Sacramento Purple Martin Nesting Population:

Decline Continues and New Predation Threat Emerges

 Daniel A. Airola,
Northwest Hydraulic Consultants
3950 Industrial Blvd #100c
West Sacramento, CA 95691

Dan Kopp,
California Department of Parks and Recreation

One Capital Mall, Suite 410
Sacramento, CA 95814

            We report on the status of the 2013 nesting population of the Purple Martin (Progne subis) in the Sacramento area. This area has supported the last sizable nesting population in California’s Central Valley, where it was once widespread (Airola and Williams 2008). The Purple Martin is recognized by the California Department of Fish and Wildlife (DFW) as a species of special concern due to substantial reductions in the species’ geographic range and numbers (Airola and Williams 2008).  

            We previously reported a consistent decline in the Sacramento Purple Martin nesting population from 2003 through 2009 (Airola and Kopp 2009) and then a slowing of the rate of decline during 2010-2012 (Airola and Kopp 2011). Here we report on a dramatic decline in the nesting population in 2013, and a new predation threat that may accelerate the decline in the future.  We also discuss statewide implications of this and other apparent declines in other populations in California.

Figure 1. Number of nesting pairs of Purple Martins at colonies in the Sacramento Region, 2002-2013.


             As we have annually since 2002, we surveyed for nesting pairs of Purple Martins at bridges in the Sacramento region (Sacramento, Yolo, and western Placer counties) that had been previously identified as occupied or suitable for use by the species. Colony locations and other suitable sites, and the criteria used to define them, were described by Airola and Grantham (2003), Leeman et al. (2003), and Kopp and Airola (2007).

            To count nesting pairs, we mapped holes in which martins nested and recorded diagnostic breeding behaviors (i.e, carrying food to nests, removing fecal sacs, begging by nestlings, and nestlings perched at hole entrances;  Airola and Grantham 2003, Leeman et al. 2003). These methods provide a consistent and repeatable basis for estimating the nesting population. We confirmed breeding of 96% counted pairs through observation of diagnostic breeding behaviors in 2013. Nesting by the other 4%, which apparently failed before nestlings or diagnostic behaviors could be detected, was inferred based on nest building and subsequent frequent hole entry by pairs (see Airola and Grantham 2003).


 Colony Occupancy and Nesting Populations Status

             In 2013, Sacramento region Purple Martins nested at seven colony sites (Table 1) the fewest number since systematic monitoring of the population began in 2002 (Airola and Kopp 2011).  This number compares to 11-12 colonies occupied during 2003-2008 and nine colonies during 2009-2012. Two occupied sites that were abandoned in 2013 were 20th St (in Highway 50 in Sacramento) and Taylor Rd (in Highway 65, Placer County). Our historical research and monitoring has shown that 20th St was occupied regularly as early as 1976 Airola and Grantham 2003) and continuously since we began annual monitoring in 2001 (Airola and Kopp 2011, Airola unpub data). The nesting population there, however, had been in steadily declining since 2005 (Table 1).  Taylor Road was abandoned in 2013 after this previously occupied and abandoned site had been re-colonized by a pair with a second-year (SY) male in 2012.

            Forty-six pairs nested at Sacramento colonies in 2013, representing a 28% population decline from 64 pairs in 2012. This decline is the highest rate of annual loss in the Purple Martin population since intensive monitoring began in 2002 (Figure 1). Overall, the martin nesting population in Sacramento has declined by 73% from its high of 173 pairs in 2004.

            Declines in nesting pairs between 2012 and 2013 occurred at five colonies. Small increases (1 pair each) occurred at two colonies and one colony had stable numbers (Table 1). Two colonies, Sutterville and Redding Avenue, support 50% of the remaining population. Three colonies supported nesting populations of three pairs or less. In the past, such low numbers at a colony tend to be associated with site abandonment in subsequent years (Table 1).

 Kestrel Predation at Nesting Colonies

             In addition to the dramatic decline in the nesting population, Purple Martins in 2013 encountered a new threat in the form of predation by American Kestrels (Falco sparverius). Kestrels have been known to nest annually immediately adjacent to the Sutterville and Redding Road colonies since at least 2009 (Airola and Kopp, unpublished records in, but no predation of martins by kestrels had ever been observed. We discuss predation observations and martin response at each of these two colonies.

            Redding Avenue Colony. American Kestrels were observed regularly during the breeding season at the Redding Avenue Purple Martin colony since 2009, and had nested for several years prior to 2013 in a hole in a nearby building. Although martins often sounded alarm calls when kestrels approached martin nesting colonies, we observed no evidence of martin predation by kestrels in over 60 visits annually to the colony before 2013.

            In 2013, Kopp observed kestrels regularly after martin colony monitoring began on 16 March. He first observed kestrels entering "weep hole" entrances to the bridge nesting chambers on 26 May, when various martin pairs were nest-building and some were likely incubating. Martins attacked kestrels beneath or adjacent to the colony on at least 15 subsequent visits.  Over the season, kestrels were seen entering at least six weep holes that were not occupied by martins, and one kestrel was observed attempting to enter an active nesting hole, but was driven away by martins and human monitors. On 23 June, Kopp found the clipped wings of a adult male martin, as is typically done by raptors. The next day he found four dead nestlings below another hole. Mass "fallouts" such as this typically happen when adults stop attending the nest. Thus, it appears that predation on at least one of the nesting pair caused nesting failure.  At least two other nests failed after the nest building period, although their loss cannot be unambiguously attributed to predation. 

            The Redding colony also supports a sizable nesting population of White-throated Swifts (Aeronautes saxatalis).  Some of the holes not used by martins that were entered by kestrels likely were occupied by swifts.  Kestrels were observed capturing or feeding on swifts during five visits in 2013.

            Sutterville Overpass Colony. Kestrels were first noted as nesting adjacent to the Sutterville colony in 2010 and were observed there during the 2011 and 2012 nesting seasons.

            In 2013, Airola first observed an American Kestrel near the Sutterville colony on 16 March, and observed a pair copulating on 18 March. Kestrels were seen regularly during subsequent monitoring in April through June, and a pair successfully nested in a wooden utility pole 60 m north of the Sutterville colony.  Characteristically, martins were first seen at the colony in early April 2013 and began nest-building at the colony in early May.

            On 26 June, Kopp observed a kestrel entering an active martin nest hole at Sutterville and then flying out with a nestling. The adult martins chased the kestrel as it flew away with the nestling and later attacked it when it tried to re-enter the same nest hole. Kopp observed a kestrel trying to enter the same hole daily during 27 June through 30 June, and flying under the colony as if looking for other occupied holes.

            In the morning of 28 June, Airola saw martins entering seven nest holes; nestlings could be heard calling in two holes (indicating they were more than about 12 days old) and adults carried food to nestlings in two others. A kestrel was present, but we observed no attacks on martins during 30 min at the site. On 29 June six holes were active (two with food deliveries and one other where removal of a nestling fecal sac was seen). A kestrel carrying a lizard was attacked by five martins and took the prey to a recent kestrel fledgling nearby. It then returned and flew beneath the martin nesting colony twice, being attacked by 10-15 martins each time. Later that evening, Kopp observed an adult martin carry food into a nest hole and then a kestrel attempt to enter the same hole; the kestrel was driven off by martins and Kopp.  Subsequently, martins attacked a kestrel under the nesting colony on 2 July.  On visits on 2, 4, 7, and 14 July, no martins were present at the colony.

            The lack of martins at the Sutterville colony after 2 July suggests that nesting was disrupted. The stage of nesting on 29 June, as indicated by timing of previous nest building and observations of nest behavior (i.e., older young seen or heard in nest holes), indicates that young from at least three pairs may have been close to fledging age at this time, but that many of the other seven pairs present likely were not. One family group of at least one adult and three HY martins were seen on 7 July about 1.9 km WSW of the colony likely represents a successful nesting from Sutterville, since it is considerably closer than the next closest colony (the California Railroad Museum, 5 km away). The lack of martins at the colony during early July when many of the nests at Sutterville likely had nestlings that were too young to have fledged, and when many nests at other Sacramento colonies still had nestlings, suggests that kestrel predation either eliminated nesting directly through predation or caused nest abandonment before fledging occurred in as many as seven nests.

            Attempt to Control Kestrels. Soon after we observed predation, we arranged with several experienced raptor biologists to capture the kestrels, using a captive Great Horned Owl (Bubo virginianus) as a lure, for relocation away from the colonies. We then contacted both DFW and  U.S. Fish and Wildlife Service (USFWS) and requested approval to capture and relocate the kestrels. Both agencies rejected the request. DFW said that under existing law it could not authorize predator removal to benefit a species not listed as threatened or endangered (K Thomas, pers. comm.). A USFWS' Migratory Bird Permit Office representative also stated that it "cannot authorize removal of predators to protect non-threatened/endangered species", and "We prefer not to interfere with natural predator-prey relationships even with threatened/endangered species and would rather let natural selection occur" (J. Brown, pers. comm.).


 Population Decline

             The rapid decline in the number of Purple Martin nesting colonies and the size of the  nesting population in 2013 increases concern for this small, isolated, and remnant population. The population is at its smallest size  in 16 years of monitoring since 1992 (Airola and Grantham 2003, Airola and Kopp 2011). We calculated a rate of decline of 60% for the period of 2004-2009 and projected that at the average annual rate of decline of 16%, the population would be extirpated within 22 years (Airola and Kopp 2009). Although the rate of the decline slowed to an average of 2.9% annually during 2010-2012 (Airola and Kopp 2011), the 28% decline in 2013 brought the average rate of decline over 2004-2013 back to 16%. Therefore, the Sacramento Purple Martin population remains on a downward trajectory.  If this rate of decline continued at this rate, the nesting population would decrease to only 8 pairs within 10 years, and complete extirpation would occur within 17 years (i.e., by 2030). This rate of decline, however, does not consider the likely higher mortality rates at colonies as they became smaller and especially effects of the kestrel predation which only manifested itself late in the 2013 nesting season. Therefore, the future rate of population decline is likely to be even more rapid than this projection suggests.

            The combination of continued long-term decline in the nesting population, fragmentation of a number of colonies into potentially non-viable units, and the new threat of nest predation by kestrels accelerates already dire concerns about the future of the Sacramento Purple Martin population. At this point, it may well be that the population is too small to maintain resiliency against the pervasive threats. As one of only two Central Valley populations, and the only sizeable one (Airola and Kopp 2011, Sylvester and Airola 2010), the loss of this population would substantially reduce the likelihood of any future recovery of the species in the large Central Valley portion of California.

 Predation Effects

             The Sacramento Purple martin population has been shown to be sensitive to changes in habitat conditions, including increases in train and car traffic, loss of flight access to nesting areas, loss of perch sites, loss of nest material collection areas, and feral cat predation. Before 2013, predation by raptors has generally been infrequent and considered of little or no importance to the population (Airola and Kopp 2007, 2009; Kopp and Airola 2012).

            Kestrel predation in 2013 appears to have disrupted reproduction to some extent at both the Redding and Sutterville colonies.  We were unable to directly access nest sites to determine reproductive success, but our observations show that kestrels were frequently present during the nestling period and perceived by martins as threats at the colonies.  Although we do not have quantitative data, we believe that the threat of kestrel predation caused martins to spend more time in nest defense that otherwise would have been used to provision young. The observation of direct predation during our relatively brief visits to sites and the apparent abandonment of nests during the height of the nesting period suggests that more unobserved direct predation could have occurred or that predation caused abandonment of active nests.  These potential predation effects further stress this rapidly declining population.

             Therefore, in addition to a precipitous decline in number of pairs attempting to nest in 2013, kestrels appear to have caused substantial loss of reproduction in the 2013 nesting population. Studies and abundant anecdotal observations elsewhere show that nesting colonies that suffer significant predation are often abandoned in subsequent years (Cousens, pers. comm., Therefore, we expect abandonment by many Sacramento martin pairs of these two largest colonies in 2014, presumably through relocation to other colony locations.  Our work has shown that larger colonies are more persistent from year to year than smaller ones (Kopp and Airola 2011). Therefore, predation-forced dispersal of martins from these two largest colonies, to relocate at other colonies with few pairs or even to currently unoccupied sites, may accelerate decline further from the previous pre-predation rates. 

 Effectiveness of Predator Control as a Martin Conservation Tool

             Perhaps the kestrels predating on martins at the Sutterville Rd. and Redding Ave. colonies could not have been captured for removal. The inability of DFW and USFWS to grant permission to address the kestrel predation threat, however, ensured that predation was allowed to continue at these two largest remaining martin nesting colonies. In describing the purpose of designating species as state species of special concern, Shuford and Guepel (2008) noted that

            "... a high priority should be placed on protecting natural processes and species, subspecies, and distinct populations that are nearing endangerment because of declining populations or vulnerability to threats. Success will be enhanced if efforts are intensified before populations decline further..."

 Clearly, the Sacramento Purple Martin population is not depending on "natural processes", and in fact depends on artificial conditions in its highly urbanized environment. Yet they are a distinct and last sizable Central Valley population that is clearly nearing extirpation and has received minimal conservation efforts from responsible agencies. Despite our considerable private volunteer efforts to understand the species' biology and apply conservation measures, the population has not responded.

            An important question is whether the isolated Sacramento population of the Purple Martin is just too small and under too much pressure to survive. Banding studies during 2003-2009 showed that adult mortality rates were substantially higher than in other stable populations, presumably due to high rates of vehicle collisions (Airola and Kopp 2007, Airola, unpub. data). Threats to habitat suitability through development and highway bridge modification continue at several sites. The addition of kestrel predation to an already precarious situation suggests to some that the Sacramento Purple Martin is a breeding "sink" (i.e., cannot reproduce at a rate to offset mortality; E. Pandolfino, pers. comm.). Certainly, all our conservation efforts to date have not secured the population, and we did not have any immediate solutions to the longer term issues affecting the population even before the kestrel threat developed.

 Implications for the Statewide Purple Martin Population

             The continued rapid decline of the Sacramento Purple Martin population raises concern about the status of other martin populations and the species as a whole in California. At its peak in 2004, the Sacramento population represented 9-20% of the estimated statewide population of 900-1,850 nesting pairs (Airola and Williams 2008, Airola and Kopp 2009).  The decline by 127 breeding pairs there since 2004 represents a decline in the statewide population by 7-14%.  Sacramento represents one of only four Purple Martin colonies in northern California for which any long term monitoring is available to assess trends (Airola 2009). Although the other three populations showed general signs of stability when surveyed in 2008, extensive data exist only for the population at Shasta Lake (Lindstrand 2008, unpubl data). In addition, these four colonies support only about 6-12% of the remaining statewide population. Based on evaluation of occupancy of colonies between the 1990s, early 2000s, and late 2000s, Airola (2009) concluded that only those Purple Martin populations in the North Coastal region of the state appear to be numerous and healthy.

            More recent declines also have occurred during the last few years at Shasta Lake and Lava Beds National Monument.  The number of breeding pairs at Shasta Lake declined from 2012 to 2013 by 37% (from 27 to 17; L. Lindstrand, pers. comm.).  Lava Beds, where martins have been known to nest at numerous sites since the 1950s (Lund 1977, Williams 1988, Airola 2009) did not support any nesting pairs in 2012  or 2013 (A. Ellinger, pers. comm.). Therefore, although the nest colony conditions in urban Sacramento are somewhat unique, the continued decline of this martin population cannot be considered simply as an anomaly.

            If the apparent inexorable decline of Sacramento Purple Martin population cannot be reversed, the only conservation benefit of our work may be in increasing awareness of the fragility of the species' status in the state. Current status assessments of California Purple Martins consist of  a few relatively fragmented efforts conducted by a variety of private individuals with limited agency and nonprofit support. The general lack of recent information on the statewide status of the species, and the indications of declines in the few monitored populations, suggests a strong need for responsible management entities to support a coordinated effort to more fully determine the status and management needs of this largely ignored species. Previous efforts to advocate for such an assessment have resulted in only limited support from USFWS (Airola 2009).

            Using the past is a guide, it appears the Purple Martin will not receive needed protection and recovery management, and will continue to decline, unless it is listed as threatened or endangered.  Due to the lack of comprehensive monitoring, however, sufficient information may not exist to support a petition to list the species. The burdens to prove that endangered status is warranted appears to have substantially increased in recent years, and thereby slowed the pace of other species listings. Will the lack of commitment on the part of agencies and statewide conservation organizations to  determine the current California status of the Purple Martin, and the greater burden of proof required to list the species, result in no action for the species until it becomes unrecoverable statewide, as now appears to have occurred in Sacramento?


             We thank Rosa Jimenez for field assistance, Jim Estep and Dick Anderson for assistance in attempting to develop a control plan for kestrel predation at martin colonies, and agency personnel Jennifer Brown (USFWS) and Kevin Thomas and Kevin Cahill (DFW) for considering our kestrel control proposal. Bruce Cousens and Ed Pandolfino provided advice and useful discussion. Thanks to Len Lindstrand for information on martin population status at Shasta Lake and for comments on the paper and to Amy Ellinger for information on population status at Lava Beds National Monument.


Airola, D. A. 2009. Status of Purple Martin populations in Northern California: results of a pilot study to develop and apply a survey methodology. U. S. Fish and Wildlife Service, Sacramento Field Office, California.

 Airola, D. A., and Grantham, J. 2003. Purple Martin population status, nesting habitat characteristics, and management in Sacramento. Western Birds 34:235–251.

Airola, D. A., and D. Kopp. 2007. Breeding population status and mortality assessment of Purple Martins in Sacramento during 2006. Central Valley Bird Club Bulletin 10: 34-44.

Airola, D. A., and D. Kopp. 2009. Recent Purple Martin declines in the Sacramento region of California: recovery implications. Western Birds 40: 254-259

Airola, D. A. and D. Kopp. 2011. The decline in the Sacramento Purple Martin nesting population slows during 2010-2012. Central Valley Bird Club Bulletin 14:108-112.

 Airola, D. A., and Williams, B. D. C. 2008. Purple Martin (Progne subis), in California Bird Species of Special Concern: A ranked assessment of species, subspecies, and distinct populations of birds of immediate conservation concern in California (W. D. Shuford and T. Gardali, eds.), pp. 293–299. Studies of Western Birds 1. Western Field Ornithologists., Camarillo, CA, and California Department of  Fish and Game, Sacramento.

 Kopp, D., and Airola, D.A. 2007. First documented nesting by the Purple Martin in Placer County, California, in nearly a century. Central Valley Bird Club Bulletin 10:68–70.

 Kopp, D. and D. A. Airola. 2012. Occurrence patterns and behavior of Purple Martins at a breeding season roost site in Sacramento, California. Central Valley Bird Club Bulletin 15:67-74.

 Leeman, T. S., Airola, D. A., and D. Kopp. 2003. 2003 status of breeding Purple Martins in Sacramento. Central Valley Bird Club Bull. 6:61–68.

 Lund, T. 1978.  The Purple Martin in the western United States, Part 2: it's a question of holes.  Oregon Birds 4(2): 1-9.

 Lindstrand III,  L. 2008. Purple Martin distribution and nesting habitat at Shasta Lake, California. Western Birds 39:166-170.

Sylvester, V. and D. A. Airola. 2010. Purple Martins nesting in low elevation transmission towers in the San Joaquin Valley, California. Central Valley Bird Club Bulletin 13:69-75.

 Williams, B.D.C.  1998.  Distribution, habitat associations and conservation of Purple Martins in California.  M.S. Thesis, California State University Sacramento.


Figure 1. Number of nesting pairs of Purple Martins at colonies in the Sacramento Region, 2002-2013.