Martin Management Models: East and West


In the spring of 2004, Stan Kostka who lives in Washington state and is a member of the Western Purple Martin Working Group, happened to be visiting southwestern Pennsylvania, and assisted Ken Kostka and Jeff Hunt of the Purple Martin Preservation Alliance in erecting two wooden T-14 martin houses with crescent entrance holes at the Butlers Golf Course colony near Pittsburgh, PA. As we prepared to bury the T-14 posts that Jeff had assembled the night before, Jeff mentioned how he worried about a pair of Starlings or House Sparrows getting a foothold in one of these houses or in any of the housing at his sites. Even though he uses starling-proof or starling-resistant entrance holes on most of his housing and gets to them at least once a week, an occasional pair of House Sparrows or starlings manages to become entrenched in a house and dominate it until his next visit. Jeff manages eight different sites spread out over a rather large area, and it is difficult to visit them often enough to totally eliminate the threat of nest site competitors.

Stan faces a similar situation. He manages many martin sites dispersed over a large area around northern Puget Sound. But Stan said that he doesn't worry about an occasional pair of starlings or House Sparrows getting a foothold at a site and dominating the housing. Quite simply, they can't, since the western management model doesn't allow it. I'll allow him to explain.

Hi Ken,

Before we get too far along in this East West conversation, I think a little background on Western Purple Martins is necessary. First of all, there is no one western management model. Rather, western martins nest across a wide range of situations, some in cavities within completely natural landscapes, others in entirely "artificial" colonies in suburban backyards, and in many "habitats" in between. Likewise, the politics of western martin conservation range widely also, some folks espousing conservation of martins indirectly through preservation of wild land ecology , while others promote the widespread adoption of the Eastern birdhouse model. I see no reason why the future of western martin conservation needs to go exclusively one way or the other, that is, either all artificial or all natural . Seems to me both are very much justified, and each indispensable, toward the preservation of every existing colony of western birds, and future martin population expansions. Neither extreme should be promoted to the exclusion of the other. Even though I advocate for the preservation of wild lands, that can never replace nestsites forever lost along the permanently altered marine shorelines of the West Coast. Permanently altered landscapes will require some human supplied nest cavities maintained for martins, now that starlings and house sparrows are here. The model I have adopted is something of a compromise, somewhere between "wild" and "artificial", its origins go back to the first attempts at direct martin conservation here in the Northwest.

Here's my understanding of what happened. Few records exist of martins in the Northwest using birdhouses prior to 1970. The two records I am aware of exist because they were unusual enough to be noted in scientific journals (Richmond, Bunch). Except for these records, martins in the Northwest were not known to use nestboxes until after the late 1970s and early 80s, when efforts to rescue a perceived remnant population began. The arrival of European starlings into the region during the 1940s coincided with the beginning of a noted decline in martin numbers. At that time, virtually all known martins nested in or around cites and towns on the Puget Sound shoreline, in a wide variety of sites, various holes in or ledges on buildings, or anywhere some kind of nestsite could be found, in other words, all the places starlings (and house sparrows) nest today . By the mid 1970s, few known pairs of martins remained, some on buildings, and some nesting in holes in old wooden marine pilings. Single nestboxes were installed on pilings at known breeding locations, in an attempt to mitigate a perceived cavity shortage caused by increasing numbers of starlings. Although some Eastern apartment style martin houses were installed early on, they were not successful at attracting multiple pairs . While displaying the same tendencies toward colonialism as their Eastern cousins, Northwestern martins tended to be much more territorial, and preferred more space between active cavities. Since loose clusters of single nestboxes were successful at attracting numerous breeding pairs, other such clusters were installed over the years, and martin populations increased. Compared to known numbers in the late 1970s, Western Washington today has a relatively large and growing martin population, the overwhelming majority of known pairs in loose clusters of single nestboxes [and gourds] on interior marine waters.

Now back to our conversation. Abandoned marine pilings, where remnant pairs were primarily found in the 1970s and 80s, are spread randomly in clusters around various parts of Puget Sound. These old pilings are similar to (or it could be argued they are the same as) submerged snags in a natural setting, a substrate still used by martins in some areas. Individual nestboxes attached randomly to these pilings mimic cavities found in a stand of dead trees used by a variety of primary cavity nesters over several seasons. Now you should have some idea of what the colonies I manage look like. Multiple single boxes, spread out over a couple hundred feet of marine shoreline, with spaces between individual cavities ranging from a few feet to fifty feet or more. This wide spacing of nest cavities is what prevents a few pair of starlings or house sparrows from becoming a significant threat to the entire martin colony.

Across the region 2004 was much the same as 2003 with excellent weather, and an increase in martins. Numbers of pairs were up dramatically at some sites, with reports of martins at other new sites elsewhere in Washington and British Columbia. With the knowledge of so many martins at so many new sites, the time seemed right for some experimentation. In hindsight, it was no accident that I decided not to control starlings at the Camano Island colony in 2004. Having reluctantly tolerated a pair of two of starlings within the site in previous years, it was our conversation back in May of 2004 that pushed me over the edge, and I said to myself, "Lets see what happens if I do no starling control here this year." Although the negative impact of starlings on martin populations has been demonstrated (Brown) it has never been, to the best of my knowledge, specifically documented in a western colony. No active starling control, and I expected pairs of martins likely would decrease from the previous year.

The results were surprising.
Only two nesting attempts by starlings were successful, and both were likely replacement clutches. One of the starling nests fledged only one young from five eggs. Three eggs disappeared before they hatched, and one never hatched. SY martins later used that nest. No tree swallows fledged young. Martins used all four nests that originally had tree swallow eggs or young. In the end, 14 pairs of martins nested, two more pairs than the previous year.

I think the widely spaced cavities allowed early returning ASY martins to establish themselves at the colony site, in spite of the fact that starlings and tree swallows were already nesting there. This (in combination with favorable weather) enabled these ASYs to attract additional martins, whose numbers eventually overwhelmed the tree swallows, and to some extent negatively impacted the starlings.

A more detailed account of the events at English Boom in 2004 is available at


You say that "abandoned marine pilings...are spread randomly in clusters around various parts of Puget Sound [and] are similar to ...submerged snags in a natural setting".  In essence, western martins haven't undergone the tradition shift (with regard to nest proximity) that eastern martins have, making this semi-colonial nesting arrangement the norm. The pilings installed by timber companies along waterways were no doubt responsible for allowing martins to transition from one loosely clustered medium (old snags) to another loosely clustered medium (old pilings)! In the east, where the transition occurred hundreds of years ago, the forests were largely removed, and the only nesting opportunities were in the tightly clustered gourd colonies or martin houses provided by Native Americans and European colonists respectively.

Not only are the western martins distinct in this regard, but it seems their habitat is fairly unique as well. The closest style habitat to Puget Sound in the east would have to be somewhere like Chesapeake Bay.  The typical eastern colony site is located at a golf course, farm field, suburban backyard, state park, or other far inland location. I am not suggesting that it would be impossible for martins to adapt to a more loosely colonial nesting arrangement in their eastern haunts. There are plenty of inland locations that could lend themselves to a western style layout. I have, for several years, managed two Tree Swallow colonies on lock walls along the Allegheny River. An interesting thing occurred this past summer at one of these Tree Swallow colonies.  An SY-M martin had been hanging around an uncolonized T-14 placed in a grassy lawn area about 75 feet from the lock wall, where the Tree Swallows are using gourds spaced 30 ft. apart (with 90% occupancy). While the SY-M martin spent most of his time on the T-14, at one point he flew to the end of the lock wall and checked out the Tree Swallow gourd there, then proceeded to work his way up the wall, checking out each Tree Swallow gourd in turn. The Tree Swallows, in turn, harassed him a bit, but he did not seem flummoxed in the least. Eventually he went back to the T-14, his curiosity satisfied. But what would have prevented him (or an SY-F) from nesting in one of these gourds? Indeed, I have seen martins land on and inspect Tree Swallow or Bluebird boxes. In northeastern Pennsylvania, I have seen them nest in gourds placed only 4 feet off the ground. These gourds had been placed for Tree Swallows in among traditional martin colony houses and gourd racks.

 It is interesting to note that eastern Tree Swallows have undergone dramatic increases due to their propensity to utilize Eastern Bluebird houses. If people are willing to erect individual nesting boxes for Bluebirds and (unintentionally) Tree Swallows, I'm sure there would be those who would be willing to do the same for martins. It would certainly be interesting to take a typical well-populated eastern style martin colony, in a T-14 for example, and break it down (between seasons) into 7 two unit houses on 7 different posts, spaced 5-10 feet apart. Would the martins abandon or spread out?  Interesting stuff! Someone should try it.  But I digress.

Back to the question of nest site competitors. There is no doubt that having the housing spread out would allow starlings and House Sparrows less opportunity to dominate the housing intended for martins. In fact, isn't the reason starlings and House Sparrows act aggressively or destructively (i.e., driving martins off, or breaking eggs, etc...) because they are repelling, what are to them, nest site competitors - the martins! I once heard an old martin enthusiast say that he built one house for the starlings and House Sparrows, and another for the martins, to avoid the competition for the martin house. It occurs to me now that by providing a number of different houses, this is essentially what you are doing in the west.  And for someone who is managing a lot of different sites spread out over a large area, where starlings and House Sparrows cannot be controlled daily, its not a bad idea! Perhaps the scarcity of martins in large parts of the Northeast is partly due to the “tightly clustered” houses that are too easily dominated by abundant nest site competitors that exclude martins.

But for the time being, I would have to say that western martins are uniquely suited to their unique habitats…freeway overpasses and coastal saltwater inlets sprinkled with wooden pilings. Don't the one or two truly inland colonies, located in suburban backyards in Spanaway WA and in Scappoose OR (Dave Fouts territory) seem more like the eastern model…with two or three poles holding traditional eastern style martin housing? Perhaps it is because the average landlord just doesn’t want seven or ten poles in his backyard.


Hi Ken,

    Yes, no doubt there are some nesting situations here that are very much like an Eastern colony, and yes certainly one big factor in how martin housing is set up will more or less always be human convenience. After all, that's what started this discussion, ease and convenience, for us, of controlling nest site competitors. The question is, for managers' of remote sites, what is the best cavity configuration to enable a colony of martins to do that on their own ?

Just to clarify here, neither of the sites you mention are truly "inland" in a Northwest sense. Scappoose is on the lower Columbia River, and Spanaway is low elevation and only 15 miles from the waters of Puget Sound. Both sites are well within the eco-region referred to as the Puget Trough, west of the Cascade mountains. "Inland" commonly refers to areas east of the Cascades, away from the wet coastal weather influence . Spanaway would be accurately described as upland, but not inland. The geography and resulting climate differences between coastal and inland areas on the West coast are much different than the East coast, because of how the prevailing winds hit the mountains.

Another site that comes to mind is the Ziak Refuge in Clatsop county Oregon, which is pictured in Eric Horvath's 1998 survey, showing a gourd rack. In any case, yes the housing is set up like an Eastern site, but again, it is in the backyard of the person managing it, so the problem of controlling nest site competitors at a distant site, is not a factor.

Martins in freeway overpasses ? I think we had better save that topic for another day.