Hi Kevin, welcome to Lightroom Forums!
What you have discovered is noise!
In digital imaging two overall types types of noise exist:
Luminance noise, and colour noise.
Luminance noise is the result of several factors, the most ubiquitous being the quantum nature of light - simply put the amount of light hitting a particular sensel is not exactly the same as the amount of light hitting an adjacent sensel even though the source of that light has the same intensity and colour. The easiest way to appreciate this is look at an image with a lot of blue sky at 1:1 in Lightroom - you will see a slight speckled effect.
Colour noise has a different genesis. Sensors do actually record light as colour, merely the intensity of light. However, immediately above the sensor in your camera is a filter called a Bayer array. The Bayer array is a patchwork quilt of coloured filters in a specific specific pattern of two green for every one blue and red.
In order for your camera, or a raw converter (such as Lightroom) to convert a grey-scale image to a colour image a process called demosaicing is used. There are many different algorithms but the essentials of the process requires that in order to deduce the colour that any particular sensel should display requires the sampling of adjacent sensels so that the full RGB colour data can be assembled. Colour noise is only one of artefacts that can be introduced into an image as a result of the demosaicing process.
Both colour noise and luminance noise are more obvious in the shadow areas of an image. The reason why is simple: signal-to-noise ratio (S/N ratio). What this means is that if the signal generated by light hitting the sensels is not much more than the variations introduced by noise (whatever the cause) then the visual effects of the noise become easier to see.
Noise, in all its forms, is present in the highlight areas of an image as well but, because the S/N ratio is so high it is much less obvious visually.
Lifting the shadow areas in Lightroom (or any other raw converter for that matter) is just a form of electronic amplification. Noise is amplified along with the true signal and therefore becomes more obvious than before.
This is consistent with your very own observations.
How to deal with noise?
The process needs to start in-camera.
Shoot in raw. Derivative images have limited scope for dealing with noise reduction.
If one shoots raw then employ a process called ETTR (Expose-To-The-Right). ETTR simply means employing the full capabilities of the camera sensor. What one does is to expose an image such that the brightest parts of that image where detail is desired should be recorded as highlights just short of actual clipping. An image shot in this fashion will often look washed out and over-exposed when first visualised but this can be easily dealt with in Lightroom by reducing the exposure to restore the correct tonal relationships. The advantage is seen in the shadow areas of the image where, because the S/N is greater noise becomes much less obvious. Also, better detail, and colour, is obtained in those shadow areas as well.
ETTR is predicated on the basis of shooting at the base ISO of the sensor.
Get light onto the sensor! ETTR is a variation of this principle, but not its only application. In low light situations use fast lenses wide open. Use a tripod to allow longer exposure than would be possible hand-holding a camera.
Shoot at the lowest ISO possible. Boosting ISO is just amplifying the electronic signal, along with the noise!
Late model cameras have a much lower noise floor than older cameras. The latest sensors in these cameras reduce the electronic and heat-related components of noise generation dramatically. However, shot noise (explained above) is purely a function of the physics of light and cannot be eliminated by electronics. Nonetheless, the term "ISO invariant" has been coined to describe the behaviour of the best sensors. In these cases there will be no more visible noise in an image shot at ISO 100 compared to ISO 800 as long as the exposure (shutter speed and aperture) remain the same, once the images are normalised in a raw converter.
The size of the sensor matters! As a general rule the bigger the sensor the less the noise that is generated.
Dealing with noise in post-processing:
Noise reduction in Lightroom and other raw converters is much better than it used to be.
In order to understand and employ noise reduction logically it is crucial to realise that sharpening and noise reduction are two sides of the same coin. In general terms both noise reduction and sharpening algorithms manipulate fine detail.
What this means is that increasing the amount of sharpening will tend to make noise more obvious and increasing the noise reduction will reduce the fine detail in an image.
Therefore, because sharpening and noise reduction have opposing effects, neither should be applied in isolation to the other.
With specific reference to your situation, I am not sure whether I am looking at the entire image or just a small section of it. What has been posted is markedly underexposed. What actually could be done with it depends on whether this is the whole image or not. What is clear is that there is no detail in what is posted and so boosting the shadows only has the effect of accentuating noise. If there is more to this image re-post the whole image. It may even be an option to send me the raw image so that I can directly view it and develop it in Lightroom.
I will continue the post once I know more about the image.
Tony Jay