Hi Gerry and welcome to UKAPS.
I don't think it's a good idea to put aqua soil on top of gravel...it'd probably end up a mess...on its own would be fine, or aquatic compost capped with gravel yes
http://www.ukaps.org/forum/threads/the-soil-substrate-or-dirted-planted-tank-a-how-to-guide.18943/.
Regarding compatibility of an UG filter - I've gown plants very successfully in soil over an UG filter. I used soil substrate over interlocking UG filter plates (similar to these
http://www.allpondsolutions.co.uk/a...tank-filter-for-aquariums-23-x-12-uf-16p.html) powered by large powerheads. The soil - a layer about 40mm thick - was sandwiched between 2mm grade plastic mesh, which held it firmly in place and capped with a couple of cm of sand. At the time - as probably now still - it was definitely regarded as a no no if you wanted to grow plants successfully.
I can only speculate why it worked so well...but I've an inkling the success was largely to do with vastly increased flow and distribution over pant roots perhaps similar to the way hydroponics works. After a while there were definitely a lot of fibrous roots penetrating the void below the soil. Incidentally, it was the only means of filtration in a very well stocked and extremely healthy tank, which ran for years with very few water changes and without any trouble whatsoever.
Also I think the under-soil flow and distribution of O2 would most likely have created another oxidised microzone in addition to the one at the soil surface. Just to put that in perspective, the oxidized microzone is the very thin highly aerobic surface layer of soil that interfaces with the water column. It is of huge importance in a soil substrate tank since it supports the hive of microbial activity needed to neutralise toxic substances and unlock nutrients. This combined with the associated increase in flow and distribution also probably helped to release more CO2 from the increased decomposition of organic matter in the soil.
The soil was mostly moss peat; its fibrous nature goes a long way to prevent clogging of the filter plates. Plus I doubt that there was much, if any flow, through the actual soil itself - it'll be too densely packed - most of the flow was from around the edges which were composed of gravel or sand, again preventing soil from clogging the plates. It's this flow that was important in producing the theorized double microzone.
Anyway, probably more information than is needed or healthy but, I had an urgent need to share
😀
