Avian communities in baylands and artificial salt evaporation ponds of the San Francisco Bay estuary

Transcription

1 Hydrobiologia 466: , J.M. Melack, R. Jellison & D.B. Herbst (eds), Saline Lakes Kluwer Academic Publishers. Printed in the Netherlands. 317 Avian communities in baylands and artificial salt evaporation ponds of the San Francisco Bay estuary John Y. Takekawa 1, Corinna T. Lu 1,2 &RuthT.Pratt 3,4 1 U.S. Geological Survey, Western Ecological Research Center, San Francisco Bay Estuary Field Station, P.O. Box 2012, Vallejo, CA 94592, U.S.A. Tel: 707/ Fax:707/ john_takekawa@usgs.gov 2 U.S. Geological Survey, California Cooperative Research Unit, Humboldt State University, Arcata, CA 95521, U.S.A. 3 U.S. Fish and Wildlife Service, Ecological Services, Endangered Species Branch, P.O. Box 2012, Vallejo, CA 94592, U.S.A. 4 Present address: 307 North Street, Sausalito, CA 94965, U.S.A. Key words: salt evaporation ponds, waterbirds, San Francisco Bay, migratory birds, baylands, salt ponds Abstract San Francisco Bay wetlands, seasonal and tidal marshes between the historic low and high tide lines, are now highly fragmented because of development during the past 150 years. Artificial salt pond systems in the Bay are hypersaline and typically support simple assemblages of algae and invertebrates. In order to establish the value of salt ponds for migratory waterbirds, we used datasets to conduct a meta-analysis of avian communities in the baylands and salt ponds of San Pablo Bay. Fifty-three species of waterbirds in the salt ponds represented six foraging guilds: surface feeders, shallow probers, deep probers, dabblers, diving benthivores and piscivores. The total number of species and the Shannon-Weiner diversity index was higher in baylands than in salt ponds during all four seasons. However, overall bird density (number/ha) was higher in salt ponds compared with baylands in the winter and spring, primarily because of large concentrations of benthivores. Cessation of salt production in 1993 and subsequent reduction in water depth resulted in a decline of some diving duck populations that used the salt ponds. Introduction The San Francisco Bay estuary is one of the most important staging and wintering areas for migratory waterfowl and shorebirds in the Pacific Flyway (Harvey et al., 1992). San Francisco Bay is recognized as a site of hemispheric importance because it supports at least 30% of some individual shorebird populations in the flyway (Harrington & Perry, 1995) and up to half of many diving duck populations (Accurso, 1992). Many migratory waterbirds use the baylands, which consists of the area between the historic high and low tide lines and comprises ha in the estuary (Goals Project, 1999). More than 90% of the historic wetlands in the baylands have been lost to urban development, agriculture, and salt production (Josselyn, 1983; Nichols et al., 1986), and many tidal marsh species are now threatened or endangered. The remaining bayland wetlands are isolated from tidal action by the construction of levees or water control structures and comprise a fragmented landscape of non-tidal salt, brackish and fresh water wetlands, agricultural lands, seasonal ponds, vernal pools, riparian scrub and artificial salt evaporation ponds (Goals Project, 1999). Historically, salt pans were a common wetland type in the baylands (Goals Project, 1999). Salt pans and seasonal wetlands likely provided foraging habitat and invertebrate prey for migratory waterbirds, as well as protected roosting areas during high tides or winter storms. All of the salt pans and most seasonal wetlands were lost to development during the past century, but salt evaporation pond systems developed for commer-

2 318 cial manufacturing of salt (Ver Planck, 1958) replaced some of the habitat values formerly provided by these wetlands (Harvey et al., 1988). Artificial salt ponds have existed in the estuary for more than 150 years (Ver Planck, 1958), and these habitats have become an integral part of the landscape as well as critical habitats for a large proportion of the waterbirds in the estuary both during the winter and migration periods (Anderson, 1970; Swarth et al., 1982; Accurso, 1992). These artificial hypersaline ponds vary seasonally in salt content from brackish to saturated, range from a few centimeters to a few meters in depth, and are composed of simple but productive assemblages of algae and invertebrates (Carpelan, 1957; Lonzarich, 1988; Lonzarich & Smith, 1997). Some conservation organizations have proposed converting the salt ponds into tidal wetlands to aid in reversing the decline of tidal wetland species and to mitigate for development. Under such a scenario, only a few hundred hectares of the more than ten thousand hectares of salt ponds in the estuary will likely remain through the next century (Goals Project, 1999). Although Anderson (1970) has compared waterbird use of salt ponds and tidal mud flats in the estuary, no studies have compared the value of salt ponds with the mix of bayland wetlands that will likely replace salt ponds following conversion. We conducted a meta-analysis from existing datasets to examine seasonal differences in populations of migratory waterbirds in salt ponds compared with other bayland wetlands (hereafter baylands) in San Pablo Bay, part of the northern San Francisco Bay estuary. We compared total species, density and diversity in baylands and salt ponds and grouped species by feeding behaviors to examine functional differences by foraging guilds. Finally, we examined how waterfowl changed their use of salt ponds in the North Bay during the past decade following cessation of commercial salt production. Study area We examined baylands and salt ponds in the North Bay subregion (37.90 N N; W W) of the San Francisco estuary (Fig. 1). These included 77 major seasonal wetlands in the baylands identified on U.S. Fish and Wildlife Service National Wetland Inventory (NWI) in Alameda, Marin, Napa, Solano, or Sonoma counties (R. Pratt, unpubl. data). Bayland habitats excluded artificial salt ponds but included both natural and diked wetlands including non-tidal salt, brackish and fresh water marshes, agricultural lands, seasonal ponds, vernal pools and riparian scrub. The majority of these wetlands were seasonally inundated. Use of wetlands in the baylands by waterbirds was compared with use in salt ponds located 5 km northwest of Vallejo, California (38 10 W, N), comprising about 4000 ha. The salt ponds were acquired in 1994 by the California Department of Fish and Game for inclusion as part of the Napa-Sonoma Marshes State Wildlife Area. Although salt production ceased in 1993, the hypersaline system remained intact with lower salinities found in primary ponds with muted tidal flow from the North Bay and higher salinity in ponds found farther inland (Miles et al., 2000). Methods Bird surveys We used the only existing datasets from three bird surveys in the North Bay subregion which met our criteria of including counts by species of several bayland wetlands or salt pond areas of known size with replication over time. Our datasets included bird surveys of bayland wetlands from the U.S. Fish and Wildlife Service Diked Baylands Wildlife Study conducted from 1982 to 1989 (Pratt, 1997), salt pond surveys from the U.S. Geological Survey Napa-Sonoma Salt Ponds project initiated in 1999, and the U.S. Fish and Wildlife Service Midwinter Waterfowl Surveys conducted in January between 1988 and Although a few restoration projects were initiated during this period, we assumed that there was little change in the overall avian community of the baylands. Diked baylands wildlife study The Diked Baylands Wildlife Study (R. Pratt, unpubl. data) was conducted from 1982 to 1989 to determine wildlife habitat values of diked, seasonal wetlands in the San Francisco Bay region. Both ground and aerial surveys were completed, although the majority of surveys were conducted by ground observers. Volunteers counted the number of birds by species at selected wetland sites up to 2-times each month during the wet season (November through April) and once a month, depending on availability of observers, during the dry season (May through October). Most of the

3 Figure 1. Baylands and salt ponds in the northern San Francisco Bay estuary. Diked baylands are indicated in gray, while salt ponds are denoted by diagonal stripes. 319

4 320 volunteers were experienced in bird identification and were members of local chapters of the National Audubon Society. On larger wetlands, the proportion of a wetland that was counted varied from 5% to 100% of the total area depending on visibility and landowner access. We limited our analyses to waterbird species, primarily shorebirds and waterfowl, which were readily counted in surveys of the North Bay subregion in 74 of the 77 wetlands surveyed where the size of the survey area was known. Napa-Sonoma salt ponds We initiated monthly complete counts of 5 of the 11 salt ponds (Ponds 1, 2, 3, 4, 7) within the Napa- Sonoma Marshes selected from a range of salinities in January 1999 (Miles et al., 2000). The ponds ranged in size from 127 to 534 ha, and salinities varied from 22 g l 1 to 225 g l 1, similar to the range found in low and mid salinity ponds in commercial salt production systems. Observers conducted counts of species with binoculars and spotting scopes from vantage points at the edge of ponds during the first week of each month from January through December, and locations of waterbirds were placed in a geographic information system coverage into 250 m 250 m grid squares (Miles et al., 2000). Surveys were conducted during the day within 3 h of the highest high tide when the largest number of waterbirds were roosting in the salt ponds. Foraging guilds We separated the waterbirds into 6 guilds to examine differences among foraging groups rather than differences among individual species. These foraging guilds included: (1) surface feeders obtained prey from the surface such as American avocet (Recurvirostra americana), black-necked stilt (Himantopus mexicanus) and red-necked phalarope (Phalaropus lobatus); (2) shallow probers foraged in the top layer (<3 cm) of sediments such as western sandpiper (Calidris mauri) and dunlin (Calidris alpina); (3) deep probers reached several centimeters into the substratum including marbled godwit (Limosa fedoa) and long-billed dowitcher (Limnodromus scolopaceus), (4) dabblers fed in the upper water column such as northern pintail (Anas acuta) and northern shoveler (Anas clypeata); (5) diving benthivores fed in deeper water on benthic invertebrates including lesser and greater scaup (Aythya marila, Aythya affinis), ruddy duck (Oxyura jamaicensis) and eared grebe (Podiceps nigricollis); and (6) piscivores fish consumers such as American white pelican (Pelecanus erythrorhynchos), Caspian and Forster s terns (Sterna caspia, Sterna forsteri), and double-crested cormorant (Phalacrocorax auritus), and (7) other incidental species including Canada geese (Branta canadensis) and ring-billed gulls (Larus delawarensis). Most of the major waterbird species that use salt ponds are not endangered, but some species seen infrequently in the North Bay such as western snowy plover (Charadrius alexandrinusnivosa) and least terns (Sterna antillarun browni) are federal or state listed. Analyses The total number of waterbirds counted at a particular site varied widely depending on the number of surveys and water levels, especially in bayland wetlands where few or no birds were counted if the wetlands were dry. Thus, we used maximum daily counts and peak densities (birds ha 1 ) of species found at each site to examine their greatest importance for waterbirds during each season. We examined differences by wetland type (bayland or salt pond) among foraging guilds during the winter (Nov Jan), spring (Feb Mar), summer (Apr Aug) and autumn (Sep Oct) seasons. Months were assigned to seasons to encompass the major migration chronology in spring and autumn. Monthly or semi-monthly surveys across years were treated as repeated measures, species were combined into guilds, and means were derived for each site. We examined differences in densities of the foraging guilds by wetland type, season, and the type-by-season interaction with two-way multivariate analysis of variance (MANOVA) techniques (Johnson & Wichern, 1988; SAS Institute, 1990). We conducted univariate analysis of variance tests for unequal sample sizes for each guild when significant multivariate results were detected, followed by pairwise comparison to locate differences among wetland types or seasons with Fisher s protected Least Squares Difference Test (Milliken & Johnson, 1984). Overall species diversity for baylands and salt ponds was assessed with the Shannon Weiner index (Krebs, 1989). Results were deemed significant when P Midwinter waterfowl survey The U.S. Fish and Wildlife Service midwinter aerial waterfowl survey indices were used to examine changes in waterfowl populations during the last 10 years as salt production ended in the North Bay ponds.

5 321 Table 1. Total number of waterbird species documented in the baylands and salt ponds of the North Bay subregion of the San Francisco Bay estuary Season Baylands Salt Ponds Winter Spring Summer Fall Overall The midwinter waterfowl index, a complete count of waterfowl species along pre-selected routes of major use areas in early January, was conducted in the estuary by two observers in a twin-engine Partenavia aircraft flying at 100 km h 1 at 90 m elevation (see Accurso, 1992). Data were compiled separately for each identifiable wetland unit in the North Bay subregion, including counts for each low and mid salinity salt pond in the Napa Sonoma salt ponds. Trends were determined with a general linear regression model. Results Density differences We recorded 82 species of waterbirds in the diked baylands and 53 in the salt ponds (Table 1). The number of species representing the common foraging guilds (Table 2) included surface feeders (3), shallow probers (8), deep probers (6), diving benthivores (10), dabblers (8), piscivores (10) and other (8) incidental species which were not analyzed further. Overall densities were higher in salt ponds compared with baylands in all seasons (Table 3). In both areas, waterbird numbers were highest in the winter and spring. For some guilds such as surface feeders, densities were consistently greater in the salt ponds (Table 4). However, the proportion of guilds in the overall composition (Fig. 2) of both the baylands and salt ponds was variable among the seasons. We found a significant relationship between the density of foraging guilds by wetland type and season (Wilks λ = 0.88, F 18,682 = 1.74, P = 0.03). The density of surface feeders was significantly higher in salt ponds than in baylands (F 1,246 = 39.8, P < ). Shallow probers had similar densities in salt ponds and in baylands. However, their densities differed by season (F 3,246 = 5.05, P = ) with largest populations in the winter and spring, and lowest numbers in the summer and fall. The density of deep probers was higher in baylands in the fall, winter, and spring, but more were found in salt ponds in the summer. However, differences in densities by season or wetland type were not significant (F 7,246 = 1.64, P = 0.13). Dabblers in baylands had consistently greater densities than in salt ponds, and no dabblers were found in the salt ponds during the fall or summer. We found a significant difference in density of dabblers by season (F 3,246 = 6.60, P = ), with densities in the spring and winter greater than the summer and fall. We found a significant interaction (F 3,246 = 5.95, P = ) between seasons and wetland types used by benthivores. Benthivores dominated the salt ponds in the winter and spring in comparison to the baylands, but were rarely present in either area during the summer and fall. Piscivores were more abundant in baylands than salt ponds in all seasons. However, we found no significant differences between seasons or wetland types (F 3,246 = 0.27, P = 0.60) because of the small and variable densities of piscivores. Species diversity The total number of species present during all seasons was greater in the baylands (Table 1). We found 50% more bird species in the baylands overall. The lowest number of species was observed in the salt ponds during the summer and fall. The number of species varied less between seasons in the baylands compared with the salt ponds. We also examined the diversity of the two systems with the Shannon Weiner diversity index (Krebs, 1989) to weigh the species occurrence by abundance (Table 5). The diversity indices for both habitat types were similar despite more species in baylands, because most of the species unique to baylands were not abundant. Overall diversity was higher in the baylands than salt ponds, and the diversity index was similar throughout the year. In contrast, the diversity in the salt ponds was highest in the winter and summer, but less diverse during migration periods. Waterfowl trends The midwinter waterfowl index for the salt ponds varied from 6500 to ducks during the past decade. Although we found no overall trend (r 2 = 0.09, F 1,7 =

6 322 Figure 2. Seasonal comparison of the mean density per hectare of waterbirds by guild for the seasonal wetlands of the northern diked baylands and the salt ponds 1, 2, 3, 4 and 7 in the Napa Sonoma Marshes State Wildlife Area, California. 0.06, P = 0.82) for total waterfowl (Fig. 3a), there were changes by foraging guilds. Dabbling duck numbers (Fig. 3b) increased (r 2 = 0.68, F 1,7 = 6.06, P = 0.04), and while the decreasing trend in benthivores was not definite (r 2 = 0.62, F 1,7 = 3.75, P = 0.10), they declined from in 1989 to 3600 in 1997 (Fig. 3c). The benthivore species with the largest decrease in numbers (r 2 = 0.48, F 1,7 = 6.36, P = 0.04) was the canvasback (Fig. 3d). The number of canvasback counted in midwinter surveys decreased nearly 400% from 8000 to 2000 individuals in the past decade. Discussion Value of salt ponds for waterbirds We found that a large number of waterbird species both roosted and foraged in the salt ponds in the North Bay, especially during the winter and spring. While salt ponds supported fewer species than baylands, salt ponds provided habitat for up to twice the waterbird density. Salt ponds provided important habitat for surface feeders, shallow probers, and diving benthivores. Alternative habitats such as salt ponds are particularly important for feeding and roosting shorebirds, because mud flats in the North Bay were unavailable several hours twice each day when inundated by high tides. Timm (1997) found highest densities of waterbirds on

7 323 Table 2. Waterbird species of the major foraging guilds in the baylands and salt ponds of the North Bay subregion, San Francisco Bay estuary. Species unique to baylands surveys ( B )are indicated Table 2. Continued Guild Common name Latin name Guild Common name Surface Feeders American avocet black-necked stilt red-necked phalarope Wilson s phalarope B Shallow Probers black-bellied plover dunlin killdeer red knot semipalmated plover snowy plover spotted sandpiper B least sandpiper western sandpiper Deep Probers Lesser B, greater yellowlegs long-billed curlew long-billed, short-billed dowitcher marbled godwit whimbrel willet Diving Benthivores Barrow s goldeneye B bufflehead canvasback common goldeneye Clark s grebe eared grebe horned grebe lesser, greater scaup pied-billed grebe redhead B red-necked grebe B ring-necked duck B ruddy duck surf scoter B western grebe Dabblers American coot American wigeon blue-winged teal B Latin name Recurvirostra americana Himantopus mexicanus Phalaropus lobatus Phalaropus tricolor Pluvialis squatarola Calidris alpina Charadrius vociferous Calidris canutus Charadrius semipalmatus Charadrius alexandrinus Actitis macularia Calidris minutilla Calidris mauri Tringa flavipes, melanoleuca Numenius americanus Limnodromus scolopaceus, griseus Limosa fedoa Numenius phaeopus Catoptrophorus semipalmatus Bucephala islandica Bucephala albeola Aythya valisineria Bucephala clangula Aechmophorus clarkii Podiceps nigricollis Podiceps auritus Aythya affini, marila Podilymbus podiceps Aythya americana Podicpes grisegena Aytha collaris Oxyura jamaicensis Melanitta perspicillata Aechmophorus occidentalis Fulica Americana Anas americana Anas discors cinnamon teal Eurasian wigeon B gadwall Dabblers green-winged teal mallard northern pintail northern shoveler Piscivores American white pelican black-crowned night heron Caspian tern common merganser double-crested cormorant Forster s tern great blue heron great egret green heron B hooded merganser B red-breasted merganser snowy egret Other American bittern B black rail B Bonaparte s gull California clapper rail B California gull Canada goose common loon common snipe B greater white-fronted goose B glaucous-winged gull B greater flamingo herring gull B mew gull B osprey ring-billed gull Ross goose B snow goose B sora B tundra swan B virginia rail B wandering tattler B western gull wood duck B Anas cyanoptera Anas penelope Anas strepera Anas crecca Anas platyrhynchos Anas acuta Anas clypeata Pelecanus erythrorhynchos Nycticorax nycticorax Sterna caspia Mergus merganser Phalacrocorax auritis Sterna forsteri Ardea herodias Casmerodius albus Butorides striatus Lophodytes cucullatus Mergus serrator Egretta thula Botaurus lentiginosus Laterallus jamaicensis Larus philadephia Rallus longirostris Larus californicus Branta canadensis Gavia immer Gallinago gallinago Anser albifrons Larus glaucescens Phoenicopterus ruber Larus argentatus Larus canus Pandion haliaetus Larus delawarensis Chen rossi Chen caerulescens Prozana carolina Cygnus columbianus Rallus limicola Heteroscelus incanus Larus occidentalis Aix sponsa

8 324 Table 3. Mean density (birds ha 1 )of waterbirds in the baylands and salt ponds of the North Bay subregion, San Francisco Bay estuary Season Baylands Salt Ponds Winter Spring Summer Fall Overall Table 5. Shannon Weiner diversity index for waterbirds in the baylands and salt ponds of the North Bay subregion, San Francisco Bay estuary Season Baylands Salt Ponds Winter Spring Summer Fall Overall Table 4. Mean density (birds ha 1 ) of waterbirds by foraging guilds in the wetlands of the baylands and salt ponds in the North Bay subregion, San Francisco Bay estuary Guild Season Baylands Salt Ponds Surface Feeders Winter Spring Summer Fall Shallow Probers Winter Spring Summer Fall Deep Probers Winter Spring Summer Fall Dabblers Winter Spring Summer Fall Diving Benthivores Winter Spring Summer Fall Piscivores Winter Spring Summer Fall saline wetlands compared with freshwater wetlands in Australia. Velasquez (1993) found similar results at the Berg River estuary in South Africa, where he observed highest densities of birds in supratidal salt pans compared with other wetlands. He found that the value of salt pans as foraging habitat for waterbirds depended on both the abundance of benthic macrofauna, primarily determined by salinity, and the availability of the macrofauna, which was related to water level. Velasquez (1993) estimated that some migratory species required the shallow salt ponds to satisfy their daily energy demands, and waterbirds reduced the availability of invertebrates 8 14 times more in salt pans than on mud flats. The datasets we analyzed were collected in different years for different lengths of time, but we were unaware of any major change in the waterbird community during this period. A survey of the San Francisco Bay estuary in the mid-1960s by Bollman et al. (1970) reported findings similar to results in our study with higher densities of birds in salt ponds compared with other wetlands. However, Bollman et al. (1970) suggested that other wetland types with more plant cover were not as well-counted as salt ponds because of visibility bias. We only used surveys where observers delineated the extent of the site that was counted to allow for direct comparison of waterbird densities between salt ponds and baylands. Thus, the affect of visibility bias should have been minimized in our analyses. In addition, the waterbird densities we reported for salt ponds were likely conservative, because prior to our surveys, water levels had begun to decrease in the salt ponds which may have reduced benthivore numbers. We found similar numbers of waterbird species in the North Bay salt ponds (53) as had been reported by Anderson (1970) for South Bay salt ponds (55). We were not surprised to find that the number of waterbird species (Table 1) and Shannon Weiner diversity indices (Table 5) were lower in salt ponds than in the surrounding baylands. However, we were

9 325 (Botaurus lentiginosus) and California clapper rail (Rallus longirostris) were found in baylands alone because they inhabited heavily vegetated areas. Diversity indices were similar among baylands and salt ponds because these species were rarely encountered. Observations of some of the rare species in the baylands were likely an artifact of the meta-analysis because: (1) bayland surveys conducted across several years provided more opportunity for sighting unusual species such as blue-winged teal (Anas discors); (2) observations were recorded by volunteer bird-watchers who typically concentrate on locating rare species; and (3) groupings differed from the salt pond surveys where the large flocks of some species were identified to genera. Differences by foraging guild Figure 3. Total waterfowl, dabbling ducks, diving benthivores and canvasbacks present in the Napa Sonoma Marshes States Wildlife Area from 1989 to The solid line indicates the population count while the dotted line presents linear trend of the data. surprised that diversity indices for baylands and salt ponds were not greatly different (Table 5) since 27 species were unique to the baylands (Table 2). Baylands represented several wetland types with a diversity of plant communities and corresponding wider habitat complexity used by different avian specialists. Some secretive species such as the American bittern Surface-feeding American avocets and black-necked stilts were abundant in the North Bay salt ponds. In salt ponds on San Diego Bay, Terp (1998) found that these two species used salt ponds more than expected compared to their use of mud flats. Harvey et al. (1988) reviewed older surveys in the estuary which suggested that both of these species were not present in the estuary in large numbers historically and may have actually increased in abundance since the introduction of salt ponds. Phalaropes (Phalaropus spp.) were present in large numbers in the South Bay salt ponds during migration (Anderson, 1970; Harvey et al., 1988; Goals Project, 1999), but phalaropes were rarely counted in the North Bay salt ponds. The presence of surface-feeding phalaropes in the South Bay may relate to the greater number of high salinity salt ponds that likely produced a more reliable supply of surface invertebrates including large populations of brine shrimp (Artemia salina) and brine flies (Ephydra cinerea: Anderson, 1970). The density of shallow probers was similar in both baylands and salt ponds, and the density of deep probers was higher in baylands in the winter and spring. However, we often observed larger total numbers of shorebirds in salt ponds, especially shallow prober species, because of the large amount of suitable habitat. Shorebirds primarily used areas of shallow water with limited vegetative cover, but these areas seemed to be less abundant in baylands. Some of the salt ponds were ringed with strips of shallow water (<10 cm) habitat along the edges or islands and levees inside the ponds which were heavily used. The size of

10 326 these habitat strips were highly variable depending on rainfall. Benthic invertebrates were abundant in salt ponds at low to mid salinities (Miles et al., 2000), but they were not always accessible to smaller shorebird species because water levels were often too deep. Terp (1998) found that although up to 30% of the western sandpipers and willets were feeding in salt ponds on San Diego Bay, few shorebirds were probing in the substratum. Instead, she found that shorebirds pecked in the water column to obtain invertebrates such as brine shrimp or speared prey on the ground or in the air. Western sandpipers were observed hovering to catch brine flies along the edge of the salt ponds in south San Francisco Bay (S. Warnock, pers. comm.). Dabblers were more numerous in baylands likely because they consume seeds of aquatic plants (Martin et al., 1951), although large numbers of northern shovelers were often seen feeding in the salt ponds. Waterbirds in the benthivore guilds had higher densities in salt ponds than in baylands. Benthivores were generally found in ponds with deeper water where they could dive to feed, thus their limited use of baylands was likely related to the lack of open water areas deep enough for foraging or large enough for diving birds to take flight. In contrast, piscivores were seen in the low salinity salt ponds where fish species diversity was 2.5 times greater and numbers were higher but varied seasonally with changes in salinity and depth through the year (Lonzarich & Smith, 1997). Piscivore densities were higher in bayland areas where salinities and prey fish abundance remain more constant. Seasonal variation The largest number of waterbirds were counted in both salt ponds and baylands during the winter. In comparison, the largest numbers of waterbird species were counted in the baylands in the winter, but overall densities were half those found in salt ponds. A larger number of species (Table 1) were found in the baylands in all seasons. We found that the density of benthivores was 4-times greater in the salt ponds compared with the baylands in the winter and spring. Salt ponds provided these species with large expanses of water available for taking flight, and the lack of disturbance on the ponds and the shallow, sheltered impoundments likely created a favorable microclimate for roosting. Davidson & Evans (1986) felt that it was critical to examine seasonal changes in use of man-made wetlands by shorebirds since these areas may be vital for survival of shorebirds during severe weather. We found that salt ponds were heavily used during migration, and populations of waterbirds were higher in the spring than in the fall, possibly because invertebrate populations increased in response to rains during the winter. Studies of western sandpipers (Warnock & Takekawa, 1996) confirmed that this species used salt ponds more during the spring. Shallow probers were found to be more dense in the spring, primarily because of the migration of large numbers of western and least sandpipers. San Francisco Bay has a Mediterranean climate with rains beginning late in the fall (Oct Nov). Thus, the salt ponds generally decrease in depth and increase in salinity from the summer through the fall, which may reduce the invertebrate biomass and foraging value for many waterbirds. In salt ponds, the largest densities of waterbirds were seen in the spring, with the next largest number of birds observed in the winter. Most of the migratory bird species were not present in the estuary during the summer, and we counted fewer birds in the fall compared with the spring. Waterfowl trends following cessation of salt production The midwinter waterfowl survey provided the only dataset where we could examine trends in waterbird populations as salt production ended in the North Bay salt ponds. These surveys indicated a decline in diving ducks beginning in 1993 (Fig. 3C) following the end of salt production. Diving ducks numbers declined because of reduced water levels in the ponds, while dabbling duck numbers increased in some ponds because their invertebrate prey were more available. The canvasback, an indicator species for the estuary (Goals Project, 1999), was the primary reason for the decrease in numbers of diving ducks. In 1990, the midwinter index for canvasbacks in the Pacific Flyway was about birds (U.S. Fish and Wildlife Service, unpubl. data). Sixteen thousand birds (29%) were in open water areas of the San Francisco Bay, and half of those were counted in salt ponds. In 1999, there were birds in the midwinter index, with (48%) of all canvasbacks in the Pacific Flyway counted in the estuary. However, only about 2500 canvasbacks (13%) were counted in the salt ponds. Thus, while the proportion of the Pacific Flyway canvasback

11 327 population wintering in the estuary increased, the use of salt ponds by canvasbacks decreased. Most likely, this was because of the decreased water depth in the North Bay salt ponds during the winter, making it less suitable for benthivores. Salt pond conservation and value for waterbirds In describing conservation of salt ponds in South Africa, Velasquez (1993) noted that the... concept of conserving only wetlands which are hydrologically pristine is almost utopian... Given the extensive urban development in the San Francisco Bay estuary (Nichols et al., 1986), recreating historic habitats may not be the best model for guiding current wetland conservation. Artificial salt ponds have existed in the estuary since the mid-1800s (Ver Planck, 1958) and the North Bay salt ponds are more than 40 years old. Salt ponds are now an integral part of this estuary and provide migration and wintering habitats for waterbirds, habitats critical for management of migratory birds (see Rappole & DeGraaf, 1996). Development of coastal zones has resulted in fewer areas available for migratory waterbirds in the flyway, and alternative wetlands may not exist outside of the San Francisco Bay estuary to compensate for loss of waterbird habitats in the ecosystem. Restoring or rehabilitating diked baylands in the estuary is likely beneficial for waterbirds most of the time. Historic wetland habitats that were converted to agriculture or urban development now have limited value for waterbird species, with the exception of areas inundated seasonally during the winter and spring (Haas, 1994). However, converting from one wetland habitat type to another, such as converting salt ponds to tidal marsh, may benefit some species at the expense of others. Most shorebirds do not prefer tidal marsh plain habitats (Warnock & Takekawa, 1995) and do prefer more open habitats. Our analyses show that waterbirds are found in higher densities on salt ponds than in bayland wetlands. Thus, a greater amount baylands will be required to compensate for loss of salt ponds to maintain current waterbird populations. Conclusion Wetland restoration is an emerging field, and the success of restoration projects has not been proven (Race, 1985). While restoration or rehabilitation has become a primary goal for endangered tidal marsh species in the estuary, these species may not benefit from increased amounts of man-made wetlands that are fragmented or differ in structure from natural marshes. For some tidal marsh species, predation, hybridization, or other causes may be controlling populations. Race (1985) suggests that policies encouraging exchanges of natural wetlands with man-made replacements should proceed with caution. More specifically, Davidson & Evans (1986) recommend that existing wetland features should not be altered without satisfactory compensation, creating new areas of similar value developed with a lead time of 5 years. Conversely, the cost to maintain a hypersaline system for waterbirds without commercial salt production is high. Thus, future efforts should be directed at developing suitable types of managed wetlands to replace values provided by salt pond systems and to maximize the value of the saline ponds that are developed. Different management options such as controlling water levels during periods when food resources are not readily available may improve the value of artificial wetlands for waterbirds (Velasquez, 1993). Acknowledgements Thanks to S. Fregien, S. Wainwright and M. Eagan (USGS), L. Wyckoff, T. Huffman, J. Schwennesen, T. Maatouck, K. Haggard and A. Crout (California Department of Fish and Game), and R. Laird and J. Lament (Ducks Unlimited), L. Allen and W. Bonnet (Can Duck Club), and C. Hickey and N. Warnock (Pt. Reyes Bird Observatory) for assistance with field surveys and analyses. The salt pond project was sponsored by the U.S. Geological Survey conducted in collaboration with K. Miles, D. Schoellhamer, W. Duffy and M. Saiki. L. Vicencio, J. Albertson and J. Buffa provided assistance with aerial survey data. We appreciate the assistance of G. Downard on figures and comments of R. Laird, S.W. De La Cruz and three anonymous reviewers on drafts of the manuscript. References Accurso, L. M., Distribution and abundance of wintering waterfowl on San Francisco Bay Master s Thesis, Humboldt State Univ., Arcata, CA: 252 pp. Anderson, W., A preliminary study of the relationship of saltponds and wildlife South San Francisco Bay. Calif. Fish and Game 56: Bollman, F. H., P. K. Thelin & R. T. Forester, Bimonthly bird counts at selected observation points around San Francisco

12 328 Bay, Febuary 1964 to January Calif. Fish and Game 56: Carpelan, L. H., Hydrobiology of the Alviso salt ponds. Ecology 38: Davidson, N. C. & P. R. Evans, The role and potential of manmade and man-modified wetlands in enhancement of the survival of overwintering shorebirds. Colon. Waterbirds 9: Goals Project, Baylands ecosystem habitat goals. A report of habitat recommendations prepared by the San Francisco Bay Area Wetlands Ecosystem Goals Project. U.S. Environmental Protection Agency, San Francisco, CA and S. F. Bay Regional Water Quality Control Board, Oakland, CA: 209 pp. Haas, J. E., Waterfowl use of grazed and ungrazed wetlands, Napa and Sonoma counties, California. Master s Thesis. Calif. State Univ. Sonoma. Rohnert Park, CA: 116 pp. Harrington, B. & E. Perry, Important shorebird stating sites meeting Western Hemisphere Shorebird Reserve Network criteria in the United States. Unpubl. Rep. to the U.S. Fish and Wildlife Serv., Portland, OR: 121 pp. Harvey, T. E., P. R. Kelly, R. W. Lowe & D. Fearn, The value of saltponds for waterbirds in San Francisco Bay and considerations for future management. Unpubl. Rep. at Wetlands 88: Urban Wetlands and Riparian Habitat. Oakland, CA. Harvey, T. E., K. J. Miller, R. L. Hothem, M. J. Rauzon, G. W. Page & R. A. Keck, Status and trends report on the wildlife of the San Francisco Bay estuary. EPA Coop. Agreement CE Final Report, U.S. Fish and Wildlife Service, Sacramento, CA: 283 pp. Johnson, R. A. & D. W. Wichern, Applied multivariate statistical analysis, 2nd edn. Prentice Hall, Englewood Cliffs, NJ. Josselyn, M., The ecology of San Francisco Bay tidal marshes: a community profile. FWS/OBS-83/23, U. S. Fish and Wildlife Service, Washington, DC. Krebs, C. J., Ecological Methodology. Harper & Row, New York: 654 pp. Lonzarich, D. G., Temporal and spatial variations in salt pond environments and implications for fish and invertebrates. Master s Thesis, San Jose State Univ., San Jose, CA: 81 pp. Lonzarich, D. G. & J. J. Smith, Water chemistry and community structure of saline and hypersaline salt evaporation ponds in San Francisco Bay, California. Calif. Fish and Game 83: Martin, A. C., H. S. Zim & A. L. Nelson, American Wildlife and Plants: a Guide to Wildlife Food Habits. McGraw-Hill Book Co., New York, NY: 500 pp. Miles, A. K., J. Y. Takekawa, D. H. Schoellhamer, C. T. Lu, W. G. Duffy & M. K. Saiki, Science support for wetland restoration in the Napa-Sonoma salt ponds, San Francisco Bay estuary, 1999 progress report. Unpubl. Prog. Rep., U.S. Geological Survey, Davis and Vallejo, CA: 57 pp. Milliken, G. A. & D. E. Johnson, Analysis of Messy Data, Vol. 1: Designed Experiments. Van Nostrand Reinhold, New York. Nichols, F. H., J. E. Cloern, S. N. Luoma & D. H. Peterson, The modification of an estuary. Science 231: Race, M. S., Critique of present wetlands mitigation policies in the United States based on an analysis of past restoration projects in San Francisco Bay. Environ. Manage. 9: Rappole, J. H. & R. M. DeGraaf Research and effective management of neotropical migrant birds. Trans. n. am. Wildl. Natur. Resour. Conf. 61: SAS Institute, SAS Procedure Guide, Release 6.04 Edition. SAS Institute, Cary, NC. Swarth, C. W., C. Akagi & P. Metropoulus, The distribution patterns and ecology of waterbirds using Coyote Hills salt ponds. U.S. Fish and Wildlife Service, San Francisco Bay National Wildlife Refuge, Newark, CA. Terp, J. M., Habitat use patterns of wintering shorebirds: the role of salt evaporation ponds in south San Diego Bay. Master s Thesis, San Diego State Univ., San Diego: 79 pp. Velasquez, C. R., The ecology and management of waterbirds at commercial saltpans in South Africa. Ph.D. Dissertation, Percy FitzPatrick Institute of African Ornithology, Univ. of Cape Town, South Africa: 248 pp. Ver Planck, W. E., Salt in California. Calif. Div. Of Mines Bull. No Warnock, S. E. & J. Y. Takekawa, Habitat preferences of wintering shorebirds in a temporally changing environment: western sandpipers in the San Francisco Bay estuary. Auk 112: Warnock, S. E. & J. Y. Takekawa, Wintering site fidelity and movement patterns of western sandpipers Calidris mauri in the San Francisco Bay estuary. Ibis 138: