Ionospheric D Layer absorption attenuates radio signals, mainly below 10 MHz. Here is how it is caused and when.
In our ionosphere, the lowest formal layer is D, roughly between 60 to 90 kilometers above earth. Because solar radiation is weakened as it gets closer to ground, the amount of ionization is much lower than in the higher E and F layers. As you can see above left, free electron density in the D layer is well below the F layer, by a factor of 106.
Radio waves move through ionosphere by pushing very light free electrons around. At higher layers, the more electron density, the more signal refraction at higher frequencies. Hence, the MUF. So, as hams and SWLs, we focus mainly on the F layer as the enabler of long distance propagation.
With its much lower electron density, the D layer does not refract signals above 250 kHz. It just absorbs. And, it only exists during daytime, unlike the higher layers where electrons and ions recombine much more slowly, if at all.
Also, at lower altitudes, our atmosphere has much higher molecular density. You will find that this results in a high rate of collisions between free electrons, ions, and particularly neutral particles, i.e. molecules. As shown above left, in black, the collision frequency increases significantly below 100 kilometers. D layer is a sweet spot with somewhat high electron density and really high collision rates. (These collisions basically turn radio signals into heat.)
To summarize, most HF radio emissions just pass through the D layer on their way to E and F, with a bit of loss.
How Much D Layer Absorption Do We Experience
D layer absorption depends on electron density, collision rates and signal frequency. Your basic formula is κ=1.15E-2•(Ννƒ-2). In this formulation, κ is dB of absorption per kilometer, Ν is electron density, ν is the collision rate and ƒ is the frequency. If you plug in typical values of N and ν for the D layer, you get the result shown above right between 2-15 MHz. Calculation as follows. Ν=1E9 electrons per cubic meter. ν=1.2E8 collisions per second. ƒ=5 MHz. ∴ κ=5.6 dB attenuation per kilometer travelled in the layer.
As you can see, the absorption is very high for medium waves, and lower HF, especially the 160 and 80 meter ham bands. But as this absorption declines by the square of frequency, it is pretty much gone by 20 meters.
Now, it is very easy for a signal to spend 75 kilometers inside the D layer for each hop. So, for a signal at 5 MHz spending 75 kilometers inside D layer absorption, total signal attenuation of more than 100 dB is easy for us to imagine. Of course, this loss goes away at night when the D layer disappears.