ACTA CARSOLOGICA	29/2	16	213-230	LJUBLJANA 2000
COBISS: 1.08
ARE COLLAPSE DOLINES FORMED ONLY BY COLLAPSE?
ALI SO UDORNICE ZGOLJ POSLEDICA UDORA? FRANCE [U[TER[IČ1
1 University of Ljubljana, Department of Geology, Aškečeva 12, SI-1000 LJUBLJANA, SLOVENIA, E-mail: france.sustersic@ntfgeo.uni-lj.si
Prejeto / received: 7. 9. 2000
Izvleček	UDK: 551.435.84(497.4)
France Šušteršič: Ali so udornice zgolj posledica udora?
Razprava teče o udornicah, enem na videz najbolje definiranih površinskih kraških pojavov. Kljub temu je v literaturi zaslediti zelo različne poglede, nekateri vidiki njihove geomorfogeneze pa so bili doslej prezrti. Predvsem gre za za nadalnji razvoj globeli v kraške površju, potem ko so prepadne stene udornice že uničene, pobočja oblikuje samo še žepasto preperevanje, denudacija pa se je zajedla že globoko pod dno nekdanjega jamskega rova. Take globeli imenuje avtor fantomske udornice. V nadalnjem obravnava pet v Sloveniji najbolj običajnih tipov udornic in jih klasificira s pomočjo terminologije splošne teorije sistemov. Zaključuje, da sam udor stropa ostaja bistven del celotnega procesa, je pa volumsko lahko zelo podrejen. Ključne besede: udornice, vrtače, kraško površje, kraške globeli, spelogeneza, Slovenija.
Abstract	UDC: 551.435.84(497.4)
France Šušteršič: Are "collapse dolines" formed only by collapse?
The paper concerns collapse dolines, which appear to be one of the best-defined surface karst phenomena. Despite this appearance, one may find quite different views in the literature, and some the aspects of their morphogenesis have been overlooked completely. Among these aspects the most obvious is the question of the ongoing development of the closed depression. After the perpendicular walls have disappeared, the slopes are reshaped only by pocket weathering, and denudation penetrates deep below the former level of the pre-existing cave floor. Dolines at this stage of development have been termed phantom collapse dolines. Five of the most common collapse doline types found in Slovenia are considered in terms of general systems theory, leading to a conclusion that cave roof collapse remains the crucial event in a collapse doline's development. However, the collapse event itself may be relatively subdued in terms of the volume of free fallen mass involved.
Key words: collapse dolines, dolines, karst surface, karst depressions, speleogenesis, Slovenia.
INTRODUCTION
A number of définitions within the karst sciences have tended to be rather volatile, but that of the "collapse doline" has remained essentially unchanged for more than a century, because the underlying concept appears to be unequivocal. This paper sets out to show that, though the central idea remains unchanged, field study has revealed a variety of evidence that makes the general understanding of the expression "collapse doline" much wider. Though application of a rudimen-tary general systems theory terminology has proved to be the most effective tool in attempting to find a pattern among various possible outcomes, the paper does not set out to rearrange wider knowledge about collapse dolines in these terms.
Nineteenth century recognition of medium-scale, more or less circular, closed depressions wit-hin the karst surface, which were immediately referred to as "dolines", inevitably raised questions about their origin. Initially, sudden collapse of cave roofs appeared to be the most likely explana-tion. However, it soon became apparent that two fundamental types of doline exist, i.e. solution dolines, formed gradually by locally enhanced dissolution of the surface bedrock layers, and col-
Sl. 1: Preseki udornic, kot jih navajajo različni avtorji.
lapse dolines, bound somehow to some sort of catastrophic collapse event. The idea that the bulk of the collapsed material must be removed by the main underground drain, rather than by tributary vadose water and seepage, appears to be ubiquitous, though (in the present author's opinion) was never expressed explicitly.
During the Twentieth Century, monograph descriptions of collapse doline sections soon began to become very much alike. (Fig. 1). The principle of ergodicity was accepted tacitly, without question. Thus, the Davisian concepts of "young" and "old" landscapes and landforms appeared to be self-evident. Without any profound discussion about their development, dolines with a greater share of bare rock slopes (walls) were accepted as younger, whereas those without walls (if ever recognised as being collapse dolines) were accepted as "older". The differences between collapse and solution dolines are obvious when dealing with "young" examples. However, the idea of equi-finality appears always to be present among "older" examples, with both types showing a similar appearance, as echoed in Mc H. Connely and J.M. Horn's (1972) paper.
A short glance at Cvijic's (18951, 42) drawing of the section of the Črna jama entrance chamber (Postojnska jama system) (Fig. 2) makes it clear what H. Cramer (1944) meant to say when providing the definition (o.c., p.327): "A collapse doline can contain open connections to the un-derlying karst-cave and is as such recognisable without problems. The local collapse of the ceiling of a cave is a result of lack of stability in the cave's arch. This is triggered by continuous erosion of the karst-surface... (= exposure of karst-cavities due to denudation of the land surface."2 Conside-ring the idea of ergodicity (taken as self evident), dolines without open entrances to caves, even without perpendicular walls, were considered to be "older" stages of the same phenomenon.
Fig. 2. Cvijic's (1895, 42) drawing of the Črna jama. Note that the underground river's elevation value is wrong, and should read 508 rather than 408.
Sl. 2. Cvijiceva (1895, 42) risba Črne jame. Višina vodne gladine je napačna, približno prav bi bilo 508 in ne 408.
1	Non-Slav readers are referred to the (1893) German edition of essentially the same book.
2	"Die Einsturzdoline kann offene Verbindungen zur unterlagernden Karsthöhle aufweisen und ist sodann als solche ohne weiteres erkennbar... Der lokale Deckenversturz einer Höhle erfolgt bei mangenlender Standfestigkeit der Höhlenfirste. Er wird ausgelöst duch fortschreitende Abtragung der Karstoberfläche ... (= Freilegung der Karsthohlräumen durch Oberflächen-denudation)."
ABOUT THE PROCESSES
Intuitively, the basic processes3 involved appear to be:
Table 1.
RM	removal of fallen material	VOID (NEGATIVE MASS) FORMATION
FC	formation of large caverns	
AP	approaching of the surface / to the surface	
CL	collapse / opening to the surface	DOLINE CREATION
SR	slope retreat	SHAPE (TRANS) FORMATION
DD	gradual disintegration due to denudation	
Detailed discussion of the processes can be found within the cited literature. In the following consideration only those aspects that presently appear to be less well known, or are crucial to the understanding of other sections, are discussed.
Logically, opening to the surface is the step that transforms an underground phenomenon into a surface one. The act itself, however, is not essentially karstic. Rather, in the sense of F. Šušteršič (1996), it is cosmopolitan, as it does not form part of any specific geomorphic system. In other words, it is a reaction of an unstable rock mass, regardless of why and how its mechanical equili-brium has been disturbed.
The idea that the original cave chamber must be very large at the time of the formation of the doline appears to be a corollary. Nevertheless, many field measurements (F. Šušteršič, 1973, 1974, 1997) reveal that even very large collapse dolines may evolve from cave chambers of relatively modest ground plan. This can be a consequence of:
•	existence of vertically oriented negative masses (voids) within the parent rock, or/and
•	prolonged mass removal, continuing far into the time of the doline's shape transformation.
The former explanation has been demonstrated in the case of Vranja jama (884, north of Planinsko polje). The collapse appeared at the exact location of a phreatic jump between two inception horizons (F. Šušteršič, 1994). Similar events are at least strongly suggested at several locations within the nearby system of Najdena jama (259). It appears that they are likely to exist where relatively small cave channels are cut by a sudden breakdown.
The latter explanation is obvious in cases where extremely large dolines lie at the intersection of regional underground water flow and local strike-slip faults (F. Šušteršič, 1997). In both cases,
3	Generally, but not inevitably, starting at the top of the Table 1.
4	Numbering is in accordarne with the central cave register of Slovenia, maintained by the Speleological Association of Slovenia and the Karst Research Institute ZRC SAZU.
tectonic loosening of the parent rock has enhanced, if not governed, the collapse process.
Removal of mass by system drains has been mentioned as inevitable. However, it is not the only influence upon doline volume and shape. In some cases, like Dolec in the Najdena jama system, a permanent draught of cold air enters the cave from between collapsed boulders, regard-less of the season. On the other hand, no significant blowing holes appear in the floor of the collapse doline about 90m higher, even in the coldest and snowiest winters. It is clear that the effects of winter cooling of the collapse material can sustain the sucking of surface air into the cave throug-hout the year. Warm, water-saturated summer air is thus cooled down, yielding a quantity of condensation water. The base of the collapse is accessible only from one side, in one of the most inaccessible parts of the cave, which is flooded for several months every year, so systematic logging of the meteorological conditions has not yet been done. However, estimates based on sporadic observations and existing literature, suggest that the potential role of condensation water in corro-
ding fallen material cannot be ignored (V.N., Dublyansky, & Y.V., Dublyansky, 2000).
*
Though generally not mentioned explicitly, the last step in the logical chain5 is the gradual disintegration of the doline's slopes (and consequently the gradual loss of the doline's identity within the surface) due to denudation. Recently, surface caving has revealed several evidently "old collapse dolines" with unroofed caves within their slopes but a solid rock bowl in the floor. Evi-dently, their present appearance reflects the denudational decay of a "normal" collapse doline and its surroundings.
Denudation of the surface around and within the doline is constantly active. Being relatively slow, it may be neglected, at the time while mass removal by an underground drain and/or condensation water and parallel slope retreat are operative. Later, surface denudation becomes the only agent that operates upon the doline, slowly equalising vertical differences and making the doline shallower. After any scree material on them has been removed, the doline slopes are reshaped too. Pocket weathering does not differ from that in areas outside the doline, and the doline slopes beco-me more and more integrated into the neighbouring surface. Field observations at Vodiska dolina, east of Ivanje Selo, show that the scree material in the centre decays faster than the surrounding slopes, and for some time a characteristic central lowering is evident. Eventually the now-subdued former collapse doline becomes just one among a variety of hollows on the land surface. Only outcrops of cave sediments (i.e. related to intersected caverns) within the slopes testify to the origin of the hollow. Such hollows are described as phantom collapse dolines.
Eventually, all negative masses (voids) of whatever origin within the denuded parent rock mass would evolve into similar depressions. Thus, a proportion of karst surface hollows that can-not be explained directly, or which display no genetic relationship to observable processes or structures, are deduced to be phantoms of former underground/negative relief forms.
5 It has to be asked why, to the author's knowledge, this final stage has been completely ignored in the literature.
SOME MOST FREQUENT OUTCOMES
A widely but tacitly accepted idea that, within a doline, the listed processes (see Table 1) operate in a simple time succession (i.e. only one can operate at a time) appears to be obvious. However, field evidence does not support this oversimplification. Instead, a number of combina-tions may occur, resulting in a variety of outcomes that may seem contradictory.
The processes may either follow one another in a strict succession or (at least some of them) can operate simultaneously. Figures 3 and 4 and the following text, describe some of the more frequently met examples, based on the evidence of field observations. The situations shown in the upper parts of the figures represent the time when a particular process actually operates (bold lines), whereas the sections at the bottom give an impression of the general aspect of the doline at various significant stages.
However, timescale is inevitably arbitrary. Though field observations do not support the idea of a sudden instantaneous collapse of the cave roof, one may speculate that, relatively speaking, collapse takes hardly any time, compared to the time needed for a big cave chamber to develop. Similarly, the time needed to transform a steep-walled depression into an inverted cone-shaped doline is not negligible. Even this time is relatively short, however, compared to that needed for denudation6 to annihilate the doline, i.e. to let down the surrounding surface below the level of the rock floor of the pre-existing cavern.
Considering negative mass transport (non-denudational removal by the underground drain) of breakdown material as the main process, collapse dolines may be either closed or open systems. Nevertheless, it must be stressed that there is complete equifinality among collapse dolines (phantom dolines), though their volumes may differ significantly due to the state of the system.
The text relating to the figures does not set out to describe the full the development story, but instead comments only upon the more interesting details.
Case C1: totally closed system, stable cave roof (Fig. 3, C1).
The expression totally closed system indicates that mass removal by the underground drain ceases long before establishment of a stable roof in the big chamber. In this case gradual break-down of the ceiling and consequent accumulation of fallen material on the floor, will initially insulate the void from any neighbouring cave channels. The process should stop as soon as a stable arch (A. Scheidegger, 1961) is achieved.
Some (perhaps exceptional) field examples demonstrate that this does not happen immedia-tely. During this process, adjusting of the cavern to the geostatic stress field will gradually bring about mechanical equilibration of the ceiling, i.e. formation of a parabolic arch (Ph. Renault, 1967) and rounding of the ground plan of the cave, making it more or less circular, or at least elliptical (F. Šušteršič, 1974).
To gain a basic impression, the reader is referred to work by J. Kortnik and F. Šušteršič (2000 / this volume, p. 152, Fig. 1). The sub-circular ground plan of Brezno pri Medvedovi konti (2330) testifies that it must have "migrated" a significant distance upwards. At present its ceiling seems to have achieved a stable arch morphology. In this particular case some roof spalling due to
61. Gams (1964) suggested 65m/Ma for the Ljubljanica river catchment area.
winter freezing is still active. However, it is induced from outside the cave and it may be assumed that - as in cases where cavern roofs are covered by flowstone - it will not collapse until denudation acting alone destroys it. On the basis of current denudation rates (I. Gams, o.c.), this will take some 102 Ka. See J. Kortnik and F. [u{ter{i~ (o.c.) for more detailed information about the process of cavern roof decay.
FŠ 20«)
Fig. 3: Collapse dolines explained in terms of closed system. Sl. 3: Udornice interpretirane kot produkti zaprtega sistema.
Case C2: totally closed system, unstable cave roof (Fig. 3, C2).
In some cases it appears that a cave roof cannot achieve stable arch morphology. Though the cavern's ground plan may be at least isometric, the roof looks very uneven and free of flowstone cover, and collapsed material on the floor looks fresh. The chamber appears still to be active, in the sense that the process of collapse is still active. Such cave roofs are likely to continue to decay until collapse transforms the cave into a doline. Because the bulk volume of broken rock is larger than the volume of the solid source rock (F. Sušteršič. 1973), such a process brings about a reduction of the cave volume7.
Cases C1 and C2 differ in the way that the arched cave roof arrives close to the surface - that is, how the ceiling becomes too thin to support itself. In the second case, both cave roof block spalling and denudational surface lowering operate. In the first case, only the latter operates and the bulk process might be much slower. Though no direct measurements of the rate of roof spalling in caves of this type exist, the process appears to act much more quickly than simple denudation, perhaps by a factor of 10 or more.
An alternative history of cases C1 and C2 may follow from a phreatic jump situation (F. Sušteršič, 1996). In this case, the cave chamber will be little wider than neighbouring "horizontal" passages. Volume will be distributed vertically, and it would not be expected that the roof would be unstable. After denudation has lowered the surface sufficiently, a relatively narrow but deep, ver-tical-walled, depression will appear, soon to transform into a more "normal" collapse doline of smaller dimensions. It appears that the entrance "shaft" of the Gradišnica cave (86 / M. Marussig, F. Velkovrh, 1959) is of this type. Further development, however, will not differ significantly from that just described, except that during the "time" of phantom doline development the unroofed cave will appear only on one side of the doline at the same time.
Case C3: Hypothetical outcome: partly open system, closure at the time of collapse
(Fig. 3, C3).
P. Habič (1963) noticed that the floors of various collapse dolines in the area south of Ljubljanski vrh (20 km SW from Ljubljana) are grouped within relatively few absolute level-determined groups, though their surface lips are at different elevations. He concluded that this is the consequence of two facts:
•	that the caves below were formed at clearly defined levels, and
•	that removal of fallen mass lasted until the roof eventually collapsed.
For many years the validity of the former idea was in doubt, because only fragments of phreatic systems8 were known in the general neighbourhood. Now that several caves have been explored fully the idea can be refuted. Reliable field evidence (M. Brenčič, 1992; F. Sušteršič, 1994) con-firms that caves at discrete levels neither exist nor existed in this area, and that there is no need to
7	If the ceiling is thick enough the chamber could eventually be filled completely (the cave space consumed) due to relative mass increase of the fallen material (F. Sušteršič, 1974). In this latter case, however, the process will stop before the collapse reaches the surface, and no doline will appear (F. Sušteršič, o.c., 30, Fig. 1b).
8	However, this considers the absolute elevation only very approximately, i.e. in the same sense as S. Wort-hington's (1991) "tiers".
postulate them. In contrast, the latter idea implies that all cave roofs are of approximately equal thickness at the moment of collapse. Even if such coincidence seemed in any way reasonable, the underlying supposition - that at the precise moment of collapse the debris would block the underground stream and halt mass removal - sounds highly unlikely. Additionally, Habi~ (o.c.) overloo-ked the fact that, if this were the situation, the dolines at higher "levels" would be "older". In this case, the best that can be said is that the field evidence is unconvincing.
Case O1: totally open system, shallow underground stream (Fig. 4, O1)9.
In the Rakov Skocjan (karst) valley, the cave system has developed relatively close to the surface. Locally, the variations between massive reef limestone and bedded limestone make litho-logical conditions for cave development very heterogeneous. Consequently, relatively small cave channels formed within the massive rock alternate with large cave chambers formed in bedded rock.
Due to the proximity of the surface, the ceilings of the chambers will spall down progressively until breaking through to the surface. Meanwhile, the flow of underground stream water below is
Fig. 4: Collapse dolines explained in terms of closed system. Sl. 4: Udornice interpretirane kot produkti odprtega sistema.
9 Note that the last stages (phantom dolines) are omitted from this figure (cases O1 and O2).
strong enough to remove the collapse material simultaneously. Consequently, the slope retreat of the newly formed dolines is unconstrained and they merge together. Eventually, a string of collap-ses will change into a canyon that will gradually be transformed into a karst valley. Denudation, however, cannot lower the bottom of the depression below base level, and consequently, the valley will evolve into a wide depression with gentle slopes, which might be compared to a small karst polje.
It must be stressed that significant tectonic structures are unnecessary and relatively large cave chambers can develop due only to the effects of lithological contrasts, later to be transformed into collapse dolines. The underground stream10 retains its primary position because it can cope with removal of the total amount of collapsed material.
Case O2: totally open system, deep underground stream (Fig. 4, O2)11.
The present volume of the Brezno pri Medvedovi konti chamber is 62.0 x 104 m3 (F. Šušteršič, 1973), which appears to be "enormous" for a cave. The doline that will evolve from it, however, will not look spectacular, compared with larger examples described in this paper. It follows that if the development history is as described, the mere existence of a collapse doline suggests the preexistence of a large cave chamber. The partial collapse of the relatively large chamber, and the related small collapse doline in Črna jama (Fig. 2) are a good example.
By reference to the example of the Rakovska kukava, F. Šušteršič (1997) demonstrated that very large collapse dolines may evolve from relatively small cave chambers. However, a much more impressive example, is provided by Laška kukava. This is nearly 100 m deep and its volume surpasses 4 Mm3 (Table 2). There are two active foci of recent material removal in its floor (F. Šušteršič, 1974).
A clue about the formational mechanism of this type of collapse doline is found not far away in Riba jama (248). The whole of the accessible cave is a single, c. 40 m-deep, cavern with an amoe-ba-like vertical section. Evidently the cavern formed by the simple settling down of tectonic crush within the broken zone of a local strike-slip fault. A similar situation is found in all of the kukave12, which are all crossed by similar faults. The explanation is that underground water finds such zones difficult to break through. Consequently, once a route was opened, flow along it would persist, even if the passage was being repeatedly obstructed by periodically collapsing tectonic crush. Such material being unstable, the process would continue until a Riba jama-like cave appeared at the surface.
If the karst stream flow persisted long enough in the same position, the "cave" would even-tually evolve into a doline, which would increase in volume until the stream flow ceased (start of closed system conditions), or until it appeared on the surface. The time span required for the latter outcome to be realised appears to be longer than the time needed to rearrange the cave system, and field examples are rare.
10However, the underground stream will soon change into a surface one.
11Note that the scale is different in each example, and that in this particular case (O2) the volume of the
doline might be as much as 100 times larger than that in the previous one.
12Singular kukava. A local name for very large, evidently collapse, dolines. The expression appears to be derived from the Latin expression concavus = inflected inwards.
It must be stressed that the volumes of the largest dolines (kukave) surpass those of the largest cave chambers by a factor of about 20, and the development of cave chambers of approximately similar volumes are mechanically impossible (F. Sušteršič, 1973).
Table 2: Volumes of collapse dolines (popularly named kukave) and volumes of the largest known cave chambers in the area (F. Sušteršič13, 1973).
"kukave"			Big chambers		
Laška kukava	4.17	Mm3	Gradišnica / Blatna dvorana14	37.5	x 104 m3
Smrkovca	1.6	Mm3	Najdena jama / Putickova dv.	7.2	x 104 m3
Rakovska kukava	1.35	Mm3	Najdena jama / Sulčeva dv.	5.0	x 104 m3
Dolga dolina	1.1	Mm3	Logarček / Blatna dvorana	4.2	x 104 m3
Gladovec	0.92	Mm3	Jama za Bukovim vrhom	3.9	x 104 m3
Ivanjska kukava	0.85	Mm3	Mačkovca / Velika dvorana	2.4	x 104 m3
Cerkniška kukava	0.53	Mm3	Logarček / Podorna dvorana	1.2	x 104 m3
*
The "cases" examined above are not the only possibilities. Combining the duration of the main processes and the possibilities of closed or open system conditions, other outcomes may be hypot-hesised. The options discussed here are those that appear to be supported by field observations or, in other words, the ones that are needed to explain, and allow understanding of, the collapse doli-nes of south-central Slovenia.
CONCLUSIONS
•	The models presented demonstrate that a collapse doline is an underground-rooted surface karst form. More rigorously, it may be considered as a non-karstic projection of a karst void onto the karst surface.
•	Among the major types of mass movement (M. Summerfield, 169, Tab. 7.5) during collapse doline development and transformation, cavity collapse (free fall), rock fall (from vertical faces), rock block slide, debris slide, rock slump and talus creep - vaguely equivalent to collapse and its direct consequences - appear always to be present. Thus the very act of collapse remains the crucial event. But it may not be very significant, it may be less than spectacular, or it may even be reduced almost to become negligible.
13After more detailed measurement some of the volumes were later corrected/increased. The data in Table 2
are the latest available.
14In Slovene, the expression dvorana means a very big room.
ACKNOWLEDGEMENT
Thanks to Dr. David J. Lowe for smoothing the text, and many little suggestions which impro-
ved the contents.
REFERENCES
Bren~i~, M., 1992: Koselevc (Summary15). Nase jame 34, 41 - 51, Ljubljana.
Central cave register of Slovenia, maintained by the Speleological Association of Slovenia and the Karst Research Institute ZRC SAZU.
Chorley, R.J., Kennedy, B.A., 1971: Physical geography, a systems approach. Prentice-Hall International, 1 - 370, London.
Cramer, H., 1944: Die Systematik der Karstdolinen. Neues Jahrbuch für Mineralogie, Geologie, und Paläontologie, Beilage Band, Abt. B, 85, 293-382.
Cvijic, J., 1893: Der Karstphänomen. Geographische Abhandlungen, 5, 217 - 329, Wien.
Cvijic, J., 1895: Karst, geografska monografija. Stamparija kraljevine Srbije, 1 - 173, Beograd.
Dublyansky, V.N., Dublyansky, Y.V., 2000: The role of condensation in karst hydrology and speleogenesis. In A.B. Klimchouk, D.C.Ford, A.N. Palmer, W. Dreybrodt (eds.): Speleogene-sis. National spelological society, 100-112, Huntsville.
Ford, D. C., Williams, P.W., 1989: Karst geomorphology and hydrology. Unwin Hyman, 1 - 601, London.
Gams, I., 1966: Factors and dynamics of corrosion of the carbonatic rocks in the Dinaric and Alpine karst of Slovenia (Summary). Geografski vestnik 38, 11 -68, Ljubljana.
Habi~, P., 1963: "Dolines" en forme de puits, dites "kolisevke", et le cours d'eau soutterain (Résumé). Treci jugoslavenski speloloski kongres. Speleoloski savez Jugoslavije, Sarajevo, 1-272, Sarajevo.
Jennings, J., 1975: Doline morphometry as a morphogenetic tool: New Zealand example. New Zealand geographer, 31, 6 - 25.
Kortnik J., Sustersi~ F.: Modelling the stability of a very large cave room Case study: Brezno pri Medvedovi konti. Acta carsologica, (this volume), Ljubljana.
Kunaver, J., 1960: Brezno pri Medvedovi konti na Pokljuki (Summary). Nase jame, 2, 1-2, 30 - 39, Ljubljana.
Marussig, M., Velkovrh, F., 1959: Gradisnica (Zusammenfassung). Nase jame, 1 (1), 24 - 28, Ljubljana.
Mc Connell, H., J.M., Horn, 1972: Probabilities of surface karst. In: R.J. Chorley (Ed.): Spatial analysis in geomorphology. Harper & Row, 111-133, London.
Renault, Ph., 1967: Contribution a l'étude des actions mécaniques dans la spéleogenese. Annales de spéléologie, 22, 1, 5 - 596.
Scheidegger, A. E., 1961, Theoretical geomorphology. Springer, 1-333, Berlin.
15The titles of summaries/abstracts (if they exist) are given just to show the foreign reader the contents of the original texts, which are, however, considered in the whole.
Summerfield, M.A., 1991: Global geomorphology. Longman Scientific & Technical, 1 - 537, Singapore.
Šušteršič, F., 1973: On the problems of collapse dolinas and allied forms of high Notranjsko (South-
central Slovenia) (Summary). Geografski vestnik, 45, 71-86, Ljubljana. Šušteršič, F., 1974: Some metric problems on the collapse dolinas (Summary). Geografski vestnik, 46, 27-46, Ljubljana.
Šušteršič, F., 1983: A simple model of the collapse dolines transformation (Summary). Acta carsologica, 12, 1 - 32, Ljubljana. Šušteršič, F., 1994: The Kloka cave and speleo-inception (Summary). Naše jame, 36, 9 - 30, Ljubljana.
Šušteršič, F., 1996: The pure karst model. Cave and karst science, 23 (1), 25 - 32. Šušteršič, F., 1997: Rakovska kukava - collapse or tumour doline?. Acta carsologica, 25, 251-289, Ljubljana.
Šušteršič, F., 1998: Interaction between tha cave system and the lowering karst surface. Case study:
Laški Ravnik. Acta carsologica, 27 (3), 115 - 138, Ljubljana. Šušteršič, F., 2000: Speleogenesis in the Ljubljanica river drainage basin, Slovenia. In A.B. Klimchouk, D.C.Ford, A.N. Palmer, W. Dreybrodt (Eds.): Spelogenesis. National speleological society, 397 - 406, Huntsville. Worthington, S.R.H., 1991: Karst hydrogeology of the Canadian Rocky Mountains. Unpubl. PhD thesis, Mc Master University, 1 - 227.
ALI SO UDORNICE ZGOLJ POSLEDICA UDORA?
Povzetek
V krasoslovju je malo jasnih in ostrih definicij. Vsekakor pa spada pomovanje udornic med tiste redke izjeme, katerih podstat se zdi nedvoumna in se jim zato definicija že več kot stoletje ni spremenila. Namen tega prispevka je pokazati, da tudi temu ni ravno tako. Četudi ostaja osnovna zamisel bolj ali manj enaka, kažejo terenska opazovanja precej divergentno podobo. Tako se seveda širi tudi pojem udornice. V nadalnjem besedilu razvrščam osnovne tipe udornic na osnovi splošne teorije sistemov; ni pa osnovni namen tega članka intepretirati udornice na tak način.
Že sam pogled na Cvijicevo (1895, 42 ) risbo prereza Črne jame v sistemu Postojnske jame (Sl. 2) zadostuje, da nam je jasno, kaj je hotel povedati H. Cramer (1944), ko je zapisal (o.c., p.327): "Udornica lahko kaze odprte povezave s podaj ležečo jamo in jo kot takšno spoznamo brez težav. Krajeven udor jamskega stropa je posedica izgube stabilnosti v jamskem oboku. Le to sproži stalno zniževanje kraškega površja.... (= odpiranje kraških jam zaradi denudacije zemeljskega površja. ).
Krasoslovne monografije dvajsetega stoletja lepo odsevajo značilno enotnost osnovnih pojmovanj (Sl. 1). Med temi kaže poudariti mimogrede privzet "aksiom", da dvorane pod udornicami neogibno izvotli sistemski odtok16. Enako brez prave osnove se je uveljavilo načelo ergodičnosti,
16Mišljeno v smislu odvajanja vode iz sistema po freatičnih ali epifreatičnih kanalih. Nasprotje votlinam, ki bi nastale kot posledica delovanja preniklih padavinskih voda v neprežeti coni.
kot neposredna posledica pa so vstopili zna~ilno davisianisti~ni pojmi kot "mlade" in "stare" udor-nice. Brez globlje misli o procesih so tiste z ve~ stenami obveljele za "mlade", tiste z malo ali celo brez sten pa za "stare."
Osnovni procesi, ki (pre)oblikujejo jamsko dvorano v udornico in to dalje v njen "fantom"17 so: Preglednica 1.
RM	odnašanje podorne gmote	NASTAJANJE VOTLINE (NEGATIVNE MASE)
FC	nastajanje jamske dvorane	
AP	bližanje dvorane površju / zniževanje površja	
CL	udor / odprtje na površje	NASTANEK UDORNICE
SR	vzopreden umik pobočij	(PRE)OBLIKOVANJE
DD	postopno brisanje zaradi denudacije	
Odprtje na površje je tisti logi~ni korak, ki podzemski kra{ki pojav (dvorano) prevede v povr-{inskega. Sam dogodek ni bistveno kraški - udor je kozmopolitski proces, ki ni vezan na nek dolo-~en geomorfni sistem (F. [u{ter{i~, 1996). Z drugo besedo, to je reakcija kamninske gmote, ki je bila vržena iz mehanskega ravnotežja, ne glede na to zakaj in kako je bilo porušeno.
Misel, da mora biti izvorna jamska dvorana ob nastajanju udornice zelo velika se zdi pravilo. Mnoga terenska merjenja (F. [u{ter{i~, 1973, 1974, 1997) pa, nasprotno, kažejo, da se lahko velikanske udornice razvijejo iz kraških votlin, katerih tlorisi niso prav veliki. To bi lahko bilo posledica:
•	obstoja navpi~no orientiranih negativnih mas (votlin v mati~ni kamnini), ali/in
•	stalnega odnašanja gmote, ki se potegne {e dale~ v ~as, ko se udornici že (pre)oblikujejo pobočja.
Prvo misel podpira položaj vhodne udornice v Vranjo jamo (8818) severno od Planinskega polja. Do zrušenja je prišlo prav na mestu freatičnega skoka med dvema začetnima horizontoma (F. [ušteršič, 1996). Podobni slučaji so vsaj zelo verjetni še drugod v sistemu Najdene jame (259). Druga razlaga pride v poštev tam, kjer ležijo zelo velike udornice prav na mestih, kjer podzemski tokovi prečkajo krajevne strižne prelome (glej F. [ušteršič, 1997).
Učinkovanje sistemskega odtoka kot dejavnika odnašanja mase smo sprejeli kot nujno. Ni pa rečeno, da je edino. V nekaterih slučajih, kot npr. Udornica Dolec v sistemu Najdene jame, se zdi, da je danes najučikovitejši faktor odnašanja kamnine korzija kondenzne vode.
Čeprav v literaturi večinoma ni omenjeno, je poslednje dogajanje v zgodovini udornice postopno izgubljanje njene identitete zaradi denudacije. Le ta deluje stalno in učinkuje enako na dno
17Glej dalje!
18Krepko tiskane številke v oklepaju so katastrske (identifikacijske) številke jam po katastru jam Slovenije, ki ga vodita Jamarska zveza Slovenije in Inštutut za raziskovanje krasa, ZRC SAZU.
udornice kot na okoliško površje. Sčasoma ga zniža tako, da tudi jamski rovi, ki so botrovali nastanku udornice, preidejo najprej v brezstrope in nato v fantomske jame (F. Sušteršič, 1998). Ker je denudacija do neke mere kaotičen proces, moremo pričakovati, da v daljšem času popolnoma zabriše tudi samo globel v kraškem površju. Dokler je še zaznavna, površje pa leži že nižje od nekdanje jamske dvorane in moremo njen izvor ugotoviti le posredno, lahko po analogiji s fantomsko jamo tako udornico imenujemo fantomska udornica. Lahko pričakujemo, da so različne globeli v kraškem površju, ki jim neposredno ne moremo določiti izvora, pravzaprav fantomi nekdanjih votlin.
Na prvi pogled se zdi, da procesi, našteti v Preglednici 1, delujejo v izrazitem časovnem zaporedju in bolj ali manj posamič. Terenska opazovanja pa te poenostavitve me podpirajo. Se več; kažejo, da so med naštetimi procesi možne številne kombinacije, katerih izidi so med seboj lahko tako različni, da je enotna shema nemogoča
Navedeni procesi si res lahko sledijo v jasnem zaporedju, lahko pa delujeju tudi v drugačnem vrstnem redu ali celo po več naenkrat. Na slikah 3. in 4. ter v spremljajočem besedilu prikazujem nekaj kombinacij, ki pojasnjujejo večino udornic v slovenskem prostoru. Na zgornjem delu posamične slike kažejo odebeljene črte čase, ko delujejo posamezni procesi (kratice so angleške), spodaj pa so preseki udornic ob posameznih pomebnejših trenutkih. Časovna lestvica je seveda precej poljubna.
Če privzamemo odnašanje gmote s pomočjo sistemskega odtoka za bistveno, so udornice glede na ta proces lahko odprti ali zaprti sistemi. Poudariti pa je treba, da je končni rezultat - fantomske udornice - vedno enak, ne glede na predzgodovino. Zelo različne pa so lahko prostornine, pač v odvisnosti od tega, v kakšnih okoliščinah se je sistem zaprl.
Izid C1: popolnoma zaprt sistem, stabilen strop (Sl. 3, C1).
Z izrazom popolnoma zaprt sistem imamo v mislih, da se je odnašanje zrušene gmote kočalo davno prej, kot se je jamski strop uravnotežil v parabolični lok. Dokler se strop še ruši, zadobi dvorana bolj ali manj okroglasto obliko, podorni material pa jo odreže od preostalega jamskega spleta. Ker je strop mehansko uravnotežen (A. Scheidegger, 1961), se nadalnje preoblikovanje ustavi, dokler denudacija dvorane ne odpre z vrha. Osnovni vtis o tej vrsti jame/dvorane lahko bralec dobi pri J. Kortniku in F. Sušteršiču (2000 / v tej knjigi, str. 152, Sl. 1)
Izid C2: popolnoma zaprt sistem, nestabilen strop (Sl. 3, C2).
V nekaterih primerih se zdi, da se jamski prostor ne more uravnotežiti. Četudi je tloris do neke mere izometričen, je strop dvorane zelo nepravilen, na njem manjka sige in podor na jamskem dnu pa je videti svež. Kaže, da se take jame podirajo, dokler ne dosežejo površja in ne preidejo v udornice.
Alternativni poti k izidoma C1 in C2 vodita iz freatičnega skoka. (F. Sušteršič, 1996). V tem primeru bo tloris "dvorane" komaj kaj večji od sosednjih jamskih rovov, postornina pa bo razporejena vertikalno. Jamski strop verjetno ne bo zelo nestabilen; udor se bo zgodil šele, ko bo denuda-cija dovolj stanjšala strop. Nastala bo ozka in globoka udornica. Vzporedni umik pobočij pa bo opravil svoje in kmalu bo povsem podobna ostalim. Zdi se, da je take vrste udor vhodno "brezno" Gradišnice (86 / M. Marussig & F. Velkovrh, 1959).
Izid C3: Hipoteti~en slu~aj: delno odprt sistem; zapora, ko se dvorana odpre na površje.
(Sl. 3, C3).
P. Habi~ (1963) je opazil, da so dnesa udornic južno od Ljubljanskega vrha urejena v nekaj izrazitih višinskih pasov, ne glede kako visoko so njihovi robovi. Sklepal je, da je to posledica dveh dejstev:
•	da so nekdanje jame nastale v izrazitih nivojih in
•	da je odnasanje podora trajalo do trenutka, ko so udornice zazijale na povrsje.
Kasnejsa opazovanja (M. Bren~i~, 1992; F. Susterstë, 1994) so prvo postavko ovrgla, saj na tem ozemlju ni najti drugega kot odlomke freati~nih kanalov19. Druga misel tiho privzema, da so stropi udornic v ~asu zrusenja ve~ine stropa približno enako debeli. Habi~ (o.c.) je tudi spregledal, da bi morale biti udornice v visjih nivojih starejse kot v nižjih, ~esar pa terenska opazvanja ne potrjujejo.
Izid O1: popolnoma odprt sistem, plitev podzemski tok (Fig. 4, O1)20.
V Rakovem Skocjanu poteka jamski sistem blizu povrsja. Precej spremenljiva litologija (menjavanje neplastovitih in plastovitih apnencev) je povzro~ila nastanek sorazmerno velikih dvoran in tesnih prehodov med njimi. Zaradi bližine povrsja se stropi dvoran rusijo in odpirajo na povrsje, vodni tok pa uspe bolj ali manj sproti odstranjevati rusevine. Vzporeden umik pobo~ij udornice siri in te postopoma prerasčajo v kanjon, ta pa dalje v krasko dolino.
Izid O2: popolnoma odprt sistem, plitev podzemski tok (Fig. 4 O2)21.
Na primeru Rakovske kukave sem (F. Susteretô, 1997) pokazal, da iz sorazmerno majhnih dvoran lahko nastanejo zelo velike udornice. Se ve~ji primer je Laska kukava. Globoka je okrog 100 m, njena prostornina pa je ve~ja od 4 Mm3 (Preglednica 2). Na njenem dnu sta danes dve zelo aktivni žarisči odnasanja (F. Susteretô, 1974). Po predloženi razlagi se podzemska voda težko prebija preko strižnih prelomov, ki potekajo po daljsih oseh kukav in kljub stalnemu sesipanju stropa vztraja bolj ali manj na istem mestu. Ce proces traja dovolj dolgo, tudi manjsa jamska dvorana polagoma preraste v zelo veliko udornico in vodni tok se kon~no pojavi na povrsju. Taksni primeri pa so seveda redki.
Poudariti je treba, da so prostornine največjih udornic (kukav) lahko celo za faktor 20 večje od največjih jamskih dvoran v okolici. Dvorane velikostnega reda kukav so mehansko nemogoče. (F. Sustersič, 1973).
19Ki pač niso vezani na neke nivoje, ampak na višinsko dosti bolj ohlapno urejene svežnje (S. Worthington, 1991)
20Zadnja "stopnja" razvoje udornice (fantomska udornica ) je na tej sliki izpuščena (izida Ol in O2).
21Merili obeh slik sta zelo različni. Prostornine udornic drugega tipa (O2) so lahko celo stokrat večje kot v prvem primeru.
Preglednica 2: Prostornine kukav in prostornine največjih znanih jamskih dvoran v okolici (F. Šušteršič22, 1973).
Kukave			Velike jamske dvorane		
Laška kukava	4.17	Mm3	Gradišnica / Blatna dvorana	37.5	x 104 m3
Smrkovca	1.6	Mm3	Najdena jama / Putickova dv.	7.2	x 104 m3
Rakovska kukava	1.35	Mm3	Najdena jama / Sulčeva dv.	5.0	x 104 m3
Dolga dolina	1.1	Mm3	Logarček / Blatna dvorana	4.2	x 104 m3
Gladovec	0.92	Mm3	Jama za Bukovim vrhom	3.9	x 104 m3
Ivanjska kukava	0.85	Mm3	Mačkovca / Velika dvorana	2.4	x 104 m3
Cerkniška kukava	0.53	Mm3	Logarček / Podorna dvorana	1.2	x 104 m3
*
Prej našteti izidi niso edini možni. S kombiniranjem različnih dolžin trajanja posameznih procesov in zapiranjem sistema ob drugačnih pogojih, lahko dobimo še druge teoretične modele. Navedeni pač zadostujejo, da razložimo večino udornic v porečju kraške Ljubljanice. Povedano lahko povzamemo:
•	Prikazani modeli so pokazali, da so udornice površinski kraški pojavi, ki pa koreninijo v podzemlju. Bolj izostreno bi jih lahko opisali kot nekraške projekcije kraških votlin na površje.
•	Med glavnimi tipi premikanja gmot, kot jih navaja M. Summerfield (1991, 169, Tab. 7.5) se v času nastajanja in nadalnjega razvolja udornic lahko pojavijo vsi, ki jih pokriva ohlapna streha "podor". Tako samo zrušenje jamskega stropa ostaja odločilen dogodek. Ni pa nujno, da bi bilo zelo opazno - po prostornini vpletene gmote je lahko skoraj zanemarljivo.
ZAHVALA
Dr. Davidu J. Loweju se zahvaljujem za natančno glajenje angleškega besedila in drobne napotke, ki so ga izboljšali tudi vsebinsko.
22Ponovljene, natančnejše meritve so pokazale, da so resnične prostornine večje od prvotno ocenjenih. Podatki v Preglednici 2 so zadnji razpoložljivi.