Bromwich and Evans*
suggest that this wonder is describing where the Severn
estuary is tidal and the river meets it in spate [causing] a whirlpool and a very
great noise
. They note that the description may be of flooding on
the English side of the Severn in Gloucestershire, noting Camden's*
description
(he actually claims to quote William of Malmesbury's De gestis pontificum Anglorum of 1125 CE
[biography])
of the bore wandering onto the plains of Gloucestershire.
However, they take the evidence of Culhwch ac Olwen to suggest the
mouth of the River Usk (Photo). The first evidence is a story about a giant Salmon, which may
represent the bore, who lives
at the (possibly) synonymous Llynn Llyw [*line 891] that he travels with the
tide upstream to Gloucester. The second is that the giant boar, Twrch Trwyth,
the chasing of which is a major part of Culhwch ac Olwen, runs
between Ystrat Yw (Ystrad Yw: area around
Tretower and Crickhowell*) and Euyas (Ewyas: eastern Black Mountains*), and heads down across the Severn between
Llynn Lliwan and Aber Gwy ("the mouth of the Wye") [*line 1179: See Map of chase], suggesting the two are close. The only factor working against this is
that the Usk (Wysg) has an old and well known name (see Williams*;
Isca, Ravenna Cosmography, 7th C, but also the town of Iscae in the
Antonine Itinerary of 2nd C, see
Other Works) and it seems strange this wasn't used in Culhwch explicitly if that area was meant.
However, other alternatives include the mouth of the River Taff: Marie Trevelyan*, in 1909, reports the following folklore...
The whirlpool of the River Taff at Cardiff forms a small lake when the bed is almost dry. In the days of old this whirlpool was called one of the "seven wonders of Glamorgan." People said it was fathomless, and in its cavernous depths a monstrous serpent dwelt, and gorged on the unfortunate victims that were drowned in the river and sucked into the whirlpool.
However, she gives neither source nor location for this. It was possibly near the mouth of the Taff, however, given the dock building there from 1839-1907, and the Taff diversion in 1853* the likelihood of this depends on whether her reference is contemporary or oral history. The Taff is not within the current area effected by the Bore, making it unlikely if the Bore is being discussed.
The Wye (Photo) also seems possible. The mouth of the Wye has a massive advantage in that it is the current location of the start of the Bore, and would therefore match with the Salmon of Culhwch ac Olwen being a embodiment of the feature and details of the Bore).
Another possible location is Goldcliff. The light shining on the local rocks at sunset makes them glow gold, hence the name* - if we hold with Bromwich's derivation of the name from "lake shining" (Note) we might draw a very thin link. Goldcliff has been a site of notable inhabitation since at least the Iron Age*, and there are considerable historical records for the area*, though none seem to suggest the use of the name Llyn Liuan.
Other suggestions include* Llymon Brook which runs from Cross Ash to join the River Trothy. This is actually a quite strong candidate, as we know the Trothy joined to the only historically recorded feature called Liuan from a boundary (c. 970 CE*; *p.241) in the Book of Llan Dâv (See Other Works). The name of the town near the mouth of the Llymon, Llanfihangel Ystern Llewern, and the naming of a nearby farm (on 1880's maps; 500m east of the river mouth on the B4233) as Pant-y-Pwll (Hollow of the Pool) are all suggestive.
However, it is hard to be definitive about the Llymon. One of the Latin boundary descriptions in the Book of Llan Dâv (c.780 CE*; for Cair Riov, *p.210; ) also appears to cover the area, and on the one hand it gives the Llymon's name as the latinated Liminan, which seems closer to the modern version, but on the other includes the "fount of Baraliuen" and mentions a pool, which seem indicative. Either way the mention of the Severn in our wonder description, and the location of the pool as the place where the Trwch entered the sea makes it very unlikely to be the location of our wonder.
The same argument also applies to Pwll Lifan, also on the Troggy (the text of the translations and parenthesised names are by Evans* p.228; dated c.876 CE*):
Eccl.a.GUEITHIRIN [LL.VETHERIN] From Oaper [Aber: mouth of] Pwll Lyfann on the Trodi (Trothy), along the Gwver of the Pwll straight to its source. From the source straight over the Cecin as far as the Cinluin. Along the Cinluin upwards as far as the boundary of Tref Pedir (Town of Pedyr) in the pant [valley/hollow] to the right. Along the pant as far as the Carn Litan (Garn Lydan) above the Allt (the Arwellt)[though Allt = hillside; cliff; wood]. From the Carn Litan, to the right, straight down as far as Rit or Euic (Rhyd yr Ewig: the Hind's Ford), on the Atguedauc (Atguedog). From the ford straight over the Cenin to the right as far as the spring of Colwyn. Along the Colwyn downwards as far as the wood. From the wood to the Crug. From the Crug along the wood as far as the Trodi, for the ford. From the ford, as the Trodi leads downwards, as far as Pwll Lifan, where the (boundary) began.
A few of these places are certainly reminiscent of locations at the upper end of the Trothy: Celyn (Cecin?) Arwellt (as suggested by Evans) Pant y Colyn (Colwyn?).
The location in the Book of Llan Dâv of the Gueithirin Trodi text with its Tref Pedir so close to the Taroci text with its Tref Peren (See Note) almost has one wishing the rest of the Trodi text was more ambiguous, but the identification of Gueithirin as Llanvetherin rather than anywhere on the Severn seems sound.
Another strong contender must be the parish of Ystrad Hafren ("Vale/Flat of the Severn"). The Book of Llan Dâv has the following boundaries (c.703CE*; *p.174 and and c.878 CE*; *p.229 respectively: translations and suggested names in parentheses are from Rees*, names in text are from the latin of Evans*) :
Its boundary is from the wood to the sea and as far as the podium (mansion) of Ceuid, and with all its liberty in field and in woods, in water and in pastures. Its boundary is, from the upper part of the grove Ili to the sea, and from Glasguern (Glasgwern) ["Green (/Blue?) meadow"] to Louern.
From the summit of the grove of Ili to the sea, and from Glasgwern to Longuern (Longwern) ["Meadow lane"?]
The mutation of Louern to Longuern is fairly hefty, but might allow for an initial Liuan.
Ystrad Hafren has been suggested as being Tidenham, just up the Severn from Caldicot, apparently on the strength of the suggestion that Podium Ceuid is Lancaut* (Book X). A Saxon survey of the manor of Tidenham gives the boundary as the Severn, the Wye, and then a line running from opposite Tintern Abbey to Oakhill, and then down to Horse Pill on the coast downstream of Stroat* (Book X). There are a number of advantages in this parish being identified as the wonder-site, including the fact that the name of a number of the features in the area (Pig Hole Pill; Sturch (Twrch?) Pill) would be resolved.
However, it is not certain that Ystrad Hafren is Tidenham, let alone the location of the wonder. For example, a further boundary from the Book of Llan Dâv (c.860 CE*; *p.173) includes Ystrad Hafren, thus (translation and suggested names/additions in parentheses are from Evans*):
LLAN CÚM The boundary is - from Oaper [Aber: "confluence"; "mouth"] Nant ["brook"; "gorge"] Bis on the Bis (Evans seems to suggest this should be Bic, in line with the end of the boundary), as the Bis leads upwards (throughout) its length as far as its source, to the road. From the road to the Ardd ["hill"; "height"], along the Ardd towards the west as far as the Cecin ("ridge"* [cefnen?]) of the Allt ["hillside"; "cliff"; "wood"], from the Allt downwards as far as Rhyd yr Onnen (Ash-ford) on Nant Broueni (Broveni), along it as far as Nant foss (ffos ["ditch"]) pluum [plwm? "lead"], along (it) upwards to Istrad Hafren. From the top of (Ystrad Hafren) as far as the Cecin, along the Cecin as far as the road, to the Ardd, along the Ardd as far as the licat ("spring")["eye"] of the finnaun ("well")["well-spring"], along (it) downwards until it falls into Bic. Along Bic downwards to Aber Bis, where the boundary began.
Given that everyone from Rees* onwards has identified Llan Cúm as Llangwm it isn't especially obvious why Tidenwell is particually suitable for Ystrad Hafren, as the obstacle of the Wye isn't obviously mentioned in the description. Indeed, "Flat of the Severn" would better describe the Gwent Levels than Tidenwell, if this is the appropriate translation. There is no obvious Bic/s brooks in Llangwm, and the boundary seems to place it outside Ystrad Hafren, rendering it unlikely to relate to Bishton (south east of Woodcroft). Equally it is unlikely to relate to Bishton near Magor, which is probably "Bishopstone", or, indeed, the report that a meadow just inland of Portskewett was known as Bisditch in the 17th C, as in the 13th it was "Biestediche" (Beast Ditch: see Note). Judgement is generally reserved about the possible Louern reference within the Ystrad Hafren boundary description until more information on the Bis/c and other features comes to light.
Pughe* has under “Llion”: "Llyn Llion – a small pool at Chepstow which ebbs and flows contrary to the tides.” Sadly, all of Pughe’s details for “Llyn Llion” seems to derive from the unreliable Iolo Morganwg (see note) and there are apparently no other records of the pool being at this location. Bradney*, who finds three of the Wonders near Chepstow, doesn't mention it (the nearest being the place livurnell, 1683*). It is possible that Pughe’s source was thinking of the ebb-and-flow well at “Woodfield” (a house near Chepstow bridge)* - however, although this did act in opposition to the tide, it is hard to imagine it inspiring the violent description of the wonder.
Finally, the Severn Estuary has changed significantly over the period since the 'Dark Age'*, with some 111.20km2 of land reclaimed on the Gwent Levels* (between Cardiff and Caldicot). Most of the Gwent Levels are below the current maximum tide at 4.5 to 6.1m (mean spring tide high water 6.2-6.4m OD, though frequently as high as 7.3m), and would be flooded if not for the sea defenses*. However, there has also been considerable silting - for example, Magor Pill [possibly once called Abergwaitha* "Mouth of the Works"*], Collister Pill (Map) and Redwick (Map) have largely silted up**. So it is of course possible this site has either been reclaimed or lost.