Birds and Bugs
- edgeoutfitting
- Oct 13
- 6 min read
Updated: 1 day ago
Many of you who’ve fished with me know that birds can be a serious distraction. And some of you are equally distracted by our avian friends. But guides all over the world are known to be amateur birders, with particular knowledge of how birds can help them understand and interpret what’s happening beneath the water’s surface. For example, this April while in southern Belize, my guide (who I'd classify as a professional birder) and I cruised the shoreline with 10 and 11 wt rods looking not at the water, but to the skies for pelicans. Diving pelicans meant bait balls. Bait balls meant rolling tarpon. And tarpon on the fly meant eternal glory.
Birds are important indicators to me and my work on the Upper Madison as well. Barn, violet-green, and tree swallows on a cloudy day portend a great dry fly bite. Kingbirds hovering over the bank then quickly darting back to cover in early August mean the mimic moths are out. And nobody fishes those…But perhaps the most exciting bird sighting is the common American Robin bouncing along the shoreline in June. That means something very special is beginning. Salmonfly emergence.
There aren’t many guides left who exclusively pursue the salmonfly dry bite. We’re a stubborn lot. It’s risky business. Guides are hired to help catch fish. And the idea of consistently and reliably putting people on large trout with a size 2 or 4 dry fly everyday for almost 3 weeks could be considered insane and often exhausting. Both angler and guide are physically and emotionally maxed during the day. From a guide’s perspective, the strategy of when and where and how to fish this hatch is part science and part voodoo. But those of us who are dry fly only know the signs of an impending good bite. Observation of what's happening all around you is critical. For myself, most indicative is the ubiquitous American robin. Not a particularly flashy bird. Most wouldn’t give it a second glance. But over the years, I’ve noticed that the robin is very closely associated with salmonfly emergence. While adult salmonflies are conspicuously large, they aren’t always easy to see in the grass from a moving drift boat. But the breast of a gorging American robin is very easy to spot. So I keep track of that throughout each day. Because even if I can't see the bug, I know it's why that Robin is bouncing in the grass and I'm in the right spot.
Given my background in wildlife research and my bird nerdiness, the internet algorithm knew to target me with ads about this new device that records bird calls, and uses artificial intelligence to detect species. And of course I bought it because I’m a degenerate. It’s made by Birdweather, and testing and modeling suggest it’s extremely accurate. Then one day while talking to Dr. Albertson at MSU who is conducting research on salmonflies and measuring variables surrounding emergence timing on the Upper Madison, I had the idea to place this recording device at the same location as where she was monitoring salmonfly emergence. The site was at a good friend’s cabin on the upper Madison near Raynold’s Pass. Because there was someone checking for salmonfly emergence every day before, during, and after the hatch, if I could place this device out before the hatch began maybe I could see if there were any interesting trends in bird activity during the hatch. And boy was there.
Beginning in June, before the hatch, Dr. Albertson or my friend checked the area for salmonflies or their shucks each day. My recording device was placed on June 4th. No adult salmonflies were present on the Upper Madison at that point. I would hopefully get a couple of weeks or more of data on bird activity pre hatch, then during the hatch, and on afterwards. In addition, you can set how accurate or confident you want your results to be. For me, I selected that I wanted the device to be a minimum of 80% confident in its species accuracy. A quick look at the data shows that the majority of recorded and classified detections were far higher than 80% confident. I took down the device on October 1.
What I found was incredible. The first thing I noticed was 12,549 individual detections of violet-green swallow calls. That’s a lot. The last one was on August 28. Then I looked above the device and a swallow nest was about 5 ft above it. So, in the data, I occasionally offer two graphs of data, one with and one without swallows as their detections related to salmonfly emergence is questionable because, well, the nest. But it's interesting that they all split and headed south before September.
Between June and October, here’s the total list of bird species (72) detected by the device.
Species List
American Crow
American Goldfinch
American Kestrel
American Robin
Bald Eagle
Belted Kingfisher
Black-billed Magpie
Black-capped Chickadee
Black-headed Grosbeak
Brewer’s Blackbird
Brewer’s Sparrow
Broad-tailed Hummingbird
Brown-headed Cowbird
Canada Goose
Cassin’s Finch
Cedar Waxwing
Chipping Sparrow
Clark’s Nutcracker
Cliff Swallow
Common Grackle
Common Nighthawk
Common Raven
Dusky Flycatcher
Eastern Kingbird
Eurasian Collared-Dove
European Starling
Evening Grosbeak
Golden-crowned Kinglet
Gray Catbird
Great Blue Heron
Green-tailed Towhee
Hairy Woodpecker
Horned Lark
House Finch
House Sparrow
House Wren
Killdeer
Lark Sparrow
Lazuli Bunting
Lincoln’s Sparrow
MacGillivray’s Warbler
Mountain Bluebird
Mountain Chickadee
Mourning Dove
Northern Flicker
Northern Rough-winged Swallow
Orange-crowned Warbler
Osprey
Pine Siskin
Red Crossbill
Red-tailed Hawk
Red-winged Blackbird
Rock Wren
Rufous Hummingbird
Sandhill Crane
Savannah Sparrow
Sharp-shinned Hawk
Song Sparrow
Spotted Sandpiper
Steller’s Jay
Tree Swallow
Vesper Sparrow
Violet-green Swallow
Warbling Vireo
Western Meadowlark
Western Tanager
Western Wood-Pewee
White-crowned Sparrow
Willow Flycatcher
Wilson’s Warbler
Yellow Warbler
Yellow-rumped Warbler
Here are some key highlights from the dataset:
Date range: June 4, 2025-October 1, 2025 (120 days)
Total detections: 47,272
Unique species detected: 72
Mean detections per day from June 4 to October 1: 394
First observed salmonfly at the site: June 27, 2025
Last observed salmonfly at the site: July 8, 2025
Observed salmonfly emergence: 12 days
Observed salmonfly emergence window bird detections: 14,590 (31% of all detections)
Observed salmonfly emergence window mean bird detections/day: 1,216
Observed salmonfly emergence window unique bird species: 41
Total American Robin detections: 6,042
Robin detections during salmonfly emergence period: 4,632 (77% of all robin detections!)
Total House Wren detections: 3,630
House Wren detections during salmonfly emergence period: 1,881 (52% of all house wren detections)
Here are a few graphs to illustrate the data.
What’s interesting about Robins is that they were still very abundant well into October. It’s not as if they migrated after salmonflies. But this data would seem to suggest a resource pulse that brings Robins to riparian habitat intensely for a short duration of time, confirming my observations over the years. Other birds had similar responses to salmonfly emergence (e.g., house wrens) but none of them even came close to the robin in intensity of the response.
The average daily number of detections spiked 3 fold for all bird detections during the 12 days of observed salmonfly emergence, driven largely by robins. But still, over half of all unique species at the site were detected during the emergence. Word got out about the bugs, that's for sure.
Detections were super high right out of the gate compared to later in the summer. I’m curious what they were like in early spring. I’ll find out this year as I plan on deploying it in April. I suspect early season hatches like caddis and mayflies drove the early spike. Also interesting was the big dip on July 3. Why is that? From my notes that day on the water, we had epic winds and clouds. Both are known to depress bird vocalizations. So, more than likely birds were still present in high numbers enjoying salmonflies, they just weren’t vocalizing as much. Also, high winds may have muted vocalizations more than usual. Then, later in July another slight peak. I have my suspicions about what drove that response.
The ecological suggestions of the data are fascinating. We tend to think of the health of the river only in terms of the fish that swim in it. Clearly the importance of its health extends beyond the riffles and into the wings of birds and likely beyond. It reminds me of a famous quote attributed to John Muir: "When one tugs at a single thing in nature, he finds it attached to the rest of the world." The quote is slightly wrong, though. What he actually wrote in his memoir was: "When we try to pick out anything by itself, we find it hitched to everything else in the universe." But you get the point. More and more, the evidence suggests that salmonflies are a keystone species. Who knows what else they’re hitched to.
Comments