Mexican Ducks in summer

[Owen Robertson]

Hi CoBirders!

Perhaps there are few more dreaded phenomena in Colorado birding than late summer Anas duck identification. Mallards molt into their "alternate" plumage (it's still termed "alternate", even though the birds are dull) from roughly June-October, and during those months every pond in the state is teeming with brown-bodied, yellow-billed ducks that may or may not contain a smidgen of diazi DNA. I've been hearing about this problem for a while now (looking at you, Ted), and took the opportunity of a freeform statistics final project to do some analysis on the wonderful eBird data that you all have contributed to! I found that, despite no (known) seasonal movements in Mexican Ducks, they are reported almost half as often in late summer than they are elsewhere in the year. This pattern is extremely strong (p < 0.0001), holds true for all four states where MEDU are regular in the ABA (Texas, Arizona, New Mexico, and Colorful Colorado), and is really quite striking (check out the boxplots and line chart at the bottom of this email for a visualization). 

Which brings me to the (slightly) chastising part of this email: male and female Mexican Ducks are eminently identifiable year-round - they look dramatically different from any worn or molting Mallard you could imagine. So why are we not finding them? The answer is that birders are lazy! (And yes, I'm including myself in that statement). We see a flock of five dozen Anas drifting through the heat haze at John Martin Reservoir in August and plop them on the eBird list as Mallard, but in December that one chocolate brown male with a yellow bill really pops, and bam - Bent County pulls another MEDU. (By the way, the disparity in reports of male and female Mexican Ducks must be wild - somebody wanna check that out? And I don't even want to think about the hybrids in summer...) Since Mexican Ducks aren't known to undertake seasonal movements, they should be out there, ready and waiting to be found by the first birder (that's you!) to take a closer look. (And if they do turn out to migrate or disperse in some way, more rigorous eBird data could help confirm that!). In sum - Mexican Ducks being overlooked in summer/fall is a real problem for eBird data, and the solution is wonderfully simple: just look for them!

Good birding!

Owen

[Mike Thompson]

Owen - from a review perspective this is a rather risky approach. Youth and exuberance can often confuse laziness with wisdom and caution. 

I have a completely different perspective from Montezuma County where MEDU are more abundant in our waterways than anywhere else in Colorado: the dip in the graph reflects careful birders and a solid, accountable review team. Not the other way around. Wise summertime submissions of MEDU/MALL easily fill the dip in the graph. 

And you did not mention the most problematic ID issue with local summertime Anas; young male MALL, not eclipse MALL, are typically quite dark and give a strong vibe for the MEDU-complex. Young male MALL are the primary confusion issue with summertime MEDU reports, not eclipse MALL. And both young and eclipse MALL are confusion issues for hybrids. Yes an experienced birder can separate them, but not usually as carefully as needed in a distant scope view.

A message from a review perspective to all observers is to always positively identity what you can, get the best documentation possible, and use extra caution when reporting MEDU and MEDUxMALL during non-breeding season. 

Mike Thompson 

m...@reardonsteel.us mobile

--
--
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To post to this group, send email to cob...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/cobirds
* All posts should be signed with the poster's full name and city. Include bird species and location in the subject line when appropriate.
* Join Colorado Field Ornithologists https://cobirds.org/membership/
---
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobirds+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cobirds/73a81814-4348-4f8b-99e1-51fa1f0a1d65n%40googlegroups.com.

[Owen Robertson]

Mike,

Thanks for your response. I want to clarify that I’m not in favor of reckless/undocumented Mexican Duck reports - such a program would not improve this gap in data I’m describing. I’m simply saying that in addition to overreporting of MEDU, underreporting (as a result of a lack of observer effort) is perhaps an even greater problem. There will always be Anas ducks that trip us up (the Waneka bird last year is a great example) no matter how well-photographed they are, or how many experienced observers chase them, but such birds will be problematic regardless of season. Conversely, the many well-documented and unimpeachable Mexican Duck records that Colorado possesses should be acceptable as such in June just as much as in January. And similarly, a dubious Mexican Duck in January should not be accepted simply because it is January - Mallard molt timing is variable, and odd plumages can occur at any time of year. I firmly believe that the kind of data gap we see in late summer is fixable - perhaps not to the levels of certainty that we have about MEDU in other times of year, but it is not unimprovable. This improvement, in my opinion, can most strongly begin with observers taking the time to thoroughly check Anas flocks in summer, documenting oddballs, and reporting them - either as MEDU, a hybrid, or as MEDU/MALL (or, perhaps, Anas sp.). If errors are made in identification, they can be corrected, but a lack of effort is not so easy to remedy. If experienced observers can separate MEDU reliably (albeit with care) from odd Mallards, then our goal should be to get birders out there checking duck flocks and learning about the ID, rather than giving up the ghost from the start. Perhaps we’re in total agreement about that, but treating an ID like something that cannot be done, or is reserved for those with decades of experience, is (in my opinion) deeply harmful, both to the Colorado birding community and to the eBird data we’re both striving to improve.

Best,

Owen

[Eric DeFonso]

Hi Owen,

Just a few nitpicks about your otherwise engaging posting.

First, regarding the data used to illustrate interstate comparison....are these the total number of individual human observations of confirmed MEDU in each state, or the total number of instances of confirmed individual MEDU in each state? I'm assuming the latter, in which case it's probably also true that in Colorado at least the vast majority of initial identifications of a persisting confirmed MEDU are made by very experienced observers. Or, in the few cases where the observer is less experienced, it's quickly confirmed by a very experienced observer or reviewer, and then chased and photographed by less experienced observers in order to gain more experience for the next rare occasion when a potential MEDU is located. How do I know this? Well it's basically my own story. I've had to learn as I go in tiny incremental steps on how to identify MEDU in the field, which also requires knowing how to identify hybrids, but I'll get to that more in a moment.

Second, it's worth noting that although the state box plots are all very comparable-looking in terms of proportions of confirmed MEDU by season, the y-axis is indeed dramatically different, and should not be overlooked. CO MEDU number no more than 10% of the next least populous MEDU state (NM) in these graphs, despite the significantly larger human population (and birding population) of CO compared to NM. I would be careful about extrapolating too much information about what MEDU movements are (or aren't) in CO based on such small number statistics. Especially since CO is the only state in this comparison where the species is out of range.

My last nitpick had to do characterizing MEDU ID as simple given how "dramatically" different they are from MALL. Well, that type of hyperbole understates the difficulty of the problem, and seeing as how ABA articles have had to be written by esteemed Colorado birders (Mlodinow, Leukering, Bushong) on how to identify MEDU, calling the process of separating the MALL complex comparable to separating Northern Shovelers from American Wigeons feels a bit denigrating to those of us who have been struggling for years to do things correctly. Does the process of MALL/MEDU/MALLxMEDU separation start to get easier and more straightforward the more you observe and practice it? Certainly, and perhaps even to a point where an experienced observer does feel as if the visual difference between these birds is to them dramatic and obvious. But learning this is still a process for the community as a whole and only keeps moving when we help each other continually with how to separate these birds. I would argue that given the already very low numbers of MEDU likely present in the state at any given time, and given my previous argument that the MEDU-observers with ID skill are already catching the majority of MEDU likely to be found anyway, it probably serves the community better at this point to just assist them with improving their ID skills instead of criticizing their past efforts. Efforts which, as Mike Thompson aptly pointed out, are already being reviewed by a solid and accountable review team. For my part, I want to continue improving my own recognition of the field-identifiable traits, and to improve communication of that information to the birding community. The more confident birders can feel when looking over a distant MALL flock, the more time they will actually spend doing so. It has less to do with laziness and much more to do with encouragement. I also know this from experience. :)

Eric

 

-------

Eric DeFonso
Boulder County, CO

[Mike Thompson]

Thanks Owen - With the limitations outlined below, experienced and inexperienced birders alike cannot reliably distinguish MEDU and hybrids from deceptive MALL without excellent looks and documentation. And experienced birders will more frequently report MEDU/MALL from May through October, when MALL plumage deteriorates.

I can think of several good reasons for a dip in MEDU reports that are not due to lack of effort:

1. The plumage discussion mentioned below; leading to cautious reporting.
2. The uncertainty of Mottled Duck in the mix of eastern ducks
3. Harsh, high-angle spring and summer light producing miserable shimmer in long distance views. Compared with low-angle fall and winter light, which is generally more favorable in the scope.
4. Seasonal behavior patterns of dabblers. It is far more feasible to find a MEDU outlier in a flock of dabblers concentrated in a one-acre feedlot pond which may be the only unfrozen water for miles. Contrasted with the scattered, widespread nature of dabblers from May-October.
5. MEDU are expanding their range northward, making a comparison with long term fixed populations in AZ and TX moot. We do not have a handle on their seasonal movement but which does seem to resemble seasonal frequeny of MALL. The following graphs illustrate this point:

Check out the statewide frequency graph for MALL and how it resembles the same patterns as MEDU and MEUDxMALL.

Now check out MEDU/MALL:

The peak frequency of MEDU/MALL reports occurs during the lull of MEDU and MEDUxMALL reports - mas o menos. From my perspective, that's awesome hand-in-glove data. It strikes me that collectively, we aren't really missing anything. MEDU researches using our data will be aware of reporting limitations during nonbreeding season and can connect the dots on either side.

With all that said, I fully support encouraging "well-documented and unimpeachable Mexican Duck records" around the clock. Amen to that. The best way to do that in my mind is to educate folks on ID points and the data will fall into place naturally. 

Thank you for bringing up the issue, it's very interesting.

Mike