Premise
The antenna gain of a transmitting antenna relates its radiation intensity U(θ,ϕ) to its input power as shown in the following formula:
$$G(θ,ϕ)=\frac{U(θ,ϕ)}{P_{IN}/4\pi}$$
Where the radiation intensity in far field zone is related to the frequency domain poynting vector $$\vec{S}(r,θ,ϕ) = \frac{1}{2}\cdot \vec{E} \times \vec{H}^*$$ by the equation:
$$U(θ,ϕ)=|\vec{S}(r,θ,ϕ)|\cdot r^2$$
The radiation intensity does not depend on the distance because of the inverse square behaviour of the Poynting vector, so it is only a representation of how the antenna distributes the power in the different directions (θ,ϕ).
Observation
The Poynting vector is not only defined in the frequency domain, but also in the time domain. Its definition is this:
$$\vec{S(t)}(r,θ,ϕ) = \frac{1}{2}\cdot \vec{E(t)} \times \vec{H(t)}^*$$
It represents the instantaneous power flowing out from the antenna.
Instead, the frequency domain Poynting vector is a vector whose real part is the average power transmitted by the antenna. In the case of harmonic (sine wave) fields, there is the factor 1/2 that comes from the averaging operator. This explains the reason why the frequency domain Poynting vector is defined with such a factor.
Question
The antenna gain is defined from radiation intensity which is defined from the frequency domain Poynting vector. So, it seems to me that the antenna gain means how an antenna "amplifies" (or "attenuates", depending on if it's + dBi or -dB) the average input power along a direction (θ,ϕ).
But physically (not from the definition convention), is this concept of antenna gain a function of time?
The time dependence of the transmitting power (due to the time dependence of the time domain Poynting vector) represents the electromagnetic power flowing out from the antenna, like shown in the following picture (see here to watch the animation):
So, is physically the antenna gain animated? Is it a time varying quantity?
The antenna gain is by definition time - independent (since it involves a frequency domain quantity,) but is the antenna ratio between instant transmitted power and instantaneous input power time dependent?