Research and Resource Management at Ardeidthe

◗ Heron and egret

nesting colonies

disturbance

patterns

◗ Invasive Spartina

marsh protection

◗ Acorns and

ecosystems oak

woodland

restoration

◗ A management

challenge raven

predation

◗ Sudden Oak Death

understanding

impacts 2002 the ARDEID 2002

In this issue A Safe Place to Nest: Disturbance patterns in heronries ◗ by John P. Kelly ...... page 1 Invasive Spartina: The challenging changeling ◗ by Katie Etienne...... page 4 Acorns and Ecosystems: Fourteen years of oak woodland restoration at the Bouverie Preserve ◗ by Rebecca Anderson-Jones ...... page 6 Resistible Forces? Raven predation in heronries ◗ by John P. Kelly ...... page 8 Sudden Oak Death Understanding the impact of a new epidemic on our forests and woodlands ◗ by Daniel Gluesenkamp...... page 10 In Progress: Project updates ◗ ...... page 13

Cover photo: Great Blue Heron by Philip L. Greene ◗ Ardeid masthead Great Blue Heron ink wash painting by Claudia Chapline

Audubon Canyon Ranch The Watch Research and Resource The following list includes ACR field observers and habitat restoration Management volunteers since the previous Ardeid. Please call (415) 663-8203 if your name should have been included in this list. Executive Staff Skip Schwartz, Executive Director PROJECT CLASSIFICATIONS: John Petersen, Associate Director B = Breeding Bird Study at Bouverie Preserve C = Common Raven Study Yvonne Pierce, Administrative Director H = Heron/Egret Project M = Livermore Marsh Monitoring N = Newt Staff Biologists Survey P = Photo Points R = Habitat Restoration S = Shorebird Project W = Tomales Bay Waterbird Census John Kelly, Ph.D., Director, Research & Resource Management Dawn Adams (HS); Sarah Allen (S); Jennie (H); Art Magill (R); Lyn Magill (R); Roger Rebecca Anderson-Jones, Bouverie Preserve Anderson (H); Marie Anderson (R); Marlowe (W); Chris Mc Auliffe (H); Pat Katie Etienne, Research Coordinator Christina Attwood (R); Audubon Canyon Mclorie (N); Diane Merrill (H); Mike Ranch - Bouverie Preserve Juniper Club George (R); Frank Herrick (R); Hannah Daniel Gluesenkamp, Ph.D., Habitat Protection and (N); Autodesk employees (R); Bob Baez Hindley (R); Stuart Hindley (R); Nita Hon Restoration Specialist (SW); Norah Bain (H); Michele Baker (R); (R); Jim Hubberts (R); Leena Ilmarines (R); Gwen Heistand, Preserve Sharon Bale (R); Melissa Banks (S); Tom Barbara Janis (R); Rosemary Jepson (R); Art Land Stewards Baty (W); Anne Baxter (H); Caitlin Bean Magill (R); Lyn Magill (R); Roberta Maran (W); Gordon Bennett (S); J. Bjork (S); (R); Sandy Martensen (R); Bill Masters (R); Bill Arthur, Bolinas Lagoon Preserve Louise Bielfelt (N); Gay Bishop (HS); Sara Diane McLure (R); JoAnne Meroney (R); David Greene, Tomales Bay properties Blauman (H); Mike Blick (R); Julie Peter Meyerhof (R); Jean Miller (H); James John Martin, Bouverie Preserve Blumenthal (W); Len Blumin (R); Patti Miramontez (R); Anne Murphy (H); Dan Blumin (HR); Ellen Blustein (M); Noelle Murphy (W); Karen Nagle (BH); Wally Field Biologists Bon (HN); Janet Bosshard (HS); Phillip Neville (H); Terry Nordbye (S); Gordon Ellen Blustein Bowman (R); Emily Brockman (HSW); Nunnaly (R); John O’ Connell (H); Nathan Farnau Janet Bruno (R); Ken Burton (W); Denise Brandon Orr (R); Julie Osborn-Shaw (R); Cadman (H); David Cain (R); Pam Cain (R); Peggy Osterhamp (R); Mari Ortwerth (HS); Lauren Hammack Barbara Carlson (N); Bill Carlson (N); Kate Richard Panzer (H); Patagonia employees Mark McCaustland Carolan (S); Mike Caselli (R); Ann Cassidy (R); Joelle Peebles (R); James Peebles (R); Mari Ortwerth (H); Dave Chalk (R); Roberta Chan (R); Kathy Peterlein (S); Myrlee Potosnak (H); Marie Claire Chapman (R); Steve Cochrane Grace Pratt (N); Jorge Presser (S); Nancy Research Associates (R); Carole Connell (H); Jaye Cook (S); Phil Proctor (R); Clark Rector’s Boy Scout Troop Jules Evens Cordle (R); Cynthia Corn (R); Helen (R); Jeff Reichel (H); Linda Reichel (H); Grant Fletcher Cravens (R); Catherine Cumberland (R); Jerry Ricard (R); Lucy Ricard (R); Robin Rachel Kamman Rig Currie (S); Nancy Dakin (R); Karen Richard (R); Rudi Richardson (W); Cheryl Davis (H); Melissa Davis (HS); Nick Riley (H); John Risberg (R); Sally Risberg Flora Maclise Degenhardt (H); Nancy DeSousa (R); (R); Brian Roberts (R); Barbara Rosen (R); Helen Pratt Carolyn Dixon (H); Roberta Downey (R); Glenda Ross (R); Liz Ruellan (R); Ellen Rich Stallcup Joe Drennan (W); David Easton (H); April Sabine (HS); Don Sanders (H); Marilyn Resource Management Associates Eberhardt (R); Ronnie Estelle (S); Serge Sanders (H); Fran Scarlett (H); Marie Etienne (HS); Cathy Evangelista (H); Jules Schabroeck (R); Elisabeth Schlosser (R); Len Blumin Evens (SW); Nathan Farnau (HSW); Katie John Sciocchetti (R); Craig Scott (SW); Roberta Downey Fehring (SW); David Ferrera (W); Binny Sierra Club “Outings” (R); Paul Skaj (W); Fischer (H); Grant Fletcher (HPSW); Ginny Joe Smith (HW); Craig Solin (S); Betsy Scientific Advisory Panel Fletcher (HPS); Johina Forder (R); Carol Spann (R); Anne Spencer (S); Rich Sarah Allen, Ph.D. Fraker (H); Jamie Freymuth (R); Audrey Stallcup (S); Jean Starkweather (H); Peter Connors, Ph.D. Fry (R); Randy Fry (R); Tony Gilbert (HSW); Elizabeth Stephens (S); Michael Stevenson Jules Evens Keith Gish (H); Beryl Glitz (H); Dohn Glitz (SW); Sandy Stoddard (R); Ron Storey (H); (H); Ellen Goldstone (N); ) Brindy Greene Grace Svitec (R); Cathleen Swanson (R); Kent Julin, Ph.D. (W); Philip Greene (H); Lauren Hammack Lowell Sykes (HSW); Judy Temko (HS); Greg Kamman (M); Madelon Halpern (H); Diane Hichwa Janet Thiessen (H); Gil Thomson (H); Rachel Kamman (BHS); Philip Hughes (H); Jeri Jacobson Dodie Thomson (H); James Titus (N); Sue Chris Kjeldsen, Ph.D. (H); Gail Kabat (W); Lynnette Kahn (HS); Tredick (H); John Trott (R); Norma Vite (R); Patty Karlin (H); Carol Keiper (SW); Marion Tanis Walters (HS); Sarah Warnock (S); Sue Doug Oman, Ph.D. Kirby (N); Richard Kirschman (SW); Ellen Weingarten (R); Eric Eisenberg (R); Heather Helen Pratt Krebs (H); Carol Kuelper (S); Alexis Lee (H); White (H); Rosilyn White (W); Tom White W. David Shuford Laura Leek (W); Gail Lester (W); Keith (W); Diane Williams (S); Phyllis Williams Rich Stallcup Lester (W); Robin Leong (H); Eileen Libby (H); Ken Wilson (HW); Jon Winter (H); Wesley W. Weathers, Ph.D. (H); Eric Lichtwardt (HS); Flora Maclise Rollye Wiskerson (H); Angie Wolfow (H). 2002 the ARDEID page 1

Disturbance patterns in heronries A Safe Place to Nest

by John P. Kelly

Table 1. Types of human or other disturbances observed or inferred at northern San Francisco Bay area heronries, 1990-2001.

Rainstorm Windstorm Power line interference Building construction Fence construction Removal of nest tree(s) Removal of nests Tree trimming or chipping Logging and tractor activity Firecrackers Army helicopter Hot-air balloon Spraying nests with garden hose Reflective (Mylar) ribbons as deterrents Shooting with 22 caliber rifle Skeet shooting Car doors slamming Trucks in area Interested human visitor Incidental human activity Investigator disturbance PHILIP L. GREENE Dog n explosion of flapping wings proximity to people, and direct intrusion Domestic or feral cat breaks the quiet morning air at and indirect disturbance caused by Common Raven A ACR’s Picher Canyon. The herons human activities have resulted in adverse American Crow ascend quickly into the air, circling, necks impacts on nesting. Red-tailed Hawk extended. Some land in nearby trees Since 1990, ACR’s regional Heron and Red-shouldered Hawk before gliding gradually back to their nest Egret Project has documented many Swainson’s Hawk sites. Herons and egrets are most likely to sources of disturbance to heronries across Golden Eagle exhibit fly-ups early in the nesting season, five northern counties of the San Bald Eagle in apparent response to perceived danger Francisco Bay area (Table 1). Any intense Western Gull but often with no obvious threat or stimu- or repeated disturbance can cause birds Unknown owl lus. Because a safe place to nest is a fun- to abandon a colony site permanently. Osprey damental requirement for successful However, it is not yet clear whether Unknown avian predator breeding, the conservation of herons and heronries are more strongly affected by Non-native red fox egrets must consider potentially adverse human interference or by other sources of Raccoon effects of nesting disturbance. To what disturbance. To complicate the matter, Unknown predator extent does human or other disturbance the likelihood of disturbance by native or threaten heronries? introduced nest predators may be Nesting colonies of herons and egrets enhanced or diminished by human activi- as you approach a colony, and retreat if are spectacular in their beauty. They have ty or human alteration of habitats. there is any such suggestion of distur- attracted human interest for thousands of To avoid investigator disturbance, ACR bance. Our best approach is to treat these years, inspiring admiration as well as observers at heronries follow these guide- beautiful birds with the cautious respect worldwide exploitation of their feathers lines during field investigations: Our first one should assume when encountering for decoration and their eggs and young concern is for the birds. Be cautious.Watch other cultures, nations, or worlds. for food. Many heronries occur in close for alert postures and listen for alarm calls Continued on page 2 page 2 the ARDEID 2002

Testing disturbance thresholds simple. The variability among sites is Academic Press). Herons and egrets nest- everal years ago, we measured the impressive (see 95% percentile ranges in ing in open habitat or in isolated trees intraseasonal pattern of disturbance Figure 1). At some sites, herons and egrets tend to react earlier and more intensely to Sresponses at 23 nesting colonies exhibit considerable tolerance to human disturbance. However, this general pat- across the Bay Area. During each trial, an activity and can be approached at close tern cannot reliably predict the sensitivity observer approached a colony on foot at a range—even by walking directly under of particular colonies. Heronries in open steady pace. We marked the position of nest trees—without exhibiting a distur- and isolated patches of trees in Suisun the approaching person when the first bance response. In contrast, birds in Marsh and northern Sonoma and Napa heron exhibited alert behavior and also other colonies will flee if humans counties vary substantially in their when the first heron flew from a nest site. approach within 200 m or more. These responses to approaching humans. We then measured the distance of each differences might partly reflect a capacity Buffer zones established to protect marked location from the colony. We for habituation. Some investigators have nesting herons and egrets from human repeated the trials at monthly intervals, even argued for systematically increasing activity are critical to the effective man- but avoided the early courtship period human activity near colony sites to stimu- agement of many heronries. Several sci- when arriving birds are extremely sensi- late habituation and resilience to distur- entific investigators have attempted to tive to disturbance (respond at greater bance events (Nisbet 2000, Waterbirds 23: identify general rules of thumb for estab- distances). Our results indicated that the 312-332). However, habituation has not lishing buffer zones, based on distur- responses of birds varied with stages in been adequately studied in herons and bance distances such as those shown in the nesting season (Figure 1). egrets. Therefore, its importance remains Figures 1 and 2. In general, minimum rec- A similar pattern of responses to other hypothetical and differences in tolerance ommended buffer zones of at least 200 m types of intrusions was found by Diana to disturbances among heronries cannot (based on behaviors exhibited in Vos and others (1985, Colonial Waterbirds be attributed clearly to habituation. response to two approaching humans; 8: 13-22) of Colorado State University, One pattern, however, has become Erwin 1989, Colonial Waterbirds 12: 104- although they did not interpret the appar- clear. The reactions of breeding herons 108) agree well with our estimates in the ent mid-season increases in sensitivity and egrets to disturbances depend Bay Area. Effective buffer zones should be (Figure 2). Studies of different types of strongly on the habitat structure of the based on upper 95th percentile of disturbance have shown consistently that colony. In our study, observers often observed (standard normal) flushing dis- heronries are less disturbed by approach- approached incubating herons at close tances plus 40-50 m because birds ing boats than by terrestrial intrusions range in heronries with densely vegetated become agitated before intruders cause a (Rodgers and Smith 1995, Conservations habitat without causing them to flush response (Rodgers and Smith 1995), plus Biology 9: 89-99). from their nests. The importance of dense an additional 100 m to protect colony Predicting the effects of disturbance at vegetation as a barrier to disturbance has sites early in the nesting season when any given colony site, however, is not so also been reported for herons along the birds are first courting and establishing upper Mississippi River (Thompson 1977, nest sites (Erwin 1989). Although such Proc. Colonial Waterbird buffers seem to provide a conservative Group 1: 26-37), in Colorado zone of protection, they remain arbitrary (Vos et al. 1985), in Florida and fail to address differences in toler- (Rodgers and Smith 1995), ance among colony sites. and in Europe (Hafner 2000, Becky Carlson and Bruce McLean pp. 202-217 in Kushlan and (1996, Colonial Waterbirds 19:124-127), of Hafner, Heron Conservation, John Carroll University in Ohio, found

Figure 1. Average distance associated with responses of Great Blue Herons to an observer approaching on foot at 23 heronries in the San Francisco Bay area. Error bars indicate 95th percentiles of standard normal distances. Colonies are most easily disturbed when at least some individuals are still in the pre-laying/courtship phase (March). Birds become more site-tenacious as they settle into the incubation phase (March-April). As nestlings grow large and begin to thermoregulate, adults may temporarily flee or alter their Figure 2. Average distance at which experimental intrusions caused behavior without significantly neglecting their young (May). Toward two or more Great Blue Herons to fly from nests at Fossil Creek the end of the nesting season, adults are rarely present at their Reservoir, Colorado; adapted from Vos et al. (1985, Colonial Water- nests; nestlings are large and alert to closely approaching observers birds 8:13-22). Number of trials is indicated above each bar. but unwilling to flee the safety of the nest (June). 2002 the ARDEID page 3

times. A colony site in Vedanthangal, India, has been actively protected since 1790 (Hafner 2000). Herons and egrets have nested at ACR’s Picher Canyon since at least 1941 and may have nested there for as long as 100 years (Pratt 1983, Western Birds 14:169-184). At other colony sites in our region where nesting herons or egrets are not disturbed when observed at close range, we have been pleased to encourage carefully managed educational opportunities for local residents. In other cases, intense disturbance has resulted only in temporary responses without causing nest failure or colony site abandonment. Unfortunately, very few studies have actually measured the effects of disturbance on reproductive success or seasonal occupancy of colony sites. We have seen colony sites in the Bay Area abandoned after interference involving construction activities, tree trimming, and harassment by Golden Eagles, Common Ravens, and humans. In west- ern Santa Rosa, a Snowy Egret and Black- crowned Night-Heron Colony moved four PHILIP L. GREENE times in five years in response to inten- tional harassment by humans. To what that the type rather than the width of turbance are suitable. A good example of extent does occasional disturbance affect buffer zone was most strongly associated this can be seen at the Great Blue Heron the productivity of heronries in the Bay with nest success in Great Blue Herons. colony on “Heron Island” in Stowe Lake, Area? We may soon have accumulated The most productive heronries were in San Francisco’s busy Golden Gate Park. enough data to find out. those that were isolated by moat-like Nevertheless, Bryan Watts and Dana Disturbances once considered to be water barriers or fencing, presumably Bradshaw (1994, Colonial Waterbirds minor could become major. Recent popu- providing protection from predators or 17:184-186), of the Center for lation growth of ravens across the San human intruders, rather than those iso- Conservation Biology, College of William Francisco Bay area, possibly related to lated simply by greater distances to and Mary, Virginia, found that herons urbanization and habitat alteration by human activity. So, occasionally, small tend to establish colonies away from humans (see Ardeid 2001), suggests areas near human activity can provide human activity (Figure 3). Their results regional increases in opportunistic nest suitable nesting habitat if barriers to dis- seem convincing: potential disturbance predation. If so, adverse effects may by humans influences the become more likely after otherwise minor distribution of heronries. disturbances that temporarily flush adult Will the undeveloped por- herons and egrets from their nests. tions of our landscape Even the most careful management of remain adequate for the heronries cannot guarantee their local long-term needs of these stability. A fundamental characteristic of birds? heron and egret colonies is that they shift locations over time, often in response to Living with herons local disturbances. These shifts often ome heronries exhibit result in the formation of “satellite” considerable tolerance colonies close to or within a few kilome- Sto humans. Black- ters of disturbed sites. Therefore, a region- crowned Night-Herons, Little al management perspective may be cru- Egrets, and Chinese Pond cial not only in protecting wetland feed- Herons have nested undis- ing areas, but also in providing suitable turbed near a village near alternative habitat for nesting near exist- Hong Kong for over a century ing colony sites. Because these beautiful apparently because they species depend widely on the landscapes Figure 3. Average density of buildings at varying distances from were thought to bring good in which we live, they remind us that con- 53 Great Blue Heron colony sites and at 58 random locations in luck (Hafner 2000). In some servation is a central challenge in manag- the lower Chesapeake Bay, Virginia; adapted from Watts and Brad- parts of the world, herons ing our world. ■ shaw (1994, Colonial Waterbirds 17:184-186). Areas are defined have been respected and by non-overlapping concentric radii. *Asterisk indicates significant protected since ancient difference between colony and random locations (P < 0.01). page 4 the ARDEID 2002

The challenging changeling Invasive Spartina by Katie Etienne

he invasion of non-native Spartina species (cord Tgrass) represents one of the most alarming threats to coastal marsh systems. Although Spartina is a relatively new chal- lenge for northern (Figure 1), we must respond promptly to avoid the serious biological consequences that have accompanied Spartina inva- sions along the coasts of Britain, France, Washington State, and the San Francisco Estuary. The introduction of non- native Spartina alterniflora to Washington State began in the Spartina alterniflora. 1870’s, when trains and cargo ships used non-native cord grass for packing material. By the 1950’s, S. these plants were alterniflora dominated 400 acres of tidal actually hybrids of mudflats in Willapa Bay, according to S. alterniflora and local oyster growers. In 1995, following S. foliosa (Daehler Figure 1. Distribution of introduced Spartina species 2000-2001. Data courtesy the discovery of two more non-native and Strong (1997, of Invasive Spartina Project httt://www.spartina.org. species (S. anglica and S. patens) and American Journal mounting evidence of biological and eco- of Botany 84: 607-611). higher levels of hydrogen sulfide, and a nomic impacts, the Washington legisla- Ongoing research by Debra Ayres and wider range of salinity (Howes et al. 1986, ture declared the Spartina invasion an colleagues at the UC Davis Bodega Marine J. Ecology 74:881-98). “environmental emergency.” Private and Lab have refined a variety of molecular Hybridization between native and public landholders reduced the size of tools for identifying genetic differences non-native species can represent one of some infestations by prioritizing projects between native and non-native Spartina the greatest threats to ecosystems. For and coordinating efforts. However, in and their hybrids. Their research shows example, S. alterniflora and hybrids can 2001, over 5,000 acres of Washington tide- that several generations of crossing have modify the physical characteristics of lands had been invaded by non-native occurred and that some hybrids have vari- marsh habitat and reduce tidal circula- Spartina species, and recently S. densiflo- able morphology and greater reproductive tion. Spartina clones have a dense net- ra was discovered in Gray’s Harbor—the vigor than either parent. The proliferation work of rhizomes that stabilize the soil. only coastal deep water port in of these hybrids also reduces the probabil- Above ground, the large stems grow close- Washington State. ity that pollen from native plants will fer- ly together, which reduces the velocity of This pattern of slow growth followed tilize remnant populations of native S. tide water and facilitates the deposition of by rapid spread has also been observed foliosa (Ayres and Strong 2002, Aquatic sediment. In Willapa Bay, the diameter of in San Francisco Estuary. In 1975, S. Nuisance Digest 4: 37-39). the average clone increases by approxi- alterniflora was imported from the Being able to distinguish hybrids from mately 75 cm per year. If left uncon- Atlantic seaboard to prevent erosion and parent species has also revealed that trolled, Spartina invasion has the poten- reclaim a marsh near Fremont, and hybrids tolerate a wider range of condi- tial to convert the salt marshes and open transplants were used to restore the San tions than either parent. The invasion of mud in San Francisco Bay into vast stands Bruno Slough. By 1990, at least 650 circu- subtidal habitat by S. alterniflora and its of hybrid and invader cord grass (Ayres lar patches of S. alterniflora were com- hybrids is facilitated by the presence of and Strong 2002). Based on the mean peting with the California native S. foliosa aerenchyma tissue. The aerenchyma form tidal range of S. alterniflora, it is predicted and other native plants (Callaway and tubes in the stem that increase oxygen that 65% of the mudflat in Bodega Bay Josselyn 1992, Estuaries 15: 218-26). A transport to below-ground tissue which could be covered if this vulnerable site subsequent study revealed that most of helps plants survive anoxic sediments, were invaded by S. alterniflora or its 2002 the ARDEID page 5

hybrids (Daehler and Strong 1996, Biological Conservation 78:51-58). An ounce of prevention is worth… There are many serious consequences much more than an extended period of Integrated Weed of transforming exposed mudflats into Management (IWM). emergent marsh vegetation. Invasion by Posted on the Invasive Spartina Project (ISP) web site is an excellent summary non-native Spartina species can adversely of IWM methods, which demonstrate that the selection of appropriate methods impact native plants and animals. depends on the scale of the problem. For example: Research in English marshes showed that ❚ the spread of S. anglica resulted in Digging can be 100% effective for removing small, isolated clones, but every seed and portion of the rhizome must be removed. Because root material decreased shorebird abundance (Goss- may extend 1.2 meters below the surface of the mudflat, this approach is not Custard et al. 1995, J. Applied Ecology 32: feasible when clones aggregate to form large patches or meadows. 337-51), and competition with eel grass ❚ can reduce food for certain herbivorous To contain a population or prevent hybridization, it is possible to control seed waterbirds (Way 1991, Washington Sea production by clipping seed heads to prevent pollination and seed dispersal. Grant). In Willapa Bay, vast swards of S. However, this requires constant attention because species flower over an extended period of time from April-December. alterniflora can restrict fish to narrow channels and limit access to open water ❚ To remove small-to-medium patches (up to 36 feet in diameter), all plant except during periods when tide water material must be completely covered with geo-textile fabric or plastic for two rises above the non-native Spartina— growing seasons. which tends to grow taller than the native ❚ Mowing can control infestations of any size, except small seedlings. Howev- S. foliosa (Cordell et al. 1998, Proc. 8th er, because mowing can initially invigorate the plant and promote root devel- International Zebra Mussel and Other opment, it must be repeated at least 4-6 times per year for a minimum of Nuisance Species Conf., Sacramento). two growing seasons. The invasion of channel margins by ❚ Amphibious equipment has been developed for mechanical smothering and non-native S. densiflora and its hybrids ripping of large infestations. This aggressive approach should be conducted can restrict flow, cause widening of the during the fall or winter and is likely to degrade surrounding habitat or water flood plain, and reduce or eliminate forag- quality. ing and nesting habitat for the federally ❚ Herbicides are 0 to 100% effective, depending upon the conditions, timing and state endangered California Clapper and method of application and may have deleterious effects on non-target Rail (Rallus longirostris obsoletus). At high- organisms. The use of aquatic pesticides currently requires a revocable-per- er marsh zones, non-native S. patens grows mit from the State Water Resources Control Board with extensive monitoring in dense “cowlicks” and competes with and reporting responsibilities. native pickleweed (Salicornia virginica), which provides critical habitat for the salt Therefore, if you see a suspicious Tomales Bay shoreline. At the time, there marsh harvest mouse (Reithrodontomys plant, please flag it and mark the location were no genetic tests available to confirm raviventris). Other plants that do not com- on a map. Report this information to Katy the identification of 64 S. densiflora plants pete well with S. patens are salt grass Zaremba, Field Biologist for the Invasive growing along the east shore of Tomales (Distichilis spicata) and the federally listed Spartina Project (ISP), at (510) 286-4091. If Bay, so we decided to carefully remove soft bird’s beak (Cordylanthus mollis). it is not possible to identify the plant in these plants based on their physical char- What can we do to protect the field, it will be genetically tested. The acteristics. S. densiflora tends to grow in coastal marshes from invasion? exact location of all invasive Spartina dense tufts or mounds in the mid-to-high t is crucial to be on the look-out for plants will be recorded with a GPS unit so marsh zone, but we found these plants at invasive Spartina. However there are the site can be monitored for resprouts. the interface between open mud and important reasons why we shouldn’t ISP was started two years ago by Debra marsh vegetation. S. densiflora usually I Smith and others with support from the flowers between April and July—earlier rush out and pull up suspicious plants. First, it can be difficult to identify both California Coastal Conservancy. The than S. foliosa which flowers between native and non-native plants, particularly Project has an ambitious mission to study June and September. early in the season, because there is con- Spartina distributions, evaluate treatment In November 2001, a single clone of S. siderable overlap in size and other field methods, and develop management alterniflora was discovered at the north characteristics that can vary with local strategies to eliminate non-native end of Bolinas Lagoon. Since then, five conditions. Please visit the Invasive Spartina in the San Francisco Estuary. The populations of S. alterniflora or hybrids Spartina Project web site (http:// ISP team has also helped to monitor and have been discovered around Drakes www.spartina.org) to download their control the first non-native Spartina dis- Estero. Although many of these plants excellent, full-color field identification covered in coastal estuaries along the grew in round clones, some sparse guides. Second, it is extremely important Marin County coastline (Figure 1). seedlings were found growing in narrow that all plant material is removed, partic- The first S. densiflora plants in Tomales bands that are characteristic of S. foliosa ularly seeds and root fragments. Finally, it Bay were detected by Doug Spicher in and did not have all of the field character- is essential that we know where non- 1999. Local monitoring began in 2001, istics of the non-native. These variations native plants have been discovered so when the ISP team assisted local property in growth and morphology underscore these sites can be consistently monitored owners and biologists from Golden Gate the value of genetic identification and the in the future. National Recreation Area, importance of conducting thorough sur- National Seashore, and Audubon Canyon veys each year. Ranch in surveying almost all of the page 6 the ARDEID 2002

Fourteen years of oak woodland restoration at the Bouverie Preserve

Acorns and Ecosystems

by Rebecca Anderson-Jones

or decades, land managers with acorn production (blue oaks). have noted declines in the They inferred that blue oak regen- Fregeneration of blue oak eration should benefit when con- (Quercus douglasii) and valley nectivity between trees is oak (Q. lobata) throughout enhanced. This kind of connectivi- California. Some evidence sug- ty is one of the tangible benefits of gests that woodland fragmenta- ACR’s efforts to support habitat tion may impair acorn produc- protection beyond the borders of tion (see below). Woodland oaks our sanctuaries. also face crucial competition The oak planting project from introduced grasses. Oak and managers use a variety of woodland conservation can seem approaches to restore oak daunting! Success requires us to woodlands. In 1988, John support acorn production and L Petersen took the direct route, address recruitment challenges working with Bouverie Preserve while bolstering the survival of Fellow Bruce de Terra, dedicated mature trees.

COURTESY BOUVERIE PRESERVE DOCENTS BOUVERIE PRESERVE COURTESY volunteers, area children, and their Ecological challenges Docent Maxine Hall and students Amanda Hall, Emily Davis, and Nathan teachers to implement an oak Competition. Managing the land Todhunter plant an oak at Bouverie Preserve. planting project. Their goal was to to enhance oak survival and restore the lower field at Bouverie regeneration is complicated by to the kind of plant community interactions between native oaks and by thatch build-up (e.g. declines in light likely to have existed before it was cleared introduced annual grasses. During the and water penetration), but they can also for ranching more than a century earlier. growing season, annual grasses are fierce have detrimental impacts on oaks. Cattle This was a wonderful learning opportuni- competitors for light, water and nutrients consume seedlings and—except in very ty. It was also a hopeful beginning for near the surface, catching water before it small numbers—compact the soil. Alien what has become a long-term restoration percolates to deeper soil profiles where it weedy species and nitrogen enrichment project at the preserve. can be used by native perennial grasses are introduced to the system in their Initially, the project was a straightfor- and mature oaks. By intercepting water, wastes. Cattle browse oak branches heavi- ward attempt to mitigate for historic oak annual grasses compound the stresses ly and can break or crush saplings or top- clearing, partially motivated by concerns affecting these magnificent trees. ple exclosures intended to protect trees. about low levels of natural oak regenera- Ubiquitous grass competitors also create Cattle also need free access to water tion statewide. Since then, our restoration significant problems for oak seedling and throughout the grazing range to avoid goals have changed subtly with our sapling survival. overgrazing near existing water sources. understanding of how they are affected Regular monitoring and careful manage- by, and affect, the extensive vernal wet- Regeneration. Early in his tenure with ment are needed to prevent or correct for land system in the lowlands (to learn ACR, John Petersen, then Bouverie these impacts. While other grazing ani- more about the vernal wetlands at Preserve’s Biologist, observed low rates of mals or management techniques may be Bouverie Preserve, see the 2001 Ardeid). oak recruitment with concern, and con- more suitable for alien grass control, each Fourteen years after the oak planting pro- ducted an age-class census in the pre- brings challenges that must be managed ject’s inception, the restored oak wood- serve’s mature oak woodland. Compari- to limit adverse effects. land is showing early evidence of matura- son of seedling numbers in his study area tion that bodes well for the future. in spring and fall indicated a loss of near- Habitat fragmentation. Lack of connec- ly three-quarters over the four-month tivity among woodlands may also con- What the data tell us period corresponding with annual graz- tribute to the low regeneration of wood- he oak planting project includes a ing to manage grass biomass. This is a land oaks in California. Eric Knapp, Kevin two-part monitoring protocol. The familiar story in California’s oak wood- Rice and Michael Goedde, from UC Davis, Tfirst involves direct growth meas- lands, with many ramifications. studied the role of tree density in wind- urements. The second uses a breeding Grazing animals can be a tremendous mediated pollen transfer in blue oaks. bird census to track the maturation of the benefit in controlling the growth of annu- They found that pollen density correlated woodlands as avian habitat, relative to al grasses and reducing problems caused with distance from the source plant and mature oak woodland elsewhere on the 2002 the ARDEID page 7

for most surviving trees beyond 1997, many have continued to thrive. Twenty-one trees of four species have grown above the livestock browse line (approx. 1.5 m), attaining a total height of 2m or greater, with nine of these greater than or equal to 3m in stature. Of these, three valley oaks stand at approxi- mately 4m (Figure 2)! Interestingly, while all plant- ings were of blue, coast live, black or valley oak species, three Oregon oak trees (Quercus garryana) currently thrive in the study area. Given the ten- dency of species in the white Figure 1. Survival of original oak stock nine years after oak subgenus to hybridize and planting. Note similar rates of sapling survival and relatively the lack of acorns to distinguish strong acorn survival in blue and valley oaks. Mortality of Figure 2. Twenty-one oak trees of four species seedlings in all species and acorns in black oaks was 100%. Oregon from valley oaks in this (including trees showing evidence of hybridization with immature woodland, variability other species) show vigorous growth above the live- in leaf shape may have resulted stock browse line (1.5 m). Valley oaks are best repre- preserve. Despite problems with imple- in misidentifications. However, field sented among the taller trees. mentation of the monitoring program, we notes suggest a more compelling expla- can draw some useful conclusions from nation. In 1997, a blue oak tree was our data. noted to have significant branch die- are incidental visitors. Over time, we Due to the unfortunate loss of location back, showing only partial resprouting, expect increased use of the restored tags, 1997 was the last year growth data while a smaller, healthy Oregon oak woodland by oak woodland specialists could be tracked to each tree’s original seedling grew within the same exclosure. and a decline in use by grassland birds, source stock. Re-plantings were carefully At this time, the Oregon oak had reached but continued use by wetland associates, documented, allowing us to track trees 23 cm in height. By 2002, the Oregon oak such as the Red-winged Blackbirds. from the initial cohort through 1997. Oaks had obtained a height of 70 cm, and the What we learned and what we are known to have high mortality rates in blue oak had died. The blue oak’s death would do differently the first years after planting or transplant- may have resulted from an intolerance of roject results have helped us modify ing regardless of the type of propagules shade or other competition between the methods for use in future oak used. The five-year survivorship of oaks two trees, and its presence may have woodland restoration efforts. For grown from acorns in our project is 25% aided in the establishment of the more P example, future projects should track tree for blue oaks and 47% for valley oaks. shade-tolerant Oregon oak. The site locations with GPS and mark trees rather These rates are higher than first-year sur- preparation and maintenance at that than exclosures, prevent rodent damage vival rates found by other researchers for location almost certainly did. If Oregon with protective sleeves before trunks can valley oaks and only slightly lower than oaks have been recruited from seed trees be girdled, and include more rigorous published first-year survival rates for blue in nearby woodlands, as these observa- weeding during the growing season. oaks. When contrasted with published tions suggest, the restoration project may Nevertheless, we have evidence that the rates of survivorship after five years, our be supplying other habitat values that project has been successful. Although results are strong for both species. Trees support recruitment. For example, direct growth measurements are no initially planted as saplings showed high saplings in the restored woodland may longer the most useful indicators of suc- survivorship, with very similar data for provide roosts for birds, such as jays and cess, they did help us track establishment valley and blue oak. (Figure 1). However, woodpeckers, which are capable of dis- of young trees. We will continue to moni- no trees planted as seedlings survived, tributing seed. Unassisted recruitment of tor the maturation of the woodland as and all black oaks (Q. kelloggii), planted oaks into the woodland is a sign of suc- habitat for breeding birds and to manage only from acorns, died. cess, and an indicator that the system is the system to promote oak growth. The Despite lost tags, we were able to track maturing. monitoring protocol is being revised to a small number of trees from 1997 to Breeding bird territories in the mature improve its usefulness as part of a long- 2002, and many of these have shown upper oak woodland were delineated and term program for monitoring breeding strong growth in stature or canopy devel- tallied annually from observations of birds across the Bouverie Preserve. The opment during this interval. Three coast singing males. This census was intended results will benefit other researchers live oaks increased 59, 62, and 98 cm in to provide a reference for comparison interested in oak woodland management height, with trunk-to-dripline increases of with breeding bird use of the lower, and restoration. The challenges of restor- 24, 41, and 51 cm respectively, while one restored woodland. To date, our observa- ing whole ecosystems make this an excit- valley oak increased in height by 138 cm tions indicate that only Red-winged ing time to manage oak woodlands in with a trunk-to-dripline increase of 102 Blackbirds use the lower, restored wood- California. ■ cm. Although we could not track growth land for breeding, although other species page 8 the ARDEID 2002

Raven predation in heronries Resistible Forces? by John P. Kelly

anaging wildlife is extremely dif- are scavengers of failed nests and do ficult because natural processes not threaten the persistence of Mthat determine animal abun- heronries. At most heronries, at least dance and behavior are often complex for now, managing the predatory activities of ravens is unnecessary. and mysterious. Among the most myste- 2001), questions have emerged about KELLY JOHN P. However, with raven populations growing rious are the influences of Common potential increases in raven predation. rapidly in apparent response to the Ravens on nesting herons and egrets. Resident ravens have become special- expansion of agriculture, roads, garbage Ravens occasionally prey on active heron ized nest predators at some regionally dumps, and urbanization (see Ardeid and egret nests, but more typically they important heronries, such as Marin Islands National Wildlife Refuge near San Do ravens threaten heron and egret colonies? Rafael and ACR’s Picher Canyon heronry at Bolinas Lagoon Preserve (see box at For observers of heron and egret colonies, nest predation by Common Ravens can be left). If ensuring the persistence of partic- a dramatic example of nature “red in tooth and claw.” Ravens are expert egg preda- ular nesting colonies depends on reduc- tors, but their greatest threat to heronries seems to be in the predation of 3.5- to 5- week-old nestlings. This is the time in the nesting cycle when adult egrets begin to ing raven predation, how might such a leave their nests unattended, as both parents forage for food needed by their devel- reduction be achieved? In a recent report oping young. At this point, nestlings may be grabbed and torn apart, then eaten or to the California Department of Fish and cached for later. Game (2001), Joseph Liebezeit and Luke Contrary to popular impressions of serious threats from raven predation, preliminary George of Humboldt State University results from ACR research indicate that nest predation by ravens is unlikely at most thoroughly reviewed methods for manag- colony sites in the San Francisco Bay area—even though ravens may be near. How- ing predation by ravens, but suitable ever, resident ravens at some sites may specialize on egret eggs and nestlings for strategies for controlling their effects on food during much or all of the nesting season. Good examples of such specialization heronries remain unclear. occur in ravens nesting near the Picher Canyon heronry at ACR’s Bolinas Lagoon Pre- serve and at Marin Islands National Wildlife Refuge near San Rafael. Conditioned Taste Aversion. In recent In 1998, Great Egrets at Picher Canyon suffered severe nest predation by Common nesting seasons, we have attempted to Ravens. Most nests were lost or abandoned, and only 26 young were successfully establish “conditioned taste aversion” fledged, compared with expected production of 100-150 young. In subsequent years, (CTA) in ravens at ACR’s Picher Canyon. ravens destroyed fewer nests, apparently because of reduced food demand associat- This involves providing chemically treat- ed with their own nesting failures. The ravens failed in 1999 and destroyed only as ed prey that can produce severe illness in many as 9 (16%) of 58 Great Egret nests (we found direct evidence of raven predation ravens and, consequently, alter their at 5 nests). They failed again in 2000 and destroyed a maximum of 12 (21%) of 75 predatory behavior (Nicolaus and Lee nests (direct evidence at 6 nests). In 2001, the ravens nested late but fledged 3 young 1999, Ecological Applications 9(3): 1039- and may have destroyed as many as 33 of 85 (39%) Great Egret nests (direct evi- 1049). By associating the taste of normal dence at 8 nests). Egg predation at Picher Canyon has been relatively rare. prey with acute illness, predators develop At Island, ravens have fledged 4–5 young each year since 1999. Estimates a reflex aversion for that taste. The aver- of Great Egret nest mortality, based on samples of individually monitored nests, were sion is then stimulated during subse- higher in 2001 (31%, n = 54 focal nests) than in 2000 (19%, n = 59) or 1999 (20%, n = quent predation attempts and ultimately 45). Great Egret nest mortality seemed to be lower in 2002 (12%, n = 68), although predation rates on Snowy Egret and Black-crowned Night-Heron nests were not alters their predatory behavior. If CTA measured. We quantified egg predation at Marin Islands based on the number of could be established in territorial ravens depredated heron and egret eggs found on East Marin Island where ravens nested. at heronries, they might avoid egret nests The results suggest a dramatic increase in egg predation in 2001 and a possible while continuing to exclude other ravens decline in 2002 (Table 1). The overall frequency of raven behaviors associated with from their territory—effectively “baby-sit- the predation of heron or egret nests increased at Marin Islands and Picher Canyon ting” the colony. through 2001 (Table 2) but was not measured in 2002. In 2001, we captured both of the resi- Perhaps most alarming has been predation of adult Snowy Egrets indicated by car- dent ravens at Picher Canyon and provid- casses found near raven roosts on East Marin Island: at least 4 adult Snowies were ed each bird with a piece of meat from taken in 2000 (none found in prior years), at least 7 in 2001, and at least 15 in 2002! fallen Great Egret nestlings found dead We are currently conducting a more rigorous analysis of trends in raven predation under the heronry. The food was treated and survivorship of heron and egret nests at Marin Islands and Picher Canyon heron- with enough fenthion to cause severe ill- ries. Whether these heronries can tolerate further increases in nest predation by ness. Captive feeding helped ensure that ravens remains unknown. 2002 the ARDEID page 9

one or more renesting may not be possible. Shotguns require a Table 1. Number of depredated eggs found on East Marin Island, 1999-2002. attempts each season. If close range, which is rarely available with ravens abandon an area after ravens. Rifles allow a greater range but nest disturbance, a new nest- require greater accuracy and carry addi- Year 1999 2000 20012002 ing pair might establish a ter- tional restrictions for safe use. Human Search-days 8 3 6 4 ritory, or, in the absence of a activity in the vicinity of the ACR heronry Heron and new pair, vagrant groups of would substantially limit safe firing angles egret eggs 45 16 140 79b ravens might occupy the and positions. Other eggsa 10 16 31 21b undefended heronry. Poisoning. Removal by poisoning would Total eggs 55 32 171 100b Repellents. Other options for require a major effort to prevent herons,

a Other eggs include Western Gull and Mallard eggs. controlling nest predation by Turkey Vultures, Scrub Jays, owls, and b Preliminary results as of 6 June. ravens suggest little promise. other native species from taking poisoned There is no consistent empiri- bait. Because ravens sample new food the treated food was consumed com- cal support for the use of raven carcasses very cautiously, they might not be easily pletely by each individual, not shared or models as effigies to deter raven activi- poisoned. The overriding issue, however, with its mate or offspring, cached for ty. The use of visual, auditory, or chemical is that removal by any method would pro- later, or taken by other species. Captivity repellants to discourage raven predation is vide only temporary control as other should not adversely affect the CTA unlikely to succeed beyond an initial ravens move in to fill vacancies. process because the primary stimulus response period, and could disturb nest- (taste) is transmitted directly through a ing herons and egrets. t ACR, we have learned to appreci- medullar pathway, bypassing cognitive ate the challenge of managing Removal. Any removal of ravens would processes that later associate the ravens in heronries. It is important probably require an ongoing control pro- A response with other stimuli such as visual to remember that ravens are native birds, gram as other ravens move in to fill or location cues. When left alone (birds protected by federal law, and have valu- vacancies. If a new pair of resident ravens were observed through a remote video able ecological roles. Most options for did not immediately fill the vacancy, monitor), both ravens readily consumed managing ravens require special permits. the food ad libitum within 15-30 minutes. vagrant non-territorial ravens Before releasing the ravens, we mounted might prey on heron and egret Table 2. Frequency of raven behaviors associated with a radio transmitter to each of them so we nests at equal or greater rates. mortality of heron or egret nests or nestlings at West could monitor their movements and Removing ravens from a colony Marin Island and Picher Canyon, 1999-2001. Behaviors include flying from the colony with eggs, young, or uniden- behaviors. However, no subsequent site by any means presents con- tified food, and perching in or adjacent to a failed nest. behavioral changes were detected. siderable difficulties. Because we were not able to develop a Translocation. Ravens cannot be Number of nest predatory behaviors / 100 hrs fully controlled experiment, we could not translocated because of potential rule out the possibility of an error in CTA problems with disease transmis- Colony site 1999 2000 2001 application or treatment dosage. We sion and exporting pest species West Marin Island 6.5 13.5 17.8 remain optimistic about the potential use to other areas. of CTA as a management tool. The main Picher Canyon 2.0 4.6 7.8 problem is that ravens are very difficult to Donation to education pro- Both colony sites 3.1 6.9 10.8 capture, often requiring two or more grams. Trapped individuals could be given to wildlife educa- weeks of daily baiting and several addi- Our best hope is that important heron- tion programs, but such programs are not tional days of trapping for each bird. Cap- ries and, ultimately, heron and egret pop- generally interested in such a commit- turing individuals a second time at Picher ulations can tolerate increases in nest pre- ment. Trapping may be very difficult (see Canyon could be even more difficult. dation by ravens. This might be possible at below). Disruption of nesting behavior. Nest pre- Marin Islands, where several hundred dation at ACR’s heronry on Bolinas Trapping and euthanasia. Trapping fol- heron and egret nests are susceptible to Lagoon has been reduced in years when lowed by euthanasia may be the most fea- predation by a single pair of ravens. At raven nesting attempts have failed. sible method of removal, but some ravens Picher Canyon, nesting egrets have weath- Therefore, preventing or delaying suc- may be difficult to trap because of their ered moderate predation pressure during cessful nesting by ravens might help in cautious behavior or use of particular the last few years without serious conse- controlling nest predation. Potential habitats. We found net launchers to be quences. However, the last time ravens at methods of disrupting raven nesting the most successful tool for trapping Picher Canyon raised a brood of at least behavior include the removal of nests or ravens, but as noted above, successful four young (1998), their food demand eggs, or treatment of eggs so they will not trapping may involve weeks of effort. It peaked just as egret chicks became avail- hatch. Addling (by shaking), oiling, or may be extremely difficult to trap both able. The devastating result (see box, page puncturing eggs to prevent hatching has members of a nesting pair. Trapping both 8) suggests that a strategy for controlling the potential advantage of delaying the ravens at Picher Canyon may be even raven predation may be necessary to ravens’ detection of nest failure, and thus harder because they have already been avoid an annual game of “raven roulette” delaying or preventing subsequent re- captured once. that risks future abandonment of the colony site. For now, however, the likeli- nesting. Disruption of successful nesting Shooting. Because ravens are extremely would require annual searches for nest hood of successfully controlling raven pre- wary, they are very difficult to shoot. dation remains unknown. ■ locations and follow-up searches to locate Shooting both members of a nesting pair page 10 the ARDEID 2002

Understanding the impact of a new epidemic on our forests and woodlands Sudden Oak Death

by Daniel Gluesenkamp

Left: Coast live oak tree in a fog-shrouded forest.

bark beetles, and “golf- ball” (Hypoxylon sp.) fun- gus on the trunk. Insects and fungi do not appear to cause SOD, but rather they opportunistically exploit dying trees and are a sign that the tree is already near death. Trees are most likely killed by a disease that creates large cankers in living trunk tissue, destroying con- ductive tissues and girdling the tree. The disease agent is an oomycete (not a fun- gus) named Phytoph- thora ramorum; other Phytopthora species are responsible for Irish RICHARD BLAIR potato blight, avocado root-rot, and the near- alifornia was not always as it is ties (including Dodecatheon, Delphinium, extinction of dozens of plant taxa in today. Ten million years ago, warm and Stipa) make up about 49% of the Australia. As is the case with other major Cwinters and wet summers support- species in our state. Most of these are American tree epidemics (including ed forests similar to those in eastern found nowhere else in the world. chestnut blight, Dutch elm disease, and north America and China: maple, beech, Today we see another transformation Monterey pine pitch canker), P.ramorum ginkgos, rhodendron and sequoias. As the involving oaks, and this time the oak trees is most likely an accidental introduction Sierra Nevada transformed from hills to are in need of protection. Ancient oak from Europe, with no evolutionary history mountains the climate changed dramati- trees are replaced by developments, in North America. cally. Rainfall decreased, summers acorns are failing to produce adult trees, P.ramorum produces spores in the wet became drier, and moisture-loving plants and now a pathogen is sweeping through season that are dispersed in water, rain were pushed north into Oregon and forests and woodlands like wildfire. This splash, and mud. P.ramorum is known to beyond. Or they hid beneath oak trees. article summarizes current knowledge kill four oak species: coast live oak Every naturalist knows that oak trees regarding Sudden Oak Death (SOD), dis- (Quercus agrifolia), black oak (Q. kellog- are important refugia, with deep taproots cusses Audubon Canyon Ranch’s response gii), Shreve oak (Q. parvula var. shrevei), that draw water and nutrients 60 feet to to the threat, and offers some suggestions and tanoak (Lithocarpus densiflorus). the surface and a canopy that cools the for what we all can do to save the trees. Scrub oak, blue oak (Q. douglassii), valley weary botanist. Thousands of plant and What is known about Sudden oak (Q. lobata), and Oregon oak (Q. gar- animal species rely on oaks for habitat Oak Death? rayana) do not appear to contract the dis- and food, and ten million years ago the ease. Of the susceptible species, tanoak is udden oak death was first observed oak trees that spread across the drying most susceptible, and a high proportion in 1995 by a Marin County extension landscape provided shelter for Arcto- of exposed trees develop the disease and agent who noted that oak trees were tertiary flora retreating northward. These S die. Coast live oak is highly susceptible to dying so rapidly they were dead before refugees speciated into new forms and infection, but inoculation experiments they could drop their leaves. Other com- today are an important component of the show that about 30% of infected trees mon symptoms of SOD include dark sap flora we love; plants with northern affini- may have some resistance to the disease bleeding from the trunk, the presence of 2002 the ARDEID page 11

without bay neighbors Sudden Oak Death and Table 1: Sudden Oak Death host list as of April 2002 (from the California Oak Mortality Task Force). (Swiecke and Berndhart, Audubon Canyon Ranch 2001, Fifth Symposium on udden Oak Death has not yet been Common name Scientific name Oak Woodlands). Since detected at ACR’s coastal Marin pre- these two species may serves, and we continue to monitor coast live oak Quercus agrifolia S compete for light and these sites. The disease is present at the California black oak Quercus kelloggii other resources, produc- Bouverie Preserve and has been cultured Shreve oak Quercus parvula var. shrevei tion of oak-killing spores by J. Davidson (U.C. Davis) and S. Swain tanoak Lithocarpus densiflorus by bay laurel can be con- (Sonoma County) from coast live oak and rhododendron Rhododendron spp. sidered a form of biologi- bay laurels at several locations. Symptoms huckleberry Vaccinium ovatum cal warfare between bays of Phytophthora infection appear on bay and oaks! This curious California buckeye Aesculus californica leaves in many parts of the preserve, and relationship has impor- Pacific madrone Arbutus menziesii the pathogen is likely present in most of tant ramifications: since BP’s coast live oak stands. Dead and dying manzanita Arctostaphylos manzanita spores are coming from live oaks with symptoms of SOD are par- bay laurel Umbellularia californica resistant host species, it is ticularly abundant along the length of the California coffeeberry Rhamnus californica unlikely that coevolution Loop Trail leading up to the bark house. toyon Heteromeles arbutifolia of the oak-pathogen inter- California honeysuckle Lonicera hispidula action will reduce viru- bigleaf maple Acer macrophyllum lence of the pathogen. Plans for combating the viburnum Viburnum bodnantense disease must include these alternate hosts. It is unlikely that (Rizzo et. al., 2001, Fifth Symposium on we will find a cure for Sudden Oak Woodlands). Oak Death. Some chemical Improved detection techniques have treatments may reduce revealed P.ramorum infection in several rates of infection and additional host plant species (Table1), extend the life of and the list continues to grow. While infected individuals most of these alternate hosts are not (Garbeletto et. al., severely affected (huckleberry, madrone, 2001, Fifth and rhododendron may be killed), they Symposium on are important to the spread and persist- Oak Woodlands), ence of the disease. For example, bay lau- but the expense rel trees probably act as “Typhoid Mary” and potential for Among the vulnerable species: coast live oak (left) in our hardwood forests. Bay trees are negative side and tanoak (above). almost unaffected by the disease but pro- effects preclude duce copious quantities of P.ramorum application to wild- spores. In contrast, oak species that are lands. Since there are Sudden Oak Death is expected to sig- killed by the pathogen often fail to pro- no means for treating nificantly affect Bouverie’s coast live oaks, duce spores even in laboratory cultures. infected forests, its important that we and black oaks and madrone may also be Thus, bay trees that are unaffected by the map the disease and reduce it’s rate of killed. Assessing the scale of the problem disease produce spores that infect and spread (see below). In particular, is complicated by the notable tree diver- kill oaks; this is shown by recent experiments with seedlings sug- sity of the Bouverie Preserve, with nine research showing that oaks gest that northern red oak (Q. oak species. It is important to collect data growing near bay trees rubra) and pin oak (Q. that will reveal patterns obscured by the are more likely to be palustris) may be visual complexity of our diverse hard- infected than are highly sensitive to wood forests. Current and planned work oaks growing the pathogen (D. at the Bouverie Preserve focuses on map- Rizzo, 2001, Western ping the distribution of diseased trees International Forest and censusing forests to quantify species Disease Work composition, age structure, and health of Conference), and dis- trees. In addition to addressing SOD con- persal of P.ramorum to cerns, this work will provide valuable midwestern and eastern baseline data to inform future manage- oak forests could have ment of our forests. terrible consequences. We are also cooperating with other researchers and agencies to understand the distribution and ecology of this new disease. In the last two years, we have Species not known to be affected by Sudden Oak Death include blue oak (left) and valley oak (right). Continued on page 12 ANE CARLA ROVETTA page 12 the ARDEID 2002

What you can do to fight Sudden Oak Death 1) Do not transport infected plant material State regulations currently restrict movement of known hosts or soil from or within infected areas. Oregon, Canada, and South Korea have established quarantines on host plant material from California, and the U.S. Department of Agriculture has imposed an interim quarantine while regulations are finalized. These rules reflect common sense: material that may be infected (including soil and branches, leaves, or wood, from any of the known host species) should not be moved. Dead or dying trees should be left in place unless they present a safety hazard and, if trimmed, material should remain on site. Bay laurel wreaths and oak, bay, and madrone firewood should not be moved into uninfected areas. 2) Lovers of the natural world must use protection. Spores are easily transported by humans, and so Sudden Oak Death is most prevalent at sites with high public visitation. After leaving an area known to be infected, wash mud from shoes, bicycle tires, horses, and vehicles; this simple action also slows the spread of other problematic non-native species. Muddy cars should be run through a carwash on the drive home. Tree work or vegetation management should be done in the dry summer months, when abundance of Phytophthora spores is lowest, and tools should be sanitized after each tree. Potential vectors can be sanitized using Lysol or a solution of 1 part bleach to 9 parts water. 3) Keep your oak trees healthy If you have oaks growing around your home or business, you can protect them from disease. When marveling at a giant oak canopy, remember that most of the tree is below ground. These large root systems are very sensitive. Do not pave, disturb, or compact within the drip line (outline of branches); if possible, pamper the root zone with 4-6 inches of mulch – but NOT bay laurel mulch! California oaks evolved with dry summer conditions and cannot tolerate wet roots year round. Do not water oak trees during the summer, and avoid planting lush landscaping adjacent to mature oaks. Potential Phytophthora hosts (Table 1) should not be planted near oaks. 4) Support increased protection for oaks in California. The greatest current threat to California oaks is clearing of oaks for urban development and agriculture. Groups such as the California Oak Foundation are working to protect oaks statewide, and many local groups contribute to local oak protection ordinances. This work has added significance in light of the rapidly spreading epidemic: the oak that you save today may be great-grandfather to the Phytophthora-resistant forests of tomorrow. ANE CARLA ROVETTA 5) Stay informed Our understanding of this epidemic is improving rapidly. Sudden Oak Death was first observed in 1995, the causal agent was identified as a species of Phytophthora in 2000, and the species was isolated and named in 2001. The list of P. ramorum hosts has tripled in the last year, and the first scientific paper on P. ramorum was published in 2002. While the popular media have done a fantastic job of covering this subject, recent coverage has favored the sensational over the factual. Fortunately, current state of the science can be found on several websites maintained by researchers and managers: The California Oak Mortality Task Force: http://suddenoakdeath.org The UC Marin County Cooperative Extension: http://cemarin.ucdavis.edu/index2.html The California Oak Foundation: http://www.californiaoaks.org

seen field visits and tissue collection by can be applied seasonally, since risk from forests requires an equilibrium between researchers from UC Davis and by the P.ramorum spores is greatest during the disease and host. This could occur Sonoma County Sudden Oak Death wet season. We are re-evaluating methods through selection for resistant trees, Coordinator. In addition to assisting used in oak woodland restoration proj- selection for less virulent pathogen geno- regional studies of this disease, these vis- ects; one approach might include green- types, or regulation of the pathogen by its have helped ACR staff to identify and house inoculation of seedlings with P. other components of the community. By assess SOD at the BP. We look forward to ramorum to screen out susceptible geno- leaving sick trees in place we allow par- continued participation by cooperating types. This would reduce transplant mor- tially resistant trees the opportunity to investigators. tality in the field, hasten development of resprout, allow the less virulent P. r a m o - ACR staff are working to develop pro- SOD-resistant forests, and provide valu- rum strains to persist, and provide habitat tocols to slow the spread of P.ramorum able data regarding the frequency of dis- for birds, fungi, insects, and oomycota among sites and within preserves. ease resistance in our oaks. that may contribute to a new stable equi- Options include establishing hygiene pro- Finally, ACR is contributing to the bat- librium. In a world where more oaks are tocols in rainy season, including steriliz- tle against Sudden Oak Death by main- killed by development than by disease, ing boots before entering and leaving the taining sanctuaries where biological preservation of natural oak forests and preserve, and even closing specific trails processes can occur uninterrupted. Since woodlands is an important mission. ■ that are probable sources of infective it is unlikely that we will find a cure for spores. Fortunately, hygiene measures Sudden Oak Death, persistence of oak 2002 the ARDEID page 13

In progress: Tomales Bay waterbird article by Rebecca Anderson- Christopher DiVittorio (UC survey ◗ Since 1989-90, Jones, page 6). Berkeley), Dispersal and teams of 12-15 observers have project updates Cape ivy control ◗ Work disturbance colonization in a conducted winter waterbird conducted by Len Blumin has California coastal grassland. censuses from survey boats North Bay counties heron proven that manual removal of Peggy Fong (UCLA), Algal on Tomales Bay. The results and egret project ◗ Annual nonnative cape ivy can indicators of nutrient enrichment provide information on habitat monitoring of reproductive successfully restore riparian in estuaries. values and conservation needs activities at all known heron vegetation in ACR’s Volunteer Brenda Grewell (UC Davis), and egret colonies in five of 51 species, totaling up to Canyon. Continued vigilance in Species diversity, rare plant northern Bay Area counties 25,000 birds. Baseline status weeded areas has been persistence, and parasitism in began in 1990. The data are and conservation concerns for important, to combat resprouts mid-Pacific Coast salt marshes: used to examine regional waterbirds have been evalu- of black nightshade, vinca, and functional significance of patterns of reproductive ated and published (Kelly and Japanese hedge parsely. interplant parasitism. performance, disturbance, Tappen 1998, Western Birds). Eucalyptus removal at Emily Heaton (UC Berkeley), habitat use, seasonal timing Predation by ravens in Bouverie and Bolinas Bird communities in north coast and spatial relationships among heron and egret colonies Lagoon preserves ◗ oak-vineyard landscapes. heronries. ◗ We are observing nesting Eucalyptus from Pike County ravens in Marin County and Jodi Hilty (UC Berkeley), Picher Canyon heron and Gulch at Bolinas Lagoon measuring their predatory Carnivore use of riparian egret project ◗ The fates of Preserve, and along the behaviors at heron and egret corridors in vineyards. all nesting attempts at ACR’s Highway 12 border of Bouverie nesting colonies, with an Picher Canyon heronry are Preserve are being cut and Martha Hoopes and Cheryl emphasis on heronries at monitored and reproductive removed with incremental Briggs (UC Berkeley), Effects ACR’s Picher Canyon and success is analyzed annually. annual efforts. Stumps and of dispersal on insect Marin Islands National Wildlife Field procedures are based on resprouts will be treated by population dynamics and Refuge. Radio telemetry and methods developed by Helen methods developed in an parasitoid diversity in galls of behavioral studies focus on Pratt who initiated the project associated investigation by Rhopalomyia californica on evaluating home range varia- in 1967 and published several Dan Gluesenkamp. Baccharis pilularis. papers on heron and egret tion, behaviors at heronries, W. Joe Jones (UC Santa Cruz), Eucalyptus resprout nesting biology. and diurnal movement Population structure of the patterns. A road survey control ◗ An experiment is ◗ California roach. Livermore Marsh As conducted throughout the San being conducted to determine ACR’s Livermore Marsh trans- Francisco Bay area revealed the optimal method for Gretchen LeBuhn (CSU San forms from a freshwater concentrations of ravens in controlling Eucalyptus Francisco), The effect of system into a tidal salt marsh, some urban/suburban areas resprouts. Dan is testing the landscape changes on native we are studying the relation- and along the outer coast. relative effectiveness of bee fauna and pollination of ship between increasing tidal cutting, use of the herbicide native plants in Napa and Plant species inventory ◗ prism and marsh channel Rodeo (glyphosate), and Sonoma counties. Resident biologists maintain topography. The results are grinding stumps, to perma- Jacqueline Levy (CSU San inventories of plant species being compared with data nently kill cut Eucalyptus trees Francisco), known to occur at Bouverie Impact of butterfly from mature reference in the lower field at Bouverie and Bolinas Lagoon preserves. gardens on pipevine marshes, and will contribute to Preserve. swallowtail populations. future restoration designs. The Grant Fletcher has established results will also contribute to a database of shoreline plant Wendy Losee (Sonoma studies of changing bird use species on Tomales Bay. Visiting investigators Ecology Center), Thermal and vegetation in the marsh. monitoring, Sonoma Creek Annual Cordylanthus Elizabeth Brusati (UC Davis), watershed assessment. Newt population study ◗ survey ◗ This project Consequences of species Steven Morgan, Susan Annual newt surveys have continues earlier field investi- invasion under global climate Anderson, and others (UC been conducted along the gations on habitat and spatial change. Davis Bodega Marine Lab, Stuart Creek trail at Bouverie relationships among patches of Yvonne Chan and Peter Arcese UC Santa Barbara), Ecological Preserve since 1987. The Point Reyes bird’s beak, (University of Wisconsin), indicators in west coast results track annual and Cordylanthus maritimus palustris, Subspecific differentiation and in Tomales Bay marshes (Kelly estuaries. intraseasonal abundance, and genetic population structure of and Fletcher 1994, Madrono Lorraine Parsons (Point Reyes size/age and spatial distribu- Song Sparrows in the San 41: 316-327). The goal is to National Seashore), tions along the creek. Francisco Bay area. Long-term further address questions ◗ water quality monitoring, Shorebirds Since 1989, we about long-term stability and Jeff Corbin and Carla Walker Creek and Giacomini have conducted annual bay- biogeographic relationships D’Antonio (UC Berkeley), Wetland. wide shorebird censuses on Effects of invasive species on among discrete patches on Jennifer Shulzitski (USGS Tomales Bay. The data are nitrogen retention in coastal Tomales Bay. Golden Gate Field Station), used to investigate winter prairie. population patterns of shore- Oak restoration ◗ Planting Multi-scaled vegetation data to of native oaks at Bouverie Caitlin Cornwall (Sonoma predict wildlife species birds, local habitat values, and Ecology Center), Community conservation implications. Preserve was conducted with distributions using a wildlife the help of school children. based assessment of biological habitat relationship model. Other associated work has health of riparian wetlands in Annual monitoring involved Bibit Traut (UC Davis), involved the effects of winter the Sonoma Creek watershed. Structure storms and food availability on measurements of oak sapling and function of coastal high- energy balance and habitat survivorship and vigor as well Elizabeth Dahm (Sonoma State saltmarsh ecotones. use. as breeding bird censuses (see University), Larval amphibian survey of vernal wetlands. the Ardeid (Ar-DEE-id), n., refers to Ardeid any member of the family Ardeidae, which includes herons, egrets, and bitterns.

The Ardeid is published annually by Audubon Canyon Ranch as an offering to field observers, volunteers, and supporters of ACR Research and Resource Management. To receive The Ardeid, please call or write to the Cypress Grove Research Center. Subscriptions are available free of charge; however, contributions are gratefully accepted. ©2002 Audubon Canyon Ranch. Printed on recycled paper. Managing Editor, John Kelly. Layout design by Claire Peaslee. Research and Resource AUDUBON CANYON RANCH IS A SYSTEM OF WILDLIFE SANCTUARIES Management at AND CENTERS FOR NATURE EDUCATION Audubon Canyon Ranch BOLINAS LAGOON PRESERVE • CYPRESS GROVE RESEARCH CENTER • BOUVERIE PRESERVE

Oak forest canopy see pages 6 and 10 DANIEL GLUESENKAMP

Audubon Canyon Ranch, 4900 Shoreline Hwy., Stinson Beach, CA 94970 Nonprofit Org. Cypress Grove Research Center U.S. Postage P.O. Box 808 PAID Marshall, CA 94940 Permit No. 2 (415) 663-8203 Stinson Beach, CA