Cave Hydrology

Hydrology of the Peak Cavern system (By John Cordingley)

Catchment

The area comprising the underground watershed of the Peak Cavern system, west of Castleton, is the most northerly of the catchments in the White Peak karst. It is about 15 km2 in area and is well defined from dye tracing work, except in the south west where the exact position of the watershed is less certain. The Peak Cavern / Speedwell Cavern / Titan / JH system is the longest known cave within the area but the explored passages only occupy a small part of the eastern corner of their catchment. To the south east is the Bradwell hydrological system and there is at least some evidence that water sinking in part of the Peak Cavern catchment may also resurge at Bradwell, suggesting the possibility of a connection between the major caves in each area. The Peak system area contains the greatest length of explored cave passage of any of the catchments within the White Peak.

Inputs

The Peak Cavern system is fed by a number of discrete sinks distributed along the northern boundary of the catchment (where the lower Carboniferous limestone dips beneath the slightly younger shales and sandstones (of Namurian age) which form the elevated ridge known as Rushup Edge. The impervious Namurian caprock act as the gathering ground for streams which generally flow southwards; when they encounter the limestone they sink almost immediately. These sinks of “allogenic” water along the northern outcrop of the White Peak limestone have collectively been described as the “Peak Fault Swallets” but there is no evidence of geological faulting to explain their distribution. The sinks have traditionally been listed in terms of numbered “points of engulfment” as well as being given cave names if passages have been entered from the sinks. For example “P8” was considered as sink number 8 (going eastwards) and when it was opened up by digging the cave discovered was named “Jack Pot”. (However, in this example, cavers also refer to the cave as “P8”). It is now known that there are more sinks than those originally considered worthy of the “P” prefix. Cavers have needed to be slightly inventive in naming these, for example the sink (and cave later explored) to the west of “P1” was called “P Zero”).

Further to the south, most of the rain falling onto the limestone upland which eventually finds its way into the Peak system sinks directly into the ground without first forming surface streams. This percolation water (or “autogenic” water) accounts for a significant proportion of the discharge flowing through the Peak and Speedwell Caverns. (Note that the good visibility celebrated by divers in some of the sumps within the Peak system is due to the high content of “filtered” autogenic water.)

Resurgences

All the water flowing through the Peak system emerges to the light of day in the Peak Cavern gorge, before draining northwards into the Hope valley. The Peak Cavern stream normally surfaces at the (sumped) Peak Cavern Resurgence against the east wall of the gorge, just outside the huge Peak Cavern entrance portal. In extreme flood some of the Peak Cavern stream (augmented by some of the Speedwell water – see below) flows directly out of Peak Cavern itself. In normal conditions the Speedwell stream resurges from two springs; These are Russet Well (on the east side of the resurgent Peak Cavern stream) and Slop Moll (west of the Peak stream and close to the path leading up to Peak Cavern). The former is an immature resurgence of the Vauclusian type (associated with a minor mineral vein) and has been explored by cave divers to a depth of 25 m. Although “immature” from the perspective of its age compared with the rest of the cave system, a vast amount of water rises here in flood, creating a pressure dome up to a metre high. The nature of Slop Moll is not really known as the water rises (sometimes in great quantities) through chokes in the gorge floor derived from a combination of accumulated scree from the rock walls and debris from mining activity. However not far upstream of the Russet Well and Slop Moll resurgences there must be a bifurcation on the main Speedwell conduit which feeds them, as dye placed in the Speedwell stream emerges from both and the water chemistry at both is very similar.

There are also various mined passages resurging into the Hope valley which probably carry local autogenic drainage from artificial subcatchments within the the main Peak catchment (e.g. Pindale Sough, Peakshole Sough, Odin Sough etc) but their hydrological relationship with Peak and Speedwell Caverns (if any) is not really understood. The catchment for the Peak system does not have any good examples of warm springs such as those found at Buxton or Matlock Bath, even though there is one small thermal spring close by in the Bradwell catchment (Eden Tree Hot Spring just to the north of Bradwell village).

In exceptionally rare ultra-wet conditions the combined flow from the various risings in the Peak Cavern gorge is spectacular. The combined Peak system outlets during a flood event on 21st December 1991 submerged parts of Castleton which was impassable to road traffic for several hours. The middle of the village turned into a lake with water up to the bottom part of the windows of the “Three Roofs Cafe”close to Peakshole Water. In such conditions a large proportion of the known part of the Peak system would be flooded to the roof or carrying water moving at such speed that any visitors would immediately be drowned. It goes without saying that the flood risk in this cave system should always be borne in mind (see below). However in dry conditions the only reasonable sized stream in the system is that flowing is the Speedwell main streamway. Far Sump may stop issuing any water for months at a time, despite a permanent flow into the upstream end. There is an unidentified outlet somewhere in Far Sump which drains to Main Stream Inlet in Peak. Continual leakage of water from Far Sump in very dry conditions thus allows the level of the sump to fall by several metres and the length to shrink from 385 m to 240 m. (The normally submerged feeder within Far Sump near its upstream end, known as Far Sump Inlet Sump, has been dye traced from the second excavated shaft in the Titan dig but most of the flow probably comes from the upper Cavedale area.) In the severest drought Ink Sump also becomes static but within the choke beyond, in Doom's Retreat, gurgling water can still be heard. The ultimate destination of this audible streamlet us unknown (but may reappear from Buxton Water Inlet Sump further downstream in Peak).

Underground flow between the sinks and the Peak system

The general west to east flow of karst waters towards the Peak Cavern system is now well understood and is greatly influenced by geological controls. Only the northern boundary has well developed sinks of discrete streams because of the adjacent impervious rocks along Rushup Edge. The autogenic nature of underground water to the south of the main line of sinks is typical of limestone with no nearby impervious caprock. The Castelton catchment has a large number of west to east mineralised minor faults (locally referred to as “rakes”) and it is generally believed that most of the water flowing from west to east is collected then guided eastwards by these veins. The more westerly sinks (such as Perryfoot Cave) are captured by Coalpithole Rake but further east they leave this to appear at Speedwell's Main Rising, which is associated with the more northerly New Rake. The central and easterly sinks such as Giants Hole drain via New Rake and / or Fawcet Rake to get into Speedwell (with a bizarre flow switching system at intervals; see below).

The majority of the autogenic drainage arriving in Peak Cavern emerges either from Far Sump or from Ink Sump. Little is known of the route of the former but the Ink Sump stream is likely to be derived from flow along the nearby Dirtlow Rake. One unusual feature of the White Peak karst is the existence of lava beds and / or clay “wayboards” interbedded with the limestones. They act as “aquicludes” (a bit like a builder's damp proof course within the limestone). This sometimes prevent downward flow within the limestone resulting in springs well above base level, such as the one near the head of Cavedale. The absence of stalactites in large areas of Peak Cavern has also been attributed to a lava bed “umbrella” stopping autogenic trickles from getting into mature passages. Where impressive displays of stalactites do occur (such as in the White River Series) the lava aquiclude has been breached by mineralised faults.

Flow within the Peak system

The water arriving in the passages of Peak and Speedwell Caverns and Titan from the various inputs described above generally flows in discrete conduits (“cave passages”) as in most other British cave systems. There are two main streamways which largely follow two separate bedding planes. The higher of the two is the main Peak Cavern bedding plane (easily visible in photographs of the Peak Main streamway). This can be traced all the way from Far Sump Extension eastwards as far as the Five Arches near to Peak Cavern entrance. The main Speedwell streamway bedding plane is at lower altitude; this bedding plane is seen all the way from Main Rising and Whirlpool Rising eastwards as far as the current limit of exploration in the downstream sump. Water flow in the streamways is influenced by a general eastwards or north eastwards dip of these two bedding planes (but with localised variations in direction and steepness of dip). Both streamways were inititated and substantially developed under phreatic conditions but sections of both later suffered extensive vadose modification in response to the lowering of the Hope valley by ongoing erosion. In this way the magnificent keyhole shaped canyon between Far Sump and Surprise View in Peak, together with the western end of the Speedwell streamway, developed over a long period. Further east in these streamways (and nearer to present day resurgence level) there has been little vadose modification, leaving fine phreatic tubes such as that in the Peak main streamway just upstream from Buxton Water Sump.

The origin of the water arriving in the Peak system from the great shafts such as Titan, Leviathan and Cliff Cavern is not known. None of these is fed by discrete sinks which makes traditional dye tracing very difficult. Each must receive autogenic flow from a fairly large area because huge volumes of water cascade down in wet conditions, sometimes making these shafts impassable. The water falling down Leviathan in JH formerly flowed via a sump at the bottom (originally choked with mine tailings) into Far Sump Extension, arriving at Stemple Highway Inlet Sump 1 and then sinking nearby to reappear from an obscure inlet within Far Sump (240 m upstream of the Peak Main Streamway). This water was captured by a dam on a ledge high up in Leviathan in the early 1990s and is now piped away. It presently flows into the Speedwell streamway at the Boulder Piles. The Titan streamway water is lost in chokes in the floor of Titan itself but this has been shown to flow to the waterfall inlet in Far Sump Extension at the upstream end of Far Sump. By contrast the Bottomless Pit in Speedwell has no high level inlet and the water falling into the choked sump pool at its base mainly originates from the Speedwell streamway. (This has only been the case since the late 18th Century when the Speedwell level was mined, allowing water from the streamway to flow towards the Bottomless Pit.) This water rejoins the rest of the main streamway water in the sumped conduit between Speedwell's downstream sump and the risings in the Peak Cavern gorge (as does water tested from Blue John Cavern nearer to Mam Tor).

In several parts of the Peak system the water flows through large vein related cavities. Some formed under deep phreatic flow, long in the past (e.g. Stemple Highway in Peak's Far Sump Extension or the Bottomless Pit in Speedwell). These have long been drained and are therefore accessible to cavers. Other vein related cavities are still actively forming under deep phreatic conditions, such as the Main Rising sump in Speedwell (explored by divers to 74 m depth). In some of these vein related cavities there is a marked increase in size underneath certain bedding planes. Such bedding planes often have thin shales or wayboards which contain iron pyrites. The pyrites oxidises to produce acidic products which greatly helps the water dissolve the limestone underneath. Good examples occur in the upper parts of Far Sump Extension and also in Joint Effort passage at the top of Cliff Cavern in Speedwell. (These are similar in nature to others in the Castleton catchment such as the bigger development below the Flats in Nettle Pot.)

The Speedwell streamway has perhaps some of the most interesting hydrological aspects of any British cave. These include the abrupt switching of major flow between Main Rising and Whirlpool Rising (normally during a big flood event) and also the remarkable siphoning effect seen at both these risings in high flow. The bulk of the water flowing in the Speedwell streamway comes either from Main Rising or from Whirlpool Rising. When the main flow suddenly changes from one to the other the one which has just lost the main flow suffers a sudden reduction in the volume of water emerging until the flow switches back again (months or even years later). When the celebrated Whirlpool Rising is siphoning the period in between the main peaks of flow might typically be around ten minutes. Under optimum conditions these pulses can arrive so quickly that they catch cavers unaware and can be quite frightening. When Main Rising is siphoning the pulses cause water to back up against the Boulder Piles restriction, forming a lake which can be seen to rise and fall by 2 or 3 metres. Further down the streamway these pulses suddenly cause water to start flowing where minutes previously a passage was dry (the best example being at Overspill Passage). The magnitude of the pulses can be so great that they are easily seen outside the system at the risings in the Peak Cavern gorge. Flood waters appearing in Lumbago Walk also show a marked siphoning action but it is not known whether this is a reflection of the pulses further upstream in Speedwell or whether it is due to an independent siphoning mechanism.

As a general principle the Speedwell streamway normally carries mainly allogenic water, largely derived from the Rushup Edge sinks whereas the bulk of the Peak main stream is mainly autogenic flow. This is known from dye test results, from studies of the water chemistry and from the “peaky” flood hydrographs associated with Speedwell compared with the slightly more smoothed out changes in flow seen in Peak Cavern. However, in flood conditions, things become rather more “interesting” for reasons explained below.

If you walk down the Speedwell streamway it's obvious that the passage is very mature to begin with but development is lost in various places such that the passage near the downstream sump is quite small (hands and knees crawling in size). Some of these outlets are now fossil passages, such as the upwards phreatic development which disappears into the Boulder Piles. The Speedwell streamway is much bigger upstream of this point than downstream because for a long time water must have flowed under phreatic conditions up into what is now the bottom of JH's Leviathan, then on via Stemple Highway to be lost in the floor of Salmon's Cavern (and who knows where after that?). However some of these outlets are still active in flood; they convey part of Speedwell's flood water into Peak Cavern via various routes and lead to catastrophic flooding, sometimes even resulting in the temporary closure of the Peak show cave in extreme conditions. In downstream order the most important of these flood outlets feeding into Peak are:

1. Treasury Sump
2. Speedwell Pot
3. an unexplored route from within Speedwell's downstream sump to Lumbago Walk

In the 1960s Speedwell Pot in Peak was capped with concrete. This was blown off in a subsequent flood. A second stronger steel and concrete cap was then fitted, which is still in place today. This famously resulted in vast amounts of water suddenly erupting from the garden of the cave custodian's house during an extreme weather event soon afterwards. The main purpose of the Speedwell Pot cap was to try and control flooding of the Peak show cave but it met with only partial success. This is because the water is now forced to back up higher in the Speedwell streamway until it overflows instead via Treasury Sump and Treasury Chamber. However this has proved useful to cavers as Treasury Sump used to be blocked with extensive gravel deposits but since the 1960s enough gravel is shifted in flood to keep the sump open most of the time, maintaining access to the Speedwell streamway from Peak via a short dive.

Preglacial hydrology

In the distant past the Hope valley floor was significantly higher than it is today. It is thought that what we now know as Peak Cavern's entrance was then at the bottom of a huge Vauclusian rising. Downcutting of the Peak Cavern gorge in response to lowering of the Hope Valley floor by erosion probably drained this rising leading to the progressive draining of the deep phreatic main conduits in Peak And Speedwell Caverns (as mentioned above). This ancient Vauclusian rising was probably more complex than just a single spring. At least one other shared rising must have existed, fed by the large ascending conduit seen in the Mendip Beer Monster's Secret Tap Room (in the roof of The Vestibule). Another large ascending phreatic conduit is known high in the Krypton Series (also in the Vestibule roof) which probably fed another nearby spring at one stage.

It seems likely that some of the water once supplying this former gigantic resurgence complex originated from major sinks west of Castleton which are now long abandoned. Eldon Hole, Windy Knoll Cave, Blue John and Treak Cliff Caverns all carried great volumes of allogenic water in the past and are likely to have contributed to the unusually large scale of development seen in Peak Cavern's entrance and gorge. It is tempting to suggest that some of this earlier drainage from the Treak Cliff Hill area may once have been conveyed into Peak via the fragment of phreatic conduit seen in Suicide Cave in the Winnats Pass (but there is little known evidence to substantiate this at present).

Flooding in the Peak system and caver safety

In many British caves heavy rain results in a fast response as underground streams rise. In the wake of a flood the flows normally then fall fairly quickly and return almost to their pre-flood flows within a few days. The water in the Peak system can certainly rise impressively fast after heavy rain but only in certain conditions, depending on the time of year – or more correctly, the distribution of rainfall within the months preceding the flood event in question. Another factor influencing flooding characteristics is the growth of surface vegetation (which is of course also dependent on the seasons of the year); actively growing plants have the capacity to absorb large volumes of rainfall in late Spring and early summer, thus preventing it from reaching cave systems. The fact that a good proportion of the water in the Peak system is of autogenic origin is also worth bearing in mind as this greatly influences the flow characteristics of the inlets which combine to form the main streams in Peak and Speedwell.

The above is easiest understood if one considers two different periods of the year; summer and winter. In a long dry summer very little water resurges in the Peak Cavern Gorge. In such conditions a major weather event (dropping, say, 30 mm of rain) may have virtually no significant effect on flows either in Speedwell or Peak. In the winter (especially after a wet Autumn) the flow emerging from the Peak system resurgences may remain high for long periods. In these conditions even a minor weather event putting down 10 or 15 mm of rain can bring the water up further to levels which become dangerous to cavers. It therefore follows that the flood risk in the Peak system depends at least in part on what time of year you visit.

However the other aspect to this flood risk is caused by the connections between Peak and Speedwell caverns mentioned above. Because there are several potential flood overflows from Peak into Speedwell there are sometimes very sudden increases in the flow in Peak. Also, because Peak carries a greater proportion of autogenic water than Speedwell, the response times of these two streamways varies. (As a general rule the maximum flow during a flood occurs earlier in Speedwell than in Peak.) It follows that the profile of a flood in the Peak system (and particularly near the entrance to Peak Cavern itself) can be very complex and unpredictable. Add to this already complex situation the possibility of further rainfall as well and it becomes clear that accurate assessments of whether the cave will flood may be virtually impossible.

In many other British cave systems it's safe to go in when the water is falling and the forecast is dry; the worst that can happen is you get to an obstacle where high water stops you, then you just come out. This simple approach to flooding in the Peak system isn't good enough. There have been many instances of cavers being caught out, even when they know the system well. On certain noted occasions this has led to cavers being trapped by ponded water (when the Five Arches sumps) for long periods. Even worse than the scenario of simply being trapped in the system is the thought of being in a section of fast water. When Hurricane Charlie struck the UK in late August 1986 the first cavers to visit Peak afterwards noticed flood foam over 3 metres up the streamway walls near Surprise View! Don't take chances with the water in the Peak system; flooding is often fast, unforseen, very serious and potentially long lasting.

Flooding Observations in the Peak Speedwell System: