and northwestern California, USA. Reprinted with permission from Jules et al. (2002), Ecology 83:3167-3181. NUMBER OF INFECTIONS

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 20 PHYTOPHTHORA DISEASE ECOLOGY

EXOTIC PATHOGENS, RESISTANT SEED AND RESTORATION OF FOREST TREE SPECIES IN WESTERN NORTH AMERICA

Richard A. Sniezko USDA Forest Service, Dorena Genetic Resource Center 34963 Shoreview Road, Cottage Grove, OR 97424 Email: [email protected]

Diana F. Tomback Department of Biology, CB 171, University of Colorado at Denver P.O. Box 173364, Denver, CO 80217 Email: [email protected]

Regina M. Rochefort USDI National Park Service, North Cascades National Park Service 810 State Route 20 Complex, Sedro-Woolley, WA 98284 Email: [email protected]

Ellen Goheen USDA Forest Service, Southwest Oregon Forest Insect and Disease Service Center 2606 Old Stage Road, Central Point, OR 97502 Email: [email protected]

Rich Hunt Natural Resources Canada, Pacific Forestry Center 506 W. Burnside Road, Victoria, BC V8Z 1M5, Canada Email: [email protected]

Jerry S. Beatty USDA Forest Service, RPC 7, Forest Health 1601 N. Kent St., Arlington, VA 22209 Email: [email protected]

Michael Murray USDI National Park Service, Crater Lake National Park Service PO Box 7, Crater Lake, OR 97604 Email: [email protected]

Frank Betlejewski USDA Forest Service, Southwest Oregon Forest Insect and Disease Service Center 2606 Old Stage Road, Central Point, OR 97502 Email: [email protected]

Abstract

Non-native invasive pathogens such as white pine blister rust (Cronartium ribicola) and Port-Orford-cedar root disease (Phytophthora lateralis) are killing trees and disrupting forest ecosystems in western North America. Populations of western white pine (Pinus monticola), sugar pine (P. lambertiana), whitebark pine (P. albicaulis), and limber pine (P. flexilis) are declining precipitously from

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 21 PHYTOPHTHORA DISEASE ECOLOGY damage by blister rust. Foxtail pine (Pinus balfouriana) and southwestern white pine (P. strobiformis) populations are also infected by blister rust in parts of their range. Phytophthora lateralis continues to spread and kill Port-Orford-cedar (Chamaecyparis lawsoniana) in Oregon and California. Because resistant individuals in all these species are rare, genetic variation may be reduced to the point where future populations may not be viable without active management. Seeds from resistant parents are now available for western white pine, sugar pine, and Port-Orford-cedar restoration for some areas. Selection and breeding programs for resistance, coupled with active ecological management, will be needed to create opportunities to restore and retain these species in forest ecosystems on federal or crown lands. Restoration strategies for maintaining these species on the landscape must include planting resistant stock and increasing any opportunities for natural regeneration until resistance characterizes populations, and they are able to continue to evolve in the continued presence of the pathogens. Scientists and the public will have difficult decisions to face regarding actions to take in wilderness areas and national parks. Problem white pines are killed, the associated ecosystems also decline, altering western forest landscapes. In addition, these white pines are all fire dependent, and have declined due to past fire exclusion on-native invasive pathogens are having large impacts on policies and resulting successional replacement (Tomback 2003; natural ecosystems in western North America (Tomback Tomback and Achuff, in preparation). et al. 1995; Campbell and Schlarbaum 2002). Two NN Three pines, southwestern white (Pinus strobiformis), prominent examples include Cronartium ribicola, which causes western white, and sugar pine, are important to local logging white pine blister rust, and Phytophthora lateralis, which causes economies (Kinloch 1984; Lowery 1984; Fins et al. 2001). The Port-Orford-cedar root disease (Hunt 1997; Hansen 1997; Jules et sugar pine and western white pine industries have already suffered al. 2002). Blister rust is rapidly killing five-needled white pines, major collapse (Graham 1990; Kinloch and Scheuner 1990). and disrupting the associated ecosystems (Kendall and Keane Whitebark pine, limber pine, both bristlecone pines, and foxtail 2001; McDonald and Hoff 2001). Port-Orford-cedar root disease pine are important high elevation species that stabilize soils and is killing Port-Orford-cedars, particularly in riparian areas (Jules et regulate snowmelt (Farnes 1990; Tomback et al. 2001). al. 2002). Both pathogens continue to spread geographically and White pine forests comprise large tracts of land in western to intensify infection levels in many locations (Kendall and Keane National Forests and Wilderness Areas and National Parks in the 2001; USDI-BLM and USDA-FS 2004). As a consequence we United States, and Crown Lands in western Canada. Disruption are losing major forest habitat types, and the biodiversity and of these ecosystems threatens local economies, alters fire regimes ecosystem services they provide (Hunt 1997; Fins et al. 2001; and ecosystem function, and degrades the aesthetic beauty of these Tomback et al. 2001; Jules et al. 2002; 2003USDI-BLM and lands (e.g., Fins et al. 2001; Tomback and Achuff, in preparation). USDA-FS 2004). In the Kootenays of British Columbia, extensive stands of western Trees resistant to these pathogens are present in the forests, white pine have been replaced by less valuable western hemlock but in many cases they are too rare and may be too widely (Hunt et al. 1985). Less than 10% of the historic five million scattered to provide adequate regeneration as well as broad acres of western white pine cover type remains in today’s Inland genetic diversity to maintain these species (Fins et al. 2001; Hoff Northwest forests (Fins et al. 2001). On a smaller scale, white et al. 2001; Kinloch et al. 2003; Kegley and Sniezko 2004). pine blister rust has killed all western white pine in the Champion Active management will be essential if white pines and Port- Mine area on the Umpqua National Forest (Sniezko, personal Orford-cedar are to continue as vital ecosystem components (Hoff communication). Similarly, dead whitebark pine is prevalent et al. 2001; USDI-BLM and USDA-FS 2004). throughout the higher elevations of Glacier National Park and the surrounding National Forests and Bob Marshall Wilderness White Pine Blister Rust Complex (Kendall and Keane 2001). White pine blister rust will make recovery of these species difficult, if not, impossible without Inadvertently introduced to the West in 1910, white pine human intervention. blister rust has spread across the range of five-needled white pines (McDonald and Hoff 2001). All eight of the western North American species of white pines are susceptible to this pathogen Whitebark Pine: a Case History (Childs and Bedwell 1948, Hoff et al. 1980). These pines occur in ecosystems from near sea level to tree line. Six of these eight Whitebark pine is the most widely distributed white pine in species have already been impacted—several severely (McDonald the western United States and Canada, inhabiting upper subalpine and Hoff 2001). Prior to 2003 there were no known cases of and treeline elevations (Arno and Hoff 1990; Tomback and bristlecone pine with blister rust infection in natural stands, but Achuff, in preparation), Because of inaccessibility, slow growth blister rust was known to occur dangerously near both the ancient rates, and its shrubby growth form, the species has not been Great Basin bristlecone pines (Pinus longaeva) in California and commercially valuable. Whitebark pine, however, provides the Rocky Mountain bristlecones (Pinus aristata) in Colorado. In keystone services as a wildlife food source and as a pioneering 2003, infection was discovered on a Rocky Mountain bristlecone species in community development after fire (Tomback et al. in southern Colorado (Blodgett and Sullivan 2004). As these 2001; Tomback and Kendall 2001). In the Greater Yellowstone

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 22 PHYTOPHTHORA DISEASE ECOLOGY

Area, whitebark pine ecosystems are designated critical habitat for Kegley 2003; Kegley and Sniezko 2004; McDonald et al. 2004; grizzly bear (Ursus arctos). Only recently, has the widespread Hunt, in press). The resistant trees are placed in seed orchards to damage and mortality from blister rust in whitebark pine produce seedling populations that are genetically diverse and rust communities gained serious attention (e.g., Tomback et al. 2001; resistant. Another strategy used for western white pine in British Tomback 2003). Columbia is seed collection from seed production areas, which Surveys since the 1990’s show moderate to heavy blister have been developed by culling cankered trees and retaining the rust infection and mortality in many parts of the range of putative resistant ones (Meagher et al. 1987). Screening programs whitebark pine in both the U.S. and Canada (reviewed in: for disease resistance have recently begun for whitebark pine and Kendall and Keane 2001; Tomback 2003). For example, since the should be feasible for the remaining susceptible pines as well 1930’s approximately 26% of the whitebark pine has died from (Mahalovich and Dickerson 2004; Sniezko, personal blister rust in Crater Lake National Park in the southern Cascades communication). In addition, information is required on seed of Oregon. By 2050, it is estimated that the decline in mature transferability of the high elevation white pines — that is, how far whitebark in the western portions of the Park will be 46% seeds can be moved within a species’ range and still maintain (Murray and Rasmussen 2003). In a recent survey along the good survival and adaptability. Common garden studies have Pacific Crest National Scenic Trail on the Umpqua National been used to establish seed movement guidelines for western Forest, over half the trees present had been infected with blister white pine and sugar pine (Campbell and Sugano 1987; Campbell rust (Goheen et al. 2002). In the Northern Rockies, the average and Sugano 1989). infection rate in 337 plots was 84% (Kendall and Keane 2001; Techniques now utilized in evaluating western white pine Tomback 2003). For many whitebark pine populations in areas of and sugar pine for resistance can likely be applied to other white moderate to high infection, rust-resistant seedlings will be needed pine species as well. In addition, tools from molecular genetics to restore the species. may make it simpler and less expensive to identify parent trees with natural resistance (Ekramoddoullah and Hunt 2002). Restoration Options: White pines Port-Orford-cedar Root Rot Several experimental projects in recent years have explored techniques for restoring white pines. A pioneering series of Phytophthora lateralis was introduced to the native range of demonstration projects for whitebark pine restoration were Port-Orford-cedar around 1952. This root pathogen is killing all initiated by Keane and Arno (2001) in west-central Montana. size classes of Port-Orford-cedar, particularly in riparian areas of These projects are being monitored for successful tree recruitment northwest California and southwestern Oregon (USDI-BLM and over time. The techniques used by Keane and Arno (2001) USDA-FS 2004). Port-Orford-cedar is an important component of comprise the basic restoration strategy for all declining white these forest ecosystems, in addition to being a valuable species for pines. They include (1) silvicultural thinning and in some cases timber and specialty products (Hansen et al. 2000). In areas of use of prescribed fire to provide regeneration opportunities for high disease incidence, we are unlikely to see many old growth these pines; allowing wildfire to burn where possible; (2) planting trees again unless action is taken. Private landowners are unlikely seeds and seedlings with high probabilities of rust resistance; and to replant Port-Orford-cedar without the availability of resistant (3) using silvicultural tools (e.g. pruning [Hunt 1998], risk hazard seedlings, thus decreasing species diversity over the landscape. rating [Hunt 1983]) to help minimize the impact of the disease in high priority forest stands. Restoration Options: Port-Orford-cedar Prescribed fire and silvicultural thinning are more likely to be successful and more cost efficient in areas of lower disease The major management strategies for Port-Orford-cedar are: hazard or in stands with low infection levels of blister rust. management to slow the spread of the disease (e.g. road closures), However, these techniques alone are not sufficient to increase the and use of seeds or seedlings from the resistance program to frequency of rust-resistant individuals on the landscape. It is restore areas of high mortality where large Port-Orford-cedar is inevitable that rust-resistant stock will need to be planted in order desired. Activities to slow or prevent the spread of Port-Orford- to establish white pine populations that will continue to co-exist cedar root disease have received major emphasis in the past, and with blister rust (Fins et al. 2001). More experimental work is will continue to be the primary focus (USDI-BLM and USDA-FS needed to develop specialized and effective techniques for each 2004). Other management activities such as planting resistant five-needled white pine species. The necessity for planting Port-Orford-cedar will potentially have an important role where resistant populations and implementing restoration techniques for disease is already present or where new infestations occur (USDI- white pine species is now well recognized (USDA Forest Service BLM and USDA-FS 2004). 2003). The frequency of natural resistance to Phytophthora Over the last 40 years, genetically rust-resistant stock has lateralis may be too low and scattered in native Port-Orford-cedar been developed for some western white pine and sugar pine ecosystems for successful natural regeneration in areas of highest populations by evaluating thousands of trees to find the rare, disease incidence (USDI-BLM and USDA-FS 2004). Since 1997, naturally occurring resistant trees (Fins et al. 2001; Sniezko and the operational resistance program has made significant progress

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 23 PHYTOPHTHORA DISEASE ECOLOGY in finding trees with natural resistance, establishing seed orchards review of the text; Heather May and Cherry Schofield for help in and producing seed for some areas (Sniezko et al. 2000; Sniezko formatting the paper. We would also like to thank Dr. M. and Hansen 2003; Sniezko et al. 2003; USDI-BLM and USDA-FS Southworth and Dr. J. Williams for their review of an earlier 2004; Sniezko et al. this proceedings). Traditional methods of version of this manuscript. selection and breeding allow us to bring together rare resistant Port-Orford-cedar trees for cross-pollination. Seeds from these Literature Cited pollinations can generate populations of genetically diverse, adapted, and resistant trees for restoration. However, the Arno, S.F. and R.J. Hoff. 1990. Pinus albicaulis Engelm. resistance program is relatively new and more work is needed to Whitebark pine. Pages 268-279 in, R.M. Burns and B.H. provide resistant populations for all areas (USDI-BLM and Honkala, editors. Silvics of North America, Vol. 1, USDA-FS 2004). Guidelines are being developed to aid managers Conifers. Agriculture Handbook 654. USDA Forest in determining where resistant seedlings could be used and where Service, Washington, D.C. to limit their use (USDI-BLM and USDA-FS 2004). Field trials Blodgett, J.T. and K.F. Sullivan. 2004. First report of white pine have been established to monitor the effectiveness of resistance on blister rust on Rocky Mountain bristlecone pine. Plant an array of sites. Disease 88:311. A Dilemma Campbell, F.T. and S.E. Schlarbaum. 2002. Fading forests II: trading away North America’s natural heritage. Healing Stones Foundation. The fact that National Parks and designated wilderness Campbell, R.K. and A.I. Sugano. 1989. Seed zones and breeding areas are also severely impacted by these pathogens raises zones for white pine in the Cascade Range of Washington pressing management issues (See Tempel et al. [2003] for and Oregon. Res. Paper PNW-RP-407. USDA Forest discussion on research needs for managing non-native species in Service, Pacific Northwest Research Station, Portland, wilderness areas). Traditionally, these lands are considered Oregon. reasonably intact ecosystems without need for active management Campbell, R.K. and A.I. Sugano. 1987. Seed zones and breeding (e.g., McCool and Freimund 2001). We face the quandary of zones for sugar pine in southwestern Oregon Res. Paper doing nothing and watching the destruction of white pine and PNW-RP-383. USDA Forest Service, Pacific Northwest cedar ecosystems, or, with public and government support, we Research Station, Portland, Oregon. evaluate on a case-by-case basis the need to restore white pines Childs, T.W. and J.L. Bedwell. 1948. Susceptibility of some and Port-Orford-cedar in wilderness areas against other wilderness white pine species to Cronartium ribicola in the Pacific values. Similar considerations will be needed in National Parks. Northwest. Journal of Forestry 46:595-599. We recognize that both of these pathogens and fire exclusion are Ekramoddoullah, A.K.M. and R.S. Hunt. 2002. Challenges and anthropogenic in origin, which could support some level of opportunities in studies of host-pathogen interactions in management action. forest tree species. Canadian Journal of Plant Pathology 24: 1-8. Conclusion Farnes, P.E. 1990. SNOTEL and snow course data: describing the hydrology of whitebark pine ecosystems. Pages 302- Ecologists, pathologists, geneticists, silviculturists, land 204 in: W.C. Schmidt and K.J. McDonald, compilers. managers, and the public will have to work together to reverse Proceedings— symposium on whitebark pine ecosystems: population declines and restore ecosystems damaged by these ecology and management of a high-mountain resource. introduced pathogens. The development of resistant tree General Technical Report INT-270. USDA Forest Service, populations offers an opportunity to counter some of the effects of Intermountain Research Station, Ogden, Utah. these pathogens. It will be a long-term process, but with Fins, L., J. Byler, D. Ferguson, A. Harvey, M. F. Mahalovich, G. concerted efforts, responsible land stewardship can be McDonald, D. Miller, J. Schwandt, and A. Zack. 2001. accomplished. Intervention to restore more natural conditions in Return of the giants: restoring white pine ecosystems by wilderness may be evaluated case-by-case, and weighed against breeding and aggressive planting of blister rust-resistant other wilderness values. Organizational and implementation white pine. Station Bulletin 72. University of Idaho, strategies that are developed for managing white pine blister rust College of Forest Resources, Moscow. and Port-Orford-cedar root disease can provide a starting point for Goheen, E.M., D.J. Goheen, K. Marshall, R.S. Danchok, J.A. work involving other introduced pathogens. Restoration work with Petrick, and D.E. White. 2002. The status of whitebark these species should provide insights that will be useful in dealing pine along the Pacific Crest National Scenic Trail on the with other non-native invasive insects and pathogens. Umpqua National Forest General Technical Report PNW- GTR-530. USDA Forest Service, Pacific Northwest Acknowledgments Research Station, Portland, Oregon. Graham, R.T. 1990. Pinus monticola Dougl. ex D. Don. Western Leslie Elliott, Angelia Kegley, and Heather May for their white pine. Pages 385-394,in, R.M. Burns and B.H.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 24 PHYTOPHTHORA DISEASE ECOLOGY

Honkala, eds. Silvics of North America. Agricultural Phytopathology 93:691-694. Handbook 654. USDA Forest Service, Washington, D.C. Kinloch, B.B. and W.H. Scheuner. 1990. Pinus lambertiana Hansen, E.M., D.J. Goheen, E. Jules, and B. Ullian. 2000. Dougl. Sugar pine. Pages 370-379 in, R.M. Burns and Managing Port-Orford-cedar and the introduced pathogen B.H. Honkala, eds. Silvics of North America. Agricultural Phytophthora lateralis. Plant Disease 84:4-14. Handbook 654. USDA Forest Service, Washington, D.C. Hansen, E.M. 1997. Port Orford cedar root disease. Pages 6-7 in, Kinloch, B.B. 1984. Sugar pine: an American wood. FS-257. E.M. Hansen and K.J. Lewis, eds. Compendium of conifer USDA Forest Service, Washington, D.C. diseases. APS Press, St. Paul, Minnesota. Lowery, D.P. 1984. Western white pine: an American wood. FS- Hoff, R.J., D.E. Ferguson, G.I. McDonald, and R.E. Keane. 2001. 258. USDA Forest Service, Washington, D. C. Strategies for managing whitebark pine in the presence of Mahalovich, M.F. and Dickerson, G.A. 2004. Whitebark pine white pine blister rust. Pages 346-366 in, D.F. Tomback, genetic restoration program for the Intermountain West. S.F. Arno, and R.E. Keane, eds. Whitebark pine Pages 181-187 in, R.A. Sniezko, S. Samman, S.E. communities: ecology and restoration. Island Press, Schlarbaum, and H.B. Kriebel, eds. Breeding and genetic Washington, D.C. resources of five-needle pines: growth adaptability and pest Hoff, R.J., RT. Bingham, and G.I. McDonald. 1980. Relative resistance. 2001 Proceedings RMRS-P-32. USDA Forest blister rust resistance of white pines. European Journal of Service, Rocky Mountain Research Station, Fort Collins, Forest Pathology 5:307-316. Colorado. Hunt, R.S. In Press. Blister-rust-resistant white pines for British McCool, S.F. and W.A. Freimund. 2001. Threatened landscapes Columbia. Canadian Forest Service BC-X-397. and fragile experiences: conflict in whitebark pine Hunt, R.S. 1998. Pruning western white pine in British Columbia restoration. Pages 263-284 in, D. F. Tomback, S. F. Arno, to reduce white pine blister rust losses: 10-year results. and R.E. Keane, eds. Whitebark pine communities: ecology Western Journal of Applied Forestry 13:60-63. and restoration. Island Press, Washington, D.C. Hunt, R.S. 1997. White pine blister rust. Pages 26-26 in, E.M. McDonald, G.P. Zambino, and R. Sniezko. 2004. Breeding rust- Hansen and K.J. Lewis, eds. Compendium of conifer resistant five-needle pines in the western United States: diseases. APS Press, St. Paul, Minnesota. lessons from the past and a look to the future. Pages 28-50 Hunt, R.S., E. von Rudloff, M.S. Lapp, and J.F. Manville. 1985. in, R.A. Sniezko, S. Samman, S.E. Schlarbaum, and H.B. White pine blister rust in British Columbia III. Effects on Kriebel, eds. Breeding and genetic resources of five-needle the gene pool of western white pine. Forestry Chronicles pines: growth adaptability and pest resistance. USDA Forest 61: 484-488. Service, Rocky Mountain Research Station, Fort Collins, Hunt, R.S. 1983. White pine blister rust in British Columbia II. Colorado. Can stands be hazard rated? Forestry Chronicles 59: 30-33. McDonald, G.I., and R.J. Hoff. 2001. Blister rust: an introduced Jules, E.S., M.J. Kauffman, W.D. Ritts, W.D., and A.L. Carroll. plague. Pages 192-220 in, D.F. Tomback, S.F. Arno, and A.L. 2002. Spread of an invasive pathogen over a variable R.E. Keane, eds. Whitebark pine communities: ecology and landscape: a nonnative root rot on Port-Orford-cedar. restoration. Island Press, Washington, D.C. Ecology 83 (11):3167-3181. Meagher, M.D., R.S. Hunt, and D.P. Wallinger. 1987. Seed- Keane, R.E. and S.F. Arno. 2001. Restoration concepts and production areas of western white pine for interim seed techniques. Pages 367-400 in, D.F. Tomback, S.F. Arno, supply. Canadian Forestry Service, Pacific Forestry Centre and R.E. Keane, eds. Whitebark pine communities: ecology Miscellaneous Report. and restoration. Island Press, Washington, D.C. Murray, M.P. and M.C. Rasmussen. 2003. Non-native blister rust Kegley, A.J. and R.A. Sniezko. 2004. Variation in blister rust disease on whitebark Pine at Crater Lake National Park. resistance among 226 Pinus monticola and 221 P. Northwest Science 77:87-91. lambertiana seedling families in the Pacific Northwest. Sniezko, R.A., L.J. Elliott, E.M. Hansen, and D.J. Goheen. In Pages 209-226 in, R.A. Sniezko, S. Samman, S.E. Press (this proceedings). Genetic resistance in Port-Orford- Schlarbaum, and H.B Kriebel, eds. Breeding and genetic cedar to the non-native root rot pathogen Phytophthora resources of five-needle pines: growth adaptability and pest lateralis: a tool to aid in restoration in infested riparian resistance. RMRS-P-32. USDA Forest Service, Rocky areas. Proceedings of the Second Conference on Klamath- Mountain Research Station, Fort Collins, Colorado. Siskiyou Ecology. Kendall, K.C. and R.E. Keane. 2001. Whitebark pine decline: Sniezko, R.A., L.J. Elliott, D.J. Goheen, K. Casavan, E.M. infection, mortality, and population trends. Pages 221-242 Hansen, C. Frank, and P. Angwin. 2003. Development of in, D.F. Tomback, S.F. Arno, and R.E. Keane, eds. Phytophthora lateralis resistant Port-Orford-cedar for Whitebark pine communities: ecology and restoration. restoration in the Pacific Northwest. Proceedings of 17th Island Press, Washington, D.C. North American Forest Biology Workshop, July 15-18, Kinloch, B.B., R.A. Sniezko, and G.E. Dupper. 2003. Origin and 2002, Pullman, Washington. distribution of Cr2, a gene for resistance to white pine Sniezko, R.A. and E.M. Hansen. 2003. Breeding Port-Orford- blister rust in natural populations of western white pine. cedar for resistance to Phytophthora lateralis: current status

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 25 PHYTOPHTHORA DISEASE ECOLOGY

& considerations for developing durable resistance. Tomback, D.F. 2003. The rapid decline of whitebark pine Pages 197-201 in, J.A. McComb, G.E. St. J. Hardy, and communities: ecological and biodiversity implications. I.C. Tommerup, eds. Phytophthora in forests and natural Pages 31-42 in. L. Taylor, K. Martin, D. Hik, and A. Ryall, ecosystems, 2nd International Working Party 7.02.09 eds. Proceedings: Ecological and Earth Sciences in Meeting. Murdoch University Print, Albany, Australia. Mountain Areas, Sept. 6-10, 2002. The Banff Centre, Sniezko R.A. and A.J. Kegley. 2003. Blister rust resistance of Banff, Alberta, Canada. five-needle pines in Oregon and Washington. Proceedings Tomback, D.F. S.F. Arno, and R.E. Keane 2001. The compelling of the Second IUFRO Rusts of Forest Trees Working Party case for management intervention. Pages 3-25 in, D.F. Conference, 19-23 Aug., 2002, Yangling, China. Forest Tomback, S.F. Arno, and R.E. Keane, eds. Whitebark pine Research, 16 (Supplement):101-112. communities: ecology and restoration. Island Press, Sniezko, R.A., E.M. Hansen, A. Bower, D. Goheen, K. Marshall, Washington, D.C. K. Casavan, and W. Sutton. 2000. Genetic resistance of Tomback, D.F., and K.C. Kendall. 2001. Biodiversity losses: the Port-Orford-cedar (Chamaecyparis lawsoniana) to downward spiral. Pages 243-664 in, D.F. Tomback, S.F. Phytophthora lateralis: results from early field trials. Pages Arno, and R.E. Keane, eds. Whitebark pine communities: 138-140 in, E.M. Hansen and W. Sutton, eds. Phytophthora ecology and restoration. Island Press, Washington, D.C. diseases of forest trees: Proceedings from the First Tomback, D.F., J.K. Clary, J. Koehler, R.J. Hoff, and S.F. Arno. International Meetings on Phytophthoras in Forest and 1995. The effects of blister rust on post-fire regeneration of Wildland Ecosystems. Forest Research Laboratory, Oregon whitebark pine: the Sundance Burn of Northern Idaho State University. (U.S.A.). Conservation Biology 9:654-664. Tempel, D., V. Wright, and P. Landres. 2003. Research needs for USDA Forest Service. 2003. Managing for healthy white pine managing non-native species in wilderness areas. In ecosystems in the United States to reduce the impacts of formation online at The Aldo Leopold Wilderness Research white pine blister rust. Report R1-03-118. Department of Institute: http://leopold.wilderness.net/research/ Agriculture, Forest Service, Missoula, Montana. invasives/questions.htm. USDI-BLM and USDA-FS. 2004. Management of Port-Orford- Tomback, D.F. and P. Achuff. In preparation. Blister rust and Cedar in Southwest Oregon: Final Supplemental western forest biodiversity: status, trends, and restoration Environmental Impact Statement. USDI-BLM issues for white pines. Forest Ecology and Management. OregonÐWashington State Office and USDA-FS Pacific Northwest Region, Portland, Oregon.

Illustration by Bob Cremins

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 26 PHYTOPHTHORA DISEASE ECOLOGY

GENETIC RESISTANCE IN PORT-ORFORD-CEDAR TO THE NON-NATIVE ROOT ROT PATHOGEN PHYTOPHTHORA LATERALIS: A TOOL TO AID IN RESTORATION IN INFESTED RIPARIAN AREAS

Richard Sniezko USDA Forest Service, Cottage Grove, OR 97424 Email: [email protected]

Everett Hansen Department of Botany & Plant Pathology, Oregon State University, Corvallis, OR 97331 Email: [email protected]

Leslie Elliott USDA Forest Service, Cottage Grove, OR 97424 Email: [email protected]

Don Goheen USDA Forest Service, J. Herbert Stone Nursery, Central Point, OR 97502 Email: [email protected]

Abstract

Port-Orford-cedar is highly susceptible to the non-native invasive root rot pathogen, Phytophthora lateralis. This introduced pathogen continues to spread and cause mortality to native Port-Orford-cedar, particularly in riparian areas, in northwestern California and southwestern Oregon. This pathogen will not cause extinction of Port-Orford-cedar nor have large impacts on the range-wide genetic diversity of the species. However, on high-risk sites, it can reduce the frequency of Port-Orford-cedar trees that provide large, structural elements to riparian ecosystem function. Fortunately, there is natural genetic variation in resistance to P. lateralis in Port-Orford-cedar. However, resistant Port-Orford-cedar are too scattered over the landscape to allow restoration to be successful in the most severely diseased areas without human assistance. The biology of the tree species, the characteristics of the pathogen, and the cooperation of many people have led to rapid progress in development of disease resistant populations of Port-Orford-cedar. Orchards have been established for several geographic areas and seed is now available for federal lands in these areas. These orchards will produce genetically diverse and adapted, disease resistant seedlings that could be used in restoration efforts in the areas of highest Port-Orford- cedar mortality. Based on early results, disease resistant seedlings from orchards containing the most resistant parents would have more than 50% survival versus less than 5% survival of seedlings from the most susceptible families. The use of naturally occurring genetic resistance is a management tool that can be useful in areas already impacted by the pathogen, and will not diminish ongoing efforts to slow the spread of P. lateralis.

Problem cedar trees that provide large, structural elements to riparian ecosystem function (USDIÐBLM and USDA-FS 2004; Jules et al. ort-Orford-cedar (Chamaecyparis lawsoniana) is highly 2002), susceptible to a non-native invasive pathogen, The importance of management activities to slow or stop PPPhytophthora lateralis, that causes Port-Orford-cedar root the further spread of the pathogen are well recognized, however, disease (USDA-FS and USDI-BLM 2003; Hansen et al. 2000). P. lateralis has already had large impact in some areas and will This pathogen continues to spread and cause mortality to native likely continue to spread (USDI-BLM and USDA-FS 2004). In Port-Orford-cedar, particularly in riparian areas, in northwestern these cases, management tools such as planting seedlings with California and southwestern Oregon. Both federal lands (Forest genetic resistance to Port-Orford-cedar root disease may be Service, National Park Service, and Bureau of Land Management) invaluable aids to restoration. Private landowners have a keen and private lands (industrial and non-industrial) are impacted. P. interest in planting Port-Orford-cedar to diversify their species lateralis will not cause extinction of Port-Orford-cedar nor have mix, but most are reluctant to plant this species without the large impacts on the rangewide genetic diversity of the species availability of disease resistant seedlings. Fortunately, there is (USDI-BLM and USDA-FS 2004). However, in areas of high- natural genetic variation in Port-Orford-cedar in resistance to P. risk sites it can severely reduce the frequency of Port-Orford- lateralis. However, the natural frequency of resistant Port-Orford-

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 27 PHYTOPHTHORA DISEASE ECOLOGY cedar is too low across the landscape to allow restoration to be 2) evaluate the durability of resistance (Sniezko et al. 2000). successful in the most severely diseased areas without human Most of the sites used for testing had recent evidence of P. assistance. lateralis mortality. In these tests, mortality from P. lateralis often begins within the first year or two after planting. Thus far, there Resistance Program is a strong correlation between greenhouse testing and early field test results (Table 2; Figure 1), however, most field sites have only been established since 2000 and monitoring will continue to History evaluate the durability of the resistance. Phytophthora lateralis has been present in Port-Orford- Continued survival of parent trees originally selected over cedar ecosystems since the early 1950s. Natural genetic resistance 10 years ago in P. lateralis infested sites, and confirmed as to the pathogen was first identified in the late 1980s (Hansen et al. resistant in greenhouse testing, provides the best evidence to date 1989). The USDA Forest Service became involved in developing of the potential utility of resistance for forest sites. Plans are a resistance program in the late 1980s with the USDI Bureau of being made for a much larger survey of resistant parent trees, and Land Management (BLM) joining the effort in the mid 1990s. An these forest trees will be monitored over time to continue to operational program to identify natural genetic resistance began in evaluate the utility of resistance. 1997 and today includes a cooperative effort involving five Preliminary analysis of greenhouse resistance testing National Forests, three BLM Districts, Oregon State University suggests at least two types of resistance in Port-Orford-cedar: (1) and a variety of other federal, state, and local agencies as well as a small percentage of families show high survival, and (2) a small industrial and non-industrial private landowners (Sniezko and percentage of families show a slower rate of mortality. Future Hansen 2000; Sniezko et al. 2003; USDA-FS and USDI-BLM testing is underway to determine what other types of resistance 2003). Short-term greenhouse testing as well as longer term field may be present, as well as the utility of resistance in the field. testing are vital components of the resistance breeding program. Conventional breeding efforts will be aimed at increasing levels of Approximately 10,000 Port-Orford-cedar trees, from across the resistance and/or combining resistance types. The durability of range of the species, have been tested for genetic resistance. resistance will depend on the evolutionary potential of the Many of these trees were selected in forest areas of moderate to pathogen as well as the nature of the resistance available high mortality from Port-Orford-cedar root rot. (McDonald and Linde 2002). Phytophthora lateralis has a very limited genetic variation (USDA-FS and USDI-BLM 2003; Testing Results McWilliams 2000) and relatively slow dispersal which should aid Testing to date has shown that natural genetic resistance to the efforts to develop durable resistance. A first look at possible P. lateralis is rare and most trees are susceptible. Of the 10,000+ resistant mechanisms show a difference between resistant and trees initially tested through a stem or branch dip technique at non-resistant seedlings in the attraction of zoospores and root Oregon State University, approximately 10% were selected for penetration of the pathogen (Oh and Hansen 2002). The further evaluation in the resistance program (see USDA-FS and identification of resistance mechanisms and their inheritance and USDI-BLM 2003 and Bower et al. 2000, for details and as well as further information on the potential for new virulent references regarding methods). These trees are now going strains of the pathogen can be used to define breeding and through a second phase of testing at Oregon State University seedling planting strategies. using a root dip inoculation technique. In this test, the roots of six The native range of Port-Orford-cedar has been partitioned rooted cuttings from each parent tree are exposed to P. lateralis. into a series of breeding blocks and breeding zones to help ensure Approximately 500 trees have been through this second phase of adaptable seedlings are used for reforestation and restoration testing and approximately 100 of the selections showed 100% (USDA-FS and USDI-BLM 2003). From the resistance screening survival. program, disease resistant populations (orchards) are being In other testing, seedling families from the breeding established at the USDA Forest Service Dorena Genetic Resource program are evaluated using the root dip inoculation. Preliminary Center. These orchards will produce genetically diverse and analysis shows that trees such as parent PO-117490, identified as adapted, disease resistant seedlings that could be used in resistant, produce seedlings with high resistance when crossed restoration efforts in the areas of highest Port-Orford-cedar with a variety of other genotypes (see Table 1). Most seedlots mortality within its native range (Figure 2). This may be one of collected from forest stands show very high mortality (> 90%) in the only options available to begin the restoration of Port-Orford- these tests. As the second phase of testing continues more cedar in disease-infested areas. Seed is now available for some resistant trees are expected to be identified. Seedlings from some areas of federal lands (USDI-BLM and USDA-FS 2004). Based resistant parent trees (for example, CF1 and 117490) have been on early results, disease resistant seedlings from orchards extensively tested over the last decade and always show high containing the most resistant parents would have greater than 50% levels of resistance relative to most seedlots from forest survival versus less than 5% survival of seedlings from the most collections. susceptible families, or seedlots from natural stands. A series of Replicated field tests, on a range of sites, have been field tests established since 2000 will continue to provide established to 1) validate results from the greenhouse testing and information of the utility of resistant seedlings in restoration.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 28 PHYTOPHTHORA DISEASE ECOLOGY

Resistance As A Tool Literature Cited

The highest risk sites for P. lateralis are unlikely to have Bower, A.D., K. Casavan, C. Frank, D. Goheen, E.M. Hansen, K. enough resistant Port-Orford-cedar to provide large trees for the Marshall, R.A. Sniezko, and W. Sutton. 2000. Screening future. The use of naturally occurring genetic resistance is a Port-Orford-cedar for resistance to Phytophthora lateralis: management tool that can be used to help re-establish Port- Results from 7000+ trees using a branch lesion test. In: Orford-cedar in the areas most heavily impacted. Land managers Everett M. Hansen and Wendy Sutton (eds.) Phytophthora will decide whether restoration of Port-Orford-cedar is an Diseases of Forest Trees; Proceedings From the First objective in their area. Traditional breeding techniques can be International Meetings on Phytophthoras in Forest and used to generate populations of genetically diverse, adapted and Wildland Ecosystems. Forest Research Laboratory, Oregon resistant trees. The use of resistant seedlings is primarily aimed at State University. pp. 99-100. high-risk sites where high percentages of Port-Orford-cedar have Hansen, E.M., P.B. Hamm, and L.F. Roth. 1989. Testing Port- already been killed, and not as a substitute for other management Orford-cedar for resistance to Phytophthora. Plant Disease tools to slow or stop the spread of P. lateralis on federal lands. 73(10):791-794. Strategies and guidelines for the utilization of resistant seedlings Hansen, E.M., D.J. Goheen, E.S. Jules, and B. Ullian. 2000. are being developed (Sniezko and Hansen 2003; USDI-BLM and Managing Port-Orford-cedar and the introduced pathogen, USDA-FS 2004). Non-federal landowners are very interested in Phytophthora lateralis. Plant Disease 84:4-14. planting resistant Port-Orford-cedar, and seedlings are just Jules, E.S., M.J. Kauffman, W.D. Ritts, and A.L. Carroll. 2002. becoming available. Spread of an invasive pathogen over a variable landscape: a Orchards are not currently available for all breeding zones, nonnative root rot on Port-Orford-cedar. Ecology but options are being explored to extend the resistance program to 83(11):3167-3181. these other areas (USDI-BLM and USDA-FS 2004). Thousands Kitzmiller, J.H., and R.A. Sniezko. 2000. Range-wide genetic of field selections will need to be tested to locate resistant trees variation in Port-Orford-cedar (Chamaecyparis lawsoniana for additional breeding zones. The option of incorporating [A. Murr.] Parl.): I. Early height growth at coastal and resistance from other breeding zones can also be evaluated, using inland nurseries. Journal of Sustainable Forestry 10(1,2):57- results from common garden studies (Kitzmiller and Sniezko 67. 2000; USDA-FS and USDI-BLM 2003) and genetic variation McDonald, B.A., and C. Linde. 2002. Pathogen population studies. genetics and the durability of resistance. Euphytica 124:163-180. Questions For The Future McWilliams, M.G. 2000. Variation in Phytophthora lateralis. Pages 50-54 in, E.M. Hansen and W. Sutton (eds.). The use of seedlings with resistance to P. lateralis may be a Phytophthora Diseases of Forest Trees. Proceedings From key to restoring some affected ecosystems. Based on current the First International Meetings on Phytophthoras in Forest knowledge about the pathogen and resistance in the host, we are and Wildland Ecosystems. Forest Research Laboratory, optimistic about the potential use of resistant seedlings. Even Oregon State University, Corvallis. with the use of resistant seedlings it will take decades for the Oh E.S., and E. M. Hansen. 2002. Mechanisms of resistance in development of large Port-Orford-cedar in the disease infested Port-Orford-cedar to Phytophthora lateralis. Proceedings areas. Some questions remain to be answered including: 1) how of Mycological Society of America. p. 69. many types of resistance are there, and 2) how durable is the Sniezko, R.A., L.J. Elliott, D.J. Goheen, K. Casavan, E.M. resistance? Answers to these questions will affect disease Hansen, C. Frank, and P. Angwin. 2003. Development of screening and breeding options, and ultimately the utility of using Phytophthora lateralis resistant Port-Orford-cedar for genetically resistant seedlings in restoration. The resistance- restoration in the Pacific Northwest. Proceedings of 17th breeding program is in its early stages, but is dynamic and will North American Forest Biology Workshop, July 15-18, continue to incorporate new information. 2002, Pullman, Washington, p. 5-8. Sniezko, R.A., and E.M. Hansen. 2003. Breeding Port-Orford- cedar for resistance to Phytophthora lateralis: current status Acknowledgements & considerations for developing durable resistance. Pages 197-201 in, J.A. McComb, G.E. St. J. Hardy and I.C. Many have contributed to the success of the resistance Tommerup (eds.). Phytophthora in Forests and Natural program: the Forest Service and BLM staff who made field Ecosystems. 2nd International IUFRO Working Party selections, Oregon State University for screening and technical 7.02.09 Meeting, Albany, W. Australia 30th Sept. Ð 5th Oct. advice, and the various landowners for their interest and allowing 2001. Murdoch University Print. tree selections. Of particular note: Scott Kolpak, Chuck Frank, Sniezko, R.A., and E.M. Hansen. 2000. Screening and breeding Frank Betlejewski, Kirk Casavan, and Jim Hamlin. program for genetic resistance to Phytophthora lateralis in

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 29 PHYTOPHTHORA DISEASE ECOLOGY

Port-Orford-cedar (Chamaecyparis lawsoniana): early the First International Meetings on Phytophthoras in Forest results. Pages 91-94 in, E.M. Hansen and W. Sutton (eds.). and Wildland Ecosystems. Forest Research Laboratory, Phytophthora Diseases of Forest Trees; Proceedings From Oregon State University, Corvallis. the First International Meetings on Phytophthoras in Forest USDI-BLM and USDA-FS. 2004. Management of Port-Orford- and Wildland Ecosystems. Forest Research Laboratory, Cedar in Southwest Oregon: Final Supplemental Oregon State University, Corvallis. Environmental Impact Statement. USDI-BLM Sniezko, R.A., E.M. Hansen, A. Bower, D. Goheen, K. Marshall, OregonÐWashington State Office and USDA-FS Pacific K. Casavan, and W. Sutton. 2000. Genetic resistance of Northwest Region, Portland, Oregon. Port-Orford-cedar (Chamaecyparis lawsoniana) to USDA-FS and USDI-BLM. 2003. A range-wide assessment of Phytophthora lateralis: results from early field trials. Pages Port-Orford-cedar (Chamaecyparis lawsoniana) on Federal 138-140 in, E.M. Hansen and W. Sutton (eds.). lands. USDA Forest Service and USDI Bureau of Land Phytophthora Diseases of Forest Trees; Proceedings From Management, Portland Oregon.

1 Table 1. Percent survival of seedling progeny from twelve parents tested in 2000. Greenhouse root dip screening and field planting results.

Parent No. of Average Percent Survival

2 3 ID Crosses Greenhouse Field Planting

118569 3 6 11

510042 4 22 17

510044 5 34 28

118562 5 41 31

510008 2 44 33

CF1 4 47 36

510049 4 47 45

117344 6 48 30

510041 3 51 32

118573 3 53 32

CF2 4 64 45

117490 3 98 81

1 The 12 parents were from a range of resistance levels.

2 Percent survival measured 9 mos. after inoculation.

3 Percent survival at OSU raised beds recorded 11 mos. after planting (mortality includes non-P. lat causes).

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 30 PHYTOPHTHORA DISEASE ECOLOGY

1 Table 2. Correlations of mortality results among three field sites and OSU greenhouse testing in 2000.

2 3 4 Bill Creek Camas Valley OSU Raised Bed

r P r P r P

OSU Root Dip 0.65 0.0003 0.84 < 0.0001 0.74 < 0.0001

1 Pearson’s correlation coefficient and P-value testing null hypothesis r = 0. Correlations using parents used in Factorial 2000 design (non-random, selected parents).

2 3 4 Mortality at 28 months Mortality at 29 months Mortality at 11 months

Figure 1. Mortality of 26 Families in Greenhouse and Field Site From 2000 Testing esting T lat mortality) alley Mortality V (29 mos) Field Camas (incl. non-P OSU Root Dip Mortality (10 mos) Greenhouse Testing

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 31 PHYTOPHTHORA DISEASE ECOLOGY

Figure 2. Distribution of Port-Orford cedar.

Port-Orford-cedar Breeding Blocks

Highway BB3 Cities Reedsport Port-Orfrd cedar BB1 State line Breeding Block Coos Bay I-5 Roseburg

Grants Pass 140 Gold Beach Medford Ashland Oregon Brookings California Crescent Happy Yreka City Camp

Pacific 96 Shasta City

Ocean 101 Weed BB6

Eureka BB5 BB2 I-5 299 Redding 9/12/2000

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 32 WILDLIFE ECOLOGY

USING A WIDE-SCALE LANDBIRD MONITORING NETWORK TO DETERMINE LANDBIRD DISTRIBUTION AND PRODUCTIVITY IN THE KLAMATH - SISKIYOU REGION

John D. Alexander Klamath Bird Observatory, P.O. Box 758, Ashland, OR 97520 Email: [email protected]

C. John Ralph, Kimberly Hollinger and Bill Hogoboom USDA Forest Service, Redwood Sciences Laboratory, Pacific Southwest Research Station, 1700 Bayview Drive, Arcata, CA 95521

Abstract

The Klamath Bird Observatory and U.S. Forest Service’s Redwood Sciences Laboratory have developed the Klamath Demographic Monitoring Network, a comprehensive bird-monitoring network in southern Oregon and northern California. The Network integrates bird conservation objectives into the ecosystem management process by incorporating academic, scientific, management, and conservation interests to inform the management and conservation process with science. We present examples of how our approach to studying bird populations in the Klamath-Siskiyou Region provides information about bird distribution and demographics. We demonstrate how the Network provides a tool for generating management-related hypotheses and resulting studies.

Introduction standardized bird monitoring methods that contribute to the Network’s GIS-based regional data center. e have developed a comprehensive bird-monitoring network in southern Oregon and northern California, Study Area and Methods WWthe Klamath Demographic Monitoring Network. This Network promotes a science-based approach to integrating bird The Network covers the KlamathÐSiskiyou Ecoregion and conservation objectives into the ecosystem management process. the Upper Klamath Basin (Klamath-Siskiyou Region). Both stand Through partnership development we incorporate academic, out nationally as areas of conservation concern (USBR 1998; scientific, management, and conservation interests to produce DellaSala et al. 1999; Shuford 2000). Our study area information that informs the management process and encompasses this region, transcending state, agency and conservation of birds. ownership boundaries. The Network (Figure 1) extends from the Network cooperators (e.g., Ralph and Hogoboom, in Eel and Russian Rivers in northern California north to the Rogue review) study bird populations and provide information about bird River in southern Oregon, and from the Pacific Coast east to the distribution, population trends at various scales, and population headwaters of the Sacramento, Trinity, Rogue, and Klamath demographic factors that drive population change. Birds are an Rivers. effective tool for monitoring biological change because: (1) there The Network is made up of a multi-layered system for are many species in a region, each of which are easily and inventory and monitoring of landbirds on a watershed basis. The inexpensively detected using standardized sampling protocols; (2) specific methods have been codified in Ralph et al. (1993). The these species respond to a wide variety of habitat conditions and Partners In Flight Monitoring Working Group and the U.S. Forest climatic perturbations; and (3) accounting for and maintaining Service have set these methods as standards for monitoring many species with different ecological requirements can be used landbirds in North America (Manley et al. 1993). The methods to implement landscape-scale conservation strategies (Hutto are implemented throughout the year, and are especially 1998). Landbird monitoring provides excellent data on the effects concentrated during the summer breeding season (May-June) and of past and present manipulations on a watershed basis as birds during the late summer and fall dispersal and migration seasons can be site specific and show associations with measurable habitat (July-October). The Network’s monitoring strategy uses several characteristics (Alexander 1999; Hutto and Young 1999; Drapeau monitoring methods applied at multiple spatial and temporal et al. 2000; Detttmers et al. 2002). scales. Bird census techniques are used to provide information We have successfully engaged voluntary participation from about species presence/absence, richness, relative abundance, academics, federal and state agencies, non-government breeding status, population trends, seasonality, and habitat organizations, and private landowners. We promote the use of associations (e.g., Nur et al. 1999). Demographic monitoring

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 33 WILDLIFE ECOLOGY using constant-effort mist netting and nest finding compliments mixed-conifer hardwood forests of the western portions of the the census techniques adding information about productivity, area (Figure 3). survivorship, and cause of population change (Nur et al. 1999). This information is of interest to the scientific community who study similarities and differences of bird communities and Bird Census bird diversity across the landscape. The Green-tailed Towhee and Breeding season bird censuses involve conducting point Swainson’s Thrush examples begin to demonstrate how count surveys at two scales (Ralph et al. 1993). Extensive scale differences in bird communities across the region represent the a censuses are done over multiple watersheds, using on- and off- gradient of habitat types, from the eastern pine and shrub-steppe road counts in both upland and riparian habitats, thus providing systems, through the mixed conifer/hardwood forests, and into the information at a landscape level. Intensive censuses are usually major watersheds of coastal Douglas fir and redwood forests. conducted with the demographic methods described below, and Understanding general bird distribution patterns and habitat are concentrated in the riparian zones or meadow edges, or in relationships is the first step to understanding the effects of habitat association with specific management efforts. These intensive condition on birds, a topic of interest to land managers and the efforts compliment the extensive dataset with site-specific conservation community. In the case of the Swainson’s Thrush, a information. Area search censuses involve a person searching two Forest Service “Management Indicator Species,” that is declining to three small areas for 20 minutes in each area (Ralph et al. in the northwestern U.S. and Canada, our distribution data inform 1993). The method is especially applicable during the non- managers about the importance of a species of conservation breeding season. concern with regards to the probability of its occurrence on particular management units. The habitat relationships data help Constant Effort Mist Netting managers understand which Management Indicator Species are of Demographic measures are primarily collected using potential concern in a particular project within their management constant-effort mist netting (e.g., Ralph et al. 1993). The method unit. involves permanent stations, usually placed in a water-associated or meadow riparian zone to maximize capture rate and to monitor Effects of Management birds in these sensitive habitats. Nets are operated at least once To illustrate how our data apply to adaptive management we every 10 days during breeding, dispersal and migration seasons. consider the Northwest Forest Plan (USDA and USDI 1994) and By capturing birds at mist-netting stations, we are able to Late Successional Reserve (LSR) management. Federal land determine the age and sex of individuals, providing information management plans use diameter at breast height (DBH), canopy about population demographics and thus the potential causes of height, and canopy cover to describe current and desired changes in abundance. conditions. Here we consider bird species that have been Both point counts and mist netting have been thoroughly identified as old-growth associates by the Forest Ecosystem investigated in the past few years and have been the subject of Management Assessment Team (USDA et al. 1993). They include: two international symposia (Ralph et al. 1995; Ralph and Dunn in Hairy Woodpecker (Picoides villosus), Northern Flicker (Colaptes press) with the objective of determining the statistical power of auratus), Pileated Woodpecker (Dryocopus pileatus), Hammond’s specific applications of the techniques. As a partial result of these Flycatcher (Empidonax hammondii), Pacific-slope Flycatcher symposia, specific standard protocols are now in place, along with (Empidonax difficilis), Chestnut-backed Chickadee (Parus area search protocols, and are incorporated into the handbook on rufescens), Red-breasted Nuthatch (Sitta canadensis), Brown methods (Ralph et al. 1993). Creeper (Certhia americana), Golden-crowned Kinglet (Regulus satrapa), Hermit Thrush (Catharus guttatus), Warbling Vireo Results and Discussion (Vireo gilvus), Hermit Warbler (Dendroica occidentalis), and Wilson’s Warbler (Wilsonia pusilla). We hypothesized that these Bird Distribution and Habitat Relationships old growth associated birds would show a relationship with Over 15,000 census stations have been established habitat characteristics used to describe forest conditions. Forest throughout the Klamath-Siskiyou Region providing baseline point count census stations where the old-growth bird abundance information about the regional distribution of birds. For example, is greater are characterized by taller canopy heights (Figure 4). the Green-tailed Towhee (Pipilo chlorurus) is an abundant breeder Hermit Warblers show a positive correlation with canopy cover in the eastern portions of our study area, in contrast to the (Figure 5). Swainson’s Thrush (Catharus ustulatus), a common breeder in the This example illustrates our efforts to guide management by western segment of the study area (Figure 2). Habitat data focusing analyses of Network data down to a watershed scale to collected at each census station provide us with an opportunity to predict the effects of specific actions on species of conservation develop and begin testing the hypothesis that general habitat concern. By basing analyses on variables used to describe current relationships in part, drive the landscape distribution for these and desired conditions in federal land management plans we species. The towhee is associated with Pine and Juniper habitats provide managers with information to inform management that dominate our eastern landscapes and the thrush breeds in the decisions and establish ecologically based targets. These data

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 34 WILDLIFE ECOLOGY provide information to help determine if management actions considerations beyond the traditional focus on breeding benefit the species assemblages they are supposed to protect. populations. Our data also can provide information to managers helping to assure proper documentation of the potential negative Science-Based Conservation effects of management actions on species of conservation concern. The data presented have been derived from implementing Within the adaptive management and ecosystem management monitoring plans that were developed as part of the Partners In frameworks our data identify “Indicator Species” (species that Flight International Land Bird Conservation Program. In represent specific management related habitat conditions) and can implementation we have integrated our monitoring strategy with be used to develop tools for effectiveness monitoring. To promote regional land management programs. The draft Coniferous Forest the development of old-growth characteristics in LSRs, and to Bird Conservation Plan: A Strategy For Protecting and Managing reduce the threat of stand replacing fires land managers are Coniferous Forest Habitats and Associated Birds in California implementing various forest thinning projects (USDA and USDI (CalPIF 2002) is an example of how Klamath Demographic 2004). By examining the relationship of birds of management Monitoring Network efforts have resulted in products that further concern with natural variations in DBH, canopy height, and the integration of bird conservation and land management by canopy cover we make predictions with regards to the effects of providing management recommendations in an easily referenced management. format. As a result of the development of bird conservation plans Our results can then be used to set thresholds for thinning a recent Presidential Executive Order (Federal Register 2001) has projects to ensure forests are not over-thinned, causing a reduction mandated a series of Memoranda of Understanding (MOUs) of canopy cover that would be detrimental for old-growth between the Fish and Wildlife Service and various land associated landbird species. With our standardized bird management agencies. These MOUs obligate agencies to the monitoring techniques we can implement efficient and cost- International Migratory Bird Treaty Act and challenge them to effective monitoring programs. By monitoring the abundance of integrate Partners In Flight bird conservation plans, such as indicator species we can test predictions within an adaptive CalPIF (2002), and their objectives. management framework, and if species begin to decline, management targets can be revised fulfilling mandates to monitor Conclusions the effects of management on species of concern within the Ecosystem Management Framework. In developing and maintaining a conservation-based bird monitoring network, and to better understand the dynamics Demographic Monitoring between demographic parameters and distribution at various Using Network mist-netting data collected in riparian scales, it is critical that multiple methods be employed at multiple habitats in the watersheds of the Klamath, Trinity, and Rogue special and temporal scales. The Klamath Demographic rivers, Ralph and Hollinger (2003) reported adult Willow Monitoring Network provides an additional tool for generating Flycatchers breeding in and migrating through the Klamath- management related hypotheses and studies to inform the land and Siskiyou Region. During the breeding season they are captured in wildlife management. As a result we can use the birds as relatively small numbers, as compared to the much higher indicators of current and desired conditions and use the network numbers in the dispersal and migration seasons when many young as an effectiveness monitoring tool within the adaptive birds are captured (Figure 6). management framework. By using constant-effort mist netting throughout the The Network uses a science-based approach to further Network, we are providing data that increase our understanding conservation in our region by involving the key interests on which about breeding season productivity, over-winter survivorship, as successful conservation depends. The Network engages academic well as age- and sex-specific distribution and abundance during and scientific interests as it provides a tool for studying patterns of the breeding, dispersal, and migration seasons. By recapturing biodiversity across landscapes. Management interests are engaged previously banded birds we are calculating survivorship estimates. as it provides a tool for studying the effectiveness of management Willow Flycatchers are receiving management attention in actions. Conservation interests are engaged as it provides a tool the Klamath Siskiyou Region due to population declines that have for integrating bird conservation objectives into land management. been documented in the Southwestern U.S. (Sogge et al. 2003). The Network has developed a partnership that links the With a look at data regarding migrant bird distribution we get a educational resources of academic community with the needs of better appreciation of the influence of how management in our the neighboring community-at-large. It brings scientific and region can have effects that reach well beyond this area. Young educational resources to the table with community participants to Willow Flycatchers that were hatched outside of the Region inform land management decision processes with excellent concentrate along the main rivers and are dependent on riparian science. Through this partnership we connect theoretical studies to habitats during dispersal and migration seasons. By focusing the practical world of land and wildlife management to promote monitoring efforts during migration, as well as the breeding bird conservation with an understanding that stable bird seasons, the Region’s Network is broadening management populations represent sustainability at the local, regional and

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 35 WILDLIFE ECOLOGY global scales. The Network brings all interests to the table to National Forest System Lands. USDA Forest Service, demonstrate the power of human knowledge and collaboration, Wildlife and fisheries Staff, Washington, D.C. July 1993. providing models for appreciating endangered ecosystems and Nur, N. S.L. Jones, and G.R. Geupel. 1999. A statistical guide to building sustainable communities and industries. Through data analysis of avian monitoring programs. U.S. hypothesis testing and effectiveness monitoring we can learn from Department of the Interior, Fish and Wildlife Service, BTP- experience to better understand how to practice stewardship of our R6001-1999, Washington, D.C. landscape. Ralph, C.J. and K. Hollinger. 2003. The status of the Willow and Pacific-slope Flycatchers in northwestern California and Acknowledgements southern Oregon. In M.K. Sogge, B.E. Kus, S.J. Sferra, and M.J. Whitfield. (Eds) Ecology and Conservation of the The Klamath Demographic Monitoring Network has Willow Flycatcher. Cooper Ornithological Society. Studies resulted from the contribution of various partners that represent in Avian Biology. 26. federal and state land management agencies, corporations, and Ralph, C. J., G. R. Guepel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of field methods for monitoring private contributors and land owners. Countless volunteer intern landbirds. USDA Forest Service General Technical Report students have been a critical force in the development of the PSW-GTR-144. Network. This manuscript benefited from reviewer comments Ralph, C.J., J.R. Sauer, and S. Droege. Monitoring bird from Joel Pagel, Dennis Vroman and Nathaniel Seavy. populations by point counts. General Technical Report PSW-GTR-149. Pacific Southeast Research Station, Forest Service, Albany, California. Literature Cited Ralph, C.J. and B. Hogoboom. In Review. 10,000 points of light in Northern California and Southern Oregon: a cooperative Alexander, J. D. 1999. Bird-habitat relationships in the census program. In C.J. Ralph and T. Rich, eds. Klamath/Siskiyou Mountains. M.S. Thesis, Southern Proceedings of the Third International Partners in Flight Oregon University, Ashland. Conference, Monterey, California. March 20-24, 2002. CalPIF (California Partners in Flight). 2002b. Version 1.0. The USDA Forest Service, PSW General Technical Report. draft coniferous forest bird conservation plan: a strategy for Ralph, C.J. and E.R. Dunn. in press. Monitoring landbird protecting and managing coniferous forest habitats and populations with mist nets. Studies in Avian Biology. associated birds in California (J. Robinson and J. Alexander, Sogge, M.K,, B.E. Kus, S.J. Sferra, and M.J. Whitfield. 2003. lead authors). Point Reyes Bird Observatory, Stinson Ecology and Conservation of the Willow Flycatcher. Beach, California. Studies in Avian Biology 26. Cooper Ornithological Society. DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt, and D.M. Shuford, W. D. 2000. Appendix X. Key shorebird areas of the Olson. 1999. A glodal perspective on the biodiversity of the Inter-Mountain West: Klamath Basin CA/OR, in Klamath-Siskiyou Ecoregion. Natural Areas Journal 19:300-319. Intermountain West regional shorebird plan, Pages 39-41 Dettmers, R., D.A. Buehler, and K.E. Franzreb. 2002. Testing (L. W. Oring, L. Neel, and K. E. Oring, eds.) Version 1.0. habitat-relationship models for forest of southeastern United Regional report of the U.S. Shorebird Conservation Plan. States. Journal of Wildlife Management 66:417-424. Manomet Center for Conservation Sciences, Manomet, Draupeau, P., A. Leduc, J-F. Giroux, J-PL. Savard, Y. Bergeron, Massachusetts (http:/www.manomet.org). and W.L. Vickery. 2000. Landscape-scale disturbances and U.S. Bureau of Reclamation. 1998. 1998 Klamath Operations changes in bird communities of boreal mixed-wood forests. Plan Environmental Assessment. U.S. Bureau of Ecological Monographs 70:423-444. Reclamation, Klamath Basin Area Office, Klamath Falls, Federal Register. 2001. Responsibilities of federal agencies to Oregon. protect migratory birds. Executive Order 13186 of January USDA and USDI. 1994. Record of decision (standards and 10. Washington DC. Vol. 66 No. 11. guidelines). USDA Forest Service and USDI Bureau Of Hutto, R. L. 1998. Using landbirds as an indicator species group. Land Management. Portland, Oregon. Pages 75-92 in J. M. Marzluff and R. Sallabanks, eds. USDA and USDI. 2004. The healthy forests initiative and healthy Avian conservation: research and management. Island Press, forests restoration act (interim field guide). USDA Forest Covelo, California. Service and USDI Bureau of Land Management. Missoula, Hutto, R.L. and J.S. Young. 1999. Habitat relationships of Montana. landbirds in the Northern Region, USDA Forest Service. USDA, USDI, USDC and Environmental Protection Agency. USDA Forest Service General Technical Report. RMRS- 1993. Forest ecosystem management: an ecological, GTR-32. economic, and social assessment. USDA Forest Service, Manley, P.N., W.M. Block, F.R. Thompson, G.S. Butcher, C. USDI USFWS, USDC National Oceanic and Atmospheric Paige, L.H. Suring, D.S. Winn, D. Roth, C.J. Ralph, E. Administration National Marine Fisheries Service, USDI Morris, C.H. Flather, and K. Byford. Guidelines for National Park Service, USDI BLM, Environmental monitoring populations of neotropical migratory birds on Protection Agency. Portland, Oregon.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 36 WILDLIFE ECOLOGY

Figure 1. The Klamath Demographic Monitoring Network and location of coordination organizations: Klamath Bird Observatory (KBO), Humboldt Bay Bird Observatory (HBBO) and USDA Forest Service Pacific Southwest Research Station Redwood Sciences Laboratory (RSL).

Klamath Bioregion of Northern California and Southern Oregon

Klamath Bird Observatory (KBO) Ashland, Oregon

Redwood Sciences LEGEND Laboratory MIST NETTING STATIONS (RSL) and Humboldt RIVERS Bay Bird Observatory KLAMATH- (HBBO) SISKIYOU REGION Arcata, California

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 37 WILDLIFE ECOLOGY

Figure 2. Percent of bird census stations within 15 Klamath Demographic Monitoring Network study areas where Green-tailed Towhees and Swainson’s Thrushes were detected.

Green-tailed Towhee

Major Rivers

% of Stations /Species

100% 75% 50% 25% 0% Swainson’s Thrush

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 38 WILDLIFE ECOLOGY

Figure 3. Distribution of bird census stations where Green-tailed Towhees and Swainson’s Thrushes among 10 habitat type classifications.

Green-tailed Towhee

CONIFER CONIFER HARDW HARDW FIR PINE OAK JUNIPER SHRUB GRASS PERCENT OF STATIONS HARDW /CONIFER

Swainson’s Thrush

CONIFER CONIFER HARDW HARDW FIR PINE OAK JUNIPER SHRUB GRASS PERCENT OF STATIONS HARDW /CONIFER

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 39 WILDLIFE ECOLOGY

Figure 4. Average tree height at bird census stations where 0-6 old-growth associated bird species were detected. (FEET) MEAN TREE HEIGHT

Figure 5. Average number of Hermit Warblers detected at bird census stations classified into 4 canopy cover groups.

3

2

BIRDS/ 1 STATION

0

0-20% 21-40% 41-60% 61-100%

CANOPY COVER

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 40 WILDLIFE ECOLOGY

Figure 6. Relative abundance of Willow Flycatcher age and sex classes (UU-unknown age and sex, HY-hatching year unknown sex, and AF-adult female, AM-adult male,) at constant effort mist netting stations in 4 Klamath Demographic Monitoring Network study areas as reported in Ralph and Hollinger (2003).

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 41

WILDLIFE ECOLOGY

UNDERSTANDING EFFECTS OF FIRE SUPPRESSION, FUELS TREATMENT, AND WILDFIRE ON BIRD COMMUNITIES IN THE KLAMATH-SISKIYOU ECOREGION

John D. Alexander Klamath Bird Observatory, P.O. Box 758, Ashland, OR 97520

C. John Ralph and Bill Hogoboom USDA Forest Service, Redwood Sciences Laboratory, 1700 Bayview Drive, Arcata, CA 95521

Nathaniel E. Seavy1 Department of Zoology, University of Florida, Gainesville, FL 32611-8029 Email: [email protected]

Stewart Janes Department of Biology, Southern Oregon University, Ashland, OR 97520

Abstract

Although fire management is increasingly recognized as an important component of conservation in Klamath-Siskiyou ecosystems, empirical evidence on the ecological effects of fire in this region is limited. Here we describe a conceptual model as a framework for understanding the effects of fire and fire management on bird abundance. This model identifies three major pathways through which fire may influence bird abundance: habitat structure, food availability, and predation. Based on this model, we review major research questions regarding the ecological consequences of fire and fire management for bird conservation in the Klamath-Siskiyou Ecoregion. These questions illustrate that we still know relatively little about how natural and anthropogenic disturbances change these systems. We suggest that this ecological information is critical for informing fire-related management decisions in the Klamath-Siskiyou Ecoregion.

Introduction unnaturally high fuel conditions (Parsons and Botti 1996). Although goals of preservation initially led to strict policies of fire ire regimes create predictable changes in landscape suppression in National Parks, biologists later came to recognize composition. Fire induced disturbance has influenced the the ecological importance of natural fire (Leopold et al. 1963). FFevolution of life history strategies of many organisms and Today, fire ecologists recognize that (1) restoring natural fire landscape-level patterns of diversity (Pickett 1976). With fire regimes is often impractical; instead successful fire management suppression programs carried out in the last century, natural fire will require repeated management intervention tailored to meet patterns have been altered (Agee 1993). These changes are specific goals, (2) achieving these goals requires sound believed to have changed habitat composition and led to fuel information on which to base management decisions, and (3) accumulations associated with unnaturally severe fires (Arno and because fires operate at a spatial scales that encompass state, Brown 1989; Agee 1993). The social and ecological federal, and private management agencies, inter-agency consequences of severe fires reinforce the need for widespread cooperation is important (Parsons and Botti 1996). suppression efforts, yet suppressing fires without reducing fuels The Klamath-Siskiyou Ecoregion covers more than 4 may lead to larger and more intense fires (Agee 2002). This million hectares of northern California and southern Oregon. recognition has led to shifts in fire prevention strategies that aim Biologically, this region is exceptionally diverse (DellaSala et al. to reduce fire hazards with prescribed burning and mechanical 1999), with unique bird (Ralph et al. 1991; Trail et al. 1997; fuels reductions. The challenge of contemporary fire management Alexander 1999) and vegetation (Whittaker 1960) communities. lies in understanding the ecological role of fire in natural systems The historical fire regime in much of this region was one of and how this process can be restored with fire management that is mixed-severity that created and maintained high spatial habitat compatible with social, economic, and aesthetic values. heterogeneity (Atzet and Wheeler 1982; Agee 1993; Skinner The dynamic nature of fire management policies in U.S. 1995; Taylor and Skinner 1998). Today, fire management is National Parks exemplify the challenge of balancing social, increasingly recognized as an important issue in the conservation economic, and aesthetic values with natural fire regimes and of Klamath-Siskiyou ecosystems, but empirical evidence on the ecological effects of fire in this region is limited. 1 Corresponding author.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 42 WILDLIFE ECOLOGY

Considering that the ecological effects of fire are diverse process of fire as it relates to bird abundance is rudimentary at and extensive, monitoring changes in ecological conditions is best. In order for prescribed fire and other fire management difficult. Although no single group of organisms can provide a strategies to play an effective role in avian conservation, Huff et complete picture of the changes caused by fire, birds are al. (in review) identified a number of research and management recognized as useful and cost-effective tools for ecological questions that need to be addressed. Here we present a subset of monitoring of land management practices (Greenwood et al. 1993; these questions in the context of our conceptual model and discuss Hutto 1998). Hutto (1998) emphasized that birds are an effective their relevancy to avian conservation in the Klamath-Siskiyou tool for broad-based monitoring because: (1) they are easier than Ecoregion. many other vertebrates to identify and less expensive to survey, How does fire suppression affect bird abundance and (2) point-count survey methods quickly collect data on many demographics? Fire suppression may decrease the frequency with species, and (3) documenting the response of many species with which fires burn in low- and moderate-severity fire regimes. different life-history characteristics and habitat requirements When fires become less frequent, fuels may accumulate, promotes understanding at the landscape level. Here, we present potentially altering patterns of fire severity. Such changes would a conceptual model to describe the influence of fire on bird alter the magnitude of the influence fire has on vegetation abundance and then use this model to illustrate the importance of structure in our model. Such an effect would extend to bird basic research questions for understanding the ecological effects abundance, either directly or indirectly through food availability of fire and fire management in the Klamath-Siskiyou Ecoregion. and predation. The ecological effects of fire suppression in the Klamath- A Conceptual Model of Fire’s Influence Siskiyou are not well understood (Frost and Sweeney 2000), but on Bird Abundance have probably been greatest at the landscape level (Skinner 1995). Although there have been no studies that directly address how bird communities are affected by habitat changes from fire Our conceptual model of fire’s influence on bird suppression in the Klamath-Siskiyou Ecoregion, Huff et al. (in communities describes three major pathways through which fire review) noted that bird communities most vulnerable to fire- induced changes may influence bird abundance: habitat structure, suppression induced changes are likely those in low-severity, predation, and food availability (Figure 1). Perhaps the most high-frequency fire regimes (e.g., mixed-conifer, oak woodlands, obvious effects of fire are changes in habitat structure, such as the and chaparral) that are well represented in the Klamath-Siskiyou reduction in canopy cover or increases in the number of snags. Ecoregion. Because fire suppression may continue in the The abundance of many bird species in the Klamath-Sikiyou Klamath-Siskiyou Ecoregion, the long-term effects on bird Ecoregion has been documented to be correlated with habitat communities should be evaluated. structure (Alexander 1999). A simple direct effect of habitat How does prescribed fire change conditions for birds? structure on bird abundance may be one of the factors influencing Prescribed fire is increasingly recognized as a potential tool for post-fire bird abundance. Additionally, habitat structure may landscape management. Yet the ability of prescribed fire to create influence bird abundance indirectly if it changes environmental the desired ecological conditions for birds is not well understood conditions, such as food availability and predation. Variation in (Artman et al. 2001; Jones et al. 2002). In the Klamath-Siskiyou, bird abundance across habitats has also been linked to food fires historically occurred in the late summer or early fall (Taylor availability (Johnson and Sherry 2001). If insect abundance and Skinner 1998). In contrast, late winter or early spring provide changes after fire (Abbott et al. 2003), such changes may explain weather and moisture conditions when burns are more easily changes in bird abundance. Such an effect may occur controlled (Biswell 1989). Because prescribed fire can be applied independently of changes in habitat structure. Predation, in many different ways and have many different effects on especially on nests, has been documented as a factor influencing vegetation structure, food availability, and predation, bird communities (Martin 1988). If fire changes predation generalizations are not easily drawn that can be applied widely to pressure directly (independent of changes in vegetation structure), birds. Thus, specific studies from the Klamath-Siskiyou then this factor should also be considered. Ecoregion may have important management implications for bird conservation. Fire Research Questions How do fuels treatments affect bird abundance? for the Klamath-Siskiyou Region Although mechanical forest management, (i.e., logging), may produce structural changes similar to natural fire, the effects of Information on avian-fire relationships within the these disturbances are not always the same for birds (Imbeau et al. Maritime Pacific Northwest has been summarized by Huff and co- 1999). The ability of mechanical activities to mimic the effects of workers (in review). This review illustrates that the effects of fire fire may depend whether or not they replicate food availability or on bird communities in the Pacific Northwest are not well predation conditions that are created by natural fire. As fuels understood. In the Klamath-Siskiyou Ecoregion there have been reduction activities are implemented in the Klamath-Siskiyou no studies that specifically address the effect of fire on birds (Huff Ecoregion, these activities should be closely monitored with et al. in review). Thus, our understanding of the ecological respect to their effect on bird abundance and demography.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 43 WILDLIFE ECOLOGY

The challenge of understanding the relationship between fuels reduction and bird conservation is illustrated in oak Acknowledgments woodland and chaparral habitats of the Klamath-Siskiyou Ecoregion. Because high-severity fires characteristic of these habitats have the potential to cause extensive damage to homes This paper was stimulated by research funded by the and property, fires are a concern in many of these areas. In an Joint Fire Sciences Program, the USDA Forest Service Pacific attempt to reduce the landscape risk of fire and maintain habitat Southwest and Pacific Northwest Regional Partners In Flight Bird characteristics created by fire, the Bureau of Land Management Programs, the Fremont-Winema, Rogue, and Klamath National has introduced fuels reduction projects throughout large portions Forests, and the USDI Bureau of Land Management Oregon State of the Applegate Valley in southwest Oregon. These habitats Office and Lakeview District. We are grateful to Mary support a unique bird community, including species such as Acorn Rasmussen and Crater Lake National Park for administering this Woodpecker (Melanerpes formicivorus), Ash-throated Flycatcher project. Discussions with J. Heinzelmann, D.C. Lee, S. Miller, (Myiarchus cinerascens), Wrentit (Chamaea fasciata), and White- and M. Widdowson helped develop the conceptual model. breasted Nuthatch (Sitta carolinensis). The effects of habitat Comments on a draft of this manuscript were provided by J. Pagel management on these birds have been addressed by regional and an anonymous reviewer. Partners in Flight conservation plans (Altmann 2000a). In cooperation with the Medford District of the Bureau of Land Literature Cited Management, we are using these plans to identify desired habitat conditions for oak woodland birds and bird monitoring to compare Abbott, I., T. Burbidge, K. Strehlow, A. Mellican, and A. Wills. treated and untreated areas. With this monitoring, the ability of 2003. Logging and burning impacts on cockroaches, fuels reduction treatments to achieve the desired results for bird crickets and grasshoppers, and spiders in Jarrah forest, conservation can be evaluated. Western Australia. Forest Ecology and Management What are effects of post-fire salvage? Post-fire salvage 174:383-399. logging removes a range of remaining structure from burned Agee, J.K. 1993. Fire ecology of Pacific Northwest forests. areas. This structure may provide important nest sites and Island Press, Washington, D.C. foraging opportunities for many cavity-nesting and bark-foraging Agee, J.K. 2002. The fallacy of passive management: managing birds (Hutto 1995; Kreisel and Stein 2000). Although salvage for firesafe forest reserves. Conservation Biology in logging has been hypothesized to have negative consequences for Practice 3:19-25. birds that rely on this structure (Hutto 1995; Wales 2001), this Alexander, J.D. 1999. Bird-habitat relationships in the effect has not been well documented (Haggard and Gaines 2001). Klamath/Siskiyou Mountains. M.S. Thesis, Southern Effects of salvage may interact with burn severity; removing Oregon University, Ashland. structure from severely burned areas may have much different Altman, B. 2000a. Conservation strategy for landbirds in effects than the same level of extraction from less severely burned lowlands and valleys of western Oregon and Washington. areas (Saab and Dudley 1998). Considering the potential for Oregon-Washington Partners in Flight, Boring, Oregon. extensive salvage logging operations in the Biscuit Fire of 2002, Altman, B. 2000b. Conservation strategy for landbirds of the the ecological effects of this activity deserves more study. east-slope of the Cascade Mountains in Oregon and The conservation of Klamath-Siskiyou ecosystems Washington. Oregon-Washington Partners in Flight, depends in part on our understanding of natural fire regimes and Boring, Oregon. related ecological processes. Fire is clearly an important agent of Arno, S.F., and J.K. Brown. 1989. Managing fire in our forests - disturbance that influences successional processes and habitat time for a new initiative. Journal of Forestry 87:44-46. composition in these landscapes. These processes influence Artman, V.L., E.K. Sutherland, and J.F. Downhower. 2001. landbird species that are dependent on specific habitat attributes Prescribed burning to restore mixed-oak communities in created by disturbances such as fire (Brawn et al. 2001). southern Ohio: effects on breeding bird populations. Recently, directives in federal land management agencies have Conservation Biology 15:1423-1434. emphasized the need to restore habitats to conditions that are Atzet, T., and D.L. Wheeler. 1982. Historical and ecological ecologically sustainable. This need is reflected in the fact that perspectives on fire activity in the Klamath geological current landbird conservation plans and strategies (e.g., Altman province of the Rogue River and Siskiyou National 2000a, 2000b) have emphasized collecting information about the Forests. USDA Forest Service Report, R-6-Range-102. effects of fire suppression, fuels treatment, and wildfire restoration Biswell, H. 1989. Prescribed burning in California wildlands on the distribution of fire dependent bird species. There is clearly vegetation management. University of California Press, a need for information from the Klamath-Siskiyou Ecoregion that Berkeley. can be used to understand the role of fire and fire management in Brawn, J.D., S.K. Robinson, and F.R. Thompson III. 2001. The landbird conservation and the restoration of ecological processes. role of disturbance in the ecology and conservation of

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 44 WILDLIFE ECOLOGY

birds. Annual Review of Ecology and Systematics Leopold, A.S., S.A. Cain, C.M. Cottam, J.M. Gabrielson, and T.L. 32:251-276. Kimball. 1963. Wildlife management in the national DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt, and D.M. parks. American Forestry 69:32-35, 61-63. Olson. 1999. A global perspective on the biodiversity of Martin, T.E. 1988. Habitat and area effects on forest bird the Klamath-Siskiyou Ecoregion. Natural Areas Journal assemblages: is nest predation an influence? Ecology 19:300-319. 69:74-84. Frost, E.J., and R. Sweeney. 2000. Fire regimes, fire history, and Parsons, D.J., and S.J. Botti. 1996. Restoration of fire in national forest conditions in the Klamath-Siskiyou region: An parks. Pages 29-31 in, C.C. Hardy and S.F. Arno (eds.), overview and synthesis of knowledge. World Wildlife The use of fire in forest restoration. USDA Forest Fund, Ashland, Oregon. Service General Technical Report INT-GTR-341. Greenwood, J.J. D., S.R. Baillie, H.Q.P. Crick, J.H. Marchant, and Pickett, S.T.A. 1976. Succession: an evolutionary interpretation. W.J. Peach. 1993. Integrated population monitoring: American Naturalist 110:107-119. detecting the effects of diverse changes. Pages 267-342. Ralph, C.J., P.W.C. Paton, and C.A. Taylor. 1991. Habitat in, R.W. Furness and J.J.D. Greenwood (eds.). Birds as association pattern of breeding birds and small mammals monitors of environmental change. Chapman and Hall, in Douglas-fir/hardwood stands in northwestern London. California and southwestern Oregon. Pages 379-394 in, Haggard, M., and W.L. Gaines. 2001. Effects of stand- L.F. Ruggiero, K.B. Aubry, A.B. Carey, and M.H. Huff replacement fire and salvage logging on a cavity-nesting (tech. coords.). Wildlife and vegetation of unmanaged bird community in eastern Cascades, Washington. Douglas-fir forests. USDA Forest Service General Northwest Science 75:387-396. Technical Report PNW-GTR-285. Huff, M.H., N.E. Seavy, J.D. Alexander, and C.J. Ralph. In Saab, V.A., and J.G. Dudley. 1998. Responses of cavity nesting Review. Fire and birds in Maritime Pacific Northwest. birds to stand-replacement fire and salvage logging in Studies in Avian Biology. ponderosa pine/Douglas-fir forests of southwestern Hutto, R.L. 1998. Using landbirds as an indicator species group. Idaho. USDA Forest Service Research Paper RMRS-RP- Pages 75-92 in, J.M. Marzluff and R. Sallabanks (eds). 11. Avian conservation: research and management. Island Skinner, C.N. 1995. Change in spatial characteristics of forest Press, Washington, D.C. openings in the Klamath Mountains of northwestern Hutto, R.L. 1995. Composition of bird communities following California, U.S.A. Landscape Ecology 10:219-228. stand-replacement fires in northern Rocky Mountain Taylor, A.H., and C.N. Skinner. 1998. Fire history and landscape (U.S.A.) conifer forests. Conservation Biology 9:1041- dynamics in late-successional reserve, Klamath 1058. Mountains, California, U.S.A. Forest Ecology and Imbeau, L., J.-P.L. Savard, and R. Gagnon. 1999. Comparing Management 111:285-301. bird assemblages in successional black spruce stands Trail, P.W., R. Cooper, and D. Vroman. 1997. The breeding birds originating from fire and logging. Canadian Journal of of the Klamath/Siskiyou region. Pages 158-174 in, J.K. Zoology 77:1850-1860. Beigel, E.S. Jules and B. Snitkin (eds.). Proceedings of Johnson, M.D., and T.W. Sherry. 2001. Effects of food the First Conference on Siskiyou Ecology, Siskiyou availability on the distribution of migratory warblers Regional Education Project, Cave Junction, Oregon. among habitats in Jamaica. Journal of Animal Ecology Wales, B.C. 2001. The management of insects, diseases, fire, and 70:546-560. grazing and implications for terrestrial vertebrates using Jones, D.D., L.M. Conner, R.J. Warren, and G.O. Ware. 2002. riparian habitats in Eastern Oregon and Washington. The effect of supplemental prey and prescribed fire on Northwest Science 75:119-127. success of artificial nests. Journal of Wildlife Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, Management 66:1112-1117. Oregon, and California. Ecological Monographs 30:279- Kreisel, K.J., and S.J. Stein. 1999. Bird use of burned and 338. unburned coniferous forests during winter. Wilson Bulletin 111:243-250.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 45 WILDLIFE ECOLOGY

Figure 1. A conceptual model describing the effect of fire on bird abundance. Fire may change habitat structure, food availability, and predator pressure. In turn, these factors may influence bird abundance through a variety of pathways.

Fire

Predator Food density availability

Habitat structure

Predation

Foraging success

Bird species abundance

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 46

WILDLIFE ECOLOGY

POPULATION TRENDS AMONG LANDBIRDS OF THE KLAMATH-SISKIYOU ECOREGION: AN ANALYSIS OF BREEDING BIRD SURVEY DATA

Pepper W. Trail Adjunct Professor, Biology Department, Southern Oregon University, 2011 Crestview Drive, Ashland, OR 97520 Email: [email protected]

Abstract

The Klamath-Siskiyou Ecoregion of northern California and southern Oregon is an area of great conservation interest due to its diverse ecological communities and extensive wildlands. This paper examines the status of the region’s bird populations, using data summaries from the Breeding Bird Survey (BBS), the most comprehensive database on long-term population trends for North American birds. The “Southern Pacific Rainforests” unit of the BBS includes much of the Klamath-Siskiyou, providing analytical summaries of population trends for landbird species of the region. Birds exhibiting a variety of habitat preferences, nesting strategies, and migration patterns were compared. The results indicate that native landbirds of the Klamath-Siskiyou have undergone broad declines over the past 35 years: 34% of species have experienced statistically significant declines, while only 13% have exhibited significant population increases over this time span. Population declines were particularly dramatic among neotropical migrants, with 39% of species showing significant declines and only 6% showing significant increases. Permanent resident species, in contrast, showed declines and increases in roughly equal numbers: 25% significantly decreased and 20% significantly increased. The only group of birds showing broad increases over the past 35 years was introduced species: four of the five introduced species increased, three of them significantly.

Introduction interest, due to its high plant diversity, exceptional array of ecological communities, and extensive roadless and wilderness ramatic population declines among numerous species of areas (DellaSala et al. 1999; Strittholt and DellaSala 2001). The breeding birds have been documented in the United region’s avifauna includes 392 species, with 190 confirmed DDStates in recent decades (see reviews in Terborgh 1989; breeders (Trail et al. 1997). The Klamath-Siskiyou is notable for Hagen and Johnston 1992; Martin and Finch 1995). These the number of bird species that reach a range limit in the region, declines extend across diverse geographical and ecological emphasizing its importance as an ecological crossroads (Trail et categories, including birds of grasslands (Askins 1999; Peterjohn al. 1997). and Sauer 1999), eastern deciduous forests (Askins et al. 1990; Little published information is available on the status of Holmes and Sherry 2001), and western coniferous forests breeding bird populations in the Klamath-Siskiyou. The region is (Marshall 1988; DeSante and George 1994). home to 16 landbirds that have been placed on the “Watch List” of This paper examines the status of landbird species that the Partners in Flight conservation effort, identifying them as breed in the Klamath-Siskiyou Ecoregion of southern Oregon and species of special conservation significance (Partners in Flight northern California. This area, dominated by a series of rugged 2004). Ten of these birds are listed as “Threatened and mountain ranges, is delineated by its distinct geology of ancient Declining” species, including Band-tailed Pigeon, Rufous metamorphosed sedimentary rocks with granitic and ultramafic Hummingbird, Olive-sided Flycatcher, Oak Titmouse, and intrusions (Diller 1902). These substrates produce varied soil Wrentit. The remaining six birds are listed as “Range Restricted” types, many with unusual mineral compositions, supporting a species, including Allen’s Hummingbird, White-headed great diversity of plant communities and numerous endemic plant Woodpecker, and Hermit Warbler. species (DellaSala et al. 1999). In this paper, I present an assessment of bird population The region extends from the area of Roseburg, Oregon, trends among breeding landbirds of the Klamath-Siskiyou based south to the Yolla Bolly Mountains in the vicinity of Covelo, on Breeding Bird Survey (BBS) data. The Breeding Bird Survey California, and from the Pacific coast east to the western foothills is an annual bird survey effort conducted across the United States of the Cascades. The Klamath-Siskiyou Region turns inland just and Canada and administered by the U.S. Geological Survey south of the Klamath River mouth, and thus does not include the (USGS). Counts are carried out along roadside transects, or broad coastal plain of Humboldt County, California. Detailed “routes,” which are 24 miles (39.4 km) long and consist of 50 delineation of the area and maps can be found in Trail et al. evenly spaced, 3-minute point counts. All birds seen or heard (1997), Strittholt and DellaSala (2001), and Siskiyou Project within ?-mile (400 m) of the point are recorded. Each route is (2004). surveyed one morning per year during the early breeding season, The Klamath-Siskiyou is an area of great conservation usually in early June. Surveys begin 30 minutes before sunrise.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 47 WILDLIFE ECOLOGY The same routes are surveyed annually, ideally by the same California north to the Canadian border, and encompasses the observer. The BBS was initiated in the eastern U.S. in 1966 and coastal mountains and interior valleys west of the Sacramento implemented nationwide in 1968. Data from over 4000 survey Valley in California and Cascade Mountains in Oregon and routes are now maintained in the BBS database. The USGS Washington. One major part of the Klamath-Siskiyou Ecoregion maintains excellent analytical summaries of these data, which are that is outside the Southern Pacific Rainforests is the available on the Internet (Sauer et al. 2003a). Trinity/Marble Mountains, which are included in the BBS’s There are a number of well-known limitations of BBS “Sierra-Nevada” physiographic region. Maps of the BBS methodology (Robbins et al. 1986; Link and Sauer 1997; Sauer et physiographic regions can be found in Robbins et al. (1986) and al. 2003b). Rare bird species, including many predators and on the Web (USGS 2004b). habitat specialists, are not adequately sampled by a general survey The Southern Pacific Rainforests region is a far larger area methodology such as the BBS. The fact that all counts are done than the Klamath-Siskiyou. Despite this, the bird species present from roadsides introduces a bias in overall habitat representation, in the Southern Pacific Rainforests data correspond well to the and may result in inflated population estimates for bird species Klamath-Siskiyou avifauna (Trail et al. 1997), with the exception that favor disturbed and edge habitats. Bird species with low of some Klamath-Siskiyou birds of higher-elevation and dry forest detectibility (e.g., relatively non-vocal species) are likely to be habitats, such as Cassin’s Finch and Green-tailed Towhee. under-counted. Common and highly vocal species may also be Overall, the South Pacific Rainforests BBS region provides a under-recorded, due to observer efforts to avoid double-counting. better match for the Klamath-Siskiyou avifauna than do other Breeding Bird Survey data also have a number of unique partially overlapping BBS regions, such as the Sierra-Nevada, or strengths. Data covering a long time span are critical to any the Oregon or California state data. All these include many analysis of long-term population change. Many BBS routes have species not found in the Klamath-Siskiyou. been run continuously for over 30 years, a record unmatched by Recently, Sauer et al. (2003b) evaluated BBS data for use any other North American breeding bird survey effort. Other BBS with a newer system of geographic stratification developed for strengths are the consistent methodology, repeated routes, and migratory bird conservation planning by Partners in Flight “Bird high standards of data management and analysis. Conservation Regions.” This analysis demonstrated that either In balance, the BBS has proven to be an invaluable source system of stratification provides comparable results, indicating of information on population trends for many relatively common, that BBS data are not highly sensitive to the particular easily detected species, which include many songbirds of great stratification scheme used. Further comparisons of bird conservation interest. BBS data were essential for developing population trends in various geographical regions of western priorities and planning objectives in the nationwide Partners in North America are presented in the Results section. Flight (PIF) bird conservation effort (Carter et al. 2000). The This paper examines BBS regional population trend data California and Oregon/Washington Partners in Flight programs from 1968-2002 for breeding landbirds of the Klamath-Siskiyou have used BBS data to prepare detailed bird conservation plans (BBS routes in the western U.S. were first surveyed in 1968). for major habitats in their areas (California PIF 2004; This is based on USGS data summaries and trend analyses for the Oregon/Washington PIF 2004). Southern Pacific Rainforests region (USGS 2004c). Following USGS recommendations, only birds encountered on at least 15 Methods routes are included in the analyses summarized here. Inferences on population trends for less frequently detected species cannot be This paper is based on BBS data summaries and trend made from BBS data. analyses prepared by USGS statisticians and available to the I eliminated from this analysis those birds on the Southern public at the BBS website (Sauer et al. 2003a). Two methods of Pacific Rainforests regional list that rarely or never breed in the statistical analysis of population trends are provided by USGS: a Klamath-Siskiyou: Purple Martin, Red-eyed Vireo, Ruby- linear route-regression approach based on estimating equations crowned Kinglet, and Townsend’s Warbler (Trail et al. 1997). I (Link and Sauer 1994) and a curve-fitting route-regression also did not include data for waterbirds and shorebirds, which the approach based on locally-weighted least squares (LOESS) (James BBS survey methodology samples poorly. This eliminated et al. 1996). Comparative analyses indicate that these two Double-crested Cormorant, Great Blue Heron, Green Heron, methods produce similar results, but that Estimating Equations are Canada Goose, Wood Duck, Mallard, Common Merganser, generally more precise (Peterjohn et al. 1997). This paper relies Western Gull, Glaucous-winged Gull, Killdeer, and Spotted on trend estimates supplied by USGS using Estimating Equations Sandpiper. (see USGS 2004a for detailed discussion of the route-regression This process produced a list of 100 breeding landbird methodology). species of the Klamath-Siskiyou for which adequate BBS data USGS analyzes BBS data geographically by state and by were available for analysis (Appendix I). Ninety-five of these region. Regional analyses are stratified by physiographic regions species are native to the region. The remaining five are non- based on Aldrich’s (1963) map of life areas of North America. native (Ring-necked Pheasant, Wild Turkey, Rock Pigeon, Much of the Klamath-Siskiyou Ecoregion falls within the European Starling, and House Sparrow). The analyses “Southern Pacific Rainforests” physiographic region as defined by summarized below are concerned almost exclusively with the the BBS. This extends from the Monterey Bay area of central native bird species.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 48 WILDLIFE ECOLOGY

I subdivided the BBS data in several ways for analysis, using declines outnumbered significant increases by well over two to both “species groups” as defined by USGS and other groupings one among Klamath-Siskiyou native landbirds (34% significant developed specifically for the Klamath-Siskiyou. For each declines vs. only 13% significant increases). Overall, including grouping, the proportion of species with increasing and decreasing both statistically significant and non-significant trends, nearly trends was compared. The different groups to which each species two-thirds (64%) of species showed population declines in the was assigned for analysis are indicated in Appendix I. BBS data. Direction and significance of population trends for all One major category of interest in bird conservation is species are shown in Appendix I. Figure 1 shows examples of Residency Status: permanent resident, short-distance migrant, or population trend data for a significantly declining species, the neotropical migrant. I compared population trends among these Bushtit, and a significantly increasing species, the Common three groups. I also examined the Nest Site category: open-cup Raven. Note the magnitude of these trends, with Bushtit declining nesting vs. cavity-nesting. There has been considerable concern from mean counts of around eight to far less than one, and about the status of cavity-nesting birds in our region, both due to Common Raven increasing from mean counts of around two to forestry practices that have reduced the availability of snags upon over eight. Exact values for the magnitude of trends of each which many species depend (Bull 1986), and to nest-site species, with graphs, are available at USGS (2004c). competition with the introduced European Starling. How robust are these results? One way to assess that is to I also present a comparison of early-seral and late-seral see if the species’ trends hold true across adjacent and overlapping breeding birds. With the loss of most late-seral forests in the regions. In Table 1, I list the 10 most sharply declining and the 10 Pacific Northwest due to logging, there is concern for the viability most sharply increasing native landbirds (among species with of old-growth-associated species, and it is of interest to examine statistically significant trends) of the Southern Pacific Rainforests this habitat variable with the BBS data. I limited the comparison (SPRF). The species are listed in order of trend strength to birds of early seral and scrub habitats (including chaparral) vs. (strongest trends at the top). The population trends of these those of late seral habitats. Thus, grassland species (e.g., Western species in the SPRF are compared with their trends in five related Meadowlark, Savannah Sparrow) and the broad group of geographic regions. These regions are the states of California and “woodland” species as defined by BBS were not considered in Oregon; the BBS physiographic regions “Sierra Nevada” (the this comparison. “Woodland” species were excluded because the Sierra Nevada Mountains and the Marble Mountains portion of range of habitat types lumped together in this designation was so the Klamath-Siskiyou) and “Cascades” (the Cascade Mountains of broad as to obscure underlying patterns (for example, the BBS Oregon and Washington, including the northeast corner of the “woodland” group includes both the conifer-dependent Steller’s Klamath-Siskiyou); and the “Western U.S.” region of the BBS Jay and the oak-dependent Western Scrub-Jay, and both the (including the states from the Rocky Mountains to the Pacific riparian Warbling Vireo and the upland Hutton’s Vireo). coast). I developed the late-seral list used here based on the Species with declining trends exhibited considerable Northwest Forest Plan’s list of old-growth-associated bird species consistency across these regions. Nine of the top ten declining (Thomas et al. 1993), with modifications indicated by subsequent species in the Southern Pacific Rainforests also showed significant research and my own experience of bird habitat relations in the declines in the most inclusive region, the Western U.S. In nine of Klamath-Siskiyou. Some species on the Northwest Forest Plan’s 48 possible comparisons across regions, these species exhibited list (e.g., Northern Flicker, White-breasted Nuthatch) were omitted increases rather than declines, but in no case were these increasing as not truly representative of late-seral habitats in the Klamath- trends significant. Two birds, Chipping Sparrow and Olive-sided Siskiyou (Marshall et al. 2003). Others not on the Plan’s list (e.g., Flycatcher, exhibited significantly declining trends in all regions. Swainson’s Thrush, Pine Siskin) were added to more accurately The species with significantly increasing trends in the reflect the late-seral avifauna (Marshall et al. 2003). Southern Pacific Rainforests appeared to be less consistent across the other five regions. These species switched to negative trends Results in 14 of 47 comparisons, and in two additional cases exhibited neither a positive nor negative trend. Only half showed Non-native Species significantly increasing trends in the most inclusive region, the Among the five non-native landbirds in the Klamath- Western U.S. Four species switched to show significant negative Siskiyou, three increased significantly from 1968-2002: Wild trends in one or more regions (Red-breasted Nuthatch, Red- Turkey, Rock Pigeon, and House Sparrow. European Starling winged Blackbird, Steller’s Jay, and Black-headed Grosbeak). numbers also increased, but this trend did not reach the level of One species, Osprey, showed significantly increasing trends in all statistical significance. The remaining non-native species, Ring- regions. necked Pheasant, declined significantly during this period. The analyses that follow consider only the 95 species of native Trends in Relation to Residency Status landbirds with BBS data suitable for analysis (Appendix I). Population trends varied markedly in relation to residency status (Table 2). Permanent residents that showed statistically Overall Trends in Native Landbirds significant trends increased and decreased in similar proportions. Between 1968-2002, statistically significant population However, short-distance migrants (i.e., those that migrate mostly

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 49 WILDLIFE ECOLOGY within North America north of Mexico) showed more than twice cases involved a species with a statistically significant increase in as many significant declines as significant increases. This the Southern Pacific Rainforests showing a statistically significant difference was further magnified among the neotropical migrants, decline elsewhere. In general, increasing trends in the Southern in which six times more species showed significant declines than Pacific Rainforest data appeared to be less consistent across increases, and almost 80% showed some degree of population regions than declining trends. This difference may be due to the decline. The only two neotropical migrants with significant fact that the increasing trends in the BBS data, even when increases in the Southern Pacific Rainforests from 1968-2002 statistically significant, were generally not as steep as the were Common Yellowthroat and Black-headed Grosbeak. declining trends, and were thus more prone to reversal by inter- year variability. Trends in Relation to Nest Type The overall population trends of cup and cavity nesting Residency Status birds in the BBS data for the Southern Pacific Rainforests are the Neotropical migrants are the group of species identified in same: 65% of both groups showed declines (Table 3). Open-cup this analysis as suffering the greatest proportion of significant nesting birds showed somewhat more significant declines than did declines. This result is in line with other studies of North cavity-nesting birds. Within the group of cavity-nesting species, I American bird population trends (Robbins et al. 1989; Terborgh further compared those species that excavated their own nest 1989; Martin and Finch 1995). Neotropical migrants face many cavities (“excavators”) with those that depend on natural cavities hazards during their long migrations, including loss of stopover or those excavated by other species (“non-excavators”). Although habitats (Simons et al. 2000) and collisions with obstacles such as the number of species is not large, it is striking that 38% of non- towers (Morris et al. 2003) and buildings (Evans Ogden 1996). excavators declined significantly, whereas not a single “excavator” They are also vulnerable to destruction of their wintering habitats species (woodpeckers and Red-breasted Nuthatch) showed a in the tropics (Terborgh 1989). In the light of the increased significant decline. mortality risks faced by neotropical migrants, habitat management actions that increase productivity on the breeding grounds would Trends in Relation to Habitat be of great benefit to the long-term persistence of these species. The results of the comparison between early seral vs. late seral birds (Table 4) do not support the hypothesis of Nest Site Status disproportionate declines among late-seral species. Significant Cavity-nesting bird species that do not excavate their own declines were documented in 40% of early seral and scrub birds, cavities may also be particularly subject to population declines. In compared to 32% of late seral birds. Equal proportions of early this analysis, 38% of these species declined significantly, as and late seral species showed significant increases. compared to not a single significant decline among cavity-nesters that excavate. Non-excavating cavity nesters often depend on nest Discussion sites in snags, and this result suggests that the availability and especially the persistence of snags may be a critical limiting factor It is important to recognize that each bird species has a for some bird species in our region. Snag management is of unique set of habitat relations, and is subject to a unique suite of particular importance in logged and burned landscapes, where factors influencing productivity, mortality, and thus population snag dispersion and size distribution, as well as overall density, trends. Generalizations across species inevitably obscure these can have great impacts on cavity-nesting birds (Saab et al. 2002). individual differences. Nevertheless, such generalizations may enable us to recognize common factors that would otherwise Habitat Preferences in Relation to Seral Stage remain undetected. The analysis of population trends in relation to seral stage does not support the hypothesis that late-seral-associated birds are Non-Native Species at disproportionate risk of population decline. However, this While a detailed consideration of non-native species result must be interpreted with caution. One problem is that some population dynamics is beyond the scope of this paper, it is worth late-seral birds of great conservation interest in the Klamath- noting that non-native species were the only category of birds Siskiyou occur at low density and are not adequately sampled by identified in this analysis that showed a net increase from 1968- the BBS methodology. Examples include Marbled Murrelet, 2002. Non-native birds have apparently not reached equilibrium Northern Goshawk, Spotted Owl, and White-headed Woodpecker. in the Klamath-Siskiyou, and their effects on native birds may An even more fundamental issue is that BBS routes are not increase in coming years. designed to reflect overall habitat availability. For example, if BBS routes in late-seral forest were established in areas with Comparisons Across Regions relatively low intensities of logging (e.g., national parks), BBS Table 1 reveals considerable consistency in population data might show late-seral bird populations as stable, even if there trends across an array of geographical regions. In only 5% of was a large loss in late-seral habitat regionally. cases (5 of 95 possible comparisons) did a species exhibit In the light of other evidence that the amount of late-seral different significant trends in different regions. All five of these forests has declined drastically in the Klamath-Siskiyou (Noss et

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 50 WILDLIFE ECOLOGY al. 1999), it is important to recognize that late-seral bird comprehensive regional monitoring and research program to populations may have declined more severely in the region than is provide the following: suggested by the BBS data. Information on BBS route ¥ Increased monitoring of habitats and species not adequately distribution in relation to land ownership or management status covered by BBS (e.g., riparian habitats, for which the (e.g., Forest Service, National Park Service, private industrial Oregon/Washington PIF has a special species monitoring timberland) is needed to resolve these questions. plan). Conservation Priorities and Future Research Needs ¥ Better understanding of the habitat relationships of at-risk species to integrate bird conservation and management (e.g., A central objective of the U.S. Fish and Wildlife Service in fuels reduction and fire management plans). Migratory Bird Program Strategic Plan is to “keep common birds ¥ More complete data on demography, including nest common;” that is, to identify and effectively address threats to searching to verify where birds breed successfully (indexes common birds before they cause significant population declines of abundance and density may be misleading) and studies of (USFWS 2004). The actions needed to rescue a species that has survivorship vs. productivity (needed to distinguish declined to endangered status are far more difficult, expensive, breeding season vs. wintering ground effects). and uncertain than are those needed to stabilize a species just beginning to decline. Fortunately, many of these studies are under way, and are It is noteworthy that many species that have shown being coordinated by the Klamath Demographic Monitoring significant declines in the BBS data for the Southern Pacific Network (Ralph 2001). This network, anchored by the Klamath Rainforests are still generally considered to be common birds. Bird Observatory, the Humboldt Bay Bird Observatory, and the Examples include American Kestrel, Rufous Hummingbird, Redwood Sciences Laboratory of the U.S. Forest Service, extends Western Wood-Pewee, Barn Swallow, Bushtit, Chipping Sparrow, along the coast from Mendocino County, California north to Coos Song Sparrow, Dark-eyed Junco, and American Goldfinch. All of County, Oregon, and inland from Crater Lake south through the these birds have adapted well to human-modified environments, Klamath Basin, the Modoc Plateau, and the upper Sacramento and their familiarity may obscure reductions in their numbers. River drainage. It includes data from over 45 constant-effort mist- Some are neotropical migrants whose declines could be linked to netting stations and more than 7000 breeding-season point counts. habitat destruction or other factors on the wintering grounds. Further development of this network, and analyses of the data it is Others, however, are permanent residents that lack obvious risk generating, will be critical to preserving the long-term health of factors for population decline. bird populations in the Klamath-Siskiyou Ecoregion. These results reported here confirm that complacency about the population status of common birds is not appropriate, and that these species deserve more study and monitoring. If such Acknowledgements adaptable species are showing broad declines, human modifications of the environment may be having even more widespread negative This paper is based on the data collected and analyzed by effects than we realize. For example, it has been suggested that the Breeding Bird Survey, USGS Patuxent Wildlife Research Chipping Sparrow and Bushtit declines in Oregon may be linked to Center. These data are an unparalleled resource for investigations fire suppression that has allowed conifer invasion of oak woodland of North American bird population trends. The author thanks John habitats (Hagar and Stern 2001). This same process could be Alexander, Stewart Janes, Nat Seavy, and Jack Williams for their involved in other species’ declines identified in the BBS data, helpful comments on the manuscript. including Oak Titmouse and American Goldfinch. There is an urgent need for research on this and similar issues relating to the Literature Cited effects of habitat modification on bird populations. This study is a first attempt to examine population trends Aldrich, J.W. 1963. Life areas of North America. Journal of among landbirds of the Klamath-Siskiyou. I hope that it will Wildlife Management 27: 530-531. stimulate more detailed analysis of BBS data from this region. Askins, R.A. 1999. History of grassland birds in eastern North Despite the limitations of these data, they offer important insights America. Studies in Avian Biology 19:60-71. into bird population trends. The broad scope of declines Askins, R.A., J.F. Lynch, and R. Greenberg. 1990. Population documented here, involving some of our most familiar and declines in migratory birds in eastern North America. adaptable birds, is reason for great concern. Current Ornithology 7: 1-57. However, BBS data alone are not enough. They must be Bull, E.L. 1986. Ecological value of dead trees to cavity nesting supplemented with additional sources of information on bird birds in northeast Oregon. Oregon Birds 12:91-99. populations in the Klamath-Siskiyou. Much further research is California PIF. 2004. Bird conservation plans are available at: needed on bird population dynamics and habitat relations in order http://www.prbo.org/calpif/plans.html. to allow private landowners and public land managers to make the Carter, M.F., W.C. Hunter, D.N. Pashley, and K.V. Rosenberg. 2000. decisions needed to preserve healthy bird populations in the Setting conservation priorities for landbirds in the United Klamath-Siskiyou. Specifically, there is a great need for a States: the Partners in Flight approach. Auk 117:541-548.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 51 WILDLIFE ECOLOGY

DellaSala, D.A., S.T. Reid, T.J. Frest, J.R. Strittholt, and D.M. Observatory, Ashland, Oregon. Olson. 1999. A global perspective on the biodiversity of the (http://www.klamathbird.org/Newsletter/Newsletter.htm). Klamath-Siskiyou ecoregion. Natural Areas Journal Robbins, C.S., D. Brystrak, and P.H. Gessler. 1986. The Breeding 19(4):300-319. Bird Survey: its first fifteen years, 1965-1979. U.S. Fish and DeSante, D.F. and T.L. George. 1994. Population trends in the Wildlife Service Resource Publication 157, Washington, D.C. landbirds of western North America. Studies in Avian Robbins, C.S., J.R. Sauer, R.S. Greenberg, and S. Droege. 1989. Biology No. 15:173-190. Population declines in North American birds that migrate to Evans Odgen, L.J. 1996. Collision course: the hazards of lighted the Neotropics. Proceedings of the National Academy of structures and windows to migrating birds. World Wildlife Science 86:7658-7662. Fund Canada and Fatal Light Awareness Program, Toronto, Saab, V., R. Brannon, J. Dudley, L. Donohoo, D. Vanderzanden, V. Canada. Johnson, and H. Lachowski. 2002. Selection of fire-created Hagan III, J.M. and D.W. Johnston (eds.). 1992. Ecology and snags at two spatial scales by cavity-nesting birds. USDA conservation of neotropical migrant landbirds. Smithsonian Forest Service General Technical Report PSW-GTR-181. Institution Press, Washington, D.C. Sauer, J. R., J. E. Hines, and J. Fallon. 2003a. The North Hagar, J.C., and M.A. Stern. 2001. Avifauna in oak woodlands of American Breeding Bird Survey, results and analysis 1966 - the Willamette Valley, Oregon. Northwestern Naturalist 2002. Version 2003.1, USGS Patuxent Wildlife Research 82:12-25. Center, Laurel, Maryland. (http://www.mbr-pwrc.usgs.gov/ Holmes, R.T., and T.W. Sherry. 2001. Thirty-year bird population bbs/bbs.html). trends in an unfragmented temperate deciduous forest: Sauer, J.R., J.E. Fallon, and R. Johnson. 2003b. Use of North importance of habitat change. Auk 118:589-609. American Breeding Bird Survey data to estimate population James, F.C., C.E. McCulloch, and D.A. Wiedenfeld. 1996. New change for Bird Conservation Regions. Journal of Wildlife approaches to the analysis of population trends in land Management 67:372-389. birds. Ecology 77:13-27. Simons, T.R., S.M. Pearson, and F.R. Moore. 2000. Application Link, W.A., and J.R. Sauer. 1994. Estimating equations estimates of spatial models to the stopover ecology of trans-Gulf of trend. Bird Populations 2: 23-32. migrants. Studies in Avian Biology 20:4-14. Link, W.A., and J.R. Sauer. 1997. New approaches to the Siskiyou Project. 2004. Maps of the Klamath-Siskiyou Ecoregion analysis of population trends in land birds: comment. can be viewed and downloaded at: http://www.siskiyou. Ecology 78:2632-2634. org/resources/. Marshall, D., A. Contreras, and M. Hunter (eds.). 2003. Birds of Strittholt, J.R., and D.A. DellaSala. 2001. Importance of roadless Oregon: a general reference. Oregon State University Press, areas in biodiversity conservation in forested ecosystems: Corvallis. case study of the Klamath-Siskiyou Ecoregion of the United Marshall, J.T. 1988. Birds lost from a giant sequoia forest during States. Conservation Biology 15(6):1742-1754. fifty years. Condor 90:359-372. Terborgh, J. 1989. Where have all the birds gone? Princeton Martin, T.E., and D.M. Finch. 1995. Ecology and management of University Press, Princeton, New Jersey. neotropical migratory birds: a synthesis and review of Thomas, J.W., M.G. Raphael, R.G. Anthony, E.D. Forsman, A.G. critical issues. Oxford University Press, New York. Gunderson, R.S. Holthausen, B.G. Marcot, G.H. Reeves, Morris, S.R., A.R. Clark, L.H. Bhatti, and J.L. Glasgow. 2003. J.R. Sedell, and D.M. Solis. 1993. Viability assessments Television tower mortality of migrant birds in western New and management considerations for species associated with York and Youngstown, Ohio. Northeastern Naturalist late-successional and old-growth forests of the Pacific 10:67-76. Northwest. USDA Forest Service, Portland, Oregon. Noss, R.F., J.R. Strittholt, K. Vance-Borland, C. Carroll, and P.A. Trail, P.W., R. Cooper, and D. Vroman. 1997. The breeding birds Frost. 1999. A conservation plan for the Klamath-Siskiyou of the Klamath/Siskiyou region. Pages 158-174 in, J.J. Ecoregion. Natural Areas Journal 19(4):392-411. Beigel, E.S. Jules, and B. Snitkin, eds. Proceedings of the Oregon/Washington PIF. 2004. Bird conservation plans are First Conference on Siskiyou Ecology. Siskiyou Regional available at: http://community.gorge.net/natres/pif.html. Education Project, Cave Junction, Oregon. Partners in Flight. 2004. PIF Watch List is available at: USFWS. 2004. The Migratory Bird Program Strategic Plan can (http://www.abcbirds.org/pif/pif_watch_list.htm). be viewed at: http://migratorybirds.fws.gov/mbstratplan/ Peterjohn, B.G., and J.R. Sauer. 1999. Population status of North ADMessage.htm. American grassland birds from the North American Breeding USGS. 2004a. Details on the trend estimation statistics used by USGS Bird Survey 1966-1996. Studies in Avian Biology 19:27-44. can be found at: http://www.mbr-pwrc.usgs.gov/bbs/ trendin.html. Peterjohn, B.G., J.R. Sauer, and W.A. Link. 1997. The 1994 and USGS. 2004b. Maps of BBS physiographic strata can be viewed 1995 summary of the North American Breeding Bird at: http://www.mp2-pwrc.usgs.gov/bbs/StrataNames/. Survey. Bird Populations 3:48-66. USGS. 2004c. BBS Bird population data summaries for the Ralph, C.J. 2001. What is the Klamath Demographic Monitoring Southern Pacific Rainforests, 1968-2002, can be found at: Network? The Klamath Bird: Winter 2001. Klamath Bird http://www.mbr-pwrc.usgs.gov/cgi-bin/atlasr02.pl?S93.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 52 WILDLIFE ECOLOGY

Figure 1. Examples of population trend data for a sharply declining species, the Bushtit (upper graph) and a sharply increasing species, the Common Raven (lower graph). The Y-axis shows the average number of birds per count for the South Pacific Rainforest region for a given year. Both trends are highly significant (P < 0.01). Data from USGS.

Bushtit Count

Year

Common Raven Count

Year

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 53 WILDLIFE ECOLOGY

Table 1. Population trends for sharply declining and increasing native landbirds in the Southern Pacific Rainforests (SPRF) compared with other BBS regions. The upper box lists the 10 native landbirds with the steepest population declines in the SPRF from 1968-2002. The lower box lists the 10 native landbirds with the steepest population increases in the SPRF over this period. See text for description of regions. Key: - - (significantly declining); - (non-significant declining trend); + + (significantly increasing); + (non-significant increasing trend); na (sufficient data not available for analysis); 0 (no trend in data). Statistical significance level for population trends defined as P < 0.10, following USGS 2004a.

So. Pacific Sierra Declining Species CA OR Cascades Western US Rainforest Nevada Bushtit ------Common Nighthawk - - + + - + - - Chipping Sparrow ------Blue Grouse - - + + - - - - Pine Siskin ------Western Meadowlark - - - - + - - - - - American Kestrel - - - - + - - + - - Lark Sparrow - - - na - - na - Golden-cr. Kinglet ------+ - - Olive-sided Flycatcher ------

So. Pacific Sierra Increasing Species CA OR Cascades Western US Rainforest Nevada Northern Pygmy-Owl + + + + + + - + Osprey + + + + + + + + + + + + Common Yellowthroat + + + + na + + + + + Black Phoebe + + + + + + na + + Common Raven + + + + + 0 + + + + Red-breasted Nuthatch + + ------+ + Red-winged Blackbird + + + + - - - - Steller’s Jay + + + - - - - + Black-headed Grosbeak + + 0 - - + + + + California Towhee + + - + + na -

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 54 WILDLIFE ECOLOGY

Table 2. Population trends in relation to residency status among native landbirds of the BBS Southern Pacific Rainforests region. N = the number of species in each category. Species list with residency status in Appendix I. Data from USGS 2004c.

All Species Species With Signif. Trends Trends Decreasing Increasing Decreasing Increasing Permanent Residents (N = 20) 11 (55%) 9 (45%) 5 (25%) 4 (20%) Short-Distance Migrants (N = 32) 17 (53%) 15 (47%) 11 (34%) 5 (16%) Neotropical Migrants (N = 33) 26 (79%) 7 (21%) 13 (39%) 2 (6%)

Table 3. Population trends in relation to nest type among native landbirds of the BBS Southern Pacific Rainforests region. N = the number of species in each category. Species list with nest type status in Appendix I. Data from USGS 2004c.

All Species Species With Signif. Trends Decreasing Increasing Decreasing Increasing

Open-cup Nesting (N = 52) 34 (65%) 18 (35%) 18 (35%) 7 (13%) Cavity-Nesting (N = 20) 13 (65%) 7 (35%) 5 (25%) 2 (10%)

Excavator (N = 7) 5 (71%) 2 (29%) 0 (0%) 1 (14%) Non-Excavator (N = 13) 8 (62%) 5 (38%) 5 (38%) 2 (13%)

Table 4. Population trends in relation to early vs. late seral habitat preferences among native landbirds of the BBS Southern Pacific Rainforests region. N = the number of species in each category. Species list with seral habitat preferences in Appendix I. Data from USGS 2004c.

All Species Species With Signif. Trends Decreasing Increasing Decreasing Increasing

Early Seral/Scrub (N = 25) 19 (76%) 6 (24%) 10 (40%) 4 (16%) Late Seral (N = 19) 11 (58%) 8 (42%) 6 (32%) 3 (16%)

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 55

WILDLIFE ECOLOGY

APPENDIX I. Species included in this analysis of landbird population trends for the Klamath-Siskiyou Ecoregion (total = 100 species). All data are taken from summaries for the Southern Pacific Rainforest physiographic province in the BBS data (USGS 2004c). Following each species is a notation indicating population trend and significance. + + means a significant increasing trend; + means non-significant increasing trend; - - means significant declining trend; and Ð means non-significant declining trend. Statistical significance level defined as P < 0.10, following USGS (2004a). The five non-native bird species included in the analysis are noted with an asterisk before the species name.

Species were included in different analytical categories, as follows (see text for definitions): Residency: PR = permanent resident; SD = short-distance resident; NM = neotropical migrant Habitat: ES = early seral/scrub; LS = late seral Nesting: OC = open cup nesting; CV = cavity-nesting; (e) = excavator; (ne) = non-excavator

Trend 1968- Residency Habitat Nesting 2002 BIRDS OF PREY Turkey Vulture (Cathartes aura) + SD Osprey (Pandion haliaetus) + + SD Red-tailed Hawk (Buteo jamaicensis) + American Kestrel (Falco sparverius) - - SD CV (ne) Northern Pygmy-Owl (Glaucidium gnoma) + + LS CV (ne) GAMEBIRDS Blue Grouse (Dendropagus obscurus) - - PR *Ring-necked Pheasant (Phasianus colchicus) - - *Wild Turkey (Meleagris gallopavo) + + California Quail (Callipepla californica) - PR ES Mountain Quail (Oreortyx pictus) + PR ES PIGEONS & DOVES *Rock Pigeon (Columba livia) + + Band-tailed Pigeon (Columba fasciata) - NM Mourning Dove (Zenaida macroura) - - SD NIGHTJARS Common Nighthawk (Chordeiles minor) - - NM ES SWIFT & HUMMINGBIRDS Vaux’s Swift (Chaetura vauxi) - - NM LS CV (ne) Anna’s Hummingbird (Calypte anna) + PR Allen’s Hummingbird (Selasphorus sasin) - NM Rufous Hummingbird (Selasphorus rufus) - - NM KINGFISHER Belted Kingfisher (Ceryle halcyon) - - WOODPECKERS Acorn Woodpecker (Melanerpes formicivorus) - PR CV (e) Red-breasted Sapsucker (Sphyrapicus ruber) + LS CV (e) Downy Woodpecker (Picoides pubescens) - PR CV (e) Hairy Woodpecker (Picoides villosus) - PR LS CV (e) Northern Flicker (Colaptes auratus) - CV (e) Pileated Woodpecker (Dendrocopus pileatus) - PR LS CV (e)

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 56 WILDLIFE ECOLOGY

Trend Residency Habitat Nesting 1968-2002 FLYCATCHERS Olive-sided Flycatcher (Contopus borealis) - - NM OC Western Wood-Pewee (Contopus sordidulus) - - NM OC Black Phoebe (Sayornis nigricans) + + PR OC Willow Flycatcher (Empidonax traillii) - - ES OC Hammond’s Flycatcher (Empidonax hammondii) + NM LS OC Dusky Flycatcher (Empidonax oberholseri) - NM OC Pacific-slope Flycatcher (Empidonax difficilis) - - LS OC Ash-throated Flycatcher (Myiarchus cinerascens) - NM CV (ne) Western Kingbird (Tyrannus verticalis) - NM OC VIREOS Cassin’s Vireo (Vireo cassinii) - OC Warbling Vireo (Vireo gilvus) + NM OC Hutton’s Vireo (Vireo huttoni) - PR OC JAYS AND CROWS Steller’s Jay (Cyanocitta stelleri) + + PR OC Western Scrub-Jay (Aphelocoma californica) + PR OC American Crow (Corvus brachyrhynchos) + SD Common Raven (Corvus corax) + + PR SWALLOWS Tree Swallow (Tachycineta bicolor) - SD CV (ne) Violet-green Swallow (Tachycineta thalassina ) + NM CV (ne) No. Rough-winged Swallow (Stelgidopteryx serripennis) - NM Cliff Swallow (Petrochelidon pyrrhonota ) - - NM Barn Swallow (Hirundo rustica) - - NM OC CHICKADEES & ALLIES Black-capped Chickadee (Poecile atricapilla) + PR CV (ne) Chestnut-backed Chickadee (Poecile rufescens) - - PR LS CV (ne) Oak Titmouse (Baeolophus inornatus ) - - PR CV (ne) Bushtit (Psaltiparus minimus) - - PR ES NUTHATCHES & CREEPER Red-breasted Nuthatch (Sitta canadensis) + + SD LS CV (e) White-breasted Nuthatch (Sitta carolinensis) + PR CV (ne) Brown Creeper (Certhia americana) - SD LS WRENS & WRENTIT Marsh Wren (Cistothorus palustris) + SD Bewick’s Wren (Thryomanes bewickii) + SD ES CV (ne) House Wren (Troglodytes aedon) - NM ES CV (ne) Winter Wren (Troglodytes troglodytes) + + SD LS Wrentit (Chamaea fasciata) - PR ES OC STARLING & WAXWING *European Starling (Sturnus vulgaris) + Cedar Waxwing (Bombycilla cedrorum) + SD OC

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 57 WILDLIFE ECOLOGY

Trend 1968- Residency Habitat Nesting 2002 THRUSHES & KINGLET Western Bluebird (Sialia mexicanus) - - SD CV (ne) Swainson’s Thrush (Catharus ustulatus) - - NM LS OC Hermit Thrush (Catharus guttata) - SD LS OC American Robin (Turdus migratorius) + SD OC Varied Thrush (Ixoreus naevius) + SD LS OC Golden-crowned Kinglet (Regulus satrapa) - - SD LS OC WARBLERS & TANAGER Orange-crowned Warbler (Vermivora celata) - - NM ES OC Nashville Warbler (Vermivora ruficapilla) - NM ES OC Yellow Warbler (Dendroica petechia) - NM ES OC Yellow-rumped Warbler (Dendroica coronata) - LS OC Black-thr. Gray Warbler (Dendroica nigrescens) - NM OC Hermit Warbler (Dendroica occidentalis) + NM LS OC MacGillivray’s Warbler (Oporornis tolmiei) - - NM ES OC Common Yellowthroat (Geothlypis trichas) + + NM ES OC Wilson’s Warbler (Wilsonia pusilla) - NM OC Yellow-breasted Chat (Icteria virens) - NM ES OC Western Tanager (Piranga ludoviciana) + NM OC SPARROWS & ALLIES Spotted Towhee (Pipilo maculatus) + + SD ES OC California Towhee (Pipilo californica) + + PR ES OC Chipping Sparrow (Spizella passerina) - - NM ES OC Lark Sparrow (Chondestes grammacus) - - NM ES OC Savannah Sparrow (Passerculus sandvicensis) + SD OC Song Sparrow (Melospiza melodia) - - SD ES OC White-crowned Sparrow (Zonotrichia leucophrys) - - SD ES OC Dark-eyed Junco (Junco hyemalis) - - OC GROSBEAK & BUNTING Black-headed Grosbeak (Pheucticus ludovicianus) + + NM OC Lazuli Bunting (Passerina amoena) - NM ES OC BLACKBIRDS & ALLIES Red-winged Blackbird (Agelaius phoeniceus) + + SD ES OC Western Meadowlark (Sturnella neglecta) - - SD OC Brewer’s Blackbird (Euphagus cyanocephalus) - SD ES OC Brown-headed Cowbird (Molothrus ater) - - SD Bullock’s Oriole (Icterus bullockii) - - NM FINCHES & HOUSE SPARROW Purple Finch (Carpodacus purpureus) - - SD OC House Finch (Carpodacus mexicanus) - SD OC Red Crossbill (Loxia curvirostra) + SD LS OC

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 58 WILDLIFE ECOLOGY

Trend Residency Habitat Nesting 1968-2002 Pine Siskin (Carduelis pinus) - - SD LS OC Lesser Goldfinch (Carduelis psaltria) - SD ES OC American Goldfinch (Carduelis tristis) - - SD ES OC Evening Grosbeak (Coccothraustes vespertinus) + SD OC *House Sparrow (Passer domesticus) + +

Chi-square goodness-of-fit tests were carried out to compare the results for both all species and for those with significance population changes only. The null hypothesis was that each residency category would show population increases in 50% of species and decreases in 50%. For all species, the chi-square was 11.26 (p < 0.005), and for species with significant changes only, the chi-square was 10.43 (p < 0.01). Thus, these results cannot be dismissed as random variations. It is clear that the neotropical migrant category contributes more than expected to the observed declines and less than expected to the observed increases.

Illustration by Bob Cremins

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 59 WILDLIFE ECOLOGY

CONSERVATION STATUS OF AMERICAN MARTENS AND FISHERS IN THE KLAMATH-SISKIYOU BIOREGION

Keith M. Slauson and William J. Zielinski Redwood Sciences Lab, Pacific Southwest Research Station, USFS, 1700 Bayview Drive Arcata, CA 95521 Emails: [email protected]; [email protected]

Abstract

The American marten (Martes americana) was historically distributed within coastal and high elevation fir forests of the Klamath- Siskiyou Bioregion (KSB) and was represented by three recognized subspecies (M. a. caurina, M. a. sierrae, and M. a. humboldtensis). The fisher (Martes pennanti pacifica) was historically distributed throughout interior and near-coast forests of the KSB. Over the last 8 years we have conducted systematic surveys, using baited track plate stations, at 497 locations within the KSB, resulting in >35,000 days of survey effort. Survey results demonstrate that martens are absent from portions of their historical range, with the most severe loss within the range of the Humboldt marten. Martens are absent from areas of the historical range of M. a. caurina on the Siskiyou National Forest. The status of M. a. sierrae within the Salmon and Marble mountains is uncertain. The fisher remains well distributed within most of its historical distribution within the KSB, but its status north of Highway 199 is uncertain. Conservation of marten populations will require protection of areas currently occupied and evaluation of whether strategic restoration of additional habitat is warranted. The population of fishers within the KSB represents the largest native fisher population within the western U.S. and has substantial potential to contribute to fisher restoration to adjacent bioregions. Maintenance and restoration of functional landscape connectivity from the KSB to adjacent bioregions to the North and East currently lacking native fisher populations will be critical for fishers to recolonize them. Introduction appearance. Fishers also have a pointed snout and large, but well- rounded ears. Fishers have white or cream patches on the chest he American marten is a house-cat sized carnivorous and the inguinal region. The diet of the fisher is dominated by mammal and a member of the weasel family (Mustelidae). small and medium sized mammals and birds, but also includes TTMartens measure from 500 to 680 millimeters from nose some reptiles, carrion, fruits, and insects when available (Martin to tail tip in length and weigh between 500 and 1400 grams 1994). The fisher is designated as a species of special concern by (Buskirk and Ruggiero 1994, Powell et al. 2003). They are long California Department of Fish and Game, a critical species by the and thin, with tan to chocolate brown fur, and have a fox-like face Oregon Department of Fish and Wildlife, and is designated a due to a pointed snout and large pointed ears. Martens also have sensitive species by the U.S. Forest Service in both Regions 5 and an irregular yellowish-orange throat patch that may extend to the 6. The fisher has been petitioned for listing under the Federal chest. The diet of the marten varies by season, year, and Endangered Species Act three times in the western U.S., with a geographic location (Martin 1994). In general the diet of the decision on whether to list the fisher in the Pacific states currently marten includes mammals, birds, reptiles, insects, carrion, and pending. fruits, and becomes most limited during the winter when it is The marten and fisher have always been highly valued for restricted largely to small mammals (Martin 1994). The marten is their fur and both species are easily trapped (Powell 1983; designated as a species of special concern by the California Strickland and Douglas 1987). The fisher brought one of the Department of Fish and Game, a vulnerable species by the Oregon highest pelt prices of any terrestrial furbearer in North America Department of Fish and Wildlife, a sensitive species by the U.S. during the early 1900s (Lewis and Zielinski 1996). Records of Forest Service in Region 5 (includes California), and has no trapping for their fur began to be kept in the late 1700s in North special designation by the U.S. Forest Service in Region 6 America (Novak et al. 1987). No laws establishing seasons or bag (includes Oregon). limits on the trapping of martens and fishers existed before the The fisher is also a carnivorous mammal and a member 1920s, consequently trapping was virtually unregulated until that of the Mustelidae. It is approximately twice the length of time (Strickland 1994). As a result, populations of both species martens, measuring 900-1200 millimeters from nose to tail tip were severely reduced on a continental scale by the early 1900s (Powell and Zielinski 1994). Fishers weigh from 2,000 to 5,500 (Strickland 1994). In the Klamath-Siskiyou region trapping grams (Powell et al. 2003). Fishers are also long and thin, but records for martens and fishers begin to appear in the late 1800s have dark brown fur that appears black in the field; grizzled fur in California (Merriam 1890; Grinnell et al. 1937) and the early around the head and neck often gives them a gray-headed 1900s in Oregon (Anonymous 1914). Trapping records for both

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 60 WILDLIFE ECOLOGY species within the Klamath-Siskiyou region and adjacent areas of other regions in the Pacific states. Finally, we identify the Pacific states show relatively high harvests during the late conservation opportunities that will help ensure marten and fisher 1800s and early 1900s followed by strong declines (Grinnell et al. persistence within the KSB and adjacent bioregions. 1937; Seymour 1980; Verts and Carraway 1998; Zielinski et al. 2001; Aubry and Lewis 2003). Precipitated by concerns from Methods authorities that populations of martens and fishers were threatened by trapping (Jewett 1915; Dixon 1925; Grinnell et al. 1937), the Historical Distribution legal trapping of fishers concluded in 1937 in Oregon and 1946 in California. Martens were protected in northwestern California in To determine the historical (prior to 1950) distribution of 1946 and statewide by 1956. In Oregon the legal season on martens and fishers within the KSB we reviewed all the published martens was closed from 1937-1937, 1941-1944, and 1947-1949, and unpublished information on their distributions or occurrences following the sharpest declines in the numbers of animals trapped within California and Oregon. We included only records that we during previous seasons (Marshall 1994; ODFW unpubl. records). considered verifiable, i.e., museum or trapping records with Martens can still be legally trapped in western Oregon. specific location information. The marten and fisher both share several life history traits that make them sensitive to the alteration of the forest habitats in Contemporary Surveys which they live. Martens and fishers both avoid open areas that We used several methods of systematic sampling to are devoid of both overhead and escape cover (Powell 1983; Drew determine the contemporary distribution of martens and fishers. 1995). Removal of these two forms of cover by logging results in Most data were derived from regional-scale surveys based on these areas being avoided until suitable cover regenerates (Buskirk either 5-km or 10-km survey grids. At each point on these grids and Ruggiero 1994; Jones and Garton 1994; Raphael 1984). we established a ‘sample unit’ comprised of six track plate Between foraging bouts, martens and fishers select areas to rest. stations in a pentagonal array with one station in the center and These sites typically occur in cavities or on platforms provided by five around the perimeter with 1-km between adjacent stations. large diameter live trees, snags, and downed logs. The loss of We have also included Klug’s (1997) survey effort on private these elements can reduce the suitability of forested areas as timberlands in coastal northwestern California. Klug (1997) used habitat for both species. Both species have large area a grid with approximately 5 km spacing, composed of six track requirements for mammals of their body sizes, with martens plate stations spaced at 1-km intervals along roads at each grid occupying home ranges from 1-15 km2 (Buskirk and McDonald point. In three locations we used a 2-km survey grid because this 1989) and fishers from 16 - 85 km2 (Powell 1993). Both species helped us address a special need for fine-scale information in are sensitive to the loss and fragmentation of mature and late- northwestern California (Zielinski unpubl. data; Slauson 2003). successional forest at the landscape scale (Rosenburg and Raphael In these grids there were either one or two track plate stations at 1986). The marten is particularly sensitive to the loss of late- each grid point. successional forest and will avoid landscapes which have lost Each track plate station consisted of a covered enclosure more than 30-35% of mature forest (Bissonette et al. 1997; Potvin open on one end. Animals enter the enclosure crossing an et al. 2000). aluminum plate coated with carbon-soot and and sticky contact There is considerable reason for concern about the status of paper before they could reach a piece of chicken (Zielinski 1995). American marten and fisher populations in the Pacific states. Fur Animals that crossed the soot and contact paper on the way to the harvests caused local and regional extirpations and declines and bait left high-resolution positive impressions of their feet. Once decades of protection from trapping have not resulted in the established, each track plate station was run for 16 consecutive recovery of fisher populations in the Pacific states (Zielinski et al. days and checked every other day to collect tracks, replace bait 1995; Lewis and Stinson 1998; Aubry and Lewis 2003) or martens and sooted plates as necessary. A commercial trapping lure was in coastal northwestern California (Zielinski et al. 2001). applied to a nearby tree when each station was established and re- Additionally, both species show direct links to old forest attributes applied after eight days if a marten or fisher had not yet been at multiple spatial scales and these attributes have and continue to detected. Klug (1997) ran each track plate station for 22 be lost and altered due to the logging of mature and old growth consecutive days; visits to check each station occurred every forests. Multiple authorities have (Jewitt 1915; Dixon 1925; other day. Lure was not used at these track plate stations. Grinnell et al. 1937) and continue (Zielinski et al. 2001; Aubry and Lewis 2003) to voice concern about the status of marten and fisher populations in the Pacific states. Results We take a bioregional-scale approach to assessing the conservation status of the marten and fisher within the globally Historical Distribution: Marten outstanding Klamath-Siskiyou Bioregion. By comparing the The historical distribution of the American marten in the historical and contemporary distributions of the marten and fisher, Pacific states was recently summarized by Zielinski et al. (2001). we will demonstrate that the changes in distributions of martens Zielinski et al. (2001) identified 29 verifiable historical records of and fishers are paradoxical when compared to population trends in marten occurrence within the KSB. In the KSB martens were

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 61 WILDLIFE ECOLOGY historically distributed in the coastal forests of the western portion three separate populations. Martens were not detected in much of of the bioregion and in the higher true fir (Abies sp.) forests in the their historical coastal distribution in California and were not Marble-Salmon-Trinity Mountains in the southeastern portion of detected in many areas surveyed on the Siskiyou National Forest the bioregion (Figure 1; Grinnell et al. 1937; Bailey 1936; in coastal Oregon. The two coastal populations are separated by Zielinski et al. 2001). This distribution was shared by three 52 km. No surveys occurred in the true fir forests of the Marble- recognized subspecies of martens, M. a. humboldtensis (Grinnell Salmon-Trinity mountains within the historical range of M. a. and Dixon 1926), M. a. caurina (Merriam 1890), and M. a. sierrae. sierrae (Grinnell and Dixon 1926; Figure 1). The boundary between M. a. humboldtensis and M. a. caurina in the Oregon Fishers Coast Range occurs at or near the border between California and Fishers were detected at 101 sample units (20.3%) Oregon and is not readily supported by the presence of any distributed across much of the area surveyed in the KSB (Figure biogeographical boundary or by preliminary genetic results (K. 5). Detections were uncommon in interior Del Norte County in Stone, unpublished data). Records of marten occurrence are California and rare in Curry, and Josephine counties in Oregon. absent from most of the interior forests of the KSB. The narrow Fishers were detected on both sides of Highway 5 on the Shasta- gap between the interior boundary of the range of M. a. Trinity National Forest in the southeastern portion of the humboldtensis and the eastern edge of M. a. sierrae is occupied bioregion. These results are encouraging; however they do not by the Klamath River Canyon. While the river itself may be a confirm whether fishers have been able to move across Interstate barrier to movement for martens, the canyon is occupied by more Highway 5 or if it represents a barrier to movement. Fishers were xeric forest types than the near-coast forest types or high-elevation detected frequently in the near coast areas within redwood forest true fir forests historically occupied by martens to the West and types, areas that lack historical information to support their past East, respectively. Molecular investigation is currently underway occurrence in these areas. to determine whether there is genetic support for this subspecific boundary. Discussion

Historical Distribution: Fisher The results of the contemporary surveys demonstrate that We found 36 verifiable historical records of fisher American martens have declined within the coastal forests of the occurrence within the KSB representing a single subspecies, M. p. KSB, with the most severe decline in coastal northwestern pacifica (Bailey 1936; Grinnell et al. 1937; Sherrell 1970; Verts California. The coastal California population in the KSB and Carraway 1998). The fisher had a more contiguous historical represents the only known population in coastal California, distribution than the marten within the KSB, occupying almost all currently occupying an area equivalent to less than 5% of its of the bioregion, except the most near-coastal areas (Figure 2). historical range (Zielinski et al. 2001; Slauson 2003). The The lack of records of occurrence for the fisher in Oregon is a bit distribution of detections is more spatially extensive in Oregon misleading, as many trapping records are available, however only than in California, but occurrences are patchier. at the resolution of each county they were trapped in (ODFW Additional surveys will be necessary to determine the fine- unpubl. data). Importantly, the KSB is at the crossroads of several scale distribution pattern of this population. The coastal Oregon peninsular portions of the distribution of the fisher in California population of martens in the KSB is one of only two known to and Oregon. It links to the Coast and southern Cascade-Sierra occur in coastal Oregon, the second is located >125 km north on Nevada ranges of California and to the Coast and Cascade ranges the Siuslaw National Forest (Zielinski et al. 2001). The two of Oregon. coastal populations of martens in the KSB are separated by >50 km. These are considerable distances for martens to travel. Contemporary Surveys Long-term viability of these populations may require the From 1994 to 2002, a total of 497 sample units (2,130 exchange of individuals and genetic material to avoid the loss of stations) were surveyed within or adjacent to the KSB, genetic variation (Nei et al. 1975). The range of dispersal representing 35,520 days of survey effort (Figure 3). Of these, distances for 26 juvenile martens in Maine (Phillips 1994) was 457 were surveyed during the summer and fall periods from 1996- 4.9 to 35.1 km for males (n = 13; median = 14.3) and 5.5 to 27.0 2002 (Zielinski et al. 2000, Slauson and Zielinski 2003, Slauson km for females (n = 13; median = 12.0) and in northeastern 2003) and 40 during the summer and winter of 1994 and 1995 Oregon (Bull and Heater 2001) was 28 to 43.3 km for males (n = (Klug 1997). Most surveys were within the western half of the 2) and 33.3 km for females (n=1). The distance between the two KSB. Very few (< 20) of the sample units surveyed were within populations in coastal Oregon is 3 times the maximum dispersal any of the seven Wilderness areas within the KSB. distance and 4 to 10 times the median dispersal distance for juvenile martens from either study. The distance between the two Martens coastal marten populations in the KSB is 1.2 times larger than the American martens were detected at 39 of 497 (7.8%) maximum dispersal distances and is 1.5 to 4 times the median sample units (Figure 4). Detections of martens were clumped in dispersal distances for juvenile martens from these studies. We

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 62 WILDLIFE ECOLOGY have serious concerns about the viability of these coastal there are potential linkage zones to the southern Cascades in this populations of martens. They are small, patchily distributed area north of Grants Pass. Additional surveys will be necessary to populations, due to both natural distribution of suitable habitat and determine whether fishers still occupy this area. The lack of to the effects of logging, and they are separated by substantial fisher detections in areas of Curry County included in our surveys distances (Slauson 2003). is consistent with the predictions of the spatial habitat model for The situation for the population of martens on the fishers developed by Carroll et al. (1999). Siskiyou National Forest may have been exacerbated by the recent The temporal changes in distributions of martens and fishers Biscuit Fire in 2002. Wildlife is a natural and essential in the KSB are contrary to that elsewhere in the Pacific states. In component of these coastal forests and plays a role in developing most interior mountain ranges where martens occur (e.g. important habitat elements for martens (e.g., snags). However, Cascades, Sierra Nevadas) they remain fairly well distributed given that late-seral conifer habitat is already reduced in coastal (Giblisco 1994; Sheets 1993; Zielinski et al. 2000), whereas Oregon, the fire may have caused a short-term loss and fishers have severely declined or been extirpated in these areas fragmentation of suitable habitat. In the coastal forests of (Aubry and Houston 1992, Marshall 1994, Zielinski et al. 1995; California and Oregon martens use stands with dense, spatially Aubry and Lewis 2003). This is likely due in large part to the extensive shrub cover (Slauson 2003; Slauson and Zielinski 2003). unique distribution dynamics of the marten and fisher in the KSB, The shrub layers are dominated by Ericaceous species (e.g., which is inconsistent with the pattern of their occurrences in the Rhododendron macrophylum, Gaultheria shallon) which have interior mountains. In these mountains fishers typically occur at waxy leaves and are highly flammable (Agee 1993). Significant lower elevations and within a narrow elevational band (Aubry and loss of the shrub layer may reduce habitat suitability, due to Houston 1992; Zielinski et al. 1995; Zielinski et al. 1997). These reduction in prey abundance or improved access to these areas by areas were historically more accessible to trappers and logging, competitors that may otherwise be limited by dense shrubs more vulnerable to wildfire, and their narrow linear arrangement (Slauson 2003; Slauson and Zielinski 2003). was fragmented more easily (Aubry and Lewis 2003). Martens, Few surveys occurred near the historic distribution of M. conversely, occur at higher elevations in the interior mountain a. sierrae in the Marble-Salmon-Trinity Mountains and ranges which are consequently less accessible, less altered by consequently we do not know the status of marten populations in logging, and composed of a larger proportion of well-connected this area. We have received recent (2001-2003) verified evidence wilderness reserves. In these areas marten populations appear to of the presence of martens within the Salmon (skull, J. Betaso remain fairly well distributed. Within the KSB, martens have pers. com.) and Trinity mountains (photographs, E. Wier pers. fared poorly in the narrow coastal forests which were more com.); however this information does not allow us to determine accessible to trappers, more accessible to logging, more rapidly the current status of these populations. It is likely that these areas affected by fragmentation of habitat and populations, and are retain marten populations similar to their historical distributions composed of proportionately little late-successional reserves or due to their unroaded and rugged nature which would have made wilderness areas. The distribution of the fisher in the KSB is access difficult for trappers. These areas are also designated extensive, occupies historically remote and rugged terrain, and is wilderness where habitat alteration by logging has been limited. more resilient to the effects of fragmentation than the more The fisher has fared far better than the marten in the KSB. linearly-arranged habitat areas in the Sierra Nevada or southern The distribution of the fisher has remained similar to its historical Cascades. extent in the areas surveyed. The KSB fisher population is the The KSB plays an important role in the conservation of largest fisher population in the Pacific states (Aubry and Lewis martens and fishers in the Pacific states. It contains 2 of the 3 2003) and likely the largest within the western U.S. It appears known populations of martens in the coastal forests of California that the fisher may have increased its distribution into the coastal and Oregon and it contains the largest population of fishers in the redwood forests in northwestern California in recent times. Pacific states. To ensure that martens persist in the KSB, efforts Historical records in the redwood forests are rare and Grinnell et should be made to maintain all habitat currently occupied, initiate al. (1937) depicted the fisher’s distribution as distinctly interior strategic restoration activities to increase and reconnect suitable from the coast. Early trapping efforts were fairly extensive within habitat patches in the vicinity of these populations, and to restore the redwood region and many martens were trapped in redwood functional connectivity where it has been lost between these two forests, but few fishers (Grinnell et al. 1937; Twinning unpubl. populations. Site- specific restoration activities should target data). This suggests that detections along the redwood coast restoration of the structural characteristics important to martens in represent a recent expansion rather than an oversight in historical coastal forests, such as dense shrub cover and late-seral conditions accounts. While few surveys were conducted north of Highway which include large diameter live trees, snags, and downed logs 199 in Josephine County, Oregon, we are not aware of any within stands (Slauson 2003). The KSB fisher population has the verifiable contemporary records of fishers in the interior potential to recolonize adjacent areas where it has been extirpated, northeastern portion of the KSB. The presence of Highway 199 and if warranted, to act as a source population for translocations combined with the community development along much of its of individuals into other portions of its historical range in the route may act to discourage dispersal in the interior KSB. Fishers Pacific states. To give this population the best opportunity to historically occurred in the northeastern portion of the KSB and recolonize adjacent areas of its range, specific efforts to maintain,

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 63 WILDLIFE ECOLOGY enhance, and restore functional habitat connectivity to these fisher and other forest carnivores in coastal Northwestern regions will be required. Specific focus should be given to the California. CDFG contract # FG-3156-WM. natural areas of forest connectivity including: the Mount Ashland Bissonette, J.A., D.J. Harrison, C.D. Hargis, and T.G. Chapin. to Siskiyou National Monument corridor (Jackson County, 1997. The influence of spatial scale and scale-sensitive Oregon; Siskiyou County, California), the area from Dunsmuir properties on habitat selection by American marten. Pages south to Lake Shasta (Siskiyou and Shasta Counties, California), 368-385 in, J.A. Bissonette, eds, Wildlife and Landscape and the Highway 199 corridor from Grants Pass (Josephine Ecology, Springer-Verlag, New York. County, Oregon) southwest to Gasquet (Del Norte County, Bull, E.L., and T.W. Heater. 2001. Home range and dispersal of California). Attention should be given to identifying specific the American marten in northeastern Oregon. Northwest areas (e.g., forested ridges, riparian areas) that currently have Naturalist. 82:7-11. suitable habitat, that are in need of habitat restoration, and areas Buskirk, S.W. and L.L. McDonald. 1989. Analysis of variability where highway crossing structures may be required to enhance or in home-range size of the American marten. Journal of facilitate safe travel of dispersing fishers. Conscious efforts to Wildlife Management 53:997-1004. maintain and connect habitat will benefit populations of fishers Buskirk, S. W. and R. A. Powell. 1994. Habitat ecology of and martens in the KSB and will also facilitate the recolonization fishers and American martens. Pages 283-296 in, S. W. of portions of the geographic range that were formerly occupied in Buskirk, A. S. Harestad, M. G. Raphael and R. A. Powell, adjacent bioregions. eds. Martens, sables, and fishers. Cornell University Press, Ithaca, New York. Acknowledgements Buskirk, S.W., and L.F. Ruggiero. 1994. The American marten. Pages 7-37, L.F. Ruggiero, K.B. Aubry, S.W. Buskirk, L.J. We would like to thank and acknowledge the following Lyon, and W.J. Zielinski, eds. American marten, fisher, individuals and organizations who helped provide financial and lynx, and wolverine in the western United States. General logistical support: Esther Burkett, California Department of Fish Technical Report, RM-254. Department of Agriculture, and Game; Greg Holm, Redwood National and State Parks; Forest Service, Rocky Mountain Forest and Range Region 5, USDA Forest Service; Redwood Sciences Lab, Experiment Station. Fort Collins, Colorado. U.S.D.A. Forest Service-Pacific Southwest Research Station; Carroll, C., W.J. Zielinski, and R.F. Noss. 1999. Using presence- Brenda Devlin and Tony Hacking, Six Rivers National Forest, absence data to build and test spatial habitat models for the U.S.D.A. Forest Service; Lee Webb, Dave Austin, Collin fisher in the Klamath Region, U.S.A. Conservation Biology Dillingham, Siskiyou National Forest, U.S.D.A. Forest Service; 13 (6):1344-1359. Shasta-Trinity National Forest, U.S.D.A. Forest Service ; Amedee Dixon, J. 1925. A closed season needed for fisher, marten, and Brickey, Arcata Office of the U.S.D.I. Fish and Wildlife Service. wolverine. California. California Fish and Game 11:23Ð25. Rich Klug, Keith Hamm, Lowell Diller, Simpson Timber Douglas, C.W. and M.A. Strickland, 1987. Fisher. Pages 511- Company. We would also like to thank Jan Werren of the 529, M. Novak, J.A. Baker, and M.E. Obbard. Redwood Sciences Lab for providing GIS support and training Drew, G.S. 1995. Newfoundland: Why old growth? and Ric Schlexer and Chet Ogan for logistical support. And Dissertation. Utah State University, Logan, Utah. finally we must especially thank and acknowledge all the field Giblisco, C.J. 1994. Distributional dynamics of modern Martes crew members. in North America. Page 484, S. W. Buskirk, A. S. Harestad, M. G. Raphael and R. A. Powell, eds. Martens, Sables, and Literature Cited Fishers: Biology and Conservation. Cornell University Press, Ithaca, New York. Agee, J. K. 1993. Fire ecology of Pacific Northwest forests. Grinnell, J., and J.S. Dixon. 1926. Two new races of the pine Island Press, Washington, D.C. marten from the Pacific Coast of North America. Zoology Anonymous. 1914. Report of fur-bearing animals. Oregon 21(16):411-417. Sportsman 2(6):20. Grinnell, J., J.S. Dixon, and J.M. Linsdale. 1937. Fur-bearing Aubry, K.B., and D. B. Houston 1992. Distribution and status of mammals of California. Volume 1. University of California the fisher in Washington. Northwestern Naturalist 73: 69- Press, Berkeley, California. 79. Jewett, S.G. 1915. The fur-bearing animals of Oregon. Oregon Aubry, K.B. and J.C. Lewis. 2003. Extirpation and reintroduction Sportsman 3: 5Ð6. of fishers (Martes pennanti) in Oregon: implication for their Jones, J. L. and E.O. Garton. 1994. Selection of successional conservation in the Pacific states. Biological Conservation stages by fishers in north-central Idaho. Pages 377-387, 117:79-90. S.W. Buskirk, A.S. Harestad , M.G. Raphael, R.A. Powell, Bailey, V. 1936. The mammals and life zones of Oregon. North eds. Martens, Sables and Fishers: Biology and American Fauna 55:1Ð416. Conservation. Cornell University Press, Ithaca, New York. Beyer, K.M., and R.T. Golightly. 1995. Distribution of Pacific Klug, R.R. 1997. Occurrence of Pacific fisher (Martes pennanti

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 64 WILDLIFE ECOLOGY

pacifica) in the Redwood Zone of northern California and Wildlife 2000: modeling habitat relationships of terrestrial the habitat attributes associated with their detections. M.S. vertebrates. University of Wisconsin Press, Madison. thesis. Humboldt State University, Arcata, California. Sheets, T. J. 1993. Washington State Pine marten distribution. Lewis, J.C. and W. J. Zielinski. 1996. Historical harvest and Department of Wildlife, Olympia, Washington incidental capture of fishers in California. Northwest Sherrell, P.E. 1970. An annotated checklist of the land mammals Science 70:291Ð297. of Curry County, Oregon. M. S. thesis, Central Washington Lewis, J.C. and D. W. Stinson. 1998. Washington State Status State College, Ellensburg. Report for the Fisher. Washington Department of Fish and Seymour, G. 1980. Traps and trapping, Part II. Commercial Wildlife, Olympia, Washington. trapping in California today. Outdoor Calif. 41(1):15-17. Marshall, D.B. 1994. Status of the American marten in Oregon *Slauson, K. M. and W. J. Zielinski. 2003. Distribution and and Washington. Audubon Society of Portland. habitat associations of the Humboldt marten and Pacific Martin, S.K. 1994. Feeding ecology of American martens and Fisher in Redwood National and State Parks. Final Report. fishers. Pages 297-315, S. W. Buskirk, A. S. Harestad, M. USDA Forest Service, Pacific Southwest Research Station, G. Raphael and R. A. Powell, eds. Martens, sables, and Arcata, California. fishers. Cornell University Press, Ithaca, New York. *Slauson, K.M. 2003. Habitat selection by a remnant population Merriam, C.H. 1890. Description of a new marten (Mustela of American martens in coastal Northwestern California. caurina) from the Northwest coast region of the United M. S. thesis, Oregon State Universit, Corvallis. States. North American Fauna 4:27-29. Strickland, M.A. and C.W. Douglas. 1987. Marten. Pages 530- Novak, M., J.A. Baker, M.E. Obbard, B. Malloch (Eds). 1987. 546, M. Novak, J. A. Baker, and M. E. Obbard, eds. Wild Wild Furbearer Management and Conservation in North furbearer management and conservation in North America. America. Ontario Ministry of Natural Resources, Toronto, Ontario Trappers Association, North Bay. Ontario. Strickland, M. A. 1994. Harvest management of fishers and Nei, M., T. Marayama, and R. Chakraborty. 1975. The bottleneck martens. Pages 149-164, S. W. Buskirk, A. S. Harestad, M. effect and genetic variability on populations. Evolution G. Raphael and R. A. Powell, eds. Martens, sables, and 29:1-10. fishers. Cornell University Press, Ithaca, New York. Phillips, D.M. 1994. Social and spatial characteristics, and Verts, B.J. and L.N. Carraway 1998. Land mammals of Oregon. dispersal of marten in a forest preserve and industrial forest. University of California Press, Berekley. M.S. thesis. University of Maine, Orono, Maine. *Zielinski, W.J., and T.E. Kucera. 1995. Track plates. American Potvin, F., L. Belanger, and K. Lowell. 1999. Marten habitat Marten, fisher, lynx, and wolverine: Survey methods for selection in a clearcut boreal landscape Conservation their detection. Pacific Southwest Research Station, United Biology 14:844-857. States Department of Agriculture, Forest Service, General Powell, R.A. 1993. The fisher: life history, ecology, and Technical Report, PSW-GTR-157. behavior. Second edition. University of Minnesota Press, *Zielinski, W. J., T.E. Kucera, and R.H. Barrett. 1995. The Minneapolis. current distribution of fishers in California. California Fish Powell, R.A., and W.J. Zielinski. 1994. Fisher. Pages 38-73, L.F. and Game 81:104-112. Ruggiero, K.B. Aubry, S.W. Buskirk, L.J. Lyon, and W.J. *Zielinski, W.J., R.L. Truex, C.V. Ogan, K. Busse. 1997. Zielinski, eds. The Scientific Basis for Conserving Forest Detection surveys for fisher and martens in California, Carnivores: American Marten, Fisher, Lynx, and Wolverine 1989-1994: Summary and interpretations. Pages 372-392, in the Western United States. USDA Forest Service, Proulx, H. N. Bryant, and P. M. Woodard, eds. Martes: General Technical Report GTR-RM-254. Taxonomy, ecology, techniques, and management. Powell, R., S., S.W. Buskirk, and W. Zielinski. 2003. Fisher and Provincial Municipal Museum of Alberta, Edmonton, Marten. Pages, 635-649, G. Feldhamer, B. Thompson, and Alberta, Canada. J. Chapman, eds. Wild Mammals of North American. 2nd *Zielinski, W.J., R.L. Truex, L.A. Campbell, C.R. Carroll, and edition. Johns Hopkins University Press, Baltimore, F.V. Schlexer. 2000. Systematic surveys as a basis for Maryland. the conservation of carnivores in California forests. USFS, Raphael, M.G. 1984. Wildlife populations in relation to stand age Pacific Southwest Research Station, Redwood Sciences and area in Douglas-fir forests of northwestern California. Lab. Unpublished Report. Pages 259-274, W. R Meehan, T. R. Merrell, and T. A. *Zielinski, W.J., K.M. Slauson, C.R. Carroll, C.J. Kent, and D.K. Hanley, eds. Fish and wildlife relationships in old-growth Kudrna. 2001. Status of American marten populations in forests. American Institute of Fisheries Research Biologists. the coastal forests of the Pacific States. Journal of Rosenberg, K.V. and M.G. Raphael. 1986. Effect of forest Mammalogy 82:478-490. fragmentation on vertebrates in Douglas-fir forests. Pages 263-272, J. Verner, M.L. Morrison, and C.J. Ralph, eds. *Available at: http://www.rsl.psw.fs.fed.us/pubs/WILD90S.html

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 65 WILDLIFE ECOLOGY

Figure 1. Historical distribution of the American marten in California and Oregon (modified from Zielinski et al. 2001). Black dots indicate verifiable historical (1919 Ð 1924 in California, mid 1800s to early 1900s in Oregon) records of marten occurrence. Solid black lines represent ranges of recognized subspecies, M.A.C. = Martes americana caurina, M.A.H. = M. a. humboldtensis, M.A.S. = M. a. sierrae. The dotted line represents the boundary of the Klamath-Siskiyou Bioregion.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 66 WILDLIFE ECOLOGY

Figure 2. Historical distribution of the Pacific fisher in California and Oregon. Black dots indicate verifiable historical (1919Ð1924 in California, mid 1800s to early 1900s in Oregon) records of fisher occurrence. Solid black lines represents the margins of the range of M. p. pacifica (Bailey 1936; Grinnell et al. 1937). The dotted line represents the boundary of the Klamath-Siskiyou Bioregion.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 67 WILDLIFE ECOLOGY

Figure 3. Survey locations within the Klamath-Siskiyou Bioregion, 1994-2002. Each black dot represents a sample unit. The three clumped areas of black dots represent 2-km spaced sample grids.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 68 WILDLIFE ECOLOGY

Figure 4. American marten detections within the Klamath-Siskiyou Bioregion, 1996-2002. Large black dots represent sample units where martens were detected; small dots represent those where martens were not detected.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 69 WILDLIFE ECOLOGY

Figure 5. Fisher detections within the Klamath-Siskiyou Bioregion, 1996-2002. Large black dots represent sample units where fishers were detected; small dots represent those where fishers were not detected.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 70 DISTURBANCE AND CHANGE

FIRE AND VEGETATION DYNAMICS IN THE WESTERN KLAMATH MOUNTAINS

Dennis C. Odion1 Institute for Computational Earth Systems Science, University of California, Santa Barbara, CA 93106 Email: [email protected]

Evan J. Frost Wildwood Environmental Consulting, 84 4th Street, Ashland, OR 97520

Dominick A. DellaSala World Wildlife Fund, 116 Lithia Way, Ashland, OR 97520

James R. Strittholt and Hong Jiang Conservation Biology Institute, 260 SW Madison Ave. Suite 106, Corvallis, OR 97333

Max A. Moritz Environmental Science, Policy, and Management Department, Univ. of California, Berkeley, CA 94720

Abstract

Fires have been important in the Klamath-Siskiyou Region for millennia. Where burn severities are mixed, as they are in much of the Klamath-Siskiyou, fires may be instrumental in creating landscape heterogeneity. Landscape heterogeneity is linked to structural and taxonomic diversity, for which the Klamath-Siskiyou region is renowned. We recently analyzed spatial patterns of fire severity in a 98,814 ha area burned in 1987 to evaluate the role of fire in creating landscape diversity. The study area was in the Marble and Trinity Mountains area, which is dominated by Douglas-fir and tanoak forests. We found that fire severity was mostly low (59% of the area), but that patchy high severity fire (12% percent of the area) also occurred. In forests and sclerophyll vegetation where fire had been previously absent for many decades less high severity fire was found, whereas those that had burned previously in prior decades had more high severity fire. Therefore, the length of fire-free intervals can influence the production of high severity patches in the landscape. Fire interval may also determine the persistence of these patches, which return to forest vegetation following many decades without stand-replacing fire. The high degree of spatial and temporal variation in fire in our study area results in a landscape of complex and dynamic patches of differing ages. Patch dynamics differ in landscapes containing even-aged plantations. We found these burned with twice the severity of closed forests. Positive feedback between plantations and fire, and the persistence of plantations over fire intervals of any length favor the filling in of landscapes with structurally and biologically simplified vegetation. Introduction (Turner et al. 2003). Such patchy landscape heterogeneity is linked to species diversity (Baker 1992). The pre-settlement fire he importance of the Klamath-Siskiyou region as a regime in the western part of the Klamath-Siskiyou is believed to floristic center of diversity, as described in classic papers have been strikingly heterogeneous (Wills and Stuart 1994). Fires T by Whittaker (1960; 1961) and Stebbins and Major of variable frequency and intensity created landscapes containing T complex age and structure mosaics. We previously studied a (1965), has led to the ecoregion being recognized for its global biological significance (DellaSala et al. 1999; DellaSala 2004, this major fire event that occurred in 1987 to evaluate patterns of issue). This remarkable diversity is generally attributed to the severity in contemporary burning in an area centered on the likewise extraordinary geologic and climatic diversity (Wallace Marble and Trinity Mountains in the Klamath-Siskiyou (Odion et 1983), and the role the region played as a refugium during glacial al. 2004, Figure 1). Here our objective is to describe how maxima (Whittaker 1960). However, other landscape processes variation in the severity and frequency of fire, operating over time such as fire also may have played a role in promoting high levels in this landscape, will lead to complex vegetation dynamics. We of species diversity. further evaluate how human influences such as fire suppression Spatial variation in fire effects can create spatially complex and silviculture may be altering these dynamics. landscapes that favor different forms of native plant regeneration Materials and Methods 1 Address for correspondence: Odion Consulting, 670 Morton St., Our study area (Figure 1) focused on a 500,000 ha portion Ashland, OR 97520

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 71

DISTURBANCE AND CHANGE of the Klamath National Forest that in many respects was ideal for sclerophyll vegetation. It is dominated by species of manzanita our objectives. A complex of lightning ignitions produced an (Arctostaphylos) and Ceanothus, which are true shrubs, and young extensive burned landscape (98,814 ha) equally divided among hardwoods and conifers that have not yet become arborescent. roaded and unroaded portions that were very similar in terms of The amount of time it takes for tree species to become general topography. arborescent, and forests to develop, varies from about 60-100 We acquired digital maps of burn severity prepared by the years after fire (Thornburgh 1982). There are also numerous Klamath National Forest using aerial photos. Severity was Douglas-fir and mixed conifer plantations in the study area (3965 mapped as low, moderate, or high. High severity was complete ha in 1987). We classified vegetation based on satellite imagery crown scorch or crown fire that consumed foliage. These burn taken just prior to the 1987 fires (Odion et al. 2004). The classes effects are generally lethal to non-sprouting species such as identified are described in Table 1, which provides an overview of conifers. Thus, mortality was close to 100%. Low severity was vegetation structure in the landscape. based on less than 50% crown scorch and represents understory, or Climate of the study area is predominantly Mediterranean surface fire. This causes little or no direct mortality of overstory (Taylor and Skinner 1998), with warm, dry summers and cool, wet trees. Moderate severity, 50-99% crown scorch, is a fine scale winters. Precipitation ranges from about 1,100 mm to over 2,000 mixture of low severity and localized high severity fire that creates mm based on the few stations with long-term data. Ninety percent canopy gaps and selective mortality of smaller size class trees. We falls between October and May. Average maximum temperature evaluated patterns of fire severity in relation to fire history and during August in a typical Douglas-fir/hardwood dominated area o plantation data also obtained from the Klamath National Forest. near the center of the study area is 33.5 C (Stuart et al. 1993). We also evaluated severity patterns in relation to vegetation While the July average for Happy Camp, 330 m elevation in the determined by analysis of satellite (Landsat) data. Methods and north-central part of the study area is 25.7 oC (Taylor and Skinner severity descriptions are described in detail in Odion et al. (2004). 1998). We analyzed characteristics of spatial pattern in fire severity During the large wildfire events in the region that began on with FRAGSTATSª, version 2.0, a fragmentation assessment 30 August 1987, so many lightning fires ignited, control actions software. FRAGSTATSª computes numerous metrics for were ineffective in stopping them (Agee 1993). This was evaluating patchy landscape phenomena (McGarigal and Marks compounded by prolonged drought and record-breaking spring 1995). heat that desiccated fuels to unusually low levels (Reider 1988). The wet temperate, mixed evergreen and conifer forests of Much of the area that burned in 1987 had been fire-free since 1911 the western Klamath have been described by Bingham and Sawyer based on the Klamath National Forest’s fire history data. Large (1992), Sawyer et al. (1977), Sawyer and Thornburgh (1977) and fires in the region occurred in 1910 (Agee 1993). Jimerson (1990). Complex, geology and rugged, steep topography along with the range in elevation contribute to rich floristic Results diversity in the Klamath (DellaSala et al. 1999). Stuart et al. (1993) describe forests in the center of our study area as being Although fuel moisture was especially low at the time of the typical of the hot, dry Douglas-fir (Pseudotsuga menzesii) 1987 fires due to severe drought, only 12% of the landscape association, which is characterized by relatively discontinuous burned at high severity. Another 29% burned at moderate and the Douglas-fir overstories with a relatively continuous midstory of remainder (59%) at low severity. Low and moderate severity fire the evergreens tanoak (Lithocarpus densiflorus), Pacific madrone comprised a landscape matrix of forests thinned varying degrees (Arbutus menzesii) and canyon live oak (Quercus chrysolepis), as by fire. High severity patches, median size 3.78 ha, were well well as the deciduous black oak (Quercus kellogii), and bigleaf dispersed throughout this matrix (Figure 1). The average size of maple (Acer macrophyllum). Common forest understory shrubs high severity patches (18.7 ha) was much bigger than the median include poison oak (Toxicodendron diversilobum), Oregon grape due to the effects of infrequent larger patches (Figure 2). The (Berberis nervosa) and hazelnut (Corlylus cornuta). This general largest patches are associated with management activities in forest vegetation occupied about 65% of the study landscape that previous burns, as described below. burned in 1987, according to the Klamath National Forest’s The three closed forest types occupied most of the 1987- Timber Type maps. From west to east, and with increasing burned landscape (Table 1), and experienced predominantly low elevation, tanoak and other hardwoods are replaced by white fir and moderate fire severity (Figure 3). High severity fire ranged (Abies concolor) dominated forests, which accounted for ~10 % from 7.8-11.4%, with the lowest proportion occurring in tall, cover in the burned area. Ponderosa and sugar pines (Pinus multi-canopy forests. Woodlands and especially sclerophyll ponderosa and P. lambertiana) also are scattered in all these forest vegetation experienced much more high severity fire than the three types, and some drier situations are dominated by ponderosa pine closed forest types (Figure 3). Fifty percent of all the high (~5 % of burned landscape). The highest elevations are dominated severity burn acreage occurred in these two vegetation types by red fir (A. magnifica, 4% of burned landscape). although they occurred on only 17% of the burned landscape. Non-forest vegetation includes hardwoods and/or chaparral One-third (3,339 ha) of the sclerophyll vegetation that that are often intermixed. Most of this vegetation is established burned had previously burned in 1966 and especially 1977 following stand-replacing fire. For convenience, we refer to this as according to the Klamath National Forest’s fire history data. This

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 72 DISTURBANCE AND CHANGE sclerophyll vegetataion had 51% and 57% high severity in these 1994). Lindenmayer and Franklin (2002) have commented that former burn areas, respectively, compared to 20% in the remainder such naturally developed early successional habitat, with its rich of the 1987 burn landscape. The 1977 burn area is apparent in array of snags and logs and non-arborescent vegetation, is an Figure 1, just to the south of the Marble Mountains Wilderness. especially important component of biodiversity. The patchy Comparing the burn severity patterns with those occurring in the habitat is often replaced by regenerating forest (Wills and Stuart 1977 burn (mapped by the same photo interpreters using the same 1994; Skinner 1995), but this process can be interrupted by methods as the 1987 maps), we found that 60% of the area that another stand-replacing fire. Thus, the return interval for stand- burned at high severity in the earlier burn also burned at high replacing fire determines the dynamic balance between mature severity in 1987. conifer forests vs. sclerophyll vegetation containing a new Areas of forest that burned severely in 1977 were generation of conifers (Figure 6). subsequently logged with helicopters. Activity fuels (tree canopies, The combination of both spatial and temporal heterogeneity small, dead trees) left untreated after such logging could have in fire leads to exceptional variability in the return interval for contributed to the greater fire severity in this sclerophyll stand replacement (Table 2). Most fire intervals in the study area vegetation. We obtained data showing the 8 separate timber sales in recent centuries are estimated to have ranged from about 5-75 following the 1977 burn and created a map by crosswalking these years at low and mid elevations (Wills and Stuart 1994; Taylor and to the Klamath National Forest managed stands layer. The map is Skinner 1998, 2003), and about twice as long in higher elevation considered to significantly underestimate the extent of this logging forests (Stuart and Salazar 2000). These estimates are based on by the Klamath National Forest by only indicating areas initially fire scar analyses, which may considerably underestimate the harvested. Nonetheless, where this post-fire logging was length of fire free periods for portions of the landscape (Minnich documented, and vegetation classified as sclerophyllous (285 ha), et al. 2000; Baker and Ehle 2001). However, using 5-75 years as there was 74% high severity. The three largest and 6th largest high an estimate for the range of most fire return intervals, the potential severity patches were in areas that burned in 1966 and 1977. range in the interval of stand-replacing fire may exceed 1,000 Logging slash and other management influences associated with years in forests (Table 2). This interval may often be long enough the earlier burns may have elevated fire severity when they (100 yrs) to allow sclerophyll vegetation to be replaced by reburned in 1987. If the large high severity patches in the 1966 regenerating forests (Table 2). and 1977 burn areas are excluded, the amount of high fire severity Although exogenous forces (climate, weather, lightning- for the remainder of the landscape falls to 8.4%. ignition patterns) may dictate a high degree of temporal and spatial There were 3965 ha of plantations burned in the 1987 fires, variation in stand-replacing fire, endogenous properties of and they were more combustible than natural closed forests vegetation can have an important influence on this variation. The (Figure 4). Also, unlike natural vegetation, plantations did not dense foliage of many sclerophylls is receptive to combustion and decrease in combustibility with stand development because the has high rates of energy release upon burning (Agee 1993). Thus, amount of moderate severity fire increased significantly with sclerophyll vegetation may exhibit a self-reinforcing relationship plantation age (Figure 5, F = 9.03, p = 0.01). Both low and high with recurrent fire (Show and Kotok 1924). As this vegetation severity fire tended to decrease, but the relationships were weak ages, tall hardwood resprouts and young conifers begin shading (R2< 0.1). Compared to natural forests, moderate severity fire may out the pyrogenic shrubs (Thornburgh 1982), resulting in lower lead to greater mortality in plantations due to the absence of larger, fire severity in older, taller sclerophyll vegetation (Odion et al. fire resistant trees. 2004). Even lower fire severity occurs in forests in the study area (Figure 3), especially long-unburned forests (Odion et al. 2004). Discussion Thus, there is a dramatic difference in the combustion properties between forests, particularly those long-unburned, and sclerophyll Massive fires, such as those in 1987, can dominate vegetation, especially when it is relatively young. Bond and van landscape-scale pattern, creating heterogeneity that is positively Wilgen (1996) describe a number of examples of alternative states associated with structural and taxonomic diversity (Baker 1992). of vegetation that facilitates fire, and vegetation that does not, Despite a fire suppression policy, large fires continue to be occurring in dynamic balance. These are found in many parts of important in the study area and surrounding region. Those of the world, including other Mediterranean ecosystems. The recent decades have been characterized by mostly low and occurrence of fire specialized, pyrogenic vegetation in dynamic moderate severity, with patches of high severity (Odion et al. balance with non-pyrogenic vegetation leads to diverse life history 2004). Over time, this spatial heterogeneity in fire creates traits in relation to fire in these ecosystems, contributing to their complex, multi-age forest structure (Taylor and Skinner 1998), and rich floras (Bond and van Wilgen 1996). dynamic vegetation mosaics (Wills and Stuart 1994). High Plantation forestry appears to have much potential to severity burn patches in 1987 were highly variable in size and influence the natural landscape dynamics and patterns contributing shape (Figsures 1-2). to diversity. Because of an absence of larger, fire resistant trees, Patchy, stand-replacing fire in forests creates the conditions plantations will be sensitive to fire. The increase in moderate necessary for sclerophylls and a new generation of conifers to severity fire with plantation age (Figure 5), likely means that establish (Thornburgh 1982; Stuart et al. 1993; Wills and Stuart plantations will remain sensitive to fire as they age. Given the

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 73 DISTURBANCE AND CHANGE amount of fire in the landscape in recent decades plantations may landscape structure. Ecology: 73: 1879Ð1887. not succeed to relatively fire resistant forest. Once established, Baker, W.L., and D. Ehle. 2001. Uncertainty in surface-fire plantations may represent a persistent, non-indigenous patch type history: the case of ponderosa pine forests in the western as long as they continue to be replanted after fires (Figure 7). United States. Canadian Journal of Forest Research 31: Moreover, because plantations can elevate fire severity in 1205-1226. surrounding forests (Key 2000) and they are often planted after Bond, W.J., and B.W. van Wilgen. 1996. Fire and plants. high severity fire, a positive feedback relationship is possible Chapman and Hall, London. (Perry 1995). This may be described by analogy with the Christensen, N.L. 1991. Wilderness and high intensity fire: how widespread, well-documented phenomenon of exotic grass much is enough. Pages 9-24 in Proceedings of the 17th Tall invasion that leads to elevated fire severity favoring the grasses Timbers Fire Ecology Conference, May, 1989, Tallahassee, (reviewed by Mack and D’Antonio 1998). Plantations in our Florida. study area have grown to cover almost one-third of the area DellaSala, D.A. 2004. State of the Klamath Knot: how far have burned in 1987 outside wilderness areas, increasing the likelihood we come and where are we going. Proceedings of the for future positive feedback with fire. This phenomenon is not Second Conference on Klamath-Siskiyou Ecology. Siskiyou limited to the study area. Elsewhere in the Klamath-Siskiyou, Field Institute. plantation establishment following high severity fire is common DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt and D.M. practice, and it is proposed for an extensive area burned in the Olson. 1999. A global perspective on the biodiversity of 2002 Biscuit Fire, just north of the study area. Lindenmayer and the Klamath-Siskiyou Ecoregion. Natural Areas Journal Franklin (2002) note that naturally developed early successional 19:300-319. forest habitat is a threatened component of biodiversity in large Jimerson, T.M. 1990. A seral stage and successional pathway part because of this widespread management practice. model for the tanoak-canyon live oak/evergreen huckleberry ecological type on the Gasquet Ranger District, Six Rivers Conclusions National Forest. Ph.D. dissertation. University of California, Berkeley. Mixed severity fire as found in the 1987 burn areas may be Key, J. 2000. Effects of clearcuts and site preparation on fire instrumental in creating landscape heterogeneity and dynamics severity, Dillon Creek Fire 1994. M.S. thesis, Department of that are linked to structural and taxonomic diversity. The Forestry, Humboldt State University, Arcata, California. establishment of even-aged plantations in landscapes may threaten Lindenmayer, D.B., and J.F. Franklin. 2002. Conserving forest to disrupt these processes. The expansion of these non-indigenous biodiversity: a comprehensive multi-scaled approach. Island fuel complexes can be fostered by fire. This can lead to Press, Washington, D.C. landscapes having a large proportion of plantation acreage, as is Mack, M.C., and C.M. D’Antonio. 1998. Impacts of biological now found in portions of our study landscape as a result of the invasions on disturbance regimes. Trends in Ecology & 1987 fires. Evolution 13:195-198. McGarigal K., and B.J. Marks. 1995. FRAGSTATS: spatial Acknowledgements pattern analysis program for quantifying landscape structure (Fragstats NT Version 2.0). USDA Forest Service General World Wildlife Fund provided funding for this project. M. Technical Report PNW-GTR-351. Creasy (Klamath National Forest) provided helpful suggestion Minnich, R.A., M.G. Barbour, J.H. Burk, and J. Sosa-Ramírez. regarding interpretation of the results. We appreciate constructive 2000. Californian mixed-conifer forests under unmanaged feedback from M. Turner, M. Murray, D. Perry, J. Keeley, J. fire regimes in the Sierra San Pedro Mártir, Baja California, Pagel, and D. Thornburgh. Mexico. Journal of Biogeography 27:105-129. Odion, D.C., J.R. Strittholt, H. Jiang, E.J. Frost, D.A. DellaSala, and M.A. Moritz. 2004. Patterns of fire severity and Literature Cited forest conditions in the western Klamath Mountains, California. Conservation Biology (in press). Agee, J. K. 1993. Fire ecology of Pacific Northwest forests. Perry, D.A. 1995. Self-organizing systems across scales. Trends Island Press, Washington, D.C. in Ecology & Evolution 10:241-244. Bingham, B.B., and J.O. Sawyer. 1992. Canopy status and tree Reider, D.A. 1988. California conflagration—recounting the condition of young, mature and old-growth Douglas- siege of ’87. Journal of Forestry 86:5-8. fir/hardwood forests. Pages 141-149 in R. R. Harris, D.E. Sawyer, J.O., and D.A. Thornburgh. 1977. Montane and Erman and H.M. Kerner, eds. Biodiversity of northwestern subalpine vegetation of the Klamath Mountains. Pages 699- California. Wildland Resource Center Report No. 29. 732 in M.G. Barbour and J. Major, eds. Terrestrial University of California, Berkeley. Vegetation of California. Wiley-Interscience, reprinted by Baker, W.L. 1992. Effects of settlement and fire suppression on the California Native Plant Society 1988. Sacramento,

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 74 DISTURBANCE AND CHANGE

California. Management 111:285-301. Sawyer, J.O., D.A. Thornburgh and J.R. Griffin. 1977. Mixed Taylor, A.H., and C.N. Skinner. 2003. Spatial and temporal evergreen forest. Pages 359-381 in M.G. Barbour and J. patterns of historic fire regimes and forest structure as a Major, eds. Terrestrial vegetation of California. Wiley- reference for restoration of fire in the Klamath Mountains. Interscience, reprinted by the California Native Plant Ecological Applications 13:704-719. Society 1988. Sacramento, California. Thornburgh, D.A. 1982. Succession in the mixed evergreen forests Show, S.B., and E.I. Kotok. 1924. The role of fire in California of northwestern California. Pages 87-91 in J.E. Means, pine forests. U.S. Department of Agriculture Bulletin 1294. editor. Forest succession and stand development research in Skinner, C.N. 1995. Change in spatial characteristics of forest the Northwest. Oregon State University, Forest Research openings in the Klamath Mountains of northwestern Laboratory, Corvallis. California, USA. Landscape Ecology 10:219-228. Turner, M.G., W.H. Romme and D.B. Tinker. 2003. Surprises Stebbins, G.L., and J. Major. 1965. Endemism and speciation in and lessons from the 1988 Yellowstone fires. Frontiers in the California flora. Ecological Monographs 35:1-35. Ecology and Environment 1:351-358. Stuart, J.D., and L.A. Salazar. 2000. Fire history of white fir Wallace, D.R. 1983. The Klamath Knot. Sierra Club Books, San forests in the coastal mountains of northwestern California. Francisco, California. Northwest Science 74:280-285. Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, Stuart, J.D., M.C. Grifantini, and L. Fox III. 1993. Early Oregon and California. Ecological Monographs 30:279-338. successional pathways following wildfire and subsequent Whittaker, R.H. 1961. Vegetation history of the Pacific coast silvicultural treatment in Douglas-fir/hardwood forests, NW states and the central significance of the Klamath region. California. Forest Science 39:561-572. Madroño 16:5-23. Taylor, A.H., and C.N. Skinner. 1998. Fire history and landscape Wills, R.D., and J.D. Stuart. 1994. Fire history and stand dynamics in a late-successional reserve, Klamath development of a Douglas-fir/hardwood forest in northern Mountains, California, USA. Forest Ecology and California. Northwest Science 68:205-212.

Table 1. Five pre-burn vegetation formations identified by unsupervised classification of 1986 Landsat image.

Vegetation Area (ha) Successional Distinguishing features Dominant species/vegetation Class (% of total) Satus Tall, well developed, Multi-canopy 29,782 Tall Douglas-fir overtopping hardwoods. Mixed textured, closed canopy Old-growth forest (30.1) conifer and ponderosa pine. Old growth. forest. Tall, dense forest with Relatively old Douglas-fir or tanoak/hardwoods. Tall, single- 27,853 relatively uniform, closed Old-growth Mixed conifer at higher elevations. Old growth. canopy forest (28.2) canopy. Short, Lower statured forest with Dense hardwoods (tanoak, madrone, oaks), and/or 20,253 single canopy relatively uniform canopy. Mid-successional young Douglas-fir, white fir and other conifers, (20.5) forest older plantations.

Hardwood sclerophyll Garry and black oak woodlands, often with 8,893 Woodlands vegetation. Canopy not Variable Douglas-fir and other conifers. Understory of grass, (9.0) closed. scattered conifer saplings and shrubs.

Chaparral and resprouting, relatively short-statured Low stature, evergreen, 12,033 Primarily early hardwoods, especially in areas burned in 1977. Shrubland sclerophyll vegetation, (12.1) successional Mixed with conifer seedlings and saplings. Young generally dense. conifer plantations.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 75 DISTURBANCE AND CHANGE

Table 2. Estimated potential frequency of stand replacement fire based on severity in 1987 fires and reported fire frequencies for predominant, low to medium forest types in the study area (Wills and Stuart 1994; Taylor and Skinner 1998). Shrubland includes hardwoods and young conifers regenerating from fire.

Stand replacement frequency

Closed Forest Vegetation Shrubland

Stand Replacing Fire 7-16% 19-50%

Fires needed to replace area = all stands 6-14 2-5 x Fire Frequnecy 5-75+ years 5-75+ years

Initial Establishment Potential Stand Replacement Frequency to 10-375+ years 1050+ years

* Fire frequency estimates from tree ring records unavailable, so figures for forests are assumed to apply in shrublands.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 76 DISTURBANCE AND CHANGE

Figure 1. Map of the study area showing severity of the 1987 wildfires.

Unroaded Roaded High Severity Moderate Severity Low Severity

Marble Mountain Wilderness

Trinity Alps N Wilderness

KILOMETERS

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 77 DISTURBANCE AND CHANGE

Figure 2. Patch sizes for high severity fire. Only patches less than150 ha are shown. There were 12 additional patches from

203 to 1143 ha in size accounting for a half of all high severity fire. Number of Patches of Number

Size (ha)

Figure 3. Fire severity proportions in the 5 different vegetation types identified. Burned Area % Burned

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 78 DISTURBANCE AND CHANGE

Figure 4. Fire severity proportions in natural, closed forests vs. plantations. Area % Burned

Figure 5. Proportion of moderate severity fire as a function of plantation age.

Moderate Severity

R2 = 0.2653 Percent Severity

Plantation Age

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 79 DISTURBANCE AND CHANGE

Figure 6. State (boxes) and process (ovals) model depicting fire and interrelated dynamics between forests and shrublands in the study area. The size of the boxes reflect the current proportions of the two vegetation types in the landscape. Natural Vegetation Dynamics

Closed Forest Vegetation

Succession Stand-replacing Fire with no stand-replacing fire for more than ~ 100 years

chaparral, sclerophyll woodland, young conifers Stand-replacing fire

Figure 7. State (boxes) and process (ovals) model depicting fire, management and dynamics leading to an increase in plantation lands in the study area.

Natural Forest, Plantation Interactions

Closed Forest Vegetation

Stand-replacing Fire Clear cut

Plantation Stand-replacing Fire

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 80 DISTURBANCE AND CHANGE

LOSING GROUND: WILDERNESS MEADOWS AND TREE INVASION OVER A 55-YEAR PERIOD IN CALIFORNIA’S KLAMATH RANGE

Michael P. Murray Crater Lake National Park, P.O. Box 7, Crater Lake, OR 9760 Email: [email protected]

Abstract Mountain meadows within Klamath-Siskiyou Wilderness areas are renowned habitat for diverse flora and fauna. Although tree invasion has been detected with small-scale investigations in other regions, the Klamath-Siskiyou has not been examined previously. The objectives of this study were to 1) determine whether tree encroachment has occurred, and 2) measure the degree of associated meadow loss. By applying computerized image analysis to aerial photographs (1944 and 1999), meadow loss was verified. Loss varied between 2 and 13% among the six meadow sites studied. The loss of meadow acreage during the 55-year time period has important implications for the natural landscape. Likely causal agents include climate change, grazing, and fire exclusion. It is recommended that further research, monitoring, restoration, and management improvements be implemented to sustain these remarkable ecosystems. Introduction expanding canopies and shade-intolerant plants dwindle. Multiply each tree’s associated shade zone by the thousands which invade, ountain meadows are vital ecosystems of the Klamath- and we recognize how meadows are losing ground. Tree Siskiyou biotic landscape. Meadows flourish on valley establishment may have other important implications such as M bottoms, hanging slopes, rolling ridgetops, glacial altering soil nutrient ratios, fire effects, hydrology, and displacing M meadow-dependent wildlife. moraines, cirque basins, and other mountain landforms. They display an impressive diversity of flora and fauna. Lizards, birds, Throughout western North American there remains a rodents, bears, bees, and butterflies, are a small portion of the pernicious void of knowledge regarding the actual degree of animal community sustained by Klamath-Siskiyou meadows. invasion at a landscape-scale. The objective of this study is to Rare meadow plants include Siskiyou sedge (Carex gigas), examine the Klamath-Siskiyou Region to 1) determine whether Klamath gentian (Gentiana plurisetosa), wing-seed draba (Draba tree encroachment has occurred, and 2) measure the degree of pterosperma), long-haired star-tulip (Calochortus longebarbatus associated meadow loss. With rapidly improving remote sensing var. longebarbatus), and mountain hairgrass (Vahlodea and imagery technology, we can effectively assess change across atropurpurea). Richness of vascular species is known to reach broad areas. This study is a first-step to identify and understand over 200 in a single square mile. meadow loss in the Klamath-Siskiyou Region. Due to a high diversity of biophysical settings in the region, there is an impressive array of meadow communities. At least six Methods broadly defined communities exist, including sedge, sedge- bunchgrass, lush-herbaceous, corn-lily, bentgrass, and dry sparse To select potential analysis sites, topographic maps and US types (Palmer 1979; Stillman 1980; Murray 1991; Van Sickle Geological Survey digital orthophotographs were examined to find 1995). open vegetated areas above approximately 1,341 meters in Meadows have historically supported meager tree cover Ð elevation (upper montane and subalpine). Based on this criterion, either occurring in patches or a few hardy individuals peppered I acquired aerial photographs that corresponded with forty about. However, during the late 1900s observations of potential sites. Klamath National Forest staff provided full access pronounced tree establishment along forest edges and within to two sets (1944 and 1999) which are 9x9-inch at approximately meadows of the Pacific Northwest were noted (Brink 1959; 1:16,000 scale, black-and-white and color respectively. Each Franklin and others 1971). These woody pioneers can gain a photo was digitized using a flatbed scanner at 600 dpi resolution. toehold where most others have succumbed to fire, drought, deep The next step entailed randomly selecting six meadow study snow, rodents, and a variety of other tree inhibitors which sustain sites. Five of forty-two photos were selected blindly. A sixth meadowlands. But over the course of a few decades young selection, Paradise Lake’s meadow, selected for a pilot analysis, conifers have propagated in pockets across the open landscape. was not random. Initially, several selected sites were dropped This trend is also apparent in the Sierra Nevada and Rocky from analysis because they were not meadows (i.e. shrubland, Mountain regions (Butler 1986; Vale 1987). serpentine scabland, etc.). After five suitable sites were added, The recent invasion warrants our attention. Most meadow each was analyzed separately within a geographic information plants prefer sunlight which is unfiltered by invading conifers. system (GIS) utilizing ArcView software and the Image Analysis Maturing trees progressively shade more area beneath their extension (ERDAS 1998). Each photo was georectified to the

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 81 DISTURBANCE AND CHANGE

UTM coordinate system. This software also allows a customized establishment. The duration of snow cover is a reflection of both search based on greenness of tree foliage which the analyst snowfall totals and ambient temperatures. Heavy summer rainfall supervises. Thus, shadows and other shades of green associated can also increase tree establishment in coarse-soiled (dry) with meadow turf and shrubs can be avoided with careful meadows. attention. For each pair of photographs (1944 and 1999), a Another allogenic factor which can lead to tree invasion is meadow boundary was delineated. This is defined as the livestock grazing (Miller and Halpern 1998; Taylor 1990; Vale ‘meadow site’. The boundary was somewhat arbitrarily mapped 1981). Grazing reduces meadow cover, thus reducing plant from the 1944 photograph to exclude surface water, exposed rock, competition while enabling tree seeds to find soil. Continued shrubs, and forest edges. Within each meadow site, some heavy grazing tends to injure and kill seedlings, but a reduction in scattered trees were usually present in 1944. These were included stock actually allows trees to flourish. Meadows at Paradise and in the calculation of canopy cover (sq. m.) and then compared to Sky High Lakes are similar habitats with historic heavy grazing. the tree area of 1999. The resulting increase in tree cover serves Analysis of tree ages and grazing records may indicate livestock as an estimate of meadow loss. The software was able to select use as the catalyst at these two sites. individual trees representing a minimum area of 0.5 sq. m. The suppression of lightning-ignited fire also appears to The six meadow sites are from a variety of landform favor tree establishment (Vale 1981; Taylor 1990). Although vast settings (Table 1). All are within the Marble Mountain Wilderness acreages of meadows are remote and far-removed from human Area, except Morris Meadow which is in the Trinity Alps settlements, fire suppression is routinely executed by land Wilderness. management agencies in Klamath-Siskiyou wilderness areas. Morris Meadow is surrounded by abnormally dense pine-cedar FINDINGS AND DISCUSSION forest indicating a fire-starved condition. Yet many older trees exhibit fire scars. It’s logical to surmise that the meadow’s tall The six meadow sites analyzed amount to 477.5 hectares. grassland would have been maintained by repeated fires killing Meadow loss during the 55-year period varied between 2-13% young invading trees. among sites (Table 2). Although isolated tree mortality was evident, an overall gain in meadow acreage was not. Tree Recommendations establishment was observed both along meadow edges and well within their extent (Figure 1). Although shrub (willow) cover was not analyzed, similar increases were observed on the photographs. ¥ Determine if Human-caused Whereas trees colonized previously unoccupied microsites, shrubs To identify cause, historic records (e.g. climate, tended to expand from 1944 patch edges. A combination of both grazing, fire history) should be examined with tree- shrub and tree increases equate to 4-26% meadow loss. ring samples. Where tree initiation dates are similar Lack of similar investigation in the Klamath-Siskiyou and correspond to an historical allogenic factor Ð a Region and elsewhere limits comparison of these results. Skinner link can be argued. (1995) examined forest gaps composed of shrubs and herbs in determining a 39% increase in tree cover across a predominantly ¥ Estimate susceptibility (risk) forested landscape in the region. This somewhat contrasts with Different meadow habitats likely vary in their my results and perhaps suggests that mechanisms inherent to susceptibility to tree encroachment. Analysis can be meadow maintenance (e.g. plant competition, herbivory, soil performed based on field data which models risk for conditions) remain effective to a greater extent than in Skinner’s different meadow habitats. Thus, priorities for forest openings. monitoring and restoration can be set. Loss of 4-26% of individual meadows represents a reduction in habitat for diverse meadow biota with additional ¥ Monitor Meadows impacts on natural processes (fire, hydrology, nutrient cycling). Canopy cover of meadow flora and invading trees can The two Wilderness Areas studied are mandated by Congress to be periodically measured to estimate trends in “preserve natural conditions...with the imprint of man’s work abundance. Where restoration (i.e. tree removal has substantially unnoticeable.” Thus the question is posed whether occurred), response of meadow vegetation should be Klamath-Siskiyou meadow loss is human-caused. investigated. Rather than a single factor, a combination of elements has probably caused meadow area loss. Environmental agents which ¥ Restoration aid or trigger tree invasion tend to originate from outside the Tree removal is justified where human factors are the immediate meadow ecosystem (Table 3). Climatic fluctuations identified cause for meadow loss. are most commonly linked with tree invasion (Brink 1959; Franklin and others 1971; Vale 1981; Helms 1987; Taylor 1995; ¥ Manage for Ecological Integrity Rochefort and Peterson 1996; Miller and Halpern 1998). Snow Meadowlands can be self-sustaining if management cover inhibits trees, thus longer snow-free periods foster takes a light approach. Lightning-ignited fires

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 82 DISTURBANCE AND CHANGE

should be permitted and grazing needs to be carefully and grazing disturbance on tree establishment in meadows controlled. Otherwise, meadows will surely be of the central Cascade Range, Oregon, USA. Journal of reduced or sacrificed. Vegetation Science 9:265-282. Murray, M.P. 1991. Meadow vegetation change in the subalpine Conclusions zone of the Marble Mountain Wilderness. M.S. Thesis. Humboldt State University, Arcata, California. Tree encroachment within meadowlands of the Trinity Alps Palmer, J.S. 1979. Vegetation on quartz diorite in the Bear Lakes and Marble Mountains was detected. It’s estimated that between 2 Area, Trinity County, California. M.S. Thesis. Humboldt and 13% of meadow acreage has been lost to tree establishment State University, Arcata, California. between 1944 and 1999. An additional like amount has been Rochefort, R.M., and D.L. Peterson. 1996. Temporal and spatial usurped by shrubs. With up to 26% of individual meadow acreage distribution of trees in subalpine meadows of Mount Rainer lost during this period, the ecological integrity of Wilderness is National Park, Washington, U.S.A. Arctic and Alpine compromised where human factors are responsible. Further Research 28(1):52-59. investigation, monitoring, and careful management is warranted to Skinner, C.N. 1995. Change in spatial characteristics of forest sustain the extraordinary meadows of this region. openings in the Klamath Mountains of northwestern California, U.S.A. Landscape Ecology 10(4):219-228. Acknowledgements Stillman, K.T. 1980. Meadow vegetation on metasedimentary and metavolcanic parent materials in the north central Marble Mountains, California. M.S. Thesis, Humboldt State I am very thankful to the California Native Plant Society for University, Arcata, California. making this work possible with an educational grant from the Taylor, A.H. 1995. Forest expansion and climate change in the Sharsmith Fund. My thanks also go to staff of the Klamath mountain hemlock zone (Tsuga mertensiana) zone, Lassen National Forest who provided access and assistance with aerial Volcanic National Park, California, U.S.A. Arctic and photographs. Alpine Research 27(3)207-216. Taylor, A.H. 1990. Tree invasion in meadows of Lassen Volcanic Literature Cited National Park, California. Professional Geographer Brink, V.C. 1959. A directional change in the subalpine forest- 42(4):457-470. heath ecotone in Garibaldi Park, British Columbia. Ecology Vale, T.R. 1981. Tree invasion of montane meadows of Oregon. 40(1):10-16. American Midland Naturalist 105(1):61- 69. Butler, D.R. 1986. Conifer invasion of subalpine central Lemhi Vale, T.R. 1987. Vegetation change and park purposes in the high Mountains, Idaho. Northwest Science 60(3):166-173. elevations of Yosemite National Park, California. Annals of ERDAS. 1998. Using ArcView image analysis. ERDAS, Inc. the Association of American Geographers 77(1):1-18. Atlanta, Georgia. Van Sickle, V. 1995. Comparison of vegetation patch composition, Franklin, J.F., W.H. Moir, G.W. Douglas, and C. Wiberg. 1971. biodiversity and wildlife, human and livestock use of Invasion of subalpine meadows by trees in the Cascade Marble Mountain Wilderness Area Meadow Basins. M.S. Range, Washington and Oregon. Arctic and Alpine Research Thesis, University of California, Davis. 3(3):215-224. Woodward, A., E.G. Schreiner, and D.G. Silsbee. 1995. Climate, Helms, J.A. 1987. Invasion of Pinus contorta var. murrayana geography, and tree establishment in subalpine meadows of (Pinaceae) into mountain meadows at Yosemite National the Olympic Mountains, Washington, U.S.A. Arctic and Park, California. Madrono 34(2):91-97. Alpine Research 27(3)217-225. Miller, E.A., and C.B. Halpern. 1998. Effects of environment

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 83

DISTURBANCE AND CHANGE

Table 1. Characteristics of six meadow sites analyzed from the Trinity Alps and Marble Mountains.

Meadow Site Meadow Lifeform Landform Elevation (meters) Location (UTM, NAD83)

Big Ridge Bunchgrass Ð Sedge Ridgecrest 2019-2124 481270, 4610835

Cuddihy Lakes Sedge Ð Thin Cirque Basin 1994-2092 473005, 4600252 Bentgrass

Meteor Lake Sedge Ð Thin Cirque Moraine 1739-1780 471093, 4597874 Bentgrass Morris Meadow Tall Grassland Glacial Valley 1338 503935, 4535707 Bottom

Paradise Lake Thin Bentgrass Cirque Moraine 1848 482577, 4606787

Sky High Lakes Thin Bentgrass Cirque Moraine 1751-1793 485094, 4600175

Table 2. Percent meadow loss to tree establishment for six meadow sites in the Trinity Alps and Marble Mountains, 1944 to 1999.

Meadow Site Area 1944 Tree Area Meadow Site 1999 Tree Increase Meadow Loss (ha) (sq.m.) Tree Area (sq.m.)

Meteor Lake 1.4 92 398 433% 2.1%

Sky High Lakes 38.3 436 13,582 3,115% 3.4%

Paradise Lake 4.5 690 2,212 321% 3.4%

Cuddihy Lakes 3.1 625 1,936 310% 4.2%

Morris Meadow 18.3 3,812 16,791 440% 7.1%

Big Ridge 20.7 2,210 29,990 1,357% 13.4%

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 84 DISTURBANCE AND CHANGE

Table 3. Abrupt allogenic changes in the environment have initiated widespread meadow tree encroachment in the Pacific Northwest.

Changes in Environment Location Reference

Warmer (north slopes), Wetter (south slopes), Livestock Oregon Cascades Miller and Halpern (1998) Warmer and Drier (westside), Cooler and Wetter (eastside) Mount Rainer, WA Rochefort and Peterson (1996 Drier (westside), Wetter (eastside) Olympic Range, WA Woodward and others (1995)

Warmer and Drier Oregon and Washington Cascades Franklin and others (1971)

Drier Yosemite NP, CA Helms (1987)

Warmer, Livestock, Fire Exclusion Lassen Volcanic NP, CA Taylor (1995; 1990) Greater snow-free period and/or lower snow depth Coastal Mountains, BC Brink (1959) Precipitation, Temperature, Livestock, Fire Esclusion Oregon Cascades Vale (1981)

Figure 1. Paradise Lake’s meadow has lost abut 3.4% of its area to invading Shasta red fir since 1944. A popular destination for reacreationists, new campsites have been created beneath the expanding canopies.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 85 DISTURBANCE AND CHANGE

VASCULAR PLANT SPECIES OF BABYFOOT LAKE BOTANICAL AREA BEFORE THE BISCUIT FIRE, WITH COMMENTS ON POST-FIRE FLORA

Linda Ann Vorobik University Herbarium, 1001 Valley Life Sciences Building #2465, University of California, Berkeley, CA 94720-2465 Email: [email protected]

Abstract

Presented here is a pre-Biscuit fire species list for the Babyfoot Lake Botanical Area (ca. 4,000 ft /1,220 m in elevation), along with comments on living plants within the botanical area after the fire. This botanical area is located on the eastern edge of the Kalmiopsis Wilderness Area, Josephine County, Oregon, Section 31 T38S, R9W (42º 13’ 15” N Lat. 123º 48’ 21” W Long.). Species were tallied along or in the vicinity of the Babyfoot Lake Trail (No. 1124A) from the parking area to the lake. The list was augmented from an unpublished Siskiyou National Forest list each June from 1994 through 2003. Vouchers were not collected within the botanical area, but the author vouchered the majority of taxa on the list from neighboring areas. Before the Biscuit Fire, the botanical area was rich in conifer species (12 species total), including Klamath endemic Picea breweriana, and woody shrubs (22 species total), including Klamath endemic Quercus saddleriana. Vegetation before the 2002 Biscuit fire ranged from open rocky outcrops with taxa such as Arctostaphylos nevadensis, Lewisia cotyledon, Cheilanthes gracillima, and Pellaea brachyptera, to trees of mesic forest, such as Pseudotsuga menziesii, Picea breweriana, Cupressus lawsoniana, Pinus lambertiana, with understory taxa in the most mesic regions including Rhododendron macrophyllum, Clintonia uniflora, and Thelepteris nevadensis. Living plants observed after the fire included Xerophyllum tenax, several crownsprouting shrubs, including Lithocarpus densiflorus, most of the taxa of the rock-outcrop and lake outlet vicinity, and a few scattered conifers including some individuals of Picea breweriana.

Introduction stand of Brewer’s spruce (Picea breweriana) in the Siskiyou Mountains. The area is primarily forested (Figure 2), with dominant trees including Douglas-fir (Pseudotsuga menziesii), abyfoot Lake, at an elevation of ca. 4,000 ft (1,220 m), is Shasta red fir (Abies magnifica ssp. shastensis), sugar pine (Pinus located in southwestern Oregon in the heart of the lambertiana), Port Orford cedar (Cupressus lawsoniana), and Siskiyou Mountains (part of the Klamath Range) at the BB incense cedar (Calocedrus decurrens). The underlying rock types eastern edge of the Kalmiopsis Wilderness Area. I have include diorite, greenstone, layered tuffs, and some ultramafic augmented an existing Siskiyou National Forest unpublished rocks including serpentine. species list for the botanical area and the approach to it since 1994, The approach to Babyfoot Lake Botanical Area is west up while leading field trips for the Siskiyou Field Institute. Because the Eight Dollar Mountain Road from Highway 199 (south of the Biscuit Fire (July through November 2002; Selma and north of Cave Junction), up road FS 4201 for about 15 www.biscuitfire.com) burned this area, publishing this revised list, miles, left onto spur road 140, and south for about one half mile to although unvouchered from plants within the Botanical Area, the botanical area parking lot. The Eight Dollar Mountain Road seemed particularly valuable. from Highway 199 to the trailhead traverses typical Siskiyou vegetation, beginning with Oregon white oak (Quercus garryana) The Area woodlands typical of non-serpentine soils at lower elevations of ca. 1,300 ft (400 m). Next the trees become widely scattered, and serpentine chaparral alternates with grassland plant associations The Babyfoot Lake Botanical Area was created in 1963 to (Jimerson et al. 1995), with scattered trees from species such as protect Brewer’s spruce (Picea breweriana) and other endemic or Jeffrey pine (Pinus jeffreyi), western white pine (P. monticola), sensitive plants of the region, including Siskiyou fritillary knobcone pine (P. attenuata), incense cedar (Calocedrus (Fritillaria glauca), Purdy’s lewisia (Lewisia cotyledon var. decurrens), and Douglas-fir (Pseudotsuga menziesii). The drive purdyi; included within var. cotyledon in the Oregon Flora up to the trailhead crosses into and out of serpentine, respectively checklist), opposite-leaved lewisia (L. oppositifolia), and Tracy’s indicated by open mixed conifer vegetation and closed canopy desert parsley (Lomatium tracyi). Located in the center of Section forest. Many of the same species occur in both communities, but 31 T38S, R9W (42º 13’ 15” N Lat. 123º 48’ 21” W Long.) (Figure Douglas-fir is much more dominant in the closed forest, and is 1), the botanical area covers 352 acres, and includes the largest joined by other species not usually found on serpentine, such as

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 86 DISTURBANCE AND CHANGE big-leaf maple (Acer macrophyllum). As one gains elevation on shallon), red huckleberry (Vaccinium parviflorum), grouse the approach to the Babyfoot Lake trailhead, sugar pine (Pinus huckleberry (V. scoparium), and evergreen huckleberry (V. lambertiana) becomes more common, and Shasta red fir (Abies myrtillus). Labrador tea (Ledum glandulosum) occurred along the magnifica ssp. shastensis) becomes part of the forest canopy. lake margins, and also occurs along stream and lake margins at lower elevations. Methods The trail to Babyfoot Lake heads north to the lake, and before the 2002 fire, was in a cool, dark forest dominated by Species were tallied along the Babyfoot Lake Trail (FS Shasta red fir, Douglas-fir, and Brewer’s spruce, where the No.1124A), beginning from the trailhead and traveling south and understory included numerous non-photosynthetic plants of west through the forest, off the trail down a west-facing rock Ericaceae and Orchidaceae. Near the botanical area one could outcrop located due north of the outlet stream, and along the outlet find bright red snow plant (Sarcodes sanguineum), or red-and- stream to the lake (Figure 1). Total walking transect distance white striped Allotropa virgata, aptly named candy cane or sugar equals about 2 miles (3.2 km) round trip to the lake and back. stick. Within the botanical area, other non-photosynthetic Elevation of area sampled ranges from about 3,800-4,000 ft Ericaceae included pine drops (Pterospora andromedea), fringed (1,160-1,220 m). The list published here was augmented from an pinesap (Pleuricospora fimbriolata), and pinesap (Monotropa existing Siskiyou National Forest unpublished document as hypopitys). Photosynthetic herbs and subshrubs of the Ericaceae compiled by contributing botanists, primarily Veva Stansell, Anita included both species of prince’s pine (Chimaphila umbellata and Seda, Rachel Winters, Chris Reyes, and Pat Munoz. My list is C. menziesii), slender salal (Gaultheria ovatifolia), one-sided neither definitive nor exhaustive, as it was created during one to wintergreen (Orthilia secunda), and three species of true two visits per year, primarily in the month of June, from 1994 wintergreen, the genus Pyrola (P. asarifolia, P. picta, and P. through 2003. Determinations are, for the most part, based on minor). field identifications and are not vouchered, although Vorobik has Non-photosynthetic orchids within the botanical area forests collected the majority of taxa on this list from neighboring areas included spotted and western coralroot (Corallorhiza maculata (Vorobik 1985). and C. mertensiana) and the completely white phantom orchid (Cephalanthera austiniae). Photosynthetic orchids included fairy slipper (Calypso bulbosa) with its delicate pink flowers, Pre-fire Vegetation rattlesnake plantain (Goodyera oblongifolia) with its mottled leaves, and western twayblade (Listera caurina). The trail Vascular plant taxa observed before the 2002 Biscuit fire are descends to a perennial stream that was, before the fire, shaded by listed in Table 1; a total of 143 taxa were tallied. As this list large Port Orford cedar (Chamaecyparis lawsoniana) and lined represents a sample of species along a walking transect described with Sierra marsh-fern (Thelypteris nevadensis), with species such in the Methods section it does not represent all taxa present within as western rhododendron (Rhododendron macrophyllum) and the botanical area. A brief discussion of the more common or queens cup (Clintonia uniflora) nearby. Plants of Thelypteris interesting taxa follows. were the first of many ferns encountered in the Botanical area, followed by licorice fern (Polypodium glycyrrhiza), sword fern The Montane Forest (Polystichum imbricans), and fragile fern (Cystopteris fragilis) all As one ascends by road to the trailhead of Babyfoot Lake, on a large (13 ft/ 4 m) rock within the forest. lowland tree species are replaced by those more typical of montane forest: red fir (Abies magnifica ssp. shastensis), sugar West-facing Rock Outcrop pine (Pinus lambertiana), and the Klamath Mountain endemic, After the trail leaves a forest dominated by Brewer’s spruce, Brewer’s spruce (Picea breweriana). Along with many other rare, it opens onto a rocky outcrop (Figure 2) projecting west towards endemic species in the Klamath-Siskiyou region, Brewer’s spruce the Kalmiopsis Wilderness Area, including Preston Peak. Species populations are thought to be a relict of a once more widespread present before the Biscuit Fire included parsley fern distribution, in this case, in the Miocene (Waring et al. 1975). (Cryptogramma crispa), lace fern (Cheilanthes gracillima), Despite the closed canopy of the montane forest canopy, the shrub birdfoot fern (Pellaea brachyptera), and Indian dream (Aspidotis understory is also impressively diverse at this elevation (ca. 4,000 densa). Along with the ferns, this rocky outcrop included the rare ft/ 1,220 m), with oak and heath families well represented, but Lewisia cotyledon var. purdyi, L. oppositifolia, and Heckner’s with species differing from those found at lower elevations. Pre- stonecrop, Sedum laxum var. heckneri. Other more common fire oak family species within the botanical area included another species included three species of Penstemon (P. davidsonii, P. Klamath endemic, Saddler or deer oak (Quercus sadleriana), rupicola, and P. parvulus), larkspur (Delphinium sp.), and the huckleberry oak (Q. vaccinifolia), and tanoak (Lithocarpus parasitic broomrape (Orobanche uniflora) with its host plant, densiflorus var. echinoides). Heath family shrubs included broadleaved stonecrop (Sedum spathulifolium). greenleaf manzanita (Arctostaphylos patula), rhododendron or South and down about 100 ft (30 m) is the outlet stream rose-bay (Rhododendron macrophyllum), salal (Gaultheria (Figure 3), with plants adapted to more mesic conditions, such as

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 87 DISTURBANCE AND CHANGE

Menzie’s saxifrage (Saxifraga menziesii), wandering daisy brought me to a greater understanding of flora of the Siskiyou (Erigeron peregrinus), and bronze bells (Stenanthium occidentale). Mountains. I thank Veva Stansell, Charlene Simpson, Anita Seda, Before the Biscuit Fire, between the outlet stream on the Maria Ulloa, Wilbur Bluhm, Linda Mazzu, former students, and rock outcrop and the lake, there was a mesic forest with mature others for delightful and educational days in the field, Cindy sugar pines (Pinus lambertiana) with trunks 2-3 ft (0.6-0.9 m) in Roché for editorial comments, the curators of the Oregon herbaria diameter, and understory taxa which included salal, orchids, and of ORE, SOC, OSC, and WILLU (Dave Wagner, Frank Lang, and non-photosynthetic Ericaceae. Surrounding the lake were trees Ken Chambers) for use of the collections, and the UC Jepson such as Cupressus lawsoniana, Pseudotsuga menziesii, and Picea Herbarium, USFS, BLM, NPSO, and the Siskiyou National Forest breweriana, with shrubs such as Salix sp., Rhododendron and Siskiyou Field Institute for financial support of my field work occidentale, and Ledum glandulosum. in southwestern Oregon. Post-fire Comments Literature Cited Although the effects of the Biscuit Fire can be seen along the drive to the Babyfoot Lake trailhead, with completely burned Atzet, T., D.E. White, L.A. McCrimmon, P.A. Martinez, P.R. Fong, areas where all trees were killed alternating with patches that did and V.D. Randall. 1996. Field Guide to the Forested Plant not burn at all, the impact to the botanical area was severe (Figure Associations of Southwestern Oregon. USDA Forest Service 4). Most trees were still standing, although dead. The forest in Technical Paper R6-NR-ECOL-TP-17-96. October 2002 consisted of strong black trunks of trees in striking Franklin, J.F., and C.T. Dyrness. 1973. Natural vegetation of contrast to a golden carpet formed by needles fallen from scorched Oregon and Washington. USDA Forest Service Technical trees. Particularly striking were burnt Brewer’s spruce, whose Report PNW-8. U.S. Government Printing Office. black outer bark contrasted sharply with lighter yellow-orange Washington D.C. underbark exposed by exfoliating puzzle-shaped flakes (Figure 5). Hickman, J. (editor). 1993. The Jepson Manual, higher plants of Even though it had been a few months since the fire swept California. University of California Press, Berkeley. through, plants were beginning to grow again. Shrubby deer oak, Jimerson, T.M., L.D. Hoover, E.A. McGee, G. DeNitto, and R.M. huckleberry oak, tanoak, and wood rose (Rosa gymnocarpa) were Creasy. 1995. A Field Guide to serpentine plant sprouting from crowns (Figure 6). New shoots of beargrass were associations and sensitive plants in northwestern California. distributed throughout the forest floor. Although most of the trees USDA Forest Service Technical Paper R5- ECOL-TP-006. were dead or dying and the duff layer had burned, exposing Kruckeberg, A.R. 1984. California serpentines: flora, vegetation, mineral soil, still a few trees around the lake survived (Figure 7), geology, soils, and management problems. University of including Pseudotsuga menziesii, Calocedrus decurrens, and Salix California Press, Berkeley. sp. Also, sporadic patches of duff with their associated understory Peck, M.E. 1961. A Manual of the Higher Plants of Oregon. plants remained minimally damaged by the fire in certain areas Second edition. Binfords & Mort. Portland, Oregon. within the forest. Vorobik, L.A. 1985. Hybridization and reproductive isolation The rock outcrops within the forest had clearly sustained between sympatric Arabis (Cruciferae) in southwestern very high temperatures, as they were scoured of all organic Oregon. PhD. Dissertation. University of Oregon, Eugene. material (Figure 8). However, the large rock outcrop (Figure 9) Waring, R.H., W.H. Emmingham, and S.W. Running. 1975. and the outlet stream burned only lightly, and most, if not all, of Environmental limits of an endemic spruce, Picea the plants located in these sites will probably survive. Another breweriana. Canadian Journal of Botany 53:1599-1613. surprise was seeing new growth of Thelypteris nevadensis and White, D.H. 1971. Vegetation-soil chemistry correlations in other species along the stream in the deep forest, in spite of the serpentine ecosystems. PhD. Dissertation. University of fact that all the surrounding old-growth trees were killed by the Oregon, Eugene. fire. Whittaker, R.H. 1960. Vegetation of the Siskiyou Mts., Oregon Most recently the standing dead trees adjacent to the and California. Ecological Monographs. 30:279-338. botanical area have been logged (Figure 10), including the area where Allotropa virgata and Sarcodes sangunium occurred, the latter of which was seen post-fire. This general locale would Additional Resources provide excellent opportunities to compare succession in vegetation of logged with unlogged burn areas outside and within www.biscuitfire.com. Biscuit Fire Recovery website. the botanical area, respectively. www.siskiyou.org. Siskiyou Project website. www.fs.fed.us/r6/siskiyou/. Siskiyou National Forest website. Acknowledgements

I am grateful to those persons and organizations that have

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 88 DISTURBANCE AND CHANGE

Table 1. Pre-Biscuit Fire vascular plant list for Babyfoot Lake Botanic Area, Josephine County, Oregon. List compiled by Linda Ann Vorobik; * indicates Sensitive Species and ** indicates Watch/Review Species for Siskiyou National Forest. Order is alphabetical by scientific name under the following major groups: I. Ferns, II. Gymnosperms, III. Dicots, IV. Monocots. Families and most nomenclature follow that found in The Jepson Manual (Hickman 1993). Family Scientific Name Common Name

I. FERNS AND FERN ALLIES

Dryopteridaceae Cystopteris fragilis fragile fern Polystichum imbricans ssp. imbricans sword fern Polypodiaceae Polypodium glycyrrhiza licorice fern Pteridaceae Aspidotis densa Indian dream Cheilanthes gracillima lace fern Cryptogramma acrosticoides parsley fern Pellaea brachyptera Sierra cliff brake Selaginellaceae Selaginella spp. little clubmoss Thelypteridaceae Thelypteris nevadensis Sierran wood fern

II. GYMNOSPERMS

Cupressaceae Calocedrus decurrens incense cedar Cupressus lawsoniana Port Orford cedar Juniperus communis common juniper Pinaceae Abies concolor var. lowiana white fir A. magnifica var. shastensis Shasta red fir Picea breweriana Brewer’s spruce Pinus attenuata knobcone pine P. jeffreyi Jeffrey pine P. lambertiana sugar pine P. monticola western white pine Pseudotsuga menziesii Douglas-fir Taxaceae Taxus brevifolia western yew

III. DICOTS

Aceraceae Acer circirnatum vine maple A. glabrum Douglas maple

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 89

DISTURBANCE AND CHANGE

Apiaceae (Umbelliferae) Lomatium nudicaule pestle parsnip L. triternatum nineleaf biscuitroot *Lomatium tracyi Tracy’s desert parsley Sanicula graveolens Sierra sanicle Apocynaceae Apocynum androsaemifolium spreading dogbane Asteraceae Achillea millefolium yarrow Antennaria rosea rosy pussytoes A. sufftrutescens shrubby pussytoes Arnica cernua nodding arnica A. spathulata Klamath arnica Blepharipappus scaber eyelash flower Cacaliopsis nardosmia silvercrown **Erigeron cervinus Siskiyou daisy E. foliosus leafy fleabane E. perigrinus wandering fleabane Eriophyllum lanatum Oregon sunshine Hieracium albiflorum white-flowered hawkweed Luina hypoleuca silverback luina Berberidaceae Achlys triphylla ssp. triphylla vanillaleaf Berberis nervosa shining Oregon grape B. repens creeping Oregon grape Brassicaceae Arabis koehleri ssp. stipitata Koehler’s rockcress Cardamine spp. bittercress Streptanthus tortuosus jewel flower Caprifoliaceae Linnaea borealis twinflower Lonicera ciliosa orange honeysuckle Symphoricarpos albus or S. mollis snowberry Caryophyllaceae Moehringia macrophylla bigleaf sandleaf Silene campanulata ssp. glandulosa catchfly S. lemmonii Lemmon’s or darting campion Crassulaceae **Sedum laxum var. heckneri Heckner’s stonecrop S. spathulifolium broadleaved stonecrop Ericaceae Allotropa virgata candycane or sugar stick Arctostaphylos nevadensis pinemat manzanita A. patula greenleaf manzanita Chimophila menziesii little prince’s pine

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 90 DISTURBANCE AND CHANGE

C. umbellata greenleaf manzanita Gaultheria ovatifolia slender wintergreen G. shallon salal Monotropa hypopytis pinesap Orthilia secunda sidebells pyrola Pleuricospora finbriolata Sierra or finged pinesap Pterospora andromedea pinedrops Pyrola asarifolia alpine pyrola, bog wintergreen P. picta white-veined wintergreentergreen P. minor lesser wintergreen Rhododendron macrophyllum Pacific rhododendron Sarcodes sanguinea snow plant Vaccinium myrtillus dwarf bilberry V. parvifolium red huckleberry V. scoparium grouse whortleberry Fabaceae Lupinus sp. lupine Thermopsis rhombifolia golden pea Fagaceae Chrysolepis chrysophylla golden chinquapin Lithocarpus densiflorus var. echioides tanoak Quercus chrysolepis canyon live oak Q. garryana var. breweri Brewer’s oak Q. sadleriana deer oak Q. vaccinifolia huckleberry oak Garryaceae G. fremontii Fremont’s silk-tassel bush Grossulariaceae Ribes roezlii var. cruentum shinyleaf gooseberry Hydrophyllaceae Romanzoffia sp. mistmaiden Lauraceae Umbellularia californica Oregon myrtle, California bay Orobanchaceae Orobanche uniflora naked broomrape Papaveraceae Dicentra formosa var. formosa bleeding heart Philadelphaceae Whipplea modesta modesty, yerba de selva Polemoniaceae Phlox gracilis slender phlox Polygonaceae Eriogonum compositum arrowleaf buckwheat E. nudum barestem buckwheat E. ternatum ternate buckwheat Portulacaceae Calyptridium umbellatum pussypaws *Lewisia cotyledon ssp. purdyi Purdy’s lewisia *L. oppositifolia opposite-leaved lewisia Montia parvifolia

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 91 DISTURBANCE AND CHANGE

Primulaceae Dodecatheon jeffreyi tall mountain shooting star Trientalis latifolia western starflower Ranunculaceae Aquilegia formosa red columbine Delphinium nuttallianum Nuttall’s larkspur

Rosaceae Amelanchier alnifolia serviceberry Fragaria vesca wild strawberry Holodiscus discolor creambush oceanspray Potentilla glandulosa sticky cinquefoil Rosa gymnocarpa little wild rose R. ursinus Pacific blackberry or dewberry Sorbus sitchensis mountain ash Salicaceae Salix spp. willow Saxifragaceae Mitella pentandra alpine mitrewort Saxifraga ferruginea rusty saxifrage S. howellii Howell’s saxifrage S. mertensiana Merten’s saxifrage Scrophulariaceae Penstemon parvulus small azure beardtongue P. davidsonii Davidson’s penstemon P. deustus hotrock penstemon P. rupicola cliff penstemon Violaceae Viola cuneata wedgeleaf violet V. hallii Hall’s Violet V. sempervirens redwoods or evergreen violet

IV. ANGIOSPERMS: MONOCOTS

Cyperaceae Carex mendocinenesis Mendocino sedge Iridaceae Iris bracteata Siskiyou iris I. innominata golden iris Juncaceae Luzula comosa Pacific woodrush

Liliaceae Allium siskiyouense Siskiyou onion *Calochortus howellii Howell’s mariposa lily Clintonia uniflora queen’s cup Disporum hookeri fairy bells

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 92 DISTURBANCE AND CHANGE

*Fritillaria glauca Siskiyou fritillary Lilium sp. (aff. columbianum) Columbia lily Smilacina racemosa false solomon’s seal Stenanthium occidentale bronze bells Streptopus amplexicaulis var. americanus twisted stalk Trillium ovatum trillium, wakerobin Veratrum insolitum Siskiyou false hellebore Xerophyllum tenax beargrass Zigadenus micranthus small flowered deathcamas Orchidaceae Calypso bulbosa fairy slipper Cephalanthera austiniae phantom or snow orchid Corallorhiza maculata spotted coralroot C. mertensiana western coralroot Goodyera oblongifolia rattlesnake plantain Listera caurina northwest twayblade Poaceae Festuca sp. fescue

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 93

DISTURBANCE AND CHANGE

Figure 1. Map of Babyfoot Lake Botanical Area. Section lines indicate square miles.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 94 DISTURBANCE AND CHANGE

Figure 2. View to Babyfoot Lake basin from ridge to the west. Lake is located in north-facing cirque near center of photograph; rock outcrop discussed in text is seen in the left center portion of the image. Photograph by L.A. Vorobik. .

Figure 3. View south down rock outcrop towards point where outlet stream from Babyfoot Lake emerges from the forest. Image taken after the 2002 fire. Photograph by L.A. Vorobik.

Figure 4. View south from Fiddler Mountain after the 2002 Biscuit Fire. Babyfoot Lake basin is dark area in upper left portion of image. Photograph by L.A. Vorobik.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 95 DISTURBANCE AND CHANGE

Figure 5. (Left) Burnt trunks of Brewer’s spruce (Picea breweriana—on the left) and Douglas fir (Pseudostuga menziesii—on the right). Photograph by L.A. Vorobik. .

Figure 6. (Below) Crown-sprouting Saddler oak (Quercus saddleriana) in October 2002, after the Biscuit fire. Photograph by L.A. Vorobik.

Figure 7. (Right) Post-Biscuit Fire view of west (outlet) end of Babyfoot Lake. A few living trees can be seen on right side of image. Photograph by L.A. Vorobik.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 96 DISTURBANCE AND CHANGE

Figure 8. Small (approximately 12 m wide) outcrop in north-facing Brewer’s spruce (Picea breweriana) forest, taken after the Biscuit Fire. Intensive heat burned all organic matter at this site. Photograph by L.A. Vorobik.

Figure 9. Maria Ulloa-Cruz examining living vegetation on large west-facing rock outcrop in October 2002 after the Biscuit Fire. Although surrounding trees and some outcrop vegetation was burned in this area, many plants were living, indicating the fire passed over relatively quickly. Photograph by L.A. Vorobik.

Figure 10. Logging remains near the junction of the Babyfoot Lake and Fiddler Mountain roads. Photograph by R. Skar.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 97

VEGETATION ECOLOGY

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY

VEGETATION ECOLOGY

THE OREGON FLORA PROJECT

Linda K. Hardison and Scott Sundberg Dept. of Botany and Plant Pathology, Oregon State University, 2082 Cordley Hall, Corvallis, OR 97331-2902 Emails: [email protected]; [email protected]

Abstract

The Oregon Flora Project (OFP) is an extensive effort to produce a new set of references for identifying Oregon plants and characterizing their distributions, morphology, and appearance. The major facets of the Oregon Flora Project are: 1) a Checklist, which provides a consistent nomenclature for all species, subspecies, and varieties of plants in the state; 2) a Flora, or plant identification manual; 3) an online Photo Gallery with multiple images of each taxon; and 4) an online Oregon Plant Atlas, which allows users to create customized distribution maps from data contained in the 300,000+ -record Atlas database. There is a critical need for up-to-date information about Oregon’s diverse plants. The Oregon Flora Project OFP provides resources for a broad spectrum of users, from novice plant enthusiasts to professionals seeking data for restoration, invasive plant, or rare plant projects. OFP members frequently receive reports of unusual plant sightings and are on most occasions able to confirm their significance and relay the information to appropriate groups. The online version of the Flora is designed to take full advantage of the interrelatedness of all facets of the project; this organization, and the ability to continually improve and update the information presented, make it a flexible and timeless source of knowledge. In all applications, the OFP will provide the ability opportunity to present information to a depth that is appropriate for the user’s needs, thereby establishing itself as a useable, and useful, reference of Oregon’s plant life. Introduction each splitting and lumping to different degrees. For example, the Checklist recognizes nine taxa for the four species Peck listed in cological studies, restoration and conservation work, and the blue-eyed grasses (genus Sisyrinchium). On the other hand, investigations of insect- or animal-plant interactions all Peck’s four taxa of yarrow (genus Achillea) were reduced to one EErequire a fundamental knowledge of a region’s plant life. in the Checklist. Traditionally, a regional flora is the prime reference used for The difficulty of assigning correct names to Oregon plants is assigning names to plants and facilitating their identification. The compounded by the fact that none of the more modern Oregon Flora Project (OFP) is addressing the need for a northwestern floras covers the entire state. Three of these—The comprehensive, up-to-date reference on the vascular plants Intermountain Flora (Cronquist et al. 1972-1994), the Jepson growing without cultivation in Oregon by producing a checklist, a Manual (Hickman 1993), and the Flora of the Pacific Northwest flora, a photo gallery, and an atlas of plant distributions. These (Hitchcock and Cronquist 1973)—are frequently used to identify data will be available to the public over the Internet at the OFP plants from this state. Unfortunately, these references do not website, www.oregonflora.org. include all of Oregon’s taxa. Additionally, nomenclature is There is a critical need for current fundamental information inconsistent among these and other floras. This can lead to about Oregon’s plants. The most recent state flora was published confusion in assigning a name to a specimen, and difficulties in in 1961 (Peck 1961) as an update to a 1941 edition. Since that conducting comparative studies over geographic areas that extend time, biosystematic research and botanical exploration has added beyond a single flora’s range. enormously to our knowledge of Oregon plant diversity. The Oregon Flora Project was initiated in 1994 when Scott Currently, the Oregon Plant Vascular Checklist (Checklist) Sundberg brought together interested people with the intent of catalogs 4507 species, subspecies, and varieties (taxa) of vascular developing a checklist and, ultimately, a new flora. As public plants in Oregon. In contrast, Peck’s flora included 3974 taxa. awareness and interest in the effort grew, the OFP group The difference of 533 taxa between the two sources is due recognized that the available plant data included a wealth of primarily to discoveries of taxa that were previously not known to information that would be immensely useful to botanists, occur in Oregon. Over 350 exotic taxa (primarily escaped ecologists, land use planners, and conservationists, as well as ornamentals and introduced weeds) and over 200 native taxa are nonprofessional plant enthusiasts. The Oregon Flora Project listed because of extensions of the taxon’s range into Oregon or eventually grew to encompass four major facets. These are: the because they have been recently described. In addition, nearly Vascular Plant Checklist, a flora of Oregon, the Photo Gallery, and 40% of the names in Peck’s flora are accepted under other names the Plant Atlas. in the Checklist. Much of this is due to discrepancies between the Most features of the Oregon Flora Project will be available taxonomic frameworks of the Checklist and Peck’s flora, with over the Internet. Presenting the data in digital form offers several

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 98 VEGETATION ECOLOGY notable opportunities. First is the ability to routinely edit and included in the three tables in the checklist database. update the material. This gives a critical advantage to the digital A complete rough draft of the Oregon Checklist is currently Atlas Project over any paper-published atlas version, allowing undergoing revision. Contributors to the Checklist revise new database records to be incorporated and erroneous data to be printouts from this draft after examining specimens and reviewing evaluated. New species lists and specimen data can be floras and other pertinent literature. Draft treatments are reviewed incorporated into the online Atlas shortly after it is received by the and edited by Sundberg and others for inclusion into the final project. Secondly, the accessibility of the Flora Project’s project’s version of the Vascular Plant Checklist. Treatments for 64% of efforts through the Internet makes it available to the general the 4507 taxa have either been completed or are in an advanced public. This allows us to share botanical knowledge with many draft form. The Asteraceae portion of the Checklist has been who might otherwise not receive it and, in the process, foster an published as a stand-alone booklet (Chambers and Sundberg appreciation for Oregon’s natural resources. Lastly, software 2000). Upon completion, the entire Checklist will be available connecting the different components of the OFP provides a online in a searchable format. The draft Checklist has been used powerful tool to access a broad spectrum of plant data that to help guide nomenclatural decisions for a number of projects, previously were unavailable or difficult to compile. many of which have produced publications [e.g., Simpson et al. In this paper we will describe each facet of the Oregon 2002 (whose records represent 39% of Oregon taxa), Guard 1995; Flora Project, explain how the data are being prepared, and Mansfield 2000; Gilkey and Dennis 2001; Gilkey and Packard illustrate how each resource can be applied to a variety of needs. 2001, Guard 1995, Mansfield 2000]. In addition to the Oregon Checklist, the Flora project has compiled taxon lists for each Oregon Vascular Plant Checklist county, based on data in the Atlas database.

Compiling a checklist of the plants of Oregon marked the The Oregon Flora—On Paper and Online beginnings of the Oregon Flora Project. The list is used to account for all taxa growing without cultivation in the state. It also The pinnacle of the Oregon Flora Project will be a new provides a nomenclatural framework that can be used to Flora, or identification manual. Building on the nomenclatural “translate” taxon names found in other taxonomic references. A framework established in the Checklist, it will provide technical key figure in the development of the checklist was the late Dr. descriptions of each species and dichotomous keys to permit Karl Urban, an instructor at Blue Mountain Community College identification of all species, subspecies, and varieties. Line and Forest Botanist at Umatilla National Forest. Between 1982 drawings of plants and distinguishing features will be included for and 1991 Urban compiled into a database a list of the plants of the most, if not all, taxa. Botanists nationwide will be solicited for the Pacific Northwest. A year-long sabbatical in 1988-1989 was species preparing taxon descriptions. The Flora project is spent in the Oregon State University Herbarium updating and committed to providing useable and accurate keys; this will be refining the synonymy of the taxa in the Forest Service’s Pacific met through careful testing of keys on plant material, and through Northwest Region. Urban was unable to further pursue revisions feedback from both experienced and novice botanists. The format of his checklist and was delighted when, in 1994, the nascent of the paper version of the new flora Flora will resemble that of Oregon Flora Project approached him for permission to use his the Jepson Manual (Hickman 1993). Entries will also include database and asked him to join the group. He later correctly notes on each plant’s status as native or introduced exotic, its stated that, “[These] efforts…were to serve as one of several rarity, and potential horticultural or ethnobotanical uses. catalysts that would help the Oregon Checklist and Flora projects The digitized version of the Flora will be available on the get underway” (Urban 1995). OFP website. With this type of key users can start keying with The current database structure for the Oregon Vascular Plant any of a number of characters that are readily identifiable. As Checklist serves a number of purposes not foreseen in Urban’s these are entered, the keying program eliminates unlikely taxa and seminal database. Table 1 summarizes data included in the three presents the user with an increasingly narrowed list of possible tables in the checklist database. identifications. Comparisons of the unknown specimen to A checklist entry provides a great deal of information: the photographs in the Photo Gallery, illustrations, morphological accepted scientific name with authorities; common names; descriptions, and distribution maps allow the user to confidently indication of native or introduced exotic status; and pertinent facts assign a definitive name, often without the heavy reliance on which allow the taxon to be distinguished from others. A valuable technical terminology commonly found in dichotomous keys. feature is the inclusion of synonyms. Diverse floras may ascribe Current technology allows the Oregon Flora Project to offer different names to a plant and, as a consequence, multiple a wealth of information on Oregon’s plants. The Flora project staff scientific names can apply to the same taxon. The Checklist is collaborating with the Northwest Alliance for Computational accounts for all such names that occur in ten floras or species lists Science & Engineering (NACSE; at Oregon State University) to that cover all or parts of Oregon, either as accepted names, or as develop software to that will link together all components of the synonyms. It thereby serves as a valuable cross-reference to these Flora. Each facet of the Flora OFP—the Checklist, the Flora, the sources, and prepares users for the nomenclature that will be Photo Gallery, and the Atlas—exists in digital form, each with applied in the new Flora of Oregon. Table 1 summarizes data information in a relational database (Figure 1). These facets will

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 99

VEGETATION ECOLOGY be available free of charge on the OFP website information from all products of the Oregon Flora project (www.oregonflora.org). Thus, synonyms, identification keys, increases tremendously their value. photographs, and distribution maps—all available data on Oregon plants—will be readily accessible to the public. Oregon Flora Photo Gallery NACSE is also developing software for the Personal Digital Field Guide (DFG) aspect of the online Flora. Users will be One of the most striking facets of the Oregon Flora project allowed to download components of the digital Flora, Atlas, and is the Oregon Flora Photo Gallery. This is an online collection of Photo Gallery, to a personal computer or a personal digital photographs of Oregon plants. While initial efforts are focused on assistant (PDA) and produce a DFG tailored to their needs. This posting at least one image of as many taxa as possible, the goal of powerful tool will offer a “slice of the pie,” a customized the Photo Gallery is to provide multiple shots of each taxon. selection of the data the Flora Project project contains. Following These would include photographs that illustrate a plant’s preferred are some examples of how the DFG can be applied: habitat, its general growth form, close-ups of features that aid in its identification (such as flowers, hairs, seeds), and an image of ¥ A botanist will be conducting rare plant surveys during an an herbarium specimen. As with the other facets of the Flora extended backcountry field trip in alpine areas. She will be Project, the information in the Photo Gallery will be incorporated backpacking, and needs to be able to confirm her into a searchable database. identifications without collecting specimens. Using the The challenge of compiling, obtaining permissions for use DFG, she indicates that she wants data on rare plant taxa in the Photo Gallery, and digitizing thousands of images, and from the vicinity of her survey area, and selects distribution along with coupling them with their relevant data, is significant. maps, photographs, species descriptions, morphology data, Original slides, prints, or digital photographs or a digital image and multiple entry key capabilities software for her target areis typically loaned or donated to the Flora project or OSU species. She downloads these to her PDA, thereby Herbarium for use in to the Photo Gallery. Data supplied with the providing herself with resources previously available only image is entered into a database table, along with a unique code. in multiple references and photographic collections. A digital copy of each slide is made, if necessary, and the file name is linked to its assigned code in a database. The physical ¥ A middle school teacher develops a science project for his slides that have been donated are labeled with the code, and class that includes identifying the plants along a nearby archived in transparent slide holder sheets. Archival copies of all creek. Rather than handing his 12-year-olds a thousand- images are retained on CD-ROMs, and those used in the public page flora with black and white drawings and technical version of the Photo Gallery are uploaded to the website server. terminology, he gathers the data from Willamette Valley Maintaining a library of photographic images gives the Photo riparian entries in the DFG, and presents his class with a Gallery editors flexibility: as multiple photos of a taxon are manageable subset of species photographs, illustrations, and obtained, those on display in the Gallery can be rotated or added keys. They use the DFG to identify the several-dozen to, giving users a varied resource over time. plants they will likely encounter. More than any other component of the Oregon Flora Project, the Photo Gallery appeals to the general public. The ¥ A habitat restoration biologist needs to know which taxa are accessibility of photographs, and the ability of all persons to relate most appropriate to plant in a site of a forest fire in to them, will attract users who seek aesthetic, artistic, and Josephine County. This requires knowing which plant taxa gardening inspiration, rather than taxonomic facts. Their use of are native to the vicinity, the habitats they typically occupy, the Photo Gallery and exposure to the other facets of the Oregon and their growth characteristics. She queries the online Flora Project can help foster a greater awareness and appreciation Flora for native plants that have been found in upland of the plants of our state. riparian habitats in the same ecoregion and within 30 miles Just as the online product connects with the public, so does of the site. She selects native perennial herbs and shrubs the opportunity to develop the Photo Gallery. Most of the images that are not rare and fine-tunes the list by removing a in the Photo Gallery have been contributed by individuals, both number of species. For each taxon she downloads with and without botanical training. The ability to provide a photographs, habitat information, and descriptions. She also scientifically informative and aesthetically pleasing photograph is downloads maps showing distributions of each of the taxa a gift that many non-botanists have given to the Oregon Flora in the vicinity of the site to help guide a search for Project, and one that can be appreciated by scientists and particular species. With these data in a laptop computer she generalists alike. visits the area and finds most of the common species she had selected. Oregon Plant Atlas

These examples illustrates how DFG’s will be provide an The Oregon Plant Atlas Project was initiated in April 1995, invaluable addition to the references currently available to with the goal of producing digital maps of Oregon plant localities. botanists, educators, and land use planners. Linking together The Atlas allows users to map distributions of any of the taxa in

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 100 VEGETATION ECOLOGY the Oregon checklist on a background map (base map) of their problematic records are flagged and evaluated further. Each map choice. Four base maps of Oregon are currently available; these generated has a link that lets users notify the Atlas project staff of illustrate landforms, rainfall, ecoregions, and county lines. Up to records they feel need examination. This allows the quality of the six taxa may be selected for display on one map. Taxa can be Atlas database to be constantly improved by individuals who often selected using common or scientific names. The user can also have an intimate knowledge of specific regions and their flora. restrict mapping records for a particular time period, collector, or The Atlas mapping program is being released to the public only vouchered specimens. Once options have been chosen, a dot in Spring 2004. It will be launched with approximately over map is generated (Figure 2). The user can then move a cursor and 300,000 records in the Atlas database, representing 98% of click on a dot to view additional information about the plant Oregon’s 4507 taxa. The Atlas is online at: www.oregonflora.org/ occurrence record. oregonplantatlas.html. Below are the predominant sources of Maps are generated using a Java applet developed by plant locality data in the Atlas database: Clayton Gautier, who is an Atlas project leader. Data for the mapping program are drawn from the Atlas database. The ¥ Herbarium specimen records—The OSU Herbarium, primary data gathered for use in the Atlas database is entered into together with the Oregon Flora Project, received federal tables using the Microsoft Access program. This database is funding to database and georeference the approximately periodically converted to a Sybase format for use in the online 156,000 specimens of Oregon plants over the next three Atlas. Plant occurrence records include plant name, years (2003-2005). Thousands of additional specimens are collector/observer, and date, and information (as available) such as housed in other institutions and private herbaria throughout associated species, elevation, and presence of flowers or fruits. the state. . The OFP currently has collaborations established There are two basic kinds of data: vouchered, which are taken to share data with herbaria in educational institutions and in from herbarium specimen labels; and unvouchered, which are Federal agency offices. Herbarium specimen label derived from quality species lists or observations and lack a information is the most valuable source of data because the specimen. records are readily verifiable; therefore special efforts are In the hundreds of thousands of records handled by the underway to include the data from these collections in the Atlas project, the precision used to describe localities spans a wide Atlas database. gamut, from highly precise latitude-longitude coordinates provided by Global Positioning Systems (GPS) units, to vague ¥ Lists prepared by individuals and non-governmental references such as “eastern Oregon.” The mapping program organizations—The Native Plant Society of Oregon requires decimal latitude and longitude coordinates to place a dot (NPSO) has provided hundreds of species lists compiled on a map, so researching locality information is critical. The Atlas during field trips. Lists prepared for research projects, project is investing extensive effort into georeferencing, or county and regional checklist activities (e.g., Hopkins et al. assigning coordinates, to the data. To facilitate this OFP staff 1993; Simpson et al. 2002), and other targeted efforts have developed a database that links locality descriptions and been contributed. Numerous class projects, masters theses, Township-Range-Section values with a latitude-longitude value publications, and unpublished data sets from the Ecology coordinates readily useable by the mapping program. In addition, program in the Botany and Plant Pathology Department at a scale of relative certainty, or “fuzziness” is assigned to each OSU contribute approximately 1,200 species lists and record. This is modeled after an index developed by the Ohio species composition data from diverse habitats. Individuals Lepidopterists (1992), and distinguishes precise locality data (such have also adopted “blocks” or designated areas of the state as GPS values) from generalized information (e.g., names of as focal points for gathering data. landforms or the nearest town). The “fuzz” value ascribes to the latter class a degree of precision for each record, such as “plus or ¥ Lists prepared for agency projects on government-owned minus 0.5 mile.” land—With approximately 56 percent of the land in Oregon A challenge to the Atlas project is to insure that the data in under federal or state ownership (Loy et al. 2001), a wealth the Atlas database are of the highest quality that is practically of information has been gathered by governmental workers possible. This is met by evaluating data to be included in the in the process of conducting environmental surveys, rare, database for their probability of accurate identification, and by threatened, and endangered plant surveys, habitat indicating the source of each record. Users can then select only evaluations, silvicultural studies, and other research records that are appropriate for their needs. Approximately one projects. The sharing of species lists from these efforts is fourth of the fields in the Atlas database relate to the quality of the mutually beneficial, as the providing agency sees their data original data or to quality control. An additional step that assures expressed in the larger context of the complete Atlas the quality of Atlas data is the inspection of distribution maps for database and the distribution maps that it produces. each taxon. A map with an unlikely record, such as a coastal species mapped in an alpine region alerts the Atlas project staff ¥ Oregon Natural Heritage Information Center (ORNHIC) that an error might have occurred in the data entry, or that the data—This program maintains a database of rare, plant identification for that record may be incorrect. These threatened, and endangered plants of Oregon. ORNHIC is in

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 101 VEGETATION ECOLOGY

the process of sharing over 11,000 of their records, which opportunity to improve and update the data, which is inherent in represent approximately 23% percent of the native plant the design of the Flora project products, is a rewarding challenge taxa of Oregon. It is important to note that the Atlas for its staff members and contributors. Their continuing efforts mapping program has controls which prevent the release of are creating a dynamic resource that will serve as a contemporary sensitive locality data on rare, threatened and endangered botanical reference of Oregon for decades to come. plants. Individuals desiring such information are referred to ORNHIC staff. Literature Cited

Project Organization Chambers, K.L. and S. Sundberg. 2000. Oregon vascular plant checklist: Asteraceae. Second printing with changes and The Oregon Flora Project leadership consists of the additions. Oregon Flora Project, Oregon State University, Director, Scott Sundberg, who also directs coordinates most Corvallis. aspects of the project, 18 additional project leaders, nine checklist Cronquist, A.C., A.H. Holmgren, N.H. Holmgren, J.L. Reveal, and advisory board members, and nine Atlas project regional P.K. Holmgren (eds). 1972-1994. Intermountain flora; coordinators. Collectively, participants’ backgrounds encompass vascular plants of the intermountain west, U.S.A. Vol. 1 plant taxonomy, GIS, cartography, software development, ecology, published by Hafner Publishing Co., Inc., New York, NY; and statistics both at professional and lay levels. As discussed Vol. 3, part B, Vol. 4 and Vol. 5 by the New York Botanical above, NACSE software developers contribute enormously to the Garden, Bronx, NY; Vol. 6 by Columbia University Press, project. Nearly all of the professional floristic botanists in the New York, New YorkY. state participate in the project in some capacity. In addition to Gilkey, H.M., and L.J. Dennis. 2001. Handbook of Northwestern their role in advisory positions, volunteers are integral to the daily plants. Revised edition. Oregon State University Press, progress of the Flora project. Over 200 individuals of all ages Corvallis. and skill levels have contributed their efforts to all facets of the Gilkey, H.M. and P.L. Packard. 2001. Winter twigs. Revised Oregon Flora Project, providing thousands of work hours. It is edition. Oregon State University Press, Corvallis. indeed a grassroots endeavor, unique in that it is being created Guard, B.J. 1995. Wetland plants of Oregon & Washington. Lone through the combined efforts of amateurs and professionals. Pine Publishing, Redmond, Washington. Likewise it is a resource that can be effectively utilized by both Hickman, J.C. (editor). 1993. The Jepson manual. Higher plants groups of users. of California. University of California Press, Berkeley. Funding for the Oregon Flora Project has come from both Hitchcock, C.L., and A.C. Cronquist. 1973. Flora of the Pacific public and private sources. Since its inception, the Flora project Northwest. University of Washington Press, Seattle. has been sponsored by the Native Plant Society of Oregon, both Hopkins, L., J. Fosback, M. Carlson, and M. Thiele. 1993. Flora financially, with annual grants, and intellectually, by organizing distribution survey. 1978 to 1993. Second edition. Douglas numerous field trips and work parties to gather data for the Atlas County Museum of History and Natural History, Roseburg, database. Small grants from societies, foundations, and businesses Oregon. have also been instrumental in funding the Flora project research. Loy, W.G., S. Allan, A.R. Buckley, and J.E. Meacham. 2001. Sizeable additions to this shoestring budget came as awards Atlas of Oregon, second edition. University of Oregon between 2001 and 2003 in the form of a one-year grant from the Press, Eugene. Bureau of Land Management, and two grants from the National Mansfield, D.H. 2000. Flora of Steens Mountain. Oregon State Science Foundation for efforts that provide data for the Atlas and University Press, Corvallis. software development for the digital Flora; these allowed the Ohio Lepidopterists. 1992. The comprehensive survey of Ohio Flora project to hire its first full-time paid staff. As the Oregon moths and butterflies. Project Nno. NGSCW- 91-11, Ohio Flora Project continues to work towards its goals, the support of Division of Natural Resources, Columbus, OhioH. both the individual and the institution are critical to its success. Peck, M.E. 1961. A manual of the higher plants of Oregon, 2nd Information on the Oregon Flora Project is shared with the edition. Binfords & Mort, Portland, Oregon. public through its website, www.oregonflora.org, the Oregon Simpson, C., J. Koenig, J. Lippert, R. Love, B. Newhouse, N. Flora Newsletter, and through lectures and articles by the Friends Otting, S. Sundberg, D. Wagner, and P. Warner. 2002. of the Oregon Flora Project, a committee of the NPSO. The Vascular plants of Lane County Oregon; an annotated purpose of the Friends of the Oregon Flora Project is to raise checklist. Emerald Chapter Native Plant Society of Oregon, funds and provide outreach for the project; benefits of Eugene. membership in the Friends include a subscription to the Oregon Urban, K. 1995. Requiem for a laptop: seeds of the Oregon Flora Newsletter. Checklist Project. Oregon Flora Newsletter 1(2):5-7. Certainly now, as work is underway on each facet, and even in future years, when the full spectrum of resources will be available, the Oregon Flora Project is a work in progress. The

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 102 VEGETATION ECOLOGY

Table 1. A description of the tables in the Checklist database and representative information within each table.

Table Description Number of fields Content examples

scientific name, family, common name, hybrid status, names and associated origin (native or exotic), ORFLORA information for all accepted 47 taxonomic notes, treatment taxa contributor name, management information

names and associated information that have been scientific name and family, SYNONYMS used for Oregon taxa in other 37 taxonomic notes, name of references but are not accepted reference using the name in the Oregon Checklist

taxon names for species scientific name and family, previously reported for Oregon EXCLUDED 19 justification for rejecting the but not known to have occurred name, report source in the state

Figure 1. Data from each facet of the Oregon Flora Project will be accessible over the Internet. Software will interconnect the components, allowing users to select diverse information about any given plant.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 103 VEGETATION ECOLOGY

Figure 2. Distribution map of Collinsia parviflora generated from the Oregon Plant Atlas mapping program. Circles indicate unvouchered observations, and squares represent vouchered specimens.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 104 VEGETATION ECOLOGY

ANALYSIS OF PACIFIC YEW HABITAT IN NORTHWEST CALIFORNIA

Thomas M. Jimerson USDA Forest Service, Six Rivers National Forest 1330 Bayshore Way, Eureka, CA 95501 Email: [email protected]

Stanley Scher Native Yew Conservation Council, P.O. Box 2928, Berkeley, CA 94702 Email: [email protected]

Abstract

In the Klamath Mountain province, interactions between abiotic and biotic factors help to define habitat requirements of Pacific yew (Taxus brevifolia). This study identifies environmental gradients, plant associations and seral stages of overstory canopy species as candidate determinants of Pacific yew tree as opposed to shrub habitat in forested watersheds of the Six Rivers and Klamath National Forests. Discriminant analysis of 2291 plots within the study area provides a basis for developing predictive models of Pacific yew distribution in this physiographic province and a rational approach to a conservation strategy for Pacific yew in northern California. An abbreviated version of this paper was presented at the International Yew Resources Conference, Berkeley, California, March 12-13, 1993. Introduction ~100-8000 ft. (30-2450m). Slopes are generally steep, ranging from 0 to 95%. The major geologic bedrock of the study area range from Mesozoic ultrabasic intrusive, Jurassic-Triassic he Pacific yew (Taxus brevifolia Nutt.) is a slow-growing, metavolcanic, pre-Cretaceous metasedimentary, upper Jurassic long-lived, shade-tolerant conifer. It achieves maximum marine sedimentary formations to Mesozoic granitics. TTdevelopment as an understory or mid-canopy species in The climate of the study area has previously been late-seral forest communities with low fire frequency in close characterized (Albert 1979; Shumway 1981; Parsons and Knox proximity to water. Pacific yew is often found on or near stream 1984). Seven forest series dominate the study area. Their banks, river margins, canyon bottoms, wet shaded ravines and distribution is largely determined by elevation, available moisture, other riparian habitats (Crawford and Johnson 1985; Scher and and parent material (Jimerson 1993). Jimerson 1989). This small to mid-sized tree or shrub occurs most frequently on public lands from northern California to Methods southeastern Alaska. This paper reports studies on the abiotic and biotic This study was conducted in conjunction with the environment of Pacific yew in northwest California and compares ecosystem classification program on the Six Rivers and Klamath habitats where it occurs in tree form, shrub form, or is absent. We National Forests in northern California. Sampling was restricted describe Pacific yew occurrence in selected plant associations, and to the major conifer series within the study area. Late-seral stage vegetation seral stages as well as the environmental gradients stands (old-growth), mid-seral stands (mature), and early-seral along which it is found. Discriminant analysis is used to develop stands (pole and shrub/forb) were stratified and randomly selected predictive models for environment and vegetation where Pacific as study sites. yew is likely to occur. Sampling methods generally followed those described by Study Area Hall (1970), Moir and Ludwig (1983) and Allen and Diaz (1986). Sample plot locations were restricted to forested stands with The study area is located in the Klamath Region of homogeneous vegetation, seral stage, soils, geology, and landform northwestern California. It is bounded on the north by the Oregon (Pfister and Arno 1980). For herbs, graminoids, and seedling border, on the south by Mendocino County, on the east by the trees, we used a plot size of 0.1 acre and of variable radius for Klamath River and South Fork Mountain, and on the west by the shrubs and overstory trees (Bitterlich 1947). For each plot the Six Rivers National Forest boundary. Within the study area, the following descriptive information were recorded: plot number, dominant topographic features include the Coast Range and national forest, ranger district, forest map number, forest timber Siskiyou Mountains drained by the Smith, Klamath, Mad, Van compartment, township, range, section, latitude, latitude minutes, Duzen, and Eel rivers and their tributaries. Elevation ranges from longitude, and longitude minutes.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 105 VEGETATION ECOLOGY

dependence of Pacific yew cover on environment and vegetation Environment Variables variables. The set of points generated by this program form a The physical environment was defined by elevation, aspect, nonparametric regression of y on x. The robust feature prevents percent slope, landform (general topographic characterization), outliers from distorting the curves. micro-position (position of the plot on the slope), horizontal Two programs, Compartment Inventory Analysis (CIA) and micro-relief (slope shape parallel to the contours), vertical micro- Forest Inventory Analysis (FIA) (USDA 1986), were used to relief (slope shape across the contours), bare ground percent, analyze the tree data. Information generated by these programs surface gravel percent, surface rock percent, potential annual included: standing cubic volume, cubic volume growth, basal area, radiation (total annual radiation received for a given aspect, slope basal area growth, site class, and trees per acre by species and dbh and latitude) (Frank and Lee 1966), transformed aspect (aspect class. transformed to a linear variable from 1-8) (Lewis 1982), and SPSSPC+ (Norusis 1990) was used to develop descriptive radiation index (ratio of total annual radiation on a given aspect statistics (mean, standard deviation, standard error), calculate and slope to total annual radiation received on a flat surface for a stand density index (SDI) (Reineke 1933), and to test for given latitude) (Frank and Lee 1966). differences in group means (t-test, p=0.05) or frequencies (chi- At each site, data were recorded on: percent surface litter square, p=.005) (Norusis 1990). Stepwise discriminant analysis cover, litter thickness, parent material, parent material origin, total (Jennrich and Sampson 1990) was used to select the combination soil depth (to a maximum of 40 inches), rootability (whether the of variables that best predicts Pacific yew occurrence. soil can be penetrated by roots), A horizon thickness, texture, and Due to the lack of a separate validation data set and the percent coarse fragments (using 2 mm sieved soil samples), A tendency of the original model to be overly optimistic in its horizon color (hue, value, and chroma using Munsell color charts) successful classification, jackknifed cross-validation is used to (Munsell 1975), sub-surface texture and percent coarse fragments, validate the stepwise discriminant function (Efron 1982). Due to color, soil drainage class, available water-holding capacity (AWC) the use of vegetation species as variables—which are often for the top 20 inches of soil, soil name (classified to family), pH missing from a plot—the assumptions of stepwise discriminant of the surface horizon (using a Hellige-Truog Soil Reaction analysis, samples from a multivariate normal population and Tester), and pH of the sub-surface horizon. equality of covariance matrices were not met. The authors however, do not believe that this invalidates the results of this Vegetation Variables analysis, and therefore, the results will be presented here. At each plot, total percent cover by moss, forbs, graminoids, shrubs and trees was estimated and recorded. All plants were Results identified to species where possible; (nomenclature follows Munz 1973). Abundance was recorded for the herbaceous and Of 2291 plots analyzed in this study, 258 contained Pacific graminoid layers only (Allen and Diaz 1986). yew. A summary of environmental and vegetation variables for Plots were classified to series, sub-series, and plant plots containing Pacific yew is presented in Table 1. association using the methods described in Jimerson (1993). Fidelity, the degree to which a species is associated with a Abiotic Environment classification group, was calculated for all classification groups Pacific yew occurred most frequently between ~1,000 and containing Pacific yew. Plots were also assigned a moisture index ~4,000 ft. elevation; mean elevation was 3,050 ft. (Table 1). Tree- rating from 1to10 based on their available moisture. For example, form Pacific yew cover was highest at ~1,000 ft. and decreased plots found in ridgetop positions with shallow soils and high soil sharply with increasing elevation (regression correlation -.16, coarse fragments would tend to be dry and would receive a R2=0.03, sign.= 0.08). Above ~3,000 ft. cover increased slightly. moisture index rating of 1, while plots located in lower 1/3 slope Mean elevation for Pacific yew differed significantly by form positions adjacent to streams with deep soils would tend to be (F=71.1, sign.= 0.0000). Pacific yew in tree form was found at a moist and would receive a moisture index rating of 10. mean elevation of 2,700 ft., while the shrub form had a mean Estimates of number of trees per acre by diameter class, tree elevation of ~3,316 ft.; plots without Pacific yew averaged height and standing basal area (basal area factor 20 or 40) were ~3,700 ft. in elevation. recorded at three points per plot using a Spiegel relaskop. In Within the study area, Pacific yew occurs primarily on addition, diameter at breast height (dbh), total tree age, 10 and 20- northerly and easterly aspects with topographic shading in the year radial growth were recorded for a minimum of one dominant draws, and in the lower one-third-slope position where moisture tree per point. conditions are highest (Table 2). Comparisons of slope positions at sites with Pacific yew in tree form, shrub form, and without Data Analysis Pacific yew showed significant differences (chi-square=170.4, Environment and vegetation data were analyzed using the sign.= 0.0000). The tree form of Pacific yew was found in the following programs and statistical packages: Robust locally lower 1/3 slope position 63% of the time, compared to 38% for weighted regression (Cleveland 1979) was used to assess the the shrub form; plots without yew occurred at a 19% rate of

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 106 VEGETATION ECOLOGY occurrence. maple/sword fern (22%), tanoak/vine maple-salal (72%), and Slope shapes containing Pacific yew were primarily concave tanoak/vine maple (25%). and linear and differed significantly by frequency of sites with tree Comparisons of mean stand age and stand age frequency by form, shrub form, and without Pacific yew (chi-square=34.7, forest series showed significant differences (F=32.9, sign.=0.0000 sign.= 0.0000). The tree form of Pacific yew occurred with & chi-square=264.8, sign.=0.0000) (Table 4). The Port-Orford highest frequency (57%) on concave slopes; the shrub form had its cedar series had the oldest mean stand age of 366 years, followed highest frequency (49%) on linear slopes, and sites without yew by the tanoak series 263 years, with the oak woodland series had the highest frequency on convex slopes (17%). having the youngest mean stand age of 150 years. The Port- Pacific yew cover was highest on gentle to moderate slopes Orford cedar series also had the highest frequency of stands above (0-20%), decreasing in cover with increasing slope until 60% 400 years stand age (35%) and the highest frequency of stands slope (regression correlation -25, R2= 0.06, sign.= 0.0006), but above 300 years stand age (65%). also increased in cover on steep slopes (60-75%). Cover and constancy (the percentage of times a species Surface rock and gravel on Pacific yew sites were minimal occurs in association with Pacific yew) for trees and shrubs are (2-5%) (Table 1). Soils were rootable, well drained, primarily listed in Table 5. Within the tree layer, Douglas-fir had the highest skeletal, moderately deep to deep, very dark grayish-brown to constancy of all Pacific yew associates (99%), followed by tanoak dark yellowish-brown in color, and derived primarily from (53%), white fir (39%), Chinquapin (Castanopsis chrysophylla metamorphic (54%) and igneous intrusive (35%) parent material Dougl.) (37%), dogwood (Cornus nuttallii Aud.) (32%), and Port- of colluvial and bedrock origin. Orford Cedar (34%). The shrub layer was dominated by dwarf Available water-holding capacity (AWC) for the top 20 Oregon grape (Berberis nervosa Pursh.) (70%), red huckleberry inches of soil was moderately high to high (3.5 inches of water) (Vaccinium parvifolium Sm.) (49%), wild rose (Rosa gymnocarpa) (Table 1). Measurements of surface and sub-surface soil samples (43%), blackberry (Rubus ursinus Cham. & Schlecht.) (41%), indicated moderately acidic conditions (pH 5.6 & 5.8) (Table 1). salal (Gaultheria shallon Pursh.) (37%), hazelnut (Corylus cornuta Marsh.) (39%), and others with lesser constancies (Table Biotic Environment 5). In stands containing Pacific yew, vegetation cover, generally The herb layer was dominated by prince’s pine (Chimaphila exceeded 95%; and the tree overstory generally exceeded 75% umbellata (L.) Barton) (61%), rattlesnake plantain (Goodyera (Table 1). Of the seven major forest series sampled, the Port- oblongifolia Raf.) (59%), vanilla-leaf (Achyls triphylla (Sm. DC.) Orford cedar and tanoak series showed the highest frequency of (55%), twinflower (Linnaea borealis Torr.) (55%), western Pacific yew occurrence—23% and 18% respectively—while the modesty (Whipplea modesta Torr.) (53%), sword fern Jeffrey pine and oak woodland forest series contained no Pacific (Polystichum munitum Presl.) (52%), and numerous other species yew (Table 3). Within the Port-Orford cedar series, the Port- of lower constancy (Table 5). Orford cedar-Douglas-fir sub-series had a frequency of Pacific Moisture index frequency comparisons showed significant yew occurrence of 41% (Table 3). This sub-series is represented differences between plots with and without Pacific yew (chi- by one plant association, the Port-Orford cedar-Douglas- square=108.9, sign.=0.0000). Plots with Pacific yew had their fir/huckleberry oak type. The Port-Orford cedar-shrub sub-series highest frequency in the wet moisture indices 7-10 (59%), and the with 29% Pacific yew occurrence includes the Port-Orford lowest frequency in the intermediate sites 4-6 (16%). Plots cedar/salal (25%), Port-Orford cedar/rhododendron-salal (22%), without Pacific yew had their highest frequency on dry sites 1-3 and Port-Orford cedar/western azalea (43%) plant associations. (39%) and their lowest frequency on wet sites 7-10 (27%). The Port-Orford cedar-white fir sub-series had a frequency of 22% On sites containing Pacific yew, stand age ranged from Pacific yew occurrence. It includes the Port-Orford cedar-white approximately 78 to 641 years, with a mean of 334 years (Table fir/Herb (14%), Port-Orford cedar-white fir/huckleberry oak (27%) 1). Significant differences in mean stand age and age class and Port-Orford cedar-white fir/Sadler oak (28%) plant frequency were found in Pacific yew growth form. Mean stand associations. age for Pacific yew tree form was 350 years, while shrub form In the tanoak series, the tanoak-Port-Orford cedar sub-series averaged 322 years, and stands without Pacific yew averaged 239 had the highest frequency of Pacific yew occurrence 44%. It years (F=71.1, sign.= 0.0000). The tree form of Pacific yew had includes the following plant associations and frequencies of its highest frequency of occurrence in stands between 300-399 Pacific yew occurrence: tanoak-Port-Orford cedar-California years old, the shrub form between 200-299 years, while plots bay/evergreen huckleberry (36%), tanoak-Port-Orford without Pacific yew had their highest frequency between 100-199 cedar/evergreen huckleberry-western azalea (78%), tanoak-Port- Orford cedar/evergreen huckleberry (35%), tanoak-Port-Orford years old (chi-square=162.0, sign.=0.0000) (Table 6). cedar-/dwarf Oregon-grape/twinflower (40%), tanoak-Port-Orford Stands containing Pacific yew were found primarily (88%) cedar-alder/riparian (50%), tanoak-Port-Orford cedar/vine maple in the old-growth seral stage with 12% in the mature seral stage. (43%), and tanoak-Port-Orford cedar/red huckleberry (50%). The Sites without Pacific yew were also included with highest tanoak-bigleaf maple sub-series had a frequency of Pacific yew frequency in the old-growth seral stage (66%) and mature seral occurrence of 37%. The plant associations and frequencies of stage (30%). They were determined to have significantly different Pacific yew occurrence within it include: tanoak-bigleaf frequencies by seral stage (chi-square=90.9, sign.=0.0000).

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 107 VEGETATION ECOLOGY

Pacific yew cover increased with standing basal area, with parent material with slightly acidic pH. These factors contribute to 2% cover at 120 ft.2 to 6% cover at 350 ft.2 per acre, decreasing to deep colluvial soils with high AWC, retention of soil moisture, and 3% cover thereafter. No significant difference was found in high potential for conifer growth. Pacific yew occurs most comparisons of mean basal area for Pacific yew tree form, shrub frequently on these high productivity sites. The conditions, which form, or stands without Pacific yew. produce superior tree growth, deep, and moist soils and cool Analysis of Pacific yew occurrence by tree density using a temperatures, also provide optimum conditions for Pacific yew stand density index (Reineke 1933) demonstrated a narrow range reproduction and survival. of optimal habitat, with highest cover at 300-450 for stand density Pacific yew cover increases with basal area and stand index. No significant difference was found in comparisons of tree density index. As a sub-canopy species, Pacific yew is shade form, shrub form, and stands without Pacific yew by stand density tolerant; however, it responds to markedly reduced light intensity. index. Analysis of Pacific yew occurrence by site class (Dunning Above 340 ft.2 basal area, 460 stand density index, the amount of 1942) shows a significantly higher frequency of tree form Pacific light reaching the forest floor is markedly reduced, and Pacific yew on high productivity sites (chi-square=21.2, sign.= 0.02) yew cover decreases. (Table 7). On class 1A sites, tree form Pacific yew occurred with Forest habitats in which Pacific yew occur can be a frequency of 35% compared to 23% for the shrub form and sites characterized as mesic, with dense vegetative cover often in excess without Pacific yew; on site class 4 lands, the frequency was 3% of 95%. The tree overstory generally exceeded 75% canopy cover. for tree form, 14% for shrub form, and 8% for lands without The dense canopy contributes to a cool, moist environment, which Pacific yew. serves to reduce moisture loss, and fire frequency. On sites with Within stands containing Pacific yew, the number of yew low overstory canopy cover (< 50%), less soil moisture is retained, trees/acre by size class increases with increasing age (Table 8). In and Pacific yew showed foliage discoloration, suggesting loss of stands below 200 years old, Pacific yew tree size was < 6” dbh, chlorophyll(s). while those above 300 years contained trees of all size classes with Pacific yew was found primarily in the Port-Orford cedar Pacific yew trees as large as 11-17.9” dbh beginning to appear and tanoak series, the two forest series with the oldest mean stand after 200 years stand age. age, in plant associations located along stream banks and canyon Stepwise discriminant analysis based on presence or absence bottoms, where stand age was significantly higher in stands of Pacific yew was used to develop a linear combination of containing yew compared to stands without yew. Here abundant predictor variables of Pacific yew habitat. Several of the variables moisture conditions favor mesic species such as Port-Orford cedar discussed earlier, such as stand age, moisture index, elevation, and (Whittaker 1960; Zobel et al. 1985) and help to reduce the micro-position, along with a group of vegetation species indicating frequency of stand-replacing wildfires. Corresponding distribution high available moisture, including twinflower, dogwood, salal, of Pacific yew and Port-Orford cedar in Research Natural Areas, vine maple, and bigleaf maple were found to be significant such as the Coquille River Falls and Port-Orford cedar in the coast predictors of Pacific yew habitat. A summary table of these ranges of Oregon, support these findings (Franklin et al. 1972). significant predictor variables with U-statistic and F-value are Compared to most other conifers, Pacific yew is highly contained in Table 9. Stand age had the highest F-value (F=144.1) sensitive to heat damage (Crawford and Johnson 1985; Scher and and was the first variable to enter the equation, it was followed by Jimerson 1989), possibly because of its thin bark. Both high twinflower, moss cover, moisture index, and elevation. A temperatures and duration of exposure influence survival and seed comparison table of mean values for significant discriminant germination . variables is found in Table 10. The success of the discriminant Fire frequency in northwestern California suggests a second function in predicting Pacific yew habitat was 72.8% for plots factor responsible for the unequal distribution of Pacific yew by containing Pacific yew and 80.9% for plots without yew (Table growth form. Stand-replacing fires occur with higher frequencies 11). The jackknifed cross validation of the discriminant function at higher elevations (Viers 1980) where Pacific yew occurs with had very similar results, 71.2% with Pacific yew and 80.7% lowest frequency. Such fires occur every 500-600 years at low without yew (Table 12). elevations, 150-200 years at intermediate sites, and 33-50 years on high elevation sites. Our studies support the findings of Viers Discussion (1980). Series mean ages decreased with incresing elevation from In northwestern California, occurrence of Pacific yew 366 years old in the Port Orford cedar series at low elevations growth form varies significantly by habitat. The tree form of where Pacific yew grew in tree form occured with highest Pacific yew is found most frequently between ~1,000 to ~4,000 ft. frequency to 234 years old for the Douglas fir series at elevation, on sites with northerly aspects or topographic shading. intermediate elevations, to 228 years old in the red fir series at Pacific yew sites are located primarily in the draws or lower one- high elevations where Pacific yew occured in the shrub form and third slope position on gentle, concave or linear slopes, with the with low frequency. tree form having the highest frequency. These conditions suggest A key characteristic of old-growth forests is the association a cool, moist microclimate. of long-lived seral dominant species such as Douglas-fir with a The soils here are derived from fine textured metamorphic shade-tolerant understory: westerm hemlock or Pacific yew

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 108

VEGETATION ECOLOGY

(McCune and Allen 1985; Franklin et al. 1988; Busing et al. forests (in western Oregon and Washington), long-term studies of 1995). Since fire risks are very low in old-growth Douglas-fir forest recovery following logging, and other data support the stands, the density of Pacific yew populations increases with hypothesis that Taxus is a widespread but predominantly late- Douglas-fir age to ~500 years. As we have demonstrated here, successional species, sensitive to fire, and slow to recover from Pacific yew is more common in old-growth forests than in disturbance. These authors conclude that Taxus attains maximal younger stands. These findings strongly suggest that long-lived basal area and adult stem density in old forests. Our studies of temperature-sensitive species such as Pacific yew may serve as a Pacific yew in the Klamath mountain province of northwestern useful indicator of old-growth forests. California are consistent with this hypothesis. A key characteristic of old-growth forests is the association of long-lived seral dominant species such as Douglas-fir with a Acknowledgements shade-tolerant understory: westerm hemlock or Pacific yew (McCune and Allen 1985; Franklin et al. 1988; Busing et al. We thank Max Creasy, Steve Matthews, Ken Simeral, 1995). Since fire risks are very low in old-growth Douglas-fir Misha Schwarz and David Imper for their assistance in collecting stands, the density of Pacific yew populations increases with data, and David Jones for reviewing the manuscript. Douglas-fir age to ~500 years. As we have demonstrated here, Pacific yew is more common in old-growth forests than in Literature Cited younger stands. These findings strongly suggest that long-lived temperature-sensitive species such as Pacific yew may serve as a Achuff, P.L. 1989. Old-growth forests of the Canadian Rocky useful indicator of old-growth forests. Mountain National Parks. Natural Areas Journal 9(1): 12- The development of a predictive discriminant model of 26. Pacific yew habitat was moderately successful, classifying Albert, G. 1979. Map of average annual precipitation. USDA approximately 73% of the plots correctly. Some potential reasons Forest Service, Region 5, Six Rivers National Forest, for this include: 1) the low density of Pacific yew in areas Eureka, California. containing the species and the likelihood of encountering it in the Allen, B.H. and D.V. Diaz. 1986. R-5 Ecosystem Classification plot size used here; 2) the sensitivity of Pacific yew to fire and the Handbook. USDA Forest Service, Pacific Southwest possibility it may have been eliminated from potential sites by Region, San Francisco, California. wildfire; and 3) failure to measure the appropriate set of predictor Atzet, T.A., and D.L. Wheeler. 1984. Preliminary plant variables. associations of the Siskiyou Mountain Province. USDA Studies such as this one that develop species specific, Forest Service, Pacific Northwest Region, Portland, Oregon. habitat characterizations, could be used to develop conservation Bitterlich, W. 1947. Measurements of basal area per hectare by strategies for Pacific yew. means of angle measurement. Allgemeine Forst Holzwirtschaft Zeitung 58:94-96. Conclusions Bollsinger, C.L., and A.E. Jaramillo. 1990. Taxus brevifolia Nutt. Pacific yew. Pages 573-579 in Silvics of Forest Studies of Pacific yew distribution in northwestern Trees of North America. R.M. Burns and B.H. Honkala, California suggest that the major determinants of its distribution Technical Coordinators. USDA Forest Service, Agricultural and growth form are: Handbook No. 654, Vol. 1 Conifers, Washington, D.C. 1. old stand age, Brockway, D.G., C. Topik, M.A. Hemstrom, and W.H. Emmingham. 1983. Plant associations and management 2. low fire frequency, guide for the Pacific Silver Fir Region, Gifford-Pinchot 3. elevation, National Forest. USDA Forest Service, Pacific Northwest 4. slope position, Region, R6ECOL130a1983. Portland, Oregon. 5. slope shape, Busing, R., C. Halpern and T. Spies. 1995. Ecology of Pacific 6. proximity to water, yew (Taxus brevifolia) in western Oregon and Washington. 7. high available water-holding capacity, Conservation Biology 9(5):1199-1207. Cleveland, W.S. 1979. Robust locally-weighted regression and 8. associated conifer series (Port-Orford cedar and Douglas- smoothing scatterplots. Journal American. Statistical fir), Association 74(368): 829-836. 9. vegetative cover, Cooper, S.V., K.E. Neiman, R. Steele and D.R. Roberts. 1987. 10. basal area, Forest habitat types of northern Idaho. USDA Forest 11. stand density, and Service. General Technical Report INT-236. Ogden Utah. 12. site class. Crawford, R.C. 1983. Pacific yew community ecology in north central Idaho with implications to forest management. Busing and others (1995) have noted that surveys of natural Ph.D. dissertation. University of Idaho, Moscow.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 109

VEGETATION ECOLOGY

Crawford, R.C. and F.D. Johnson. 1985. Pacific yew dominance Statistical Software Manual Volume 1. University of in tall forests, a classification dilemma. Canadian Journal California Press, Berkeley: 339-358. of Botany 63: 592-602. Jimerson, T. M. 1993. Preliminary plant associations of the Curtis, R.O. 1982. A simple index of stand density for Douglas- Klamath Province, Six Rivers and Klamath National fir. Forest Science 28(1):92-94. Forests. USDA Forest Service, Six Rivers National Forest, Daubenmire, R.F., and J.B. Daubenmire. 1968. Forest vegetation Eureka, California. of eastern Washington and northern Idaho. Agricultural Lewis, J. 1977. Soil moisture and temperature regimes in the Experiment Station Technical Bulletin 60 Washington State Mad River District of Six Rivers National Forest. MS. University, Pullman. 104 p. thesis, Humboldt State University, Arcata, California. 187p. Dunning, D. 1942. A site classification for the mixed-conifer McCune, B., and T.F.H. Allen. 1985. Will similar forest develop selection forests of the Sierra Nevada. USDA Forest on similar sites. Canadian Journal of Botany 63:367-376. Service. California Forest and Range Experiment Station. Moir, W.H., and J.A. Ludwig. 1983. Methods of forest habitat Research Note 28. Berkeley, California. 21 p. type classification. Page 6 in, Workshop on Southwestern Efron B. 1982. The jackknife, the bootstrap and other resampling Habitat Types, Albuquerque, New Mexico. plans. CBMS-NSF Regional Conference Series in Applied Munsell Color Corporation. 1975. Munsell soil color charts. Mathematics. Society for Industrial and Applied Kollmongen Corp. Baltimore, Maryland. Mathematics, Philadelphia, Pennsylvania. Munz, P.A. 1973. A California flora. University of California Frank, E.C., and R. Lee. 1966. Potential solar beam irradiation on Press. Berkeley. slopes: Tables for 30¼ to 50¼ latitude. USDA Forest Service, Norusis, M.J. 1988. SPSS/PC+ For the IBM PC/XT/AT/PS2. Research Paper RM-18. Rocky Mountain Forest and Range SPCC Inc. Chicago Illinois. Experiment Station. Ft. Collins, Colorado. 116 p. Parsons, A.M., and E.G. Knox. 1984. Soils of the Six Rivers Franklin, J.F., and C.T. Dyrness. 1972. Natural vegetation of National Forest. Under the direction of S.R.Miles. Copy on Oregon and Washington. USDA Forest Service General file at USDA Forest Service. Region 5, Six Rivers National Technical Report PNW-8. Portland, Oregon. Forest, Eureka, California. Franklin, J.F., F.C. Hall, C.T. Dyrness, and C. Maser. 1972. Pfister, R.D., and S.F.Arno. 1980. Classifying forest habitat types Federal research natural areas in Oregon and Washington, A based on potential climax vegetation. Forest Science Guidebook for scientists and educators. USDA Forest 26(1):52-70. Service. Pacific Northwest Forest and Range Experiment Pierce, D.J. 1984. Siras moose forage selection in relation to Station. Portland, Oregon. Page code: CO-2, PO-2. browse availabity in north-central Idaho. Canadian Journal Franklin, J.F., and D.S. DeBell. 1988. Thirty-six years of tree of Zoology 62 (12):2404-2409 population change in an old-growth Pseudotsuga-Tsuga Reineke, L.H. 1933. Perfecting a stand density index for even- forest. Canadian Journal of Forest. Research 18:633-639. aged forests. Journal of Agricultural Research 16:627-638. Franklin, J.F., W.H. Moir, M.A. Hemstrom, S.E. Greens and B.G. Scher, S., and T.M. Jimerson. 1989. Does fire regime determine Smith. 1988. Forest communities of Mount Rainier the distribution of Pacific yew in forested watersheds? National Park. US National Park Service Scientific USDA Forest Service. Pacific Southwest Research Station. Monograph. Series No. 19. General Technical Report. GTR-PSW-109. Nerkeley, Hall, F.C. 1970. An ecological classification proposal and its California. importance in land measurement. In, Range and Wildlife Shumway, S.E. 1981. Climate, In Forest soils of the Douglas fir Habitat Evaluation. A Research Symposium. USDA region P.E. Heilman, H.W. Anderson, and D.M. Miscellaneous Publications 1147. 210-217. Baumgartner (eds.) Washington State University, Pullman. Hemstrom, M.A., W.A. Emmingham, N.M. Halvorson, S.E. 87-91. Logan and C. Topik. 1982. Plant associations and Spies, T.A., and J.F. Franklin. 1988. Old growth and forest management guide for the Pacific Silver Fir Zone, Mt. dynamics in the Douglas-fir Region of western Oregon and Hood and Willamette National Forests. USDA Forest Washington. Natural Areas Journal 8(3):190-201. Service R6-ECOL-100-1982A. Portland, Oregon. 104 p. USDA. 1986. Timber management plan inventory handbook. Hemstrom, M.A., and J.D. Franklin. 1982. Fire and other USDA Forest Service, Pacific Southwest Region, R-5 FSH disturbances of the forests in Mount Rainier National Park. 2409.21b. San Francisco, California. 525 p. Quaternary Research 18:32-51. Viers, S.D., Jr. 1980. The influence of fire in coast redwood Hemstrom, M.A. and S.E. Logan. 1986. Plant associations and forests. In Proceedings of the Fire History Workshop, management plan, Siuslaw National Forest. USDA Forest Laboratory of Tree Ring Research, University of Arizona, Service, Pacific Northwest Region. R6-ECOL-220-1986A. Tucson. October 20-24. pp. 93-95. Portland. Oregon. Wani, M.C., H.L. Taylor, M.E. Wall, P. Coogen, and A.T. Jennrich, R. and P. Sampson. 1990. Stepwise Discriminant McPhail. 1971. Plant antitumor agents. VI. The isolation Analysis. Pages 339-358 in, W.J. Dixon, editor. BMDP and structure of taxol, an antileukemic and antitumor agent

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 110 VEGETATION ECOLOGY

from Taxus brevifolia. Journal of the American Chemical Zobel, D.B., L.F. Roth, and G.L. Hawk. 1985. Ecology, Society 93: 2325-2327. pathology and management of Port-Orford-Cedar Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, (Chamaecyparis lawsoniana). USDA Forest Service, Oregon and California. Ecological Monographs 30(3):279- General Technical Report. GTR-PNW-184. Pacific 338. Northwest Forest and Range Experiment Station. Portland,

Table 1—Environment and vegetation data for plots containing Taxus brevifolia, Six Rivers and Klamath National Forests, northwestern California.

Variable Mean Std. Dev. Minimum Maximum Elevation (meters) 930 301 305 1585 Slope (%) 42 24 2 75 Surface Gravel (%) 2 5 0 5 Rock (%) 5 12 0 10 Soil Depth (inches) 34 10 19 > 40 Coarse Fragments (%) 38 9 14 80 AWC (inches water) 3.5 2.5 1.3 7.1 Surface pH 5.6 1.8 5.0 7.5 Sub-surface pH 5.8 1.8 5.3 8.0 Vegetation Cover (%) 95 6 80 100 Tree Cover (%) 79 18 60 98 Stand Age (years) 334 115 78 641

Basal Area (ft.2) 301 107 120 587 Stand Density Index 450 153 126 920 Relative Density 64 26 16 156

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 111

VEGETATION ECOLOGY

Table 2. Frequency comparison of sites with tree form, shrub form, and without Pacific yew by landscape position.

Frequency (%) Landscape Position Tree form Shrub form Without Yew Ridge, Summit, Crest 1 4 12 Upper One Third Slope 15 20 42 Middle One Third Slope 22 38 27 Lower One Third Slope 63 38 19

Table 3. Fidelity table for series, sub-series, and plant associations containing Pacific yew.

Vegetation Type Plots With yew Frequency (%) Tanoak Series 583 104 18 Tanoak-Shrub Sub-series /Evergreen Huckleberry 30 1 3 /Evergreen Huckleberry-Salal 25 6 24 /Rhododendron-Evergreen Huck. 10 3 30 /Rhododendron-Salal 13 2 15 /Salal 17 2 12 /Dwarf Oregon grape 45 9 20 /Poison Oak-Hairy Honeysuckle 17 1 6 /Rhododendron-Huckleberry Oak 6 2 33

Tanoak-Maple Sub-series 62 23 37 /Bigleaf Maple/Sword fern 32 7 22 /Vine Maple-Salal 18 13 72 /Vine Maple 12 3 25

Tanoak-Port-Orford cedar Sub-series 75 33 44 /CA. Bay/Evergreen Huckleberry 14 5 36 /Evergreen Huckleberry-Azalea 9 7 78 /Evergreen Huckleberry 26 9 35 /Dwarf Oregon grape/Twinflower 10 4 40 /Alder/Riparian 4 2 50 /Vine Maple 7 3 43 /Red Huckleberry 4 2 50

Tanoak-Canyon Live Oak Sub-series /Rockpile 20 1 5 /Evergreen Huckleberry 19 3 6 /Dwarf Oregon Grape-Salal 14 2 14

Tanoak-Chinquapin Sub-series /Salal 20 2 10 /Rhododendron-Salal 21 6 29 /Rhododendron-Beargrass 21 2 10 /Dwarf Oregon Grape 34 3 9 /Evergreen Huckleberry-Salal 4 1 25

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 112 VEGETATION ECOLOGY

Vegetation Type Plots With yew Frequency (%) Port-Orford cedar Series 166 39 23 Port-Orford cedar/Shrub Sub-series 48 14 29 /Salal 16 4 25 /Rhododendron-Salal 18 4 22 /Azalea 14 6 43 Port-Orford cedar -White Fir Sub-series 50 11 22 /Huckleberry Oak 11 3 27 /Herb 21 3 14 /Sadler Oak 18 5 28

Port-Orford cedar-Red Fir Sub-series /Sadler Oak-Thinleaf Huckleberry 17 3 18

Port-Orford cedar-Douglas-fir Sub-series /Huckleberry Oak

Port-Orford cedar-Western White Pine Sub-series /Huckleberry Oak 9 2 22

White Fir Series 605 33 5 White Fir-Shrub Sub-series /Wild Rose-Snowberry 33 1 3

White Fir-Red Fir Sub-series /Sadler Oak 18 1 6

White Fir-Douglas-fir Sub-series /Sadler Oak 35 4 11 /Sadler Oak-Huckleberry Oak 33 8 24

/Mountain Maple 6 1 17 /Vine Maple 9 1 11 /Wild Rose/Twinflower 14 6 43 /Wild Rose-Snowberry 77 2 3 /Hazelnut 17 1 6 /Thimbleberry 6 3 50 /Western Modesty 16 7 44 White Fir-Brewer’s Spruce Sub-series /Sadler Oak-Thinleaf Huck. 11 2 18

White Fir-Chinquapin Sub-series 21 4 19

Red Fir Series 203 3 1 Red Fir-White Fir Sub-series /Vanillaleaf-Prince’s Pine 11 1 9 /Bracken Fern 17 1 6

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 113 VEGETATION ECOLOGY

Vegetation Type Plots With yew Frequency (%) Red Fir-Brewer’ Spruce Sub-series /Sadler Oak-Thinleaf Huck. 16 1 6

Douglas-fir Series 623 79 13 Douglas-fir-Shrub Sub-series /Huckleberry Oak 26 6 23 -California Bay/Poison Oak 8 1 13 /Hazelnut 29 2 7 /Big-leaf Maple/Sword fern 13 5 38 /Vine Maple-Dwarf Oregon-grape 4 3 75 /Huckleberry Oak-Dwarf Tanbark 12 1 8

Douglas-fir-Chinquapin Sub-series -Tanoak 24 7 29 /Rhododendron-Dwarf OR.-grape 14 3 21 /Sadler Oak/Bear grass 57 7 12 -Tanoak/Oregon-grape 14 4 29 Jeffrey Pine 76 0 0 Oak Woodland 35 0 0 Total 2291 258 11

Table 4. Frequency of stand age distribution for the primary forest series in the study area with mean age and 2 standard errors (SE).

Mean Age Age Class Frequency (%)

Forest Series Age (2SE) <100 100-199 200-299 300-399 >400

Tanoak 263 (9) 6 22 38 23 11

Douglas-fir 234 (13) 13 27 30 21 9

Port-Orford cedar 366 (23) 0 11 24 30 35

White Fir 238 (8) 8 38 28 16 10

Red Fir 228 (15) 5 46 30 12 7

Jeffrey Pine 243 (32) 9 41 20 18 12

Oak Woodland 150 (62) 40 40 11 9 0

Total 249 (5) 9 29 30 21 11

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 114 VEGETATION ECOLOGY

Table 5. Plant cover and constancy for associated tree and shrub species, Six Rivers and Klamath National Forests, northwestern California.

Species Cover Std. Dev. Min. Max. N Const. Trees: Pseudotsuga menziesii 40 17.9 2 90 255 99 Lithocarpus densiflora 23 20.9 1 80 136 53 Taxus brevifolia 5 5.0 1 30 113 44 Abies concolor 21 18.8 1 95 101 39 Castanopsis chrysolepis 12 12.8 1 60 96 37 Chamaecyparis lawsoniana 29 19.6 1 90 88 34 Pinus lambertiana 8 6.9 1 35 86 33 Cornus nuttallii 5 5.8 1 40 83 32 Arbutus menziesii 5 4.2 1 20 79 31 Acer macrophyllum 8 8.0 1 35 60 23 Libocedrus decurrens 9 7.6 1 30 49 19 Shrubs: Berberis nervosa 7 9.6 1 60 180 70 Vaccinium parvifolium 4 4.5 1 20 127 49 Rosa gymnocarpa 2 1.3 1 7 110 43 Rubus ursinus 1 0.9 1 5 107 41 Corylus cornuta 4 5.6 1 35 101 39 Gaultheria shallon 30 26.9 1 95 95 37 Quercus vaccinifolia 18 21.6 1 85 71 28 Symphoricarpos mollis 2 2.6 1 20 63 24 Rhododendron macrophyllum 21 21.9 1 80 61 24 Quercus sadleriana 10 14.6 1 80 56 22 Acer circinatum 14 19.6 1 96 55 21 Vaccinium ovatum 33 27.1 1 90 48 19 Paxistima myrsinites 2 3.7 1 25 47 18 Rubus parviflorus 3 6.6 1 40 34 13 Rhus diversiloba 2 2.5 1 12 34 13 Holodiscus discolor 3 2.9 1 15 34 13 Rhododendron occident. 11 11.9 1 45 24 9 Amelanchier alnifolia 2 1.6 1 7 24 9 Gaultheria ovatifolia 4 4.4 1 20 23 9 Lonicera hispidula 2 1.7 1 7 19 7 Acer glabrum var. torreyi 3 2.9 1 10 17 6 Vaccinium membranaceum 9 17.6 1 65 16 6 Rubus leucodermis 1 1.0 1 5 16 3 Arctostaphylos nevadensis 3 2.2 1 8 13 5

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 115 VEGETATION ECOLOGY

Species Cover Std. Dev. Min. Max. N Const.

Herbs: Chimaphila umbellate 4 5.2 1 40 157 61 Goodyera oblongifolia 1 0.1 1 2 152 59 Achyls triphylla 6 8.3 1 50 143 55 Linnaea borealis 10 14.9 1 80 141 55 Whipplea modesta 5 8.2 1 45 136 53 Polystichum munitum 3 5.1 1 40 133 52 Trientalis latifolia 1 0.6 1 4 110 43 Disporum hookeri 1 1.5 1 15 101 39 Pyrola picta 1 0.1 1 2 91 35 Pteridium aquilinum 2 3.3 1 30 85 33 Xerophyllum tenax 3 5.9 1 35 75 29 Iris sp. 1 0.5 1 3 71 28 Smilacina racemosa 1 0.3 1 3 69 27 Vancouveria hexandra 3 7.1 1 50 63 24 Viola sempervirens 2 2.3 1 15 60 23 Chimaphila menziesii 1 0.5 1 4 56 22 Trillium ovatum 1 0.3 1 2 55 21

Table 6—Stand age comparison for plots with tree and shrub form yew and stands without Pacific yew.

Stand Age Class Frequency (%)

Category < 100 100-199 200-299 300-399 > 400 Tree Form Yew 0 4 24 44 32 Shrub Form Yew 1 16 30 24 29 Without Yew 10 32 30 18 10

All Plots 9 29 30 21 11

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 116 VEGETATION ECOLOGY

Table 7—Frequency comparison of sites with tree form, shrub form and without Pacific yew by Dunning site class (base age 300 years).

Frequency (%) Site Class Tree Form Shrub Form Without Yew 1A 35 23 23 1 23 28 27 2 23 24 24 3 14 10 14 4 3 14 8 5 2 1 4

Table 8—Stand table of Pacific yew trees/acre by size class and stand age category with 95% confidence interval (CI).

Pacific yew trees/acre (95% Cl) Age Category N < 2” 2-5.9” 6-10.9” 11-17.9” < 100 1 10 100-199 28 27 (1.6) 60 (10.9) 200-299 71 49 (9.2) 80 (16.9) 58 (57.5) 32 (4.5) 300-399 80 52 (8.8) 108 (18.1) 93 (22.4) 52 (9.1) > 400 78 67 (12.1) 124 (23.2) 108 (27.3) 25 (11.5)

Table 9—Summary table for significant discriminant variables.

Step Variable U-Statistic F-Statistic Degrees of Freedom 1 Stand Age 0.9405 144.057 1.0, 2278.0 2 Twinflower 0.9094 113.375 2.0, 2277.0 3 Moss 0.8836 99.933 3.0, 2276.0 4 Moisture Index 0.8707 84.486 4.0, 2275.0 5 Elevation 0.8587 74.841 5.0, 2274.0 6 Red Huckleberry 0.8490 67.362 6.0, 2273.0 7 Pacific Dogwood 0.8396 62.008 7.0, 2272.0 8 Micro-position 0.8325 57.006 8.0, 2271.0 9 Salal 0.8286 52.186 9.0, 2270.0 10 Vine Maple 0.8257 47.906 10.0, 2269.0 11 Chinquapin 0.8237 44.138 11.0, 2268.0 12 Sugar Pine 0.8219 40.941 12.0, 2267.0 13 Huckleberry Oak 0.8196 38.371 13.0, 2266.0 14 Jeffrey Pine 0.8170 36.246 14.0, 2265.0 15 Bigleaf Maple 0.8154 34.176 15.0, 2264.0

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 117

VEGETATION ECOLOGY

Table 10—Means for significant discriminant variables in the Pacific yew predictive model.

Variable With Yew Without Yew Total Micro-position 3.2 2.4 2.5 Elevation (ft.) 2968 3502 3442 Micro-relief Vertical 2.3 1.9 2.0 Horizontal 2.4 2.0 2.1 Stand Age (years) 356 246 258 Twinflower (%) 5.5 1.1 1.6 Moss (%) 10.5 3.4 4.2 Moisture Index 6.4 4.7 4.9 Red Huckleberry (%) 2.2 0.6 0.8 Pacific Dogwood (%) 1.7 0.5 0.7 Salal (%) 11.0 3.9 4.7 Vine Maple (%) 3.0 0.4 0.7 Big-leaf Maple (%) 1.8 0.7 0.8 Chinquapin (%) 4.4 2.6 2.8 Port-Orford-cedar (%) 9.7 2.8 3.6 Huckleberry Oak (%) 4.8 3.1 3.3 Sugar Pine (%) 2.6 3.4 3.3 Evergreen Huckleberry (%) 6.2 2.2 2.7 Total Shrub (%) 39.5 23.0 24.8 Rhododendron (%) 5.0 2.4 2.7 Jeffrey Pine (%) 0.1 1.0 1.0

Table 11. Classification matrix for Pacific yew presence or absence using discriminant analysis.

Percent Number of cases classified Group Correct With Yew Without Yew With Yew 72.8 187 70 Without Yew 80.9 386 1637 Total 80.0 573 1707

Table 12. Jackknifed classification matrix for Pacific yew presence or absence using discriminant analysis.

Percent Number of cases classified Group Correct With Yew Without Yew With Yew 71.2 183 74 Without Yew 80.7 391 1632 Total 79.6 574 1706

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 118 VEGETATION ECOLOGY

WOODY PLANT DISTRIBUTIONS IN WESTERN OREGON RIPARIAN FORESTS: INSIGHTS FOR RESTORATION AND MANAGEMENT

Daniel A. Sarr Klamath Network-National Park Service, 1512 E. Main Street, Ashland, OR 97520 Email: [email protected]

David E. Hibbs Department of Forest Science, Oregon State University, Corvallis, OR 97331 Email: [email protected]

Abstract The Klamath-Siskiyou Region presents a challenge to riparian restoration due to its diverse species composition and steep environmental gradients. Here, we present data on distributions of native woody plant species collected in a riparian inventory extending from the eastern Siskiyou Mountains to the Coast Range of northwest Oregon. We collected data with a nested sampling design using 16 1-ha plots in each of four watersheds: (1) the Applegate Basin in the eastern Siskiyous, (2) the South Umpqua Basin of the southern Cascades, (3) the McKenzie Basin of the western Cascades, and (4) the Alsea Basin of the western Coast Range. Within each of the 64 hectares, we collected vegetation and environmental information in 18 (40 m2) sampling plots arrayed along topographic transects. We analyzed species presence/absence data for the entire dataset (1152 (40 m2) samples) to determine weighted averages of environmental variables where a species occurred, and the range (standard deviation) in these values. Species sorted independently along the gradients measured with species showing associations with specific climatic, topographic, and disturbance-associated settings. Some generalist species, such as Pseudotsuga menziesii (Douglas-fir), occurred across most of the gradient. Other species, such as Picea sitchensis (Sitka spruce) and Pinus ponderosa (ponderosa pine) were restricted to wet and dry climate areas, respectively. Still other species, such as Salix sitchensis (Sitka willow) showed relatively broad distributions on the climate gradient, but strong association with fluvially disturbed habitats. These data are presented to provide insights into the distributions of native riparian species and to inform riparian restoration planning in the Klamath-Siskiyou Region. Introduction Methods

lthough ecological restoration of riparian forests is a Study Area common activity for volunteer and professional groups in In the summer of 1999, we sampled riparian vegetation in AAthe Klamath-Siskiyou Region, relatively little quantitative four watersheds (hereafter termed study basins) of western Oregon information is available describing species distributions in native chosen to characterize the steep regional climate gradient: (1) the riparian forests of the region. Studies in the Pacific Northwest Applegate in the eastern Siskiyou Mountains; (2) the South and elsewhere have demonstrated that riparian vegetation Umpqua in the southern Cascades; (3) the McKenzie in the central composition diverges from slope forests towards more frequently western Cascades, and (4) the Alsea in the western Coast Range. flooded streamsides and from headwater streams towards larger Sampling was conducted in low to moderate elevation areas of rivers (Vannote et al. 1980; Hupp 1986; Gregory et al. 1991; roughly similar mean annual temperature (9-10.5 ∞ Celsius), but Mouw and Alaback 2002), presumably due to variation in light, with mean annual precipitation ranging from ca. 600 mm in the flood duration, and soil characteristics along these gradients. A eastern Applegate Basin to over 2500 mm in the western Alsea corresponding understanding of the regional variation in riparian Basin. Summer climate accentuates the differences in water vegetation is lacking for the Klamath-Siskiyou Region and, balance with warm, dry conditions prevailing in the southern indeed, for most regions, although studies elsewhere suggest it can interior and cool, moist, and foggy conditions characterizing the be substantial (Collins et al. 1981; Alcaraz et al. 1997). coastal mountains (Western Regional Climate Center 2003). In this paper, we address this need for regional scale information by providing species distribution data from a study of Riparian Inventory four major watersheds in western Oregon that span a steep climate In 1999, we sampled vegetation and environmental gradient from the relatively warm and xeric southern interior to characteristics at sixteen streams randomly selected from a the comparatively cool and mesic northwest coast. Our objective population of 1st though 5th order stream reaches in each of the is to provide quantitative information about the distributions of four study basins. Stream order was determined for each stream woody riparian plant species along measured environmental at the sample reach from USGS 1:24,000 topographic quadrangle gradients to inform riparian restoration. maps using the Strahler method (Strahler 1957). For each basin,

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 119 VEGETATION ECOLOGY we interviewed staff at the appropriate land management agency estimates of the breadth of species distributions across the (USDA Forest Service or USDI Bureau of Land Management) to measured environmental gradients. Selected environmental determine management histories of each potential sample reach. variable were chosen to develop ordination diagrams illustrating We confined field sampling to reaches with no timber harvest or species distributions. We developed two sets of direct ordinations other management activities within the last 30 years and which for woody species occurring more than 5 times in the 1152 sample occurred within 3-4 km from an access road or trail (Figure 1). In plot dataset. each basin, this yielded 30-50 potential reaches from which First, we plotted all the common species in a bivariate space sixteen were selected at random for field sampling. All plots were defined by mean annual precipitation and height above stream. placed at least 50-200 meters upstream or downstream from a This two dimensional space was selected to illustrate the road or trail crossing. Although the distribution of stream orders geographic and local habitat optima for a given species. We differed substantially among the study basins, most were 2nd or overlaid curves of predawn plant moisture stress (Sarr 3rd order streams. unpublished data) upon the direct ordination diagrams to link the At each sample reach, we established a one-hectare (200 m geographic and topographic variation to a physiologically relevant x 50 m) plot that straddled the stream, but did not include the metric (Waring and Cleary 1967; Waring 1969; Hawk and Zobel stream channel. All distances were slope distances measured 1974; Zobel et al. 1976; Giordano and Hibbs 1993). Plant along the ground surface with a fiberglass tape. moisture stress was measured with a pressure chamber on live On each side of the stream we established six 25 x 33.3 m conifer seedlings under 2 m tall arrayed across a topographic subplots (thus 12 total in each hectare plot) with the long side gradient from streamside to hillslope (see Waring and Cleary parallel to the stream. The first 24 m of each subplot was divided 1967, Waring 1969 for methodology). The portrayal of predawn to produce three contiguous 8 x 33.3 m sampling cells at plant moisture stress in the diagrams illustrated both the steepness increasing distances from the stream edge. One of each of the six of the transriparian moisture gradient and the actual physiological pairs of opposing subplots was randomly selected for sampling of stress at a given height above the stream in different climates of vegetation and environmental characteristics. Data were collected western Oregon. Since the predawn plant moisture stress curves using 4 x 10 m (40 m2) sample plots arrayed with the long side were derived from measurements at only a single stream in each parallel to the stream. All three sampling plots within a subplot of the four basins, they should be interpreted as approximations. were placed the same random distance from 0 to 23 m into the Second, we ordinated understory species in a bivariate long dimension of the subplot and from 0 to 4 m into the short space defined by canopy cover and soil litter depth. This second dimension of each sampling cell (Figure 1). set of ordinations was intended to illustrate species distributions Vegetation and environmental characteristics were measured along a disturbance/succession gradient that we observed in most within each 40 m2 sampling plot. We determined the presence of of our sites. This gradient ranged from open, skeletal areas where all woody species and estimated total percent cover of each signs of fluvial activity were frequent, to densely forested areas species in the plot. Ground distance from the plot center to the with well-developed soils and little evidence of disturbance. Tall water’s edge at the time of sampling was recorded, and slope in shrubs were ordinated with overstory cover and litter depth, degrees from the plot center to the stream edge was estimated whereas low shrubs, subshrubs, and ferns were ordinated with using a clinometer held 1 m above the ground. Slope data were total foliar cover (overstory cover + additional projected foliar used to compute height above stream. Overstory cover was an cover of tall shrubs) and litter depth. ocular estimate of tree cover straight up from a point 1.75 m above the ground viewed through a 20 cm diameter ring held Results vertically at arms length (ca. 45 cm). Total cover was estimated by applying the same method with the observer seated and When species weighted means were plotted in a two included any shrub layers greater than 1.2 m above the ground, in dimensional space of mean annual precipitation and height above addition to overstory cover. the stream, species showed varied distributions based on their Soil characteristics were determined by digging a 30 cm soil moisture requirements and flood tolerance (Figure 2a-d). In pit in the middle of each 40 m2 sample plot. Litter depth, and, particular, it is apparent that some species vary considerably in where a bedrock layer was encountered, depth to bedrock were both their mean positions and the amplitude of their distributions measured in centimeters. In all, eighteen plots were sampled across the two gradients. Field measurements also suggest that within each of the 16 sites for a total of 288 samples per basin variation in plant moisture stress across the topographic gradient (1152 sample plots overall). is greatest in the driest climates and that sites with low moisture stress extend to greater heights above the stream in wet climates. Data Analysis From our entire dataset of 1152 sampling plots, we Trees calculated weighted means of selected variables (i.e., the mean Conifers were well distributed across the climate gradient value of that variable for all plots where the species occurred) to (Figure 2a), with Picea sitchensis (Sitka spruce) occurring only in directly ordinate species along environmental gradients. We also the wettest climates (mean precipitation = 211.8 cm ± 10.8 cm) calculated standard deviations in the environmental variables as (weighted mean ± s.d.) and Pinus ponderosa (ponderosa pine)

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 120 VEGETATION ECOLOGY occurring in the driest climates (Precipitation = 94.7 cm ± 17.3 (Figure 2c). cm). Pseudotsuga menziesii (Douglas-fir) showed the broadest Tall shrubs also ranged along gradients in overstory cover climatic amplitude (as indicated by standard deviation in weighted and litter depth, but there was broad overlap in standard deviations mean precipitation) of all the conifer species (Precipitation = 135.4 of both factors for most species. Weighted means of species along cm ± 44.0 cm). Most conifer species occurred > 3 m above the axes of overstory cover and litter depth were positively correlated stream, with the exception of Chamaecyparis lawsoniana (Port (Figure 3a), suggesting species sort meaningfully on a complex Orford cedar), which primarily occurred on the floodplain of the gradient from shaded forest sites with well developed soils to open only creek where we encountered it (Height Above Stream = 1.7 sites with more skeletal substrates. This gradient of conditions m ± 0.6 m) (Figure 2a). describes the effects of fluvial or other soil disrupting disturbances Hardwood species weighted means were also well in riparian forests, but it also describes the changes from forested distributed across the climate gradient, but showed greater areas with deep soils to rocky outcrop areas with sparser canopies, topographic differentiation than conifers (Figure 2b). Alnus rubra which were fairly common in the riparian forests of the South (red alder) showed the highest weighted mean for precipitation and Umpqua and Applegate basins. Species with lowest weighted broad climatic amplitude (Precipitation = 174.4 cm ± 41.1 cm) means for both litter depth and overstory cover included the whereas the two deciduous oaks, Quercus garryana (Oregon white riparian dependent S. sitchensis (Sitka willow), as well as the more oak) (Precipitation = 85.6 ± 8.7 cm) and Quercus kelloggii xerophytic species associated with rocky areas, including (California black oak) (Precipitation = 86.5 cm ± 9.9 cm), were Ceanothus sanguinus (redstem ceanothus), and Ribes roezlii restricted to the driest environments sampled. Populus (Sierra gooseberry) (Figure 3a). Species associated with greater trichocarpa (black cottonwood) showed the lowest weighted mean overstory cover and litter depth included Rhododendron height above stream of all the species (1.0 m ± 0.5 m), followed macrophyllum (Pacific rhododendron), Gaultheria shallon (salal), by A. rhombifolia (white alder) (1.5 m ± 1.6 m), and A. rubra and Vaccinium membranaceum (Figure 3a). (1.9 m ± 2.3 m). Alnus rubra and A. rhombifolia occupied roughly similar microelevations, but the latter species only Low Shrubs, Subshrubs, and Ferns occurred in the drier climate areas. The broadleaved evergreen This group also showed considerable variation across species Castanopsis chrysophylla (Dougl.) DC. (chinquapin), gradients of climate and microelevation (Figure 2d). Species with Quercus chrysolepis (canyon live oak), and A. menziesii (Pacific highest weighted means for precipitation included the fern species madrone) occupied high topographic positions (Height Above Athyrium filix-femina (lady fern) and Dryopteris austraica (wood Stream > 5.4 m) in moderately dry climates (Precipitation < 140 fern). Polystichum lonchitus (holly fern) and Polystichum cm). imbricans (imbricate sword fern) occurred in the driest climates, Most of the conifers appeared to occur in plots with with the latter occurring in rocky areas at high elevations above moderate litter depth (Appendix 1a.). P. sitchensis showed the streams (Height Above Stream = 8.7 m ± 4.1 m). A. filix-femina lowest mean depth (Litter Depth = 3.4 cm ± 3.8 cm) and C. occurred at the lowest microelevations, followed by D. austraica lawsoniana showed the highest weighted mean (Litter Depth = (Figure 2d). 17.3 cm ± 12.9 cm) (Appendix 1a). Hardwoods showed greater Along axes of total cover and litter depth (Figure 3b), P. variation in weighted means. P. trichocarpa had the lowest imbricans showed lowest weighted means for both factors (Litter weighted mean litter depth (Litter Depth = 1.7 cm, ± 2.8 cm), Depth = 5.8 cm ± 5.6cm, Total Cover = 58.8% ± 31.3%), whereas A. glabrum (mountain maple) (Litter Depth = 12.8 ± 9.3) reflecting its preference for open rocky areas. Highest values for had the highest mean. total cover and litter depth were recorded for the evergreen subshrubs Rubus nivalis (dwarf bramble), Chimaphila menziesii Tall Shrubs (little prince’s pine), and Berberis nervosa (dwarf Oregon grape) Weighted means for tall shrubs also sorted across gradients (Figure 3b). of climate and microelevation (Figure 2c). Species with highest Weighted means and standard deviations of selected weighted means on the precipitation axis included Menziesia environmental variables for the 40 most commonly occurring ferruginea (false huckleberry), Sambucus racemosa (red woody plant species in the Applegate Basin (the only study basin elderberry), and Rubus spectabilis (salmonberry) (Figure 2c). entirely within the Klamath-Siskiyou Region) are included in Toxicodendron diversilobum (poison oak), snowberry Appendix 1. Weighted means of remaining species may be Symphoricarpos albus (snowberry), and Holodiscus discolor obtained from the senior author upon request. (oceanspray) occurred in the driest climates and at relatively high microelevations (Figure 2c). Salix sitchensis (Sitka willow) showed relatively broad climatic amplitude, but was restricted to Discussion the lowest microelevations (Height Above Stream = 0.8 m ± 0.6 m). Physocarpus capitatus (ninebark) and Cornus sericea ssp. Although this study extended from the eastern Siskiyou occidentalis (red-osier dogwood) occurred near streams in Mountains to the northern Oregon Coast Range, we propose that relatively dry climates, whereas Ribes bracteosum (skunk currant) the quantitative relationships developed here might also apply, at occupied the lowest microelevations in the wettest climates least in part, for species distributions across the similarly steep

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 121

VEGETATION ECOLOGY east-to-west climate gradient in the northern Klamath-Siskiyou A preliminary step before beginning a restoration project Region. Certainly, many species are common to both regions. might be to determine the annual precipitation of the site from Going from the dry interior to the wet coast, we would expect a published maps or on-line sources. With this information, a general change from relatively stress tolerant species such as pines reader could use Figures 2a-d to assemble a potential list of and oaks to fir, hemlock, and spruce (see also Whittaker 1960, species for the site. Using Figures 2a-d, one could trace a path Waring 1969). A more subtle, but equally important shift in from streamside up to 10 m above the stream on each of the species along streamsides (e.g. Alnus rhombifolia (white alder) to figures to develop a list of potential woody plant species for a Alnus rubra (red alder)) would likely also occur. Riparian forests restoration project. would also grow taller, darker, with a milder moisture gradient Within a given site, one would next look for more fine scale from streamsides to hillslopes. As Figures 2a-d suggest, such environmental variation due to canopy cover, litter depth, and soil climatic variation would likely to be associated with substantial texture, drainage or other factors to refine placement of species. changes in woody riparian vegetation in both streamside and Although a common restoration planning approach is to locate and higher topographic positions survey a reference riparian site that is as similar as possible to the Within the riparian zone, species respond individualistically damaged area and record local distributions of the species seen, to environmental variation. Species varied both in their mean that option may not always be available. Therefore, we refer the position on an environmental gradient but also differed greatly in reader to Figures 3a-b, and Appendix 1 for additional information their range on any particular gradient. Because most of the about where native riparian species occur along the environmental species encountered in this study are important riparian gradients measured in this field study. It is important to mention components throughout the Klamath-Siskiyou Province, a review that the distributions of species reported in this paper comprise all of some examples of this phenomenon have direct application to ages of individuals for each species and the natural distributions this region. recorded may vary from the optimal conditions for juvenile plants. We found that some species show high specialization for Moreover, riparian plantings, especially container stock, may have particular climate zones Picea sitchensis (Sitka spruce), Menziesia slightly different site requirements. It is also possible that ferruginea (fools huckleberry)) and relatively broad topographic severely disturbed sites may present unique establishment distributions while others are linked to specific streamside conditions (e.g., soil compaction, toxicity) that were not measured environments (Populus trichocarpa (black cottonwood)) but are in these natural riparian forests. Nonetheless, these distributions widely distributed across the climate gradient. Still other species from a large sample of riparian sites probably contain important appear to be associated with different degrees of canopy cover and heretofore unpublished information about the natural habitat soil depth. Evergreen species such as Rhododendron associations of our woody riparian species. macrophyllum (rhododendron), Gaultheria shallon (salal), and From our field observations, we can add an additional Vaccinium spp. (huckleberry) were most common in forested areas insight for a restorationist or manager in the Klamath-Siskiyou with well developed soils, whereas other species were most Region - the notion of life history strategies. Grime (1977) argued common in open disturbed environments with skeletal soils, that plants evolve toward three major adaptive strategies: (1) a usually near streamsides (e.g., Salix sitchensis (Sitka willow)). competitive strategy, which combines high growth rates, large Rocky outcrop areas far from the stream edge also had distinctive sizes, high shoot to root ratios, shade tolerance, and often species (e.g., Polystichum imbricans (imbricate sword fern). longevity to dominate resource rich sites; (2) a stress tolerant Gaining both a conceptual and quantitative grasp of this habitat strategy, involving slow growth, low shoot to root ratios, and complexity is the first task of a restoration practitioner or evergreenness, evolved on relatively resource poor habitats, and manager. One needs to understand not just where on a resource, (3) a ruderal strategy, characterized by light seeds, shade- climate, or hydrologic gradient a species might occur, but how intolerance, and rapid juvenile growth, specialized for the sensitive the species is to that gradient. ephemeral establishment opportunities created by disturbance at Across the east to west climate gradient of the Klamath- Siskiyou Region, drought and competition are likely to be first resource rich sites. considerations for the survival of planted riparian vegetation at These three strategies appear to be well represented in dry and wet ends, respectively. In the interior regions, where riparian forests of the region. Near streams, where disturbance is summer drought is severe, planted species should be placed with frequent, light and moisture are abundant, and soils are skeletal, careful consideration to aspect and height above stream to prevent ruderal genera, such as Alnus (alder), Populus (cottonwood), and dessication. Species that naturally occur in drier climates should Salix (willow) predominate. On floodplains with fertile soils and form an important part of any planting mixture. Approaching the less frequent disturbance, and in mesic climates generally, coast, moisture stress decreases, and position along the competitive species such as, Tsuga (hemlock), Thuja (cedar), and topographic moisture gradient is probably less critical (though Abies (fir) dominate. Sequoia sempervirens (redwood) probably light availability or flooding may still drive some topographic also fits in this group, and alluvial redwood stands are among the segregation of species). Vigorous vegetation growth in coastal most productive, massive, and long-lived forests in the world. At climates probably limits the establishment success of slow somewhat drier sites, where drought stress becomes more growing or less shade tolerant species. important, stress tolerant genera such as Pinus (pine), Quercus

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 122 VEGETATION ECOLOGY

(oak), or Arbutus (madrone) and various xerophytic shrubs are Collins, S.L, P.G. Risser, and E.L Rice. 1981. Ordination and important elements of the riparian flora. It seems prudent to classification of mature bottomland forests in northcentral consider species with all three strategies when selecting species Oklahoma. Bulletin of the Torrey Botanical Club for restoration at a site at all except mild coastal sites, where stress 108(2):152-165. tolerators would likely compete poorly. Ruderal species are well Gregory, S.V., F.J. Swanson, W.A. McKee, and K.W. Cummins. adapted to streamsides throughout the region. Competitive 1991. An ecosystem perspective of riparian zones. species are best adapted to floodplains where they can grow to BioScience 41(8): 540-551. great sizes and yield future benefits of shading and large woody Grime J.P. 1977. Evidence of the existence of three primary debris. Stress tolerators are ideally adapted to harsher sites in the strategies in plants and its relevance to ecological and interior, especially south facing slopes, or dry, rocky areas, where evolutionary theory. American Naturalist 111:1169-1194. they ameliorate conditions and facilitate establishment of less Giordano. P., and D.E. Hibbs. 1993. Morphological response to hardy species. competition in red alder: The role of water. Functional Moreover, vegetation communities are not stable. The Ecology 7:462-468. selection and placement of riparian plant species, therefore, must Hawk, G.M,, and D.B. Zobel. 1974. Forest succession on alluvial anticipate the temporal changes in abiotic and biotic conditions at landforms of the Mackenzie River Valley, Oregon. a site. Thus, including species that are resilient after disturbance Northwest Science 48(4):245-265. (ruderal or stress tolerant species at wet or dry sites, respectively) Hupp, C.R. 1986. Upstream variation in bottomland vegetation provides for recovery in the open, potentially stressful conditions patterns, northwestern Virginia. Bulletin of the Torrey of early succession. Including shade tolerant or long-lived Botanical Club 113(4):421-430. (competitor) species will help establish and maintain a desired Mouw, J.E.B., and P.A. Alaback. 2002. Putting floodplain vegetative cover in the future, when competitive pressures become hyperdiversity in a regional context: an assessment of most important. Joint consideration of short-term goals, such as terrestrial-floodplain connectivity in a montane site stabilization or amelioration of microclimate, and long-term environment. Journal of Biogeography 30:87-103. goals, such as stream shading and large wood debris recruitment, Sarr, D.A. In Prep. Multiscale controls on woody riparian is desirable. vegetation: Distribution, diversity, and tree regeneration in four western Oregon watersheds. Ph.D. Thesis. Department The quantitative relationships and insights in this paper of Forest Science, Oregon State University, Corvallis. provide just a glimpse into the complexity of riparian forests in Strahler, A.N. 1957. Quantitative analysis of watershed the Klamath-Siskiyou Region. A diversity of climates, geologic geomorphology. American Geophysical Union Transactions conditions, and species create a challenging, but rewarding 38:913-920. template for riparian restoration and management. We encourage Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell, and further research and communication to develop a sound basis for C.E. Cushing. 1980. ‘The river continuum concept’. conserving, restoring, and sustainably managing these globally Canadian Journal of Fisheries and Aquatic Science 37:130- important ecosystems. 137. Acknowledgements Waring, R.H. 1969. Forest plants of the eastern Siskiyous: their environmental and vegetational distribution. Northwest Science 43:1-17. This work was funded by the Cooperative Forest Ecosystem Waring, R.H., and B.D. Cleary. 1967. Plant moisture stress: Research (CFER) Program and Shaubert and Tarrant Fellowships evaluation by pressure bomb. Science 155:1248-1254. to the senior author from Oregon State University. Rick Momsen Western Regional Climate Center. 2003. Online historical data and Ronasit Maneesai assisted in field data collection. An earlier summaries for Oregon cities 1971-2000 (accessed at draft of this manuscript was greatly improved by review http://www.wrcc.dri.edu). comments from Jack Williams and Michael Parker of Southern Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, Oregon University. Oregon and California. Ecological Monographs 30:279- 338. Literature Cited Zobel, D.B., A. McKee, and G. Hawk. 1976. Relationships of environment to composition, structure, and diversity of Alcaraz, F., S. Rios, C, Inocencio, and A. Robledo. 1997. forest communities of the central western Cascades of Variation in the riparian landscape of the Segura River Oregon. Ecological Monographs 46:135-156. Basin, SE Spain. Journal of Vegetation Science 8:597-600.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 123 VEGETATION ECOLOGY

Figure 1. Diagram illustrating the location of: (a) the hectare plot relative to an access road and sample stream, and (b) the arrangement of sampling cells, and 40 m2 sampling plot, within a hectare.

25 m Hectare Plot

B

Str eam

2 eam 40m

str sample plot 50+ m 200 m Access Road

Sampling cells A ot 8 m 33.3 m Subpl

250m2 sample belt

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 124 VEGETATION ECOLOGY

Figure 2 a-d. Scatterplot of weighted species means (± s.d.) in a direct gradient space defined by mean annual precipitation and height above stream. Contours of predawn plant xylem potential stress are overlaid based on empirical measurements at four locations along the climate gradient in September. (a) conifer trees; (b) hardwood trees; (c) tall shrubs; (d) low shrub, subshrubs, and ferns. The contours represent preliminary relationships for interpretive purposes only.

Figure 3 a-b. (a) Scatterplot of weighted species means (± s.d.) for common tall shrub species in a direct gradient space defined by mean litter depth and overstory cover. (b) Scatterplot of weighted species means (± s.d.) for common low shrub, subshrub and fern species in a direct gradient space defined by mean litter depth precipitation and total foliar cover. Note a positive relationship between weighted means for litter depth and the two cover indices.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 125 VEGETATION ECOLOGY

Appendix 1. Table of weighted means and standard deviations for selected environmental variables for most common tree and shrub species occurring in the Applegate basin, southwest Oregon: (a) 15 most frequently encountered tree species; (b) 25 most frequently encountered shrub species. Note that the number of occurrences (n) used to calculate means (out of 1152 possible) is provided at left of table and refers to frequency for entire four-basin dataset.

SCIENTIFIC NAME COMMOM NAME n

Trees

Abies grandis (Dougl.) Forbes grand fir 146 Acer macrophyllum Pursh. bigleaf maple 306 Alnus rhombifolia Nutt. white alder 24 Alnus rubra Bong. red alder 234 Arbutus menziesii Pursh. Pacific madrone 25 Calocedrus decurrens (Torr.) Florin. incense cedar 84 Castanopsis chrysophylla (Dougl.) DC. chinquapin 74 Cornus nuttallii Aud. Pacific dogwood 64 Fraxinus latifolia Benth. Oregon ash 22 Pinus lambertiana Dougl. sugar pine 12 Pseudotsuga menziesii (Mirbel) Franco Douglas-fir 431 Quercus chrysolepis Liebm. canyon live oak 116 Quercus garryana Dougl. Oregon white oak 25 Quercus kelloggii Newberry California black oak 30 Taxus brevifolia Nutt. Pacific yew 186 Tall Shrubs, Low shrubs, and Ferns

Acer circinatum Pursh vine maple 442 Amelanchier alnifolia Nutt. Saskatoon serviceberry 71 Berberis aquifolium Pursh tall Oregon grape 112 Berberis nervosa Pursh low Oregon grape 452 Chimaphila menziesii (R. Br.) Spreng prince's pine 13 Corylus cornuta var. californica Marsh California hazelnut 318 Gaultheria shallon Pursh salal 214 Holodiscus discolor (Pursh.) Maxim. ocean spray 89 Linnea borealis L. twinflower 256 Lonicera ciliosa (Pursh) DC. orange honeysuckle 65 Lonicera hispidula (Lindl.) Dougl. California honeysuckle 154 Pachistima myrsinites (Pursh) Raf. Oregon boxwood 76 Philadelphus lewisii Pursh mockorange 103 Polystichum lonchitis (L.) Roth. holly fern 20 Polystichum munitum (Kaulf.) Presl sword fern 842 Physocarpus capitatus (Pursh) Kuntze ninebark 34 Rhamnus purshiana DC. cascara 156 Rosa gymnocarpa Nutt. bald hip rose 220 Rubus parviflorus Nutt. thimbleberry 143 Rubus ursinus Cham. & Schlecht. trailing blackberry 628 Satureja dougasii (Benth.) Briq. yerba buena 26 Symphoricarpos albus (L.) Blake snowberry 215 Symphoricarpos mollis Nutt. creeping snowberry 24 Toxicodendron diversilobum (Torr. & Gray) Fosberg poison oak 140 Whipplea modesta Torr. whipplevine 132

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 126 VEGETATION ECOLOGY

Height Above Overstory Cover Mean Annual Conifer Cover Litter Depth (cm) Stream (m) (%) Precipitation (cm) mean s.d. mean s.d. mean s.d. mean s.d. mean s.d.

4.3 3.2 66.4 23.6 68.6 34.4 10.0 7.7 126.0 19.2 3.0 3.0 65.6 27.5 43.8 39.6 9.0 9.3 133.9 42.3 1.5 1.6 58.3 23.4 31.5 30.7 8.3 10.1 105.6 11.8 1.9 2.3 59.0 28.8 30.2 37.4 5.8 8.6 174.4 41.1 6.1 3.9 63.6 19.5 67.0 40.9 9.2 9.3 96.3 14.0 5.0 3.7 65.7 24.3 66.7 37.4 8.3 7.7 126.3 41.5 5.9 3.5 72.3 23.1 69.7 36.1 11.8 8.6 139.9 33.6 4.6 3.4 76.7 18.7 65.9 37.3 11.1 9.8 145.0 36.1 2.2 3.1 61.1 24.4 49.3 38.0 9.0 8.4 102.7 34.6 7.4 5.0 62.5 27.2 65.4 36.3 9.0 9.9 118.2 24.2 4.3 3.6 67.7 24.2 65.8 36.9 9.3 10.3 135.4 44.0 5.4 3.9 59.1 27.6 53.6 38.0 8.1 8.1 104.4 9.0 5.1 3.0 66.6 24.4 60.8 35.6 7.0 5.7 85.6 8.7 5.4 3.7 69.7 17.8 55.7 35.7 6.7 5.6 86.5 9.9 4.2 3.5 70.3 23.8 72.7 34.3 9.6 8.1 138.0 34.4

3.7 3.2 64.7 25.8 65.0 39.4 9.1 10.1 154.5 38.3 5.0 3.9 60.0 24.6 64.6 36.3 8.7 8.7 115.0 32.1 4.8 3.5 66.4 21.2 69.0 34.4 6.9 5.8 105.1 30.4 4.6 3.4 70.2 21.8 72.6 33.4 10.6 8.9 144.5 39.5 7.4 3.4 78.7 13.3 81.5 23.8 11.2 9.7 153.1 34.7 4.3 3.2 66.4 22.6 64.7 36.4 10.1 8.4 137.5 45.4 4.1 3.3 72.4 22.1 66.8 36.6 10.5 9.3 165.5 37.2 5.8 4.0 58.7 26.5 63.0 38.1 6.7 6.6 110.3 27.0 4.6 3.5 68.4 23.6 81.0 28.4 11.0 11.5 143.1 33.6 4.8 4.1 61.6 26.6 66.2 35.8 7.8 7.4 123.5 36.9 5.2 3.9 64.3 22.6 70.6 32.4 7.4 8.1 109.9 36.1 5.4 3.5 68.1 20.6 70.9 35.2 9.0 8.2 125.0 37.8 4.2 3.8 64.3 25.5 56.2 37.5 7.0 7.1 112.8 37.6 3.9 3.7 64.5 23.1 50.5 35.1 9.4 6.5 92.3 9.9 3.8 3.4 63.9 26.5 54.3 41.3 8.5 9.2 164.6 44.6 1.5 1.6 57.5 29.3 45.0 41.4 5.0 6.9 123.4 32.6 4.5 3.6 65.4 25.3 66.0 39.3 8.8 8.8 170.8 44.8 4.9 3.5 64.4 23.1 70.3 35.6 9.3 8.5 121.9 36.7 3.5 3.4 60.4 29.2 45.6 41.7 7.0 7.9 151.8 43.8 3.5 3.1 64.4 26.1 60.4 39.5 9.1 10.0 143.5 40.5 3.2 2.6 70.4 23.8 61.0 34.4 8.3 8.3 113.4 41.8 3.9 3.4 62.2 24.7 57.3 37.5 8.3 8.4 103.1 24.5 4.0 4.0 65.1 22.9 61.7 39.6 13.2 11.7 134.6 48.5 5.3 3.9 62.3 25.6 62.2 38.2 7.7 6.7 95.7 13.8 6.1 3.6 64.2 24.7 92.9 18.6 8.6 9.7 146.1 32.6

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 127 VEGETATION ECOLOGY

CONIFERS OF THE KLAMATH MOUNTAINS

John O. Sawyer Department of Biological Sciences, Humboldt State University, Arcata, CA 95521 Email: [email protected]

Abstract

Botanists have long celebrated the extraordinary conifer richness of the Klamath Mountains, but the number of species that grow in the region remains an enigma. The Klamath Mountains are one of the most exceptional temperate coniferous forest regions in the world at many different levels. An explanation of its conifer richness involves not only its paleoendemics and widespread relicts, but its neoendemics. The difficulty in identifications suggests that modern conifer populations have continued to evolve and respond to the current environmental conditions. Localized areas with an unusually high number of conifer taxa (species, subspecies, and varieties) are found where seed sources are locally available and substrates heterogeneous. Introduction most part, distinctive climates, vegetation, soils, water, and other resources” (Hunt 1967). The Klamath Mountains is an area of distinctive Paleozoic he Klamath Mountains of southwestern Oregon and and Mesozoic basement rocks surrounded by rocks of younger northwestern California comprise one of the richest areas age. The western boundary of the province is represented by the of conifers in the world. Using the definition of the TT South Fork fault and the eastern boundary by the Cenozoic Klamath Mountains by Irwin (1966), I consider that the regional volcanic rocks of the Cascades (Irwin 1966). The Coast Ranges to number to be 35 (Table 1). This region compares favorably with the west are differentiated from the geomorphic province mainly others known for their high conifer richness. Central China has by rocks of the Franciscan formation. 50, Mexico’s Sierra Madre Occidental 64 and Sierra Madre McKee (1972) included younger rocks, the Dothan and Oriental 43 (Ricketts et al. 1999). What is surprising is the Hornbrook formations, and associated rock types, in his treatment difficulty in identifying these conifers. Depending on how a of the Klamath Mountains of Oregon. Rocks of the Dothan person classifies the region and its trees, the conifer richness in formation are equivalent to those of the Franciscan formation in southwestern Oregon and northwestern California can vary from California. Coleman and Kruckeberg (1999) argued for an area 33 to 42 taxa. larger than that proposed by Irwin. The geomorphic province This richness is further highlighted by localized areas that should include “all the older rocks” and the “accreting dynamic have an usually high number of taxa, even by regional standards. continental margin” rocks. They did not present a map or Forests near Bear Basin Butte in the Siskiyou Mountains contain boundaries for a new area. 16 conifer species: Alaska yellow cedar (Xanthocyparis Conservation biologists used these expanded concepts in nootkatensis), Brewer spruce (Picea breweriana), common juniper defining a Klamath-Siskiyou Ecoregion (Ricketts et al. 1999). I (Juniperus communis), Douglas fir (Pseudotsuga menziesii), found the differences in three of the maps in the theme issue of incense cedar (Calocedrus decurrens), Jeffrey pine (P. jeffreyi), the Natural Areas Journal (Williams 1999) — the one by Coleman Port Orford cedar (Chamaecyparis lawsoniana), knobcone pine and Kruckeberg (1999), that presents the Klamath Mountains (Pinus attenuata), lodgepole pine (P. contorta), mountain hemlock geomorphic province of Irwin; the one by Noss et al. (1999), that (Tsuga mertensiana), noble fir (Abies procera), ponderosa pine (P. considers an area for conservation; and the one by DellaSala et al. ponderosa), sugar pine (P. lambertiana), western white pine (P. (1999), that places the ecoregion in a global perspective — useful monticola), white fir (Abies concolor), and yew (Taxus brevifolia). in understanding why estimates of conifer richness vary. The Coleman and Kruckeberg map, like that of frontispiece Why the Variation in Number of Taxa? in Wallace’s (1983) book, The Klamath Knot, is very similar to that of Irwin’s geomorphic province. Its boundaries include the Authors have various concepts of the Klamath Mountains, least area of the three maps. The maps of Noss and DellaSala so it is not surprising that different conifer totals exist. Geologists expanded the ecoregion boundaries beyond the geomorphic have long considered the Klamath Mountains to be a province. The Noss map included parts of the Cascades and Coast physiographic province (Diller 1903), one of several that make up Ranges north of Redwood Creek in California. The DellaSala map the western cordillera of North America. While geologists was similar in Oregon to that of the geomorphic province, but in characterize provinces in geological terms, they can be considered California included lands west of the crest of the Coast Ranges as ecoregions with a “..... distinctive structural framework giving rise far south as Snow Mountain and as far west as the eastern range to distinctive landforms expressing their structure and, for the of Sequoia sempervirens. Much of the Eel River watershed was

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 128 VEGETATION ECOLOGY included, but lands immediately along the coast were excluded. Cupressus and Juniperus, the Cypresses and Junipers They were included in the Noss map. Identification of our cypresses and junipers is not as Biogeographers (Brown and Lomolino 1998) have long complicated, but the number of taxa is unresolved. Baker cypress argued that larger areas of comparable environmental (Cupressus bakeri) grows in the Siskiyou Mountains and in the heterogeneity should contain more species. As expected, the Cascades (Munz 1957; Peck 1961; Lanner 1999). Those in the conifer richness differs in the three areas according to size. The Siskiyous (C. bakeri var. matthewsii) have darker foliage and geomorphic province has 33 conifers; the larger (DellaSala map) larger cones whose scales lack an umbo (protuberance). Some has 35, and the largest (Noss map) has 38 (Table 1). The Noss populations in the Cascades have lighter foliage and smaller cones ecoregion includes California juniper (Juniperus californica) and whose scales have a conspicuous umbo (C. bakeri var. bakeri). excludes Sargent cypress (Cupressus sargentii), and the DellaSala Griffin (1993), Eckenwaler (1993), and Stuart and Sawyer (2001) ecoregion does just the opposite. Both ecoregions include did not recognize the variety, but in this treatment I will include mountain juniper (Juniperus occidentalis ssp. australis). The Noss C. bakeri var. matthewsii for this discussion (Table 1). boundaries include shore pine (Pinus contorta ssp. contorta) and Knowing the number of varieties of Juniperus communis western red cedar (Thuja plicata) that are lacking in the in the region is another problem. This circumboreal species has geomorphic province (Griffin and Critchfield 1972; Burns and been treated many ways by botanists. Farjon (1998) includes more Honkala 1990; J. Sawyer personal observations). than 25 names in synonymy. Adams (1993) recognized three varieties in North America, two of them occurring west of the Conifer Identification Rocky Mountains. Plants that form dense mats (J. c. var. montana) are found in Trinity Alps and in other mountain ranges Abies, the Firs that reach subalpine elevations, but plants at lower elevations on White fir taxa in southern Oregon and northwestern the Josephine ophiolite have long, trailing, ground-hugging California present several problems that are reviewed in the Flora branches (Juniperus communis var. jackii). Smith and Sawyer North America (Hunt 1993). The determination of silver fir (Abies (2003) recognize the taxon and I put in Table 1. amabilis) and subalpine fir (A. lasiocarpa) is straightforward. Grand fir (Abies grandis) grows along the coast and in adjoining Pinus, the Pines mountains where many trees are better considered as hybrids Identification of pines in southwestern Oregon and between A. concolor and A. grandis. Such trees are extensive in northwestern California offers its own problems. Washoe pine the mountains of southwestern Oregon at low elevations, less so in (Pinus washoensis) has been recently reported from the eastern northwestern California. Montane trees are most often considered Siskiyou Mountains (Callahan 2000). This species was first A. concolor (Munz 1959; Peck 1961; Griffin 1993; Stuart and described from the eastern Sierra Nevada, and the largest Sawyer 2001), or if assigned to lower rank, as A. concolor var. population is in the Warner Mountains. Genetic work suggests that lowiana (Lanner 1999). Hunt (1993) reports that trees of A. P. washoensis arose from gene exchange between the western P. concolor (A. concolor var. concolor), A. lowiana, and A. concolor ponderosa var. ponderosa and the eastern P. ponderosa var. x A. lowiana hybrids grow in southwestern Oregon and scopulorum that grows in the Rocky Mountains (Niebling and northwestern California. Depending on interpretation, Conkle 1990). Most botanists recognize the species (Munz 1959; southwestern Oregon and northwestern California has four to Griffin 1993; Kral 1993; Lanner 1999; Stuart and Sawyer 2001; seven white fir taxa. The jury is still on the varieties of A. Smith and Sawyer 2003), but Farjon (1998) considers it concolor, so I only listed Abies amabilis, A. concolor, A. grandis, conspecific with Pinus ponderosa var. ponderosa. I placed Pinus and A. lasiocarpa (Table 1). washoensis in Table 1. Identification of red fir taxa in the higher elevations is Variation in Pinus contorta populations suggests additional equally tenuous. Traditionally most montane and subalpine relationships with the Rocky Mountains. Two of the four populations were considered Shasta fir (Abies magnifica var. subspecies occur in southwestern Oregon and northwestern shastensis), while those in the Siskiyou Mountains were A. California (Critchfield 1957; Peck 1961). Pinus contorta ssp. procera (Munz 1959; Peck 1961; Griffin 1993; Lanner 1999; contorta grows along the coast. This subspecies shares the Stuart and Sawyer 2001). Hunt (1993) and Smith and Sawyer characteristic of having open cones at maturity with P. contorta (2003) considered high-elevation firs in southern Oregon and ssp. murrayana of the Sierra Nevada. Trees on the Josephine northern California as A. x shastensis, a hybrid between A. ophiolite have cones that are closed at maturity, a character they procera and A. magnifica. Parker (1963) considered that A. share with lodgepole pine (P. contorta ssp. latifolia) of the Rocky procera, A. magnifica var. shastensis, and hybrids between A. Mountains. Critchfield refer to these populations as “the Del Norte procera and A. magnifica var. shastensis were all found in the race.” They have been found to be morphological and chemically region’s mountains. Red fir (Abies magnifica var. magnifica) variable, sharing leaf characteristics with plants of the coast and grows in the southern Cascades and in the Sierra Nevada. Rocky Mountains (Bivin 1986, Oliphant 1992). These populations Therefore, it is possible to consider that the region has up to three are included in Pinus contorta ssp. contorta in this discussion. red fir taxa. I placed A. procera and A. x shastensis in Table 1. The fourth subspecies, Bolander pine (P. contorta ssp. bolanderi),

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 129

VEGETATION ECOLOGY grows near Fort Bragg, California (Lanner 1999; Smith and continuing adaptation of the groups to the current environmental Sawyer 2003). I included two subspecies (Table 1). conditions (Sawyer in press). Instead of looking mostly back in time, consider that adaptive radiation in the recent geological past Tsuga, the Hemlocks may explain much of the plant and animal richness the Klamath There is sufficient regional variation in Tsuga mertensiana cones Mountains. to justify the recognition of a second subspecies (Tsuga mertensiana ssp. grandicona) based on specimens from the What are the Causes of Localized Conifer Siskiyou Mountains (Fajon 1998). He also reports this subspecies Richness? from the northern Sierra Nevada. Not all trees in Siskiyou Mountains have large cones; those within the size range of the Explanations for localized areas with an unusually high typical subspecies (Tsuga mertensiana ssp. mertensiana) also number of conifer taxa might be included in what Eckert and occur here. I recognize both subspecies (Table 1). Sawyer (2002) call the marginal hypothesis. High conifer richness results when ecological interactions between plants are reduced. What are the Causes of Regional Conifer Richness Let me cite some examples. Seedlings of shade intolerant pines and Variation? can grow up to trees near shade tolerant fir trees in a forest where the canopy is always open. Rocks in a glacial moraine restrict individual trees to favorable microsites and forest canopies are Whittaker (1961) argued that because of their long and open (Sawyer and Thornburgh 1988). Only a few widely-spaced complex history, the mountain ranges in southwestern Oregon and trees establish in nutrient-limited soils, including ultramafic northwestern California have amassed a diverse flora from many substrates (Kruckeberg 1992; Oline et al. 2000). Extreme climates sources. Some taxa evolved in the region, but others migrated into create open patches of krummholz at timberline (Ryerson 1983). the area, since it is centrally located and continuous with adjacent Cyclically disturbed habitats maintain forests with seral species regions at montane elevations. Its diverse climate, topography, and (Murray et al. 2000). geology provided many different environments. Hence the flora is However, Eckert and Sawyer (2002) found that seed source the result of a collection of many taxa with long enduring lineages availability and substrate heterogeneity appear to be important in from many different sources (Coleman and Kruckeberg 1999). explaining rich stands, such as those at California’s Bear Basin Wilken (1993) reported that “At least 50% of all modern Butte. In comparing both rich stands with simple ones of Pinus genera and species in the California flora today can be traced with balfouriana in the northwestern California and the Sierra Nevada, some certainty to the taxa of the Paleogene forests.” These relict we found that simple stands were neither more mesic nor more taxa represent those of the Arcto Ð Tertiary flora, including Picea, nutrient-rich, nor were the rich stands found only on extreme sites. Sequoia, and Torreya (Raven and Axelrod 1978). Stebbins and Instead, we found that an alternate hypothesis, similar to the Major (1965) emphasized the endemic aspect of the California mountain island effect hypothesis of Hamrick et al. (1994), better flora, pointing out that a notable number of these relicts, including explained these results. The simple stands were the product of Chamaecyparis lawsoniana and Picea breweriana, were species-specific elevation limits. Abies magnifica in the Sierra paleoendemics (plants that in the geological past had a much Nevada grows over a considerable elevational range. There upper larger range). However, other relict conifers, such as Pinus montane Abies magnifica forests are replaced by subalpine Pinus balfouriana, Sequoia sempervirens, and Torreya californica, are contorta forests as elevation increases. Individual fir trees occur in not restricted to the region. Many other conifers, such as Abies the lower elevations of the subalpine belt, but rarely at higher concolor, Pinus ponderosa, and Pseudotsuga menziesii, grow elevations where stands are almost exclusively Pinus albicaulis or throughout the West. Taxus brevifolia ranges from Alaska south to Pinus balfouriana. This is not the case in the compressed forest San Francisco Bay and the central Sierra Nevada. Pinus belts found in mountains of northwestern California, where lambertiana ranges from California to central Oregon. Most of the individual Abies x shastensis can be found on the highest ridges. region’s conifer richness comes from these generalists, with Conifer richness varies throughout the mountain ranges of ranges that go well beyond the boundaries of the Klamath northwestern California, where it is greatest in the northern-most Mountains, not from local relicts. stands. The stands in the Yolla Bolly Mountains were well south Stebbins and Major (1965) and Smith and Sawyer (1988) of the range of Pinus albicaulis and Pinus contorta ssp. pointed out that the Klamath Mountains also include many newly murrayana and at the range limits for Pinus monticola and Tsuga recently evolved taxa, including neoendemics (taxa arising mertensiana, thereby reducing the species pool in the southern recently in the region). Cupressus bakeri ssp. matthewsii and mountain ranges. In the richer stands of the Marble, Salmon, and Juniperus communis var. jackii stand out as conifers that fill that bill. Might the Del Norte race of Pinus contorta be next? Trinity Mountains and Trinity Alps, the subalpine conifers mix An unappreciated part of the explanation of regional conifer with montane ones, such as Abies concolor, Calocedrus richness is their continuing evolution. Confusion over the number decurrens, and Pseudotsuga menziesii. In the Sierra Nevada, the of conifer taxa is the result of high genetic diversity and the limits of these montane species are well below the range of Pinus balfouriana. These findings suggest that the local seed availability

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 130 VEGETATION ECOLOGY is important in explaining the richer stands. In rich forest stands University of California Press, Berkeley. with open tree canopies, safe sites for seedling establishment are Hamrick, J. L., A. E Schnabel, and P. V. Wells. 1994. nearby for all conifer species. In simple forest stands with closed Distribution of genetic diversity within and among tree canopies, safe sites for seedling establishment are limited to populations of Great Basin conifers. Pages 147-161 in, K. T. one or a few conifer species. The presence of local seed sources, Harper, L. L. St. Clair, K. H. Thorne, and W. M. Hess, eds. enhanced by substrate heterogeneity, best explains conifer species Natural History of the Colorado Plateau and Great Basin. richness patterns in the Klamath Mountains at the local scale. The University of Colorado Press, Niwot. extreme conifer richness at Russian Peak, where 17 conifer Hunt, C.B. 1967. Physiography of the United States. W.H. species can be seen on a hike to the ridges above Duck Lake, is Freeman and Company, San Francisco, California. the ultimate example (Sawyer et al. 1970). Hunt, R.S. 1993. Abies. Pages 354-362 in, Flora North America Editorial Committee. Flora of North America. Volume 2, Literature Cited pteridophytes and gymnosperms. Oxford University Press, New York. Irwin, W.P. 1966. Geology of the Klamath Mountain province. Adams, R.P. 1993. Juniperus. Pages 412-420 in, Flora North Pages 19-38 in, E. H. Bailey, editor. Geology of America Editorial Committee. Flora of North America. northern California. Bulletin 190. California Division of Volume 2, pteridophytes and gymnosperms. Oxford Mines and Geology, San Francisco, California. University Press, New York. Kral, R. 1993. Pinus. Pages 373-398 in, Flora North America Bivin, M.M. 1986. A fifth subspecies of lodgepole in northwest Editorial Committee. Flora of North America. Volume 2, California and southwest Oregon. M.A. thesis. Humboldt pteridophytes and gymnosperms. Oxford University Press, State University, Arcata, California. New York. Brown, J.H. and M.V. Lomolino. 1998. Biogeography. Second Kruckeberg, A. R. 1992. Plant life of western North American edition. Sinauer Associates, Inc., Sunderland, ultramafics. Pages 31-74 in, B. A. Roberts and J. Massachusetts. Proctor, eds. The ecology of areas with serpentinized rocks: Burnes, R.M., and B.H. Honkala. 1990. Silvics of North a world view. Kluwer Academic Publishers, The America. Volume I, Conifers. USDAAgriculture Handbook Netherlands. 654. Washington, D.C. Lanner, R.M. 1999. Conifers of California. Cachuma Press, Los Callahan, F. 2000. The word for the world is conifer. Mountains Olivos, California. and Rivers 1:12-15. McKee, B. 1972. Cascadia, the geological evolution of the Coleman, R.G., and A.R. Kruckeberg. 1999. Geology and plant Pacific Northwest. McGraw-Hill Book Company, New life of the Klamath-Siskiyou ecoregion. Natural York. Areas Journal 19:320-340. Munz, P.A. 1959. A California flora. University of California Critchfield, W.B. 1957. Geographic variation in Pinus contorta. Press, Berkeley. Maria Moors Cabot Foundation Publication 3. Harvard Murray, M. P., S.C. Bunting, and P. Morgan. 2000. Landscape University, Cambridge, Massachusetts. trends (1753-1993) of whitebark pine (Pinus albicaulis) DellaSala, D.A., S.B. Reid, T.J. Frest, J.R. Strittholt, and D.M. forests in the West Big Hole Range of Idaho/Montana, Olson. 1999. A global perspective on the biodiversity of U.S.A. Arctic, Antarctic, and Alpine Research 32:412-418. the Klamath-Siskiyou region. Natural Areas Journal 19:300- Niebling, C.R. and M.T. Conkle. 1990. Diversity of Washoe pine 319. and comparison with allozymes of ponderosa pine races. Diller, J.S. 1903. Klamath Mountain section. American Journal Canadian Journal of Forest Research 20:298-308. of Science. 4th Series 15:342-362. Noss, R.F., J.R. Strittholt, K. Vance-Borland, C. Carroll, P. Frost. Eckenwaler, J.E. 1993. Cupressus. Pages 412-420 in, Flora North 1999. A conservation plan for the Klamath- Siskiyou America Editorial Committee. Flora of North America. ecoregion. Natural Areas Journal 19:392-411. Volume 2, pteridophytes and gymnosperms. Oxford Oline, D. K., J. B. Mitton, and M. C. Grant. 2000. Population University Press, New York. and subspecific genetic differentiation in the foxtail pine Eckert, A.J. and J.O. Sawyer. 2002. Foxtail pine importance and (Pinus balfouriana). Evolution 54:1813-1819. conifer diversity in the Klamath Mountains and Oliphant, J.M. 1992. Geographic variation of lodgepole pine in southern Sierra Nevada. Madroño 49:33-45. northern California. M.A. thesis. Humboldt State Farjon, A. 1998. World checklist and bibliography of conifers. University, Arcata, California. The Royal Botanic Gardens, Kew. England. Parker, E.L. 1963. The geographic overlap of noble fir and red Griffin, J.R. and W.B. Critchfield. 1972. The distribution of fir. Forest Science 9:207-216. forest trees in California. Research Paper 82/1972. Peck, M.E. 1961. A manual of the higher plants of Oregon. USDA Forest Service, Berkeley, California. Second edition. Oregon State University Press, Corvallis. Griffin, J.R. 1993. Pinaceae. Pages 115-121 in, Hickman, J.C. editor. The Jepson manual: higher plants of California. Raven, P.H. and D.I. Axelrod. 1978. Origin and relationships of

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 131

VEGETATION ECOLOGY

the California flora. University of California Publications Smith, J.P. and J.O. Sawyer. 1988. Endemic vascular plants of in Botany 71:1-134. northwestern California and southwestern Oregon. Ricketts, T.H., E. Dinerstein, D.M. Olson, C.J. Loucks, W. Madroño 35:54-69. Eichbaum, D. DellaSala, K. Kavanagh, P. Hedao, P.T. Smith, J.P. and J.O. Sawyer. 2003. A checklist of the vascular Hurley, K.M. Carney, R. Abell, and S. Walters. 1999. plants of northwestern California. Miscellaneous Terrestrial ecoregions of North America, a conservation Publications No. 2 (18th edition). Humboldt State assessment. Island Press, Covelo, California. University Herbarium University, Arcata, California. Ryerson, D. 1983. Population structure of Pinus balfouriana Available online at: http://www.humboldt.edu/%7Eherb/ Grev. & Balf. along the margins of its distribution area in Stebbins, G.L. and J. Major. 1965. Endemism and speciation in the Sierran and Klamath Regions of California. M.S. thesis, the California flora. Ecological Monographs 35:1- 35. Stuart, J.D. and J.O. Sawyer. 2001. Trees and shrubs of California State University, Sacramento, California. California. University of California Press, Berkeley. Sawyer, J.O. and D.A. Thornburgh. 1988. Montane and Wallace D.W. 1983. The Klamath knot, explorations of myth and subalpine vegetation of the Klamath Mountains. Pages 699- evolution. Sierra Club Books, San Francisco, California. 732 in, M.G. Barbour and J. Major, eds. Terrestrial Williams, C. 1999. Theme issue: the Klamath-Siskiyou vegetation of California. California Native Plant ecoregion. Natural Areas Journal 19:294-420. Society, Special Publication Number 9, Sacramento, Whittaker, R.H. 1961. Vegetational history of the Pacific coast California. states and the central significance of the Klamath Sawyer, J.O., D.A. Thornburgh, and W.F. Bowman. 1970. Region. Madroño 16:5-23. Extension of the range of Abies lasiocarpa into Wilken, D.H. 1993. California’s changing climates and flora. California. Madroño 20:413-415. Pages 55-58 in, Hickman, J.C. (ed.). The Jepson Sawyer, J.O. In press. The varied landscapes of northwestern manual: higher plants of California. University of California California. University of California Press, Berkeley. Press, Berkeley.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 132

VEGETATION ECOLOGY

Table 1. Distribution of conifer taxa in southwestern Oregon and northwestern California. Taxa occur in the Klamath Mountains unless followed by a ‡. The overall distribution includes ∉œ = taxa whose southern most range limits are in the region, ∈Œ = taxa whose northern range limits are in the region, ⊂à = taxa whose western range limits are in the region, n. = northern, nw. northwestern, e. = eastern, s. = southern, w. = western, CA = California, OR = Oregon. The common occurrence of Abies concolor indicates the lower boundary of the montane and the common occurrence of Tsuga mertensiana the subalpine.

SPECIES OVERALL DISTRIBUTION DISTRIBUTION REGIONALLY ABUNDANCE

Abies amabilis Pacific NW ∉ e. montane, subalpine Local

Abies concolor Western NA montane, subalpine Regionally common

Abies grandis Pacific NW coastal, w. low elevation, w. Locally common montane Abies lasiocarpa Western NA ∉œ e. montane, subalpine Local

Abies x shastensis nw. California, s. Oregon montane, subalpine Regionally common

Abies procera Pacific NW ∉œ w. montane, subalpine Locally common

Calocedrus decurrens California, s. Oregon low elevation, montane, Varies with parts of region subalpine Chamaecyparis lawsoniana Near endemic w. low elevation, w. montane, Locally common subalpine Cupressus bakeri ssp. Endemic montane Local matthewsii Cupressus macnabiana ‡ California ∈Œ inland at low elevation Local

Cupressus sargentii‡ inland at low elevation Local California ∈Œ

Juniperus californica‡ California ∈Œ inland at low elevation Locally common

Juniperus communis var. jackii Endemic w. low elevation Local

Juniperus communis var. Western NA montane, subalpine Local montana Juniperus occidentalis ssp. California ⊂Ã montane Local australis‡ Juniperus occidentalis Great Basin ⊂Ã e. low elevation Locally common ssp. occidentalis Picea breweriana Endemic montane, subalpine Local

Picea engelmannii Western NA ∉œ e. montane Local

Picea sitchensis Pacific NW coastal Locally common

Pinus albicaulis Western NA subalpine Locally common

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 133

VEGETATION ECOLOGY

SPECIES OVERALL DISTRIBUTION DISTRIBUTION REGIONALLY ABUNDANCE

Pinus attenuata CA, s. OR ∈Œ low elevation, w. montane Locally common

Pinus balfouriana var. Endemic sublpine Locally common balfouriana Pinus contorta Pacific NW coastal, w. low elevation Locally common ssp. contorta Pinus contorta California ⊂Ã montane, subalpine Locally common ssp. murrayana

Pinus jeffreyi CA, s. OR low alpine, montane, subalpine Locally common

Pinus lambertiana California, s. Oregon w low elevation, montane, Regionally common

Pacific NW low elevation, montane, Pinus monticola Locally common subalpine Pinus ponderosa low elevation, montane, Western NA Regionally common var. ponderosa subalpine

Pinus sabiniana CA, s. OR ∈Œ low elevation Locally common

Pinus washoensis Great Basis ⊂Ã montane Local

low elevation, montane, Pseudotsuga menziesii var. Pacific NA Regionally local menziesii subalpine

Sequoia sempervirens CA, s. OR w. low elevation Locally common

low elevation, montane, Taxus brevifolia Pacific NW Regionally local subalpine

Thuja plicata ‡ Pacific NW ∉œ coastal Local

Torreya californica California ∈Œ e. low elevation Very local

Tsuga heterophylla Pacific NW coastal, w. low elevation Locally common

Tsuga mertensiana var. California montane, subalpine Locally common grandicona Tsuga mertensiana var. Pacific NW montane, subalpine Locally common mertensiana

Xanthocyparis nootkatensis Pacific NW ∉œ w. montane, subalpine Local

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 134

VEGETATION ECOLOGY

BIODIVERSITY BELOW GROUND: MYCORRHIZAL FUNGI AND OREGON WHITE OAKS (QUERCUS GARRYANA) Lori Valentine, Carolyn Petersen, Heather Tugaw, Aaron Hart, Mariah Moser, Jonathan Frank, Harold Berninghausen, and Darlene Southworth Southern Oregon University, Department of Biology, Ashland, OR 97520 Email: [email protected]

Abstract

Although plants associated with oak savannas and woodlands are visible above ground, another association lies hidden below ground: mycorrhizal fungi and oak roots. The relationship is mutualistic in that fungi take up water and minerals from soil and transfer them to the tree that, in turn, sends carbohydrates to the fungi. We investigated diversity of mycorrhizal fungi associated with Oregon white oak at Whetstone Savanna Preserve in southern Oregon. How many species of mycorrhizas associate with Oregon white oak? Can mycorrhizas on oak roots be predicted by the fungi (Basidiomycota and Ascomycota) that form fruiting bodies above ground? We described mycorrhizal diversity by morphotyping using microscopic characters and by molecular techniques to identify fungi associated with oak roots. About 40 ectomycorrhizal morphotypes were described from Whetstone Savanna in southern Oregon. Five were found in 5% or more of soil samples; the rest occurred in one or two soil cores. Preliminary data indicated that Oregon white oaks and ponderosa pines shared few fungi. Mycorrhizas were also sampled from oaks growing on paired serpentine and non-serpentine soils. Oaks on both soil types had abundant mycorrhizas indicating that heavy metals do not inhibit mycorrhizal fungi. Of 81 mycorrhizas, 48% were unique to serpentine soils. On Whetstone Savanna, we trapped small mammals that ate hypogeous fungi and spread fungal spores. The value of this research is baseline data with which to compare below-ground mycorrhizal communities in a variety of growth conditions including fragmented and degraded habitats. Introduction Different forms are called morphotypes because often ectomycorrhizas cannot be identified to species by microscopic he savannas and woodlands of Oregon white oak characters (Agerer 1991; Goodman et al. 2002). In an attempt to (Quercus garryana) and its associated flowering plants, identify morphotypes to species, we turned to DNA analysis, TTbirds and mammals form a familiar ecosystem in low- comparing particular DNA fragments from mycorrhizas to elevation habitats of the Pacific Northwest. But another corresponding fragments from identified fruiting bodies including community in this ecosystem, less familiar, but nevertheless gilled mushrooms, truffles, and cup fungi collected near Oregon important to the health of oaks lies hidden below ground. This is white oak trees. the community of mycorrhizal fungi growing in association with Although conifer mycorrhizas in the Pacific Northwest are oak roots. The mycorrhizal relationship is symbiotic and well studied (Molina and Trappe 1982), oak mycorrhizas have mutualistic in that the fungus provides nutrients to the tree, and received little attention. Our questions concern biodiversity. How the tree provides carbon to the fungus (for review: Smith and many species of fungi form mycorrhizas with Oregon white oak? Read 1997). Which fungal species are they? Which are the common species? The structure that develops from an interaction between Can we predict which fungal species form the mycorrhizas with fungus and root is a mycorrhiza—a fungus root. Most oak roots below ground by identifying the species of sporocarps mycorrhizas on oaks are ectomycorrhizas—shortened roots above ground? containing oak and fungal tissue. In ectomycorrhizas, the fungal Answers to these questions may be useful in several ways. component consists of a mantle of fungal cells that coat the root For example, fungi may form a common mycorrhizal network in surface, hyphae extending out from the mantle into the soil, and an oak woodland. One tree may be connected to another tree by hyphae growing into the root and forming a network among the fungal hyphae, in which case individual oaks could share carbon root cortical cells. This network, called the Hartig net, is thought and nutrients. There is evidence for a common mycorrhizal to be the site of carbon and nutrient exchange between tree and network between conifers and birches (Simard et al. 1997) and fungus (Smith and Read 1997). between non-green flowering plants and trees (Taylor and Bruns This paper is an overview of our work on mycorrhizas of 1997; Bidartondo and Bruns 2002). The first step in assessing the Oregon white oaks in southern Oregon. We have found an existence of mycorrhizal networks among oaks is to determine if enormous diversity of mycorrhizas associated with Oregon white oaks share the same fungi. If the same fungal species are on roots oak roots (Valentine et al. 2002, in press). They can be black, rust of different trees, there is potential for linkage (Amaranthus and colored, yellow, or shiny from air trapped at the surface. They Perry 1994). Also where another species such as ponderosa pine vary in branching pattern and in extent of emanating hyphae. occurs with oaks, we can ask if the two species share the same

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 135

VEGETATION ECOLOGY fungi on their roots? Could one tree species be able to inoculate The sites for the study of mycorrhizal relations between the other? Could a mycorrhizal network link the two species? oaks (Q. garryana) and pines (Pinus ponderosa) are in the The diversity of morphotypes on oaks from one site create Cascade-Siskiyou National Monument (CSNM), Jackson County, baseline data with which to compare oak mycorrhizas in other (42¡ 2’ N, 122¡ 23’ W) and on two private ranches near Ruch, soils. Our primary site is Whetstone Savanna on the Agate Desert Jackson County, Ellis (42¡ 15’ N; 123¡ 8’ W) and Gordon (42¡ 15’ alluvium north of Medford. To determine how the ctomycorrhizal N, 123¡ 2’ W). These are all sites where canopies of oaks and pine community is affected by soil composition, we sought sites with overlapped. soils that differed drastically from the alluvium at Whetstone Methods for sampling roots in soil and for DNA methods Savanna. For example, Oregon white oaks occur on serpentine followed Valentine et al. (2002, in press). Briefly, we used a soil soils that are low in nitrogen, phosphorus, and potassium (Brooks corer (1 inch diameter, 10 inches long) to remove soil, including 1987). Since poor soils generally promote mycorrhizal formation, roots. We washed the roots, picked out mycorrhizas, and we might predict that serpentine soils would have abundant examined them microscopically. mycorrhizas. However, because serpentine soils contain To assess the similarity between sites or between trees, we potentially toxic elements such as copper, chromium and nickel, computed Sorenson’s Index of similarity, S (Brower et al. 1998). such substrates might inhibit mycorrhizal formation or fungal s growth. Serpentine soils generally have sparse vegetation and a high number of endemic species (Kruckeberg 1984). We DNA Methods compared oak mycorrhizas on serpentine with those on non- For DNA analysis, we ground mycorrhizas or pieces of serpentine soils. fungal fruiting bodies and extracted DNA (Gardes and Bruns As part of our survey of fungal fruiting bodies under oaks, 1993; Valentine et al., in press). Following established we searched for truffles belowground. Two of these hypogeous procedures, we amplified small segments of DNA using fungal- genera, Tuber and Peziza, formed mycorrhizas with oak roots specific primers via the polymerase chain reaction Then we cut (Valentine et al. in press). In conifer forests, flying squirrels and the amplified pieces of DNA into fragments (RFLPs) using red-backed voles eat truffles and spread their spores (Fogel and commercially available enzymes, separated fragments on Trappe 1978; Johnson 1996). This leads to the hypothesis that a acrylamide gels, and analyzed sizes of fragments on gels using comparable situation exists in oak savannas. Do small mammals oneDscan software. eat truffles and spread spores via defecation? If so, small mammal mycophagy might be important to consider in Small Mammal Trapping reforestation and management of fragmented woodlands. To trap small mammals, Sherman live traps, (3 x 3.5 x 9.5 We used morphotyping, DNA analysis, and small-mammal in) baited with rolled oats, were set every other week for three trapping to investigate patterns of distribution, species richness consecutive nights of trapping from April through June. Animals and commonality of mycorrhizas associated with Oregon white were identified in the field and released. A kill trap was used to oaks. capture one pocket gopher. Fecal pellets were examined for spores. Methods Results Sites and Sampling All study sites reported here are in Oregon. Our principal Diversity of Ectomycorrhizas on Oregon White Oaks study site is Whetstone Savanna Preserve (42∞ 25’ N 122∞ 54’ At Whetstone Savanna, 200 soil cores from the top 6-10 W) that is located on the western edge of the Agate Desert in the inches of soil around 25 trees contained over 40 morphotypes or Rogue Valley, Jackson County. The Agate Desert is an alluvial species. The most common morphotype, Cenococcum geophilum fan capped with a shallow layer of clay loam over cemented hard (Figure 1), a non-fruiting ascomycete, occurred in 75% of soil pan. The site is characterized by patterned ground with mounds cores. The next most common four morphotypes occurred in 5- and vernal pools. 17% of cores. Most occurred rarely—once or twice in our soil For the studies comparing mycorrhizal communities with cores. oaks on serpentine soils and nearby non-serpentine soils, we selected three pairs of sites. These are Brick Pile Ranch, south of Comparison of Fruiting Body Species with Mycorrhizal Species o Jacksonville, Jackson County, serpentine (42¡ 08’ N, 122 51’ W) Fungal fruiting bodies were collected at Whetstone Savanna and non-serpentine (42¡ 08’ N, 122¡ 52’ W); Sexton Mountain, over three years. Three types were recognized and partially south of Placer, Josephine County, serpentine and adjacent non- identified: truffles (hypogeous fungi) exposed in the top inch of serpentine (N 42¡ 37’ N, 123¡ 19’ W); and Eight Dollar Mountain, mineral soil by raking away leaf litter, gilled or pored mushrooms west of Selma, Josephine County, serpentine (Q. garryana var. (epigeous fungi), and resupinate fungi (“supine” means “lying breweri, 42¡ 17’ N, 123¡ 41’ W) and non-serpentine (Q. garryana down”) that are flat on the surface of logs. Fungi were identified var. garryana, 42¡ 17’ N, 123¡ 42’ W). to genus, some to species. These are Tuber species, Peziza

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 136

VEGETATION ECOLOGY infossa, and several other truffles and large fleshy fungi such as although we have no direct evidence to support that theory. Russula spp., Clitocybe odora, Hygrophorus eburneus, Boletus Many ectomycorrhizas are formed by hypogeous spp. ascomycetes, including some new species that have not been Soil samples collected from fruiting body sites contained described yet. Sporocarps and mycorrhizas are overlapping sets. many mycorrhizas. DNA analyses showed that the mycorrhizas Not all fleshy fungi match an ectomycorrhiza, and not all did not necessarily match the fruiting body directly above. ectomycorrhizal tips match a fruiting body. Presence of a fungal DNA analyses identified mycorrhizas of Tuber spp., Peziza fruiting body does not correlate with presence a mycorrhiza of infossa, Scleroderma cepa, Xerocomus zelleri and Hygrophorus that species below ground. eburneus. In addition, DNA sequences identified four fungi Overlapping roots of ponderosa pine and Oregon white oaks forming mycorrhizas: Boletus sp., one in the Pezizales, and two share few mycorrhizas. Degree of similarity varies among sites. resupinates, Antrodiella semisupina and Tomentella sp. Our preliminary conclusion is that mycorrhizal linkages between Ponderosa pines and Oregon white oaks are variable and site Common Mycorrhizas between Ponderosa Pine and Oregon White specific. This situation differs from that between conifers and Oak madrones (Amaranthus and Perry 1994) and between conifers and In soils below overlapping canopies of Oregon white oak birches (Simard et al. 1997). and ponderosa pine, relatively few morphotypes were found on Oak mycorrhizas are abundant on serpentine soils in the both host trees (Table 1). Similarity of morphotypes between pine Klamath-Siskiyou region providing evidence that heavy metals do and oak varied with sites. In the Cascade-Siskiyou National not inhibit mycorrhizal formation. Morphotypes occurring only Monument, similarity was low (0.17), but at Gordon Ranch, on serpentine soils are candidates for endemism. Little other similarity was high (0.69). For some mycorrhizas, the host tree information on mycorrhizas and serpentine soils has been could not be distinguished. reported. Maas and Stuntz (1969) compared abundance and diversity of fungal fruiting bodies from coniferous habitats on and Mycorrhizas on Serpentine Soils off serpentine in the Cascade Mountains of Washington. Of 212 Mycorrhizal species richness on serpentine soils was fungi, 19% were found only in serpentine soil areas. The total comparable to that on non-serpentine soils (Table 2). Of 81 total number of fungal species on serpentine soils was about half of morphotypes on paired serpentine and non-serpentine sites, 39 that on non-serpentine soils. Mycorrhizas were not examined. (48%) occurred on serpentine sites only, 28 on non-serpentine Preliminary data provide evidence that small mammals eat only, and 14 on both serpentine and non-serpentine. Cenococcum hypogeous fruiting bodies and defecate spores into the soil, an geophilum (Figure 1) was the only ectomycorrhiza to occur on all interaction that disperses spores with the potential to inoculate sites. new roots. This resembles the pattern of small mammal dispersal of ectomycorrhizal inoculum in coniferous forests and in other Small Mammal Trapping habitats such as dry Eucalypt forests in Australia (Fogel and Four species of small mammals were trapped at Whetstone Trappe 1978; Johnson 1996). Savanna: harvest mouse (Reithrodontomys megalotis), white- Baseline data generated here may be useful in regeneration footed deer mouse (Peromyscus maniculatus), California vole and restoration efforts. Establishment of seedlings or transplants (Microtus californicus), and western pocket gopher (Thomomys may be limited by mycorrhizal status (McCreary 2001; Jones et al. mazama). At least six spore types were present in fecal pellets. 2003). Mycorrhizas are essential for plant growth and survival: they take up water and nutrients and transfer these to the host plant. Yet, greenhouse-grown plants may not form mycorrhizas or Conclusions may form a simpler mycorrhizal community. For oaks, soil inoculum may be patchy or absent in grasslands and at distances Diversity of mycorrhizas is relatively high with 40 different from oak woodlands. Restoration projects involving oak seedlings types on oak roots at Whetstone Savanna. The below-ground need to consider the importance of mycorrhizas. community is complex and patchy in distribution. It is partly mirrored by fungi above ground. In order to tell which mycorrhizas are there, you have to dig. Some mycorrhizal fungi Acknowledgements do not form fruiting bodies; others fruit seasonally or not even every year. This research was funded by National Science Foundation Trees share few fungi. Cenococcum geophilum is the most Grants DEB-9981337 through the Biocomplexity Program and common mycorrhiza, as it is in virtually all ectomycorrhizal Research at Undergraduate Institutions and DBI-0115892 to the communities. Because of its ubiquity, Cenococcum is a candidate Biotechnology Center at Southern Oregon University. We thank for formation of a possible common mycorrhizal network The Nature Conservancy for access to Whetstone Savanna.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 137

VEGETATION ECOLOGY

Literature Cited 422. Kruckeberg, A.R. 1984. California serpentines: flora, vegetation, Agerer, R. 1991. Characterization of ectomycorrhizae. Methods geology, soils, and management problems. University of in Microbiology 23:25-73. California Press, Berkeley. Amaranthus, M.P., and D.A. Perry. 1994. The functioning of Maas J.L., and D.E. Stuntz. 1969. Mycoecology on serpentine ectomycorrhizal fungi in the field: linkages in time and soil. Mycologia 61:1106-1116. space. Plant Soil 159:133-140. McCreary, D.D. 2001. Regenerating rangeland oaks in Bidartondo, M.I., and T.D. Bruns. 2002. Fine-level mycorrhizal California. University of California Agriculture and Natural specificity in the Monotropoideae (Ericaceae): specificity Resources Publication 21601, Oakland. for fungal species groups. Molecular Ecology 11: 557-569. Molina, R., and J.M. Trappe. 1982. Patterns of ectomycorrhizal Brower, J.E., J.H. Zar, and C.N. von Ende. 1998. Field and host specificity and potential among Pacific Northwest laboratory methods for general ecology. McGraw-Hill, conifers and fungi. Forest Science 28:423-458. Boston, Massachusetts. Simard, S.W., D.A. Perry, M.D. Jones, D.P. Myrold, D.M. Durall, Gardes, M., and T.D. Bruns. 1993. ITS primers with enhanced and R.Molina. 1997. Net transfer of carbon between specificity for basidiomycetes - application to the ectomycorrhizal tree species in the field. Nature 388:579- identification of mycorrhizas and rusts. Molecular Ecology 582. 2: 113-118. Smith, S.E., and D.J. Read. 1997. Mycorrhizal symbiosis, 2nd Brooks R.R. 1987. Serpentine and its vegetation. Dioscorides edition, Academic Press, New York. Press, Portland, Oregon. Taylor, D.L., and T.D. Bruns. 1997. Independent, specialized Fogel R., and J.M. Trappe. 1978. Fungus consumption invasions of ectomycorrhizal mutualism by two (mycophagy) by small animals. Northwest Science 52:1-31. nonphotosynthetic orchids. Proceedings National Academy Goodman, D.M., D.M. Durall, J.A. Trofymow, and S.M. Berch Science USA 94:4510-4515. (eds). 2002. Concise descriptions of North American Valentine, L.L., T.L. Fiedler, S.R. Haney, H.K. Berninghausen, Ectomycorrhizae [Internet]. Victoria (British Columbia) and D. Southworth. 2002. Biodiversity of mycorrhizas on Mycologue Publications and Canada-BC Forest Resource Garry oak (Quercus garryana) in a southern Oregon Development Agreement, Canadian Forest Service [cited savanna. In R.B. Standiford, D. McCreary, K. L. Purcell 2002 May 9]. (tech coord.). Proceedings of the fifth symposium on oak http://www.pfc.cfs.nrcan.gc.ca/biodiversity/bcern/manual/in woodlands: Oaks in California’s changing landscape.USDA dex_e.html. Forest Service Gen Tech Rep PSW-GTR-184, pp 151-157. Johnson, C.N. 1996. Interactions between mammals and Available from http://danr.ucop.edu/ihrmp/proceed/sym ectomycorrhizal fungi. Trends in Ecology and Evolution proc16.html. 11:503-507. Valentine L.L., T.L. Fiedler, A.N. Hart, C.A. Petersen, H.K. Jones, M.D., D.M. Durall, and J.W. Cairney. 2003. Berninghausen, and D.Southworth. In press. Diversity of Ectomycorrhizal fungal communities in young forest stands ectomycorrhizas associated with Quercus garryana in regenerating after clearcut logging. New Phytol. 157: 399- southern Oregon. Canadian Journal of Botany.

Table 1. Similarity of mophotypes on oaks and pines at three sites.

a Site Ponderosa pine only Oregon white oak only Both hosts Ss Uncertain host Total morphotypes

CSNM 4 19 2 0.17 9 34

Ellis 4 3 1 0.29 3 11

Gordon 6 7 4 0.61 4 21

a Sorenson’s Similarity Index; a value of 1.0 indicates identical species on two sites.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 138

VEGETATION ECOLOGY

Table 2. Species richness of morphotypes with oak roots on serpentine and non-serpentine soils.

Morphotypes on Morphotypes on non- Morphotypes on both soil Site Total morphotypes serpentine soils only serpentine soils only types Whetstone N/A 40 N/A 40

Brick Pile 17 8 6 31

Sexton 18 15 3 36

Eight Dollar 9 17 5 31

Figure 1. Cenococcum geophilum, an ectomycorrhizal fungus with roots of Oregon white oak. Photo by L. Valentine.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 139

CULTURAL ECOLOGY

RESTORING INDIGENOUS HISTORY AND CULTURE TO THE KLAMATH-SISKIYOU ECOREGION: CONSERVATION, RESTORATION, AND WOOD FIBER PRODUCTION IN THE FOREST MATRIX

Dennis Martinez P.O. Box 495, Douglas City, CA 96024 Email: [email protected]

“We believe that as a community of ecologists living in times of unprecedented ecological change, we can no longer afford the questionable luxury of working solely within our own traditions if we are to learn to live sustainably. Conserving our options means, in part, conserving the diversity of ways of thinking about problems.” -Jesse Ford and Dennis Martinez (from Invited Feature on Traditional Ecological Knowledge in Ecological Applications, vol.. 10, October 2000, by the Ecological Society of America, Introduction , p. 2.)

Abstract

Native Americans inhabiting the Klamath-Siskiyou Ecoregion managed cultural resources with selective burning at virtually all elevations. Following their relocation in 1855, the relatively open vegetation structure that had persisted for millennia, partly due to their use of fire and partly lightening caused, began to close up, increasing fire hazard, lowering biodiversity, decreasing productivity, and diminishing ecological integrity. I argue that protected habitat reserves, while necessary, are not sufficient for habitat protection without restorative management of the forest matrix with timber revenues as a byproduct of a primary restoration and conservation focus. I suggest the development of reference ecosystems for restoration that incorporate historic Indian fire regimes in their seasonality, frequency, and spatial selectivity. ethnography and ecology in the construction of historical Introduction reference or baseline ecosystems which could assist in the recovery of ecological integrity by including Native cultural practices and their effects on key ecological features, along with ndigenous peoples of the Klamath-Siskiyou Ecoregion were Western science-based approaches to restoring ecosystem engaged in intensive and extensive forest management for function, throughout the forest matrix. Matrix restoration, I IImillennia in order to maintain the plant and animal resource, conclude, could be accomplished through the use of an integrated which sustained tribal cultures and economies. Intentional fire, historical-functional model in designing multiple entry timber supplemented by selective harvesting of plants and animals, was harvest prescriptions that primarily promote conservation and the primary management tool. I argue that Native American restoration while yielding an economic byproduct. inhabitants of the Klamath-Siskiyou Ecoregion, like in many other North American biomes, burned in most vegetation types at virtually all elevations in order to modify, regenerate, and increase Methodological Limitations of Western Ecological fruit and seed production of a wide range of plant species for cultural use as well as to rejuvenate culturally important animal Sciences and Ethnography habitat. I argue further that the frequency, and especially the Both Western ecological and ethnographic methodologies spatial selectivity, of Indian fire—in addition to lightening fires— was practiced long enough by relatively large populations of possess important strengths and weaknesses. Their integration in Native peoples to have had an important impact on forest the task of constructing reference models for conservation and structure, composition, and function over millennia, and which restoration could overcome their respective limitations. As I will likely affected the genetics, distribution, and maintenance of discuss in more detail below, ecological restoration is a balancing species and vegetation types that are presently poorly distributed, act between historical fidelity and ecological functionality, i.e. missing, or undergoing detrimental successional changes. I between historical accuracy and how well any restored process, suggest that Indians of the region were a keystone species ecosystem, or species will work in terms of persistence, resilience, (Bonnickson, et al. 1993; Kay 1994), and that their removal in and stability (Higgs and Martinez 1996; Falk et al. 2001). 1855 began a process—accelerated later by Euro-American stock overgrazing, exotic plant invasions, industrial forestry, and fire Western Ecological Sciences suppression—of ecological degradation that continues today. I Western scientific methodology in ecology is based on then offer examples of how to integrate the sciences of quantitative analysis and experimental replicability with the goal

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 140 CULTURAL ECOLOGY of explaining and predicting natural phenomena. While some (1999) tells the story of the gradual separation, in scientific and ecologists of late have rejected rigid mechanistic methodologies— U.S. National Park management thinking, of indigenous peoples indeed the prevailing ecological paradigm is in such a state of flux from their role in managing their homelands in Dispossessing the that it is difficult to adequately and completely categorize—I think Wilderness: Indian Removal and the Making of the National that it is also true that the main thrust of the ecology of the last Parks. See also Martinez (2003a). half century, under the influence of modern technologically The problem with reductionist methodology for our sophisticated economies and the cyber revolution of the 1970s and purposes here lies in its exclusion of factors thought to be external 1980s, has been to reduce the complexities of natural systems to to any given experimental focus. There is always a tradeoff the simplified and abstract bioeconomics of food chains, niches, between information content and reliability. Replicable productivity, yields, etc., and simplified ecosystem analyses and experimental tests are only reliable or rigidly determinate when indicators like focal species and species absence\presence models. the scope of their questions has been greatly limited (Kraus 1974; Generally speaking, modern ecology has been individualistic in its Ehrenfeld 1978). Putting ecology on an exclusively quantifiable conception of plant-to-plant and animal-to-animal interactions and basis necessarily leaves out qualitative analyses like the part that mechanistic in its explanation of plant community development— human culture and history have played in the development of many denying the ecological realty of “community” and the forest structure and composition. Anchoring reference ecosystems social-cultural dimension of animals (although the new science of in real past time, at least as a beginning point in forest restoration behavioral ecology is already confirming some indigenous analysis, could overcome the predictive limitations of Western understandings of animal social behavior (Gleason 1926; Elton reductionist methodologies. We are currently experiencing a 1927; Tansley 1935; Lindeman 1942; Agee 1993; Worster 1994; gigantic and unprecedented experiment in secondary succession as Fox 2003). a result of industrial forest practices, overgrazing, and fire Analysis of vegetation in terms of individual species suppression, the outcome of which is highly unpredictable even moving along a temperature-moisture-soil gradient—and not with the most sophisticated computer modeling. communities with an co-evolutionary history and future—could lead in some cases to minimizing the significance of blended Ethnography ecotones between two or more vegetation types; and which could Ethnographic research, which forms the bulk of my be at least partly composed of modal species that only occur in argument for the current ecological relevance of past indigenous abundance in an intermediate part of a light-shade continuum, for land management practices in the Klamath-Siskiyou Ecoregion, example, savanna modal species which are communities and not can provide the cultural and historical context which could just individual transitional species between prairie and forest suggest hypotheses that the ecological sciences can test with their (Pimm 1991). As I will argue below, Indian burning kept experimental methodologies. The fundamental methodological vegetation development in many plant communities in a state of problem with historical disciplines like ethnography and arrested seral succession. This could be viewed from the ethnohistory is that, like the reconstruction of climate history, they perspective of the prevailing ecological paradigm as a negative rely mainly on indirect or proxy lines of evidence. example of forest fragmentation and not as the maintenance of In this paper I hypothesize that Native American tribes of stable forest conditions or modal ecotone species assemblages that the Klamath-Siskiyou Ecoregion would have had to burn a are not targeted for conservation because they are not defined as a considerable number of plants to maintain their economies and community. This is not to deny the detrimental effects of cultures. But my hypothesis assumes much larger pre-European fragmentation by roads and staggered-setting clearcuts in some settlement (precontact) populations than agreed upon by all habitat types. I am only suggesting that we need to be more scholars. Ethnographer James Mooney made the first to attempt discriminating about our evaluation of what we perceive as the an estimate of North American (U.S. and Canadian) Native negative effects of fragmentation. This actually happened in the populations in 1910. His estimated 1.15 million in this region upper Midwest when most ecologists were denying the existence (Denevan 1976). Mooney’s estimate stood until Henry Dobyn’s of oak-tallgrass prairie savanna ecosystems and of modal species (1966) controversial estimate of 10 million for North America, associated with an intermediate place on the light spectrum; the and 90 to 112 million for the Western Hemisphere (Dobyn 1983). matter was finally resolved when Steve Packard of The Nature Even Dobyn’s most vehement critic, David Henige (1983), came Conservancy found modal savanna species as a result of historical up with numbers significantly higher than Mooney. The higher research, and successfully restored them on sites in the greater numbers currently have the most scholarly support. Chicago metropolitan area (Stevens 1995). Modal ecotone Dobyns used a novel method of estimating population size communities may not be limited to the pine or oak savanna base on educated guesses of the number of Native persons killed vegetation types but could be disappearing in middle to high by European diseases. He calculated 95% losses. His bichronic elevation mixed conifer forests due to shading out, with one result method took into account the numbers lost to disease before the being the elimination of natural evolutionary processes where original local counts were made years after first European “natural” includes human management. Unfortunately, North contacts. He then took the lowest point in indigenous populations American ecologists have traditionally separated humans from (e.g., 500,000 in 1900 in North America), and, assuming 95% nature when doing ecosystem studies. Historian Mark Spence losses, calculated backwards to a population of 10 million.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 141 CULTURAL ECOLOGY

We will never know with certainty the actual precontact restoration is the process of assisting the recovery and population numbers, but it is fairly certain, based on recent management of ecological integrity. Ecological integrity includes scholarship, that Mooney’s original estimate is far too low. a critical range of variability in biodiversity, ecological processes Modern scholarship has pushed both Indian populations and and structures, regional and historical context, and sustainable length of time spent in this hemisphere significantly higher and cultural practices (emphasis added).” (See SER website for longer than heretofore believed (Dobyns 1983; Fiedel 1987; extended discussion of this definition at). Ramenofsky 1987). The development of a reference ecosystem model requires Recent research indicates that European diseases may have balancing historical fidelity with ecological functionality (Higgs spread rapidly enough to decimate Indian populations in the and Martinez 1996; Martinez 2002; Higgs 2003). The Pacific Northwest between 1550 and 1600. Pullen (1999) quotes preindustrial historical model, which includes as much of key ethnographer Melville Jacobs, Notebook 125, in which the Indian ecological structure, composition, and processes (i.e. function) as informant refers to the time before the last great famine (1830’s?): is possible to retrieve using the tools of the newly emerging field “Long ago lots of people lived in Applegate River in one village of historical ecology (Egan and Howell 1999), can serve as an there…” He goes on to explain how inter-marriage with the initial guide to a reference ecosystem and as a conceptual tool Shasta allowed the Da-ku-be-te-de, inhabiting the Applegate with which to assess current forest conditions. How close we are Valley, to go to the Shasta, near Yreka, California, and obtain food able to come to historical reference conditions at the stand level is and avoid starvation. This suggests a larger population than the usually determined by conditions of the larger forested landscape usual low estimates noted in the literature. Deserted villages were in which the stand is embedded. Because of fragmentation, fire commonly seen. (See Reg Pullen, p. IV - 2, Melville Jacobs, suppression, exotic plant invasions, etc. some level of periodic Notebook 130). Pullen (p. IV-3) quotes early southwestern human intervention will probably always be necessary. Because Oregon resident Dr. Lorenzo Hubbard who speaks about the of changed environmental conditions at the landscape scale, it Tututni of the lower Rogue River in 1856: needs to be integrated with ecological functionality. Function can be measured with standard scientific tests, e.g. net primary “According to tradition, many years ago they were far productivity, nutrient cycling, etc. (Covington 1997). more numerous than at the present time, wars and Forest conditions are constantly changing over time. Our diseases having in some instances destroyed whole point of reference in the past is not static. Rather than a snapshot tribes. The marks of old towns and large settlements in time, we are rerunning a long moving sequence of ecological everywhere found, now entirely deserted, are strong variation resulting from periodic disturbance events that were evidence of their traditions.” historically bounded in intensity and frequency. Native management is part of that circumscribed natural flow. By Finally, much of ethnography is what Western scientists call attempting to approach preindustrial conditions, we hope to get anecdotal, and is not usually accepted as hard or reliable evidence. the rate and extent of change back within this historical range of But, in combination with physical evidence from ecology, it could variability (HRV). The question is not change, but what kind of be used in a mutually reinforcing way. That in fact is the change. Like medical doctors, ecological restorationists “work principal objective of this paper: how to use the somewhat sketchy with natural process, intervening no more than necessary to nudge ethnographic record of southwest Oregon (and the more complete nature just enough to change its natural trajectory from a human record of northwest California) to provide explanatory hypotheses caused downward spiral to one that is potentially positive” where physical evidence of precontact forest structure and (Martinez 2003b). composition is missing or ambiguous. For example, see Pullen (1995) for ethnographic descriptions of forested lands capes under Indigenous Land Management Practices in the Indian management. Techniques of the newly emerging field of Klamath-Siskiyou Ecoregion and their Relevance to historical ecology as well as conventional ecology can then test Conservation and Restoration those hypotheses. Anecdotal information, in the history of science (my original academic field of study), has often played an Given the apparent longevity of the indigenous societies of important role in suggesting explanations of natural phenomenon the Klamath-Siskiyou Ecoregion (perhaps as long as 12,000 years that were later experimentally tested. and at least 4000 years), what kinds of resources were managed, and how extensively and intensively were they managed in order to ensure cultural survival? Would intentional fire in lower Balancing Historical Fidelity with Ecological elevation valleys and foothills where permanent villages were Functionality in Reference Ecosystem Models for located have been sufficient for all of their fire-dependent cultural Restoration and Conservation needs? What kinds of important cultural resources would then have gone unmanaged? Would these resources have required fire management to have been culturally useful? Would lightning fires The Society for Ecological Restoration International (SER- have been sufficient to prepare culturally important plants at I) has defined ecological restoration as follows: “Ecological higher elevations for intensive and prolonged human use?

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 142 CULTURAL ECOLOGY

Ethnobiologist M. Kat Anderson and ethnographer Thomas Report for World Wildlife Fund\U./S. Forest Service Upper Glade C. Blackburn point out the enormous quantities of burned plant National Pilot Stewardship Project, Dennis Martinez (2001); see material which would have been required to maintain the material also Reg Pullen (1996). culture of California Indian societies and conclude “that only Basketry and Cordage plants include beargrass (Xeropyllum careful and effective management could have supplied the tenax), hazelnut (Corylus californica), Iris chrysophylla and other phenomenal quantities of raw materials required to support such a Iris species; milkweed, (Asclepias spp.), and dogbane (Apocynum community over long periods of time” (Blackburn and Anderson androsaemifolium). These species are found in scattered locations 1993). Cordage was certainly one of the critical needs. up to around 5500 ft. [1676 m.] (Barbara Mumblo, Applegate Blackburn and Anderson note that just one 40-foot deer net Ranger District botanist, pers. comm., 2001). Iris chrysophylla containing 7000 feet of cordage required the harvesting of 35,000 was on the California Review List until 1994 (Report by Richard plant stalks of milkweed (Aesclepias spp.) or dogbane (Apocynum Brock and Richard Callahan, 1994, on file with the Applegate spp.) burned the season before. When all of the fire-modified Ranger District). Beargrass, the most important basket plant, is plant material in ten major cultural use categories are considered, rarely found in quantity. Fire suppression and shading out by including deer and elk habitat, it is clear that significant burning encroaching conifers constitute the primary reason for the scarcity was a prerequisite for tribal survival. of beargrass as well as other basket and cordage plants (author’s Since it is generally accepted that Indians burned lower observations). elevation valleys and foothills regularly, as often as every year, to Food plants include higher elevation nuts, the most enhance culturally important plant foods like corms, acorns, grass important of which were hazelnut, black oak (Quercus kelloggii), and forb seeds, basket plants, etc., I will focus on cultural and Sadler’s oak (Quercus sadleriana). These latter two species resources at middle and upper elevations in the Applegate and produced two of the three most popular acorns. But without Rogue River watersheds of the eastern Siskiyous. (For more detail regular management by fire, hazelnut patches and oak stands on lower elevation burning see Reg Pullen [1996] and Jeff stagnate, acorn production declines, and regeneration eventually LaLande [1995]). ceases due to shading out (author’s observation). Other foods Conventional wisdom on the fire history of the Klamath- include geophytes (corm producing herbaceous plants) like Lilium, Siskiyou Ecoregion holds that lightning was the primary ignition Fritillaria, Camas, Tritelia (two species of which are on ONHP source for middle to high elevation mixed evergreen/conifer Review List and California Watch List), Perideridia, and forests (Frost and Sweeney 2000). I will argue that Indian Calochortus species, which grew historically on high elevation burning was in fact done in these higher elevation forests, and that meadows but which are disappearing because of brush and tree both lightning and Indian fire shaped forest structure and encroachment (author’s observation). Only one patch of camas— composition. I will base this argument on the kinds of cultural the most prized edible corm—is known in the eastern Siskiyous plants and animals that needed to be fire-managed in predictable (Chant Thomas, pers. comm., 2001). and effective ways for the material culture to be maintained. Important higher elevation berries include greenleaf Lightning fires are not predictable in their timing, location, extent, manzanita (Arctostaphylos patula), elderberry (Sambucus and effectiveness. While lightning fires did affect forest cerulea), thimbleberry (Rubus parviflorus), serviceberry conditions, the result was far too random for Native resource (Amelanchier alnifolia), gooseberry (Ribes spp.), and huckleberry managers who required regular fires at specific intervals in (Vaccinum spp.). Ribes marshallii is on the ONHP List 2 and particular places (patches or “yards”), usually in a rotational California Watch List. Again, fruit production and patch pattern of varying fire return intervals, in order to meet different size\abundance have declined because of shading out due to lack resource needs dictated by a variety of environmental and societal of fire (author’s observation). imperatives (Lewis 1973; Bonnickson et al. 1997). Medicinal plants are well represented at higher elevations. Osha root (Angelica arguta) is found at elevations of 5000 to 6000 Critical Middle to High Elevation Cultural Plants ft. [1524 to 1829 m.]. It is the most important Indian medicine and Animal Habitat of the Klamath-Siskiyou that I know of in the Klamath-Siskiyou ecoregion. I collect and Ecoregion use it regularly as an anti-microbial agent. Lomatium nudicaule (also known as Indian consumption plant, or Indian celery), Here I will briefly examine several categories of critically another osha species is found in forest openings at middle important cultural plants which would have to have been elevations, as is L. triternatum (Barbara Mumblo, pers. comm., extensively used—and therefore maintained in great abundance— 2001). Mules-ears (Wyethia angustifolia) is another very popular by relatively high populations of precontact Indians, but which are medicine plant found at middle elevations which is poorly now as a whole fairly scarce or poorly and unevenly distributed. distributed. The ranges of the two tobacco species (Nicotiana While a few of these species also grow at lower elevations, most spp.) have severely contracted (Donn Todt, pers. comm., 2001). (and these were usually the most desirable) grow at middle to high The most popular tobacco species, N. quadrivalvis, was grown at elevations. All require regular light ground fires to be productive higher elevations. I have personally never found it in the or culturally useful. The ranges of all of these species have been Klamath-Siskiyou Ecosystem. Neither has Donn Todt. reduced. More detailed information can be found in the Final Deer and elk habitat was burned every 3 to 5 years to

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 143

CULTURAL ECOLOGY rejuvenate browse. As is typical of Indians inhabiting interior about seasonal harvesting rounds across all elevations and the far- montane forests, deer and elk hunting were more important than reaching effects of even a single fire event, both in spatial extent salmon and steelhead fishing. It is difficult to overestimate the and in its effects on non-cultural plants and habitat. importance of deer to tribal economies in the interior Klamath- I tend to find greatest plant diversity at present on Siskiyou Mountains. Besides providing meat, deer supplied sinew serpentine opens — most in the process of closing up with for sewing, leather for clothes, and a vast miscellany of bone and chaparral and invading trees — and on roadcuts within middle antler tools and implements. High deer populations were elevation coniferous forests when I do Threatened, Endangered, maintained by creating forest openings (yards) through intentional and Sensitive (TES) vegetation surveys. The spottiness of regular fire, burning hundreds of acres in one fire, thus enhancing the low intensity fires was the key factor in creating productive natural carrying capacity of the range. Ridges were travel ecotones. Major stand-replacing fires — probably rare in the corridors and were kept open to facilitate deer and elk drives and precontact Klamath-Siskiyou Ecoregion due to lack of fuel to provide easy access for packing large quantities of meat back to buildup —burn more uniformly and thus, as Lewis and Ferguson villages where it was smoked for winter use. Ridges also were (1973: 84) note: “the very ‘spottiness’ and much higher frequency important beargrass sites—another reason for regular burning of of very localized Indian burning seem to have affected a much the ridges. And when stored food supplies were getting low in the more complex overall ecosystemic pattern than would have been hunger moons of late winter, before salmon runs and Indian the case with only natural fires.” greens were available, new palatable browse from the early fall burns of the previous year attracted elk and deer to middle elevation ridges and forest openings. Up to three hunts were Fire in the Klamath-Siskiyou Ecoregion undertaken per year; 150 deer snares and nets were utilized per hunt. Notably, a single 40-foot deer net took 35,000 stalks of Some ecologists have questioned the view that the Klamath- milkweed or dogbane, which had been burned the year before Siskiyou Ecoregion was characterized by mostly low to moderate (Pullen 1996). severity fires. They claim that it was a region of mixed-severity Agroecological diversity of various sizes and shapes of fire that included at least some high severity fire and extensive forest openings created a complex mosaic of repeating edges or stand-replacing events. This view is part of the ongoing debate transitional ecotones which in turn created diverse kinds of animal over fuels versus weather as the principal driver of fire events and habitat as well as ecological niches which supported a rich flora of has political overtones. both conservative and generalist species supporting complex food Evan Frost has stated that “…conserving ecological webs. Agroecological diversity contributed to cultural resiliency integrity in the Klamath Mountains depends on the extent to (Turner et al. 2003). Ecosystem function, then, would have been which fire is allowed to play its essential role in the enhanced by Indian fire. As pioneer ecologist Eugene Odum ecosystem…[and]…that to be effective over the longterm, (1971) noted: “…it seems likely that ecotones assume greater ecosystem management strategies for federal lands should restore importance where man has greatly modified natural communities” fire using historic patterns of frequency, intensity, seasonality and (see also Lewis 1973). In this way, climatic climax, resource-poor spatial extent.” Frost also notes that “relatively few fire history (i.e. poor in cultural resources, species richness, and wildlife studies have been completed in the Klamath Mountains compared habitat) coniferous forests contained “hot spots” or islands of to neighboring forest regions…[and]…only a few different sites biodiversity through fire-driven arrested seral succession. have been studied, and these fail to represent the full range of Many places were left unburned for a variety of reasons, regional variability that exists within these forests.” He further including animal habitat needs (e.g., thermal cover for deer and notes that fire frequency is better studied than fire size and elk) and the fact that it was unnecessary to burn everywhere in severity (Frost and Sweeney 2000). order to satisfy cultural needs. Spatial selectivity of Indian fire Clearly, we need more fire studies in the Klamath-Siskiyou (i.e.. where they burned) would have been sufficient to create Ecoregion. Still, the overwhelming evidence to date is in favor of refugia for animal and plant species which otherwise would have low to moderate severity fires. Tom Atzet, Rogue-Siskiyou been rare in montane coniferous forests. Riparian zones are only National Forest regional ecologist, has estimated the mean fire 2% of the land area of typical watersheds, but are used by 80% of event to have been 49 acres in size (Atzet, pers. comm., 1995). species at some point in their life cycles. On the Klamath Natural Forest, estimates range historically from Spatial extent (number of acres burned) is difficult to one to 1000 acres per event (Martinez 1995). ascertain, but spatial selectivity is really the more germane Weather-driven stand-replacing fires can certainly occur in ecological consideration because selective fire favors ecotone most fuel-load categories, but it seems to me that these occurred creation and maintenance that is productive for wildlife and relatively infrequently in the Klamath-Siskiyou Ecoregion. Fuels humans out to proportion to its actual size. Geographer Thomas (including fire ladders and contiguous tree crowns), all other Vale has estimated that North American Indians actually occupied factors (e.g. topography, weather) being equal, set the upper limits only 0.02% of the continent’s area, and therefore, he concludes, of severity. And, compared to other regions, fires of high severity Native American fire affected little land area and had virtually no still occur less often. In the 1987 Silver Fire in southwest Oregon, effect on the landscape (Vale 2000). Dr. Vale apparently forgot for example, 9% was high severity, 32% was moderate severity,

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 144 CULTURAL ECOLOGY and 59% was low severity; the patches of high mortality were less refugia for sun-loving, fire-dependent plants and the animals that than several hundred acres each. Similarly, the 1994 Dillon Fire this managed habitat sustained. Larger meadows below the and the Big Bar and High Fires in 1999 (Klamath and Shasta- subalpine zone were similarly maintained for deer and elk forage Trinity National Forests respectively) were mostly of low severity Guessing at where suitable habitat occurs, and then creating (Frost and Sweeney 2000). reserves to protect that assumed habitat—given our poor Because fuel loads were much lower before Indian predictive capabilities at present and our lack of knowledge of the removal and fire suppression, most burning was probably of low life requirements of most at-risk species—is not enough. The to moderate intensity and severity. In fact, one important reason, matrix can influence dispersal and recolonization rates, provide among many, for burning was to reduce fire hazard. Fuel loading suitable habitat, and affect the type and magnitude of edge effects was so low that fires were set at the bottom of slopes and allowed (Lindenmayer and Franklin 2002; Noss 2002; Rabinowitz 2002; to burn uphill (Warren Corbett, Mountain Maidu elder; Frank Fox 2003; Stokes 2003). Lake, Karuk, pers. comm.). Fires did get away occasionally, of The matrix can also provide connectivity. Connectivity is course, especially when weather was more of a factor than fuels. species specific. A map of vegetation cover, based on the Patch- If we are going to “restore historic patterns of frequency, Corridor- Matrix or Landscape Continuum models derived from intensity, seasonality, and spatial extent” (and selectivity), we need Island Biogeography theory may not correspond to a map of to include Native burning regimes to the extent that they are connectivity for a given species. Therefore, as Lindenmayer and retrievable. Here is a good example of the usefulness of both Franklin (2002) conclude, silvicultural practices which favor ecological and ethnographic knowledge sources when major gaps conservation\restoration in the matrix may provide better occur in the data. Indian burning patterns, based on indirect lines connectivity than wildlife corridors for species that disperse of evidence which include maintaining considerable cultural randomly. resources, created a patchy landscape which, along with the Lindemayer and Franklin (2002) suggest a variety of natural diversity of topography, parent rock material and soil types silvicultural approaches to restoration. They call this “risk- of the Klamath geological province, favored high biodiversity. As spreading”: “the implementation of a range of strategies at Frost and Sweeney (2000) conclude: “the patchiness associated different spatial scales. Management for diversity calls for with moderate severity fires has been instrumental in promoting diversity of management…” (emphasis added). They then suggest species and habitat diversity in the Klamath-Siskiyou region.” the “use of knowledge of disturbance regimes in natural forests to guide mature management” (emphasis added) (Lindenmayer and Integrating Conservation Biology and Ecological Franklin 2002). While the authors are sometimes ambivalent on Restoration with Wood Fiber Production in the the question of indigenous peoples and their management Matrix activities as a natural part of ecosystems, they do state so in one citation (Lindenmayer and Franklin 2002). I would add that cultural practices—ancient or modern—are “natural” if they are Laura Jackson and Dana Jackson (2002), in their recent ecologically familiar to the forest over long stretches of time, i.e. book, The Farm as Natural Habitat: Reconnecting Food Systems they are within the historical range of variability in their To Ecosystems, argue that species and habitat are not going to be frequency, intensity, and extent. saved unless we find a way to grow food which is more ecologically appropriate. I believe that the same is true for wood fiber production. We need to design timber harvesting Variable Density Management in the Matrix prescriptions which reconnect with forest ecosystems. A reference ecosystem model which includes Indian burning patterns (fire seasonality, frequency, spatial extent, intensity, and In advocating restoration and conservation matrix site-selectivity) as a supplement to lightening burning regimes management with timber revenues as a byproduct, I want to also needs to be translated into harvest prescriptions which further emphasize that reserves are still necessary—just not sufficient. conservation and restoration in the matrix. We have seen above in The nuts-and-bolts of multiple-entry restorative thinning, the section on indigenous management the enormous quantities of prescription fire, collecting and sowing seeds of missing native fire-modified cultural plants that were required to support the plant species, etc. can be found in my World Wildlife Fund report material culture of large Indian populations in precontact times. (2001) and unpublished but available Collaborative Learning Indians of the Klamath-Siskiyou Ecoregion made seasonal rounds Circle funded draft Holistic Restoration Forestry Manual (2003). from snowmelt to snowfall while tending and harvesting resources See also Lindenmayer and Franklin (2002) for a complete at virtually all elevations. discussion of variable density management. Their rotational burning pattern (fire intervals ranging from The objective of variable density management or risk- one to 20 or more years) created a mosaic of patches of varying spreading is to restore sufficient redundancy in vegetation types, sizes and shapes and including a variety of vegetation types and seral stages, vegetation structure and composition, light and shade, age-classes at the landscape level. It was a forest perforated by denseness and openness, etc. in order to provide suitable habitat in openings that served simultaneously as cultural plant patches and places where we usually lack specific information about the life

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 145 ABSTRACTS requirements of target species. Wood fiber is provided while Fiedel, S. J. 1987. Prehistory of the Americas. Cambridge doing structural restorative thinning, fire reintroduction, and University Press. Cambridge, United Kingdom. composition recovery. The primary focus of harvest prescriptions Fox, D. 2003. More than meets the eye: behavior and is on restoration and conservation. Economy follows ecology. conservation. Conservation In Practice 4(3):20-29. The Native historical reference model for the Klamath- Frost, E.J., and R. Sweeney. 2000. Fire regimes, fire history and Siskiyou Ecoregion suggests the restoration and maintenance of a forest conditions in the Klamath-Siskiyou region: an perforated forest matrix at middle to high elevations, in Mixed overview and synthesis of knowledge. World Wildlife Fund. Conifer\Upper Montane forest types, of openings of various sizes Ashland, Oregon. and shapes and including meadow restoration in the Upper Gleason, H.A. 1926. The individualistic concept of the plant Montane\Subalpine transition where meadows are being invaded association. Bulletin of the Torrey Botanical Club. January by brush and trees. In sum, millennia-long fire management of 1926:7-26. upper elevation forests and meadows contributed to the creation Henige, D. 1983. Numbers from nowhere: The American Indian and maintenance of a distinctive vegetation structure and contact population debate. University of Oklahoma Press, composition which may have fostered the survival of many modal Norman. species and communities, and maintained adaptive co-evolutionary Higgs, E. 2003. Nature by design. Princeton University Press. processes between plants, animals, and indigenous humans. We Princeton, New Jersey. have long recognized the role of humans in the modification of Higgs, E., and D. Martinez. 1996. Minutes of the board of plants and animals in traditional old world agriculture; we need directors of the Society for Ecological Restoration: revision only now recognize the same dynamics in new world agroecology of definition of ecological restoration by science and policy and restore not only indigenous history and culture to the working group co-chaired by Eric Higgs and Dennis Klamath-Siskiyou Ecoregion, but renew that ancient relationship Martinez. Rutgers University, New Jersey. Available Online. as part of our own native heritage. www.ser.org. Jackson, D., and L. 2002. The farm as natural habitat: Literature Cited reconnecting food systems with ecosystems. Island Press. Washington, D.C. Agee, J.K. 1993. Fire ecology of Pacific Northwest forests. Kay, C.E. 1994. Aboriginal overkill: the role of American Indians Island Press. Washington, D.C. in structuring western ecosytems. Human Nature 5:359-398. Blackburn, T.C., and M.K. Anderson (eds). 1993. Before the Kraus, E. 1974. The unpredictable environment. New Scientist wilderness: environmental management by native 63:649-652. Californians. Ballena Press. Menlo Park, California. LaLande, J. 1995. An environmental history of the Little Bonnicksen, T.M., et al. 1997. Native American influences on the Applegate watershed. U.S.D.A. Forest Service, Medford, development of forest ecosystems. Pages 439-470 in, N.C. Oregon. Johnson. A.L. Malk, R.C. Szaro, and W.T. Sexton, eds. Lewis, H.T. 1973. Patterns of Indian burning in California: Ecological stewardship: a common reference for ecosystem ecology and ethnohistory. Ballena Press. Ramona, management. Elsevier Science, Ltd. Oxford, United California. Kingdom. Lewis, H.T., and T.A. Ferguson. 1998. Yards, corridors and Covington, W. et al. 1997. Restoring ecosystem health in mosaics. Human Ecology 89:79-83. ponderosa pine forest in the southwest. Journal of Forestry Lindeman, R. 1942. The trophic-dynamic aspect of ecology. 95(4):23-29. Ecology 23:399-417. Denevan, W. 1976. The native population of the Americas in Lindenmayer, D.B., and J. Franklin. 2002. Conserving forest 1942. University of Wisconsin Press, Madison. biodiversity: a comprehensive multiscaled approach. Island Dobyns, H.F. 1983. Their numbers become thinned: Native Press. Washington, D.C. American population dynamics in eastern North America. Martinez, D. 2001. Final report World Wildlife Fund\U.S. Forest University of Tennessee Press, Knoxville. Service Upper Glade National Pilot Stewardship Project. Doak, D., and L.S. Mills. 1994. A useful role for theory in World Wildlife Fund, Washington, D.C. conservation. Ecology 75: 615-626. Martinez, D. 2002. Putting ecology into restoration forestry. Egan, D., and E. Howell. 1999. Historical ecology. Island Press, Distant Thunder, Journal of the Forest Stewards Guild. Washington, D.C. Spring 2002:1, 2, 11. Ehrenfeld, D. 1978. The arrogance of humanism. Oxford Martinez, D. 2003a. Introduction to holistic restoration forestry: University Press. Oxford, United Kingdom. forested landscapes of southwestern Oregon and northern Elton, C. 1966. Animal ecology. Rev. Edition. New York. California. Unpublished teaching manual, Douglas City, Falk, D., E. Knapp, and E.O. Guerrant. 2001. An introduction to California. restoration genetics. Position paper for Society for Martinez, D. 2003b. Protected areas, indigenous peoples, and the Ecological Restoration. Available Online. www.ser.org . western idea of nature. Ecological Restoration, December

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 146 CULTURAL ECOLOGY

2003: 247-250. removal and the making of the national parks. Oxford Noss, R. 2002. Context matters. Conservation In Practice University Press, Oxford, United Kingdom. (3)3:10-19. Stevens,W. K. 1995. Miracle under the oaks: the revival of native Odum, E. 1971. Fundamentals of ecology. 3rd edition. W.B. in America. Pocket Books, New York. Saunders Co., Philadelphia, Pennsylvania. Stokes, D., and P. Morrison. 2003. GIS-based conservation Pullen, R. 1996. Overview of the environment of native planning: a powerful tool to be used with caution. inhabitants of southwestern Oregon, later prehistoric era. Conservation In Practice 4(1):38-41. Bureau of Land Management\U.S.D.A. Forest Service Tansley, S.G. 1935. The use and abuse of vegetation concepts publication. and terms. Ecology 16:284-307. Rabinowitz, A. 2002. Ground truthing conservation: why Turner, N.J., I.J. Davidson-Hunt, and M. O’Flaherty. 2003. biological exploration isn’t history. Conservation in Practice Living on the edge: Ecological and cultural edges as sources (3)4:20-25. of diversity for social-ecological resilience. Human Ecology Ramenofsky, A. 1987. Bodies of evidence: reconstructing history 31:439-463. through skeletal analysis. John Wiley and Sons. New York. Vale, T. (ed.). 2002. Fire, native peoples, and the natural Scott, J. M., P.J. Heylund, M.L. Morrison, J.B. Haufler, M.G. landscape. Island Press. Washington, D.C. Raphael, W.A. Wall, and F.B. Samson. 2002. Predicting Worster, D. 1994. Nature’s economy: a history of ecological species occurrences: Issues of accuracy and scale. Island ideas. Second Edition. Cambridge University Press, Press. Washington, D.C. Cambridge, United Kingdom. Spence, M.D. 1999. Dispossessing the wilderness: Indian

Illustration by David Hicks

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 147 ABSTRACTS

Abstracts

Breeding system and seed production of an endangered plant, Fritillaria gentneri, in southern Oregon

Kelly Amsbury and Bob Meinke Plant Conservation Biology Program, Oregon Department of Agriculture and Oregon State University, 2082 Cordley Hall, Corvallis, OR 97331 Emails:[email protected] and [email protected]

Fritillaria gentneri Gilkey (Gentner’s fritillary), one of Oregon’s most beautiful native wildflowers, is listed as endangered by both the Oregon Department of Agriculture and the U.S. Fish and Wildlife Service. A Recovery Plan for this species, including recommendations for population augmentation and reintroduction projects is currently being finalized by the USFWS from a draft plan developed by the ODA. Our three-year study evaluates the seed production capability of native populations of F. gentneri, both to determine the potential for the use of wild-collected seed in recovery projects, and to learn more about the breeding system of this unusual lily. Seed production has been reported intermittently since the species’ description in 1951, but doubts have been raised as to the parentage of the seed observed, as well as the specific identity. (F. gentneri is virtually indistinguishable from F. recurva when in fruit.) By performing controlled pollination treatments on positively identified plants, we hoped to unequivocally document the potential for seed production by F. gentneri. To prevent herbivory by deer, study plants were enclosed in large-mesh wire cages at six southwest Oregon study sites prior to the initiation of manual pollination treatments. In the first year of our study, no fruits were produced in response to 189 conspecific pollination treatments Ð treatments included selfing, open-pollination, and pollination with pollen from conspecific plants that were near neighbors, far neighbors, and from another population. However a capsule was produced on one of the seven plants pollinated with pollen from sympatric plants of Fritillaria affinis. In 2001, conspecific and interspecific pollination treatments were completed on 273 flowers; unfortunately drought prevented fruit set on all study plants. In the final year of the study, fruits were produced by 52% of F. gentneri flowers pollinated with pollen from F. recurva, 12% of F. gentneri x F. affinis crosses, and only 2.3% of conspecific matings. Open-pollinated plants also occasionally produced capsules, but as both F. gentneri and F. recurva are visited by hummingbirds, and these two species are sympatric in some sites the pollen parent of these fruits cannot be determined. Fruits contained an average of 87.3 seeds, with a mean of 17.5 of these containing apparent embryos. Pollination treatment did not affect the number of seeds produced per fruit, or the number of embryo-containing seeds. Fertility also varied in relation to the female parent; some plants were consistently fertile regardless of the pollen source with which they were treated, while others never produced capsules. This patchy fertility indicates that populations may be made up of individuals with varying chromosome numbers Ð further cytological and molecular studies are needed to answer this question.

These unexpected results suggest that F. gentneri may be of recent hybrid origin, with F. recurva and F. affinis as the putative parents. Future studies focusing on artificially hybridizing these two species, cultivating and observing the progeny produced and documenting pollinator movement between them will provide data essential for evaluating the possibility that F. gentneri originated for a recent interspecific hybridization event. Low seed production, combined with the inability to determine the male parent in open-pollinated progeny, limit the value of collecting native seed for recovery projects Ð as a result, development of transplant cultivation protocols should focus on asexual bulblet propagation.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 148 ABSTRACTS

Effects of fire on rare plants and vegetation in serpentine savanna and fen communities in southern Oregon

Darren Borgias The Nature Conservancy, Medford Field Office, 33 N. Central Avenue, Suite 405, Medford, OR 97501 Email: [email protected] Nathan Rudd The Nature Conservancy, 821 SE 14th Avenue, Portland, OR 97214 Len Eisenberg The Nature Conservancy, Medford Field Office, 33 N. Central Avenue, Suite 405, Medford, OR 97501

The Nature Conservancy and the US Forest Service cooperated in evaluating effects of a prescribed, fall-season fire on a Jeffrey pine savanna, woodland, and a Darlingtonia fen, in the serpentine Siskiyou Mountains at Cedar Log Flat Research Natural Area on the Siskiyou National Forest. The project was motivated by concern over perceived low abundance of the rare plant species, hypothesized to result from increased shrub and graminoid cover in the uncharacteristically long interval since the last burn on the site (ca. 90 years). The prescribed fire was conducted over September 28-30, 1997 under relatively mild conditions, with temperature ranging from 67oF to 85oF, relative humidity ranging as low as 27%, and winds light to 5 mph. The burn crew moderated fire behavior by adjusting ignition pattern to allow a relatively heterogeneous burn pattern, minimize large tree mortality and to maintain fire-line control. We measured changes with pre-burn data (1996, 1997) and post burn data (1998, 1999, and 2000). For the rare plants, Microseris howellii and Senecio hesperius, and their associated understory species we evaluated change in frequency using replicated (n=4) split block experimental design, and used a paired t-test of difference in frequency (burn - control) between 1998 and 1997. To evaluate changes in coarse-scale vegetation structure we used a point intercept approach (1064 points) on paired, pre and post-burn low-elevation, color aerial photos to determine cover by structural class. We also used the photos to determine the fate of 177 prominent individual trees. In the fen, we counted plants and measured the size of Epilobium oreganum in a pair of burn and no-burn plots.

Post-hoc power analysis indicated that for the rare plants and associated species our methods were able to detect with 90% power differences in treatments for species with at least 15-20% frequency. Perennial forbs showed the greatest positive response to the burn. Reproductive and juvenile stages of Microseris howellii increased in response to fire. Other perennial forbs to respond positively to the fire included Hastingsia serpentinicola, Galium ambiguum and Ranunculus occidentalis and Iris chrysophyllis. Carex spp. also responded positively to fire. Species demonstrating relative declines in burn plots included Senecio hesperius adults, Horkelia serricata and Danthonia californica. The short-term analysis generally demonstrates short-lived changes for these species, followed by trend toward returning to pre-burn condition. No impact of the fire could be detected for Festuca idahoensis, the most frequently occurring species in all plots. In the fen, Epilobium oreganum abundance increased in both burned and unburned plots; the percentage increase was larger in the burned macroplots for all but reproductive stages. For the coarse structure in the woodland and savanna, changes within 1.5% to 13.4% of the pre-burn cover estimates were detected with 90% confidence. In the lower slope Jeffrey Pine - Incense Cedar / Whiteleaf Manzanita woodland and Jeffrey Pine / Idaho Fescue savanna, live shrub cover was significantly reduced, cover of dead shrubs increased. Snags were significantly increased, yet tree cover was not significantly changed. Of the prominent trees tracked, 14% were converted to snags or down logs. Recruitment of new snags outpaced loss of existing snags for a 4-fold increase. In the mid-slope Jeffrey Pine - Incense Cedar / Huckleberry Oak woodland the tree, live shrub, and down log cover were significantly reduced, while the cover of dead shrubs, bare ground, and snags increased. The herbaceous cover did not change. Total tree cover was reduced nearly 58%, snags increased significantly, and tracked prominent live trees were reduced by 24%. The herbaceous cover, down logs, and bare ground and rock did not change significantly. For the upper slope with a Jeffrey Pine Ð Sugar Pine / Huckleberry oak Ð Pinemat Manzanita - Common Beargrass woodland, the tree and shrub cover were reduced, while the herbaceous cover, bare ground and rock, and dead shrub components increased. Total cover of trees was reduced by nearly 50%, while <10% of the tracked prominent trees were killed.

Increased herbaceous cover in response to fire did not occur as hypothesized in the lower slope woodland and savanna communities. Instead, resprouting shrub species, which dominated the understory, quickly began to recover their aerial canopy. Non-native weedy

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 149 ABSTRACTS species were not observed to be invading any of the units, and this is presumed to result from the atypical serpentine soil and the remote character of the site.

This study provides finer resolution insight into short-term vegetation response to fire in Jeffrey pine savannas, woodlands, and fens. Further research should investigate the extent to which live roots of the burned shrubs occupied the soil below the burned canopy, and the interactions below ground with propagules of herbaceous species. Other research should investigate the compatibility of spring season burns during a restoration phase for similar systems that have missed fire events over several return cycles.

Postglacial vegetation and fire history near Bolan Lake in the northern Siskiyou Mountains of Oregon

Christy Briles and Cathy Whitlock University of Oregon, Department of Geography, 1644 Arthur Place, Eugene OR, 97402 Emails: [email protected] and [email protected].

Analysis of pollen and high-resolution macroscopic charcoal from Bolan Lake, Oregon was used to reconstruct the postglacial vegetation and fire history of the northern Siskiyou Mountains. The Siskiyou Mountains are interesting biogeographically because they harbor high levels of floristic diversity resulting from an array of climate conditions and soil types. This heterogeneity supports many plant species and allows for the meeting of several vegetation boundaries. To understand the history of Siskiyou forest diversity, it was necessary to reconstruct postglacial vegetation and fire regimes and assess their response to past variations in climate. During the late-glacial period, cooler and drier conditions than present supported subalpine parkland vegetation and few fires. Wetter conditions after 14,500 years ago resulted in a closed forest of Abies and Tsuga mertensiana and more fires than before. Warm, dry conditions in the early Holocene led to the development of open Pinus and Quercus woodland and more frequent fires than before or at present. Cool and wet conditions in the late Holocene allowed Abies and Picea to become more abundant and resulted in low fire frequency. Modern forests of Abies and Pseudotsuga and the present-day fire regime were established ca. 2000 years ago. Variations in Holocene fire frequency at Bolan Lake are generally similar to those recorded in western Oregon and northern California and suggest a widespread response to large-scale changes in climate.

The return of the wolf to the Klamath-Siskiyou Region: implications for ecosystem conservation

Carlos Carroll Klamath Center for Conservation Research, P.O. Box 104, Orleans, CA 95556 Email: [email protected]

After an absence of several decades, the gray wolf has recently returned to Oregon. The return of this top predator to the state will strongly affect land management policy and ecosystem function. I use the results of habitat and viability models to answer questions such as: Will wolves be able to colonize and survive in the Klamath/Siskiyou ecoregion? What ecological role did they play historically, and can they be restored in sufficient numbers to restore that role? What land use changes will be needed to ensure their survival? What will their return mean for other species within the ecosystem, be they prey (deer and elk), competitors (coyote), forest carnivores (fisher), or other species such as birds that may be indirectly affected by the impact of wolves on ungulates and vegetation? Are wolves a good focal or umbrella species for conserving biodiversity? Although wolves are often associated with wilderness, they also appear to thrive in the industrial forest of the north central U.S. However, results from the western U.S. suggest that although much of the wolf population will be found outside protected areas, these areas remain the key to persistence of viable populations. Short-term prospects for wolf recovery in the Klamath/Siskiyou are good, but the rapid rate of landscape change due to development in southern Oregon threatens their long-term viability. Because they avoid rugged terrain, wolves will probably be most abundant in the northern portion of the ecoregion. Protecting wolf habitat would therefore do little to ensure survival of fisher populations. It is also unlikely that wolf conservation needs will indirectly protect hotspots of rare endemic species. Although it is unlikely that focusing on the wolf’s requirements alone will protect all facets of biodiversity, the wolf can help us identify landscape linkages that preserve landscape connectivity between the Klamath/Siskiyou and neighboring ecoregions.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 150 ABSTRACTS

The impacts of the Quartz Gulch Fire on fish and aquatic amphibian population dynamics in Glade Creek, Oregon

Jake Chambers Aquatic-Riparian Effectiveness Monitoring Program (AREMP), USDI Bureau of Land Management, 4077 Research Way, Corvallis, OR 97333 Email: [email protected]

Glade Creek is a tributary of the Little Applegate River, located in the Siskiyou-Klamath Mountains of Southern Oregon. In October 2000, a survey crew from the Aquatic and Riparian Effectiveness Monitoring Program (AREMP) conducted an extensive survey of the Glade Creek Watershed that included an electrofish survey of fish and aquatic amphibians. The data collected includes the abundance and length of fish and amphibians at ten randomly selected sample reaches in the Glade Creek Basin. In August 2001, lightning ignited the Quartz Gulch Fire outside the watershed boundary, which subsequently spread into the Glade Creek drainage, burning approximately 43% of the basin area. In October 2002, an AREMP survey team returned to Glade Creek and resurveyed the sample reaches to determine whether we could detect change in channel morphology, habitat and biota following a natural disturbance. The purposes of this study is to examine the impacts of natural disturbance on the spatial distribution and relative abundance of fish and amphibian communities over a short temporal scale, and to determine if there is a shift in size-classes of fish within each reach. Results indicate a change in the spatial distribution of fish and aquatic amphibians across all sample reaches, while the frequency of different size-classes of rainbow trout (O. mykiss) shifted to an uneven distribution. Analysis also shows a decrease in relative abundance and it appears to be correlated with fire intensity within the active mosaic of the burn.

Potential pollinators and insect visitors to threatened and endangered Fritillaria gentneri (Liliaceae) and closely related species Fritillaria affinis and Fritillaria recurva

Kathleen Donham 3776 Devils Garden Road, Medford, OR 97501 Email: [email protected]

Carol Ferguson Department of Biology, Southern Oregon University, Ashland, OR 97520 Email: [email protected]

Gentner’s fritillary is endemic to a restricted area in SW Oregon and listed as “endangered” by the US Fish and Wildlife. Efforts to artificially pollinate Fritillaria gentneri have not resulted in fruit or seed set, and there is speculation that the species is a hybrid of related sympatric species Fritillaria recurva and Fritillaria affinis. DNA analyses have not been completed, and F. gentneri pollinators have not been documented. Observations of flower visitors to F. gentneri and two putative hybrid parents, F. affinis and F. recurva, were made during the flowering season (February-April) 2002. Anna’s hummingbird (Calypte anna) was photographed moving from F. recurva to F. gentneri, and three mining bees were captured, each visiting one of the three fritillary species. The andrenids visiting F. recurva and F. gentneri were both carrying pollen on the ventral thorax. Queen bumblebees, Bombus vosnesenskii, were frequently seen exploring for nesting sites in the immediate vicinity, but were not observed visiting flowers. Dipteran and lepidopteran larvae were discovered feeding on flower parts in F. affinis and F. gentneri, respectively. Preliminary observations using fluorescent pollen analogue dye revealed no movement of dye from F. recurva to either F. gentneri or F. affinis. Further information about the habitat requirements of pollinators may be critical for establishing long-term recovery plans if F. gentneri is found to be capable of sexual reproduction.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 151

ABSTRACTS

A rare orchid, Cypripedium fasciculatum (Orchidaceae), breaks the rules for pollination

Carol Ferguson Department of Biology, Southern Oregon University, Ashland, OR 97520 Email: [email protected]

Kathleen Donham 3776 Devils Garden Road, Medford, OR 97501 Email: [email protected]

Cypripedium fasciculatum, the clustered lady’s slipper, is one of three Cypripedium species endemic to the western United States. C. fasciculatum is federally listed as a ‘species of concern’ and in Oregon as ‘critically imperiled’. Prior to this research the pollinator of Cypripedium fasciculatum was unknown. Since 1998 we have studied the pollination biology of this orchid at field sites in southwestern Oregon. Our research sought to determine the pollinator(s) of C. fasciculatum and to elucidate the relationship between the pollinator and the orchid. Floral phenology and insect activity were monitored between orchid emergence and fruit set. C. fasciculatum pollen was collected and characterized using light microscopy and scanning electron microscopy. Our results suggest a parasitic wasp (Order Hymenoptera, Family Diapriidae, Subfamily Belytinae, Genus Cinetus) pollinates C. fasciculatum in southwestern Oregon orchid sites. Stingless wasps are a large and diverse group of insects, and as a rule, have not been considered important as pollinators. Diapriid life history is not well known but these wasps have been documented to parasitize fungus gnat (Order Diptera, Family Mycetophilidae, Family Sciaridae) larvae and pupae. Sciarids have consistently been the most abundant insect captured in the Oregon Cypripedium field sites. Despite their abundance, fungus gnats have never been collected with C. fasciculatum pollinia on them. Diapriid wasps have consistently been the only insects collected with C. fasciculatum pollinia adhered to their thorax. Floral phenology of C. fasciculatum was characterized as having six distinct phenological stages based primarily on the condition of the orchid’s reproductive structures and to a lesser extent the condition of the perianth. C. fasciculatum flowers appear to be most receptive to pollinators during the third phenological stage when the reproductive structures of the flowers are situated to enhance the transfer of pollen from pollinator to receptive stigma surface and from an anther to the diapriids thorax when exiting the posterior openings of the labellum. Diapriid activity appears correlated with floral phenology for example diapriids were most abundant when the largest percentage of C. fasciculatum flowers was in the third phenological stage. C. fasciculatum offers no nectar reward. We hypothesize C. fasciculatum lures the wasp to enter the orchid labellum by emitting chemicals that mimic the odor of the diapriids’ host, the fungus gnat. C. fasciculatum’s floral morphology ensures that only those insects of the correct size (less than 3 mm) will successfully maneuver the one-way path through the flower. Future research will focus on achieving a better understanding of the natural history of the diapriid wasp, its fungus gnat host, and the relationship of both insects to C. fasciculatum. We will also conduct further research on characterizing the site of odor production in the orchid and characterizing the chemical constituency of the odor.

Turnover, rate of species accumulation, and persistence of species rich assemblages of butterflies in the Marble Mountain Wilderness, Siskiyou County, California

Robert F. Fernau Geography Graduate Group, University of California, Davis, CA 95616 Email: [email protected]

I present a new bootstrap resampling procedure to estimate rates of species accumulation (both absolute and proportional) with two summary statistics that may be used to: 1) identify which study stations have highly dynamic assemblages of mobile organisms vs. which stations are comparatively stable, and 2) estimate the adequacy of the sample size and the incremental benefit of additional samples. The results are compared with turnover and persistence equations to assess the consequences of a heterogeneous landscape for dynamic populations of butterflies in the Marble Mountain Wilderness. Broad ranges of species richness and temporal dynamics across study stations are exhibited. Cryptic biodiversity hotspots and non-cryptic biodiversity hotspots constitute the extremes of a temporal continuum. Cryptic biodiversity hotspots are localities with high species richness of mobile organisms whose occurrences require comparatively longer time frames to identify than non-cryptic biodiversity hotspots. Non-cryptic biodiversity hotspots are localities that require comparatively modest amounts of time to identify, but are nonetheless diverse. Results from rapid assessments provide a comparatively accurate characterization of non-cryptic localities. The data from this research is based upon 17 years of fieldwork in the Marble Mountain Wilderness.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 152 ABSTRACTS

Distribution and environmental/ habitat relations of five endemic plants associated with serpentine fens

Evan Frost Wildwood Environmental Consulting, 84 4th Street, Ashland, OR 97520 Email: [email protected]

Serpentine wetlands, commonly known as California pitcher plant (Darlingtonia californica) bogs or fens, provide the primary habitat for five rare plant species endemic to the Klamath-Siskiyou region: Waldogentian (Gentiana setigera), Oregon willow-herb (Epilobium oreganum), large-flowered rush-lily (Hastingsia bracteosa), purple-flowered rush-lily (Hastingsia atropurpurea) and western bog violet (Viola primulifolia ssp. occidentalis). The primary objectives of this study are to: 1) investigate habitat relations of all five taxa at both coarse (fen) and fine (plot) scales, and 2) determine what environmental or biological variables may be most important for conservation of the rare species and their habitat. Fen-(n= 27) and plot-scale (n=1,200) data were collected in 2001 and 2002 from sites located across southwest Oregon and northwest California. Initial data analysis suggests that the presence and/or abundance of target species at the fine scale are correlated with soil moisture, soil texture, topography and vegetative cover variables. In contrast, fen-scale variables appear to be poor predictors of target species presence. The implications of these findings for development of a multi-species fen conservation strategy will be discussed.

Phytophthora ramorum: the cause of Sudden Oak Death

Ellen Goheen USDA Forest Service, Southern Oregon Forest Insect and Disease Service Center, 2606 Old Stage Road, Central Point, OR 97502 Email: [email protected]

What is now known as Sudden Oak Death was first reported in 1995 in Mill Valley, California where rather striking and seemingly spontaneous death of tanoaks was observed. Since that time, thousands of tanoaks, coast live oaks, and black oaks have died in the forests of twelve coastal California counties. The disease was detected in July 2001 in southwestern Oregon in Curry County. Approximately 24 disease centers totaling 52 acres have been identified in a nine-square-mile area east and north of the community of Brookings. The disease has been found throughout Europe in nursery plants and landscape plantings. It has recently been found on nursery stock in both Oregon and California.

The causal agent, Phytophthora ramorum, is new to science and of unknown origin. Its host list is large and growing, including at least 25 species in 12 plant families. Plants currently known to be affected are: tanoak (Lithocarpus densiflora), coast live oak (Quercus agrifolia), California black oak (Q. kelloggii), Shreve’s oak (Q. parvula var. shrevei) canyon live oak (Q. chrysolepsis), Pacific madrone (Arbutus menziesii), California bay laurel/Oregon myrtle (Umbellularia californica), evergreen huckleberry (Vaccinium ovatum), manzanita (Arctostaphyllos manzanita), Pacific rhododendron (Rhododendron macrophyllum) and other rhododendron cultivars, California coffeeberry (Rhamnus californica), cascara (R. purshiana), toyon (Heteromeles arbutifolia), poison oak (Toxicodendron diversiloba), bigleaf maple (Acer macrophyllum), California buckeye (Aesculus californica), California hazelnut (Corylus cornuta), salmonberry (Rubus spectabilis), honeysuckle (Lonicera hispidula), coast redwood (Sequoia sempervirens), Douglas-fir (Pseudotsuga menziesii), western starflower (Trientalis latifolia), Kalmia (Kalmia latifolia), Viburnum spp. and Pieris spp.

Phytophthora ramorum appears to infect only aboveground plant parts (i.e., leaves, branches, and/or stems). It is not considered a root disease. Across the range of hosts, three different types of diseases can be described: a) Ramorum leaf blight resulting in non-lethal foliar infections, b) Ramorum tip blight which describes small branch infections that may or may not result in whole plant mortality, and c) Sudden Oak Death, whereby large cankers are found on the main stem of oaks and tanoaks and cause tree death.

Phytophthora ramorum is associated with moderate climates and moist conditions. It produces thick walled resting spores called chlamydospores, and deciduous sporangia containing swimming zoospores. Several understory hosts, such as California bay laurel and Pacific rhododendron, produce prolific inoculum on their leaves and are considered important in localized disease spread. Phytophthora ramorum has been isolated from infected plant materials, soil, rain water associated with infected overstory trees, stream water, bark, and wood. Pathways of spread include the movement of infected plant material and wind driven rain. Quarantines are in place to prevent the artificial movement of the pathogen. Oregon has undertaken an aggressive eradication and monitoring program against Phytophthora ramorum.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 153

ABSTRACTS

Fish and the Biscuit Fire

Tom Halferty NOAA Fisheries, 2900 NW Stewart Parkway, Roseburg, OR 97470 Email: [email protected]

Fires have been a part of the Klamath-Siskiyou bioregion for millions of years. All life forms either adapted to it or perished. Fish have survived quite well. In a conflagration some individuals may die from excessive heat, while others escape up or downstream. The winter following a big fire can be challenging as soil exposed to the weather erodes after duff layers are burned off and canopies no longer intercept raindrops. Fine sediments washed into streams can also smother eggs. However, those fines contain nutrients that increase the number of microfauna and invertebrates which may benefit juvenile fish the following spring. Large woody debris increases as burnt trees fall into streams, helping to trap gravel and cobbles which provide spawning habitat. Water flowing over large wood scours out pools that provide increased summer habitat. Within about three to five years, dead tree roots will have decayed, resulting in landslides that bring in more large trees, boulders, cobbles, gravel, and fines. The fines can again cause problems to redds, but once they are washed out by winter storms he habitat can be greatly improved over pre-fire conditions. In general, fire can be a rejuvenator of fish habitat.

When comparing the effects of the Biscuit Fire on fish with another large fire, the Hayman Fire of Colorado, a couple of geographic considerations come into play. Areas further east such as Colorado, Utah and Idaho get much of their rain during summer thunderstorms where large amounts of rain come down in very short time periods with great intensity. In SW Oregon, most of the rain comes in the winter over longer time periods with less intensity. The Hayman Fire occurred in an area of decomposed granite that is highly erodible. The summer after the fire, thunderstorms washed granitic sands down slopes generating debris flows that sent walls of mud down stream channels killing fish on a large scale and filling reservoirs with mud, sand, and burnt debris. Much of the Biscuit Fire area is very rocky which helped to protect thin soil layers from the massive types of erosion that occurred after the Hayman Fire. Thus in some respects, SW Oregon was lucky that due to climate and geology the fish were spared the massive sediment movement that often occurs after fires in other parts of the country.

Many times, the impacts of fire suppression do more harm to fish than the fire itself. Soil disturbance from bull dozer-created fire-lines, toxic fire retardants that land in the streams, burnout operations that get out of hand, and in general the impact of thousands of people working on the fire takes a toll. On the Biscuit Fire, to prevent the spread of Port-Orford-Cedar root-rot (Phytophthora lateralis), water from infected streams was treated with chlorine. Three separate fish kills were caused as water-tenders pulled away from the streams where they obtained water and accidentally spilled treated water back into the streams. Hopefully, over time, fire suppression techniques will become more environmentally friendly. Already, a different manner of filling tenders and adding chlorine is planned to avoid polluting streams. Non-toxic retardants could be developed. Even now, resource advisors from the local area work with the over-head teams who often come from out of state so that local resource needs are addressed as the fire suppression team does its job. As manual fuel reduction and prescribed fire are introduced to the landscape, wildland fire use can be considered, allowing fires to burn in some areas and be controlled at other places where fuel breaks are maintained to stop fire from reaching communities. People who build homes in the forest need to use non-flammable building materials and create a defensible space around their dwellings. Fire is a natural part of the ecosystem in SW Oregon. As humans, we can adapt to live with fire returning now and then.

Broad-leaved noxious weed abundance and distribution across the Cascade-Siskiyou National Monument, southern Oregon

Paul Hosten Bureau of Land Management, 3040 Biddle Road, Medford, OR 97504 Email: [email protected]

Analysis of the abundance and distribution of broad-leaved weeds indicates that roads, timber harvest, and high livestock grazing intensity contribute to weed proliferation across the Cascade-Siskiyou National Monument. Actual weed population counts were higher than expected weed population counts in close proximity to roads and livestock watering points. Observation of higher actual versus expected counts was limited to within 100 m of roads and trails, and to within 300 m of livestock watering points. Higher actual counts of weed populations were observed in the three highest livestock utilization classes. Actual weed population counts were not significantly higher within acutely disturbed areas relative to undisturbed. However, more recently treated conifer pole stands showed higher weed abundance

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 154 ABSTRACTS than old growth or mature timber stands. This indicates that recovery by woody vegetation shades out noxious weeds. The starburst shape of Canada thistle invasion at higher elevations indicates an active invasion process. Linear extensions along roads and creeks indicate pathways of spread. The association of weeds with roads, watering points, acute disturbance, and high livestock utilization ascertain the need for disturbance for the establishment of new weed populations. Based on the identification of soils susceptible to invasion, Canada thistle and yellow starthistle have not achieved their full possible range across the CSNM. The existence of disturbed and as yet uncolonized areas indicates an ongoing colonization process.

Native American management and prescribed burning of riparian zones

Frank K. Lake Pacific Traditional Ecological Knowledge Program, Oregon State University, 334 Strand Ag. Hall, Corvallis, OR 97331 Email: [email protected]

For centuries Native Americans of the Klamath-Siskiyou bioregion have lived in and managed riparian zones. Horticultural practices and prescribed burning influenced the productivity and diversity of the plant and animals species found in various riparian zones spatially and temporally. Historical and contemporary management and burning practices of Native Americans and the relevance to restoration and conservation of biodiversity will be discussed. Various riparian zones ranging from lowland valley bottoms to upper elevation forested creeks and springs will be reviewed. Fire effects in relation to ecological processes affected by western fire management, cultural management and burning, and impacts to flora and fauna will be covered.

Forgotten birds of the riparian system: monitoring stream-associated birds as a metric for watershed quality

Sherri L. Miller and C. John Ralph USDA Forest Service, Pacific Southwest Research Station, Redwood Sciences Laboratory, Arcata, CA 95521 Emails: [email protected] and [email protected]

Our study objective was to develop and test a survey technique and watershed scale design to monitor trends in aquatic foraging bird densities. The resulting methods and design will be used to examine the efficacy of quantifying the association of stream and watershed quality with trends in the bird population. In 2002, we integrated the study methods into an avian monitoring project on Trinity River, a major tributary to the Klamath River. We have completed over 120 km of surveys along randomly-chosen 2-km stream reaches in the Smith River watershed on Six Rivers National Forest in northern California, and over 480 km of bird surveys along the mainstem of the Trinity. We recorded measurements of stream and habitat characteristics, as well as observations of dippers and other stream species, such as Common Merganser, Belted Kingfisher, and Spotted Sandpiper. In addition to census and foraging data, we collected a large variety of physical and biological measurements of the stream and bank habitats to identify the factors most related to abundance. Dippers were the most consistently observed birds and are the most likely to provide a metric for monitoring watershed quality. Surveys were conducted on all stream orders and we found highest dipper densities on larger streams and main stems of the river. Dippers were present in riffles and rapids significantly more often than expected by the proportion of each habitat.

We also conducted an intensive color-banding and census effort along a 7-km reach of Hurdy Gurdy Creek, a Smith River tributary, where over three years we have banded most of the resident breeding population. We survey the stream once per month during breeding and dispersal seasons and have identified territories, located seven nests, documented triple-clutching, nest fidelity, and nest attendance by three or more birds. The knowledge we are gaining on how Dippers utilize their habitat will help us to design research on other watersheds.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 155 ABSTRACTS

Pacific lamprey redd densities in the Illinois River of southwestern Oregon

Richard K. Nawa Siskiyou Regional Education Project, 9335 Takilma Rd, Cave Junction, OR 97532 Email: [email protected]

Greg Bennet Siskiyou Research Group, Cave Junction, OR 97532

Lamprey declines have been documented in the industrialized areas of the Northern Hemisphere, primarily in the United States and southern Europe, but none of the 34 species have become extinct. Similar to Pacific salmon species, Pacific lamprey (Lampetra tridentata) show a declining trend in the southern and eastern portions of its range where human impacts to freshwater habitat are severe and cumulative. Lamprey redds in the Illinois River basin in southwestern Oregon were concentrated in lower reaches of tributary streams with perennial flows. Redd densities of over 20/mile were recorded in stream reaches with summer stream temperatures too high for optimum salmonid rearing (>23 degrees C). Lamprey redd densities averaged 6.1 redds per/mile on private lands. Public lands in cool temperature headwater areas averaged only 0.5 redds per/mile. Extended freshwater rearing of larval lamprey for up to six years and lack of mobility makes this species extremely dependent on perennial flows throughout the summer rearing period. Perennial flow is the most critical factor for juvenile lamprey rearing in valley bottom streams prone to losing surface flow due to drought or water withdrawal.

An ecological study of Preston Peak’s flora: establishing baseline data for climate change research on subalpine vegetation Jamie O’Donnell Environmental Education Program, Southern Oregon University, Ashland, OR 97520 Email: [email protected]

Mountain ecosystems show greater sensitivity to climate than many other ecosystems because of the high rate at which climatic factors shift along mountain gradients. These ecosystems, therefore, act as important indicators for climate changing. Based on a combination of palaeobotanical and plant physiological studies, researchers suggests that mountainous plant species will respond to changing climate by migrating upward along altitudinal gradients to stay within suitable conditions where they can successfully compete for resources. This project establishes baseline data on Siskiyou Mountain flora, a highly diverse flora possessing numerous species endemic to the region as well as species at both their northern and southern extension. Species composition and cover data was collected from 83 100-m2 plots sampled from a 240-acre study site on the upper west face of Preston Peak in the Siskiyou Wilderness. A comprehensive species list was also generated for the entire site to include species not found in plots and to establish a record of species located on the upper 50 meters of the summit where inaccessible terrain precluded the establishment of plots. A total of 245 species of vascular plants were documented from the study site with 197 documented from 100-m2 plots. Many species were found in only one or two plots demonstrating high diversity and variance as a result of complex spatial patterns of plant distribution. Multivariate analyses indicate that elevation plays a significant role in community structure and demonstrates an inverse correlation with species richness. Hierarchical Cluster analysis of plot data resulted in a community classification possessing nine groups that describe typical plant species assemblages on the landscape. The site’s diversity and variability complicate predicting impacts of climate change on plant distributions in the Siskiyou Mountains. Edaphic factors and microhabitats may buffer some community types from changing climate, while other community types may shift altering community interactions and ultimately species composition. Regardless, climate change will impact plant distribution and diversity. The magnitude and temporal scale of the climatic change in conjunction with the specific characteristics of terrain and vegetation in the Siskiyou will determine the types and extent of change to vegetation.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 156 ABSTRACTS

Analysis of environmental factors limiting Pacific yew occurrence and regeneration in southwestern Oregon

Stanley Scher Native Yew Conservation Council, P.O. Box 9456 Berkeley, CA 94709 Email: [email protected]

The Pacific yew (Taxus brevifolia), a small to medium, shade-tolerant, understory tree, ranges from northern California to southern Alaska. During the early 1990s, bark harvested from Pacific yew on federal, state, and private lands in the Pacific Northwest served as a primary source of the anti-cancer agent paclitaxel (Taxol¨). Early yew bark harvests approached 1.6 million pounds per year, and resulted in a marked reduction in native yew populations. Natural stands of Pacific yew continue to be threatened by commercial harvests. The USDI-BLM recently issued permits to harvest four million pounds of Pacific yew boughs from late successional reserves in southern Oregon for production of generic Taxol. Although regeneration programs were advocated to restore Pacific yew populations to pre-harvest levels, previous regeneration studies were limited to a small number of sites in the Klamath region of northern California Forest. To extend our understanding of Pacific yew regeneration, study sites on the Rogue River National Forest (RRNF) in southwestern Oregon were selected. We report here, a preliminary analysis of abiotic and biotic environmental factors that promote or limit Pacific yew regeneration on riparian forest sites.

Studies on the Six Rivers National Forest provide evidence that proximity to water, soil water-holding capacity, seral stage of the overstory canopy, and other environmental factors influence the probability of yew occurrence in forested watersheds. Branches falling from adjacent overstory trees may facilitate clonal regeneration of yew. On the Ashland Ranger District and at study sites in the Klamath region, ungulate browsing strongly diminishes the probability of yew regeneration. Harvesting overstory canopy trees and adjacent shrub species may alter yew habitat by increasing access to browsing ungulates. Where Pacific yew shares habitat with Port Orford cedar (Chamaecyparus lawsoniana) or other hosts to Phytophthora, sensitivity to fungal infections may also play a role in yew occurrence and regeneration. In developing a predictive model, the ability to assign weights to each environmental factor facilitates calibrating the model for different physiographic provinces. This approach is expected to enhance the analysis and assessment of Pacific yew occurrence and regeneration across its range.

Impacts of the Quartz Gulch Fire on stream channel characteristics within Glade Creek

Ted Sedell Aquatic and Riparian Effectiveness Monitoring Program (AREMP), USDA Forest Service, 4077 SW Research Way, Corvallis, OR 97333 Email: [email protected]

In October of 2000 and 2002, the Aquatic Riparian Effectiveness Monitoring Program (AREMP) performed extensive surveys to catalog major watershed characteristics in the Glade Creek. Glade Creek is a tributary of the Little Applegate River located in the Klamath- Siskiyou Mountains in Southern Oregon. The frequency of log jams, pools ≥1 meter in depth, and bank stability were among the major characteristics documented. Bank stability was measured as the frequency of areas that were affected by bank erosion and mass failure. Pools were measured for maximum depth of ≥ 1 meter and log jams were classified as containing ≥ 3 pieces of wood with a minimum size of 0.3 meter DBH and 3.0 meters long. In August of 2001, lightning ignited the Quartz Gulch Fire, which burned approximately 43% of the Glade Creek watershed. A comparison of the two surveys revealed an increased frequency of log jams and bank instability, whereas frequency of pools of ≥1 meter decreased. Spatial analysis of the data indicates that the majority of the fire’s impact on these variables occurred in the low to medium fire intensity areas.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 157 ABSTRACTS

Matrix habitat effects on the health of patchy plant populations: serpentine plants in a changing landscape

Priya Shahani Department of Ecology and Evolutionary Biology / EMS A316, University of California, Santa Cruz, CA 95064 Email: [email protected]

The loss and fragmentation of habitat through land conversion are primary causes of reduced ecological health and increased biodiversity loss. However, little work has explored the implications of these alterations for population health of species restricted to habitat patches embedded in heavily manipulated landscapes. My work focuses on a system of serpentine rock outcrops embedded in a highly varied landscape in the Siskiyou National Forest, ranging from managed forests in various stages (from uncut old-growth Douglas fir forest to 15 year old ‘seed tree retention stands’) to less-harsh serpentine soils. Through this work, I am examining the ways in which human alterations of forested lands may exert strong, but indirect, cascades of effects in neighboring serpentine rock outcrops. Specifically, I am interested in how plant population health in undisturbed serpentine rock outcrops may be impacted by management of surrounding matrix lands, through changes in pollinator behaviors and communities and also in abiotic conditions. I explore how contrasting matrix lands affect pollinator community composition and behavior, subsequent reproductive success of serpentine plants, individual plant growth and survival, and overall plant population health. I have focused my work on Sedum laxum, a common local-endemic serpentine plant. Understanding how the effects of habitat changes propagate across landscapes will enable better predictions of the broader consequences of localized land use plans, and is critical for informed management decisions that better merge resource use and preservation in many varied landscapes. While I focus this work on one relatively common serpentine plant species, my overarching goal is to provide an awareness of the ways in which many other patchy plant population may be impacted by such matrix effects, thereby improving management plans for a range of serpentine plant species in the Klamath-Siskiyou region as well as in other patchy habitat systems.

Atlas of rare plant species on Lake Peak, Oregon: baseline data for global climate change monitoring

Lori Sims Medford District BLM and Siskiyou Field Institute, 3040 Biddle Rd, Medford, OR 97504 Email: [email protected]

Global climate change is expected to cause massive shifts in plant distribution from increased temperature, changes in precipitation, and snowmelt. Rare plants, ecosystems, and species interactions are all threatened by these shifts. Studying the effects of global climate change is hindered by the lack of baseline data. To create baseline data for global climate change research, this study was designed in two parts. This study was a comprehensive inventory of vascular plants and bryophytes of the Lake Peak Study Area. This study, in addition, mapped rare species with the intent of providing baseline data for future monitoring. I identified 343 species of vascular plants and 99 non-vascular species at Lake Peak. I mapped 16 rare plant species and created an atlas of these maps. Basic abiotic background information of the area was included in the atlas and used to analyze abiotic habitat of each species. I determined that 15 of the 16 species were immediately threatened by global climate change. Global climate change may threaten the remaining species at some point in the future. Monitoring these populations will determine if elevation shifts in distribution caused by global climate change are occurring in the Siskiyou Mountain flora.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 158 ABSTRACTS

The Biscuit Fire: vegetation changes and post-fire rehabilitation

Cecile Shohet Illinois Valley Ranger District, 26568 Redwood Hwy, Cave Junction, OR 97523 Email: [email protected]

Wayne Rolle Rogue River National Forest, P.O. Box 520, Medford, OR 97501 Email: [email protected]

This summer’s half-million acre Biscuit fire was the largest wildfire in Oregon in the last 100 years. Short-term effects to native plant species and communities are described. Initial rehabilitation of 300 miles of fire line and some sensitive interior areas has already been completed. Rationale used in choosing areas, treatments types, and plant materials for the initial rehabilitation is presented. Proposed longer-term ecological restoration activities, timber salvage and monitoring are being evaluated now by the agencies.

Uncertain harvest: patterns of acorn production in interior southwest Oregon and far-northern California

Donn L. Todt Ashland Parks Department, 621 Altamont Street, Ashland, OR 97520 Email: [email protected]

Nan Hannon Landmark Research

A longitudinal study of acorn production along the Oregon-California border demonstrates the variability of the acorn crop produced by two species of oaks (Quercus kelloggii and Q. garryana). Variability in production correlates with masting cyclicity that is, in turn, influenced by climatic trends. Understanding the diachronic variability and trends of oak masting has many practical applications in the fields of wildlife management, landscape ecology, conservation biology, and landscape restoration. Our primary focus, however, is to report a fourteen-year regional record of variability in acorn production and to suggest its applicability to understanding the cultural ecology of regional pre-contact Native American peoples.

Overview of the history of public lands and the Siskiyou National Forest: the early years and the Oregon land fraud

Greg Walter P.O. Box 1668, Cave Junction, OR 97523 Email: [email protected]

This presentation chronicled the establishment of the laws and organizations that divided and disseminated the landscape of the American West to public and private stakeholders. The development of the General Land Office and Department of Agriculture were outlined as well as the ultimate establishment of US Forest Reserves which were withdrawn for conservation purposes. I discussed how the Oregon Land Fraud Trials impacted the establishment of the Siskiyou National Forest and other national forests in the West. The land fraud trials resulted in the indictment of three of the four US representatives, the US land commissioner and the attorney general for the state of Oregon. This chain of events precipitated the Siskiyou Forest Reserves’ eventual designation as a national forest. The presentation provides relevant history that gives a context for public land laws and how they pertain to forest management, fire suppression history, mining, grazing and water rights.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 159 ABSTRACTS

Basin-scale restoration planning and monitoring in southwest Oregon

Debra Whitall U.S. Forest Service, 645 Washington St., Ashland, OR 97520 Email: [email protected]

Implementing restoration models that build social capital are crucial for the long-term stability of ecosystems, industries, communities, and ultimately, our society. Social capital refers to connections among individuals and the trust and reciprocity that arise from them. The Rogue Basin Restoration Technical Team, comprised of community and government agency representatives, has been chartered to develop a comprehensive, strategic plan for identifying high priority restoration needs across jurisdictional boundaries. The team developed this strategy and is currently involved in the implementation and analysis of an Ecological Management Decision Support (EMDS) model. The Rogue Basin Restoration Technical Team will use the EMDS model to identify and prioritize restoration and monitoring opportunities throughout the Rogue River and south coast basins in southwest Oregon. The results of this effort will be available to any stakeholders wishing to use it and will be accessed via inquiry and online.

Guiding principles of ecologicallyÐsound natural resource management in fire-adapted ecosystems of the West: developing a scientific consensus Cindy Deacon Williams Headwaters, 84 4th St., Ashland, OR 97520 Email: [email protected]

Jerry F. Franklin College of Forest Resources, University of Washington, Seattle, WA 98195 Email: [email protected]

Jack E. Williams AuCoin Institute, Southern Oregon University, 1250 Siskiyou Blvd., Ashland, OR 97520 Email: [email protected]

Dominick A. DellaSala World Wildlife Fund, 116 Lithia Way, Suite 7, Ashland, OR 97520 Email: [email protected]

During the summer of 2002, the Klamath-Siskiyou Ecoregion of southwestern Oregon and northwestern California, like much of the western United States, experienced extensive forest fires. These large-scale fires triggered social and political interest in developing and implementing forest management strategies to prevent future fires and avoid the destruction that much of the public and their political representatives perceive is caused by such extensive forest fires. In response to these interests and perceptions, a cross-disciplinary meeting of scientists was convened for the purposes of developing ecologically-sound principles to 1) guide forest management efforts to restore fire to its appropriate ecological role, 2) guide management responses to fire once it has ignited, and 3) guide post-fire recovery efforts. While the guiding principles for ecologically sound natural resource management discussed during this scientific workshop focused on issues in the Klamath-Siskiyou Ecoregion, the fundamental concepts are applicable to many fire-adapted ecosystems with mixed severity fires throughout the West.

In contrast to the widespread public belief that forest fire results in ecological damage, it was the consensus of the scientific group that fires in this region generally burn in a mosaic of low, medium and high severity, are essential to maintaining the biological integrity of natural forests, and can be particularly important to the integrity and function of riparian and stream communities. Forest management decisions, whether for actions before, during, or after a fire, should reflect the reality that forest fires will occur in the future and are a desirable disturbance at the landscape scale. It also was the consensus of the group that ecological principles should be the primary guide

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 160 ABSTRACTS for forest fire management in more wildland settings, whereas the social desire for defense of human communities appropriately should be the primary guide for fire management in urban interface zones. The group also found that few post-fire landscapes are allowed to recover naturally, despite the fact that seeding, planting and other post-fire restoration practices generally are unnecessary and often are ecologically counterproductive.

Potential impact of the Sudden Oak Death pathogen (Phytophthora ramorum) on the Siskiyou flora

James W. “Djibo” Zanzot Plant Pathology Department, University of California, Davis, 1 Shields Avenue, Davis, CA 95616 Email: [email protected]

Jennifer L. Parke Oregon State University, Department of Botany and Plant Pathology 2082 Cordley Hall, Corvallis OR 97331

Phytophthora ramorum, the causal agent of Sudden Oak Death, has been found near Brookings, Oregon, prompting eradication efforts and implementation of a quarantine on a nine-square-mile area in Curry County. In California, this pathogen has a broad host range that currently includes 22 species of trees, shrubs, and herbs. The potential for SOD spread in Oregon raises questions about disease epidemiology relative to the situation in California. The Klamath-Siskiyou Range defines the northern edge of the California floristic province, with a highly diverse flora and many new potential host species. To assay the potential susceptibility of plant species to P. ramorum foliar blight, we employ a detached leaf assay. Leaves are dipped half way into a suspension of zoospores, petiole down, and incubated in a moist chamber for seven days (dH20 control). Necrotic lesion area is then assessed with image analysis software. Known hosts are included in the assays as positive controls. Priority of screening goes to species found in associations where the pathogen has been found (tanoak associations with high annual precipitation and mean annual temperature), and to rare ericaceous plants of Curry, Josephine and Jackson Counties.

Many species develop disease symptoms in this assay, indicating some level of susceptibility to Phytophthora ramorum. The question of how environmental conditions will affect disease progress in situ cannot be answered by these data. However, the data reinforce the need to look at species other than known hosts when surveying for new disease loci in Oregon.

PROCEEDINGS OF THE SECOND CONFERENCE ON KLAMATH-SISKIYOU ECOLOGY 161