Cactus Wren Campylorhynchus brunneicapillus
Few birds are as dependent on a specialized habitat as the San Diego Cactus Wren is on cactus thickets. The birds remain in their stands of cholla or prickly pear year round, maintaining their nests for roosting. San Diego County has two subspecies of the Cactus Wren, C. b. anthonyi, widespread and fairly common in and near the Anza–Borrego Desert, and the San Diego Cactus Wren, C. b. sandiegensis, localized and seriously threatened in the coastal lowland. The survival of the San Diego Cactus Wren is one of the county’s greatest challenges in bird conservation. Development or fires threaten the wren’s remaining habitat, and that habitat is so reduced and fragmented that its ability to support the birds over the long term may already be impaired.
Breeding distribution: In San Diego County, the San Diego Cactus Wren is concentrated in four regions: southern Camp Pendleton/Fallbrook Naval Weapons Station (about 70 pairs), Lake Hodges/San Pasqual (90 pairs), Lake Jennings (25 pairs), and Sweetwater/Otay (extending from Dictionary Hill on the north to Otay Mesa on the south, from Euclid Avenue on the west to Upper and Lower Otay lakes on the east; 80 pairs). Other San Diego County sites combined contribute probably fewer than 50 individuals. Population estimates, from Mock (1993), are based on data extending back to the 1980s and are not comparable with our atlas results, so trends in the short term are unclear. From 1997 to 2001, high counts in one day in one atlas square were of 30 in the Fallbrook Naval Weapons Station (E6) 8 April 2000 (W. E. Haas), 14 at Lake Hodges (K10) 13 August 1997 (R. L. Barber), 12 in the Wild Animal Park (J13) 3 June 1999 (D. and D. Bylin), 12 at Lake Jennings (O14) 19 June 1998 (M. B. Stowe), and 12 at Sweetwater Reservoir (S12) 9 June 1998 (P. Famolaro). Because these figures are not the results of surveys for the Cactus Wren specifically, they are not comparable with the larger numbers yielded by Weaver’s earlier focused surveys: 20 at Lake Hodges 25 February 1989, 47 in the Wild Animal Park and at San Pasqual Battlefield in March 1990, and 33 at Lake Jennings 9 March 1990. Also, some sites on private property where Cactus Wrens were known before 1997 could not be surveyed for this atlas because of lack of access. This problem rather than habitat loss accounted for our missing the Cactus Wren near Ramona (K13/K14) and on Mother Miguel Mountain (T13), though urban sprawl is extending into the latter area.
Rea and Weaver (1990) listed 77 sites in San Diego County and mapped the species as extirpated in the 1980s at 26 of these. Subsequent records submitted to the California Natural Diversity Data Base and San Diego County’s Multiple Species Conservation Plan added 16 sites, and field work for this atlas, 1997–2001, added about 18 further sites and relocated the species near six sites where Rea and Weaver (1990) thought it extirpated. The “population” at several of these sites, however, consists of as little as a single individual and therefore may be ephemeral or not viable. Very isolated sites, especially likely to fall into this category, are lower Los Peñasquitos Canyon (N8), very well covered, with only one record, of one singing male 7 June 1998 (P. A. Ginsburg), and Sandrock Canyon, Serra Mesa (Q9), with one singing male 2 May–15 June 2000 (J. A. Martin).
The cactus thickets on which the San Diego Cactus Wren depends are restricted to stands of open sage scrub at elevations below 1500 feet on south- and west-facing slopes, at the bases of hillsides within a quarter mile of river valleys. Along San Mateo Creek and the Otay River, they occur in broad dry washes. The wrens especially favor hillside gullies in which cacti can grow especially tall. The birds’ territories range in size from 0.8 to 2 ha, averaging 1.3 ha, thus tending to be smaller than those of Cactus Wrens in Arizona (Rea and Weaver 1990).
In and near the Anza–Borrego Desert, subspecies anthonyi is most common on well-drained soil with abundant cacti (counts in the breeding season up to 31 on Mescal Bajada, J25, 12 June 1998, M. and B. McIntosh; 20 near Indian Hill, R28, 6 May 1998, J. O. Zimmer; 20 near Jacumba, U28, 20 March 1998, C. G. Edwards). It is fairly common in the town of Borrego Springs, where houses are scattered and native desert plants are used widely in landscaping (up to 12 on 15 March 1997, P. D. Ache). But it is lacking from valley floors, where cacti cannot grow in the alkaline soil, from badlands almost lacking in vegetation, and from the sandy region around Ocotillo Wells. Cactus Wrens extend well up the mountains’ east slope, nearly to the head of San Felipe Valley (H20; two in April and May 2000, A. P. and T. E. Keenan) and Boulevard (T26; one on 31 May 1998 and 31 May 1999, D. S. and A. W. Hester). Their elevational range extends up to about 3900 feet (Rea and Weaver 1990). We discovered two sites a short distance onto the coastal slope in southeastern San Diego County, in semidesert scrub with considerable snake cholla: Miller Valley (S24; up to three on 9 May 2000) and 4 miles east of Cameron Corners (U24; two singing males 10 May 1999, L. J. Hargrove). Cactus Wrens, anthonyi on the basis of Weaver’s observations and photographs, also occur on the coastal slope in Dameron and Oak Grove valleys (C15/C16/C17), in sage scrub dominated by flattop buckwheat and deerweed with abundant cacti and yuccas (up to seven, including fledglings, in Dameron Valley, C15, 7 June 1997, K. L. Weaver). Here, in a habitat shared with several other typically desert species, the wrens appear to nest exclusively in the cane (or valley) cholla.
Nesting: The Cactus Wren’s nest, a hollow football-shaped structure with the entrance hole at one end, is unique. Subspecies sandiegensis builds in chollas or prickly pears almost exclusively; of 584 nests observed by Rea and Weaver (1990), only two, in yellow bush penstemons, were not in cacti. But the birds evidently do not discriminate among the coast cholla and either species of prickly pear (Opuntia littoralis or O. oricola). Rea and Weaver (1990) found the median height of 98 cacti in which the wrens were nesting to be 138 cm, range 74–226 cm. Cacti lower than this range do not offer suitable habitat, presumably because the nest is too accessible to predators.
Though also usually selecting cacti, subspecies anthonyi is slightly more flexible in choice of nest site. We noted four nests in Mojave yucca, one in a palo verde, one in a deformed ocotillo, and one apparently hidden in the skirt of a California fan palm, besides many in four species of cholla.
Because Cactus Wrens maintain nests for roosting year round, observations of nest building and occupied nests are not valid clues to the species’ breeding schedule. Likewise, old nests do not confirm breeding. Nevertheless, because the species is almost completely sedentary, the difference between the distribution of nests and the distribution of breeding wrens is minor. The most conspicuous possible exception was at Barrett Junction (U18), where I found a nest but no Cactus Wrens 13 February 2000. Repeated checks of the site through the following spring never revealed any birds, only a gradually deteriorating nest.
Our observations from 1997 to 2001 suggest that mid March to early June is the main season for Cactus Wrens to lay in San Diego County. Forty-three collected egg sets from San Diego County (all but three of the San Diego Cactus Wren) range in date from 7 March to 10 July. Bent (1948) reported a California date as early as 2 March. Our only observations outside this window were in the wet spring of 1998, when fledglings in Culp Valley (G23/H23) 23 and 24 March implied laying as early as mid February (M. L. Gabel).
Migration: Dispersal of Cactus Wrens away from their breeding sites is minimal, but the rare instances of such dispersal may be vital to the species’ viability. Guy McCaskie’s two coastal records of vagrants well away from breeding habitat, from Point Loma 14 October 1967 and Mission Bay 14 February 1970, have not been replicated since. A juvenile found at the south base of Cowles Mountain (Q11) 23 June 2001 (SDNHM 50564) shows that juveniles may leave their natal territories soon after independence from their parents. The known nesting pairs nearest this site are about 1.9 miles to the southwest below Lake Murray and an equal distance to the northeast near Cowles Mountain’s northeast base in southern Santee.
Winter: Winter numbers differ little from those in the breeding season. High winter counts of sandiegensis were of 23 in the Wild Animal Park (J13) 29 December 2001 (K. L. Weaver) and 21 at Lake Jennings (O14) 7 February 1998 (M. B. Stowe). High winter counts of anthonyi were of 22 in north Borrego Springs (F24) 19 December 1999 (P. K. Nelson et al.) and 20 in Box Canyon (L23) 10 January 1998 (S. D. Cameron). The only winter record more than 3 miles away from a known colony was from about 2800 feet elevation in Rancho Cuca (F14; one on 26 and 27 December 1999, S. and J. Berg). Seemingly isolated occurrences in Pamo Valley (J15; two on 2 January 2000, W. E. Haas), along Highland Valley Road about 5.5 west of Ramona (L13; one on 2 January 1999, J. McColm), and in the canyon of the San Diego River above El Capitan Reservoir between Boulder and Isham creeks (M17; one on 24 January 1998, R. C. Sanger) reflect sites possibly occupied only intermittently or sites previously known but missed during the breeding season because of access problems.
In the Anza–Borrego Desert we noted Cactus Wrens in winter in seven squares where we did not see them in the breeding season, suggesting occasional winter dispersal into marginal habitat. In two squares we saw nests but never the birds themselves at any season.
Conservation: Because of its restriction to stands of chollas and prickly pears, the San Diego Cactus Wren has always had a rather patchy range (Bancroft 1923). Nevertheless, it was formerly widespread at elevations below 1000 feet in coastal San Diego County, especially in the area now covered by the inner city of San Diego (Rea and Weaver 1990, specimens in San Diego Natural History Museum). Habitat destruction in the form of urban sprawl threatens the San Diego Cactus Wren gravely. W. L. Dawson recognized this as early as 1923, and since then the threat has only intensified. Cactus Wrens occurred formerly on the south-facing slopes just north of northern San Diego County’s coastal lagoons but were eliminated from them during the 1980s. The pressure from urbanization is especially strong in the range of the Sweetwater/Otay population. The environmental impact statement for Highway 125 specifies elimination of 11 Cactus Wren territories (V. Marquez pers. comm.). Recent public acquisitions of significant tracts of coastal sage scrub for San Diego County’s multiple-species conservation plan include few if any Cactus Wren sites, lying largely too far inland. With the population so reduced and fragmented, the long-term viability of what remains is an open question (Mock 1993).
Rea and Weaver (1990) also identified fire as a threat to the San Diego Cactus Wren, citing Benson (1969) in calling fire “the chief limiting factor in the distribution of cacti in southern California.” The long time required for a burned cactus thicket to regrow to a height sufficient for nesting Cactus Wrens can result in the species’ dying out in burned habitat. One year after the Laguna Canyon fire in the San Joaquin Hills, Orange County, the population of Cactus Wrens was down 72% (Bontrager et al. 1995). The threat was dramatized in February 2002 when the Gavilan fire burned much of the Cactus Wren habitat in the Fallbrook Naval Weapons Station.
Habitat fragmentation may compound the negative effect of habitat destruction. Rea and Weaver (1990) noted that during the 1980s all 26 sites where they documented the bird’s disappearance had supported fewer than five pairs and that at 18 of these sites the extent of the habitat still appeared sufficient to support at least one pair. If the habitat is adequate, however, rather isolated populations may persist. At Malcolm X Library in southeast San Diego (S11), in partly degraded sage scrub isolated for decades from the rest of the Sweetwater/Otay population, I found that about six pairs persisted from 1997 through 2001. At this site and on Dictionary Hill (R12/S12) Cactus Wrens make some use of spiny garden plants around houses adjacent to colonies in native cactus thickets, so the birds’ ability to colonize appropriate landscaping should be explored further, as a means of extending—not replacing—populations in natural habitat.
Other recommendations listed by Rea and Weaver (1990) for conservation of the San Diego Cactus Wren apply now more than ever. The protection of all remaining sites is critical, even if they fall outside multiple-species conservation plans and in the path of proposed highways. Degraded and burned areas need restoration, through planting of cacti. The subspecies’ precise numbers are not clear because of difficult access to many sites; some sites on private property could not be surveyed by atlas observers. A thorough census and continued regular monitoring of the population is essential.
The San Diego Cactus Wren has benefited to some extent from the formal listing of the California Gnatcatcher as threatened, since almost all of the wren’s sites also support the gnatcatcher, but because of the wren’s specializations listing remains important. The petition to list the San Diego Cactus Wren was denied only on the basis of unpublished letters disputing the subspecies’ validity and the statement “no apparent morphological or other morphometric differences have been detected to date that distinguish coastal birds from other cactus wrens” (Beattie 1994), in disregard of the evidence of Rea and Weaver (1990).
Taxonomy: Rea and Weaver (1990) analyzed the subspecies of Cactus Wren in southern California thoroughly, detailing the seven characters in which C. b. sandiegensis Rea, 1986, differs from both C. b. anthonyi (Mearns, 1902), and C. b. bryanti (Anthony, 1894) of northern Baja California. C. b. sandiegensis differs from anthonyi primarily in its more extensively barred tail (especially on rectrices 4 and 5), larger spots on the belly, paler ochre background color on the belly, and less concentration of the breast spots into a dense patch. All known specimens of sandiegensis are distinguishable from anthonyi by at least one character, 87% by three or more characters. A genetic study (Eggert 1997) found evidence of population isolation at a level even finer than the subspecies evident on the basis of external characters. Rea and Weaver (1990) described a difference between the subspecies in song—slower, lower, and raspier in sandiegensis—that has yet to be studied in detail. On the basis of freshly collected fresh-plumaged topotypes, Rea (1983) supported the distinction of anthonyi of the Sonoran Desert from C. b. couesi Sharpe, 1881, of the Chihuahuan Desert.
Rock Wren Salpinctes obsoletus
Like the Canyon Wren, the Rock Wren is a bird of rocky canyons and boulder-covered slopes. But the Rock Wren is more flexible, living also in badlands of eroded earth, the beds of partially filled reservoirs, and sparsely vegetated areas of little relief. It is more dispersive than the Canyon Wren, wintering regularly if uncommonly in small rock outcrops and expanses of disturbed bare dirt where it does not breed. From 1997 to 2002 we saw this more opportunistic lifestyle illustrated by striking fluctuations in numbers in the Anza–Borrego Desert. Rock Wrens increased sharply during the wet winter of 1997–98, remained common for one year, then decreased in the following dry years.
Breeding distribution: The Rock Wren is most widespread and numerous in the Anza–Borrego Desert, especially on rocky slopes. Daily counts in the breeding season ranged up to 50, including 30 singing males, around Indian Hill and Carrizo Palms (R28) 6 May 1998 (J. O. Zimmer), 40 on a bajada southwest of Halfhill Dry Lake (J29) 10 April 1998 (L. J. Hargrove), and 40 at Split Mountain (L29) 18 April 1998 (G. Rebstock, K. Forney). The convergence of these high counts in a single year is no coincidence but followed a winter of unusually plentiful rain. The number of Rock Wrens reported per hour in the eastern third of San Diego County was at least 50% greater in 1998 than it was in any other of the atlas period’s five years. In 1998, Rock Wrens were common also in the Borrego and Carrizo badlands (25 near Five Palms Spring, G29, 16 May 1998, G. Rebstock, K. Forney; 29 in Arroyo Seco del Diablo, N28, 16 May 1998, R. and S. L. Breisch). Even on the flat floor of the Borrego Valley (F24) Rock Wrens breed occasionally (four, adults feeding young, 27 May 2001, M. L. Gabel).
On the coastal slope Rock Wrens are more localized but their numbers are more consistent from year to year. Here they are most numerous in the rocky foothills near the Mexican border, then decrease toward the northwest. The largest number reported from the coastal slope was 20 on the south side of Otay Mountain (V15) 25 May 1999 (D. C. Seals). In northwestern San Diego County the only sites of reports of more than three Rock Wrens in a day were on the southwest slope of Double Peak (J9; 12 on 14 March 1998, four on 15 May 1998, J. O. Zimmer) and near Willow Spring (A5) on the Riverside County line (six on 10 and 22 June 1999, K. J. Winter). We seldom found the Rock Wren above 4000 feet elevation, but the record at the highest elevation was of a nest with nestlings near the summit of Hot Springs Mountain (E20) 8 June 2001 (K. L. Weaver). The Rock Wren is rare along the coast, where the only records in the breeding season are of a singing male 1 mile inland from the Santa Margarita River mouth (G4) 16 April and 1 May 1999 (P. A. Ginsburg), a pair at Torrey Pines (O7) 22 April 1997 (D. G. Seay), and two pairs, each with one fledgling, in Goat Canyon (W10) 12 June 1999 (W. E. Haas).
Nesting: Rock Wrens nest usually in crevices among or under rocks, sometimes in rodent burrows in earthen banks. Several nests reported by atlas participants were furnished with the “porch” of small rock chips, unique to this species. Most nests were in rock piles, banks, or road cuts; one at Morena Dam (T21) 12 April 1997 was in a hole in the dam’s concrete (R. and S. L. Breisch).
Our observations from 1997 to 2001 imply egg laying beginning in early March (fledglings in Blair Valley, L24, 2 April 1998, R. Thériault) and continuing through early June (adult disposing of nestling’s fecal sac between San Dieguito Reservoir and Mount Israel, K9, 28 June 1998, L. E. Taylor), exceptionally to early July (nest with nestlings at upper end of El Capitan Reservoir, N16, 1 August 2000, J. R. Barth). The span falls well within the range 5 February–28 July given for 77 California egg sets by Bent (1948). The wide range suggests multiple broods when the food supply is favorable.
Migration: The Rock Wren shows up in only small numbers away from sites where it could breed, so its migration is inconspicuous. On Point Loma (S7), where it is unknown as a breeding bird, it is recorded from 16 September to 22 February (Unitt 1984). From 1997 to 2001, the latest record from an atlas square where the species was not found during the breeding season was of one at Torrey Pines State Reserve (N7) 3 March 1999 (K. Estey).
Winter: Any winter influx of Rock Wrens into San Diego County is only slight. Numbers in winter are not conspicuously higher than in the breeding season (maximum count 40 near Indian Hill 8 February 1998, J. O. Zimmer). The winter distribution follows a pattern similar to the breeding distribution. We found the species in winter in 99 atlas squares where we did not find it during the breeding season, however, so there is at least considerable dispersal of the local population. The species still occurs rarely at high elevations (one near the summit of Hot Springs Mountain 13 February 1999, K. L. Weaver; one at Mount Laguna, O23, 21 January 2002, E. C. Hall, J. O. Zimmer). It is appreciably more frequent in northwestern San Diego County in winter but still rather rare along the coast, with only 13 records 1997–2002, none of more than two birds.
The Rock Wren’s fluctuations in abundance in the Anza–Borrego Desert in winter were similar to those in the breeding season. The number reported per hour in the last three winters of the project was only one third that in 1997–98.
Conservation: No changes through history in the Rock Wren’s status in San Diego County are known. The Rock Wren benefits to a small degree from human modifications of the environment, using road cuts, riprap, quarries, and areas of disturbed bare dirt—more in winter than in the breeding season.
Taxonomy: Rock Wrens in San Diego County, as throughout the mainland of North America north of southern Mexico, are S. o. obsoletus (Say, 1823).
Canyon Wren Catherpes mexicanus
No bird sound evokes a feeling of wilderness more than the spiraling, echoing song of the Canyon Wren. Cliffs, talus slopes, desert gorges, rocky ravines, and boulder-studded chaparral are the Canyon Wren’s habitat. A year-round resident, the Canyon Wren is uncommon, living in well-spaced territories.
Breeding distribution: The Canyon Wren’s range is constrained by the birds’ need for rugged topography. Such conditions are widespread in San Diego County’s foothills, mountains, and desert, but even here the wren’s distribution is quite patchy. San Diego County’s mosaic of geology also affects the Canyon Wren: mountains formed of granite, such as Woodson or Corte Madera, offer far more exposed rock and so more Canyon Wren habitat than those formed of gabbro, such as Viejas or McGinty, which lack extensive rock outcrops and are more uniformly clothed in chaparral. The higher counts on the coastal slope are from granitic terrain like the San Diego River gorge near Dye Mountain (K18; nine on 9 July 2000, L. J. Hargrove), Lyons Peak (S17; seven on 11 July 1999, J. R. Barth), and the gorge of Cottonwood Creek below Barrett Dam (T18; seven on 8 June 2000, L. J. Hargrove). Nevertheless, the wrens still occur in fair numbers on some steep gabbro peaks like Otay Mountain (four on the north slope, U15, 31 May 2000, L. J. Hargrove).
Other than a single singing male near the mouth of Horno Canyon (E3) 30 May 1998 (R. and S. L. Breisch) we did not find the Canyon Wren during the breeding season within about 5 miles of the coast. The most coastal localities were in San Mateo Canyon (B3; two singing males 28 May 2000, P. Unitt), the canyon of San Marcos Creek (J8; pair 28 June 1997, J. O Zimmer), and Mission Gorge (Q11; up to six on 11 July 1999, N. Osborn). Elevation is not a concern to the Canyon Wren in San Diego County; the species occurs near the summits of both Hot Springs Mountain (E20; singing male 19 May 2001, K. L. Weaver, C. R. Mahrdt) and Cuyamaca Peak (M20; two singing males 23 May 1998, G. L. Rogers). Canyon Wrens reach their peak abundance in the steep canyons descending the east face of the mountains: the maximum count was 14, including eight singing males, in Borrego Palm Canyon (F23) 5–8 July 2001 (L. J. Hargrove). Farther out into the desert the birds are fewer but well distributed in suitable terrain, with up to seven along Alma Wash near Starfish Cove (K28) 24 March 2000 (L. J. Hargrove).
Nesting: The Canyon Wren typically nests in rock crevices, in caves, or on sheltered rock ledges. One nest near Indian Flats (D19) 23 May 2001 was in a vertical crevice just below some White-throated Swift nests (J. R. Barth). The Canyon Wren takes advantage of man-made nest sites too. One nest near the rocky slopes of the south fork of Moosa Canyon (G9) 30 May 1998 was in an outdoor light on a patio (C. Cook); another at Lake Morena (T21) 31 May 1999 was inside an equipment shed (R. and S. L. Breisch).
Bent (1948) reported that 68 California egg sets ranged in date from 28 March to 11 July. The breeding season we observed in San Diego County 1997–2001 agrees with this almost exactly, except that a bird between Lake Poway and Mount Woodson (L12) carrying insects 4 April 1998 (M. and B. McIntosh) implies young hatched from eggs laid by about 19 March. Several species nested exceptionally early in the wet spring of 1998.
Winter: The Canyon Wren engages in only minor dispersal away from its breeding territories in winter. Furthermore, most locations where the species was seen in winter but not the breeding season may be suitable breeding habitat, perhaps occupied only intermittently. The only site where the species occurred in winter more than one atlas square away from where it was found in the breeding season was Ysidora Gorge (G5), along the Santa Margarita River in Camp Pendleton (one on 26 December 1998 and 23 December 2000, P. Unitt). There is little if any retreat from the higher elevations in winter. One individual was near the summit of Hot Springs Mountain 5 February 2000 (K. L. Weaver); others were around 5600 feet elevation near Garnet Mountain (N22) 19 and 26 February 2000 (G. L. Rogers).
Conservation: The Canyon Wren has become a village and even city bird in places in Mexico. At San Javier in Baja California Sur it enters the old stone mission and takes advantage of the church’s acoustics to amplify its song. But in San Diego County it shows little sign of urban adaptation: the walls of packed adobe common in Mexico are a microhabitat no longer duplicated in the United States. Most Canyon Wren habitat is insulated from development by its very ruggedness. The only possible change in the species’ distribution we detected was its absence 1997–2002 from the bluffs at Torrey Pines City Park (O7), where J. L. Dunn noted it occasionally from 1974 to 1976 (Unitt 1984).
Taxonomy: Canyon Wrens in San Diego County are C. m. conspersus Ridgway, 1873, which ranges from the Great Basin south to northwestern Mexico.
Bewick’s Wren Thryomanes bewickii
Bewick’s Wren is something of an enigma, a cavity nester that is common in chaparral and sage scrub, habitats with no trees offering cavities. The answer is the wren’s flexible definition of a cavity: rocks, caves, holes in the ground, and the litter of humanity serve for nest sites just as well as holes in trees. Bewick’s Wren thus can occupy any habitat with at least moderately dense shrubbery: sage scrub, the understory of oak, riparian, and pinyon–juniper woodland, desert-edge scrub, and desert washes, as well as chaparral, the habitat where it is most abundant. Though a year-round resident, Bewick’s Wren spreads uncommonly in winter even into the sparsest desert scrub.
Breeding distribution: The Bewick’s Wren is one of San Diego County’s most widespread birds. In the breeding season it occurs over the entire coastal slope except the Coronado Peninsula, ranging as high as the summit of Hot Springs Mountain (pair with fledglings 15 July 2000, K. L. Weaver, C. R. Mahrdt). Its abundance on the coastal side follows no strong pattern except that numbers are low in extensively urbanized or heavily forested areas. Numbers can be high in both the coastal lowland (e.g., 50 in Vista, H8, 16 May 1999, J. O. Zimmer) and in the mountains (e.g., 33 in Noble Canyon, O22, 22 April 1997, P. Unitt). Counts in the Pine Valley area (O21/O22/P21/P22) 1993–97 suggested that in mature chaparral the only species exceeding Bewick’s Wren in abundance are the Wrentit, Spotted Towhee, and Black-chinned Sparrow (Cleveland National Forest data).
Field work for this atlas revealed Bewick’s Wren to be considerably more widespread than expected on the desert slope. Even in the breeding season it is lacking only from the most sparsely vegetated valley floors and badlands. It is one of the more common birds on scrubby rocky slopes and among pinyons and junipers in desert mountains, with numbers up to 29 on the northwest slope of Whale Peak (L25) 25 June 1998 (R. Thériault). Bewick’s Wren is rare in the town of Borrego Springs itself but resident in the thicket of mesquites on the Borrego Valley’s floor (G25; up to eight on 8 April 1997; pair entering possible nest crevice in a mesquite 11 March 1997, R. Thériault).
Nesting: Bewick’s nests in tree cavities like those used by the House Wren; atlas participants noted three nests in coast live oaks and one in a willow. Trees isolated or scattered in chaparral offer Bewick’s Wrens nest sites in areas too sparsely wooded for House Wrens. But more often we found nests in man-made structures: in crevices of wooden buildings, in pipes, under roof tiles, and in birdhouses. Bewick’s Wrens take advantage of discarded trash that offers cavities, and in treeless chaparral such trash may provide the best nest sites. We noted Bewick’s Wren nests in a discarded hub of a truck wheel, in a cardboard beer carton, in a coffee can, and in an abandoned bullet-riddled car. One nest was reported from a yucca, and it is likely the wrens nest among the leaf bases of Mojave yuccas as well as in woodpecker-excavated cavities in them. Caves, rock crevices, and brush piles have also been noted as nest sites for Bewick’s Wren (Bent 1948).
Our observations from 1997 to 2001 suggest that Bewick’s Wrens lay mainly from mid March to mid June, in agreement with 54 egg sets collected in San Diego County from 1887 to 1937. Occasional pairs may start earlier, building nests as early as 15 February (2001, Pauma Valley, E12, K. Fischer). Nest building may take over three weeks to complete, however (Kennedy and White 1997), so such dates may not be good guides to dates of egg laying. Reports of Bewick’s Wrens “feeding young” as early as 18 March could refer to males feeding incubating females. But fledglings near Ross Lake (B7) 26 March 1999 (K. L. Weaver) must have come from a clutch in which incubation began about 24 February.
Migration: Because there are so few areas in San Diego County where Bewick’s Wren is not resident, there is almost no information on its migration here. At places in the Anza–Borrego Desert where it is not suspected to breed, Bewick’s Wren has been seen as late in spring as 26 March (1999, one in June Wash, M27, R. Thériault). Away from its few breeding locations in the Salton Sink Bewick’s Wren has been recorded from 29 August to 17 April (Patten et al. 2003).
Winter: Bewick’s Wren is even more widespread in San Diego County in winter than in the breeding season. Over most of the county winter numbers are very similar to those in the breeding season (maximum daily count 50, north of Morena Reservoir, S21, 16 February 1998, S. E. Smith, and along Espinosa Trail, R19, 23 February 1999, G. L. Rogers). The birds remain through the winter at least in small numbers at high elevations, being recorded from the summits of Hot Springs Mountain (E20) and Monument Peak in the Laguna Mountains (O23). Even there the impression of numbers lower than in the breeding season could be due simply to less calling and no singing in the winter. Though territorial pairs may be sedentary, there is appreciable movement, perhaps mainly by young birds. We noted Bewick’s Wren in winter in almost every atlas square in the Anza–Borrego Desert where it was absent during the breeding season. Numbers of such dispersers into sparse desert scrub are usually small, but daily counts were sometimes as high as nine, along Fish Creek Wash (M27) 2 February 2000 (M. B. Mulrooney) and at Split Mountain (L29) 2 December 1999 (P. D. Jorgensen).
Conservation: Despite its heavy use of man-made artifacts as nest sites, Bewick’s Wren is not an urban adapter. Apparently its need for extensive shrubbery for foraging, or a dependence on insects of native habitats for food, keeps it in territories with a component of native scrub. But it is relatively tolerant of this habitat being fragmented. Bolger et al. (1997) and Crooks et al. (2001) identified Bewick’s Wren as a species of low sensitivity to fragmentation, and this is corroborated by our atlas results. In central San Diego, Bewick’s Wren is missing from only those squares that have been totally urbanized, lacking any significant native scrub. Of the eight chaparral birds analyzed by Crooks et al. (2001), the Bewick’s Wren was the least sensitive, predicted to have a 95% chance of surviving for 100 years in fragments as small as 13 hectares.
In the eastern U.S., Bewick’s Wren has experienced serious population decline and range contraction, attributed to the increase and spread of the House Wren, which competes for nest cavities and destroys the eggs of other birds (Newman 1961). Even though the House Wren is increasing in San Diego County, a concomitant decrease of Bewick’s Wren seems unlikely here. The treeless chaparral in which Bewick’s Wrens are so common is unsuitable for breeding House Wrens. Unless development of the entire coastal slope penetrates far deeper into foothill chaparral than it does now, into the Cleveland National Forest and other open-space reserves, it seems that large numbers of Bewick’s Wrens will remain well isolated from both the inroads of the House Wren as well as from direct loss of habitat.
Taxonomy: Bewick’s Wrens in San Diego County are T. b. charienturus Oberholser, 1898, in which the upperparts are dark chocolate brown and the central rectrices are so dark as to mute the blackish bars. T. b. correctus Grinnell, 1928, is a synonym of charienturus, based on the comparison of newer unfoxed and older foxed specimens (Rea 1986). With a range encompassing southwestern Alta California and northwestern Baja California, this subspecies—darker than others in the western United States— is another taxon characterizing the San Diegan District of the California Floristic Province.
House Wren Troglodytes domesticus
The House Wren breeds commonly in San Diego County’s woodlands of oak, sycamore, and conifers, but here it is only beginning to take on the city-dwelling habits that justify its name over so much of the United States. Within its historic range, almost coextensive in San Diego County with that of the coast live oak, the wren has long patronized birdhouses as well as natural nest holes. But only since the 1990s has it spread into suburbs built over former sage scrub. It is on the increase as a winter visitor or year-round resident too, possibly in response to climatic warming.
Breeding distribution: The House Wren is widespread on San Diego County’s coastal slope, most numerous in oak woodland and riparian woodland with large sycamores. It is common from low elevations (100 in Horno Canyon and along the south fork of San Onofre Creek, D3, 31 May 1998, K. Perry, D. Gould; 135 in Los Peñasquitos and Lopez canyons, N8, 11 April 1999, P. A. Ginsburg et al.) to the higher mountains (13 near the summit of Hot Springs Mountain, E20, 18 June 2000, K. L. Weaver, C. R. Mahrdt; 25 at Mount Laguna, O23, 13–14 June 1998, C. G. Edwards). The eastern edge of the House Wren’s range follows the eastern edge of the coast live oak’s range almost exactly. The exception is along San Felipe Creek, where the wren extends in riparian woodland east to Sentenac Ciénaga (J23; up to three, all singing males, 14 June 1998, R. Thériault). Also, there is one record of the House Wren nesting in the cottonwoods at Butterfield Ranch in Mason Valley (M23; two on 17 April 1999, P. K. Nelson).
In the city of San Diego the House Wren has begun colonizing eucalyptus groves and urban trees away from natural habitat, using birdhouses, man-made structures, and possibly crevices behind loose strips of eucalyptus bark for nest sites. The spread is going from north to south. The largest concentration found in nonnative habitat is in Pottery Canyon, La Jolla (P7), with up to 11 on 27 May 1999 (L. and M. Polinsky). South of Mission Valley House Wrens are still uncommon through the breeding season but probably nesting at Point Loma College (S7; courting pair 10 May 1997, J. C. Worley) and confirmed nesting in the Tijuana River valley (W10; nest with nestlings 8 May 2001, T. Stands, S. Yamagata) and at the west edge of Otay Mesa (V12; male singing while female carried nest material into a horizontal pipe, 13 April 2001, P. Unitt).
Nesting: As a cavity nester, the House Wren takes advantage of a wide variety of nest sites. Most frequently, it uses old woodpecker holes and natural cavities in large trees, especially coast live oaks and western sycamores. Other trees in which atlas observers noted House Wren nests were cottonwood, willow, Engelmann oak, and black oak. The House Wren is also San Diego County’s leading customer of birdhouses. Electrical boxes, open-ended pipes, and street lamps were other man-made nest sites reported repeatedly. We noted Cliff Swallow nests taken over by House Wrens on three occasions.
The House Wren offers one of the strongest examples of a breeding season shifting earlier in the year. Fifty-six egg sets collected in San Diego County from 1890 to 1942 range in date from 14 April to 13 June. The range for 119 sets collected throughout California reported by Bent (1948) is 11 April–26 June. Our observations from 1997 to 2001, however, show that House Wrens now begin laying regularly about 1 April. These observations include several of nests with nestlings, adults disposing of fecal sacs, and adults carrying insects as early as 13 April, as well as a nest with eggs in a box along the Sweetwater River near Highway 94 (R13) 9 April 1997 (A. Mercieca). Occasionally the birds may lay even as early as about 16 March, as implied by young already fledged at the upper end of Sweetwater Reservoir (S13) 14 April 1999 (P. Famolaro). The change has taken place only at the beginning of the season, as we observed nests with nestlings as late as 12 July, meaning egg laying as late as about 14 June.
Migration: Because the House Wren occurs in San Diego County year round, its migrations are difficult to define exactly. It is not clear whether the population turns over completely from summer to winter or whether some breeding birds remain as permanent residents. Spring migrants return to their breeding territories at low to moderate elevations in March; even as high as 4400 feet (Upper Green Valley, M21) the birds may be singing and paired as early as 21 March (1997, P. D. Jorgensen). In the Anza–Borrego Desert, away from sites where it winters, the House Wren is recorded as an uncommon migrant from 13 March (1999, two in Borrego Palm Canyon, F23, A. G. Morley) to 13 May (1995, one in Hellhole Canyon, G23, H. A. Wier). The 15 spring migrants reported by Massey at Agua Caliente Springs (M26) 4 April 1994 were exceptional; from 1997 to 2001 we did not note any concentration in the Anza–Borrego Desert greater than four.
Postbreeding dispersal begins as early as 29 July (1992, one in Culp Valley, H23, M. L. Gabel), though fall migration is generally not obvious until late August.
Winter: Wintering House Wrens are most numerous in the coastal lowland, where daily counts can be as high as 22 along the Santa Margarita River north of Fallbrook (C8) 23 February 2002 (K. L. Weaver), 21 in San Pasqual Valley (K12) 29 December 2001 (C. G. Edwards), and 21 in lower Los Peñasquitos and Lopez canyons (N8) 2 December 2001 (D. K. Adams). With increasing elevation, the birds become scarcer, though as high as 4100 feet elevation we noted numbers up to six around Twin Lakes (C18) 24 January 1999 (P. Unitt). The highest winter House Wren locations were around 5300 feet in the Laguna Mountains (N22, 12 December 1998, G. L. Rogers; P23, 14 January and 21 December 1999, E. C. Hall, J. O. Zimmer) and at nearly 5500 feet at the Palomar Observatory (D15, one on 20 December 2001, K. L. Weaver).
On the east side of the mountains, wintering House Wrens are regular in San Felipe Valley (up to five near San Felipe, H20, 21 December 1998, I. S. Quon, and 27 February 1999, A. P. and T. E. Keenan). But at lower elevations in the Anza–Borrego Desert they are quite uncommon (no count of more than two per day) and restricted almost exclusively to oases and the irrigated floor of the Borrego Valley.
Conservation: The House Wren’s primitive breeding distribution followed that of oak woodland closely, in a pattern much like that of the Oak Titmouse or Acorn Woodpecker. Since the 1970s, breeding House Wrens have become ever more widespread in the coastal strip from which they were once absent. It is unclear why a bird that occupies suburban backyards over much of North America should have been so slow to take advantage of the city of San Diego. The wren’s spread has followed that of the Nuttall’s Woodpecker, one of the primary excavators of cavities in which House Wrens nest.
The House Wren’s range and numbers have increased considerably in winter as well. In spite of Emerson’s (1887) report of one collected and another seen on Volcan Mountain (I20) on 24 and 28 January 1884, and another collected at Witch Creek (J18) 13 December 1904 (FMNH 144799), Willett (1912) and Stephens (1919a) said that only “a few” wintered in the coastal lowland and mentioned no wintering at higher elevations. The House Wren’s status as a widespread if uncommon winter visitor in San Diego County’s foothills and mountains emerged only as a result of the field work for this atlas. Numbers on San Diego Christmas bird counts have increased, the total per party-hour more than doubling from 0.052 from 1966 to 1975 to 0.132 from 1992 to 2001. Increasing winter low temperatures may be enabling more House Wrens to remain through that season. As a hypothesis for testing, I suggest that an increasing proportion of the breeding population is failing to migrate, and these now permanent residents are responsible for the breeding season shifting earlier in the year.
Taxonomy: Only the grayish western subspecies of the House Wren, T. d. parkmanii Audubon, 1839, is known from California. The name T. aedon (Vieillot, 1809) has been shown clearly to have been proposed after T. domesticus (Wilson, 1808) (Oberholser 1974, Rea 1983, Banks and Browning 1995). Continued use of aedon is based on tradition rather than conformity to the international code of zoological nomenclature.
Winter Wren Troglodytes troglodytes
The Winter Wren reaches the southern limit of its range in San Diego County, where it is a rare winter visitor. The wren’s apparent rarity is exacerbated by the difficulty of finding it: the birds hide in dense woodland undergrowth, revealing themselves only with their double “chick” note. One summer occurrence of the Winter Wren on Palomar Mountain was far outside the species’ breeding range and one of the least expected events brought to light by the field work for this atlas.
Winter: Most of the published records of the Winter Wren for San Diego County are from well-birded sites around metropolitan San Diego, especially Point Loma and the Tijuana River valley. The more uniform coverage of the county directed toward this atlas, however, yielded Winter Wrens only in the foothill and mountain zones, likely reflecting the species’ distribution more accurately. Three of the nine reported from 1997 to 2002 were around Palomar Mountain, and together with previous records for the area (e.g., Beemer 1949) suggest this is the most likely area for Winter Wrens in San Diego County. Nevertheless, at least 45 Winter Wrens have been reported in the county, south to Smuggler’s Gulch (W10), site of the southernmost record of the Winter Wren along the Pacific coast. In the Anza–Borrego Desert the Winter Wren has been reported only from Coyote Creek Canyon (D23; 7 December 1977 and 16 February 1978, AB 32:400, 1978) and Pinyon Mountain Road (K24; 26 November 1983, B. Massey).
Winter Wrens have been seen in San Diego County only singly, except for two in Marion Canyon (D12) 6 March 1949 (Beemer 1949). The number reported per winter during the atlas period varied from zero in 1997–98 to five in 1999–2000; the only earlier year with more was 1983–84, with six.
Migration: Dates for the Winter Wren in San Diego County range from 30 September (1981, Point Loma, S7, R. E. Webster, AB 36:218, 1982) to 6 March (cited above), except for one at Warner Springs (F19) 3 May 1999 (C. G. Edwards) and the single summer record.
Breeding distribution: The sole summer record of the Winter Wren in San Diego County is the only one from anywhere in California south of Santa Barbara County and the Sierra Nevada. Clark R. Mahrdt and Edward C. Hall observed one in Jeff Valley (E15), at 4900 feet elevation on Palomar Mountain, repeatedly from 28 May to 12 July 1999. It sang territorially, and on the final date was carrying twigs, but since only one bird was ever seen, presumably it was an unmated male. Hall tape-recorded the song and sent the recording to K. L. Garrett at the Los Angeles County Museum of Natural History; Garrett confirmed the song as typical of western Winter Wrens.
Taxonomy: The Winter Wren has yet to be collected in San Diego County, so the subspecies occurring here is (are?) uncertain. Both the dark rufous T. t. pacificus Baird, 1864, and the less rufous T. t. salebrosus Burleigh, 1959, have been collected well to the south of their breeding ranges (Rea in Phillips 1986). One in the Tijuana River valley 3–15 December 1990 was "thought to be a bird from the eastern population (i.e, T. t. hiemalis Vieillot, 1819) on the basis of its call" (D. M. Parker, AB 45:322, 1991). Winter Wrens giving the call of hiemalis have been heard in Death Valley and the Imperial Valley (Patten et al. 2003) and probably photographed in southeastern Arizona (Monson and Phillips 1981), but the subspecies has yet to be collected in winter west of New Mexico.
Marsh Wren Cistothorus palustris
The Marsh Wren is well named, for it is rarely seen outside a marsh. It occurs in San Diego County in two roles. As a year-round resident it inhabits freshwater and brackish marshes mainly along and near the coast—home of subspecies C. p. clarkae, narrowly restricted to coastal southern California. As a winter visitor from farther north it is more widespread, invading salt marshes, wet grassy areas, and marshes too small to support a resident population. As a conservation issue the Marsh Wren is complex too: much of its primitive habitat has been destroyed, yet in San Diego County, over the second half of the 20th century, its breeding range spread considerably.
Breeding distribution: The core of the Marsh Wren’s distribution in San Diego County is the “Mesopotamia” between the Santa Margarita and San Luis Rey rivers. The birds inhabit many lakes and ponds in these river valleys as well as marshes in the river channels themselves, upstream along the Santa Margarita to north of Fallbrook (C8; pair with nestlings 24 July 2001, K. L. Weaver) and along the San Luis Rey to Couser Canyon (E10; two on 6 and 13 June 1998, K. Aldern, M. Bache). Numbers in this area are as high as 30 at Guajome Lake (G7) 31 May 1999 (D. and C. Wysong), 20 at Whelan Lake (G6) 19 April 2001 (J. Smith), and 15 at O’Neill Lake (E6) 5 June 1998 (P. A. Ginsburg). Marsh Wrens are also resident in the coastal wetlands, even small ones, from San Onofre (C1) to Los Peñasquitos Lagoon (N7). At the larger lagoons, numbers are as high as 50 at San Elijo (L7) 6 June 1999 (B. C. Moore) and 27 at Los Peñasquitos 2 May 1999 (D. K. Adams).
Elsewhere in the county breeding Marsh Wrens are quite localized. The largest colonies outside the core range are in La Jolla Valley (L10; up to 10 on 7 May 2000, K. J. Winter), along the Sweetwater River near Interstate 805 (T11; 28 on 18 April 1999, W. E. Haas), and at the Dairy Mart pond in the Tijuana River valley (V11; eight on 9 May 1999, P. Unitt). At other places numbers are small—no more than five reported in one atlas square per day. Outside the coastal lowland there are only a few breeding-season records, and the species is apparently inconsistent at any site. On the coastal slope such reports are from near Warner’s Ranch (G19; one on 17 June 2000, J. D. Barr) and a pond straddling the line between squares T23 and U23 1.25 miles northeast of Cameron Corners (one singing male 16 June 2001, L. J. Hargrove). On the county’s desert slope, the Marsh Wren has never been confirmed breeding but has occurred during the breeding season along Coyote Creek at both Middle Willows (C22; two, including a singing male, 6 May 2001, P. D. Jorgensen) and Lower Willows (D23; one on 19 May 1999, B. Peterson; one on 21 May 2001, M. L. Gabel; three on 12 June 1994, Massey 1998), in San Felipe Valley near Paroli Spring (I21; two singing males 13 June 1999, J. O. Zimmer), and at Carrizo Marsh (O29; one on 4 May 1978, P. D. Jorgensen; one on 17 April 1998, M. C. Jorgensen).
Nesting: Using strips of marsh plants, Marsh Wrens build a characteristic globular nest with a side entrance, lashing it to several erect leaves, typically of cattail or tule, emerging from water. Placement over water rather than concealment helps protect the nest from predators. Males build several nests; for the eggs, females select one the male’s nests or build yet another of their own. The sedentary Marsh Wrens of western Washington use their nests for roosting through the winter (Verner 1965), and a report of an “occupied nest” at the Santa Margarita River mouth (G4) 12 February 1998 (B. Peterson) suggests that San Diego County’s breeding Marsh Wrens, also sedentary, may do so as well.
Marsh Wrens enjoy a long breeding season in San Diego County. Two families of newly fledged young at Buena Vista Lagoon (H5) 15 April 2000 (J. Determan) must have hatched from eggs laid as early as 20 March; nestlings in the Santa Margarita River north of Fallbrook (C8) 24 July 2001 (K. L. Weaver) must have hatched from eggs laid as late as the first week of July. Thus the breeding season we observed extends the 29 April–20 June range of 22 San Diego County egg sets preserved at WFVZ and even the 24 March–22 July range of 113 sets from throughout California reported by Bent (1948).
Migration: Occasional Marsh Wrens show up away from breeding localities as early as August (one in the Tijuana River valley—before the species colonized this area—14 August 1978, P. Unitt; one at the Borrego sewage ponds, H25, 23 August 1998, P. D. Jorgensen). Most likely these early birds are short-distance dispersers of the local population—possibly from the Salton Sea in the latter case. The main influx of long-distance migrants from east of the Sierra Nevada does not begin until the third week of September (Unitt et al. 1996). In spring, our latest record during the atlas period of a Marsh Wren away from any suspected breeding site was of two at the Borrego sewage ponds 4 April 1997 (H. L. Young, M. B. Mosher). But in previous years the Marsh Wren had been noted at San Diego as late as 25 April (1965, G. McCaskie), and this date agrees with migrants’ schedule elsewhere in southern California and Baja California (Unitt et al. 1996). A climatic shift toward warmer winters could result in the migratory subspecies of the Marsh Wren departing earlier—and, because their migration is facultative, reaching southern California in smaller numbers.
Winter: San Diego County’s local population of Marsh Wrens is greatly augmented in winter by migrants from the north and northeast. The migrants mix with the local subspecies clarkae but show up at many additional places. In the coastal lowland numbers in such places range up to 20 in Vista (G8) 23 December 2001 (M. Lesinsky), 10 at San Dieguito Reservoir (K8) 28 December 1997 (J. Determan), and 10 at the upper end of El Capitan Reservoir (N17) 6 February 2001 (D. C. Seals). Marsh Wrens occur uncommonly and locally at higher elevations as well, where 10 at Swan Lake (F18) 18 December 2000 (G. L. Rogers) was a high count. No more than four were reported from any other foothill or mountain location, but these locations ranged in elevation as high as 4600 feet at Lake Cuyamaca (M20; up to three on 18 February 1999, A. P. and T. E. Keenan), 4650 feet at Doane Pond (E14; one on 24 December 2000, G. C. Hazard), and 5100 feet in Crouch Valley, Laguna Mountains (P22; two on 31 December 1998, P. Unitt). In the Anza–Borrego Desert the Marsh Wren is uncommon, very local, and irregular except possibly at Middle and Lower Willows in Coyote Creek Canyon. The highest desert counts 1997–2002 were of four at ponds in the northern Borrego Valley (E4) 19 December 1999 (P. R. Pryde) and 18 February 2001 (P. D. Ache).
Conservation: With the elimination of almost all coastal wetlands from Los Angeles and Orange counties, southern California’s endemic subspecies of the Marsh Wren undoubtedly experienced a huge population decline. Presumably Orange County’s vast “Gospel Swamp,” whose birds were never adequately documented before the swamp was destroyed (Hamilton and Willick 1996), was once the center of this subspecies’ population. Cistothorus p. clarkae is now recognized as a species of special concern by the California Department of Fish and Game. In San Diego County, however, until 1953 the only sites where Marsh Wrens were known to breed were Guajome Lake, the county’s largest natural freshwater marsh (20 egg sets in WFVZ), and nearby San Luis Rey (G6; Sharp 1907). Even Sharp’s “San Luis Rey” may have actually been Guajome Lake. Willett (1912) wrote that A. M. Ingersoll had noted the (presumed) local subspecies in “early spring” at Lindo Lake (O14/P14), but without a specimen or precise date we cannot be sure the birds were not migrants. Marsh Wrens apparently first colonized the San Pasqual Valley and Mission Valley in 1978, the Tijuana River valley in 1980 (Unitt 1984). Other likely nesting sites such as the Sweetwater River emerged only during the field work for this atlas.
Because of near total human control of water in coastal San Diego County, the Marsh Wren is at the mercy of wetland management. But it has apparently benefited more from the installation of ponds and reservoirs than it has suffered from elimination of wetlands. In Mission Valley, where vegetation was removed along the San Diego River in 1988 and 1989, then allowed to regrow, Marsh Wrens recolonized in 1993. Increased siltation of the coastal lagoons, combined with reduced tidal flushing, converts saltwater habitat into freshwater habitat suitable for nesting Marsh Wrens, and this is the most likely factor that permitted the birds to spread (Unitt et al. 1996).
A trend toward a dryer climate is a concern for a bird as dependent on wetlands as the Marsh Wren. The drying up in 2001 of the Dairy Mart ponds in the Tijuana River valley rendered unsuitable what had become one of the species’ more important local sites.
Taxonomy: The subspecies of the Marsh Wren breeding in coastal San Diego County is C. p. clarkae Unitt, Messer, and Théry, 1996, distinguished from other subspecies by its small size, dark rufous rump and scapulars, and extensively black crown (brown usually reduced to a small patch on the forehead). Its known range is confined to the coastal strip from Los Angeles to San Diego, except for two singing males, presumably clarkae, at Lagunita El Ciprés near Ensenada, Baja California, 6 July 1999 (Erickson et al. 2001). The east end of Batiquitos Lagoon (J7) is the type locality. For many years, California’s coastal Marsh Wrens were called C. p. paludicola Baird, 1864, but that subspecies, with an extensively brown back and crown, apparently ranges no farther south than Lincoln County, Oregon (Unitt et al. 1996). Summering birds in the Anza–Borrego Desert are most likely C. p. aestuarinus (Swarth, 1917), as this is the subspecies resident in the Imperial Valley (Unitt et al. 1996). It is somewhat larger and paler than clarkae, with more brown than black on the crown and more white streaking on the back. No specimens of aestuarinus have yet been collected in the county, however. Winter visitors are C. p. pulverius (Aldrich, 1946), from the western Great Basin and Columbia Basin, and C. p. plesius Oberholser, 1897, from the intermountain region east of pulverius. These partially migratory subspecies are longer-winged than the sedentary ones and, pulverius especially, have whiter underparts and paler tawny scapulars and rump.