As always, the grounding threads are quite interesting, and this one is over a year old. But as somebody else pointed out, do not confuse RF grounds with safety and DC grounds. They are two different things. Safety grounds never carry current except in the instance of an electrical fault. RF grounds carry current, sometimes at very high voltages, and are never to be bonded or connected to safety grounds. This is despite what some inspectors who know nothing about radio will try to claim.
The purpose of safety and DC grounds for lightning are fairly self explanatory and NEC outlines the requirements well.
RF grounds are harder to understand. Earth ground is not ground to RF. RF flows in the earth readily, although earth presents high impedance to RF current. Consider an electrically "complete" balanced antenna like a dipole. It has one element that is grounded, the other is "hot". If the dipole is not resonant and tuned up with a matching network with high standing waves on the feeder, the ends of the dipole alternate back and forth at very high voltage, one end to the other, alternating at the frequency of the radio signal. The grounded element of the dipole is connected to earth ground at the generator (transmitter) because the DC ground of the radio is integral with the RF signal generator's ground. But the RF signal the transmitter generates is like any AC circuit and in order for it to be complete, like any AC circuit has to be in order for it to work, the RF current must return to the generator. If you shunt the RF current from the grounded element of the dipole to earth ground via a ground rod that is also bonded or connected to your DC and/or safety grounds, you just created a ground loop. You provided an alternative low-impedance path for potentially high-voltage RF current to return to the generator by flowing over all your equipment that is connected to that safety ground rod.
With vertical antennas, they will normally be mounted on a metal mast or tower, which is DC grounded for lightning protection. A small vertical can have an elevated ground plane, which mirrors the other half of a dipole to make it electrically complete. The ground plane of this vertical will be normally connected to earth ground because it is connected to the metal mast or tower. So it is safety grounded for lightning.
Ground mounted verticals, like my 72 foot free-standing tower which is loaded up as a 160m Vertical Mast Radiator, can be either grounded or ungrounded. In the case of my tower it is grounded. In the case of AM broadcast towers they are usually ungrounded with an air-gap safety ground for lightning. The air-gap safety ground consists of a couple balls - one connected to the insulated Vertical Mast Radiator, the other connected to earth ground - and when the potential on the tower reaches dangerous levels due to EMP from direct or nearby strikes the spark jumps the air-gap on the two balls to earth ground.
So how can a grounded tower be loaded up as a vertical radiator, like my 72 footer? Because of the principle I noted in the first paragraph - earth ground is high impedance to RF but not a dead short like it is for low frequency AC power or lightning (DC). So I can load that tower up with a slant wire going part way up it and make it radiate just like your elevated ground plane vertical. In order to make the vertical mast radiator electrically complete, the RF current flows in the earth back to the generator. To provide a lower impedance path for the RF current returning to the source I have 90 ground plane radials, each 130 feet long buried in the earth about 1/2" deep. These are all bonded together at the base of the tower to a ground rod that is separate from the tower lightning grounds, and from there back to the generator (transmitter). This RF ground system can NOT be connected or bonded to the safety grounds of the system. Why? Because it has very high voltages on it and carries current as part of the normal operation of the circuit. When my loaded tower is transmitting there is thousands (around 6,000 volts) at the base of the tower, even though it is grounded. If somebody would touch it, it would ruin their day big time. So I have to have a safety fence around the tower base to keep people and critters away from it.
How can that tower have high voltage on it when it is grounded? Because the tower itself becomes a capacitor when it is loaded up at RF, providing series capacitance in the antenna system. It is in essence a 144 foot long folded dipole "hot element" with the other half of the dipole in the ground plane. The impedance presented by the lightning ground rods (six of them) is a dead short to lightning, but very high to RF. If I did something stupid like connect the RF ground system of that mast radiator to safety grounds at the shack, I now have thousands of volts flowing across all my equipment. I provided an alternative low-impedance path for RF current to flow back to the generator instead of returning on the relatively high-impedance path of the feeder. My feeder "ground" is still grounded - at the generator, but not before. The system I have will not work with the RF ground plane bonded to DC grounds any better than if you take an elevated ground plane vertical and put a jumper wire on it from the radiator to the ground plane elements.
If you want a system where the RF ground plane is connected or bonded to the base of the radiator (tower) then you have to use a different method of loading called a gamma match. The gamma rod forms a capacitor and drives the vertical radiator thru capacitive coupling. The gamma would be ~40 foot long on my 72 foot tower for 160 meters, and is not practical. So I chose a (quite common) method to load it, used in the AM broadcast industry for better than 70 years, and also used by many hams, to drive it. But the method chosen involves very high voltages.
This is where broadcast radio, and some applications in ham radio, part ways with NEC. NEC says to ground everything to one point and then have thousands of volts on your equipment that will knock you clear across the room when you key down the transmitter. Although there is a couple parts of NEC that deal with broadcast transmitters, they do not deal with RF grounds at all and are largely slanted towards residential and business/commercial AC grid power safety and lightning grounding.
So there's an easy rule of thumb you can use to ground any station that is also a transmitting station (not just SW listening). Draw out a schematic of your ground wiring. Then study it. If you can find more than one path for the RF part of the circuit to return to the transmitter then you have a ground loop. And the equipment in your station is going to be hot from common mode current flowing on it. That RF ground is already grounded at the transmitter and it is the only ground it can have on systems running thousands of volts on the antenna. On lower voltage systems you can ground it externally to a safety ground rod and nothing bad will happen, other than you still have a ground loop. But if you someday put in a high-voltage antenna system like my loaded tower, you will have to learn (hopefully not the hard way) about the difference between RF and DC/low voltage AC grounds.