Professional Documents
Culture Documents
I Volume 10
Number 2
BeRA
June 1983
Gunung Sewu
Java
Scallops in
Norway
Peak
Speleothem
Dates
Articles for publication in the Transactions may cover any aspect of speleology
and related sciences, such as geology, geomorphology, hydrology, chemistry; physics,
archeology and biology. Articles on technica l matters such as caving techniques, equipment,
diving, surveying, photography and documentation are also accepted for publication as well
as expedition reports, historical and biographical studies.
These notes are intended to help authors to prepare their material in the most advantageous way so as to expedite publication and to reduce both their own and editorial labour.
It saves a lot of time if the rules below are followed.
All material should be presented
in as close a format as possible to that of the Transactions. Text should be typed doublespaced on one side of the paper only.
If typing is impractical, clear neat handwriting
is essential. Subheadings, sectional titles etc., within an article should follow as far
as possible the system used in the Transactions.
In any c ase, they should be clearly
marked, and a system of primary, secondary and tertiary subheadings , if used, should be
clearly indicated and double-checked before submission.
All material should be accompanied by an abstract stating the essential results of
the investigation for use by abstracting, library and other services.
References to previously published work should be given in the standard format used in
the Transactions.
In the text the statement referred to should be followed by the relevant
author's name, the date, and sometimes page number, in brackets. Thus: (Bloggs, 1999, p. 99).
All such references cited in the text should be given in full, in alphabetical order, at the
end. Thus:
Bloggs, B. 1999. The speleogenesis of Bloggs Hole. Bu lleti n X Caving Assoc.
vol. 9, pp 99-199.
Books should be cited by author, date, title, publisher and where published.
Periodical titles should be appreyiated in World List of Scientific Periodicals format if
possible.
Acknowledgments:
dnyone who has given a grant or helped with the investigation, or the
preparation of the article, should be acknowledged briefly. Contributors in Universities and
other institutions are reminded that grants towards the cost of publication may be available
and they should make appropriate enquiries as early as possible.
Expedition budgets should
include an element to help publication, and the editor should be informed at the time of
submission.
Illustrations: line diagrams and drawings must be in BLACK ink on either clean white
paper or card, or on tracing paper or such materials as kodatrace.
Anaemic grey ink and
pencil will not reproduce~
Illustrations should be designed to make maximum use of page space.
If photo-reduction is contemplated all lines and letters must be large and thick enough to
allow for their reduction.
Letters must be done by stencil, letraset or similar methods,
not handwritten . Diagrams should be numbered in sequence, Fig. 1, Fig. 2, etc., and referred
to in the appropriate place in the te xt by inserting (Fig.l) etc., in brackets. Captions
should be typed on a separate sheet if they are not an inherent part of the diagram.
Photographs are welcome.
They must be good clear black and white prints with sharp
f o cus, and not too much con trast.
Prints about 15 x 10 cm (6 x 4 inches) are best.
Experienced
authors may make up their complete photo pages (Plate s ) with captions printed or electro-typed
in, but other authors sh ou ld lightly pencil the photo number on the back, type the caption on
a separate sheet and indicate in the text the point where the photo is referred to: Thus:
(Pho t o 1 ) etc.
If any text, diagrams or photos have been published elsewhere, it is up to the author
to clear any copyright or acknowledgment matters.
Speleological expeditions have a moral obligation to produce reports (contractural in
the cases of recipients o f awards from the GharParauFoundationl.
These should be concise
and cover the results of the expedition as soon as possible after the return from overseas,
so that later expeditions are informed for their planning. Personal anecdotes should be kept
to a minimum, but useful advice such as location of food supplies, medical services etc.,
should be included.
Authors may order reprints ~f their contribution for their own private use. The order
must be notified t o the edito r at the time of submission. Orders after publication cannot
be accepted .
If yo u have any pr Oblems regarding your material, please consult the editor in advance
o f submissi o n.
(Dr. T.D. Fo rd, Geology Department, University of Leicester, Leicester LEi 7RH.
Ph one 0533 - 554455 e x t. i~l 0 r 0533 -715265).
CAVE SCIENCE
TRANSACTIONS OF THE BRITISH
CAVE RESEARCH ASSOCIATION
June 1983
CONTENTS
The caves of Gunung Sewu, Java
A.C. Waltham, P.L. Smart, H. Friederich,
A.J. Eavis
&
T.C. Atkinson . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
&
B.Wilkinson . . . . . . . . . . . . . . . . . . . 97
Cover photo:
&
CAVE SCIENCE
Trans. British Cave Research Association. vo~.~o, no.2, pp.55-96. June 1983
:....0.:.
_.,., , _...
v ......
'lldian Ocean
Figure 1
//
100 Ill.
Merapl Volcano
II
II
III
f,
,' "
\ \
lawoe Vo lcano
VOGV"K""U
"
,
/
,/
,/
/1'1
"
Wono.arl Plateau
WOlloa"II,
,...... ,......
~IIAC'''''"TOIIO
"""
~'""'r~~'-'"
~~".....~~~
0'
...--
Co k.r.,
........
le r....... '
. -"'-"
Dr, Ha,
"...,
._'""" ....
~
r--..
",.-....~,
,-...
,-..
r'""
f"'"""<\'-"
,......
~,-...~
..........
,......,
r"',.........
..........
,-..
~~;-....
Figure 2
... r."
400 r .,.1
Gunung Sewu
Wono.arl Plateau
.oe
o
Watw table
/
Figure 3
eout"
Indian Ocean
commonly very thinly bedded. Elsewhere, a crude and massive bedding may be
recognisable, bu t many outcrops have an almost structureless appearance.
Biohermal structures, with reef maSses and forereef breccia slopes are
identifiable in some places. A distinctive feature of the Gunung Sewu
limestones is its content of isolated, lenticular or irregular, masses of
vOl canic ash. Individual ash pockets may be from centimetres to tens of
metres across, and it would appear that they may be dense enough in some of
the eastern areaS to cause perching within the karst aquifer;
the clay
matrix of some of the limestone breccias may also contribute to this effect.
In most areas, the base of the limestone is not seen and in Central Sewu it
would appear to be well below sea-level.
SURFACE KARST
Gunung Sewu is an extremely well developed cone karst (Lehmann, 193 6).
The landscape is entirely dominated by the conical limestone hills. There
are probably 10,000 of them, all of remarkably consistent size, the mean
being around 200 m in diameter and 50 m high. The lack of forest cover means
that the cones are clearly visibl e and they stretch to the horizons in almost
monotonous uniformity. The dominant shape is a hemisphere, though some are
more sinusoidal, others are purer cones with uniform slopes and only rounded
tops, while a few are more irregular. Terraces, cliffs, overhangs and notches
occur irregularly and appear to have no more than local significance. The soil
cover is thin and patchy and there is negligible scree; bare rock and various
karren forms constitute most of the slope surfaces. Artificial terracing is
almost ubiquitous, preserving narrow soil strips between steps of bare rock or
hand-built rubble walls. The cones are not simply related to geological
structure and are certainly not s imple, exhumed reef-Knolls. The only
geological influence is the deve lopment of flatter, more rounded cones in the
chalkier limestones of the facies transition zone towards the Wonosari Plateau.
Between the cones, va lleys are floored with clay which may be 10 m deep . In
some cases there is only a low rocky col between adjacent cones, but linear,
terraced, clay-floored depressions , integrated into dendritic dry valley systems)
are much more common. Some of these systems feed to the coast or sinkholes
along the edge of the Wonosari Plateau; others terminate in closed basins
within the cone karst, and these mayor may not have conspicuous open sinkholes.
Artificially dammed lakes, Known as telagas, are formed on the thicker clay
floors. Without the ubiquitous terracing, surface run-off would cause major
soil erosion in these valleys. The proportion of conical hill to valley floor
does vary, though through most of Gunung Sewu the limestone hills occupy two to
three times the are of the clay-floored valleys . In some of the higher central
parts, the valleys are reduced to narrow winding strips between the cones,
while in parts of northeast Sewu and close to the coast the cones are only
isolated hills in wider al luvi al tracts.
A feature of Gunung Sewu is the general concordance of cone summit altitudes,
rising from each flank towards the central east-west crest line (Fig. 3). Even
the finer details of this "summit surface" match closel y to the geological
structure, and suggest the possibility of development of the cones by dissection
of an anticlinal stratimorph . The integr at i on of the valley systems indicates
the importance of early fluvial erosion, but the precise origin of the cones is
open to debate , contributions to which were n o t among the objectives of the
1982 field work. Nevertheless, the Gunung Sewu cone ka rst remains a remarkably
impressive product of limestone erosion.
CAVE EXPLORATION
Cave entrances abound in Gunung Sewu. Fossil caves, commonly with profuse
stalagmites, open into the sides of the conical hills, but f a r more important
are the sinkholes sited on the edges of the valley floors between the cones.
From almost any point in Sewu it is possible to just walk downhill and arrive
at a sinkhole within a kilometre or so. Most entrances descend steeply or
vertically, and many have been modified by loc a l people who have built stone
wall s, behind which floodwaters drop sediment, building up flat fields right
to the lip of the entrance drop. A few entrances have been blocked, but
sinkho l es are still a major component of the Sewu karst.
Villagers have exp l ored some caves in search of dry - season water suppl i es .
Most hori zontal entrance have been explored, though commonly only as far as
the first water and they a ppear always to have stopped at deep water.
57
GUNUNG
Sffi~U ,
JAVA
2. Dam around the Pule ireng sinkhole on the e dge of the renovated tel aga
(Waltham)
3. Andy Eav is and Sudiyono rig the entrance drop of Luwang Jalak Bromo
58
nvaltham)
Vertical drops have foiled them through lack of equipment, though they have
managed some spectacular climbs and have used bamboo ladders for drops of up
to about 10 metres where water was visible below. The dominance of shaft
entrances has reduced the overall impact of these explorations.
Of early foreign visitors to the area , Danes was the most significant. His
enthusiasm was considerable and his writings (1915) describe many of the cave
entrances, but he nowhere explored far beyond daylight. Since then, Balazs
(1968) and many others, including British, karst geomorphologists have visited
Gunung Sewu, but they either ignored or did not notice the caves. Indonesian
cavers, from the Specavina national group, together with various foreign guests
have visited a handful of the Sewu caves, but have done little s y stematic
exploration.
In summer 1982, after a brief reconnaissance in 1981, the authors of this
paper explored many of the caves as part of a groundwater exploration project
(see below). The project identified 250 entrances, explored 170 of them and
surveyed 62 of the caves with a combined surveyed length of nearly 28 kilometres.
Details of this work appear in an unpublished report (Waltham et al , 1981),
copies of which are held in the BCRA and RGS libraries. Only the major caves
are described in this paper, though brief notes on all known sites are in the
Gunung Sewu Luwang Register as an appendix.
In 1982, a group of Belgian cavers from the Verbond van Vlaamse Speleologen
en Alpinisten , led by Denis Wellens, also visited Sewu on a filming project.
They made the first exploration and survey of Luwang Grubug.
THE CAVES AND CAVE HYDROLOGY
The caves of Gunung Sewu fall into a number of reasonably definable groups.
One group comprises the fossil stalagmite caves within individual cone hills;
these are common, but are unlikely to be of any considerable length, and due to
their low water resource potential were almost totally ignored by the 1982
project.
A major group of caves lies along the northern margin of Sewu where it
borders the Wonosari Plateau . Surface rivers from the Plateau sink where they
meet the outcrop of the cavernous Sewu limestones, and there is also a
considerable underground water input due to southward leakage from the Wonosari
Plateau aquifer. The result is a suite of active river caves which in the wet
seaSon must be nearly all impassable or flooded to the roof, and in the dry
season still contain some sizeable flows. Great lengths of explorable stream
passages do not exist because nearly all the known caves descend rapidly to a
flooded zone between 10 and 30 m above sea level. Dye tracing of the two major
river sinks has proved a connection over more than 15 kID, right beneath the
Gunung Sewu ridge, to the Baron resurgence on the coast. It is likely that all
of the caves in this group drain to Baron (Fig. 4). The flow at the Baron ranges
between about 6 and 30 cumecs, and it appears to account for the drainage of a
very large proportion of Gunung Sewu. The resurgence is on the beach , and behind
the boulders of the entrance collapse only 100 m of river passage is explorable
up to a deep sump. The curved line of the major uhderground link on Fig. 4 is
based on the assumption that the major zone of cave conduits controls the
position of a conspicuous groundwater trough recognisable on a contoured water
table map. Details of this map, its compilation and its implications, are in
the main report (Waltham et al, 1981) and will also appear in a future
publication by the same authors.
In contrast to the large cave passages associated with the major sinks of the
northern margin, the central area of Sewu is characterised by steeply descending
shaft systems. They a re located in the valleys and depressions between the
conical hills and mostly have quite small catchment areas. In the dry seaSon
the entrances are all dry, but some do progressively pick up water at increased
depths;
in the wet season they are important active drains and some back up
water to the surface. Most have little horizontal extent before terminating in
either static pools, active sumps or clay chokes formed of the vast amounts of
inwashed surface sediment. Only a few reach sub-horizontal conduits, and these
rapidly sump. Only one shaft cave, Buhputih, has been dye tested , again to the
Baron resurgence (Fig. 4) . It is likely that most drain direct to Baron, or
alternatively northwards into the marginal groundwater trough and then to Baron;
those further to the southeast probably drain to other coastal springs all of
which are much smaller than Baron.
The fourth, and rather less well defined, group of caves comprises a number
of more generally horizontal systems in the north and east of Gunung Sewu.
59
There is some spacial overlap with vertical shaft systems which also occur
within this area. The larger number of horizontal caves is at least partly
due to the geology, in that ash beds within the limestone increase in number
to the east and must reduce the vertical permeability of the aquifer. Dye
tests have defined the hydrology of the area (Fig. 4). The underground Kali
(River) Bribin is an anomalously large, major conduit at high level within the
cone karst. It must have a large catchment, and three cave streams have been
dye-traced to it (F ig . 4) . Downstream it has been traced through Ngreneng, into
the marginal groundwater trough and thence to Baron. The edge of the Baron
catchment has been partly defined by the dye trace from the Sodong (Mudal) cave
to the Pracimantoro spring on a low level plain across a facies boundary
comparable to that onto the Wonosari Plateau. There is no accessible cave at
Pracimantoro itself.
The following descriptions of the caves are intentionally very short. They
define just the character of the caves, and the. surveys provide the details.
All the smaller caves are only referred to in the appendix register.
conventional notation is used on the cave surveys, e xcept for three additional
items added to evaluate the water resources which were the original purpose
of the surveys. These are:
1) Figures beside water - flow arrows refer to dry season flows in litres/second
2) Figures in square boxes refer to the depth in metres below the entrance
3) Figures in rounded boxes refer to pool capacities in cubic metres.
STREAM CAVES OF THE NORTHERN MARGIN
GUA SEMULUH
Length 1250 m Depth 52 m
Grade 5 survey (Fig. 5).
A large, seasonally dry, level passage extends from the northern entrances
to the main entrance just at the start of a canal which is heavily used for
water supply. Beyond the canal, the large passage continues with gours and
stalagmites and then cont inues unsurveyed to a partial mud blockage. An entrance
in the next depression to the south drops into a large unexplored passage which
may be the continuation and may connect with the upstream inlet in Gua Bribin.
From the gour passage, the Semuluh water drains down a smaller, younger canyon
to a terminal sump.
LUWANG CEBLOK
Length 600 m
Depth 92 m Grade 5 survey (Fig. 6).
Two 25 m entrance pitches drops into a roomy passage leading to a third pitch.
This gives access to two large sloping boulder chambers separated by a
spectacularly low duck. The lower chamber ends in a wide and deep te~minal lake
which has a waterfall dropping straight into it from an inaccessible roof passage.
The cave is quite short, but very varied and extremely pleasant;
it may in the
future be used for water abstraction from either the entrance or terminal pool.
GUA SUCI
Length 260 m Outline survey (Fig . 7).
The Kali suci river sink is one of the most impressive in Sewu, but it does
not live up to its promise. A powerful stream flows in a passage 5 to 10 m high
and wide, which connects through to two large collapse windows, but not 100 'm
from daylight the third segment of cave ends in a deep sump pool overlooked by
massive stalagmite.
The sump is however very short, and connects to the half kilometre long Gua
Buri Omah. This cave has a short tributary canyon draining into about 400 m of
large gently graded river passage which sumps at both ends. Downstream it drains
to Luwang Grubug, at almost the same level.
LUWANG GRUBUG
Length 2290 m Depth 161 m Outline survey (Fig. 7).
Grubug is the finest cave yet found in Sewu. It was originally explored, in
1982, by Denis Wellens and his Belgian colleagues, who are publishing their own
description and survey. The outline survey in Fig. 7 is a compilation of the
writers' own low grade surveys with the downstream passage simplified from the
Belgians' survey.
The Grubug entrance is a dramatic 64 m free hanging pitch into the centre of
a chamber crossed by the underground Suci river. The narrow surface opening
creates magnificent visual effects when sunbeams, filtered through overhanging
trees, and picked out by a thin mist, strike the floor of the chamber and slowly
60
0'\
I-'
00
/:
/--,'
Indian
- ,'
Ocean
..
:'
... -- -~
,-
Lebak Bareng
J.
//
t-Y~------
/ ,'
'9
'5
..
'. /
.... .
t... \
'-
,./
Puniran.
./
Figure 4
' ..
/ .C;::;::
... .
----
Ngiralan
..
'7
Ngepoh
Jomblang
.. ~
I.
'.
.,
.. /
/ ';
,I
Principal road
Cave entrance
River
.........
r'.
:",
/.-...---'
.~
I ('
~~
~)
. "
\BuhpUtl~cr'-'II'"
/
.. /- ....
.
/
./
Jomblang
. '"
S<m>erejo
" ' .
..
----...-'.
. .. -
-.-->'
..n "..
~SidoIlardjo
Jero
. .
{j:O . .
....'.: ..
;
ct(\'t 'P
--- .
~:~.
.. ,~'~blo.k
~~--+-.
. . ' .'
,Grubug:
.... ... ~ . . . . . . .
..
')J~~.
Bedesan
_ _. - .,.., Seropan
// ~ 0t~ , M~
/ ....
.
Nglngrollg
".
------,"","",'
---
kllomelres
........~
Major village
Contour In metres
Underground drainage
\.-
.,
F.."
Daren.
, . ..... ..:' .. /
.. -....!
Baran
.. ~. "
- _______ ._.":1,
..
.:,;~i'
05
'0
'5
Q)
l\J
Road
-40
-20
~'~
+40
Cl.
Sump
~.:::;:::;0--_. ..
PROJECTED
CAVE
SECTION
DOWNSTREAM
PLAN
l00mo....
1982
/1'9
Figure 5
J:::'"~ ,
~
~.. "*- ~ ,.~
r=
</'
60
GR 644139
j ---
H.
SITE 63
JAVA
SEMULUH
,~
Sump
""....
Upa.r.am caye
GUA
"'\
..............
~.-
I/,.....
:,...,
'";;~ ...
NI
PLAN
COMPLETE
......
100 Ir
CAVE
move across it during the day. Three passages radiate from the chamber. A wide,
dry, high level tunnel leads to the la~ge Jomblang entrance, the easiest way in
with a broken 40 m pitch. Upstream to the north is over half a kilometre of
wading or swimming in a wide passage. The downstream canyon has over half a
cubic metre per second of white water in an extremely sporting passage.
Cascades, rapids, pools and wa"terfalls alternate as far as a terminal sump,
from where the water has been dye-tested to the Baron resurgence.
LUWANG SEROPAN
Length 650 m Depth 65 m Grade 5 survey (Fig. 8).
A broken entrance shaft and a short canyon passage lead directly to a length
of river passage. This is mostly 8 m wide and a lmost level, characterised by
deep pools and low ducks as far as sumps both up and downstram. The source of
the water is unknown, and downstream it flows into the Bedesan cave.
LUWANG BEDESAN
Length 1025 m Depth 104 m Grade 5 survey (Fig. 10).
A narrow, meandering canyon passage descends a series of pitches and climbs,
to junction with a large stream passage. Upstream this is wide and almost level
to a sump at about the Same level as the continuation in the Seropan cave.
Downstream there is a magnificent flight of large gours and cascades as far
as a deep, depressing and very muddy terminal sump pool, before which there are
several tributary passages. The terraced gour pools make this cave one of the
more spectacular and exci t ing yet found in the S ewu area.
LUWANG SERPENG 2
Length 220 m Depth 96 m Grade 5 survey (Fig. 12).
Serpeng 2 has the smallest entrance of a trio of holes near the Serpeng
village. A 47 m shaft drops into a large dry passage which is unexplored
upslope. Downstream a series of small climbs and pitches lead down a large
canyon to the edge of a large and impressive terminal lake which is a potential
target for a future abstraction borehole. A short distance east of the entrance
lies the enormous crater-like Serpeng 1 pothole with a sloping, crumbling 60 m
pitch to a short passage and sump. Adjacent to this, a deep blind valley ends
at the massive entrance of Gua Serpeng which sumps hardly out of daylight.
GUA MULO AND GUA NGINGRONG
Lengths 170 and 380 m Depth 74 m Grade 5 survey (Fig. 9).
The Mulo caves are in two parts. Valleys from the north end at Gua Mulo
which has a low, wide, level passage through to an exit into a large blind
depression, partly ringed by rocky cliffs, indicative of some collapse in its
origin. Another valley system enters it from the south, and in its east wall
is the tall arch entrance to Gua Ngingrong. The large and .impressive passage
leads to a series of pitches each into a deep pool, and then into a large
chaotic chamber and a terminal sump complex - a spectacular piece of cave but
again one which does not live up to the scale of its entrance.
GUA SUMURUP
Length 1435 m Depth 58 m Grade 5 survey (Fig. 11).
With its entrance in the dry gorge just beyond the sumped river sink of the
Kali Tegoan, Gua Sumurup surprisingly and unfortunately does not lead back into
the underground river. Its impressive entrance passage leads to a 24 m pitch
into a deep lake, beyond which lies a spacious chamber. But the further passages
are just roomy level tunnels with abundant mud from their annual complete
flooding, and they all end in murky sumps. The main river sink has been
dye-tested to the Baron r esurgence.
POTHOLES OF THE CENTRAL AREA
LUWANG BLEKONANG
Depth 134 m Grade 5 survey (Fig. 15).
A succession of dry shafts provide a steep descent in an ever diminishing
passage to a depth of 134 m. Beyond that point a narrow rift continues.
LUWANG BUHPUTIH
Length 850 m Depth 200 m Grade 5 survey (Fi g . 14).
A series of climbs, rope pitches and rifts drops into a large dry canyon
which continues to the head of a pitch. From that point the cave spirals down
63
.,.en
CI a
I \~
lIoor
f.y
&I
,/
10
loll' .. '
I
OPln?/'
'8Wllel
I,"",,..,,,,~
I r
l'
.,/
Wat.r'a" "tnl.t
Track Junction
LUWANG
CEBLOK
01"'\ ;"'\
,~
Pool--1
IdVO--'
..)
) I~.....L
liopi
I 8urlici
olmF )
Track
Figure 6
1+131
Telaga floor
PLAN
&0
Pool
Entrance
100 Metr . .
Dry yaney
Percolation Inl.t l
Entrance P23
OFFSIl
FLOOR DETAIL
PROJECTED SECTION
~::.~:i\'----"
.....J III.
Telaga
-----------
- 100
- 80
- eo
- 40
-20
-0
+20
U1
0">
'N
ProJected .ectlon
3:f;i\
-
.-----z.
'0
:f!
'-
'\
650
......,.
Luw,na
'0
....
W
Jomblan:g
500 metr
Figure 7
- - - - - - - --
>::::2 0
8wt Omah
.~
....
8ucI
Luwang 80nglat
Luwana ~rubua
to aaron
THE suel
. UI
.0
...
-=- '0:
200
Kall 8uel
0'1
0'1
-40
+20
It
Entrano.
vr
0' ~
PLAN
..
Figure 8
"AI' 222222228;;$
/;'
100 ... tr
.ump
".,
U""... ".,
~
SEMI-EXTENDED SECTION
c.,~
~~
10
,
LUWANG SEROPAN
""
-..J
1N
PLAN
Collapse depression
1/11//
1/11111
Dry valley
FllIIJ(e"
RII
I
Sump
50
~.,
100 Melres
I'
,---
__
\I
"" -----------A
' ",
P7~~/.../,
---------
PROJECTED SECTION
G.R. 549126
Gua Nglngrong
SITE 170
- 80
-60
- 40
-20
+20
+40
+60
A......... I . . .
'r ..L
110M
.......c.
,....
.
~?1j
"'?
y.-::l'
a.d n
,~--~-------.~~
... tr
a.,op.n
"OE
...----:l1ii
AREA MAP
z
PLAN
LUWANG BEDESAN
.t,..
0L'__........__..L...---''--...L__.J5,O.:....______________.:..:;1C?0 M
eo
up.t, ........,.
PROJECTED SECTION
100
Figure 10
68
(Waltham)
5 . The f o ssil passage in Luwang Grubug, looki ng out to the Jomblang ent rance
(E avis)
(IvaI tham)
69
-..J
Gorge
Snk
P24
Lake
'
MId Hal
1082
--=--
Figt.l'e 11
_==========::;:::==
~_
----- \,, __ _
Entrance
JAVA
- - =s\
QR 811120
_~~~~~ __ ~
--w-
-40
~--
+40
SITE 218
GUA SUMURUP
Dry veley
____
.L_
50
100 metres
Downstream SUTl>
--------~-
EXTENDED SECTION
Flood level
/25
ff
Clay Het
Downstream SlIl1>
*;s==--=::::::::;;
.....::::...:;;:.~;:::-.~ Clay Het
I~
~2=75=m=omItt=e=d=====;::=;:==--
.....
PLAN
15~~
1-361~
-40
.,.en
CI a
I \~
lIoor
f.y
&I
,/
10
loll' .. '
I
OPln?/'
'8Wllel
I,"",,..,,,,~
I r
l'
.,/
Wat.r'a" "tnl.t
Track Junction
LUWANG
CEBLOK
01"'\ ;"'\
,~
Pool--1
IdVO--'
..)
) I~.....L
liopi
I 8urlici
olmF )
Track
Figure 6
1+131
Telaga floor
PLAN
&0
Pool
Entrance
100 Metr . .
Dry yaney
Percolation Inl.t l
Entrance P23
OFFSIl
FLOOR DETAIL
PROJECTED SECTION
~::.~:i\'----"
.....J III.
Telaga
-----------
- 100
- 80
- eo
- 40
-20
-0
+20
U1
0">
'N
ProJected .ectlon
3:f;i\
-
.-----z.
'0
:f!
'-
'\
650
......,.
Luw,na
'0
....
W
Jomblan:g
500 metr
Figure 7
- - - - - - - --
>::::2 0
8wt Omah
.~
....
8ucI
Luwang 80nglat
Luwana ~rubua
to aaron
THE suel
. UI
.0
...
-=- '0:
200
Kall 8uel
0'1
0'1
-40
+20
It
Entrano.
vr
0' ~
PLAN
..
Figure 8
"AI' 222222228;;$
/;'
100 ... tr
.ump
".,
U""... ".,
~
SEMI-EXTENDED SECTION
c.,~
~~
10
,
LUWANG SEROPAN
""
-..J
1N
PLAN
Collapse depression
1/11//
1/11111
Dry valley
FllIIJ(e"
RII
I
Sump
50
~.,
100 Melres
I'
,---
__
\I
"" -----------A
' ",
P7~~/.../,
---------
PROJECTED SECTION
G.R. 549126
Gua Nglngrong
SITE 170
- 80
-60
- 40
-20
+20
+40
+60
-..J
Gorge
Snk
P24
Lake
'
MId Hal
1082
--=--
Figt.l'e 11
_==========::;:::==
~_
----- \,, __ _
Entrance
JAVA
- - =s\
QR 811120
_~~~~~ __ ~
--w-
-40
~--
+40
SITE 218
GUA SUMURUP
Dry veley
____
.L_
50
100 metres
Downstream SUTl>
--------~-
EXTENDED SECTION
Flood level
/25
ff
Clay Het
Downstream SlIl1>
*;s==--=::::::::;;
.....::::...:;;:.~;:::-.~ Clay Het
I~
~2=75=m=omItt=e=d=====;::=;:==--
.....
PLAN
15~~
1-361~
-40
pH
P7
...
,.,
"
.....
\
I
I
"
"
.....
\,\
~"\
~,,""
"
\,
Entrance
'...
....
" ,
"
. ..
~ ~
... . -
,/
~,!" ,
.l
...
Figure 12
SITE 184
PLAN
,Ii
GR 557127
10 metr
-100
011"", _ _ - . .......,.
l~
.'
,q,~ ,
.. )~.
,;., .., /i
,/
.r-
, - - - .....
"
EXTENDED SECTION
a..ta.,.
.......
. , ~ .. I'"
_.......... .
/,~ ,
\,.-.
0,1
..... . . . .......,........,'
..... .'
~~
10
~.,
. ~. "
LUWANG SERPENG 2
Entrance
'""" .../
-120
- 80
-40
pe6
0-- -- #'
"\
II
Figure
pool
.hol.
r.:-
13
8u .. p
PLAN
N
.........
Pool
'~~5
.
.'
10 metr
pe
Pool
oL
low
EXTENDED SECTION
8URFACE
"."" ,..::::.fj"
'
8ump
TJJ!?,,'
JAVA 1982
~:"1~
~ ............
'.:: -"~ ..~.:: '"
.--::
,"
GR 742998
. D
c ::~prm.dl.t
SITE 179
pH
P7
...
,.,
"
.....
\
I
I
"
"
.....
\,\
~"\
~,,""
"
\,
Entrance
'...
....
" ,
"
. ..
~ ~
... . -
,/
~,!" ,
.l
...
Figure 12
SITE 184
PLAN
,Ii
GR 557127
10 metr
-100
011"", _ _ - . .......,.
l~
.'
,q,~ ,
.. )~.
,;., .., /i
,/
.r-
, - - - .....
"
EXTENDED SECTION
a..ta.,.
.......
. , ~ .. I'"
_.......... .
/,~ ,
\,.-.
0,1
..... . . . .......,........,'
..... .'
~~
10
~.,
. ~. "
LUWANG SERPENG 2
Entrance
'""" .../
-120
- 80
-40
pe6
0-- -- #'
"\
II
Figure
pool
.hol.
r.:-
13
8u .. p
PLAN
N
.........
Pool
'~~5
.
.'
10 metr
pe
Pool
oL
low
EXTENDED SECTION
8URFACE
"."" ,..::::.fj"
'
8ump
TJJ!?,,'
JAVA 1982
~:"1~
~ ............
'.:: -"~ ..~.:: '"
.--::
,"
GR 742998
. D
c ::~prm.dl.t
SITE 179
-...J
N
--
40
- 200 -
- 180 -
-160 -
-140 -
-120 -
-100 -
-60 -
- 60 -
1-
-20 -
EAST
PROJECTED SECTION
NORTH
Low.r
~P40
l.tj, .=
P40
I~"" "
Figure I.
~-----------i~~"
iP2~
1P'4 ,
'J
18Y.I-~..I!.\\
SITE 48
Duck
Duck
PLAN
50
G.A. 644082
JAVA, 1982
Sump
100 Metres
"
-140
-120
-100
.-80
-il0
-40
-20
08
P48
. \
Tor,..,..
o
I
10
I
20
,
jN
30 "'elres
P28
Figure 15
Continue mall
G.R. 607024
JAVA, 1982
CAVE SURVEY
GUNUNG SEWU
SITE 16
L. BLEKONANG
EXTENDED SECTION
Trav.rae
O~-~----
,,
Contlnu down
1'41
ao "' ,
GR 684066
Figure 16
P71
-100
Contlnu low
... a
GR 728031
JAVA 1982
EXTENDED SECTIONS
SITE 137
LUWANG GANDEK
SITE 119
JALAK BROMO
LUWANG
-100
Field
Call,o ..
0
.' , \
Tel
,,
"\
\
P1 0
,
I
I
P10
P ..
P3a
P20
,
"
\ narrow
-100
-100
LUWA'NG KENTENG
GR 538068
SITE 9
GR 621004
SITE 10
ao ",et,.,
LUWANG JOMBLANG
GR 613032
EXTENDED SECTIONS
Figure 17
- - - - - - - - - - - 1
I
I
\
JAVA, 1982
....
G.R. 638058
-20
-40
p,e
--r
,.J
L,dge
-eo
/1
-eo
(~(
~~v
~~'
P90
-100
P8
FLOOR DETAIL
OFFSET
EXTENDED SECTION
-120
o
L '_ _~_ _L-~
. ~I
__
__
110 Met,
~'
-140
-1eo
Sump
74
Figure 18
beneath itself for more than 130 m. A trio of spacious shafts leads to a small
st~eamway which contains a fine succession of gaur pools, steepening into
climbs and pitches and two deeper shafts. Various inlets combine to give a
substantial flow down the gently graded lower streamway, which has extensive
flowstone deposits and a series of long pools some of which have minimal
airspace, before the terminal sump. Buhputih is the deepest cave yet explored
in Sewu, and its fine shafts and lower streamway combine to make it an
excellent system. There is the possibility of upstream extension from the
foot of the main shafts.
LUWANG DAREN
Length 240 m Depth 122 m Grade 3 survey (Fig. 13).
An impressive free-hanging 65 m entrance pitch is followed by another pitch
in the same shaft, and then a passage broken by short climbs which spirals
round beneath the entrance chamber. This ends .at a drop overlooking a noisy
streamway, and two alternative descents ] ead to upstream and downstream
sections neither of which can be followed far.
LUWANG GANDEK
Depth 138 m Gr a de 3 survey (Fig. 16).
Two fine shafts each about 70 m deep drop to a very low bedding plane
passage which continues half full of water.
LUWANG GOPLAR
Depth 85 m Grade 3 survey (Fig. 17).
An impressively large opening has completely overhanging walls in a superb
bell shaft with a minimum drop of 70 m. This lands on a sloping boulder floor
and there is no wayan.
LUWANG JALAK BROMO
Depth 105 m Gra de 3 survey (Fig. 16).
A spiralling succession of short pitches leads to a . much larger shaft over
40 m deep, from the foot of which a climb descends to the edge of a ~urther
undescended dry shaft of unknown depth.
LUWANG JERO
Depth 151 m Grade 3 survey (Fig. 20).
A fine descent of 78 m in a large entrance shaft leads to a short dry
descending passage ending at a second pitch. This immediately bells out into
a free hanging drop down the centre of a chamber, the floor of which slopes
down boulders at one end to a terminal sump pool.
LUWANG JOMBLANG (10)
Depth 95 m Grade 3 survey (Fig. 17).
A single cylindrical shaft drops via a 93 m pitch to an impassable floor of
mud and boulders.
LUWANG JOMBLANG (1 75)
Depth 106 m Grade 3 survey (Fig. 23).
A spectacular 77 m entrance shaft has a single outlet which rapidly diminishes
in size. This has another pitch and a small canyon which continues to the top
of an undescended drop of 20 m; this enters a large chamber which appears to
continue as a s i zeable passage.
LUWANG KARANG
Length 325 m Depth 94 m Grade 3 survey (Fig. 19).
An entrance pitch of 44 m drops to the head of a meandering canyon with a
phreatic roof tube in its first section. Short climbs break the steady descent
to where mud shows the level of annual flooding, 10 m above the normal sump leveL
LUWANG KENTENG
Depth 72 m Grade 3 survey (Fig. 17).
A steeply descending dry canyon has a series of pitches followed down 72 m
to the lip of an undescended 20 m pitch beyond which the passage appears to
continue as a narrow rift.
GUA LEBAR BARENG
Length 470 m Depth 166 m Grade 5 survey
(Fig. 22).
75
-..J
0'\
-100
\
\
------
w.t on
-----
c3
--
~------------------ ---------
EXTENDED SECTION
wat.r l.v.1
----
P44
Entr.nc.
Dolin.
__
__
/ ~ ~~
Figure 19
---
--lL...
Dolin.
PLAN
~~
Chamber
Ch.mb.r
Foaall phre.tlc 1 1
- --
-L-
,~r f~
c?'~_
Entr.nc.
SITE 98 GR 501047
LUWANG KARANG
eo
lump
v~
Sump
m.t,..
A large and inviting entrance yields an easy wa lkin g pas s age which barely
descends . This ends at the top of a spectacular shaft system which drops 140 m
with unbroken vertical walls. At its foot is a deep pool under a showerbath,
and an outlet streamway which, though pleasant and decorated, ends prematurely
in a deep clear sump pool . Even though of no great length, Lebak Bareng
provides some exciting vertical caving .
LUWANG NGEPOH
Depth 182 m Grade 3 survey (Fig. 21).
An unassuming entrance climb and a low, wide , boulder-strewn chamber lead to
the top of a fine , dry, broken , shaft over 60 m deep . At its foot a small
opening reveals the blackness of another shaft, the best part of 100 m deep ,
which is characterised by a rain of dripping water . The 67 m pitch ends on what
appears to be a floor but is in fact a number of very large, loose slabs of
dubious stability . From the foot of the next pitch a few metres of canyon ends
at an undesended 20 m pitch where there appears to be a way on below. It is
likely that , with very little effort, Ngepoh could become the deepest cave in
Sewu .
LUWANG NGIRATAN
Length 325 m Depth 168 m Grade 5 survey (Fig. 18) .
A broken entrance drop is followed by a tortuous and constricted passage to
the head of a 106 m shaft. This is a magnificent vertical feature, draped in
stalactites and floored by a deep pool. The continuing passage features
stalagmites, flowstone , gours and pools but unfortunately ends in a sump after
little further descent.
LUWANG PUNIRAN
Depth 100 m Grade 3 survey (Fig. 23).
A staircase of five pitches f l oors a steeply descending canyon . A further
climb ends at the lip of a n undescended 20 m pitch into a pool which appears
that it may be a sump.
LUWANG SETRO
Length 250 m Depth 140 m Grade 3 survey (Fig . 24) .
A large high canyon passage has a series of short drops followed by two
deeper pitches and a fina l staircase into a terminal sump pool .
LUWANG TONG POCOT
Length 900 m Depth 142 m Grade 5 survey (Fig. 25) .
The roomy entrance shaft soon narrows into a fossil rift, so that the only
way on is in a small youthful overflow passage . This has a succession of nine
short drops, interspersed with an interesting swim and a pair of thought provoking squeezes, before opening out into a horizontal gallery . This contains
a series of long pools, and has plenty of length though unfortunately a general
shortage of standing height. A pair of waterfalls provide interest near to the
halfway point to the terminal sump. The combination of shafts and stream cave
make Tong Pocot a fine system, but the restricted passage sizes make it more
memorable for its sporting chal lenge than for its grandeur .
CAVES OF THE NORTHEASTERN AREAS
GUA BRIBIN
Length 3900 m Depth 33 m Grade 5 survey (Fig. 26).
An uninspiring entrance passage , blac kened b y the soot from paraffin torches ,
eOnds at a T- junction with a major river galle r y . The cave river is dammed by a
substantial masonry structure with overflow channels, and large pumps feed a
pipeline to a water distribution scheme for the surrounding vi ll ages. upstream
of the dam , the cave is occupied by a l ake one kilometre long , most of which is
deep and so requires a monumental amount of swimming. Beyond the lake a short
section of river passage ends in an upstream sump which has been proved to drain
a wide catchment area towards the east. A large dry tributary passage has been
followed for 500 m in a northerly direction;
it continues unexplored an d may
relate to the fossil passage in Gua Semuluh.
Downstream of the dam lies a kilometre of gently descending and very fine
river cave. Foaming r ap ids and deep, smooth, fast-movin g water alternate with
canals and lakes , bordered by a variety of shingle beaches and rocky terraces.
The mostly black wallS are broken by just patches of white stalactites and
77
-..J
~:.:
\/
: I
,..)
\
"'1
I I
I I
""
I \
PLAN
8ump
JAVA 1982
iN
GR 612088
20
P.8
Figure
EXTENDED SECTION
P78
-120
-80
-40
o --
Entrance
SITE 106
LUWANG JERO
50 m.lr
1-1511
-200
-100
Entranoe
P87
P28
P20
u~
50 ",elr
GR 630053
Contlnu
EXTENDED SECTION
II'
I ---?--
n---
Figure 21
P31
SITE 109
JAVA 1982
LUWANG NGEPOH
(Waltham)
(Eavis)
<----(
~
Enttanc.
.,
-----
P13
GUNUNG
SEWU
CAVE
SURVEY
JAVA
-50
P50
1982
P25
SITE
37
' GR
526085
-100
P45
SEMI-EXTENDED
SECTION
i~==::::-JJ~~
h.
50
100 ",.tr
PLAN
1/
II
//
//
1/
1/
1/ Road
II
Figure 22
80
-150
tv
(Xl
\
\
SITE 151
I
I
I
P20
-100
GR 625011
EXTENDED SECTIONS
SITE 175
LUWANG JOMBLANG
P17
Figure 23
60 ",atr..
GR 628110
LUWANG PUNIRAN
-100
P1S
" ,,,
40 matra.
G":~
P32
SITE 33
GR 523059
JAVA 1982
,,
,
\
\
I
0 1
Sump
.;-
,
\
EXTENDED SECTION
~----..?t'
Figure 24
LUWANG SETRO
Road
-120
- 80
-40
--
(X)
P5
:---~
"
\-1<401
C8
Ct:;:'"
200mto~
ComRIe.
= =
PLAN
-------.
80
Figure 25
8Ift1)
.... 0:
Poole ~ P8
'ft,
EillI~"'\
P8
"\
P7
P8,PS.C5.P7
:::::: =-
EXTENDED 8ECTION
100 melr..
OUNUNO
narrow
"\
@
SURVEY
narrow
JAVA
<=U;"'p
Poole
----~:...
OR831101
CAVE
SITE 81
SEWU
1982
P12
-120
-80
-<40
..
...
co
co
a:
CO
C
:::
"
c
>
c
III
>
a:
::;)
CO
III
>
C
u
::;)
III
CO
::
;::
CO
a:
"
0
:!
..
III
iii
C!J
z
::;)
z
::;)
"0
84
lJ1
en
....
Sump
\ --_
II
"'-
: --
SECTION
PLAN
50
Ram PIMIlPS
0-
--80
--40
200 metres
Oat r
Head pool
- - ____
Ooi'le
.' - ...:~~-
PUMP
~CTION
OF
18
PLAN
SITE
/---"
c:=---".
Oownatream aump ..
Figure 27
~..,:;'"
_~
....~
.(
Chamber / / .. .... .....~.~,
/.......
__~"""~~>:?'::7
~" "
' . Y- ( Entrance
-+40
Pipes
.,.:::;
__ /- - - - - - ")_~r---~--~
I,
-----.
.Om.""
Bamboobridge
).. . _- -? -
" ~
', ,~
Ram pump.
DETAIL
GIL A P
182
Slreamway
SITE
JAVA
Gaurs
___
;:=-4-
Top ~
GR 729192
1982
GUA
V~I-421
~8
-,.." .
..
a.
<
>
<
..
... . .....
.. ..
a
>
a:
.
:>
CJ
Z
o
Q
o
en
~
:::t
a:
"
..
..
.,. .
>
<
:>
~
;;;
"z:>
z
:>
"o
CJ
\ .... .. .........
.../ \ ..........
~/
'.
..
...........>::'. ................
---
\:-;
III
:---
. . . ....
....
....
>
..... .
:'"
..........
86
\\
lB82
SITE 133
G.R . 762112
'00
.\
200 METRES
"
PLAN
'. . (i):ff0.
"...
-10
'!'!! ,
.'
CROSS SECTIONS
ENLARGED X2
~, :::::;
- -otO
.w
\
\
\@
LUWANG SAPEN
~~O_
-8
--- ~J:;.,
~
\
\
\
\
. ..---.......,
11
"
I---:;n---+~
.0 '-----"'"
Figure 29
87
Gua Gilap
13. Sulu
Prarelan"
If
Kepleng"
34
Spring
Discharge. summer
, 1982 (lis)
Surfac e river flowing
..
II
----
,,'
.-
/" 1 /)\_ . .
'-'
Kilometre
"
"'-1> .. ,-
Figure 30
Stream sm
d connection
..... Known un d ergroun
II
'ble
II
.. _. - Poss.
with en trance
Known cave
....... D
~---, -) = cU~..--J
dry
-:>
--- ....,
II
, k
250 - -
D
_~___:'5~O~O~:~_________
To K.
"O"I:
Telagas
Contours (metres)
I nd
Alluviated Iowa
Karst upland
BCRA library, and is welcome to contact any of the writers who can provide
various additional data. Finally, future visitors to Gunung Sewu are reminded
that the above cave descriptions are based on dry seaSon descriptions, and most
exploration is probably impossible in the wet season.
THE GUNUNG SEWU CAVE SURVEY
This was the official title of the exploration project which waS carried out
by the writers in 1982. The background to the Survey is of particular
significance because it was one of the few instances where cavers have been
contracted to explore caves specifically on an economic basis , in this case
for the purpose of assessing groundwater resources.
Severe water shortages each dry season have become part of the way of life
in Gunung Sewu. Not only is there inadequate water for any economically
beneficial irrigation schemes, but during the dry seasons there is in many
areas a complete lack of surface water, and even resources for drinking supply
are desperately scarce. Traditionally, dry seaSon supplies have come mainly
from the artificially dammed telagas, most of which dry up at some time during
the season, and from a number of shallow accessible caves. Better organisation
of rainwater catchment from roofs, together with repairs to telaga dams, have
created some improvements in the recent past . Additionally, isolated schemes
to pump water from the caves have been established, for example at Bribin and
Gilap, but currently none of these pump schemes is in full working order.
Since the late 1970s, Indonesia's Ministry of Public works and Britain's
Overseas Development Association have jointly financed a large groundwater
project centered on Yogyakarta, with Sir M. MacDonald & Partners of Cambridge
as consultant engineers. Gunung Sewu lies within the area of this project .
Adrian Young is the field engineer with responsibility for Sewu, and he
initiated a programme of telaga improvements. He also realised that various
isolated attempts to utilise the groundwater had generally not been efficient;
notably a series of 13 boreholes in the limestone had mostly been dry or
produced uselessly small yields. Furthermore, overall planning for the
economic development of even the known sources could not proceed until the
water resources of the karst aquifer were properly assessed. He wanted a team
of cavers to carry out direct exploration of the sinkholes, and he therefore
contacted the Royal Geographical Society, who passed his request on to the
writers.
After various discussions at Cambridge, a one -man recce with Adrian Young
in 1981, and some tedious "political" hassles, the project waS established.
The contract waS simply that two two-man teams should explore and map as many
sinkholes as possible in a search for useable water resources. A p art - way
change over involved a fifth caver, and some overlap was invaluable to the
water-tracing programme. Excellent facilities, in terms of living and transport,
were provided at Wonosari, and searching for entrances waS eliminated by two
of the local groundwater staff, Mas Sudiyono and Mas Untung, compiling the bulk
of the Luwang Register before the cavers arrived. All this combined to leave
the cavers almost nothing to do except explore the caves. Adrian young and
Sudiyono in particular discovered . the joys of cave exploration, but most work
was done just by the teams of two experienced cavers as this waS faster with
so much vertical work.
The object of the exercise was to find water. Exploration of "likely"
passages was the prime task . Caves with water were then surveyed to grade 5,
and dry caves of any length were just surveyed to grade 3 . Extensive programmes
of dye tracing and water quality testing were carried out at the same time.
Cave geomorphology, photography and exploration of dry caves were relegated to
incidental roles.
Economically useable resources really had to be one of two types. The first
were small streams or pools generally at depths of less than 30m, which could
be exploited by direct access or hand dug and operated wells. The second were
major streams or large lakes at depths up to about 100m, which could support
abstraction schemes using boreholes and submersible pumps. The lack of finance
in the rural economy of Gunung Sewu favoured the first type in the short term,
but the latter have valuable long-term implications.
A number of useable resources of both types were found. In some caves,
improved access or diverted cave streams could make a site viable. Some well
and borehole sites were pinpointed using Bob Mackin's Molephone, where the
target cave passage waS only a few metres wide. Unfortunately the Molephone
could not be used at the greater depths due to the shielding effect of the low
91
resistance ash beds within the limestone, but grade 5 surveys were considered
accurate enough for a borehole to hit a cave lake 10 m or more in diameter.
After the first phase of explorations it was realised that the main conduits
from the northern zone of river sinks to the Baron resurgence all lie at very
low level, probably mostly in the phreatic zone; there waS therefore little
chance of gaining access to major underground rivers in explorable caves
within the central part of the Sewu karst. This did enhance the value of the
anomalously high level Bribin cave river, but its descent via the Ngreneng
cave into the deep watertable trough eliminated prospects of further downstream
access to it. The balance of exploration effort therefore shifted, with
greater importance attached to finding small perched resources within the
shaft systems of the central area; many caves were therefore left unexplored
where they descended to depths from which small resources could not be
exploited economically.
Exploration of a karst aquifer can hardly ever be considered complete, but
the 1982 programme in Gunung Sewu waS considered to be well worthwhile. A
large proportion of the sinkholes was expl9red at a cost equivalent to that of
about four boreholes. Some resources with immediate potential were discovered,
and five sites have been budgeted for development within 1983. Other long-term
resources have been discovered, and the proper assessment of the aquifer can
now permit economic planning of future water supplies throughout Sewu using
a beneficial combination of telagas, small cave supplies and larger pumped
schemes.
ACKNOWLEDGEMENTS
The Gunung Sewu Cave Survey was financed by the Overseas Development Administration of
London and the Ministry of Public Works of the Government of Indonesia, on a sub-contract to
Sir M. MacDonald and partners) Consulting Engineers/of Cambridge.
In particular, the caving
team's gratitude is extended to J.I.M. Dempster at the MacDonald head office; Adrian Young in
Wonosari; Mas Sudiyono and MasUn:ung and all the other staff on the Womosari P2AT project.
Thanks are also due to the wonderfully friendly people of Gunung Sewu who made it such a delight
to work in their country.
REFERENCES
Balazs, D. 1968. Karst Regions in Indonesia; Karszt es Barlangkutatas, vol. 5, pp.3-56.
Danes, J.V. 1915. Das Karstgebeit Goenoeng Sewoe in Java; Boehm. Ges. Wiss. Sitzungber, pp.1-75.
Lehmann, H" 1936. Morphologische Studien auf Java; Geog. Abhand. serie 3, vol. 9, pp.1-114.
Waltham, A.C., Smart, P.L., Friederich, H., Eavis, A.J. and Atkinson, T.C., 1982. Gunung Sewu
Cave Survey, 170pp, unpublished Report.
March 1983
92
APPENDIX
GUNUNG SEWU LUWANG REGISTER
All known sinkholes in Gunung Sewu have been recorded in this register, in
a sequence reflecting only the date of investigation.
For each site, the number,
name and grid reference are followed by brief details except for those caves
described in the text of this paper.
Surveys and further details of a number of
the sites are in the main report (Waltham et al, 1981).
Names are abbreviated as:
G=Gua or cave; L=Luwang or sinkhole; S=Sumber or spring.
After the grid
reference, L=explored length, D=explored depth, P=pitch and C=climb, all in
metres.
The grid and all entrance locations in the central part of Sewu are
marked on figure 4.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
96
CAVE SCIENCE
Tran . Br i tish Cave Research Association. Vol. 10, no . 2, pp.97- 102. June 1983 .
ABSTRACT
The underground outlet of Lake Glomda lsvatn at Svartisen, Northern Norway ,
has a catchment area of 27 . 7 km 2 . Cave div ing confirm ed that the cave
consis ts of a totally phreatic tube system , undu l ati n g vertically as phreatic
loops at least 23 m deep . Scallops and passage dimensions were measured in a
passage cross- section which channeled all of the wat e r.
This allowed the
volumetric discharge to be calculated. The scallop discharge overestimates
the mean annual runoff dischar ge more than three times, but is comparable to a
probable snow-melt flood discharge .
In this climatic regime, phreatic scallops
seem to reflect the upper flow regimes rather than the annual "mean". Hence,
the results compare favourab l y with other studies of dissolution kinetics and
experimental scallop generat i on.
INTRODUCTION
Scallops and flutes a re formed by the interaction be tween a turbulent moving
liquid or gas phase and a soluble or volatile solid surf a ce. Numerous field
obs ervations in caves and streams reveal a pronounced variation in scallop (or
flute) size with the variation of the apparen t rate of flow . This has been
ver i fied by experi ments (Goodchi l d and Ford, 1971; Blumberg and Curl, 1974;
Curl, 1974), where mathematical relationships between channel (conduit)
geometr y , scallop len gth and wate r velocity have been derived as in Curl 's
(197 4) formula:-
u=
(V/L 32 ) Re *
[:2 . 5
flow and runoff regimes, catchment areas and competence for sediment transport
and deposition).
Phreatic passages with actively forming scallops are by definit ion
permanently submerged (unless they function as paraphreatic (floodw ate r)
overflow fubes), and the only way to observe them is by cave diving. The
purpose of this paper is to report some preliminary results of scallop and
passage morphometry from an active phreatic system ~n Northern Norway. It
alSo demonstrates the great scientific contribution that cave diving may make
by in si.t:u observations of such phreatic processes.
AREA AND LOCALITY DESCRIPTION
The cave system studied is the underground outlet of Lake Glomdalsvatn
at SVartisen, North Norway. The lake is situated in a shallow glac i a l basin
at the junction of the two principal valleys in the area, Austerdalen and
Vesterdalen. The cave is the only known outlet for the lake, except for a
diffuse flow through glacial drift, which is probably negligible.
The lake has a catchment area of 27.7 krn 2 with both karstic a nd nonkarstic runoff (Fig. 1). In periods when the Svartisen icecap expanded,
glacier -dammed water was also diverted into the system. A drainage tunnel
was blasted to empty the glacier-dammed lake in the mid-1950s and there is
no glacial runoff in the present catchment. The extent of the present glacier
indicates that this would have been the case even if the tunnel had not been
constructed.
Fig .1. The lake Glomsdalvatn, the cave system and its catchment area.
Shaded area: marble with karst features;
(1) Glomsdalvatn;
(2) Lake outlet;
(3) Flood overflow channel at surface .
98
'"'"
SECTION
VallerGralla
/
Middle
Entrance
Surface
Diving
Survey
GLOMDALSVATN
Grode I and 3 .
Fig.2.
- 16m.
,..~--
OUTLET
A. Ive. 8 . Wilkinson
1981.
S ..E. Lauritzen
1977-82
-20m .
Gallum's
Lair
_~ake
Jordtulla
SINKHOLE (JORDTULLA
UNDERGROUND
-23m .
-17m .
MIDDLE ENTRANCES
500m.
~
~~~
Resurgence
LONGITUDINAL
Ar 8r
-8m
grade
/)
OUTLET
~r=
Cf
\
~M. N .
8.
10m.
-P
? --
Mica
Schist
- 15m.
Tinker
8ell
-13m.
Green Rising
8CRA
GLOMDALSVATN
RESURGENCE
A. ~
PLAN '
SECTION '
The lake possesses a dry surface overflow channel, which will take flood
discharge, limiting the maximum hydraulic gradient (and thus the discharge)
through the cave system. Present-day snowmelt floods exceed this limit,
which suggest that a wide range of flow conditions still exist in spite of
the loss of the pro-glacial drainage from the system.
In a previous study of the caves in the region, Horn (1947) referred to
this particular system and more recently Renwick (1962) suggested a probable
corrosion rate estimate. Since 1980, hyd rochemic al and hydrological records
have been conducted on a regular basis (Lauritzen 1981). The lake outlet
drain$ underground into a whirlpool at the marble/mica-schist contact and
emerges as a Vauclusian spring 500m downstream. From this point, the water
eventually joins with the main river of the area, Glomaga, (Fig. 1).
CAVE DIVING
The waterfilled cave may be inspected through suface shafts at three
locations along its course, (Fig. 2). In all these locations, water emerges
from upstream ascending tunnels and disappears into descending tubes downstream. In 1981 and 1982, a cave diving project was undertaken to investigate
the nature of the phreatic conduits. The objective was to explore and survey
the passages and to measure scallops . as a complementary study to the hydro chemical corrosion studies in progress (Lauritzen, in preparation).
The sinkhole is about 7m in diameter at the whirlpool and descends steeply
to at least -23 m where exploration stopped in 1981 (Fig. 2). The Middle
Entrances are phreatic tubes descending down to two "lakes" connected by a
narrow channel at the watertable. Upstream, the water emerges from a subcircular tube, rising water from at least -20 m. The middle Entrances are
also connected by a phreatic loop of 17 m depth. Further downstream, the
water descends down to at least - 16 m, where exploration was halted. The
second entrance, "Valtergrotta", has not yet been explored by diving , but it
shows the same features ("lake" and descending tubes) as the Middle Entrances.
The resurgence, or "Green Rising" (Fig. 3), descends steeply to a - 8 m and
more gently to -13 m before the 6 m diameter tube turns upwards to a watertable
intersection (Tinker Bell). The water emerges from a new loop which waS
explored to a depth of -15 m before it ascends towards the watertable. The
passage is partially controlled by the over-hanging marble/mica-schist contact,
a general tendency that has been observed elsewhere in the area (Jenkins, 1959).
So far, the cave diving confirms the impression that the system is almost
totally phreatic, possessing relatively deep (up to - 23 m) phreatic loops
(Ford, 1971 and 1977), which appear to intersect the watertable at regular
intervals; i.e. it is a State 2 looping system as classified by Ford and
Ewers (1978). The existehce of watertable features at the resurgence and in
the Middle Entrances supports this interpretation (Figs. 2 & 3).
SCALLOP AND ' PASSAGE MEASUREMENTS
The passage waS measured at the lower part of the first phreatic loop
upstream of the resurgence, a place where it possesses a sub-circular cross section, 6.0 m in diameter, (Fig.3).
At the time (August 1982) the cave was
flooded which permitted measurement of only a few (11) scallops. Theklengths
varied between 7 and 14 cm, with a few scallops of about 4 cm length in the
vicinity of a schist flake.
Based on these data, scallop length, using the Sauter-mean (L32, Curl, 1974)
waS calculated to 10.6 cm. The mean flow velocity and total vOlumetric discharge
were calculated using Curl's(197 4) equation as 43.6 cm per second and 12.3 m3
per second respectively. The maximum and minimum discharges inferred from the
scallop sample were 8.6 and 20 m3 per second. The detailed results are compiled
in Table 1.
100
TABLE 1
Scallop discharges inferred for Glomdalsvatn underground outlet (+ 50 C)
Minimum
7.0
Sauter-mean (L32)
10.58
Maximum
14 . 0
Scallops
( cm)
Flow velocity
43.6
70.7
Discharge
(m 3 s -l)
+5.23
12.3 -3.70
20.0
8.62
2) Estimated snow-melt runoff give a good fit to the discharge deduced from
the majority of scallops recorded. This result is comparable to denudation
studies in sub-arctic environments, where the majority of the annual corrosive
denudation takes place during the snow-melt period.
3) These preliminary conclusions must be further tested by extensive
exploration and meaSurements of passages and scallops in the rest of the
system. The current hydrologic and hydrochemical records in the cave will
further provide a check on the proposed relationship between discharge and
corrosion rate.
4) This work also demonstrates that quantitative results of significant
scientific value may be easily collected by skilled cave divers. Diving makes
it possible to observe fundamental cave-forming processes, and we hope this
will encourage cave divers in other areas to make similar studies . It is
likely that scallop-forming discharges will relate differently to the mean
annual runoff ' under other climatic regimes;
and we are waiting anxiously to
learn about furthe r results around the world!
Details on how to measure
scallops and passages may be obtained from S. E. Lauritzen.
ACKNOWLEDGMENTS
Glynn Bolt, Valter Glomdal, Marion Dale, Kai Lyng, Shirley and David st . Pierre assisted
in the field. The cave diving was performed by Andrew lve and Barry Wilkinson in 1981 and by
Andrew and Ruth lve in 1982. Financial support for the fieldwork was provided by
Fridtjof Nansens' and the affiliated Funds for the Advancement of Science and the Humanities;
Rana Museums-og Historielag and by the Ghar Par au Foundation. We owe them all our sincere
thanks. This is contribution No.8 of the karst research project in Norway.
REFERENCES
Blumberg, P.N. and Curl, R.L. 1974. Experimental and theoretical studies of dissolution
roughness. J. FLuid. Mech. 65, 735-751.
Curl, R.L. 1974. Deducing flow velocity in cave conduits from scallops.
NatL. Speleol. Soc .
Bull. 36, (2), 1-5.
Ford, D.C. 1971. Geologic structure and a new explanation of limestone cavern genesis .
Trans . Cave Res . Group of Great Brit . l3 (2), 81-94.
Ford, D.C. 1977. Genetic classification of solutional cave systems. Proc. 7th Int. Speleol.
Congr. Sheffield, 189-192.
Ford, D.C. and Ewers, R.O. 1978. The development of limestone cave systems in the dimensions
of length and depth. Can . J . Earth Sci. L5, 1783-1798.
Goodchild, M.F. and Ford, D.C. 1971 . Analysis of scallop patterns by simulation under
controlled conditions. J. GeoL . 79, 52 - 62.
Hagen, I. 1977. Isohydatkart, Svartisutbyggingen . Norges Vassdrags-og Elektrisitetsvesen ,
Hydrologisk avdeling , Oslo .
Hellden, U. 1974. Karst; en studie av Artfjallets karstomraade samt jamforande
korrosionsanalyser fran Vastspetsbergen och Tjeckoslovakien. Medd. Lunds Univ . Geogr. Inst.
AvhandZ. LXXII, 192 pp.
Horn, G. 1947. Karsthuler i Nordland. Norges Geol . Unders. L65, 4-77.
Jenkins, D.A. 1959. Report on the C.U.C.C. expedition to Svartisen, Norway 1958. Cave Sci.
4 (29), 206-228.
Lauritzen, S.E. 1981. A 'study of some karst waters in Norway. Spatial variation in solute
concentrations and equilibrium parameters in limestone dissolution . Norsk geogr . Tidsskr.
35, 1-19.
Lauritzen, S .E. 1982. The paleocurrents and morphology of Pikhaaggrottene, Svartisen, North
. Norway. Norsk geogr . Tidsskr . 36, 183-209.
Pizarowicz, J.A . a nd Maslyn, M. 1981. Empirical confirmation of Curl's flow velocity
calculations, 1974. Proc. 8th Int. SpeleoZ . Cong ., Bowling Green, Kentucky, U.S.A.,772-774.
Renwick, K. 1962. The age of caves by solution. Cave Sci. 4 (32), 338-350.
White, W.B. 1977 . Role of solution kinetics in the development of karst aquifers. Mem . Int.
Assoc. HydrogeoL . Z2, UAH press, Univ . of Alabama, Huntsville, Ala., pp. 503-517 .
M.S. Received 15th December 1982
CAVE SCIENCE
Trans. British Cave Research Association. Vol.10, no.2, pp .103-ll5. June 1983.
INTRODUCTION
The purpose of this paper is to record the results of some thirty 230T h/234u
determinations of speleothems from caves in the Peak District of Derbyshire, to
suggest how the results may be interpreted in the light of morphological
sequences of events both surface and unde rground and to suggest correlations of
those events with the British Quaternary succession.
Waters & Johnson (1958) have used the sequence of river terraces in the
valley of the River Derwent in Derbyshire to establish a relative chronology
for events in the later part of the Pleistocene. In the absence of any
isotopic dates from the area, dating of these events in relation to the
glacial/interglacial stages has so far been a matter for conjecture, (Johnson,
1957; Waters & Johnson, 1958) largely based on the presence of till with
Lake District erratics on the Hathersage terrace around Bakewell, and a suggested
correlation of that till with the Pleistocene sequence to the west of the
Pennines (Straw & Lewis, 1962).
West-bank tributaries of the Derwent mainly drain from the limestone outcrop
of the White Peak with its numerous cave systems (Beck 1975; Ford 1977).
Some of these caves contain bone-bearing deposits (Bramwell 1977), most of
which are of Devensian or post-glacial age , so that they throw little light
on the morpholo gical development of either the limestone landforms or the caves.
The isolated, incompletely excavated and now lost, Victory Quarry fissure,
north of Buxton, has alone shown evidence of early Pleistocene faunas, probably
of Cromerian age (Spencer & Melville, 1974), or perhaps even earlier in the
Bramertian stage of the Norwich Crag (Stuart, 1982).
CAVE MORPHOLOGY IN RELATION TO SURFACE FEATURES
. Morphological studies of the distribution of cave passages, their altitudes,
sedimentary fills and stalagmite deposits led to the establishment of a
relative sequence of events in the caves of the Eyam-Stoney Middleton area
(Beck, 1975), and later work showed that strikingly simil ar situations existed
in the Bradwell, Lathkil1 Dale, and Castleton areas (Ford, 1977; Beck,1980).
The sequences are summarised in Table 1. The location of the caves sampled
is given in Fig. 1.
The cliffs on the north side of Stoney Middleton Dale and the walls of two
tributary valleys from the north, Eyam Dale and The De1f, contain many short
caves and one large cave system, Carlswark Cavern. The latter is active in
part and continues westwards as Streaks Pot, from which it is separated by
flooded passages beneath The Delf. Four cave levels can be recognised, two of
which lie some distance above the main valley floor throughout. Each level
consists of a network of phreatic tubes, with large elliptical strike tubes
103
km
t50
'-----'
MIDDlETON
Sarah's Cave
Water Icicle
Close Cave .
o
I
Fig. 1.
Kilometres
,
,
5
I
and smaller dip-tubes of more circular cross section. The lack of extens ive
vadose passages in the upper levels suggests relatively fast abandonment a nd
capture by the next level beneath. Each phase of capture therefore appe a r s to
be related to an external erosional event which rapidly lowered the outle t level
in the vall:ey.
In Carlswark Cavern all four levels can be recognised. Access between levels
is generally via joint - oriented capture points. On the upstream side of c a pture
points there is often a vadose trench in the floor of the upper tube for a
short distance, while the unmodified tube is seen downstream. The upper levels,
comprising
Ivy Green Cave, Sarah's Cave, the Holerin-the-Wall and fragments
in Merlin's Cavern and Carlswark's First and Second Remnant Complexes (Beck,
1975) carry only small percolation inlets, but the third level, the Carlswark
Complex, lying between 175 and 184 metres O . D., carries a large stream which
falls, again via a joint - oriented capture point, into the Lower Complex at the
eastern end. This only now occurs in wet weather due to modif ication of the
drainage pattern by lead miners' soughs;
the stream is now captured by Moorwood
104
....
VI
Anglian
Hoxnian
Stillstand
Vadose
downcutting
Vadose downPost-glacial
cutting through
fill and
speleothem
deposition
locally
Devensian II
Sand, silt
and Clay- fills
Lower Complex
(167 m)
Lower Bagshaw
Complex (181 m)
Local fills
Mid-Devensian
interstadial
Speleothem
deposits
and limited
downcutting
Extensive fill
Local fills
Devensian I
Sand, Silt
and Clayfills
Bradwell Parish
Complex (203 m)
and Bagshaw
Cavern
Speleothems on
fill
Extensive fills
Ipswichian
Vadose
incision
Speleothems on
fill
Carlswark
Complex (178m)
Wolstonian
Stillstand
Hazlebadge
Complex
(209 ml
No evidence
First Remnant
Complex important. (200 ml
Cromer ian
Vadose
downcutting
Second Remnant
Complex (185 ml
Speleothems in
Sarah's Cave and
Hole in the Wall
Hartle Dale
Complex
(274 m)
Early phreatic
tubes.
Lower
Pleistocene
Bradwell
Phreatic
Solution
Eyam
stOney-Middleton
Probable
Phase
Sta~
Swallets choked in
Devensian re-excavated
(186 m)
Local fills
Speleothems in Peak,
Speedwell, Giants Hole
and Winnats Head (lower)
Incision to 218m;
Main cave systems;
Peak Gorge Initiated
Speleothems in
Winnats Head Cave.
Earliest Tubes
[above 303 m)
Phreatic development
in mineral veins.
Castleton
Limited incision
No evidence
Lathkill Dale
suggested Correlation of Developmental Stages in the Major Catchment Areas with the British Pleistocene sequence
Table 1
Fills in lower
Poole's Cavern.
Speleothem in Thirst
House Cave
Poole's Cavern
incised (350 m)
Poole's Cavern
route initiated
(above 357 m).
Karstification at
shallow depths
Speleothems in
Foxholes ?
Buxton
20 8759
..,
1_
/7/
AO.D.
700
600
SOD
....!....a . . . . . . .
:"
... .
- _
eoofl
700
_-"'""::::::;0_'---
600
Carlswor1< Comple.
lower Complltx
Fig. 2.
Cave levels of Stoney Middleton Dale, Eyam Dale and The Delf.
eleyotion
A.O D.
metres. feel
1000
300
Hartle Dale
Caves .
bench
~_ .L
knickpolnt
250
Bagshaw
Covtrn
flOor 01
rok.
Hallebadg.
/-forll.
-1 '
N.'herwot.r
200
Swallp(
Ear"1
Rake
150
500
Fig. 3.
Bradwel1
Parish
Cove
001.
107
'"
...Dl
Hop.
Terrace .
Brodwel1
n,
Lower Series
Brodw.l1
01 Bagshow
Cavern .
Resurgence
m("tn.' s
, 1
AO 0 ,
3SO
A.O.D
1100
-- -
1000
]00
900
800
700
250
200 .
Hope
600
For Sump. Ppok
500
Cav.rn .
ISO
Peakshotv
Water
Fig. 4.
Hope T.rron"
Holhersogp T(lfToce
: ;'
l.Jrw'Iam(>d
Waler ICiclE'
~r ";~:~
i:.~'
R. surgenc.
~::~IOW
spnngs
/'
;7
,/
7 777
.?777~
LolhklU
Date
Lolhkill
Lower
Head
Coles
Cov.
Dol.
Cav
Col.s Dol.
Coles Da le (hanging south west branch I
Fig . 5.
108
the outlet elevation rather than an internal feature of the cave. Recession of
the impervious cover had caused this route to be abandoned before furthe r
lowering of the outlet took place.
Changes in the nature of both the Peak and Speedwell stream passages from
narrow vadose canyon to wider passage with a greater degree of breakdown suggest
a further standstill in outlet elevation at app roximately 206m.0.D.The downstre a m
limit of vadose incision in both streamways lies at approximately 195 m 0.0.,
demonstrating clearly that the outlet also stood at this elevation for a
considerable time while the lower parts of the stream canyons were being incised.
The present risings lie at 187 m 0.0. in the Peak Cavern Gorge, the lowest
route in both caves being unable to carry the flood discharge. Water in Peak
Cavern backs up to the 195 m level to flow via its earlier route through the
present show cave, and is joined by overflow from Speedwell via a route in which
a canyon is also incised down to 195 m.
Close to the surface watershed lie the large and mainly inactive influent
caves of Blue John, Treak Cliff and Winnats Head. They lie at increasing
distances from the shale margin, Winnats Head on this basis being apparently
the oldest, Blue John the most recent. All show major vadose features throughout, but none penetrates below 259 m 0.0.
The swallet caves in th~ Rushup Vale to the west of the surface watershed,
include Giants Hole, Jackpot (P8) and Gautries Hole. Though they all show vadose
incision, particularly in Giants Hole, they all end at sumps close to the
altitude of the upstream sumps in the Peak-Speedwell Cavern system, indicating
the presence of an as-yet undrained phreatic drainage system. The altitude of
the sumps' water surface at about 244 m is thus only a reflection of the amount
of vadose incision in the middle parts of the Peak and Speedwell systems.
The resurgences at Castleton are graded to the Hope Terrace, the higher
Hathersage Terrace not being traceable upstream of Hope.
The Lathkill Dale catchment area occupies a central position in the limestone
outcrop and drains eastwards. It contains some extensive caves, but much of the
active system is flooded except in drought. Unlike the other catchments, the
caves of Lathkill Dale have no allogenic inputs other than tiny trickles from
the small lava outcrops.
Lathkill Dale begins as a dry valley network in the topographic and structural
basin centred on the village of Monyash and runs eastwards to join the River Wye
near Rowsley. The source of the River Lathkill in normal weather is Lathkill
Head Cave or small risings a little further downstream at approximately 198 m 0.0.
It appears that before miners' sough drainage lowered water levels in the a rea,
Lathkill Head Cave discharged a permanent stream, but in dry weather it is now
possible to enter the cave and reach a phreatic stream passage. In drought this
has been followed westwards for some 600 metres. In the downstream direction it
becomes too small to follow, but ends close to Lower Cales Dale Cave, which also
discharges a large wet weather stream.
Along the flanks of Lathkill Dale and Cales Dale are many small caves lying
between 235 m and 244 m 0.0. (Fig. 5). They lie close to the highest line of
cliffs at the confluence of the dales, and are often associated with a prominent
bench. They include Upper Cales Dale Cave, Lynx Cave and One Ash Cave.
The highest tube network is represented by a small cave in a field to the
east of Cales Dale, and by the large passages of Water Icicle Close Cavern to the
southwest. They lie at 305 m 0.0. and contain an extensive fill of broken
limestone, c~ay and flowstone, giving them an ancient and mature appearance.
Lathkill Dale is graded to the Hope Terrace, having cut through a thick till
sheet on the Hathersage Terrace at Alport, southwest o f Bakewell. Neither of
these terraces can be traced far enough up valley to be related to the cave levels
though there seem to be some poorly developed breaks in the thalweg (Fig. 5).
The evol~tionary development of th~ cave systems and their individual passages
from phreatic to vadose forms depends on successive lowering of resurgence points
which act as controlling base levels in the karst drainage basins. Such levels
in turn depend on successive phases of downcutting of the River Derwent, though
unfortunately the lower terraces . noted by Waters and Johnson (1958) do not
extend as far north as these areas, making correlation with th~ caves and their
deposits impossible. The developmental stages discussed herein are summar ized
in Table 1.
.
The local base level for all the caves is the head of incision of the Hope
Terrace, variously incised by up to 50 m below the till-covered Hathersage
Terrace. The conversion of cave systems from phreatic to vadose may thus have
started during the incision to Hathersage Terrace levels, being suspended during
the glacial phase represented by the till sheets and then recommenced to lower
109
levels during the incision to the Hope Terrace. Vadose modification has
continued intermittently since this phase as shown by sedimentary fills and
speleothem deposition.
CAVE SEDIMENTS
The caves themselves show certain morphogenetic features which are broadly
indicative of climatic conditions on the surface:
phreatic solution - (water-table high): early phases of development before
incision of the River Derwent, and continuing to the
present below successive water tables.
vadose incision into rock - (high energy run-off through the caves): heavy
precipitation and/or melt water, interglacial or
waning glacial climate.
gravel, sand and silt fills - (moderate energy run-off):
proglacial or deglacial conditions.
probable periglacial,
Table 2
U
Age
/o
Cave
Samp le No
Description
Analysis No
iJ2.runl
Treak Cliff
Cavern
78000
-1
(bulk )
7 .0
0.993
64
0 .682
125 6
78001
-1
(bu1 k)
9.8
1.129
121
0.713
131 4
78002
-1
(bulk)
6.2
1.046
>1000
0.692
126
-2
(s c )
20.8
1.147
187
0.843
186 7
-3
( fs top)
11 . 6
1.109
142
0.846
191 + 15
- 13
17 .6
1.182
557
0 . 828
176 + 8
35.7
1.287
84
0.076
2*
Wi nnats :Iead
Cave (u ~ ., er
series)
78004
78020
broken fs
slab contains
broken sc
broken sc
-2
(x 10 3 y B. P )
- 7
(outer)
\, i nnats Head
Cave
( lower series,
30m be l ow
78004)
80032
-2
broke n sc,
Fox chamber
(bulk)
80039
fs veneer,
Fo x cha mber
-1
(bulk)
74.8
1.168
151
0.397
54
Speedwell
Caverns
(Bung Hole
series)
80027
broken fs in
bou l ders
-1
(base)
0.8
1 .208
11
0.598
96 4
800 28
fs cementi ng
bou1 ders
-1
(top)
0.6
1.447
0.145
17
Sarah's Cave
78022
broken sc
-1
(bu1 k)
1.2
1.057
0.804
ca ve
7R023
sq/fs f rom
st reamway
-1
(base)
0. 9
1 . 334
0.198
Foxholes
Hi gh ',hee 1don
78044
fs floor,
under lyin g
Ipswichian
fauna
-1
(bulk)
1.5
0 .996
15~
0 .9 72
Pinda1e Cave
78045
sc in 10wlevel tube
-1
(bulk)
1.2
0.914
132
0.912
Peak Cavern
79006
br oke n fs in
stream canyon
-1
(bulk)
6.9
1.290
43
0.427
59
79007
eroded fs,
Victoria
Gallery
-1
(bulk)
1.0
1.473
73
0.384
51
-1
(bulk)
3.4
1. 240
28
0.500
73 ' 2
51 2
Me r 1 i n's
79008
Giant's Hole
eroded fs,
near to 79007
172 + 21
- 18
24
>350
79009
broken fs,
near to 79006
-1
(bulk)
3 .4
1.447
> 1000
0. 382
. 80031
fs on mud fill
-1
(top)
. 13.5
1.143
59
0 . 010
1.1 0 .1
80029
fs on pebble
fi 11
-1
(top )
2.6
1.099
144
0.031
3.4
80030
fs near
entrance
-1
(base)
a.5
1.035
52
0.147
17 2*
80033
f s near
Maginn's
137.6
1 .262
1A5
0'.398
54
(bulk)
-3
0.1
Rift
80034
fs, over1 i es
80034
-1
(bu1 k)
78 .3
1 .242
617
0.35 9
48 1
80037
fs at Gi ant's
Windpipe
-1
(hulk)
1.9
1.145
0.699
125 + 22
- 19
80041
sg in Upper
series
-1
(base)
1.9
1.075
12
0033
3.6 0.2
80043
fs in Upper
Se ri e s
-1
(base)
1.7
1.013
41
0 :020
2.2 0 . 2
Th i rst House
Cave
80035
fs be l ow
arc heo
Deposit
-1
(bulk )
0'.4
1 . 123
0.620
102
lIo1e - i n-the-
80042
sg on fi 11
-1
(top)
1.0
1 . 121
17
0 .85 5
195 + 14
- 13
80056
f s on wall
co ntai ns
hiatus
145 + 17
1.1::. 11
'.'ater l ci c1e
Close Mi ne
80057
sg on f1 oor
contai ns
hi atus
-1
(top )
0 .2
1.112
10
0.751
-2
(base)
0 .3
1.015
124
1 .027
>350
-1
(base)
0.2
0.990
22
1.042
>350
-2
0.1
1 .082
13
0.891
fs
f1 0wst one
-14
(top)
sg
s ta 1aogmite
sc
s ta 1act i te
225 + 64
- 41
1tl
6
II
.......
II
4
".,
~3
<l:
Z O~~-L4-~--~~~~--~~--~~~~__~~LL+-~~~~~~~~~~~~~~~~~
o
100
120
20
40
60
80
320
>350
140 160
200
240
280
Fig. 6.
Bar graph (upper part) of 23U h/ 234 u ages with ~la error
limits for Derbyshire speleothems, and histogram (lower part)
of frequency distribution of ages showing relationship to
marine isotopic stages (numbered at top). Unshaded parts of
histograms and crosses on bars indicate ages influenced by
detrital Th contamination.
Table 3
Su ested correlation of speleothem a e groups to the British Quaternar
sequence of Mitchell et al. (1973
Group
Duration (y B.P.)
o -
I - II
II
II - III
III
III - IV
IV
V
17,000
45,000
75,000
90,000
145,000
170,000
>350, 000
17,000
45,000
75,000
90,000
145,000
170,000
225,000
Stage
Flandrian to Late Devensian
Devensian II glacial
Mid - Devensian interstadial
Devensian I glacial
Last inte rglacial (? Ipswichian)
Penultimate glaciation (? Wolstonian)
Penultimate interglacial (? Hoxnian)
(? Cromerian)
1.
2.
3.
4.
170,000 to
150,000 Y B.P. (part of marine isotope stage 6)
Bakewell till deposited on Hathersage ~err ac e, clay-fill in phreatic
and vadose passages of main levels of Castleton and Stoney Middleton
cave systems. Start of incision below Hathersage Terrace during
waning phases? Penultimate glaciation (Wolstonian?).
5.
145,000 Y B.P.
(parts of marine isotope stages 5 and 6 , particularly
5e); Incision of valleys to Hope Terrace level. Vadose trenches
cut through clay and other fills. Vadose deepening of stream
passages at Castleton and capture of drainage by new phreatic
complexes at Stoney Middleton, in the Lathkill Basin, and at Bradwell.
Stalagmitic deposition in many caves. Last interglacial (Ipswichian).
6.
7.
75-45,000 Y B.P.
stages 3 and 4):
downcutting.
113
8.
9.
Although the stalagmite dates from Peak District caves are of value in
establishing a time span for the formation, infilling, abandonment, and
degradation of the major cave levels, as yet they only provide a tentative
basis for correlation of the caves with the till sheets of the area, and
with most of the terraces of the River Derwent (,Waters & Johnson, 1958).
Knick point recession has almost always stopped short of the karst area,
except on the River Wye, where there are no significant cave systems.
At Castleton, the highest terrace, the Hathersage Terrace, is uncertainly
traceable above Hope and the Peakshole Water is incised into the solifluction
sheet on the Hope Terrace, thereby being at the altitude projected for the
Hathersage Terrace at Castleton. The gorge at the entrance to Peak Cavern
shows evidence of having been a vauclusian spring (Ford, 1977) initiated
during the phreatic development of Peak Cavern, probably during the Hoxnian.
The lip of the spring was apparently cut down during the formation of the
Hathersage and Hope Terraces . Thus the deep vadose downcutting within the
cavern probably dates from the Ipswichian, with fill and re-exc avation
phases during the Devensian, as shown by speleothems of mid - Devensian age ,
deposted both on fill and on the lip of the vadose trench.
At Bradwell, the present rising lies close to the elevation of the Hope
Terrace, but this conclusion can only be reached by upstream projection of
the Hope Valley floor . At Stoney Middleton, the Hathersage Terrace lies
between 140 and 150 m O.D., well below the elevation of the lowest c ave
level. The stalagmite dates do not conflict with the general opinion that
the Hathersage Terrace was covered with till of Wolstonian age. The climatic
sequence suggested by both morphology of the caves and by the dates can
then be tentatively related to geomorphological events as in ~able 1.
The small vertical range of the four cave levels in the main areas
(Beck, 1980), coupled with the speleothem d at es greater than . 350,000 years
from the highest mature remnant caves in the central area, suggests that
underground drainage was well developed in the exposed area of limestone
before that time, but that some recession of the marqin of the shale cover
has taken place since.
.
CONCLUSIONS
It is concluded that uranium series dating of speleothems in Derbyshire
caves demonstrates a sequence of events relating to changing climatic
conditions during the Pleistocene. Progressive lowering of the water-table
permitted cave development at successively lower altitudes thou gh abandonment
of older levels is nowhere complete . A partial correlation with river
terraces can be proposed though knick- point recession has not gone far
enough for the lowest river terraces to be reflected in cave morphology.
Depositional phases of speleothems are separated by hiatuses which appear to
corr elate with glacial episodes. Some correlation with the marine isotopic
record is also possible.
ACKNOWLEDGMENTS
We thank Noel Christopher and Richard Shaw for assistance in collecting stalagmite
samples. Professors H.P. Schwarcz and D. C. Ford are thanked for providing the facilities
for speleothem dating. One of us (J.S.B.) received an NERC research studentship dur ing
par t of this work.
114
REFERENCES
Atkinson, T.C. , Harmon, R.S . , sm~~5' P~~4 & Waltham, A.C. 1978. Palaeoclimatic and
geomorphic implications of
Th/
U dates on speleothems from Britain. Nature ,
vol. 272, No. 5648, pp. 24-28.
Beck, J.S. 1975. The caves of the Foolow-Eyam-Stoney Middleton area, Derbyshire, and their
genesis. Trans Brit. Cave Res. Assoc. , vol. 2, No.1, pp 1-11.
Beck, J.S. 1980 . Aspects of speleogenesis in the Carboniferous Limestone of North Derbyshire.
Unpubl. Ph.D. Thesis. Univ. Leicester.
Bramwell, D. 1977.
Broecker, W.S. & Van Donk, J. 1970. Insolation changes, ice volumes and the 18 0 record in
deep sea cores. Rev. Geophys. Space Phys .. vol. 8, pp. 169-198.
Ford, T.D. 1977 . Limestones and Caves of the
Norwich. 469 pp.
Geo-Abstracts Ltd.,
Gascoyne, M. 1980. Pleistocene climates determined from stable isotope and geochronologic
studies of speleothems. Unpub. Ph.D. Thesis, McMaster Univ., Hamilton, Ont., Canada.
Gascoyne, M. 1981. Chronology and climate of the Middle and Late Pleistocene f.rom
speleothems in caves .in Northwest England. Quaternary Newsl. No. 34, pp. 36- 37.
Gascoyne, M., Schwarcz, H.P. & Ford, D.C., 1978. Uranium series dating and stable isotope
studies of speleothems: Part I. Theory and techniques. Trans Brit. Cave Res. Assoc.
vol. 5, no. 2, pp 91-111.
Johnson, R.H., 1957. An examination of the drainage pattern of the eastern part of the Peak
District of Derbyshire. Geogr. Stud. vol. 4, pp 46-55.
Johnson, R.H. 1967. Some glacial, periglacial and karstic landforms in the Sparrowpit-Dove
Holes area of North Derbyshire . East Midi. Geogr. vol. 4, no. 4, pp 224- 238.
Kominz, M.A., Heath, C.R., Ku, T.L. & Pisias, N.G., 1979. Brunhes time scales and the
interpretation of climatic change. Earth Planet Sci. Lett. vol. 45, pp 394-410.
Mitchell, G.F . , Penny, L.F., Shotton, F.W. & West, R.G. 1973 . A correl~tion of Quaternary
deposits in the British Isles. Geol. Soc. London. Spec. Rept. no. 4, pp 1- 99-. -Shackleton, N.J., Opdyke, N.D. 1973. Oxygen isotope and magnetic stratigraphy of equa~orial
5
Pacific core V28- 238: oxygen isotope temperatures and ice volumes on a 10 and 10 year
scale. Quat. Res. vol. 3, pp 39- 55.
Spencer, H.E.P. & Melville, R.V., 1974. The Pleistocene mammalian fauna of Dove Holes,
Derbyshire. Bull. _C;eol. Surv. G~ . No. 48, pp 45-53.
Straw, A. & Lewis, G.M., 1962. Glacial drift in the area around Bakewell, Derbyshire.
Midld. Geogr. vol 3, pp 72-80.
Stuart, A.J., 1982.
M. Gascoyne,
Atomic Energy of Canada Ltd.,
Pinawa,
Manitoba,
Canada.
115
East
John S. Beck,
Geology Department,
University of Leicester,
Leicester LEl 7RH.
NOW A VAILABLE
LIMESTONES AND CAVES OF THE PEAK DISTRICT
compiled and edited by T. D. Ford
published by "Geo-books " c / o Geo-Abstracts Ltd.
University of East Anglia , Norwich
469 pages . 11.50 soft back; 15 hard back.
Available from B. C. R. A. Sales
B.C.R.A. Sales
CAVE SCIENCE
The TRANSACTIONS OF THE BRITISH CAVE RESEARCH ASSOCIATION covers all aspects of
speleological science, including geology, geomorphology, hydrology, chemi stry ,
physics , archaeology and biology in their app l ication to caves , as well as
techno l ogical matters such as exploration, equipment , s urveying, photogr aphy a nd
documentation. Papers may be read at General Meetings held in various parts o f
Br i tain, but they may be submitted for publication without bei ng read. Manuscripts
should be sent to the Editor , Dr. T . D. Ford , Geology Dept ., University of Leices t e r,
Leicester LEl 7RH, who will be pleased to advise in cases of doubt about the
preparation of manu scripts. The Tr a n sactions i s normally iss u ed four times a yea r
to pa i d-up membe r s of the British Cave Research Association . Subscri ptions a r e
due on January 1st annually .
Full Members 10.00 per annum
Joint Members 12.50 per annum
Corporate Members & Clubs 12.50 per annum
COUNCIL FOR 1983
President:
Cha i rman :
Depu ty Chairman: J . R. Wooldr i dge, 129a Park Hil l Road , Harborne, Birmingham . B17 9 HH
Secretary: Dick Willis, Basement Flat, 6 Worcester Terrace, Clifton, Bristol. BS8 3J W
Membership Secretary: D.R.Stoddart , 23 Claremont Avenue , Bishopston , Bristol ,
BS7 8JD