Bcra 10-2-1983

Ca e Science
The Transactions of the Briti.h Cav. Rrch A.sociation
I Volume 10
Number 2
BeRA
June 1983
Gunung Sewu
Java
Scallops in
Norway
Peak
Speleothem
Dates
BRITISH CAVE RESEARCH ASSOCIATION

NOTES FOR CONTRIBUTORS
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).
ISSN 0263-7 60X
CAVE SCIENCE
TRANSACTIONS OF THE BRITISH
CAVE RESEARCH ASSOCIATION
June 1983
Volume 10, Number 2
CONTENTS
The caves of Gunung Sewu, Java
A.C. Waltham, P.L. Smart, H. Friederich,
A.J. Eavis
&
T.C. Atkinson . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Mean annual runoff and the scallop flow regime in a

subarctic environment:
Preliminary results from
Svartisen, North Norway

S-E. Lauritzen, A.lve
&
B.Wilkinson . . . . . . . . . . . . . . . . . . . 97
Speleothem dates and Pleistocene Chronology in the Peak

District of Derbyshire
T.D. Ford, M. Gascoyne
Cover photo:
&
J.S. Beck . . . . . . . . . . . . . . . . . . . 103
The 70 metre shaft in Gua Lebak Bareng by A.C. Waltham
Published by and obtainable from

The British Cave Research Association
30 Main Road
Westonzoyland
Bridgwater
Somerset TA7 OEB
~ Copyright the British Cave Research Association, 1983.

No part of this publication may be reproduced in any other
publication/used in advertising, stored in an electronic retrieval
system, or otherwise used for commercial purposes, except for single
copies for research purposes, without the prior written consent of
the authors and of the Association.
CAVE SCIENCE
Trans. British Cave Research Association. vo~.~o, no.2, pp.55-96. June 1983
THE CAVES OF GUNUNG SEWU , JAVA
by A. C . Waltham, P. L. Smart, H. Friederich, A. J. Eavis & T. C. Atkinson

ABSTRACT
Gunung Sewu is an area of spectacular limestone cone karst in southern Java. Hundreds
of cave entrances are known to exist, and many of them were explored for the first time in
1982. Eight caves have more than a kilometre of mapped passage, and sixteen reach depths
of over 100 metres. Most of the cave streams drain to a single resurgence, and some of
the caves provide valuable water resources.
THE GUNUNG SEWU KARST

Gunung Sewu lies adjacent to the south coast of central Java (Fig. 1). The
limestone hills have an area of over 1000 kID 2 and rise to altitudes of around
500 m, though most of the area is at under 300 m. The name "Gunung Sewu"
translates as "Thousand Hills" and derives from the small limestone cones
which dominate the landscape. Immediately to the north lies the Wonosari
Plateau;
the city of Yogyakarta is 30 kID away and not far beyond is the
active volcano of Gunung Merapi (Fig. 2).
Located just south of the equator, Gunung Sewu has a Warm climate, shielded
from extremes by breezes from the Indian Ocean. Average temperatures are
indicated by that of the water in the caves (27 0 C) and there is remarkably
little variation from this, day or night. Rainfall is mainly in the months of
November to May, and averages around 2000 mm per year. The dry seaSon varies
from three to seven months, and may be completely dry; August 1982 had
cloudless skies throughout. Though the climate may be ideal for caving, the
dry season creates undue hardship for the local population.
Over a quarter of a million people live within the karst. Small villages
are scattered throughout the are a, and practically everyone is occupied bt
farming. Good vOlcanic clay soils in the valley floors are intensively
cultivated while terracing on the thin soils of the limestone hills provides
further land of poorer quality . The critical restriction on economic
development is the total lack of surface water or readily available ground water throughout the dry season, and consequently the population of Sewu are
among the poorest in Java. They are however, hard-working and extremely
friendly people. The whole area is easily accessible by a dense network of
very rough roads almost completely devoid of traffic. A robust vehicle is
essential but few places are more than a kilometre from a driveable track.
Finding cave entrances is no problem, as any local person knows the way to
the nearest luwang (sinkhole).
GEOLOGY
Massive reef limestones of Miocene age support the karst features which
distinguish Gunung Sewu. They have a total thickness of at least 650 m, and
through most of the area dip very gently towards the coast, though they are
structurally more complex along their northern margin. They are underlain
by various volcanics and clastics, and are only overlain by clays mainly of
weathered volcanic ash origin which floor the valleys within the karst.
Towards the north and northeast the reef limestones show a transition into
chalky, bedded, lagoonal limestones whose largest outcrop is on the Wonosari
Plateau (Fig. 2). This boundary is complex and interdigitated but is a major
hydrological feature with considerable influence on the pattern of cave
development within Gunung Sewu. The chalky limestones of the Wonosari Plateau
are basically non-cavernous, though isolated small cave passages are known;
surface streams from the Plateau sink where they meet the reef limestones
. In detail, the Gunung Sewu limestones show a considerable range of lithology.
Compact, fine-grained, cream coloured calcilutites dominate, but autobreccia
structures are common. Some breccias are spectacular, with crystal-lined vugs.
Beds of chalky limestone occur sporadically through the main limestone, but
increase in proportion towards the northern facies boundary where they are
55
:....0.:.
_.,., , _...
v ......
'lldian Ocean
Figure 1
//
100 Ill.
Merapl Volcano
II
II
III
f,
Kar.t area. of Java
,' "
\ \
lawoe Vo lcano
VOGV"K""U
"
,
/
,/
,/
Kall Progo Plain
/1'1
"
Wono.arl Plateau
WOlloa"II,
,...... ,......
~IIAC'''''"TOIIO
"""
~'""'r~~'-'"
~~".....~~~
0'
...--
Co k.r.,
........
le r....... '
. -"'-"
Dr, Ha,
"...,
._'""" ....
~
r--..
",.-....~,
,-...
,-..
r'""
f"'"""<\'-"
,......
~,-...~
..........
,......,
r"',.........
..........
,-..
~~;-....
Figure 2
The Gunung Sewu area
... r."
400 r .,.1
Gunung Sewu
Wono.arl Plateau
.oe
o
Watw table
/
Figure 3
Cro ectlon through Gunung Sewu
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
1. Cone ka rst with an alluviated valley floor in northern Gunung Sewu

(Waltham)
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
.........
Bribln { ' / . ....
r'.
:",
/.-...---'
.~
I ('
~~
~)
. "
\BuhpUtl~cr'-'II'"
/
.. /- ....
.
/
./
Jomblang
. '"
S<m>erejo
" ' .
..
----...-'.
. .. -
-.-->'
..n "..
~SidoIlardjo
Jero
. .
{j:O . .
....'.: ..
;
ct(\'t 'P
--- .
~:~.
.. ,~'~blo.k
~~--+-.
. . ' .'
... 1-- ... ..::'
,Grubug:
.... ... ~ . . . . . . .
..
')J~~.
Bedesan
_ _. - .,.., Seropan
// ~ 0t~ , M~
/ ....
.
Nglngrollg
".
CAVES OF GUNUNG SEWU
------,"","",'
---
kllomelres
........~
Major village
Contour In metres
50m groundwaler contour
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
GUNUNG SEWU CAVE SURVEY
,~
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
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
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
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
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
GUNUNG SEWU CAVE SURVEY,

JAVA, 1982
SITE 2
G.R. 815128
LUWANG
CEBLOK
01"'\ ;"'\
,~
Pool--1
IdVO--'
..)
) I~.....L
liopi
I 8urlici
WoN "' ''''.lIonl
olmF )
AND WELL LOCATION
DETAIL OF ENTRANCE SHAFT
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
Qua 80ng Tawing
ProJected .ectlon
3:f;i\
-
Song Tawing Jomblang Orubug
.-----z.
'0
:f!
'-
'\
650
......,.
Luw,na
Qua Burl Omah
'0
....
W
Jomblan:g
500 metr
Figure 7
- - - - - - - --
>::::2 0
8wt Omah
.~
....
8ucI
Luwang 80nglat
. .. . ... . .. .. . .. .. . . . . ... ... . . . .... .. . . . . . . .... , .. .. . . . . . . .. . . .. ... .
Luwana ~rubua
GRUBUG CAVE SYSTEMS
to aaron
THE suel
. UI
.0
...
-=- '0:
200
metrM above 8M .....
Kall 8uel
0'1
0'1
-40
+20
It
Entrano.
vr
0' ~
' .........., " ,,

\
PLAN
..
Figure 8
"AI' 222222228;;$
.......... "., ,,."'M

__
I
/;'
100 ... tr
.ump
".,
U""... ".,
~
SEMI-EXTENDED SECTION
c.,~
~~
10
,
GUNUNG .S EWU CAVE SURVEY, JAVA, 1982

SITE 191
G.H.611118
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
GUNUNG SEWU CAVE SURVEY, JAVA, 1982
GUA NGINGRONG - GUA MULO
- 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

SITE 104 G.R. 574124
.t,..
0L'__........__..L...---''--...L__.J5,O.:....______________.:..:;1C?0 M
eo
up.t, ........,.
PROJECTED SECTION
100
Figure 10
68
GUNUN G SE1vU, JAVA
4. A water carrier walks to the entrance of Gua Sodong
(Waltham)
5 . The f o ssil passage in Luwang Grubug, looki ng out to the Jomblang ent rance
(E avis)
6 . The ri ve r pa ssag e in Gua Br i bin
(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
@ GUNUNG SEWU CAVE SURVEY
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
GUNUNG SEWU CAVE SURVEY,

JAVA, 1982
SITE 2
G.R. 815128
LUWANG
CEBLOK
01"'\ ;"'\
,~
Pool--1
IdVO--'
..)
) I~.....L
liopi
I 8urlici
WoN "' ''''.lIonl
olmF )
AND WELL LOCATION
DETAIL OF ENTRANCE SHAFT
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
Qua 80ng Tawing
ProJected .ectlon
3:f;i\
-
Song Tawing Jomblang Orubug
.-----z.
'0
:f!
'-
'\
650
......,.
Luw,na
Qua Burl Omah
'0
....
W
Jomblan:g
500 metr
Figure 7
- - - - - - - --
>::::2 0
8wt Omah
.~
....
8ucI
Luwang 80nglat
. .. . ... . .. .. . .. .. . . . . ... ... . . . .... .. . . . . . . .... , .. .. . . . . . . .. . . .. ... .
Luwana ~rubua
GRUBUG CAVE SYSTEMS
to aaron
THE suel
. UI
.0
...
-=- '0:
200
metrM above 8M .....
Kall 8uel
0'1
0'1
-40
+20
It
Entrano.
vr
0' ~
' .........., " ,,

\
PLAN
..
Figure 8
"AI' 222222228;;$
.......... "., ,,."'M

__
I
/;'
100 ... tr
.ump
".,
U""... ".,
~
SEMI-EXTENDED SECTION
c.,~
~~
10
,
GUNUNG .S EWU CAVE SURVEY, JAVA, 1982

SITE 191
G.H.611118
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
GUA NGINGRONG - GUA MULO
- 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
@ GUNUNG SEWU CAVE SURVEY
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
"
"
'4"' ~ ' #'
.....
\,\
~"\
~,,""
"
\,
Entrance
'...
....
" ,
"
. ..
~ ~
... . -
,/
~,!" ,
.l
...
Figure 12
GUNUNG SEWU CAVE SURVEY JAVA 1982
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
DETAil AND CAVE 8TREAMWAY
pe
Pool
oL
low
EXTENDED SECTION
8URFACE
"."" ,..::::.fj"
'
8ump
TJJ!?,,'
JAVA 1982
~:"1~
~ ............
'.:: -"~ ..~.:: '"
.--::
,"
GR 742998
. D
c ::~prm.dl.t
SITE 179
LUW ANG DAREN
pH
P7
...
,.,
"
.....
\
I
I
"
"
'4"' ~ ' #'
.....
\,\
~"\
~,,""
"
\,
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
DETAil AND CAVE 8TREAMWAY
pe
Pool
oL
low
EXTENDED SECTION
8URFACE
"."" ,..::::.fj"
'
8ump
TJJ!?,,'
JAVA 1982
~:"1~
~ ............
'.:: -"~ ..~.:: '"
.--::
,"
GR 742998
. D
c ::~prm.dl.t
SITE 179
LUW ANG DAREN
-...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
LUW ANG .BUHPUTIH
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
LUW ANG GOPLAK

SITE 3
LUWA'NG KENTENG
GR 538068
SITE 9
GR 621004
SITE 10
ao ",et,.,
@ GUNUNG SEWU CAVE SURVEY JAVA 1982
LUWANG JOMBLANG
GR 613032
EXTENDED SECTIONS
Figure 17
LUW ANG NGIRA TAN
- - - - - - - - - - - 1
I

SITE 171
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
Ooure and pool.
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
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
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
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)
A single cylindrical shaft drops via a 93 m pitch to an impassable floor of
mud and boulders.
LUWANG JOMBLANG (1 75)
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
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
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
\
\
------
Dry on w.t.r l.v.1
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
@ GUNUNG SEWU CAVE SURVEY JAVA 1982
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
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
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
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
__ , GUNUNG SEWU CAVE SURVEY
LUWANG NGEPOH
GUNUNG SEWU, JAVA
7. Main stream passage in Gua Sodong at Mudal
8. Waterfall inlet in Luwang Ceblok
(Waltham)
(Eavis)
9. The 63 metre shaft in Gua Lebak Bareng

(Eavi s)
10. Upper passage in Gua Ngi n grong (Waltham)

79
<----(
~
Enttanc.
.,
-----
'------ --~-------- - - -----~-
P13
GUA LEBAK BARENG

)
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
flowstones. A single muddy inlet passage is not completely explored but

probably connects to Luwang Sindon. Below the inlet junction a final fast
sluice takes the river into its terminal lake over 300 m long, much of which
is deep enough to require swimming. Beyond the sump the water is next seen
in the flooded rift of the Gua Ngreneng collapse, where the main conduit is
frustratingly inaccessible; and from there it has been dye-traced to the Baron
resurgence . The Bribin is a fine piece of river cave well worth a visit;
life
jackets are essential for its exploration together with some float rope for the
downstream section .
GUA GILAP
A large collapse doline has a massive, arched entrance in one wall . A thin
path snakes down a long boulder slope well into the darkness zone, right down
to the cave stream. The path is heavily used by villagers who depend on the
cave stream for their sole dry seaSon water supply . Two failed pump schemes
remain in the cave, though better engineering should one day lift the water at
least to the daylight area and Save the villagers scrambling into the darkness.
Downstream the cave sumps in the boulders of the collapse, though a higher level
route leads to an undescended pitch which may provide the way on.
Upstream the passage is' open and shortly leads to a large chamber with a floor
of mud and collapse debris but a roof still intact. Beyond this, the streamway
continues as a splendid keyhole gallery with a meandering canyon cut in the floor
of a tube over 5 m ,in diameter. The water flows ' through an endless succession
of gour pools floored with crunchy, crystalline calcite, and most of the keyhole
ledges are decorated with white stalagmites . The upstream sump is created where
the roof plunges into a pool ponded behind a gour-covered zone of collapse debris .
Upstream of its chamber, the Gilap cave is one of the best decorated in Sewu and
provides easy and most enjoyable caving.
GUA SODONG (Dadapayu)
Length 2075 m Depth 90 m Grade 5 survey (Fig . 28) .
The almost level entrance passage is normally dryas far as a pool 200 m from
daylight. This is heavily used by villagers for both washing and abstraction;
their journeys in and out using paraffin torches have left black soot over a ll
the wallS, and too often a thick stinking fog hangs in the cave. It is a
uniquely unpleasant cave environment, which should soon be eliminated by a new
well from the roadside to permit direct abstraction from the cave pool.
The cave beyond the pool was explored through a 3 . 5 m sump, which can be
free - dived with care, though it disappear s late in the dry season as the pool
level drops . Beyond the sump, the pool continues with long and very low ducks
when the level ' is high. The pool ends at a gravel bank after which the passage
continues 5 m high and wide. It keeps this size to the end of the cave, but for
the great majority of its length it has a narrow, meandering, vadose canyon cut
up to 5 m deep in its floor. The canyon is mostly a metre wide, so not difficult,
and its interminable twists and turns are relieved by a few deep pools, some
cascades and a couple of handline pitches. The canyon is only developed in some
sections as it was cut by waterfall retreat in the original stepped profile of
the cave; where the canyon is missing, the cave is spacious and quite well
decorated . The terminal sump comes as rather a disappointment;
the promise
of further lengths of gently graded passage is provided by the dye trace to the
Bribin cave sump also at high level.
.
GUA SODONG (Mudal)
Length 4290 m Depth 46 m Grade 5 survey ( Fig . 29).
A fine dendritic stream cave system with three major branches, Sodong is the
longest known cave in Sewu. Its entrcance is normally crowded with villagers who
use the water from an inlet just inside. The main flow leaves the passage and
is thought to go via . L~ang Sapen to the Northern Tributary . Heavily polluted
overflow fills pools in the passage beyond, and these constitute a very se+ious
health hazard. Once the pools are traversed the cave becomes progressively more
pleasant , as a high tunnel provides easy walking broken by two short handline
pitches and long canals, with one area of decorated chambers.
The Main Stream Passage is mostly a high canyon, with a sizeable stream along
its whole length of nearly twokilometres . The water emerges from the top sump,
flows through a succession of pools, small cascades and waist-deep canals, and
eventually feeds into the terminal sump from where it has been dye traced to the
Pracimantoro spring two kilometres away and on almost the same level.
81
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
LUW ANG TONG POCOT
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
approx. 43Om. a.sl
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---~--~
' " II.cl,Io PUlllllP I"talke
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
GUA SODONG (Mudal)

JAVA
\\
lB82
SITE 133
G.R . 762112
'00
.\
200 METRES
"
PLAN
'. . (i):ff0.
"...
-10
'!'!! ,
.'
CROSS SECTIONS
ENLARGED X2
~, :::::;
- -otO
.w
II'ROJECTED SECTION OF LUWANa S,!PEN

\
\
\
\
\
\
\@
LUWANG SAPEN
~~O_
-8
--- ~J:;.,
~
\
\
\
\
NORTHERN TRtBUT ARY
APPROXIMATE RELATION OF CAVES TO SURFACE
. ..---.......,
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:
Nor th ern Gunun g Sewu

88
Telagas
Contours (metres)
I nd
Alluviated Iowa
Karst upland
The Northern Tributrary is a long inlet, again a meandering canyon for

almost half its length, and containing deep water. In its upper reaches its
morphology becomes more complex with branches, high-level chambers and oxbows,
some of which contain spectacular displays of white c a lcite deposits including
stalactites, stalagmites, gours and cave pearls . These decorations complement
the long streamways to make Sodong a very fine system different in character
from most of the Sewu caves.
CAVE GEOMORPHOLOGY
In general the cave passages show little relationship to geological
structure, mainly due to the rather featureless nature of the Gunung Sewu
limestones. Some shafts and isolated segments of rift passage are developed
on obvious joints or faults. Bedding planes control a handful of passages,
notably in Sumurup, Tong Pocot and Grubug, but the interbedded volcanic ash
units appear to have a regional influence (for .example, in the perching of the
Bribin streamway) greater than their barely visible control on individual
passages.
The only major topographic influence on the caves is provided by the supply
of allogenic water from the Worrosari Plate au. This accounts for the majority
of the larger cave passages occurring along the northern margin of Sewu. Even
with the existence of the . inter-cone valley systems, catchment a re as for the
shaft caves in the heart of the karst are limited and most passage sections are
quite small.
In contrast, the caves show a close relationship to the water table. They
have an overall pattern of classic simplicity which relates closely to the
theoretical flow of groundwate r. Nearly all the vertical or steeply inclined
caves are essentially vadose, while nearly all the horizontal conduits have
phreatic origins. This situation existed in the past, giving the few high
level, since abandonned, horizontal caves, and continues to the present where
most drainage descends rapidly to a low-level water table even with such great
distances to the resurgences. The dominance of this pattern perhaps reflects
the poor geological influence in most of the aquifer.
The vadose shaft caves are of three types. The single shafts only occur as
such because their deeper continuations are blocked by sediment or collapse;
they may be either relatively cleanly sculpted cylindrical shafts such as
Jomlang 10, or may be collapse-modified bell shafts such as Goplak. Of the
multiple drop systems there are two types. Some were initiated on a stepped
profile with a degree of geological influence, and have developed into a
succession of separate shafts links by short canyons or rifts,for example at
Tong Pocot and Jalak Bromo. Others started as a more irregular steeply
descending route, from which deep canyons have been cut commonly developing
into a stepped floor profile, as at Setro and Puniran. With the inclusion of
the many shorter explored caves, the latter type probably dominates.
In various caves,stalagmite deposits and passage rerouting show evidence
of past phases of development, but no widespread patterns have yet been
recognised and correlation with surface changes would still be premature.
Some shafts do intersect isolated segments of very old phreatic cave.
Additionally, those shaft systems that le ad into level conduits all show
phreatic origins in their level sections;
those in Lebak Bareng and Tong Po~ot
are good examples. In most of these cases it appears that the nearly level
conduits originated just below the contemporaneous water tables, and the
subsequent fall in water tabl e levels has been relatively slight. Karang is,
however, distinctive in that it has a marked phreatic level 45 m above the
present sump level;
its location, close to the water table trough from the
northern sinks to the Baron resurgence, may indicate a relationship between
the greater water table lowering and the development of the major conduits.
Buhputih is a singular example of a cave with a more complex morphology. It
has both a high -l evel vadose canyon and also a lower conduit section, but the
intervening series of shafts is broken by another more gently graded section.
This is probably controlled by the geology but may also relate to a past phase
of phreatic development.
The caves with greater lateral extent all show evidence of phreatic
initiation. This may appear a s either fine keyhole passage profiles (as in
both the Sodongs and Gilap), abandonned high level phreatic galleries ( as in
Grubug) or relatively unmodified phreatic tunnels (as in Bribin and Sumurup).
Individual systems are generally not complex enough to identify long
histories of development easily , but the sediment sequences , including some
spectacular bone beds, in some caves appear to have a value for future research.
89
The relatively little incision from originally shallow phreatic caves in

those systems in the heart of the cone karst, notably Bribin and Sodong
(Dadapayu), suggests limited and slow water table lowering in these areas
(though the picture may be locally distorted by geologically influenced
perching). In contrast the caves of the northern marginal zone show much
deeper incision and abandonment, notably at Grubug. This appears to relate
to the m6re extensive development of caves in this zone fed by the Wonosari
Plateau drainage, followed by more rapid and greater water table lowering in
response to the increasing efficiency of the main conduits feeding directly
to the Baron resurgence.
CAVE BIOLOGY
Unfortunately the Gunung Sewu caving team did not include any biologists,
but the members did observe some features of the cave life which contrasted
with their experience in tropical caves elsewhere. Various species of small
bats occurred in most caves, and it was surprising that they penetrated to
the very ends of even the longest caves. The bats had their uSual generous
collections of parasites, .and on a number of occasions were seen to have a
marked enthusiasm for swimming (or were very incompetent at taking a drink).
Large colonies of bats were not found, and only the Serpeng and Grubug caves
were seen to contain more than a few hundred individuals. In the entrance to
Gua Serpeng, villagers use catapults to shoot the bats down from a 20 m high
roof, and then take them to the local market to sell as food. There are
relatively few cave swiftlets, although they are reported in large numbers
from caves further southeast in Sewu.
Snakes are a normal hazard of tropical cave entrances, and a few exciting
and nerve-racking encounters were experienced. Further into the caves very
few snakes were seen. The nearly static pools and lakes in the seasonally
active river caves of the northern marginal zone were distinguished by
populations of huge black eels over a metre long and as thick as a man's arm.
They had spectacular sets of teeth and a rather aggressive character, thereby
rather enlivening the swims necessary to get to and from the bottoms of ropes
on the pitches.
Of the smaller animals few spiders of any size were seen, but there are large
numbers of the evil, black, tail-less whip scorpions which range up to the size
of a man's hand. In the streams, white shrimps are common, and some caves
contained large white crayfish and semi-transparept catfish. Small white crabs
were common, as were normal darkly coloured frogs;
some frogs in the entrance
shafts exhibited amazing climbing abilities when they could progressively jump
up and cling on to a vertical wall.
POTENTIAL FOR FUTURE EXPLORATION
Many incompletely explored caves are described above or appear in the
appendix register. The Luwang Register also notes .about 50 other known but
unvisited sinkholes; Gunung Sewu provided far more than could possibly be
examined within the scale of the 1982 project. Locating any of these entrances
is easy because most villagers know by name the luwangs nearest to their homes.
The best exploration potential probably lies in the zone of river sinks along
the northern margin, where many inviting holes were left undescended because the
main objective of 1982 lay in the shafts of the central area. There are also
many other entrances which can easily be found by walking and aSking questions
in many parts of the cone karst.
The 1982 work never extended east of Pracimantoro or north of Gua Gilap.
The edges of the cone karst on its northern extension contain a number of sinks
and risings which appear to be linked by open cave (Fig. 30). East of Pracimantoro
there is a large area of karst mostly awaiting its first visit by cavers, although
Specavina members did descend the 120 m deep bell shaft of Luwang Ombo within
this region;
and Ombo was later explored by visiting French cavers to a depth
of 230 m through 2900 m of large passages. Maps of the eastern end of the karst
mark a number of sinking and rising rivers which could well repay investigation.
All cave exploration in Java is coordinated by the national body, Specavina
(currently c/o Dr. R.Ko, P.O. Box 55, Bogot,Indonesia) and foreign groups shOUld
work with Specavina as closely as possible. Cooperation should not only help
Specavina to develop, but foreigners will benefit immensely if they can be
joined by Javanese speaking cavers. Any future expedition to Sewu would also
be well advised to consult the writers' full 1982 report which is kept in the
90
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
A.C.Waltham, Civil Engineering Dept., Trent Polytechnic, Nottingham.

P.L. Smart, Geography Dept, University, Bristol.
H. Friederich, P.O. Box 39, Tutume, Botswana.
A.J.Eavis, Tidesreach, Redcliffe Road; Hessle, N. Humberside.
T.C.Atkinson, Dept. Environmental Sciences, University of East Anglia,
Norwich.
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
L Bodeh 740102 Descending passage, unvisited.

L Ceblok 615128 See text
L Goplak 538068 See text _
G Karangwetan 515099 50m passage to narrow canyon, continues
L Bendo 663113 Small entrance, unvisited
L Karang 658114 Descending passage, unvisited
L Karanglampar 659114 P40, unvisited
L Jurangjero 668115 Shaft, unvisited
L Kenteng 621004 See text
L Jomblang 613032 See text
L Song Gupit 609137 P56 to choke
L Sumbring 612148 Climb and p60 to choke
L Balong U 612030 Entrance too small
L Balong B 612030 Entrance too small
L Sumberejo 612028 Entrance too small
L Blekonang 607024 See text
L Gesik 602030 Entrance too small, draughts
L Bentar 598078 D60, Climb and pIO to large canyon and choke
L Ngrapah 588078 Choked chamber
L Glesung 587112 D70, p25, C8 and P21 to choke
L Jomblang 579111 Unvisited
L Maca~m~ti 574111 Unvisited
L Nglibeng 565067 Walled up sinkhole with draught, needs digging
L Kerwo 574063 D45, PIO and p20 to choke
L Jomblang 575057 Small shaft, choked
G Jurug 565074 Large passage to mud choke
L Karang 573051 No open cave
L Tabuhan 573082 L200, C4 and mud passage to p4 into sump
L Soroiten 543105 Narrow canyon to P20 undescended
L Krinjing 537108 D60, P55 to mud slope and sump
L Gowah 534108 D99, Pitches of 5,70,6,4 and 4 to impossible rift
L Tlogodadi 526081 p25 to chamber with choke and inlet passage
L Setro 524058 See text
L Katok 524049 D55, Climb to p5 and p15 to chamber with sump
L Jowa 514048 Canyon with P12 into chamber with sump
L Ngegong 509028 Blocked shaft
G Lebak Bareng 526085 See text
L Bamban 528087 Large shaf.t probably connects with 37
L Tlogolaran 509094 Blocked shaft
L Mbibres 512073 P8 to choked chamber
L Gelap 613020 D72, P60 to canyon to sump
L Ngoro-oro Ciut 507073 p40 on loose wall to passage and P30 undescended
L Bawongan 498073 L200,p20 and canyon to P20 undescended
G Tritis 512090 Small tourist cave, unvisited
L Gunung Beteng 607105 p40 continues vertically, floor unseen
L Puring 597107 P30 and slope to mud choke
L Bawahan 598097 Small shaft, choked
L Buhputih 644082 See text
L Sindon 640112 P34 and P20 i lto canyon and P8 undescended, may connect
to Bribin
L Ledok 634103 P12 and P5 into canyon, too narrow to continue
L Tong Pocot 631 101 See cexc
L Ledok 738101 Narrow shaft unvisited
L Cahkeri 735118 P30 unvisited
L Jumbleng 726035 P7 to choked chamber
L Gading 754036 Dl12, P68 to canyon and P25 to sump
L Sirik 746043 D48, p33 and rift to sump
93
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
L Sawah 725078 p5 to choked chamber

L Sruput 717079 065, P40 to boulder slope and choke
G Blunyah 713086 Passage to P18 and large canyon to mud choke
G Kalen 704090 L150, Low passage to p29 and P7 into canals to choke
L Ombo 727116 P8 unvisited
L Sawah 718112 Unvisited
G Semuluh 644139 See text
L Towati 690070 Blocked sink
L Sumur 693037 P40 to 100m of streamway to undescended narrow pitch
L Jurangjero 693037 Unvisited
L Ngandong 687045 Shaft, unvisited
L Ngimbes 686045 Unvisited
L Kluwangan 692012 p8 to choke
L Glaragahumbo 693001 Slope to p30 to mud choke
L Sirih 680006 Two blind climbable rifts
L Pucang 688994 P70 to ledge at top of undescended P30
L Kenteng 680982 Small blind hole
L Gondang 682995 C7 to choked chamber
L Wates 417118 Descending passage, unvisited
L Soka 416114 Descending passage, unvisited
L Ngrau 411118 Shaft, unvisited
L Ngledok 390127 Descending passage, unvisited
L Ceme 387129 Cave, unvisited
L Bandung 397128 Unvisited
L Ngurik 386123 PlO unvisited
L Legundi 376116 P15 into doline collapse and choked chamber
L Saga 376108 Descending cave, unvisited
L Bledok 37llio Double shaft, unvisited
L Ngrejek 500098 051, P28 and P12 into chamber with sumps
G Ngowe-owe 496104 120m passage to head of undescended P80
G Ngegab 492091 Canyon to P8 and choke
L Gondang 499089 P20 unvisited
L Trecep 486076 PlO unvisited
G Kenongo 491079 L160, 085, Passage then P20, P20 into choked chamber
L Bobak 487066 Unvisited
L Klepu 490053 Large passage to mud choke
L Jomblang 476057 P23 and climb to mud choke
L Santen 464060 Blocked sink
L Klumpit 440050 P15 to choke in rift
G Klumpit 439049 P20 to choke
L Banteng 432058 075, Pitches of 30,17,7,6,4 and 9m to sump
L Karang 501047 See text
L Tlempek 505041 Shaft, unvisited
L Kuang 435074 Shaft too narrow to enter
G Jurug 583137 Descending passage, unvisited
G Glendu 591139 Lloo, Canyon descends gently to sump
L Kebo 580130 Low passage, unvisited
L Bedesan 574124 See text
L Gedilan 576126 Shaft, unvisited
L Jero 612088 See text
L Pendul 602078 045, Canyon with climbs becomes too narrow
L Wuluh 623063 p8 and p25 into choked rift
L Ngepoh 630053 See text
L Toar 623045 C15 and canyon to undescended P20
L Besole 628033 P70 to boulder floor choke
L Puleireng 596034 05~, Shaft behind dam on telaga floor, with pitches of
30, 8 and 8m to undescended P12
L Mundu 628029 062, Pitches of 10, 38 and 8m to canyon and sump
L Ledok 624038 Pair of sinkholes, one has PlO into complex of rifts
not descended, other has P15 to choke
L Gebang 636045 PlO to p25 undescended with much loose rock
G Pengangson 662158 Two sinkholes to sumps
L Ngrinjing 693126 Unvisited
L Grigak 700108 Unvisited
L Jalak Bromo 684066 See text
L Nglampeng 595085 Unvisited
L Purung 628087 Unvisited
L Jambu 623088 Unvisited
L Nujo 730105 Unvisited
L Kalen Unvisited
L Jarak 735065 P30 to mud choke
L Gedawung 588125 Shaft, unvisited
94
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
L Cabe 586 125 Shaft, unvisited

L Blerong 369 142 L120, Passage to Pl0, chamber a nd duck, continues
unnamed 456040 L165,051, Series of climbs in canyon to sump
L Tjabe 591062 Blocked sink
L Bandung Sumuran 492120 P40 and climbs to choke, unvisited
G Sodong (Mudal)
762112 See text
G Songgilap 755101 Large entrance and climb to choke
L Watugajah 727048 Small entrance, unvisited
L Suh Kidul 728032 P8 to choked chamber
L Gendek 728031 See text
L Jurug Watu 575036 Shaft too narrow to enter
L Ngirowari 578036 P40 to terminal rift
L Gupakwarak 528018 Cl0 to choke
L Blubug 573070 Small cave with C5 to choke
L Tirisan 596074 D35, Climbs to choke
L Mojing 603063 Climbs and Pl0 to mud choke
L Branjang 1 588032 C25 to mud choke
L Branjang 2 588032 040,p20 and canyon becomes too narrow.
L Bete 778091 028, P16 into two choked chambers
L Gadjah 778089 Small choked shaft
S So dong 359129 Stream cave in doline, sumped both ways.
S Sungei Besar 358135 Large entrance blocked by'choke
S Ngerenean 462023 Large resurgence on beach, no cave
L Puniran 628110 See text
unnamed 533113 Air photo feature, unvisited
unnamed 527115 , Cave passage, unvisited
unnamed 539122 Air photo, unvisited
unname d 605186 Air photo feature, unvisited
L Bendo 657138 P5 and climb to low bedding plane
L Lebuh 677130 Ll00 Large entrance to canyon, narrow but continues
L Sapen 763108 L250, P36 into small streamway, sumps both ways
G Kajubang 422076 Passage to blocked chamber
L Pringwulong 432084 p8 and P35 to terminal rift
L Ngampal 458056 P25 to choked rift
unnamed 438103 Unvisited
G Ngreneng 635130 La r ge doline with choked passes and flooded rift
G Toto 619139 L925 Larqe entrance to streamway, upstream through boulder
chambers to choke, downstream to sump
G Kedokan 637063 Desc e nding passage to sump, used for water supply
G Jomblang 733158 L354, Complex of small streamways, used for water
supply, downstream continues small but promising
549126 See text
G Ngingrong (Mulo)
L Ngiratan 638058 See t ext
L Sumelap 639065 D43, p32 into streamway, continues
G Pozo 640075 Cave, unvisited
G Tritis 537072 Climb to undescended P15
L Jomblang 625011 See text
G Buri Omah 597142 See text under Gua Suci
L Ngel0 625037 Narrow canyon in sharp rock, continues
L Bohol 716018 L175, 057, p35 and climbs to passage, continues low
L Oaren 742998 See text
G Bribin 647117 See te,xt
G Sodong (Oadapayu) 681092 See text
G Gilap 729192 See text
L Serpeng 1 559128 D86, P60 in large shaft to passage and sump
L Serpeng 2 557127 See text
S Slili 564995 Short passage to large sump pool; springs on beach
S Sundak 567994 Short cave to deep sump pool; springs on beach
S Baron 500016 Major resurgence, 100m river cave to sump
S Ngobaran 453026 Climb into large chamber with stream sumped both
ways; springs on beach
G Song Tawing 603122 L355, D69, Large entrance with high levels, and
smaller canyon through pools to undescended P7
L Grubug 600127 See text
L Kenteng 731067 Sinkhole with loose boulder floor choke.
L Gunung Bolong 595316 P70, unvisited
L Ngluweng 708042 p25, unvisited
95
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
L Pengangson 377075 C8, p20 and Pl0 to sump

L Glagah 377087 L150, D48, P18 to descending rift one way and P24
other way to streamway to sump
L Macanmati 395107 P13 to streamway to undescended p8
L Seropan 577118 See text
L Besole 378081 Rift with pitches of 5,10,12 and 12m, continues narrow
S Pondjong 684178 Major resurgence, no cave
L Jati
715133 P55 to ledge and undescended P45
S Pracimantoro 787097 Major resurgence, no cave
S Piyuyon 773946 Major resurgence on beach, cave 5m to sump
G Bendungan 697095 Low passage with pools to undescended P15
L Wediwutoh 637137 035, Canyon to mud choke
L Jurang 634132 020, St~ep passage to sump
G Suci 601147 See text
unnameq 587142 Blocked sink
unnamed 574140 Choked sink
unnamed 576117 P8 to undescended pitch
unnamed 587119 Undescended P8
G Song Oangal 594119 Ll00, Large entrance to canyon to mud choke
unnamed 595111 P22 to mud choke
G Serpeng 562127 Large river sink with sump just out of daylight
L Banteng 522122 River distributary sink choked with mud
G Sumurup 519120 See text
unnamed 507131 Cave passage, unvisited
G Grengseng 536060 Short fossil stalagmite cave with small chambers
L ' Cikal
418134 Cave, unvisited
L Gondosore 412129 Cave, unvisited
unnamed 358152 Cave, unvisited
L Pongkok 585136 Unvisited
L Songjembul 707124 p40 and rift to undescended P12
G Tritis 567057 Blocked large cave entrance
G Brangking 536017 Short cave to sump pool
G Tepus 761113 Short tourist cave
S Wonoseri 656970 Rising, no cave
S Mutisari 670977 Rising, no cave
S Beton 698213 Rising, no cave
S Beton 796118 Rising, no cave
S
S
S
S
S
S
S
S
S
Teleng 699209 Unvisited

Puring 682996 Rising too narrow to enter
Sambiroto 797088 Unvisited
Gedaren 697190 Rising, no cave
Su1u 711189 Unvisited
Nggremeng 691220 Unvisited
Karangmojo 473173 Unvisited
Umbu1denkok 550149 Unvisited
Penuh 502154 Unvisited
96
CAVE SCIENCE
Tran . Br i tish Cave Research Association. Vol. 10, no . 2, pp.97- 102. June 1983 .
MEAN ANNUAL RUNOFF AND THE SCALLOP FLOW REGIME

IN A SUBARCTIC ENVIRONMENT
Preliminary results from Svartisen, North Norway
Stein - Erik Lauritzen, Andrew Ive and Barry Wilkinson
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
u = mean flow velocity

y
kinematic viscos ity of liquid

Sauter - mean of scallop l engths
Re* = Scallop Reynolds number, 2,200

~
= Friction factor , 9.4.
This formula has been checked in a f loodwater passage, where there is no

baseflow regime (Pisarowicz and Maslyn, 1 981) . Discharge calculations based on
the formula f itted the me asured flood discharge within 8 .3%, which is well
within the probab l e erro r for a natural system.
Recently, palaeocurrent flow r ates were calculated from scallops and
passage dimensions in a fossil phreatic network . The se measu r eme nts are in
accordance with the continuity equation for integrated pipe networks
(Lauritzen , 1 982) . Howeve r, it is not Known which partes) of the paleo-flow
regime is represented by the scallops.
Deep phreatic passages will transmit both basef l ow and floodwater discharges .
The driving forc e in limestone dissoluti on is the degree of undersatu r a tion
(aggressiveness) (White, 1977), which will cha n ge with the climatic regime.
Flow rates also affect the rat e of dissolu tion . Therefore, scallop sizes in
different hydraulic and chemical r egimes are difficult to pre dict . Empirical
studies from in situ situat ions are necessary to determine which part of a
flow regime the sca llops represent. Such studies are very important as a
b a ckg round for reconstruction of environmental cond itions in the past (i.e.
97
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
R. and A. Ive S. St Pierre
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
Fig.3 . The Green Rising resurgence.
-13m.
Green Rising
A Ive. S. E. Lauritzen and

8. Wilkinson
1981
8CRA
GLOMDALSVATN
RESURGENCE
A. ~
CROSS - SECTIONS '
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
(em s -1) 30.5
43.6
70.7
Discharge
(m 3 s -l)
+5.23
12.3 -3.70
20.0
8.62
MEAN ANNUAL RUNOFF

Preliminary discharge measurements indicate that approximately the same
amount of water passes through the sinkhole and the resurgence. Therefore, it
is assumed that the total discharge also passes through the measured cave
cross-section. Hence, it would be advantageous to compare these results with
the regional mean runoff for the area.
Isohydate maps of the area for the pe"r iod 1930 - 1960 have recently been
compiled (Hagen, 1977). The specific runoff for the catchment area of the lake
is 90-95 Is-1 Km- 2 giving a mean annual discharge through the cave of 2.5 m3 s-l.
DISCUSSION
The calculated discharge based on scallops overestimates the mean annu a l
runoff by at least a factor of 3.
(The smallest scallops found ( 4 cm) sug gest
a cave conduit discharge about 15 times as large as the mean annual runoff!).
This demonstrates that, at least in this location, flow rates derived fr o m
scallops by no means represent the annual "mean" in a sub-arctic region.
The hydrology in sub-arctic and alpine environments is characterised by a
strong seasonal variation with a maximum discharge during the summer snow and
ice melting period, and it is reasonable to assume that effective scallop
formation may take place during this period. In Swedish Lapland, Hellden (1974)
found that 67% of the total annual denudation took place during 3 months of
the yea~, i.e. the snow-melt period (May-July). This is also the period of
highest discharges.
Since most of the winter precipitation falls as snow, it is reasonable to
assume that 6 months' accumulation may be concentrated through the 1-2 months
of snow-melt s3ason. This gives a snow-melt discharge through the cave of
about 7.5-15 m s-1 which fits well with the scallop discharge (Table 1).
Extending this approach, the smallest scallops correspond to 8 months'
precipitation (snow), released through two weeks of intense snow-melt. Such
drastic conditions do not seem reasonable, and we suggest, as a preliminar y
hypothesis, that the occurrence of schist flakes in the passage, protruding
from the wall may cause secondary flows and eddies in the large tube cross section.
The formation of scallops is a function of the masS transfer across the
boundary layer interface under turbulent flow (Blumberg and Curl, 1 974) , which
again is determined by flow velocity and chemical aggressiveness of the water
in question. As a consequence, small scallops form faster than larger ones
under the Same chemical regime. But, on average, it is probably the discharges
of highest aggressiveness and duration that will dominate a scallop pattern
("Scallop dominant discharge"). In cases where the discharge has changed, a n
overall increase should give a quicker response with respect to scallop sizes,
than an ovetall decrease would do. This means that small scallops should be
superimposed more easily on larger ones, than vice versa.
CONCLUSIONS
Although these calculations are based on only a few scallops and the system
is not yet fully investigated, some gene ral implications may be stated:
1) Under this particular climatic regime, scallop-forming discharges are more
than three times greater than discharges deduced from regional mean annual
runoff data.
101
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
Stein-Erik Lauritzen, Department of Chemistry, Univ: of

Oslo, P.O. Box 1033, N-Blindern oslo 3 Norway.
Andrew lve, 120 westcombe Ave., Croydon CRO 3DB, U.K.
Barry Wilkinson, 421 Middleton Road, Carshalton, Surrey, U.K.
102
CAVE SCIENCE
Trans. British Cave Research Association. Vol.10, no.2, pp .103-ll5. June 1983.
SPELEOTHEM DATES AND PLEISTOCENE CHRONOLOGY

TN THE PEAK DISTRICT OF DERBYSHIRE
Trevor D. Ford, M. Gascoyne & John S. Beck
ABSTRACT
Uranium-series dating of speleothems from Derbyshire caves has demonstrated that there
have been five distinct periods of speleothem deposition separated by four periods of nondeposition since >350,000 y B.P. Morphological correlation with sequences of cave development
has indicated a time span for the formation, filling, abandonment and degradation of some
major cave levels. Limited correlation only can be suggested with surficial deposits and with
the river terrace sequence. Periods of speleothem deposition in Derbyshire are comparable
with those established in Northwest Yorkshire and the Mendip Hills. Some degree of
correlation with the marine isotopic record is also indicated.
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
'-----'
Hole in the Wall
MIDDlETON
Sarah's Cave
Thirst House Cave
Water Icicle
Close Cave .
o
I
Fig. 1.
Kilometres
,
,
5
I
Sketch map of the northern part of the Carboniferous

Limestone outcrop in Derbyshire showing caves which
have provided dated speleothems.
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~
(Figures in brackets indicate probable outlet elevation)
Swallets choked in
Devensian re-excavated
(186 m)
Local fills
Speleothems in Peak,
Speedwell, Giants Hole
and Winnats Head (lower)
Treak Cliff Cavern

stalagmites on fill.
Incision to 197 m.
Blue John Cavern

incised. Russett well
outlet formed Incision
down to 204m.
Speleothems in Giants
Windpipe.
Derived loess fill
Incision to 218m;
Main cave systems;
Peak Gorge Initiated
Speleothems in
Winnats Head Cave.
Cave Dale outlets

to phreatic cave
syste ms (above 242m).
Earliest Tubes
[above 303 m)
Phreatic development
in mineral veins.
Castleton
Lathkill Head Complex

(193-206 m)
Limited incision
Lathkill Head Complex

initiated (220 m)
Upper Cales Dale

Complex (242m).
Further Speleothem
deposits in Water
Icicle Cave
No evidence
Water Icicle tubes

(303 m) and later
Speleothem deposition.
Lathkill Dale
(after Beck, 1975, 1980; and Ford 1977)
suggested Correlation of Developmental Stages in the Major Catchment Areas with the British Pleistocene sequence
Table 1
Wye Head and Otter

Hole (304 m)
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
sough in the western part of the cave in normal weather.

The Lower Complex is only known in the vicinity of Carlswark's Resurgence
Entrance, but is thought to continue below water. It lies between 168 and 171
metres O.D.
The cave levels of Stoney Middleton Dale cannot be related directly to terrace
remnants in the valley; mining and quarrying have largely obliterated such '
features. Slight breaks of slope do occur in the long profile however. The cave
levels are shown in Fig. 2.
As the original natural resurgences to any of the cave levels are unknown,
the relationship to terraces downstream of Stoney Middleton Dale is also unknown.
The mouth of theDale, however, appears to be at an indistinct knick point at the
head of the Hope Terrace. Since a small area of till has been found overlying
the limestones in the quarries on the south side of the Dale, apparently
representing aggradation on the Hathersage Terrace, it seems likely that the
incision of the Dale covers the period of formation of these two terraces.
The Bradwell Dale catchment area lies to the north-west adjacent to that of
Stoney Middleton. Together they occupy the north - eastern portion of the
limestone outcrop. Bradwell Dale drains northwards, Namurian sandstones forming
a prominent escarpment to the east overlooking a shale valley. The limestones
rise gently from beneath the shales to form a dip-slop to the west, and it is
principally this dip-slope which is drained by the risings at Bradwell at the
north end of the dale where it joins the Hope Valley. Small swallets along the
base of the escarpment provide the only allogenic inputs, and little cave
passage is known as a result, except in the vicinity of the resurgence.
The caves again consist of strike - oriented tubes, modified to varying degrees
by vadose activity, and showing more influence by mineral veins than at Stoney
Middleton. The largest known cave is Bagshaw Cavern, a tourist cave which runs
sub - parallel with Bradwell Dale on the west side. The cave was found by mining
operations in 1812 .
.
Two cave levels are displayed in Bagshaw Cavern, the lower level being active,
but the upper level carrying overflow in flood. They are connected by two steeply descending tubes. The upper level has up-dip branches which are inactive,
and which appear closely related to caves intersected by Outlands Head Quarry
to the west. In the absence of a detailed survey, it is difficult to relate
Bagshaw Cavern to the short caves of Bradwell Dale, but it is tempting to equate
the upper level with Bradwell Parish Cave, at 204m O.D., and the lower level
with phreatic tubes found in mines on the west side of the dale at an elevation
roughly 15 metres lower.
Caves at higher levels are very fragmentary, but there appears to be a
concentration of tubes at 213 m O.D. in Hazlebadge Cave and on the east side of
the dale near Bradwell village.
In Hartle Dale, an up-dip tributary of Bradwell Dale, there is a short
phreatic cave lying at 276 m O. D . Although seen in isolation, its association
with a prominent bench on the valley side, which ends at a knick point, suggests
an early drainage route at this level.
The cave levels of Bradwell Dale are shown in Fig. 3.
Like Stoney Middleton , both Bradwell Dale and the cave systems are graded to
the Hope Terrace and traces of till on the limestone dip-slope to the west again
suggest that the incision of the Dale and the partial conversion of the caves
from phreatic and vadose was contemporary with the period of the Hathersage and
Hope Terraces.
.
In the Stoney Middleton and Bradwell areas, the surface catchment roughly
corresponds with that underground. This is not so at Castleton to the west of
Bradwell. Here, the swallet caves lie on the west side of a surface divide,
while the risings are on the east side of the watershed at Castleton. The
catchment underwent considerable modification during removal of the shale cover.
At Castleton there are no extensive phreatic networks which can be used to
define a series of cave levels . The major phreatic passages of Peak and Speedwell
Caverns appear to have operated from a very early stage, carrying water to a
vauclusian rising at Castleton, and undergoing no modification until the outlet
level was sufficiently lowered for vadose incision to commence along preferr~d
routes. The evidence for stages in the lowering of the outlet is found in the
extent of canyon incision in the Peak and Speedwell stream passages, and in the
abandoned inlet passages (Fig. 4).
One such inlet system is the Pickering's Series of Pe a k Cavern, which ends
upstream close to Middle Bank Pot, an abandoned swallet. The passage displays
vadose features down to an elevation of 22 m O.D., below which point it is an
unmodified phreatic tube. It is likely that this lower limit was controlled by
106
CAVE LEVELS OF STONEY MIDDLETON DALE .

EYAM DALE . AND THE DElF.
20 8759
..,
1_
/7/
AO.D.
Val loy lIoor
700
Phreatic cove passage
600
Vadose cow' passage
SOD
....!....a . . . . . . .
:"
... .
- _
eoofl
First R""lnont Compto.
700
_-"'""::::::;0_'---
Second R.mnont Compl(1J(
600
Carlswor1< Comple.
lower Complltx
Fig. 2.
Cave levels of Stoney Middleton Dale, Eyam Dale and The Delf.
THE CAVE LEVELS OF

BRADWELL DALE .
eleyotion
A.O D.
metres. feel
1000
300
Hartle Dale
Caves .
bench
~_ .L
knickpolnt
250
Bagshaw
Covtrn
flOor 01
rok.
Hallebadg.
/-forll.
floor of Sionlo,", Dol.
-1 '
N.'herwot.r
200
Swallp(
Ear"1
Rake
150
500
Fig. 3.
Bradwel1
Parish
Cove
001.
Cave levels of Bradwell Dale.
107
'"
...Dl
Hop.
Terrace .
Brodwel1
n,
Lower Series
Brodw.l1
01 Bagshow
Cavern .
Resurgence
RELATION SHIP OF CASTLETON CAVE LEVEL S

TO RIVER TERRACES IN I~O P E VALLEY.
m("tn.' s
, 1
AO 0 ,
3SO
A.O.D
1100
-- -
Hlghvst unmodifiPd p,r-.ollc tubvs .
1000
]00
900
800
Deppesl canyon inc~sion In wes!prn swall.ts .
700
Oe(>pest canyon Incision by longc lilr swollqts .
250
Lowest exh"nSIY breokdowr.
200 .
Lowes t conyons of Peak Speedwttll
Hope
600
For Sump. Ppok
500
Cav.rn .
ISO
Peakshotv
Water
Fig. 4.
Hope T.rron"
Holhersogp T(lfToce
Not traceable on the

Peokshol. Wa ter
Relationship of Castleton Cave Levels to River Terraces

in Hope Valley.
lHE CAV E LEVELS OF LA1HKILL

DA LE AN D CALE~, DALE .
: ;'
l.Jrw'Iam(>d
Waler ICiclE'
'::~':::' "m rwm
~r ";~:~
i:.~'
30a~ : 77-4iI?'77:7~------------'------------------------1 )~~:

1 Unnomod
Lolhk ilt
800
~ / __ . _ ____ . __ ._ .. __ _____ .... . .. ]~o.___
200 : :
-:;7;r--;7;r---;7"-/7~--:7:;;r--1/,.;\~7--;7;~=---::~/
1~>>'."I,7/---:;;r----~/--/-:;r--/-;r-c o~250
R. surgenc.
~::~IOW
spnngs
K'S,;" 10 volley floor profiles
/'
;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.
Cave levels in Lathkill Dale and Cales Dale.
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,
clay fills - (low energy run-off, temporary ponding by blocked resurgences):

glacial or cold periglacial (clay probably derived from
loess) .
stalagmitic deposits - (wet temperate climate): interglacial or interstadial.
downcutting through fill - (low energy run-off): cool wet temperate or waning
periglacial conditions.
Scanning electron microscope studies of cave sediments from North Derbyshire
currently in progress have confirmed theloessic or fluvioglacial origin of
many of the sands and silts (R.P. Shaw, personal communication).
Vadose incision into rock and downcutting into fill both result from erosive
run-off but they can be distinguished by the degree of intensity in most cases.
The 10 metre-deep entrenchment of the Peak Cavern streamway has been a result of
either longer or more intense run-off than the limited 1 metre deep trenches in
some Stoney Middleton caves;
this is not to say, however, that a fill stage
did not precede incision in many caves, with total removal of fill at the onset
of high run-off.
Details of sedimentary fills are still under study so that it is not
practical to describe them herein.
SPELEOTHEMS
Stalagmitic deposits, collectively known as speleothems, are critically
important in the elucidation of the chronology, since they can only be deposited
after the cave has been drained and under climatic conditions suitable for
percolation of surface waters through the limestone, i.e. in non-glacial
conditions. Speleothem deposition will be minimal or absent under glacial
climates owing to the lack of biogenic C02 derived from soil and vegetation.
Provided the speleothems are free from contamination by included detrital
sediment, they can be dated by the 230Th/ 23 4u method, (Gascoyne et al., 1978).
Some thirty speleothems from 13 ,caves mostly in the River Derwent catchment
have been dated using this method and the results are summarised in Table 2.
Uranium contents of these samples were generally less than 2 ppm but a few
samples were found to contain as much as 238 ppm. A few speleothems were
mildly contaminated wit~ 23~h (and, presumably, 230Th) derived from included
sediment (indicated by 30Th/232Th ratios less than 20) and the calculated ages
are therefore likely to be greater than their true ages. No attempt has been
made to correct for this contamination because of the likelihood of variations
in the initial 230Th/232Th ratio of the sediment and variations in the amount
of Th leached during laboratory preparation of the speleothem. Many of the
speleothems analysed were too small to permit adequate replication of the ages.
However, reasonable agreement between results waS found in the few cases where
replicates have been made and good age precision and concordance have been
demonstrated for analyses of other speleothems (Gascoyne, 1980).
llc
Table 2
U
Age
/o
Cave
Samp le No
Description
Analysis No
iJ2.runl
Treak Cliff
Cavern
78000
fs on co11 apsed block
-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)
* Low U or Th yie l d ( 10%)
sg
s ta 1aogmite
t Heavily contamina ted by detrital Th
sc
s ta 1act i te
225 + 64
- 41
1tl
ISOTOPIC DATING & CORRELATION

The results of isotope dating given in Table 2 support the sequence of
events deduced on morphological grounds as shown in Table 1, providin~ a
partial chronology . of those events. The dates also provide further evidence
towards the establishment of the absolute chronology of the later Pleistocene
in Britain. Comparing the data with those for north - west England (Gascoyne,
1981) and Mendip Hills, (Atkinson et al., 1978), it is clear that the Same
five groups of ages of speleothem deposition can be recognised (Fig. 6):
I) 0 to 17 , 000 y B.P.; II) 45 to 75,000 y B.P.; III) 90 to
145,000 y B.P.;
IV) 170 to 225,000 Y B.P.; and V) limited deposition more than 350,000 y B.P.,
beyond the limit of applicability of the 230Th/234u dating method. These
groupings indicate periods when the climate in these areas was neither glacial
nor periglacial. Although they closely correspond to the groupings established
for north-west England and Mendips, the presence of a period of deposition
from 60 to 65,000 y B.P., seen in the Mendip study, was not observed here.
Instead, a broader period of deposition frolU 45 to 60;000 Y B.P. was obtained .
Within the l limi ted of analytical uncertainty of the ages, these groups
correspond reasonably well with the chronologies indicated by analysis of
deep sea cores (Shackleton & Opdyke, 1973; Kominz et al., 1979) and ~oral
reef terraces (Broecker & Van Dank, 1970) for the glacial/interglacial sequence
of the northern hemisphere (isotopic stage divisions for the marine record are
also shown in Fig. 6).
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.
A correlation of these groupings with the British Quaternary stage sequence,

proposed by Mitchell et al., (1973), is given in Table 3. The correlation for
groups ' III, IV and V must remain tentative because none of the deposits in the
type sections for these ,Pleistocene stages has yet been reliably dated. The
ideal method for lithostratigraphic definition of stages by subdivision of a
continuous strata record is unlikely to be attained on currently available
evidence, but the speleothem dates presented herein support the generally acce pte d
faunal and floral criteria for stage definition.
112
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)
SEQUENCE OF EVENTS IN THE PEAK DISTRICT

A sequence of events can now be proposed for the Pleistocene evolution of
the northern part of the limestone outcrop and its contained caves in the Peak
District~
1.
pre-350,000 y B . P . (prior to marine isotope stage 9):

a) Stripping of the Millstone Grit cover from a large part of the
limestone plateau. Development of phreatic tubes at high
levels in the limestone and of solution cavities in fractures
and mineral veins, possibly 'during Pliocene times.
b) Limited incision of valleys sufficient to drain some high level
caves. Speleothem deposition in Foxholes and Water Icicle Close
Cave.
2.
350,000 to 225,000 Y B.P.

(marine isotope stages 9, 8 and part of 7).
Further incision of valleys into the limestone, .approaching the
altitude of Hathersage Terrace. No speleothem deposition. Mainly
glacial (Anglian or pre-Anglian glaciations?).
3.
225,000 to 170,000 Y B .P. : (parts of marine isotope stage 6 and 7)

fluvial formation of Hathersage Terrace; formation of the main
levels of the cave systems, e.g. main Castleton stream caves.
Speleothem deposition in high level caves, such as Winnats Head,
Sarah's and Hole in the Wall, while main caves were largely phreatic
at lower end. Penultimate interglacial (Hoxnian?).
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.
90-75,000 Y B.P. (Devensian I glaciation; parts of marine isotope

stages 5 and 4):
intermittent periglacial conditions only in the
Peak District. solifluction sheet on Hope Terrace. Knick- point
recession barely affects resurgences. Considerable sedimentar y
filling of caves.
7.
75-45,000 Y B.P.
stages 3 and 4):
downcutting.
(Mid-Devensian interstadial; parts of marine isotope

stalagmitic deposition on fill followed by
113
8.
45-17,000 y B.P. (Devensian II glaciation ; parts of marine

isotope stages 2 and 3) periglacial only in the Peak District.
Renewed gravel and sand infilling of caves near limestone margin
and sink-holes.
9.
0-17,000 y B.P. (post-glacial; marine isotope stages 1 and part

of 2) further downcutting through fills, and maturing of lowest
cave levels, stalagmitic deposition resumed to present d ay .
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.
Archeology and Palaeontology
Chapter 14 in Ford, T.D. 1977, q.v.
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.
Peak District. Geo-Books.
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.
Pleistocene Vertebrates of the
Waters, R.S. & Johnson, R.H., 1958.

vol. 2, NO . 9, pp 3-15.
Trevor D. Ford,
Geology Department,
University of Leicester,
Leicester LEI 7RH.
British Isles. Longman, London 212 pp.
The terraces of the Derbyshire Derwent.
M. Gascoyne,
Atomic Energy of Canada Ltd.,
Pinawa,
Manitoba,
Canada.
115
East
East Midi. Geogr.
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
THE BRITISH NEW GUINEA SPELEOLOGICAL EXPEDITION

compiled by Dave Brook
(Transactions British Cave Research
Asoociation
Volume 3, combined nos 3& 4, 1976 )
Available from B.C.R . A. Sales
3 (plus p & p 1)
DAN YR OGOF
by Alan Coase & Dave J udson
(Transactions of the British Cave Resear c h Ass ociat ion
Volume 4 , combined nos 1 & 2 , 1977)
(With large survey sheets )
Available from B . C. R. A. Sales
4.50 (plus p & P 1 )
Write for full price list of other B.C.R.A. publications
B.C.R.A. Sales
c/o 30 Main Road, Westonzoyland, Bridgwater, Somers et .
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 :
Dr . G. T . Jefferson, Zoology Dept ., University College , P.O . Box 78

Cardiff. CFl lXL
J . J . Rowland, Llywn yr Eos, Capel Bangor, Aberystwyth , Dyfed . SY 23 3LR
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
Treasurer : Joseph W. Dey, 2 Wyatt Avenue, Sheffield. Sll 9FN .

Editor : Dr. T.D. Ford , Geology Dept . , The University , Leicester LEl 7RH
Bullet i n Edi tor: Juan S. Corrin, 55 Osborne Terrace, Bacup, Lancs. OL13 8J Y
Librarian : ROy Paulson, 26 Ashley Road , Keyworth, Nottingham.
Fore i gn Secretary: J.R. Middleton, 2 Broad Elms Close , Sheffiel d. Sll 9ST
Conservation Officer: D. M. Judson, Rowlands House , Summerseat, Bury, Lancs . BL9 5NF
Sales Officer: B.M . Ellis, 30 Main Road , Westonzoyland , Bridgwater , Some r set . TA7 OEB
Biolog i cal Recorder : M.C . Day, 118 Whitmore Road , Harrow. HAl 4AQ
Mining Recorder : P.B . Smith, 21 Lees Hall Road, Sheffield . S8 9JH
Archaeological Recorder: Dr. J . D. Wilcock, 22 Kingsley Close, Stafford . ST17 9BT
Bookings Manager: Dr . R.G. Picknett , 28 Potters Way , Laverstock, Salisbury , Spl l PX
Distribution Manager: P . R. COUSins , 8 Giffords Croft, Lichfield , Staffs.
Ins u ra n ce Manage r : Geoff Wells, 39 Linden Road , Redland, Bristol . BS6
Adver t ising Manager : Keith Plumb , 55 Firwood Avenue , Urmston, Manchester.
Club
Representatives
D. Padley , 12 st . Anns Lane , Sutton Bonnington , Loughborough .

S . Moore , Havencourt , The Parks, Aldington, Evesham, Worcs .
Peter Robertson , 36 Tadfield Road , Romsey, Hants. 505 8AJ
Pat Devine , 12 Parkside Road , Sale , Cheshire .
Co - opted Dr . C. Wood , Yorkshire Dal es National Park , Colvend, Hebden Road,

Grassington . BD23 5LB
Back numbers of the Transactions and other publications can be obtained from
Bryan Ellis , 30 Main Road, Westonzoyland, Bridgwater, Somerset TA7 OEB .
Opinions expressed in this publication are the sole responsibility of the
individual authors and in no way may be taken as representing the official
view of the British Cave Research Association , unless otherwise stated.

Bcra 10-2-1983

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bcra 10-2-1983

Uploaded by

Copyright:

Available Formats

Ca e Science

The Transactions of the Briti.h Cav. Rrch A.sociation

BRITISH CAVE RESEARCH ASSOCIATION

ISSN 0263-7 60X

Volume 10, Number 2

Mean annual runoff and the scallop flow regime in a

Preliminary results from

Svartisen, North Norway

Speleothem dates and Pleistocene Chronology in the Peak

J.S. Beck . . . . . . . . . . . . . . . . . . . 103

The 70 metre shaft in Gua Lebak Bareng by A.C. Waltham

Published by and obtainable from

~ Copyright the British Cave Research Association, 1983.

THE CAVES OF GUNUNG SEWU , JAVA

by A. C . Waltham, P. L. Smart, H. Friederich, A. J. Eavis & T. C. Atkinson

THE GUNUNG SEWU KARST

Kar.t area. of Java

Kall Progo Plain

The Gunung Sewu area

Cro ectlon through Gunung Sewu

1. Cone ka rst with an alluviated valley floor in northern Gunung Sewu

Bribln { ' / . ....

... 1-- ... ..::'

CAVES OF GUNUNG SEWU

50m groundwaler contour

GUNUNG SEWU CAVE SURVEY

GUNUNG SEWU CAVE SURVEY,

WoN "' ''''.lIonl

AND WELL LOCATION

DETAIL OF ENTRANCE SHAFT

Qua 80ng Tawing

Song Tawing Jomblang Orubug

Qua Burl Omah

. .. . ... . .. .. . .. .. . . . . ... ... . . . .... .. . . . . . . .... , .. .. . . . . . . .. . . .. ... .

GRUBUG CAVE SYSTEMS

metrM above 8M .....

' .........., " ,,

.......... "., ,,."'M

GUNUNG .S EWU CAVE SURVEY, JAVA, 1982

GUNUNG SEWU CAVE SURVEY, JAVA, 1982

GUA NGINGRONG - GUA MULO

GUNUNG SEWU CAVE SURVEY, JAVA, 1982

GUNUN G SE1vU, JAVA

4. A water carrier walks to the entrance of Gua Sodong

6 . The ri ve r pa ssag e in Gua Br i bin

@ GUNUNG SEWU CAVE SURVEY

GUNUNG SEWU CAVE SURVEY,

WoN "' ''''.lIonl

AND WELL LOCATION

DETAIL OF ENTRANCE SHAFT

Qua 80ng Tawing

Song Tawing Jomblang Orubug

Qua Burl Omah

. .. . ... . .. .. . .. .. . . . . ... ... . . . .... .. . . . . . . .... , .. .. . . . . . . .. . . .. ... .

GRUBUG CAVE SYSTEMS

metrM above 8M .....

' .........., " ,,

.......... "., ,,."'M

GUNUNG .S EWU CAVE SURVEY, JAVA, 1982

GUNUNG SEWU CAVE SURVEY, JAVA, 1982

GUA NGINGRONG - GUA MULO

@ GUNUNG SEWU CAVE SURVEY

'4"' ~ ' #'