Sun light is about 1kW/square meter.

A 737 has a wing area of 125 square meters. To deliver 7.5MW to those wings will need 60kW/square meter.

That's a far greater energy density.

As for "time in beam" of a moving beam. Presumably one end of the beam is fixed, and depending on the target's velocity the beam might not be moving much at all (oncoming plane).

The beam’s horizontal motion speed goes down linearly with height if the beam is sent from a fixed location.

So, picking 30,000 feet for flight height and 250m/s for the plane’s speed, at 900 feet the beam’s speed would be 7½ m/s. That’s quite doable for birds.

More importantly, if the beam isn’t sent from directly below the plane, and the plane flies more or less directly towards or away from the beam’s source, the speed at which it moves horizontally goes down considerably (1/cos(α), where α is the deviation from the vertical, I think), and its width when flying through it in the direction of the beam’s source at constant height will go up considerably.

The beam’s width, and, with it, intensity, will likely be larger on the ground, but my gut feeling is this won’t be enough to correct for both these factors in all cases (corrections welcome)

I started writing a complex demonstration, but the intercept theorem, (Thales' theorem) indeeds shows quite nicely that from a ground station, tracking a plane flying at (𝒽1,𝓋1) means the beam goes at 𝓋2=𝓋1*𝒽2/𝒽1. So, 5 m/s for a bird at 200m.

That said, I was mostly imagining space-based stations.

In the above case, I talked about ground-based stations at airports. Unfortunately, those would require high angular velocities, but would already be in bird-restricted areas (airports use all sorts of tricks, including drones, sounds, people chasing them, etc, to scare away birds). You could also conceivably drastically lower the beam energy output by using multiple beams that concentrate at a single point.

Now, a fun calculation: what would be the tipping point between birds saved by not emitting carbon dioxide and birds killed by the towers, if there is one?

One last point: I'd hazard that the bird is likely to change direction if it starts running into a beam and it has time to (ie, it is not killed "instantly"). This drastically lowers the dangerous altitude range, as the bird is less likely to fly in the same path as the beam (how likely was it in the first place? One would need to go faster than the other to catch up, but not too much to actually dispense the lethal dose? I haven't done the math, but I'd say it's quite unlikely).

The point where the beam is aimed would be moving with 250m/s. (or about so) Therefore the hypothetical bird which crosses the beam would only get the energy for a fraction of a second.

I don't know if this reasoning is correct or not, but definitely has nothing to do with the speed of light in air. I'm not even sure why you are bringing it up.

I think he's referring to the motion of the beam to track the target.