Well Freddie you may well joke re a container at the base of the mast and its cost to make those alterations! But the one thing you are wrong on is the cost of the modifications vs the cost of energy needed to run the extra power from 4 to 10kw as it would very soon out way the cost of those mod's.

Naturally I don't know how much this would help for certain as this would depend on the level of loss as it stands! But one thing is for sure it would go up! And very likely arrive at the level of increase in erp equal to the move from 4 to 10kw so it speaks for its self cost wise. Again if it was me I would be looking at both and possibly impalement both at the same time.

90% of all the ILR stations of 70's and 80's ran little more than a few hundred watts but covered a good amount of land with an appropriate signal, ech of them had the TX shed, room, container at the base of the mast somewhat similar to the average setup on all the radio ships did and together with the sea acting as a fantastic earth and ground plane / radial it couldn't fail.

You mentioned 390 and its strong signal well this was / is a classic case showing how effective it can be to have a totally matched low loss antenna and grounding 390 had a maximum of 10kw available but only ever ran 8kw at most! And would reduce this to around 4kw by night as not to cause interference to distant usage of the frequency.

Radio City was another example of perfect antenna length and matching as it was only ever a maximum of 2kw but most of the time ran only 1kw, ok it wasn't as strong in London as the likes of BIG L but considering it was only very low power it was a blinding erp. So as you can see and as all of the offshore books will show its not about the power so much as having a perfect match and ground at the mast end.

My own personal experiences running both AM and FM over many years would defiantly lend its self to the later ie antenna and ground efficiency over power 100% all of the time.

Garry Stevens