2004 2004 the ARDEID Page 1

Home , Abbotts Lagoon, Audubon Canyon Ranch, Common raven, Drakes Estero, Olema Valley

Research and Resource Management at Audubon Canyon Ranch Ardeidthe

◗ behavior in

heronries

Common Raven

◗ spatial dimensions

Raven home

ranges

◗ power of rainfall

Livermore

Marsh

◗ a test case

Eliminating

Ehrharta

◗ unseen soil

dwellers

Ehrharta

underground 2004 2004 the ARDEID page 1

Common Raven behaviors in heronries Vague Consequences of Omnipresence by John P. Kelly

t’s as if the ravens are not even there. Although Common Ravens fly frequent- Ily through some heronries in search of opportunities to prey on eggs and chicks, nesting Great Egrets exhibit little or no defiance. In fact, they seem to show a complete lack of interest or concern. In addition, Great Egrets leave their nests unguarded as nestlings approach three weeks of age, several weeks before young egrets can fly—even though many of these broods are taken by patrolling ravens. New cohorts of herons and egrets continue to disperse from heronries each summer and to contribute beautifully to the life in our marshes and estuaries. But raven numbers are booming throughout most of the San Francisco Bay area (see The Ardeid 2001). Whether heron and egret populations will continue to repro- duce adequately under the growing presence of ravens or, alternatively, suffer major declines in productivity has become a common concern among observers of heronries. The particular threat posed by ravens is difficult to understand fully, because most instances of nest predation are phantom events Figure 1. Study area and heronries (circles) in the northern San Francisco Bay area, including colony sites revealed only by sudden nest vacancies. for extended watches (*) and intensive observations of raven activity (PICA: Picher Canyon; WMAR: West The wisdom of egret behaviors in the Marin Island; DRES: Drakes Estero). presence of ravens is also mysterious. Their apparent carelessness is consistent among individuals and structured by a Measuring every move surrounded by grazed prairie grasses and long evolution of adaptive responses to rom 1999 through 2003, Audubon vegetated dunes; and (3) West Marin complex ecological forces, including nest Canyon Ranch field observers meas- Island, a 2.7-acre “rock” in the Marin predation. Scientific attempts to under- Fured nest mortality and raven Islands National Wildlife Refuge just off stand how birds respond to the risk of occurrence at all of the known heronries the eastern Marin County shoreline, nest predation have failed to discover reli- in the northern San Francisco Bay area where a blend of California buckeyes and able ways to predict nest success. There- (Figure 1). At ten of these sites, we con- woody shrubs often supports the San fore, to learn more about the threat of ducted two all-day watches for raven Francisco Bay area’s largest concentration growing raven populations on heron and activity. We then selected three heronries of nesting Great Egrets, Snowy Egrets, egret nesting colonies (see The Ardeid with resident ravens for more intensive Black-crowned Night-Herons, and Great 2002), I conducted an investigation of study: (1) ACR’s Picher Canyon, near the Blue Herons (see The Ardeid 2003). predator—rather than prey—behaviors shoreline of Bolinas Lagoon, where Great At each of the three heronries selected (Kelly et al. in review). Collaborators Blue Herons, Great Egrets, and Snowy for close study, we recorded detailed included ACR’s Katie Etienne, Jennifer Egrets settle among the branches of coast information on raven occupancy, land- Roth of PRBO, several ACR field biologists, redwoods in the bottom of a narrow ings, patrol flights, and interactions with and numerous volunteer observers. canyon; (2) Drakes Estero in the Point other species. At five-minute intervals Specifically, we asked: under what condi- Reyes National Seashore, where nesting during hundreds of two-hour observation tions and to what extent do ravens exploit Great Egrets and Great Blue Herons periods, we recorded the presence of heronries for food? crowd into a small patch of bishop pines continued on page 2 the ARDEID 2004

In this issue Vague Consequences of Omnipresence: Common Raven behavior in heronries ◗ by John P. Kelly ...... page 1 The Spatial Dimensions of Raven Life: Home-range dynamics in western Marin County ◗ by Jennifer Roth ...... page 4 Eliminating Ehrharta: Bolinas Lagoon Preserve as a test area for regional conservation ◗ by Daniel Gluesenkamp ...... page 6 Ehrharta Underground: ◗ by Gwen Heistand...... page 6 The Power of Rainfall: Influences on the hydro-geomorphology of Livermore Marsh ◗ by Katie Etienne ...... page 10

Cover photo: Common Raven by Peter LaTourrette ◗ Ardeid masthead Audubon Canyon Ranch Great Blue Heron ink wash painting by Claudia Chapline Research and Resource Management Executive Staff The Watch Skip Schwartz, Executive Director John Petersen, Associate Director Volunteers for ACR research or habitat restoration projects since The Yvonne Pierce, Administrative Director Ardeid 2003. Please call (415) 663-8203 if your name should have been Staff Biologists included in this list. John Kelly, PhD, Director, Research & Resource PROJECT CLASSIFICATIONS: Management Rebecca Anderson-Jones, Bouverie Preserve C = Coastal Habitat Restoration at Toms Point E = Ehrharta erecta Removal at Bolinas Lagoon Preserve G = Grassland Management at Bouverie Preserve Katie Etienne, Research Coordinator H = Heron/Egret Project M = Livermore Marsh and Olema Marsh Surveys Daniel Gluesenkamp, PhD, Habitat Protection and P = Photo Points R = Habitat Restoration S = Tomales Bay Shorebird Restoration Specialist Project T = Turkey Research at Bouverie Preserve W = Tomales Bay Gwen Heistand, Bolinas Lagoon Preserve Waterbird Census Land Stewards Nancy Abreu (H); Ken Ackerman Ellen Krebs (H); Carol Kuelper (S); Andy Bill Arthur, Bolinas Lagoon Preserve (C,G,S,W); Drew Alden (R); Judy Allen LaCasse (H); Joan Lamphier (H); Jean- David Greene, Tomales Bay properties (C); Leslie Allen (R); Sarah Allen (S); Bob Michel Lapeyrade (G); Laura Leek (W); John Martin, Bouverie Preserve Anderson (H); Janica Anderson (H); Bill Lenarz (H); Stephanie Lennox (H); Field Biologists Autodesk employees (R); Bob Baez Robin Leong (H); Liz Lewis (M); Eileen (S,W); Norah Bain (H); Bruce Bajema Libby (H); Patricia Lonacker (R); Flora Ellen Blustein (H,S); Katy Baty (W); Tom Baty (W); Maclise (H); Phil Madden (R); Art & Lauren Hammack Gordon Bennett (S,W); Shelly Benson Lynn Magill (R); Jean Mann (R); Alan Mark McCaustland (R,W); Sherman Bielfelt (G); Gay Bishop Margolis (R); Marin Conservation Corps Michael Parkes (S); Mike Blick (G); Julie Blumenthal (R); Roger Marlowe (W); Janaea Martin (W); Len Blumin (R,W); Patti Blumin (G); Mark Mc Caustland (H,M,S); David Data Management (H,R); Ellen Blustein (H,S); Noelle Bon McConnell (H); Jean Miller (H); Ann Melissa Davis (H); Janet Bosshard (H); Tom Bradner Mintie (E); Len Nelson (H); Wally Research Associates (E); Anna-Marie Bratton (E); Emily Neville (H); Linda Nicoletto (C); Terry Brockman (H,S,W); Ken Burton (S,W); Nordbye (S,W); Richard Panzer (H); Jules Evens Phil Burton (H); Denise Cadman (H); Jennie Pardi (H); Bob Parker (G); Mike Flora Maclise Ann Cassidy (H); Bonnie Clark (S); Parkes (H,R,S,W); Lorraine Parsons (M); Helen Pratt Michelle Coppoletta (R); Rig Currie (S); Tony Paz (E); Joelle Peebles (G); Rich Stallcup Melissa Davis (H); Nick Degenhardt (H); Precious Peoples (H); Kate Peterlein Jack Dineen (H); Carolyn Dixon (H); (S,W); Jeff and Kathy Peterson (R); Resource Management Associates Giselle Downard (H); Roberta Downey Myrlee Potosnak (H); Linda & Jeff Len Blumin (C,R); Joe Drennan (W); Judy Dugan (E); Reichel (H); Rudi Richardson (W); Roberta Downey Bob Dyer (H); Cathy Evangelista (H); DeAnn Rushall (T); Ellen Sabine (H); Scientific Advisory Panel Jules Evens (R,S,W); Katie Fehring Marilyn Sanders (H); Fran Scarlett (H); (S,W); David Ferrera (W); John Finger Cindy Schafer (S); Phyllis Schmitt (H,R); Sarah Allen, PhD (R); Binny Fischer (H); Grant & Ginny Gordon & Laurie Schremp (H); Alice Peter Connors, PhD Fletcher (P,S); Carol Fraker (H); Daniel Shultz (H); Asha Setty (C); Joe Smith Jules Evens George (S,W); Tony Gilbert (H,S,W); (W); Anne Spencer (S); Bob Spofford Kent Julin, PhD Beryl & Dohn Glitz (H); Mike Green (H); Rich Stallcup (M,S); Jean Greg Kamman (W); Philip Greene (H); Bill Grummer Starkweather (H); Jim & Jeanne (H); Karlene Hall (R); Madelon Halpern Sternbergh (H); Michael Stevenson Rachel Kamman (H); Lauren Hammack (M); Fred (S,W); Ron Storey (H); John Sutherland Chris Kjeldsen, PhD Hanson (S,W); Roger Harshaw (S); (C); Lowell Sykes (H,S,W); Judy Temko Doug Oman, PhD Michael Henkes (G); Diane Hichwa (H,S); Janet Thiessen (H); Gil & Dodie Helen Pratt (H,S); Judy Hiltner (C); Peter Horn (S); Thomson (H); Peggy Thorpe (H); Lisa W. David Shuford Roger Hothem (H); Steve Howell (W); Toet (E,R); Travelsmith employee Rich Stallcup Lisa Hug (S,W); Jeri Jacobson (H,S); service day (R); Sue Tredick (H); Susan Wesley W. Weathers, PhD Rosemary Jepson (G); Rick Johnson Tremblay (G); Tanis Walters (S); Ryan (H,S,W); Calvin Jones (G); Gail Kabat Watanabe (R); Penny Watson (W); (W); Lynnette Kahn (H,S); Bob Rosilyn & Tom White (W); Phyllis Kaufman (H); Guy Kay (H); Thomas Williams (H); Jessica Wilson (C); Ken Kehrlein (G); Richard Kirschman (S); Wilson (W); Katy Zaremba (R). page 2 the ARDEID 2004

food. Because nesting ravens occupy the same home ranges across years (see article on page 3), nest predation in waterbird colonies by new resident pairs of ravens might increase as neophobia subsides. We did not find significant annual increases in nest predation, but we did find significant annual increases in raven predatory behaviors: longer patrol flights, Figure 3. Great Egret nest failures known to have more landings, more frequent movement resulted from raven predation peaked as nestlings among landing sites, and more flushing approached three weeks of age, at ACR’s Picher and harassment of nesting Great Egrets. In Canyon, 2000-2001. addition, recovered (fresh) prey remains at the Marin Islands suggested increasing predation of adult Snowy Egrets: ravens Evaluating occurrence killed at least four adult Snowies in 2000, seven in 2001, 15 in 2002, and eight in ur surveys revealed substantial 2003. On one occasion, ACR biologist nest predation by ravens at some Mark McCaustland watched in awe as a Oheronries in the region but high raven chased an adult Snowy Egret to the variability among colony sites—even ground and killed it. Such trends in behav- though ravens are common throughout ior are consistent with the attenuation of the San Francisco Bay area. We estimated neophobia in adult ravens and suggest a regional average of only about one that nest predation might increase annu- raven occurrence in each heronry every ally for some years after ravens begin to five hours and detected ravens in only occupy heronries. about one-fourth of the heronries each The notion that ravens might become year. At Picher Canyon, resident ravens more daring predators across years is were present less than 5% of the time, but consistent with previously known behav- ravens occupied colonies at West Marin iors. For example, ravens may have diffi- Island and Drakes Estero about 30% of culty overcoming neophobia by their typ- the time. The resident ravens occupied all ical means of approaching slowly Figure 2. Raven predation on Great Egret nests, three sites of intensive study more often (Heinrich 1988, Condor 90: 950-952), length of raven patrol flights, and number of raven after mid-June, often in pairs or accom- because egrets build their nests in isolat- interactions with other species increased signifi- panied by fledgling ravens. Keep in mind, ed locations in the nest-tree canopy. cantly (P < 0.05) with increased raven productivity. however, that ravens occurred only rarely Ravens also require unusually long peri- at most other heronries. ods of time, under experimental condi- Resident ravens preyed on about 27 tions, to accept carcasses of large birds ravens in and near colony sites, identified Great Egret nests per year at West Marin (Heinrich et al. 1995, Auk 112: 499-503). perching substrates, and noted their posi- Island, 20 nests per year at Picher Perhaps most importantly, however, the tions above or below the tree canopy or Canyon, and only two to three nests per risk of major injury if egrets become on open ground. We measured how long year at Drakes Estero. Great Blue Heron defensive may be enough to prolong their ravens remained at each landing site and nest mortality was less than two nests per neophobic behavior. estimated their distances to heron or year at each site and, for most of these, egret nests. Whenever a raven landed in we did not determine the cause of failure. Assessing predation risk or near a nest, we recorded the nest con- Of 11 instances of complete colony site he lack of annual increases in nest tents, nesting stage, age of chicks, and abandonment in the region, only one—at predation at the three study sites, in number of adult herons or egrets flushed a site with nesting Great Blue Herons and Tspite of annual increases in associ- from the area. For each patrol flight, we Great Egrets—was associated with repeat- ated behaviors, suggests that resident recorded the time of day, flight duration, ed heavy disturbance by ravens, and a ravens may interfere with the activities of height above nests, and alert responses of single Great Blue Heron nest was estab- other nest predators, such as raccoons, adult or nestling herons or egrets. When lished there the following year. owls, or raptors. Ravens were dominant in ravens interacted with other bird species, The low regional rates of raven occur- 93% of interactions with visiting bird we determined which individuals were rence and nest predation in heron and species such as Osprey, Golden Eagle, dominant and subordinate, the duration egret colonies were associated with recent Red-tailed Hawk, Peregrine Falcon, and of chases, and other details. And of influxes of ravens. If ravens move into an American Crow. In one instance, a Red- course, we recorded any evidence of nest area without prior experience in heron- tailed Hawk that was flushing egrets from predation. From these data, we looked for ries, they may exhibit “neophobia” for the their nests at the Marin Islands was evidence that the nest predatory behav- first few to several nesting seasons. Neo- harassed repeatedly by the resident iors of ravens might be influenced by phobia is a characteristic cautiousness ravens until it left the area. Although their foraging experience, energy needs, toward novel food items that subsides as ravens often chase visitors away, they also time of day or season, or the availability ravens learn through experience not to respond opportunistically to nest distur- of prey. fear what may later become important see Omnipresence, page 5 2004 the ARDEID page 3

Home-range dynamics in western Marin County The Spatial Dimensions of Raven Life by Jennifer E. Roth

Figure 1. Locations, centers of activity (50% occurrence), and home ranges (95% occurrence) of the female (dashed lines) and male (solid lines) ravens nesting near ACR’s Picher Canyon heronry. Unlabeled outlying areas are isolated portions of home ranges.

On one occasion, a bird that discovered our traps and escaped without being captured flew into the air, circling and calling angrily. Ravens came from all directions. We soon had a large group of ravens circling the area and calling. The warning had gone out! Conversely, imag- ine our excitement each time we handled one of these large, intelligent birds. Each bird responded differently to being captured; some birds struggled and others quietly observed us. Each release was followed by a few tense moments as we watched to see whether there would be any problems with the harnesses we used to attach the radio-transmitters to the arly one morning, a pair of ravens Capturing ravens was one of the most dif- birds. Fortunately, all radio-tagging activi- flies over the heron and egret ficult parts of the study, and we went to ties were successfully completed without Ecolony at ACR’s Picher Canyon in great lengths to outsmart these intelligent mishap, but the challenges of following search of unattended eggs or young birds. We started by identifying birds that their movements were just beginning. chicks. At Point Reyes National Seashore, we were interested in studying and then We radio-tracked each bird for one to ravens soar routinely over dune beaches watched them for several days to learn three years. Each time we detected a bird, in search of Snowy Plover nests. Mean- more about their habits and decide on a we mapped its location on a topographic while, at Tony’s Seafood Restaurant on suitable trapping location. Once we map and recorded information on behav- Tomales Bay, a pair of ravens lingers each decided to attempt a capture, we rose ior and habitat use. For example, we day in the vicinity of a nearby dumpster. several hours before dawn to place the recorded birds flying over grazed grass- What impact is raven predation having on bait and hide our traps and ourselves lands, foraging along beaches, attending heronries and rare bird species? What fac- before first light; the birds would avoid nests in cypress trees, and roosting in tors are influencing their movements and the area for days at the first hint that pine groves. We also mapped nest sites use of the landscape? These questions something was amiss. We hid in bushes and concentrated food sources in the prompted us to begin radio-tracking and ditches, covering ourselves with area. We used the data we collected to breeding ravens in western Marin County branches and dried grass. We often spent estimate the home-range size of each bird in order to learn more about their impact several hours in our makeshift blinds and to evaluate some of the factors likely on sensitive bird species and the factors before the ravens came near the bait. to affect home-range size and space use affecting their distribution (Roth et al. Ravens are surprisingly cautious around within home ranges. unknown food sources, often approach- 2004). Home-range size ing the area and backing away several Radio-tracking times before eating anything. Mean- home range is the area used by an e captured and attached back- while, we waited breathlessly as they animal during normal activities pack-mounted radio-transmit- approached the traps or moved into a Asuch as food gathering, mating, Wters to 16 adult ravens from 15 position where we could shoot a net over and caring for young. We measured each resident pairs, including both members of them. Our disappointment over near home range by calculating the area that the pair residing near Picher Canyon. misses was tangible. continued on page 4 page 4 the ARDEID 2004

season due to differences in source to their nest, and they occupied behavior between the sexes: relatively large home ranges compared to females are largely responsi- other ravens in western Marin County ble for incubation and males (Figure 2). leave the vicinity of the nest to forage while females are Space use within home ranges attending the nest. Contrary irds are likely to use some areas to our expectations, home- more intensively than others due to range size did not differ sig- Bthe patchy distribution of food and nificantly between female habitat areas suitable for feeding, caching and male ravens in our study. food, and nesting. We evaluated space use Also, we did not find a signifi- within home ranges to determine cant difference in the distri- whether ravens’ locations exhibited ran- bution of locations for the dom, uniform, or clumped distributions. female and male of the Picher We also estimated the size of core areas or Canyon pair (Figure 1). It may centers of activity, defined as areas with a be that sex roles do not differ 50% probability of an individual raven’s Figure 2. Raven home-range size increases with the distance enough during the breeding occurrence (Figure 1). We found that loca- between the nest and the nearest concentrated food source (P < season to affect overall home- tions of females were significantly 0.05). Labels indicate the home ranges of ravens occupying famil- range size in ravens. Alterna- clumped within 83% and 100% of home iar locations in western Marin County. The male raven at Picher tively, the similar home-range ranges in 2000 and 2001, respectively. Canyon was not included in the analysis. Note the log scale on sizes between male and Males were more variable in their use of both axes. female ravens may reflect space, with locations clumped in only their tendency to interact fre- 38% and 44% of home ranges in 2000 and quently with their mates 2001, respectively. encompassed a 95% probability of an through courtship, pair-bond reinforce- We used these data to determine the individual raven’s occurrence, assuming ment, increased vigilance by the male relationship between centers of activity, that this area would include all normal during egg laying and incubation, or nest sites, and concentrated food sources. activities and exclude rare excursions to other social behaviors associated with All 16 of the radio-tagged birds centered more distant areas (Figure 1). nesting (Boarman and Heinrich 1999). their activities around their nest sites. In Home-range size was highly variable Distance to food sources some cases, there was a concentrated for the ravens in our study (Figure 2). food source within that center of activity. Sizes ranged from 0.2–4.9 km2 for females n general, prey abundance and dis- Other birds had two centers of activity, and 0.3–4.9 km2 for males. The home tance to food and water influence with one centered around the nest site ranges of the Picher Canyon pair were Ihome-range size and movements in and one centered around a concentrated large compared to other pairs: 4.2 km2 for birds. Besides preying on egret eggs and food source. The Picher Canyon ravens the female and 4.9 km2 for the male. chicks, ravens in our study fed on small were unique among the individuals we Many factors may contribute to varia- mammals and reptiles; grain, calf carcass- studied in having three distinct centers of tion in home-range size among individu- es, and afterbirths found at many local activity that centered around (1) their als. For example, food supply, competi- dairies; human refuse at beaches, parking nest site, (2) the Picher Canyon heron and tion for food, territoriality, vegetation lots, and dumpsters; and eggs and young egret colony, and (3) a nearby horse ranch type, and body size might influence the of other birds in the area. On several where they could rely on a daily supply of distances birds must travel. We tested the occasions, we even saw ravens harass grain at feeding time (Figure 1). These possibility that differences in sex and dis- Turkey Vultures until they regurgitated activity centers were separated by habitat tance to important sources of food could their food. The ravens were able to catch areas that the ravens only rarely occupied. the regurgitations in mid-air! However, explain the large differences in home- Annual variation range size among ravens in western despite the wide variety in raven diets and Marin County. some unusual foraging behaviors, home- ike most things in nature, food sup- range sizes are most likely to be influ- ply, habitat quality, and other factors Sex differences enced by the locations of stable, concen- Lthat affect home-range use, may ex may influence home-range size in trated sources of food. vary from year to year, causing correspon- birds when there are differences in We determined the distance between ding changes in home-range size. We Sbody size or behavior between each raven nest and the nearest ranch, compared home-range sizes between females and males. When males and human food source, or waterbird colony years and found no significant differ- females differ greatly in size, associated in order to evaluate the influence of con- ences-large home ranges remained large differences in agility and power might centrated food sources on home-range and small ones remained small. However, allow them to compete less for particular size. We found a general trend indicating there were significant small-scale shifts in prey or perhaps increase their collective that ravens nesting farther away from home-range placement for most females menu of available prey. However, male concentrated food sources had larger (67%) and males (63%). These shifts ravens are only slightly larger than home ranges (Figure 2). For instance, the were associated with changes in nest females. Nonetheless, we expected to find Picher Canyon ravens nested about 1 km location and the distribution of desirable a difference between female and male away from the heron and egret colony, food resources between years. Given the home-range sizes during the breeding which was the nearest concentrated food continued page 5, top 2004 the ARDEID page 5

apparent association between ravens and States (Sauer et al. 1997). Their successful References cited concentrated food sources in the area, adaptation to human-dominated land- Boarman, W. I. 2003. Managing a subsidized pred- variation in the size and placement of scapes has likely been a major contribut- ator population: reducing Common Raven pre- home ranges likely reflected variation in ing factor to these population increases, dation on desert tortoises. Environmental Management 32: 205-217. grazing or harvesting rotations on ranch- and such adaptations are evident in the Boarman, W. I., and B. Heinrich. 1999. Common es, the distribution of and access to home-range dynamics of individuals Raven (Corvus corax). In A. Poole and F. Gill human foods and garbage, and the sea- ravens. Because ravens structure their [Eds.], The Birds of North America, No. 476. The sonal timing and reproductive success of lives, in part, around ours, human land Birds of North America, Inc., Philadelphia, PA. nearby waterbird colonies. use has become an important considera- Kelly, J. P., K. L. Etienne, and J. E. Roth. 2002. Abundance and distribution of the Common It may be no surprise that the way in tion for resource managers concerned Raven and American Crow in the San Francisco which ravens are distributed across the with the effects of increasing raven popu- Bay Area, California. Western Birds 33: 202-217. landscape is related to both natural and lations on sensitive species (Boarman Roth, J. E., J. P. Kelly, W. J. Sydeman, and M. A. anthropogenic features of the environ- 2003). Such concerns suggest that funda- Colwell. 2004. Sex differences in space use of breeding Common Ravens in western Marin ment. Raven populations are increasing mental changes in how we manage agri- County, California. Condor 106: 529-539. in many parts of the San Francisco Bay cultural, recreational, and urban environ- Sauer, J. R., J. E. Hines, G. Gough, I. Thomas, and Area (Kelly et al. 2002; see The Ardeid ments may be necessary to control the B. G. Peterjohn. 1997. The North American 2002) and throughout the western United effects of growing raven populations. ◗ breeding bird survey results and analysis. Version 96.4. Patuxent Wildlife Research Center, Laurel, MD.

Omnipresence, from page 2 Predation of Great Egret nests was bances by other predators or humans. most likely early in the post-guardian We did not test directly whether ravens period, when parents are absent and influence other nest predators, but our nestlings can be taken easily by ravens results suggested that ravens might (Figure 3, page 2). Interestingly, the reduce or displace the nest predatory timing of nest predation by ravens at activities of other species: raven preda- Picher Canyon did not differ from that tion on nestling Great Egrets at Picher measured for 1968–1979 (Pratt and Canyon did not differ significantly Winkler 1985), when ravens were not from 1968–1979 predation levels (Pratt present (P > 0.91). So the timing of nest and Winkler 1985, Auk 102: 49-63), predation reflects a general pattern of when ravens were not present vulnerability rather than the timing of (P=0.56). raven predation per se. One consistent pattern we found Most observers have noticed that was that ravens with large families ravens are now present throughout our engaged in more nest predatory activi- region and that their overall numbers

ty (Figure 2, page 2). This suggests that PHIL NOTT are increasing. But our closer look at the extent of raven predation depends Color-banded Common Raven at Abbotts Lagoon, Point raven behavior suggests that we on the food demand of resident Reyes National Seashore. should be cautious about assuming ravens. If so, nest predation might be increases in nest predation in heron- managed effectively by limiting raven ries. On the other hand, where ravens reproductive success. However, this inter- their nest site and the heronry (see article have been present for only a few to sever- esting possibility has not been tested. on page 3). al years, nest predation might increase At the Marin Islands, we surveyed prey In general, ravens were more likely to with continuing declines in neophobia. remains in the vicinity of the raven nest occupy heronries after 10 AM and before To further complicate things, whether and in a cache and shell dump area used 3 PM. This suggests possible early morn- ravens add to, displace, or buffer nest pre- routinely by the resident ravens. The ing foraging for other available food, dation by other species remains a mys- results of these surveys, and the timing such as road kills, or improved opportu- tery. As a result, the extent to which and rates of nest predation on Great nities for nest predation with midday ravens occupy heronries may not closely Egrets, indicated that the resident ravens declines in colony attendance by adult reflect predation risk. obtained most or all of their energy needs herons and egrets. In contrast, West For those who watch the lives of from the heronry. Such dependence on Marin Island ravens were present contin- ravens with particular interest, it should the heronry for food is consistent with the uously through the day, spending little or be no surprise that their behavior chal- apparent link between predatory activity no time patrolling other areas for food. lenges as well as inspires our understand- and food demand in resident ravens, but They were also more likely to occupy ing of nature. In assessing the risk of contrasts strongly with low rates of nest colony positions below the nest-tree raven predation, herons and egrets as well predation at most other colony sites in canopy in early morning, perhaps forag- as human observers should look beyond the region. At other sites, alternative food ing for fallen chicks. However, the overall the mere presence of these prominent sources may be more available. For exam- risk of nest predation in heronries was predators. ◗ ple, resident ravens at Picher Canyon best revealed by the prevalence of raven spend much of their time feeding at a activities (such as number of landings)— horse ranch approximately 2 km from not by how often ravens were present. page 6 the ARDEID 2004

ACR’s Bolinas Lagoon Preserve as a test area for regional conservation Eliminating Ehrharta

by Daniel Gluesenkamp

Ehrharta erecta. Ehrharta covering native habitat. Pike County Gulch.

n 1930, near the University of Califor- Ranch’s efforts to understand and defend invading western Marin, with large popu- nia Berkeley campus, an alien invader against its spread. lations in Olema Valley, along the Iescaped from a government-sponsored Ehrharta erecta is a highly invasive Panoramic Highway on Mt. Tamalpais, genetics experiment. Although the perennial grass native to South Africa. around the towns of Inverness and escapee was a genetically natural (not The species was part of an experiment at Bolinas, and in Audubon Canyon Ranch’s genetically modified) species, created by the U.C. Berkeley botanical garden inves- Bolinas Lagoon Preserve. evolution, its escape into a new land has tigating whether increasing chromosome Ehrharta erecta is a prolific seed pro- had serious consequences to the natural numbers increases invasive ability; as it ducer, and the small seeds likely are car- inhabitants of our region. In the decades turns out, Ehrharta erecta was tremen- ried to new sites via soil on shoes, on deer since, the green monster has spread qui- dously invasive without any alteration hooves, and as contamination in potted etly, wiping out resident flora and fauna whatsoever. By 1950 the species was plants. Ehrharta can thrive in an extreme- as it invades, and is now on the verge of abundant at the U.C. Berkeley campus, ly wide range of habitats, from coastal taking over much of western Marin and in 1996 the tremendous abundance dunes to closed-canopy forest, and forms County. The invader is a South African of Ehrharta erecta in San Francisco natu- robust monospecific stands under full sun grass named Ehrharta erecta, and this ral areas compelled recognition of the or in as little as 2.5% of daylight (Haubensak article tells the story of Audubon Canyon species as a major conservation concern and Smyth 2000). Once established, popu- (Sigg 1996). Ehrharta erecta is currently lations increase rapidly and Ehrharta

Ehrharta Underground: There’s more to an invasion than meets the eye by Gwen Heistand

ryptosphere—even the winged beetles, and larval environment’s physical attempt to understand the name given to the soil insects of all sorts, to name a attributes. The dense sward impact of the invasive grass Cand leaf litter environ- few. The diversity and com- of green grass sweeping up on soil fauna. ment is laced with the promise plexity of the cryptosphere has ACR’s Pike County Gulch is Hidden ecologies of discovery. What creatures been compared to coral reefs an obvious result of Ehrharta ne square meter of fer- inhabit this hidden world? and tropical rainforests. And invasion. Impacts of the tile soil can contain Strange primitive insects with- yet, we know very little about invader on soil chemistry and more individual organ- out wings (some without it. biotic communities are O isms than all the humans that eyes), springtails (named for As mentioned in the unknown. While it is not fea- have ever lived. Various stud- their anal appendages called accompanying article, sible at present to perform a ies have reported up to 1012 furcula that propel them Ehrharta erecta creates a detailed study, one piece of bacteria (1 trillion), 1012 proto- through soil pore spaces), dense thatch of vegetation the Ehrharta experiment zoa, 107 nematodes, spring- nematodes, thousands of both above and below involves collecting data on tails and mites, 107 insects, species of mites, feather- ground, thereby altering the leaf litter invertebrates in an 2004 the ARDEID page 7

immediate action is required to prevent Ehrharta from expanding, merging, and smothering the natural diversity of this singular preserve. Three lines of defense n 2003 ACR’s Habitat Protection and Restoration (HPR) Program initiated Ian ambitious project to address the threat posed by this aggressive invader. The project includes (1) scientific Ehrharta removal plot. Uninvaded versus invaded habitat. research targeted to identify techniques for controlling Ehrharta erecta; (2) eradication of Ehrharta on ACR lands; and (3) becomes the dominant herbaceous plant, We have recently discovered several collaborating with other groups and land excluding native and non-native vegeta- patches of Ehrharta erecta in each of the managers to develop regional solutions tion (McIntyre and Ladiges 1985). There three canyons that comprise ACR’s to the Ehrharta invasion. are currently no established techniques Bolinas Lagoon Preserve. While most of In June 2003 we began an experimen- for controlling the plant or for restoring the patches are still small, the Ehrharta tal assessment of Ehrharta removal tech- invaded habitat (Pickart 2000). infestation in Pike County Gulch provides niques. The experiment compares three a glimpse at one promising techniques: manual removal possible future for by pulling, dilute herbicide treatment, ACR’s flagship pre- and solarization (covering with black serve: nearly 2000 plastic). Each treatment is applied to 10 m2 of the canyon Ehrharta-dominated plots. Data collect- floor is covered ed from this experiment will allow us to with a dense determine which Ehrharta removal treat- monospecific ment is most effective and to evaluate sward, and the how each treatment influences recovery bright green of the natural communities that we are invader is climb- trying to restore. ing the sides of the In addition to the 30 experimental canyon into the plots designed to evaluate removal tech- brown leaf litter of niques, the Ehrharta experiment the natural herba- includes two different sets of unmanipu- ceous understory. lated controls with 10 plots each, Given the plant’s Ehrharta-dominated plots and un- Figure 1. After one year, all three treatments significantly reduced the percent amazing rate of invaded plots. In each of the 50 plots, we cover of Ehrharta erecta, relative to the invaded control plots, while manual spread and ability are quantifying the abundance of removal also stimulated germination from the seed bank (see text). Solid bars to thrive in a Ehrharta, vegetation composition, the indicate the percent of plots covered by Ehrharta; error bars indicate one stan- abundance of native plants and of other dard error (P < 0.001). range of habitats, continued on page 8

1,000 earthworms, on fungus, some feeders and root-feeding native plants may affect above and 20,000 km of on roots, and nematodes. Forests display a and belowground feedbacks, fungal mycelia. some on other relative abundance of fungal potentially changing soil (Neher 1999, invertebrates. feeders (Neher 1999). chemistry and altering soil Pennisi 2004). Where stud- Belowground food webs biota to facilitate invasion (van Bacteria, fungi, ied, different and ecological relationships der Putten 2002). and nematodes ecosystem are intricate and intimately Hidden consequences? are three of the types have linked to the aboveground taxonomic groups been found to environment (Wardel et al. uring February that have the lowest support different 2004, Wardel 2002). Science is through April of 2004, percentage of described assemblages of soil just beginning to delve into DDan and I worked with species, making classification biota. Grasslands have shown these connections. Few stud- several volunteers to collect of soil communities difficult at a marked abundance of ies have attempted to assess 350 1-cm2 litter samples from best. In addition, within each omnivorous soil invertebrates. soil biota in relation to either the center of each of the 50 taxonomic group, some Agricultural soils can have habitat restoration or invasive experimental plots used in the species feed on bacteria, some greater numbers of bacterial species. Invasion of non- Erharta study. Each sample page 8 the ARDEID 2004

bags were buried at each of 10 locations, and the first set of bags will be exhumed and tested for viability in November 2004. Early observations of buried seed bags provide reason for hope. Each of the marker stakes is surrounded by a constel- lation of small clusters of Ehrharta seedlings, one cluster for each seed bag! This shows that the germination rate of these buried seeds is very high and that relatively few seeds remain dormant, and suggests that control efforts may not have Ehrharta plot. Buried seed bag sprouting (Ehrharta in circles). to contend with a large and persistent soil seedbank. Although the increasing ground cover non-native invaders, important habitat manual removal actually stimulates by Ehrharta clearly reduced the cover of characteristics, and leaf-litter inverte- Ehrharta seed germination and increases native plants (Figure 2), recovery of native brate communities. abundance of the invader (Figure 1). plants one year after treatment did not Preliminary results indicate that a While Year One results show solarization differ significantly among treatments. weak herbicide solution is most effective to be moderately effective at reducing However, differences in native cover at reducing Ehrharta abundance, while Ehrharta cover, some adult Ehrharta among treatments may become signifi- plants are able to sur- cant as the small native plants mature. vive beneath the plastic Probably the most interesting variable tarp; the utility of this that we will be monitoring is the response technique will depend of leaf-litter invertebrate critters (see on how well these sur- accompanying article). Very few scientific vivors recover in the studies have quantified the impacts of next year. plant invasion on animal communities, Ehrharta control and only a handful have assessed how may further depend on animal communities are affected by inva- limiting the ability of sive plant control efforts. Terrestrial inver- Ehrharta populations tebrates are extremely important, both as to recover from buried elements of decomposition and nutrient seeds, so we are experi- cycles and as the basis of many terrestrial mentally assessing seed food webs, and ACR’s Ehrharta research longevity by burying will be one of only a few studies that small bags of seeds at assess how the terrestrial invertebrate two depths, exhuming community is affected by non-native seedbags at regular plant invasion. It will be extremely inter- intervals, and germi- esting to observe how invertebrates nating the exhumed Figure 2. Native plant cover declines with increasing cover by Ehrharta recover following Ehrharta removal! erecta. Data are from unmanipulated plots (r2 = 0.59, P < 0.001). seeds to determine While the scale of Bolinas Lagoon seed survival. Twelve Preserve’s Ehrharta invasion is daunting,

was run through a Berlese tles), 13 diplurans (primitive in the grass dominated plots data, combined with data on Funnel, which uses light to wingless insects), and 11 versus more fungal feeders in the abundance Ehrharta and drive soil fauna downward unidentified larvae—with the native forest understory. other non-native and native into a container, producing approximately two-thirds of Plots covered in black plastic plant species, as well as data 1 1 approximately /8 to /4 tea- the sample still to be sorted! might be expected to have a characterizing plant species spoon of invertebrates. Once all samples have all different suite of creatures composition for each plot, Characterization, on a broad been processed, we will ana- during and immediately after should advance our under- scale, of invertebrate assem- lyze the results to determine treatment. One aspect of what standing of how Ehrharta blages in each of these sam- whether there is any differ- we are hoping to ascertain is erecta control, eradication, ples has begun. Just to give a ence in soil fauna between the how resilient the ecologically and dispersal might affect flavor of the numbers found in three treatments (herbicide, important leaf litter commu- native biota. less than a teaspoon, one sam- solarization, and manual nity is in response to restora- On a more personal note, ple currently being analyzed removal) as well as between tion efforts. Because so little is it is difficult to describe the contains 187 acari (mites), 214 the invaded control sites and known about soil ecology on thrill of seeing a microscopic collembolans, 3 dipterans uninvaded plots. We might ACR lands, whatever we dis- fly with beaded antennae and (flies), 12 coleopterans (bee- expect to see more omnivores cover will be exciting. These perfectly formed halteres or a 2004 the ARDEID page 9

assigned 2 interns to map all Ehrharta understand how natural systems work occurrences in Marin County. As ACR’s and is creating a fruitful nexus of collabo- HPR Specialist, I communicate frequent- ration between “pure” scientists and ly with biologists from California State “applied” restorationists. The urgent need Parks, the National Park Service, and for research and restoration volunteers other agencies to share information and has fueled a joyful reconnection of citi- insights regarding Ehrharta manage- zens with nature that is similar to the ment. When ACR’s Ehrharta experiment amateur scientist revolution of 19th cen- is completed we will use the data to tury England. Most importantly, the ubiq- develop a prescription for restoring uity and impact of biological invasions is Ehrharta-invaded habitat, and we expect helping humans realize that they must be that presentation of results will further active and conscientious stewards of this Volunteers set up the experiment. increase regional Ehrharta control planet: where Rachel Carson’s Silent efforts. Spring demonstrated that human actions Conservation commitment can have global impacts, the replacement our commitment to protecting the pre- of a diverse natural community by a sin- serve’s natural diversity mandates that we he genus Ehrharta is largely gle species of grass is teaching us that act to restrain this harmful invader. In the restricted to the Cape region of inaction is not without consequences. last year we have conducted surveys to TSouth Africa, which, like California, Over decades and centuries, Ehrharta locate and map all ACR Ehrharta patches is one of 5 regions on the planet with a erecta may or may not accumulate new and have begun work to treat the high Mediterranean climate. Ehrharta erecta pathogens and herbivores and become a priority sites. ACR has followed a “stitch in has the greatest geographic range of any well-regulated and harmless part of our time” policy with regard to control work, plants in the genus, and it is interesting to coastal plant communities. In the mean- focusing on small easily- eradicated think that Ehrharta erecta may, by its time, conservation organizations such as patches of Ehrharta that have the poten- nature, be a colonizer of new sites. In a ACR remain committed to defending tial to become intractable infestations. manner reminiscent of many of California’s ancient natural diversity from Results of this first year have been prom- California’s most marvelous natives, the the negative impacts of Ehrharta erecta ising, and we expect that several years of taxon named Ehrharta erecta emerged in and other runaway invaders. ◗ concerted effort by staff and volunteers an accelerated burst of speciation stimu- will be required to control this advanced lated by the appearance of summer-arid invasion. climate in southern Africa. Ehrharta erec- References cited Invasive species such as Ehrharta erecta is not a “bad” plant but, rather, is an Haubensak, K. and A. Smyth. 1999. Erharta erecta. Species assessment prepared for Channel ta do not recognize property boundaries, amazing and tremendously vital organ- ism that is the singular result of a unique Islands National Park. and so we are working with our neigh- McIntyre, S. and P. Y. Ladiges. 1985. Aspects of the bors and with other restorationists to evolutionary history. Unfortunately, its biology of Erharta erecta. Lam. Weed Research improve Ehrharta management across aggressive spread in coastal California 25: 21-32. the region. In the last 18 months, ACR threatens to eliminate some surviving Pickart, A. J. 2000. Ehrharta calycina, Erharta erec- representatives of California’s own unique ta, and Ehrharta longifora. In: Bossard, C. C., J. staff have helped make the Marin- M. Randall, and M. C. Hoshovsky (eds): Invasive Sonoma Weed Management Area (WMA) evolutionary history. plants of California wildlands. University of aware of the threat posed by Ehrharta Harmful invaders like Ehrharta pose California Press, Berkeley, CA. invasion, and as a result the WMA con- very significant challenges, but they also Sigg, J. 1996. Erharta erecta: sneak attack in the provide important opportunities. making? California Exotic Pest Plant Council tributed significant support to imple- News 4(3):8-9. menting ACR’s Ehrharta experiment and Research on invasive species is helping us

beetle larva with a wildly spectacular, minute beings who enjoy spending Saturdays Pennisi, E. 2004. The secret life of intricate exoskeleton that and a feeling of increased and Sundays bent over micro- fungi. Science 304: 1620-1622. van der Putten, W. H. 2002. How to looks like it is made from tor- respect for all that lives scopes keying out and counting be invasive. Nature 417: 32-33. toise shell or the field of view underfoot, I can’t help but wee beasties found in the cryp- Wardle, D. A., R.D. Bardgett, J.N. under a dissecting scope wonder what we may be tosphere. Many thanks go to Klironomos, H. Setala, W.H. van completely filled with mites of doing to alter this world as Tom Bradner, Anna-Marie der Putten, and D.H. Wall 2004. different sizes and shapes. we move invasive species Bratton, Judy Dugan, Ann Ecological linkages between aboveground and belowground Knowing that these creatures around the globe and prac- Mintie, and Tony Paz. biota. Science 304: 1629-1633. live their lives navigating the tice less than sustainable Wardle, D. A. 2002. Communities pore spaces in the soil or the methods of agriculture. ◗ References cited and Ecosystems: Linking the underside of decomposing Aboveground and Belowground Neher, D. A. 1999. Soil community Components. Princeton leaves adds to my amaze- ACR is extremely fortunate to composition and ecosystem University Press, Monographs in ment at the intricacy of life have a crew of committed, and processes: comparing agricultural Population Biology. on this planet. And, along possibly slightly crazy, volun- ecosystems with natural ecosystems. Agroforestry Systems 45: with the sheer aesthetic teers who share this sense of 159-185. pleasure of viewing these respect and adventure and page 10 the ARDEID 2004

Influences on the hydro-geomorphology of Livermore Marsh The Power of Rainfall by Katie Etienne

he North Pacific Coast Railroad marsh features, constructed a levee along the such as channel TTomales Bay shoreline in the 1870s dimensions and that modified the hydrology of east-shore tidal exchange, marshes. Although wooden trestles were that are often built across major tide channels, water used to design circulation was altered and sediment marsh restora- began to accumulate behind the levees. tion projects. We Livermore Marsh at the Cypress Grove measured the Research Center is one of these wetlands same topograph- that exhibit some interesting physical and ic features to biotic characteristics. Before Audubon document Canyon Ranch (ACR) purchased the prop- changes during erty in 1971, previous owners installed the transforma- floodgates in the levee so they could use tion of Livermore the 26-acre wetland for grazing and fresh- Marsh from a water storage. In 1982, the levee was freshwater to a eroded by El Niño flood waters. ACR tidal system. decided to repair the levee and construct This study a spillway at an elevation that would pre- included month- vent tidewater from entering Livermore ly measurements Marsh. ACR excavated four ponds in the of the tidal inlet, lower marsh to maintain coveted fresh- annual surveys water habitat during dry seasons. of developing However, when the levee breached by tide channels, flood waters again in 1998, ACR decided and two detailed Figure 1. Elevations associated with tidal inundation at Livermore Marsh, in 2003. to allow the wetland to be restored natu- topographic sur- rally to a tidal marsh system. veys of the lower In 1999, grants from the Frank A. marsh in 1999 and 2003 (see photo on stable reference for depth measurements Campini Foundation and the Marin back cover). Hydrologist Lauren at 47 locations across the 96-ft bridge. Community Foundation allowed ACR to Hammack used these data to estimate One of the interesting questions we are restore trail access across the levee breach four characteristics of tidal marsh sys- exploring is whether different types of and to conduct a five-year study to docu- tems: tidal inlet area, active tide channel tidal patterns have predictable effects on ment physical and biotic changes associ- length, active tide channel volume, and the cross-sectional area of the tidal inlet. ated with the reintroduction of tidal cir- tidal prism volume (Table 1, page 12). The “Neap” tides exhibit the minimum range culation. In recent issues of The Ardeid, tidal prism estimate represents the vol- between high and low tides, while we presented reports summarizing our ume of tide water that flows through the “spring” tides exhibit the largest range research objectives (1999), watershed levee below the elevation of the mean between highs and lows. “Median” tides effects (2001) and changes in bird use higher-high water (3 ft NGVD29, or 5.4 ft range between mean-higher-high water (2003). This article examines principal NOS based on NOAA predictions). (MHHW) and mean-lower-low-water factors influencing the shape (topogra- Tidal versus fluvial effects (MLLW) and occur for several days phy) of the lower marsh (Figure 1). between the neap and spring tide cycles. idal processes and terrestrial river The physical characteristics of tidal I measured the inlet each month dur- flow are important factors that marshes have been studied extensively ing the last day of a median tide cycle and should be evaluated before apply- around the world, and some of the most T scheduled neap and spring measure- ing scientific models to particular sites. important research on the morphology of ments at least three times a year. We were This is a complex and demanding chal- diked salt marshes has been conducted in surprised to find that tidal action did not lenge, but at Livermore Marsh, we had the San Francisco Bay area (Coats et significantly influence either the cross- the advantage of being able to routinely al.1989; Coats and Williams 1990, and sectional area of the inlet or the rate of and accurately monitor changes in the Williams and Orr 2002). These engineers change in the inlet area. inlet area. We used a laser level mounted have identified correlations between key From our analysis, it appears that at either end of the bridge to provide a cumulative rainfall was the primary factor 2004 the ARDEID page 11 KATIE ETIENNE KATIE Figure 4. Six years after the levee breach, the tidal prism increased rapidly after the tide channel finally eroded the sill of the perched pond. The increased tidal circulation and winter flooding contributed to a Figure 2. The tidal inlet area of Livermore Marsh Figure 3. The tidal inlet area of Livermore Marsh is 28% increase in inlet area between 18 Dec 2003 increased significantly with cumulative rainfall (P < approaching predicted values based on studies of and 7 Jan 2004. 0.001). mature levee marshes in San Francisco Bay (square) and Tomales Bay (triangle). influencing changes in the inlet area well as an increase in tidal inlet area from tidal prism estimates in 1998 (Figure 2). In addition, the effects of rain- (Table 1). These episodic changes in through 2001. fall were not significantly influenced by channel development are probably asso- The head cut of the primary tide chan- tide conditions. Figure 3 illustrates the ciated with rainfall events and the cohe- nel progressed very slowly through the high variability in the size of the tidal inlet sive nature of the sediment, which tends middle of the marsh, compared to chan- during the first five years, although the to erode in blocks. nel banks that developed in the perimeter inlet area appears to be stabilizing around In 1999, Gian-Marco Pizzo identified of the marsh. The rate of erosion in the the values predicted from mature marsh the significance of the dense substrate, middle marsh was reduced by a remnant systems (see The Ardeid 2002). which is not easily eroded by runoff or stand of tules, because their dense stems The role of sediment tidal action. His calculation of tidal veloc- decreased water velocity and their rhi- ity in the primary channel (0.2 m/sec) zomes bound the sediment together. e calculated the volume of pri- was far less than the 1 m/sec normally Another factor was probably the percola- mary and secondary tide chan- required to cause erosion. Pizzo conclud- tion of water from the perched pond, Wnels, as well as incipient chan- ed that channel development in Liver- which maintained soil saturation in the nels that may eventually become tertiary more Marsh was primarily influenced by middle of the marsh. Saturation with channels or blind sloughs. During the first the propagation of vertical head cuts that water prevented cracking of the substrate year after the levee breach, the primary produce the stair-step shape in develop- that allows blocks of sediment to tip and channel increased in length by 128 ft. The ing channels. During the next five years, fall downstream. primary tide channel did not lengthen we continued to document the lateral Meanwhile, channel banks around the between 1999 and 2000, but there was a movement of numerous head cuts and perimeter of the marsh continued to col- small increase in channel volume as the slumping banks. Over time, head cuts and lapse into the channel, and sediment was channel became deeper and wider. This channel banks became taller and the slowly transported through the system. pattern of channel lengthening followed music of falling water became louder as Finally, after several weeks of heavy rain by widening is consistent with other turbulent flow scoured material away and the spring tide cycle of 6-8 Jan 2004, developing marshes where changes in from the base of new waterfalls. the sill between the crescent pond and width are also more rapid than changes in Sporadic changes in tidal prism the head cut eroded, and water previously depth (Coats et al. 1995). stored in the large pond was released Although there was an increase in rior to analysis, we partitioned the (Figure 4). As rain continued to fall, high channel length and volume in 2001 that topographic survey data to distin- tides began to circulate through the cres- contributed to the increase in tidal prism, Pguish between active tide channels cent pond, which increased the tidal the tidal inlet continued to fill in (Table and isolated or “perched ponds” that did prism by at least 0.36 acre-ft. This esti- 1). This inverse relationship between not drain during diurnal tidal cycles mated increase does not include the addi- changes in the tidal prism and the inlet is (Figure1). The small circular pond closest tional tidal volume resulting from the ero- different from the direct relationship in to the levee is one of four ponds con- sion of unconsolidated material in the mature marsh systems, where increased structed in 1983, when ACR was manag- pond by winter runoff and high tides. inlet area corresponds to increased tidal ing the system as a freshwater marsh. The This process of slow erosion and tide prism volume (Coats et al. 1995). larger, crescent-shaped pond is a remnant channel development indicates the Subsequent channel surveys indicated of a pre-existing channel that was importance of substrate composition. that although most channels were becom- mapped by the U.S. Coast Survey in Another major factor is the elevation of ing wider and longer, the total channel the1862 (see The Ardeid, 1999). Because the marsh plain. Survey data show that volume and tidal prism decreased in 2002. water level in these and other small most tide water never flows above the However in 2003, we measured a large ponds in the marsh did not rise and fall deepest portions of the developing chan- increase in channel volume and length as with the tides, we excluded their volumes page 12 the ARDEID 2004

Table 1. Changes in the geomorphic characteristics of Livermore Marsh (1998-2004) show a continuing The dominant factors influencing the increase in channel length and tidal prism volume, but fluctuations in channel volume and inlet area early development of Livermore Marsh (2002-2003). were fluvial action, sediment characteristics, and marsh plain elevation. We antici-

1 2 pate the effect of tidal action on physical Year Inlet area Tide channel Active tide channel Tidal prism and biological processes in the marsh will (ft2) length (ft) (acre-ft) (acre-ft) depend upon future rates of sediment transport and sea level rise. ◗ 1998 283 230 0.32 1.4 1999 176 420 0.40 2.3 2000 94 420 0.44 2.6 2001 62 560 0.52 2.8 2002 32 712 0.46 2.4 References cited 2003 95 770 0.56 2.7 Coats R.N., M. Swanson, and P. B. Williams. 1989. 2004 127 1170 0.92 3.1 Hydrologic analysis for coastal wetland restoration. Environmental Management 13(6):715- 1 below 3 ft NGVD29 727. 2 below 1.5 ft NGVD29 Coats, R.N., and P. Williams. 1990. Hydrologic techniques for coastal wetland restoration illus- trated by two case studies. P. 236-246 in J.J. Berger editor, Environmental Restoration: sci- nels, and only extreme high tides have The isolated pond near the levee exceed- ence and strategies for restoring the earth. any effect on the surface of the marsh ed the salinity of Tomales Bay during Island Press, Washington, D.C. Coats, R.N., P.B. Williams, C.K. Cuffe, J.B. Zedler, D. plain. This helps explain why salinity in summer months. These salinity differ- Reed, S.M. Waltry, and J.S. Noller. 1995. Design most areas with restricted tidal exchange ences reflect the importance of marsh guidelines for tidal channels in coastal wet- tends to remain fresh or slightly brack- plain elevation, which continues to lands. Philip Williams and Associates, San ish. From 1998-2003, salinity measure- restrict tidal influence. Francisco, CA. Williams, P. B. and M.K. Orr. 2002. Physical evolu- ments in the crescent-shaped pond tion of restored breached levee salt marshes in remained near 1 ppt, and most of the the San Francisco Bay Estuary. Restoration tide channel was brackish (4.0–9.6 ppt). Ecology 10(3):527-542.

Visting investigators

Effects of invasive species on nitrogen reten- Community based assessment of biological A comparison of carbon cycling and material tion and other issues in the ecology and health of riparian wetlands in the Sonoma exchange in grasslands dominated by restoration of coastal prairie. Jeff Corbin Creek watershed. Caitlin Cornwall, native and exotic grasses in Northern and Carla D’Antonio, UC Berkeley. Sonoma Ecology Center. California. Laurie Koteen, UC Berkeley. Carnivore use of riparian corridors in vine- Water quality monitoring at several points in Black Brant counts at Drakes Estero, Tomales yards. Jodi Hilty, UC Berkeley. Tomales Bay, including ACR’s Walker Bay and Bodega Bay. Rod Hug, Santa Creek delta. Lorraine Parsons, Point Rosa, CA. Multi-scaled vegetation data to predict Reyes National Seashore. wildlife species distributions using a wildlife Monitoring Avian Productivity and Survivor- habitat relationship model. Jennifer Bird communities in North Coast oak-vine- ship (MAPS) station at Livermore Marsh. Shulzitski, USGS Golden Gate Field yard landscapes. Emily Heaton, UC Denise Jones, Institute for Bird Station. Berkeley. Populations.

Impact of butterfly gardens on pipevine swal- Factors causing summer mortality in Pacific Tidewater goby inventory for Tomales Bay. lowtail populations. Jacqueline Levy, San oysters. Fred Griffin, Bodega Marine Darren Fong, GGNRA. Francisco State University. Lab. Conservation concerns for Cordylanthus The effect of landscape changes on native Effects of sudden oak death-induced habitat maritimus ssp. palustris. Todd Wilms, bee fauna and pollination of native plants change on vertebrate communities. Kyle Emeryville, CA. in Napa and Sonoma counties. Gretchen Apigian and Don Dahlson, UC Berkeley. LeBuhn, San Francisco State University. Strophariaceae of California. Peter Werner The differential invisibility of annual and and Dennis Desjardin, San Francisco Ecological indicators in West Coast estuaries. perennial grasslands in California by a State University. Steven Morgan, UC Davis Bodega non-indigenous invader, Foeniculum vul- Marine Lab; Susan Anderson, UC Santa gare. Joel Abraham and Jeff Corbin, UC Barbara; and numerous collaborators. Berkeley.

Consequences of species invasion under Differences in arthropod community composi- global climate change. Elizabeth Brusati, tion in native and exotic dominated grass- UC Davis. lands. Natalie Robinson, UC Berkeley. 2004 the ARDEID page 13

Tomales Bay Shorebirds ◗ Ehrharta erecta Vernal wetland botanical In progress: Since 1989, we have conducted management and surveys at Bouverie project updates annual baywide shorebird research ◗ Erharta erecta is a Preserve ◗ As part of our censuses on Tomales Bay. highly invasive perennial grass overall effort to determine the North Bay counties heron ◗ Censuses involve six baywide native to South Africa. It is ecological values of vernal and egret project Annual winter counts and one currently invading west Marin wetlands at Bouverie Preserve, monitoring of reproductive baywide count each in August County and is abundant in botanist Ramona Robison activities at all known heron and April migration periods. A ACR’s Pike County Gulch. The conducted floristic surveys and egret colonies in five team of 15-20 volunteer field goals of this project are to designed to target rare plants. northern Bay Area counties observers are needed to understand the effects of The surveys provide plant began in 1990. The data are conduct each count. The data Ehrharta invasion, develop species lists, vegetation cover used to examine regional are used to investigate winter tools for control of Erharta, and values for each species, and patterns of reproductive population patterns of shore- restore habitat invaded by exact (GPS) delineation of performance, disturbance, birds, local habitat values, and Erharta at Bolinas Lagoon wetland habitats and locations habitat use, seasonal timing conservation implications. Preserve. of rare plants. and spatial relationships among heronries. The project has Tomales Bay waterbird Olema Marsh bird census Salmarsh ice plant been recently incorporated into survey ◗ Since 1989-90, ◗ Although considerable bird removal ◗ Non-native ice the Integrated Regional teams of 12-15 observers have census work was conducted at plant is being removed from Wetland Monitoring (IRWM) conducted winter waterbird ACR’s Olema Marsh in the marshes and upland edges at program, a pilot project to censuses from survey boats 1980s and early 1990s, recent Toms Point on Tomales Bay, develop regional monitoring for on Tomales Bay. The results information on bird use is using manual removal, shading San Francisco Bay. We are provide information on habitat needed to include Olema with black plastic, and glypho- currently preparing an values and conservation needs Marsh in the Point Reyes sate. The goals are to eliminate annotated atlas of regional of 51 species, totaling up to National Seashore Giacomini invasive ice plant from Toms heronries. 25,000 birds. Future work will Wetland Restoration Project. Point and to collaborate with focus on trends and determi- Methods involve point counts other land managers to Picher Canyon heron/ ◗ nants of waterbird variation on and distance sampling, remove ice plant from other egret project The fates of Tomales Bay. conducted by Rich Stallcup. sites in Tomales Bay. all nesting attempts at ACR’s Picher Canyon heronry are Predation by ravens in Plant species inventory ◗ Eradication of Elytrigia monitored annually, based on heron/egret colonies ◗ We Resident biologists maintain pontica spp. pontica ◗ procedures initiated by Helen have been observing ravens in inventories of plant species Elytrigia is an invasive, non- Pratt in 1967, to track long- Marin County and measuring known to occur along the native perennial grass that term variation in nesting raven predatory behaviors at Tomales Bay shoreline, and on forms dense populations with behavior and reproduction. heron and egret nesting ACR’s at Bouverie and Bolinas nearly 100% cover in seasonal colonies. The field work Lagoon preserves. wetland sites. We are using Livermore Marsh ◗ As involves radio telemetry and Cape ivy control, manual removal by groups of ACR’s Livermore Marsh, on behavioral observations. We volunteers, light starvation and Volunteer Canyon ◗ Work Tomales Bay, transforms from have produced scientific solarization using black plastic conducted by Len Blumin has a freshwater system into a papers on the status of ravens tarps, and glyphosate spot proven that manual removal of tidal salt marsh, we are and crows in the San Francisco treatments. nonnative cape ivy can studying the relationship Bay area, patterns of home successfully restore riparian Eucalyptus removal ◗ between increasing tidal prism range use, and raven predatory vegetation. Continued vigilance Eucalyptus trees are being and marsh channel develop- behaviors in heronries. ment. Monitoring of winter and in weeded areas of ACR's removed, with incremental breeding bird use began in Experimental assessment Volunteer Canyon has been annual cutting, from ACR’s 1985. The data will be linked to of Wild Turkey impacts ◗ important, to combat resprouts Bolinas Lagoon Preserve, and measurements of vegetation Invasive Wild Turkeys are of black nightshade, Vinca, and along the Highway 12 border to reveal changes associated common at Bouverie Preserve Japanese hedge parsely. of Bouverie Preserve. Dan with the developing tidal and throughout most of Annual surveys and Gluesenkamp is conducting an marsh. We also monitor the Sonoma County. Dan removal of non-native experiment to determine the Gluesenkamp is measuring the optimal method for controlling depth and duration of ground cordgrass ◗ Protection of effects of ground foraging by Eucalyptus resprouts. water which strongly influ- ACR’s shoreline properties Wild Turkeys on vegetation, ences biological conditions in from invasion by nonnative Wood Duck boxes ◗ Rich invertebrates, and herpeto- the upper marsh. Spartina species is critical to Stallcup has installed and fauna in the forest ecosystem ◗ the protection of ACR lands maintains several Wood Duck Newt population study of Bouverie Preserve. The and provides a critical nest boxes along Bear Valley Annual newt surveys have results will and provide contribution to the overall Creek in ACR’s Olema Marsh. been conducted along the information that can be used to monitoring and management Stuart Creek trail at Bouverie improve management and Bluebird boxes ◗ Tony of Tomales Bay and Bolinas Preserve since 1987. The control of turkey populations Gilbert has installed four Lagoon. In addition to conduct- results track annual and by agencies. bluebird boxes in the Cypress ing surveys on ACR lands, intraseasonal abundance, and Grove grasslands with the Katie Etienne is collaborating size/age and spatial distribu- objective of providing nest on surveys of other shoreline tions along the creek. sites for two pairs of Western properties in these estuaries. Bluebirds. 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. ©2004 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

Pacific Land Surveys measured the topography of ACR's Livermore Marsh during a five-year study of ecological changes associated with the reintroduction of tidal circulation.

Livermore Marsh see page 9 KATIE ETIENNE KATIE

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