I recently returned from Peru where I spent some much needed time watching and photographing Kelp Gulls. I spent most of my time working several beaches and coastal lagoons around Lima. The subspecies of Kelp found here is nominate
L.d. domincanus - incidentally, this population is the northernmost stronghold for the species in South America.
As noted in Part 1 of this two-part series, there is a dearth of information on Kelps in the north of South America. Perhaps not surprising, there is little data on molt timings or molt strategies for populations in Peru. Howell & Dunn (2007) give molt descriptions for southern South America, while Olsen & Larsson (2004) simply assert that northern birds molt "later" than southern conspecifics.
With this being the austral spring, I was expecting to find the majority of 1st cycles with juvenile flight feathers intact, or at the most, a few advanced individuals commencing inner primary molt. Instead, I discovered a radically different flight feather molt sequence - one that, to my knowledge, has not been documented for this species.
Below, I present a series of images illustrating a molt sequence not previously described in
Larus dominicanus. Rough estimates as percentages (n=375) are provided for some of these molt patterns.
 |
A: Hatch Year (HY) Kelp Gull with all renewed, 2nd generation, rectrices and secondaries. All juvenile primaries retained. Flight feather molt apparently suspended at s1/p1. Approximately 35% of this age cohort showed a similar molt sequence.
|
 |
| B: Renewed tail feathers and more than half of secondaries already 2nd generation. Secondary molt in motion. All juvenile primaries. Approximately 15% of birds showed a similar sequence with some scaling back to a less advanced pattern to the outer secondaries (see next example). |
 |
C: Similar to the individual above, but note the outermost secondaries now take on more of a regressive, 1st generation appearance. This was noted in roughly 3-5% of individuals.
|
 |
D: This individual has roughly half of its secondaries renewed, but with no apparent active molt or molt gaps. Tail feathers 2nd generation. Seen in about 25% of individuals.
|
 |
E: Similar to the bird above, tail feathers 2nd generation (except for r5). Tertials renewed as are a few adjacent, inner secondaries. All juvenile primaries.
F: Again, tertials renewed as are a few inner secondaries. Juvenile primaries and outer wing coverts.
|
The sequence of tail feathers and secondaries being renewed before any primaries have been dropped is unlike the regimen followed by most large gulls in the northern hemisphere, where flight feather molt is typically initiated by a prebasic molt, typically beginning at p1, p2 and so forth. With this being the austral spring season - a time of transition from prealternate to prebasic molts, and a time when the two molts commonly overlap - important questions to consider are:
1) Are the 2nd generation secondaries and rectrices part of an extensive 1st prealternate molt?
2) Are they the first flight feathers to be replaced in the 2nd prebasic molt, preceding molt of primaries?
3) With the highly variable mixture of retarded upperwing coverts and scapulars, versus those that are rather advanced and adult-like, is it possible some hatch year Kelp Gulls have a preformative molt soon after hatching, followed by an extensive prealternate molt late in the austral winter/early spring?
4) The advanced, adult-like, aspect to the renewed secondaries is puzzling. Is it possible these individuals become adult-like in 3 molt cycles?
Perhaps the explanation isn't an easy or straightforward one, and maybe there are multiple pathways at work with no single answer. In those individuals undergoing active molt in the remiges, roughly 1/4 showed some primary molt. This is where matters become much more interesting and perplexing.
 |
G: The outermost secondary is retained juvenile, with p1-p2 dropped. Rectrices 2nd generation. Found in approximately 10% of individuals. With such an advanced aspect to the scapulars, such birds give the impression of a 2nd cycle at rest.
|
 |
| H: More similar to the molt sequence of a large gull from the northern hemisphere, this individual has renewed p1-p2, but has also replaced tertials and 2-5 adjacent secondaries. This molt pattern was the exception rather than the rule and only observed in a handful of birds (less than 2%). |
 |
| I: Another bird "seemingly" coming close to the flight feather molt pattern of a northern gull, although the assumption here is that molt began with the innermost secondaries and worked its way out, distally. In reality, this sequence isn't very similar to a northern gull's flight feather molt pattern. All tail feathers appear to be 2nd generation, but possibly juvenile(?). Note the contrast between s1 and p1, suggesting some time lapse. Found in less than 5% of all individuals. |
 |
| J: All 2nd generation secondaries and rectrices (although there is a possibility these are juvenile tail feathers?). Some outer primaries renewed (p5 dropped, p6-p10 2nd generation on right wing; p8-p10 2nd generation on left wing). This sequence appears to be eccentric! Found in roughly 3-5% of all individuals. |
 |
K: Similar to "J" above, but only primary molting is p9 with all others juvenile.
|
 |
L: A most unusual case. p1-p3 juvenile, p6-p8 renewed 2nd generation. P9-P10 retained juvenile (symmetric). Secondaries renewed. The tail on this bird is ambiguous to me, but may be juvenile.
|
 |
| L2: Same individual above. |
 |
| L3: Same individual above. |
The percentages given above are crude estimates based on my daily field notes and review of several thousand photographs. The values are percentages of birds in active molt, and don't account for the subgroup showing entirely juvenile remiges (a substantially small number of birds were still with complete juvenile flight feathers). These numbers are only meant to give an impression of how common (or uncommon) each molt pattern is in late October and early November for this particular population.
 |
| Limited prealternate molt has produced mostly new 2nd generation scapulars, several proximal wing coverts and a few tertials (bold white edges). But also notice the replaced p9 on the left wing (not likely systematic molt). Such birds with almost all juvenile remiges were clearly a small minority. |
Needless to say, larger samples and more study is necessary in order to get a more accurate assessment of which molt is producing which feathers. But if the 1st prealternate molt is said to include tertials, then naturally, the few inner secondaries replaced on birds such as D-F must be an extension of this molt. Further, some advanced birds apparently go on and continue to replace all secondaries (such as A). The big wrench in this new data is where the primary molt fits in. Birds such as I-L need careful analysis.
 |
| A large number of individuals - from those with some degree of renewed secondaries - resembled the bird above. This is a striking appearance that would attract attention in a flock of gulls in North America. |
 |
| Same individual above. We can be sure these young Kelp Gulls replace flight feathers beginning closest to the body and outward. Most interesting is the aspect of these new secondaries, appearing somewhat 3rd generation in nature. |
Final Thoughts
As this is a single point in time in the calender year, there's no telling if those with more extensive primary molt are individuals that have migrated from southern populations, or, are largely resident birds which are afforded a speedy and extensive molt due to the rich food source of the Humboldt Current. A question to ponder is if such birds - as adults - are confined to a 12 month breeding cycle. A population's breeding cycle is an integral component to understanding how that group regulates feather molt. My visit, being at the onset of the breeding season, is a time when many adults are building, protecting and sitting on nests. From Yorio et al (2016), we can infer that young fledge from January through February in Peru. Thus, the youngest free-flying Kelp Gulls on display during my visit were approximately 8-10 months of age. However, there are some who believe Kelp Gulls on the Peruvian coast may have asynchronous breeding schedules (Monica Paredes pers comm).
References
Howell, S.N.G., and J. Dunn. 2007. A Reference Guide to Gulls of the Americas. Peterson reference Guide Series. Boston. Houghton Mifflin.
Olsen, K.M., and H. Larsson. 2003. Gulls of Europe, Asia and North America. London: Christopher Helm.
Pyle, P. 2008. Identification Guide to North American Birds, Part 2. Slate Creek Press, Bolinas, CA.
Yorio, P. 2016. Distribution and Trends in Kelp Gull (Larus dominicanus) Coastal Breeding Populations in South America. Waterbirds 39: 114-135.
