The belief that some active orogens may be in a tectonogeomorphic steady-state is widely held but poorly vindicated. In such a state, tectonic rock uplift is balanced by erosion such that the land surface remains invariant. The strongest quantitative evidence for this equilibrium is usually a simple mass balance of estimated rates of tectonic input and of regional erosion, but such estimates are both subject to considerable uncertainty and predicated on conflicting time and length scales. We seek a geomorphic signature that will characterize the degree to which a region is, or is not, in steady-state, at the length/time scale of our choosing. An ideal candidate orogen for such a study is the Central Range of Taiwan, which is uplifting rapidly, at an estimated maximum rate of ~5 mm/y, and propagating laterally, with marine topography exhumed as the range grows southwards. We study a high resolution digital elevation model (DEM) of Taiwan, and employ surface hydrology modelling tools for drainage network extraction and topographic analysis. Our central premise is that the surface fluxes across two similarly shaped subcatchments - one abutting the range boundary and the other commencing at the range divide and only distantly connected to the range front - will be almost identical if and only if the region is uplifting and eroding at equal, spatially uniform rates, but that these fluxes will be systematically different for pre-steady-state, time-dependent topography. Therefore, mean paired pre-steady-state subcatchments should have systematically different geometries, whereas mean paired steady-state subcatchments are expected to have very similar geometries. We will show an analysis of the ensemble variation of subcatchment properties across the Central Range, for which our preliminary results suggest that part of the propagating orogen has indeed reached steady-state.
Most extensional and compressional inverse basins are characterized by very important quantities of sediments (5 to 20 km). The volumes of the sedimentary material are comparable to the local changes in the volume of the crust submitted to extension/compression and subsidence. It is logical to assume that the way in which the material is being deposited and eroded is as important as the way in which it is being deformed tectonically. Thus a kind of feedback should exist between the mechanical evolution of the basin and surface process. Using explicit numerical approach allowing for brittle-elasto-ductile rheologies, large strains, and surface erosion/sedimentation, we conducted computer simulations of simultaneous extension, erosion and subsidence in which we considered both "oceanic" or "young" (single leading lithology) and "continental" (competent upper crust, ductile lower crust, strong mantle) cases. Our results show that the surface processes can significantly change the rate and style of extension and subsidence of the basin and evolution of its shoulders. In continents, loading and unloading of the upper crust due to surface processes may provoke feedback flow in the weaker lower crust accelerating subsidence and uplift. For example, under the same boundary and initial conditions, intensive erosion/sedimentation during the rifting "creates" a much larger basin, with almost 2 times greater "apparent" coefficient of extension. This can obviously change existing interpretations of the style of extension and estimates of the rate of the post-rifting subsidence derived from the conventional models. The other obvious result is that the surface processes greatly influence the geometry and evolution of the rift shoulders, and distribution and evolution of the faults.
Sediment discharge into a basin is a first order control on sequence stratigraphic architectures, acting as a boundary condition for far-field dispersal. Consequently, understanding both the geomorphological process system in uplifting catchments and the tectonic displacement field of linked extensional faults is crucial to the modelling of the sediment flux to neighboring depocentres.
Incorporation of a landslide algorithm is essential for the generation of realistic landscapes in numerical models of basin and range topography (Densmore et al. 1998). In addition, mapping of time series of landslides in the Southern Alps demonstrates that the associated volumetric efflux is comparable to the denudation rate estimated from river sediment loads (Hovius et al. 1997). Catchments in tectonically active regions therefore, whether characterised by convergent or extensional tectonics, can be viewed as being commonly dominated by landsliding as the primary process of hillslope erosion.
A survey of upland catchments and associated basin margin fans in the American southwest shows that catchment and fan areas are related by a dimensionless parameter whose value is determined primarily by parameters relating the sediment discharge of the catchment to the accommodation space in the basin (Whipple & Trayler 1996; Allen & Hovius 1998). Known or inferred variations in the tectonic subsidence rate at the basin margin in the SW USA examples is the most plausible explanation for the observed variations in this dimensionless parameter.
We focus on a number of catchments and fans on opposite sides of Death Valley which have different rates of tectonic displacement. Using a numerical landscape evolution model which incorporates both a geomorphological rule set and a tectonic displacement field, we have produced realistic landscapes which can be compared with the morphometric properties of catchments in the Death Valley region derived from 30 m resolution DEMs. The effects of climate change and of changes in tectonic slip rates on catchment erosion rates and fan deposition rates are investigated, with emphasis on response times to the changes in the forcing mechanism. The results are of interest to those investigating the coupling between tectonic uplift/subsidence, denudation, sediment transport and deposition, but also to those concerned with the relative roles of climate and tectonics in determining sediment supply at a range of time scales.
Allen, PA & Hovius, N, Basin Research, 10, 19-35, (1998).
Densmore, AL, Ellis, MA & Anderson, RS, Journal of Geophysical Research, 103, 15203-15219, (1998).
Hovius, N, Stark, CP & Allen, PA, Geology, 25, 231-234, (1997).
Whipple, KX & Trayler, CR, Basin Research, 8, 351-366, (1996).
The concentration of in situ-produced cosmogenic nuclides in river sediment-borne quartz reflects spatially averaged rates of denudation in a drainage basin (Bierman and Steig, 1996). This technique is applicable for estimation of time-integrated mean denudation rates in small catchments as demonstrated by Brown et al. (1995) and Granger et al. (1996). In this study the method was applied to medium-sized catchments to evaluate the validation of the model beyond the local scale. Denudation rates inferred from 10Be in quartz in river sediment from middle-European rivers in medium-altitude mountain ranges were compared with conventional rates calculated from the present-day flux of daily monitored suspended and dissolved load in these streams. Denudation rates calculated from in situ-produced cosmogenic nuclides are 18-24 mm/ka for the river Regen (SE Germany), which compares to 8 mm/ka as estimated from conventional data. Cosmogenic rates are 50-90 mm/ka for the river Neckar (SW Germany), whereas conventional rates are 35-55 mm/ka. Denudation rates from the two methods agree remarkably well, given that conventional data represent sampling intervals of a few years at most, whereas cosmogenic data integrate over several 1000 years. No regional patterns are visible for samples that have been collected from upstream down to the river mouth. The fact that cosmogenic rates in both rivers are sometimes higher by a factor of two to three may be explained by variations in erosion rates, the difficulties in accurately measuring suspended loads in streams of which a significant fraction is associated with low-frequency flood events, or by the uncertainty associated with the barely constrained amount of anthropogenic pollution in the solute load. However, the concentration of in situ cosmogenic nuclides in river sediments is a valuable and accurate technique to estimate time-integrated denudation rates and applicable to medium scale catchments. The effects of climate, tectonics, lithology and topography on denudation rates may be studied by applying this method to different river basins.
Bierman P & Steig EJ, Earth. Surf. Proc. Landforms, 21, 125-139, (1996).
Brown ET, Stallard RF, Larsen MC, Raisbeck GM & Yiou F, Earth Planet Sci. Lett, 129, 193-202, (1995).
Granger DE, Kirchner JW & Finkel R, J. Geol, 104, 249-257, (1996).
The Blue Mountains region of central New South Wales forms a morphological anomaly on the southeastern Australian high-elevation rifted margin. The Blue Mountains occupy the inland part of the Permo-Triassic Sydney Basin, bordered on both the north and the south by the Palaeozoic crystalline rocks of the Lachlan and New England fold belts; the escarpment is located ~50 km further inland in the Sydney Basin than on the crystalline basement to the north and south. The offshore sedimentation history also records lateral variations in sediment flux and accommodation space along the SE Australian margin. Moreover, the Cenozoic history of landscape evolution, as recorded by mid-Tertiary basalt flows which preserve ancient landscapes, appears to vary laterally: whereas Eocene-Miocene basalts south of the Sydney Basin flowed down paleo-valleys, indicating that landscape dissection was already well underway at the time of their eruption; within the Blue Mountains Miocene basalts cap relatively flat hill-tops, suggesting that most incision post-dates their emplacement.
We have studied the controls on these apparent lateral variations using both field observations and numerical models. We have mapped the Blue Mountains basalt outcrops in detail in order to reconstruct the Miocene landscape and to quantify both the amount of sub-basalt relief and post-basalt incision rates. New geochemical and K-Ar geochronological data indicate that most of the Blue Mountains basalts are co-linear and were emplaced in a relatively short time span (20.1-14.5 Ma). We therefore suggest that their mapped bases record a single mid-Miocene landscape. Sub-basalt relief is remarkably gentle; it does not exceed 100 m for any single basalt cap and is of the order of 200 m for the entire region. This contrasts sharply with a present-day relief of >600 m in major river gorges. By comparison of the reconstructed mid-Miocene and present-day topographies, we estimate plateau lowering and river incision rates at ¾10 and 3 40 m/My, respectively.
We explain the drastic post-Miocene increase in regional relief by passage of major knickpoints up the river gorges. Estimated knickpoint retreat rates are ~900 m/My. The Blue Mountains are bounded on the east by the Lapstone structural complex; a major faulted monocline that forms the present-day escarpment. Extrapolation of the estimated retreat rates suggests that knickpoints were initiated on this structure at ~40-50 Ma. The Blue Mountains escarpment has been interpreted to result from either early Tertiary uplift on the Lapstone monocline or from passive denudation of previously tilted resistant sandstone on the monocline. We employ a numerical surface process model in order to explore these two hypotheses and conclude that both mechanisms have played a role in shaping the anomalous morphology of the Blue Mountains region.
High-elevation passive continental margins are the primary morphological features arising from continental break-up. Determining their long-term history is important in understanding the interaction between tectonics and denudation in their development, constraining numerical landscape evolution models and predicting patterns of drainage development and offshore sediment delivery. In-situ cosmogenic isotope analysis is a valuable addition to the range of techniques available to estimate rates of denudation. When coupled with longer-term, regional-scale estimates, cosmogenic isotope analysis can potentially provide critical constraints on denudation chronologies and models of landscape evolution in passive margin settings.
The main topographical elements of the central Namibian margin (south west Africa) are typical of many high-elevation passive margins. Along the margin a major sea-ward facing escarpment, with a relief of ~1000 m, is located ~150 km inland of the present coastline and separates a gently inclined coastal plain from an interior plateau. In order to assess variability in denudation rates across the Namibian margin, samples for analysis of cosmogenic 10Be and 26Al were collected from the escarpment zone and coastal plain inselbergs.
Samples from the Gamsberg, the highest and one of the clearest expressions of the escarpment in central Namibia, indicate rates of summit lowering ranging from 0.26 ± 0.06 to 0.49 ± 0.1 m/Ma and rates of escarpment retreat on the order of 10 m/Ma. Granite inselbergs on the coastal plain have summit denudation rates ranging from 2.51 ± 0.5 m/Ma to 7.43 ± 1.6 m/Ma. These data are necessarily integrated over the past ~105 a. However, given the extensive evidence for apparently stable arid environmental conditions throughout the Quaternary, and possibly throughout much of the Cenozoic, it is probable that similar rates have characterised this period.
Slow rates of denudation, and by implication slow rates of escarpment retreat, during the Cenozoic are supported by apatite fission track data for the region. The combined data are consistent with a post-break-up denudational record in which an initial phase of relatively rapid denudation occurs in the early Cretaceous, prompted by high local relief generated by rifting and the establishment of new lower base levels, was followed by a phase of much less active erosion. The data presented here are contrary to conventional theories which suggest passive margins evolve through the large-scale uniform retreat of an escarpment originally created at the continental edge at break-up. Instead, they support more recent numerical modelling studies which indicate that escarpment behaviour is controlled by antecedent topography and the location of inland drainage divides.
Bedrock landslides are important agents of hillslope erosion and sediment mobilization in actively growing mountain ranges. Sediment fluxes from mountainous catchments depend in part on the magnitude, timing, and location of bedrock landslides within the catchments, which in turn are functions of the landslide triggering mechanism. Bedrock landslides may be triggered by either tectonic or climatic events. Tectonic events subject hillslopes to seismic accelerations which are topographically focused, and so are highest near ridgecrests and hilltops. Thus, tectonically-triggered landslides nucleate near the tops of hillslopes and propagate downwards. In contrast, climatic events are likely to trigger landslides through two distinct processes: (1) channel incision and consequent lowering of hillslope baselevel, and (2) dynamic increases in groundwater flow, which produce increases in seepage body forces and changes in effective stress. Both of these processes cause climatically-driven landslides to nucleate near hillslope toes. Thus, different triggering mechanisms preferentially remove material from different parts of the hillslopes. During the evolution of a mountain range, repeated application of one or the other triggering mechanism causes landslides that consistently recur in similar mean positions on the hillslopes, yielding a distinctive signature that is characteristic of that mechanism. To quantify these signatures, we examine digital elevation models of three areas, each historically dominated by a different landslide trigger: (1) the Santa Monica and Santa Susannah Mountains, California, in which the 1994 Mw 6.7 Northridge earthquake caused widespread shallow landsliding; (2) the Santa Cruz Mountains, California, in which landslides are dominantly triggered by intense, episodic rainfall; and (3) the Finisterre Range, Papua New Guinea, in which frequent high-magnitude earthquakes cause failure of both ridgecrests and saturated hillslope toes. The distinctive signatures of these areas provide hope that similar signatures in less well-studied areas may allow quantification of probable landslide triggers, aiding hazard assessment.
More than 50 giant landslides, most of them of the rock avalanche type with volumes up to 0.4 cubic kilometres, lie in clusters in front of reverse-fault bounded mountain fronts in the southern Puna Plateau, the northwestern Sierras Pampeanas, and the Cordillera Oriental (24 - 28° S). The catastrophic landslides only occurred at mountain fronts with (I) relief contrasts larger than 400 m, (2) slope angles in excess of 20°, and are (3) composed of either granite, low-grade metamorphic rocks, or coarse clastic sedimentary rocks. These lithologies provide inherent potential planes of weakness, which act as breakaway and sliding surfaces. Furthermore, landslide formation is always (4) related to mountain-bounding faults with Quaternary activity, and (5) the largest landslide clusters exist in front of Neogene transfer faults, which were reactivated as steep reverse faults (Hermanns and Strecker, in press).
The landslides were dated by 40Ar/39Ar and 14C age determination, cosmic ray exposure dating (21Ne) combined with detailed tephrochronologic correlations. Landslides occur in two different morphologic settings: narrow valleys and broad piedmont regions abutting steep mountain fronts. The majority of landslide deposits in narrow valleys are either late Pleistocene (~ 30 ka) or Holocene (~ 5 ka) in age and correspond to periods with increased humidity. The age of landslide deposits in piedmont regions is more difficult to define but these deposits are generally older than the first two groups. Cosmogenic nuclide ages on landslide debris in the southern Puna define a cluster between 150 and 450 ka. These deposits do not have an obvious relation to climate variations; in addition, many of these slides occurred in regions where glacial effects were negligible on morphodynamic processes along mountain fronts. In contrast, the high frequency of landslides in narrow valleys may represent long-distance effects of more humid climate conditions. The erosional power of allochthonous streams with large catchment areas was increased, and sustained undercutting enhanced the conditions for landsliding. However, considering the close vecinity of all landslides with active faults, we suggest that eventual mountain-front failure was triggered by seismicity in most cases. This is especially true for landslides in the piedmont regions, where the large distances between avalanche sites and trunk streams exclude that changed climatic conditions affected the mountain fronts. For example, a ~ 33-kyr-old ash, sampled from an important landslide cluster in the reverse-fault zone along Sierra Aconquija, indicates a vertical minimum displacement of 10 m, suggesting that several strong, shallow-seated earthquakes have occurred along this mountain front.
Hermanns RL & Strecker MR, Bulletin Geological Society of America, (in press).
There happened many landslides due to severe rainfalls periodically during the past 100 years in the Rokko Mountains, Kobe, central Japan. The Kobe Earthquake which killed more than 6,000 persons occurred in January 1995 and also caused many landslides in the mountains. This may also change in hydro-erosional conditions in the mountains. In the present study, firstly, the relationship between the landslides and heavy rainfalls is summarized (cf. there is a distinct 25-30 year period of annual heavy rainfall which caused the many landslides) and, secondly, pond sediments from an experimental pond in the mountains are used to check the change in hydro-erosional conditions. Some core samples were obtained for reconstructing past erosional conditions and estimating sedimentary rate. Sediment traps have been set on the floor of the pond since the time just after the earthquake for checking the sedimentary conditions after the earthquake. Some physical and chemical properties of the sediment were analyzed to investigate changes in hydro-erosional conditions in the catchment area. 137Cs concentration was used as a time marker for the core sediments of past some dozens of years. Some sequences for physical properties of core samples show that there are two major peaks in the coarse fraction; the first major peak from the bottom surface of the pond is corresponded to the heavy rainfall time in 1967, which is supported with 137Cs concentration data, and the second one to the heavy rainfall in 1938. In the heavy rainfall times, grain density of sediments is large and loss on ignition is low, which suggests that fine detritus with low organic matter from the surrounding catchment accumulated in the center of pond bed; quartz content (SiO2) slightly increases in theperiods. Sedimentation rate for the interval before the earthquake is roughly estimated from the core sample dated with 137Cs concentration and that for the interval after the earthquake is estimated with sediment trap. Estimated result shows that the sedimentation rate after the earthquake is considerably larger than the interval without heavy rainfall although much smaller than the times of heavy rainfall. Fine particle content (Al2O3) in the sediment increases after the earthquake, suggesting some changes in hydro-erosional environment, though there is little difference in the content between periods with heavy rainfall and without before the earthquake.
As part of an integrated study of meltwater chemistry, geothermal and volcanic activity and hydrology this paper investigates the dynamics and significance of suspended sediment transport and discharge throughout two substantial jökulhlaup events in Iceland. Despite considerable research on the dynamics of meltwater discharge and quality in alpine glacierised basins there are few studies in large geothermally/volcanically active basins where jökulhlaups are regular events. This paper presents discharge and suspended sediment data from the Skaftá River in S. Iceland during part of the 1997 meltseason when two geothermally generated jökulhlaups occurred. This study is significant as these two jökulhlaups provided a rare opportunity to monitor chemistry, turbidity and discharge before, during and after the events. Departures of a number of hydrochemical species (H2S, CT, pH, Ca2+, Mg2+, K+, Na+, Cl-, SO42- and F-) from background concentrations and relationships confirmed that the floods were geothermal in origin. The continuous monitoring of turbidity and stage allowed the generation of suspended sediment concentration and discharge time series.
The larger of the two geothermally generated jökulhlaup events had a peak discharge of ~570 m3s-1 and a peak suspended sediment load of ~4,600 kgs-1. The total sediment flux over 7 days of this jökulhlaup event represented approximately 50% of the total sediment export of the entire monitoring period (Julian Day 195 - 255). Spatial analysis of sediment transport indicates that significant deposition occurs along the river course.
Geothermally generated jökulhlaups and associated volcanic activity are shown to be of great importance in suspended sediment transport. A significant amount of sediment appears to be deposited along the river course during jökulhlaups and it is then available for remobilisation by future high flow events whereas the sediment transported through the system to the nearshore zone is an important contribution to the coastal sediment budget and coastal stability. Knowledge of the dynamics of discharge and suspended sediment transport are essential if the rivers potential as a source of water or hydroelectric power are to be exploited.
Many coastal lakes were inundated by both the Storegga tsunami (7000 14C yr BP) and the mid-Holocene sea-level rise (the Tapes transgression) in western Norway. The tsunami eroded lake bottoms and deposited graded and/or massive beds of sand, rip-up clasts, and coarse plant material. By contrast, when the rising sea entered the lakes, it deposited only gyttja, silt and fine sand, without causing much erosion of the underlying lake sediments. Storegga tsunami deposits in some coastal lakes were previously interpreted as ordinary marine sediments from the Tapes transgression. Our re-interpretation of these deposits shows that the transgression maximum phase was reached after 6500 yr BP, more than a thousand years later than previously inferred for the coast of Sunnmøre. The new data cannot be combined in a shoreline diagram without showing the 6000 yr BP and 7000 yr BP shorelines as slightly warped.
This paper considers the links between tectonic activity, climatic variability, sediment supply and calcretisation processes with reference to the development of Pliocene (?) to Quaternary age pedogenic and groundwater calcretes in the Tabernas Basin, Almería Province, southeast Spain. Calcretes developed within alluvial fan and fluvial gravels are described from sites adjacent to major tributaries of the Rambla de Tabernas (Nash and Smith, 1998). Six distinct calcrete units can be identified within the basin which have variable distributions but have developed in an identifiable evolutionary sequence. Two pairs of calcrete units are widely present across the basin preserving two former higher land surfaces. Each of these former land surfaces has been planated and subsequently buried by alluvial fan or fluvial gravels. A massive groundwater calcrete unit is present at the base of each gravel sequence, immediately in contact with the underlying bedrock, with a less well developed pedogenic calcrete unit situated at the top of the gravel sequence. The lowest two calcrete units within the basin are more spatially restricted, are confined to the floors and flanks of incised drainage lines, and appear to have been formed by groundwater calcretisation processes.
The operation of pedogenic and groundwater calcretisation processes is considered in the context of the reconstruction of the early phases of landscape development, and is suggested to have been partly controlled by periods of uplift and stability within the Tabernas Basin. Pedogenic calcretes at the top of the alluvial fan or fluvial gravel sequences exhibit an increase in calcrete stage from youngest to oldest parts of the toposequence, with the best developed calcretes at highest landscape positions. This would tend to suggest that pedogenic calcretes have either been forming relatively constantly throughout the history of the basin or have developed in pulses during periods of suitable climate. The operation of groundwater calcretisation processes, however, appears to be more strongly influenced by tectonic movement, with cessation of calcretisation occurring following periods of basin instability and fluvial incision. On the evidence presented by the Tabernas Basin, groundwater calcrete formation would appear to be most closely controlled by a supply of carbonate-rich sub-surface water, the flow of which would be greatly disrupted by uplift and incision.
Nash, DJ & Smith, RF, Earth Surface Processes and Landforms, 23, 1009-1029, (1998).
Glacio-eustasy has been shown to be a primary control on sedimentation in the open ocean and along passive continental margins, but its importance in clastic-dominated deep-marine sequences at active plate margins remains poorly understood. In order to test the relative importance of glacio-eustasy at tectonically active plate margins during times of substantial polar ice, a high resolution 18O and 13C record from the planktonic foraminifera (Globorotalia inflata) was undertaken from the Plio-Pleistocene (ca 1180-600ka) Kazusa Groups, a forearc basin fill, onland SE Japan. This was combined with a high-resolution study of the magnetic susceptibility, total organic carbon, and %CaCO3 in order to evaluate the response to any glacio-eustatic changes in continental-margin sedimentary processes. The sections reveal globally recognised glacial-interglacial cycles, with sandy intervals correlating with inferred glacials, suggesting that relative sea-level changes during glacial-interglacial cycles exerted the primary control on sediment accumulation in the deep - marine forearc basin. Cross-spectral analysis of 18O and 13C data from the inter-turbidite hemipelagic and pelagic mudstones reveals Milankovitch control both at precession and eccentricity modes, with a shift in the irrelative importance at about 900 ka. The results of this study have important implications for stratigraphers and sedimentologists because they show that at times when there is substantial polar ice: (1) the main control on sediment accumulation at active plate margins is glacio-eustatic, and (2) support the sequence stratigraphic paradigm developed from passive continental-margins that global sea-level changes exert a primary control on siliciclastic deposition.
The main structure of external Jura is a flat decollement within triasic evaporites at shallow depth, 200 and 400 m below topographic surface (as shown by ECORS profile and drillings) The Seille river which crosses the external front of Jura, presents 4 main stepped alluvial terrasses visible over 15 km. The relative elevation difference between terrasses decreases from inner to outer chain and disappears in the foreland (Bresse graben). These stepped terrasses give evidence of recent deformation which may correspond to a response of several coeval loadings which act at different scales in space and time:- the continuous Alps building which is accomodated in the external Jura onto the above mentioned flat decollement but also onto the deeper faults in the basement inherited from Oligocene rifting as shown by seismicity.- the discontinuous accomodation of glacial cyclic charge/discharge which affects the distribution and the rate of deformation.
Although the partially preserved older terrasses should record the cumulative effect of the two loadings (tectonic and ice) i.e. the finite deformation. The well preserved younger one's should help for discriminating these two (tectonic and climatc) loadings. The objectives of this project are to identify and caracterize the recent behaviour of active faults, in particularly the effects of temporal loading such as ice cap and to discriminate the glacial signal from the tectonic one. To comply with these objectives, morphological and geological investigations are currently carried out: - topographic levellings across the different terrasses to determine their precise heights, - synthesis of existing lithological data, completed with 20 boreholes to caracterize the thickness and the nature of deposits,- subsurface geophysic (2-dimensional electric sounding) to spatialized the sequences of deposits found with boreholes- datation of deposits (14C, U/Th).
It is well known that the melting of the Fennoscandian ice sheet caused significant uplift of the formerly glaciated areas, and therefore induced bending of the lithosphere. In this work we evaluate the role of lithospheric flexure resulting from post-glacial rebound as a possible source of regional stress variations in the northern North Sea. We use stress information derived from earthquake focal plane mechanisms and from a variety of borehole measurement techniques to establish a stress data base which allows to constrain spatial variations of stress. Wellbore leak-off tests provide a measure for the magnitude of the least principle stress (S3) and show that S3 is consistently high about 150 km west of the current coastline but drops towards the coast. Further, focal plane mechanisms are quite compressional (thrust faulting or strike-slip faulting), where S3 inferred from borehole data is high but show a tendency towards extension (normal faulting and strike-slip faulting) closer to the coast. The pore pressure closely correlates with spatial stress changes (high pore pressures were stresses are high), which infers that deglaciation also induced pore pressure changes. The NW-SE orientation of the maximum horizontal stress (SHmax), that is very consistent throughout northwest Europe rotates in the same region where the stress magnitude variations are observed. We have developed a three dimensional finite element model to investigate the effects of glacial melting and the associated flexuring of the lithosphere on the local stress field in the northern North Sea. The comparison of the model results with the observed stresses suggests that the late Quaternary melting of the Fennoscandian ice sheet strongly influences the in situ stress field in the northern North Sea. The model shows that magnitudes of spatial stress changes due to glacial unloading are comparable to differential stresses expected from geodynamic processes (such as ridge push) and thus can markedly influence current tectonics in the region.
Contemporary seismicity presents a puzzling picture of crustal movements in northern Britain. Although some seismic activity clusters on major through-going faults, such as the Great Glen Fault and Highland Boundary Fault, in regional terms earthquakes fail to delineate tectonic structures or provinces. Instead, the region's moderate, low-magnitude (M<5) seismicity is concentrated in the western Scottish Highlands, where it is clustered in isolated pockets of activity. This regional skew in earthquake activity has recently been attributed to the effects of glaciation, since the extensive Devensian and later, more localised, Younger Dryas stadial (Loch Lomond Readvance: 11,000 - 10,000 yr. BP) ice masses were centred in the western Highlands. In particular, previously recognised associations between the distribution of contemporary seismicity, postglacial faulting and the Younger Dryas ice limits implies that the region's glacial inheritance continues to influence its present-day seismotectonics.
Preferential fault activity along former ice margins is not unexpected, since it is here that the crust responds differentially, from glacial rebound within the ice limits to minimal uplift (or even subsidence) of the ice-free areas beyond. Furthermore, zones of enhanced present-day seismicity in the western Highlands coincide with areas where significant ice thicknesses accumulated during Younger Dryas times. Analysis of high-resolution airborne imagery and fieldwork in the Kintail region shows that postglacial surface displacements are not restricted to isolated, discrete faults, as highlighted by previous workers, but instead are distributed across a broad network of faults whose postglacial displacements are considerably less than those previously documented. The geometry of these relatively minor postglacial structures, located on the flanks of glaciated valleys and lochs or along the crests of adjacent uplands, mimics the general form of the glacial troughs themselves, suggesting that they are subsidiary to larger faults buried beneath the valley floors or loch waters.
A picture emerges in which glacial/interglacial crustal loading cycles during the Quaternary are likely to have promoted the repeated reactivation of favourably-oriented basement faults. In this scheme, pre-Quaternary basement structures define a network of fault-guided valleys and sea inlets which, during glaciations, are preferentially exploited and occupied by ice. Greater ice accumulations in these glacial troughs result, on deglaciation, in differential glacio-isostatic recovery, which is accommodated by reactivation of the underlying faults, thereby imposing further structural control on the glacial landscape. The landscape framework of the Scottish Highlands, therefore, expresses the prolonged interaction between glacial loading, tectonic faulting and seismicity. Arguably the main question arising from the research is to what extent such interactions continue to be active today.
The size of the Antarctic ice sheet during the Last Glacial Maximum remains uncertain, despite attempts to produce a reconstruction consistent with global eustatic sea-level records. An approach to estimating former ice sheet dimensions that is complementary to geomorphological studies is to observe the isostatic rebound in response to deglaciation. In Antarctica, sea-level records which could be used for this purpose are extremely rare, so a geodetic method of measuring the present rate of rebound is highly desirable.
Lambert Glacier is the largest outlet glacier in East Antarctica, discharging into Prydz Bay via the Amery Ice Shelf. Moraines on adjacent mountains indicate that the change in ice thickness since the LGM varied greatly throughout the catchment. The grounding line of the glacier is also thought to have retreated from a position beyond the present front of the Amery ice shelf, indicating significant changes in the distribution of grounded ice.
Predictions of rebound rate were calculated along a north-south transect at 68° E longitude for three scenarios of ice sheet history and a range of plausible earth rheological models. The calculations indicate that the relative uplift rate between sites on the transect varies from +7 to -7 mm/yr for the different ice histories, almost independently of the earth rheology. This may be large enough to measure using long-term GPS observations of vertical velocity.
A permanent GPS instrument has operated at Australia's Mawson Station, near the north end of the transect, since 1993. A new geodetic site near Beaver Lake, on the southwest of the Amery Ice Shelf, was installed in December 1997. Initial results from the first 25 days of data indicate that it will be possible to detect isostatic rebound of less than 1 mm/yr after a few years of observations at the site. Calculating the relative vertical velocity between thesse two sites may therefore make it possible to discriminate between the various scenarios of glacial history.
The waxing and vanishing of land ice sheets are driven by climate. It generates and drives a number of gedynamic processes. The water masses stored in the ice are taken from the sea, by that driving glacial eustatic sea level changes. The lowering and subsequent rise in sea level affect the earth rate of rotation; a glacial acceleration followed by a postglacial deceleration. The rotational changes drive variations in ocean circulation affecting distribution of heat and water masses. The glacial loading deforms the geoid relief not only in the ice proximity but also in the far field. The Fennoscandian glacial deformation and rebound amounts to 830 m with a simingly total compensation by lateral mass movements in the asthenosphere (leaving little or no room for deeper loading adjustments). The central Fennoscandian uplift commenced at about 13,000 BP (i.e. long before this area was deglaciated). The deglaciation period with rates of uplift in the order of some 10 cm per year is in Sweden associated with a very high seismic activity both in seismic recurrance and in magnitudes of events. The dome-like and typical glacial isostatic rebound finished some 4500 years ago leaving a linear uplift factor that seems to originate from an associated phase boundary displacement in lower lithosphere.
Mörner, GeoJournal, 3.3, 287-318, (1979).
Mörner, Earth Rheology, Isostasy and Eustasy, John Wiley & Sons, 251-284, (1980).
Mörner, Tectonophysics, 176, 13-24, (1990).
Mörner, Terra Nova, 3, 408-413, (1991).
Mörner, Quaternary Science Review, 15, 939-948, (1996).
Mörner, Annales Geophysicae, 16, Supl. I, C54, (1998).
Crustal deformations and earthquake-related landform development dynamics may be quantified within identified and localized tectonically active areas using combined geomorphic analysis and in situ-produced 10Be dating of associated surficial features. New 10Be datings of alluvial fan surfaces affected by an active reverse fault located in the Bermejo foreland basin allow to calculate a shortening rate for the Eastern Argentine Precordillera.
The studied fault scarp, Las Tapias Fault (LTF), is located at about 31°S latitude in the eastern part of the Argentine Precordillera fold and thrust belt. The Central Precordillera and Eastern Precordillera morphological sub-provinces form oppositely verging systems that bound the Precordillera thick-skinned triangle zone on both sides of the Matagusanos Valley. Tectonic activity in the Eastern Precordillera is suggested by the occurrence of Quaternary faults as well as shallow seismicity. The 18 km-long LTF crosses a low-lying region between northward trending Sierra de Villicúm and southward trending Sierra Chica de Zonda. Considering the regional state of stress deduced from fault kinematics, this fault is purely reverse. The LTF is characterized by a W-facing, 8-15 m high scarp which may be followed all along its entire length. This indicates that the main thrust fault has reached the surface as a single fault plane. Due to arid climatic conditions, the scarp morphology does not indicate any evolution toward a smooth profile by gradual transfert of material at rates proportional to local slope (i.e., diffusion model). The fault trace crosses three alluvial fan units which are stepped on the hangingwall while they are superposed on the footwall of the LTF (Siame, 1998).
Using topographic profiles constructed across the fault scarp, upper and intermediate alluvial fan cumulated vertical displacements have been estimated at 8±4 m and 18±5 m, respectively. Using in situ-produced 10Be concentrations measured in surface boulders exposed to cosmic rays, alluvial fan surface minimum exposure ages have been calculated at 1.0±0.5 ka for the lowermost fan, 6.0±1.4 ka for the intermediate fan and 16.0±3.6 ka for the uppermost fan. These data lead to a Holocene vertical slip rate of 1.5±1.0 mm/yr on the LTF. Considering the fault dip estimates, this yields a shortening rate of about 2 mm/yr (Siame, 1998). The fault geometry as well as the calculated rate of underthrusting suggest that the LTF may represent a recent out-of-sequence development of the major Eastern Precordillera W-verging thrust. In addition, the LTF geometry and the calculated slip-rates imply that this fault may produce earthquakes having a maximum magnitude (Mw) of 6.5-7.0 with a recurrence time of about 600 years.
Siame LL, Cosmonucléide produit in-situ (10Be) et quantification de la déformation active dans les Andes Centrales, Thèse de l'Université Paris-Sud, 471 p., (1998).
Morphological analysis based on diffusive analogue is used to describe slope processes. It enables the dating of seismic scarp from its geometry, when an intial topography is assumed. When the fault scarp corresponds to a single seismic event, its intial topography is roughly standart (initial slope angle = slope of repose of material). Morphological dating can be used for such scarps in a lot of active tectonic studies (for example: Avouac, 1993). When scarps correspond to cumulative displacements, one model derived from diffusion analogue is classicaly used to infer the age of the uplift (Hanks, 1984). This model considers that the uplift is continuous, and that surface faulting is vertical and fixed through time. However, numerous examples of trench logs show that the fault can move. Moreover, many examples of cumulative scarps associated to reverse faulting present a free face at their base, suggesting a forward step of the rupture through time. Continuous uplift model can not be applied in this case. By considering several basic tectonic settings, we propose alternative methods to date cumulative reverse fault scarps. We choose numerical approach to compute seismic dislocations alterning with diffusive erosion periodes. Tectonic displacement is simulated by translation of altitudes (the rupture moves forward at each earthquake) and diffusion is linear. Experiments show four recurrent morphologies for which we test their robustness according to variations of tectonic and erosional parameters.
It appears that only one morphology can be univocaly linked to constant forward migration of the fault. A very simple criterium using geometrical parameters extracted from scarp profiles, enables to infer a maximum diffusion age of the scarp (product of diffusion coefficent diffusion and age). In order to give a more accurate value of the diffusion age, we compare slope data with an analytic solution.
We apply this method along the Gurvan Bulag thrust fault (Mongolia), where we surveyed a 20 m high cumulative reverse fault scarp. We determine a diffusion age of the scarp to be 40 m2. Assuming a diffusion coefficient for arid regions in the range of 0.5-1 m2/ka, we determine the age of the scarp to be in the range of 40-80 ka.
Avouac J-P, J. Geophys. Res., 98, 6745-6754, (1993).
Hanks TC, Bucknam RC, Lajoie KR & Wallace RE, J. Geophys. Res, 89, 5771-5790, (1984).
The Moyenne Durance Fault is located in southeastern France (Provence region), about 50 km north of Marseille. It is possibly the only fault in France for which there exist three independent types of observations supporting its present activity: modern instrumental seismicity, historical seismicity (four damaging earthquakes in the past five centuries), and palaeoseismicity (one episode of reverse dislocation by one meter between 9,000 and 26,000 yr ago). The left-lateral Moyenne Durance Fault borders the eastern part of the Provence thrust sheet. We analyzed in detail the southern portion of the fault, between Peyruis and Corbières, where its main trend exhibits a discontinuity associated with a 10-20° right bend, thus giving rise to the Manosque Anticline. The anticline is formed by Oligocene to Miocene limestones and sands, but sediments belonging to the Tortonian-Messinian (10-6 Ma), Valensole I formation, are also involved in the deformation.
Even though the Moyenne Durance Fault is one of the fastest slipping faults of southern France, it has a slip rate of less than one mm/yr, which implies that the evolution of the local landscape is dominated by the erosion/deposition cycle rather than by tectonics. Nevertheless, dislocation modelling of selected landscape and topographic features of the region suggests that uplift of the Manosque anticline can be seen as the long-term response of a local change in strike along the main trend of the Durance Fault. The buried fault that is inferred to be responsible for the growth of the anticline dips gently to the west-northwest and its geometry is consistent with the interpretation of seismic reflection profiles crossing the region from the west (Vaucluse Plateau) to the east (Valensole Plateau). The sense of slip on the inferred fault is dominated by left-lateral motion with some reverse component. In contrast, the Miocene stress field was dominated by west-northwest-trending compression. This result suggests that a major reorientation of the remote stress field must have taken place during the Pliocene.
Surface deformation produced by large earthquakes leaves unequivocal and often clearly visible tectonic imprints. In contrast, moderate earthquakes (M<6) comparable to the 1997 Umbria-Marche seismic sequence, exhibit faulting geometry and dimensions that did not express surface rupturing. In order to study the surface effect of tectonic processes, and constrain the amount of coseismic and postseismic deformation, we did repeated surveys of a levelling line across the Colfiorito normal fault, associated with the Ms 5.9 main shock of 26/09/1997. The levelling measurements performed along a moderately damaged aqueduct (designed in 1960) reveal 45 ± 5 cm of total coseismic vertical deformation. The coseismic deformation is not localised, however, and it is distributed over 2 km distance consistent with surface warping. The postseismic deformation obtained from three sets of measurements (over 13 months) shows a reversed movement that amount ~ 20% of the total vertical offset. Modelling of displacements in an elastic half space predicts a maximum of 55 cm coseismic surface displacement, for 1.0 m slip at depth on a 60° dipping fault, and with the fault tip at 3.0 km depth. Furthermore, the seismic moment Mo = 1.14.1018 Nm associated with the Ms 5.9 main shock, yields 70 cm average dislocation, and suggests a good agreement between measured and calculated coseismic displacements. Uplifted alluvial terraces, tilted Holocene deposits and subsurface faulting observed in trench, indicate that successive coseismic deformations may have been registered by both the morphology and geological units. At a larger scale, the relationships between the seismogenic fault, late Quaternary sedimentary basin and drainage system, suggest that there exist a cumulative deformation and a landscape evolution through several cycles of moderate earthquakes with M ~ 6, possibly combined with fewer large size events.
The Straits of Messina area is part of a roughly N-S oriented active normal fault belt, which extends for about 380 Km, from the northern Calabria to the Ionian coast of South Sicily (Siculo-Calabrian Rift). In the Straits of Messina area normal faults cut the pre-existing Neogene thrust belt of the Apenninic-Maghrebian Chain, and propagate southwards to the Hyblean Plateau foreland. In the Straits of Messina the Late Quaternary activity of major fault segments strictly controlled the evolution of both the sedimentary basins and the morphology of the uplifted blocks. In the Quaternary stratigraphic record, the fault activity resulted in a rapid facies change, as suggested by the development in the tectonic troughs of coarse-grained clastic fans (Ghiaie di Messina) which lie above Middle Pleistocene pelagic clay levels. The normal faults strongly controlled also the morphological evolution of the Straits of Messina. Several NNE trending fault segments control the present topography and show huge escarpment with very sharp morphology. Long-term morphological effects of normal faulting are recorded in a broad area surrounding the major border faults. The onset of the activity of the normal fault belt resulted in a sudden increase in the uplift rate of both Sicily and Southern Calabria shoulders which are characterized by the occurrence of a flight of Middle-Late Pleistocene marine terraces. Geological and morphological data combined with biostratigraphic analysis indicate that in the Straits of Messina region rifting processes occurred at least since 600 ky. The distribution of geologic and morphologic fault-related features clearly indicate the overall pattern of the major border faults on both sides of the Straits also suggesting the off-shore extent of the major fault segments. Finally the different values of uplift rates recorded on the both sides of rift zone provide useful information to better constrain the geometry of the major fault zone.
We present new evidence for the propagation processes of the North Anatolian Fault (NAF). Folding in the Dardanelles Straits region allows us to document the timing of the deformation preceding, and the finite displacement after, the passage of the propagating tip of the NAF. The accuracy of the observations is due to interplay between deformation and sea-level changes in the Mediterranean (the well-known Messinian regression followed by the Pliocene transgression). Over the long-term the kinematics around the Sea of Marmara pull-apart (total displacement of about 85 km over the past 5 m.y.) is similar to the present-day kinematics deduced from the space geodesy. Models with rigid blocks moving coherently seem appropriate to describe the kinematics of the Marmara region. Thus steady deformation localized on large faults appears to be the dominant mechanical process. However, formation of the prominent Dardanelles anticline between 7-5 Ma is well-explained by compressional strain associated to a propagating damage zone in front of the fault. At a larger scale, westward propagation of the NAF over nearly two thousand km in the past 10 m.y. appears to be associated with strain recovery, suggesting that the continental lithosphere retains long-term elasticity.
Interdisciplinary investigations of a series of basins east of Erzurum, eastern Turkey have identified a range of landforms that reflect complex, dynamic responses to past and present tectonic and climatic controls. Detailed work has focused upon the Pasinler Basin (circa 44°40'N, 46°30'E) which contains many of the features present in adjacent basins. The basins themselves reflect former, localised extensional regimes. At present, they are being overridden by thrusts from the north and south. A number of geomorphological-process domains exist:
1. Thrust blocks. Advancing mountain ranges to the north and south have thrust large blocks onto the basin margins. These are frequently back-tilted. Two groups can be identified by their impact on surface drainage. Several blocks (possibly older) deflect flow so that channels from the mountains are diverted and run parallel to the main basin axes. A second (possibly younger) group is characterised by channels which cut across the tilted blocks. The position of these channels may ante-date the initiation of local tilting.
2. Mass movement scars and deposits. Several major landslide and debris flow deposits occur in front of fault line scars. One west of Pasinler, occurs next to, and possibly partly overrides a major alluvial fan and extends more than 5 km onto the basin floor. The scale of such features, and relationship to faults suggests a seismic origin.
3. Alluvial fans. Major fans extend from the faulted guided escarpments onto the basin floor. Typically these consist of both fluvial and debris flow deposits. At present, most of the fans are deeply incised suggesting a change in both flow regime and sediment supply. The depth of incision in places and the relatively fresh appearance of the upper fan surfaces suggests this probably occurred during the Late Quaternary/Holocene but more precise dating is problematic at present. Several of the fan surfaces feature minor escarpments and breaks of slope which run approximately parallel to the main fault lines. These probably reflect ongoing tectonic activity. The fans grade into lower gradient fluvial landforms on the basin floors.
4. River terraces and floodplains. At present, many of the larger river channels exhibit low sinuosity and are braided, reflecting dominance by nival flows. Several terraces occur at various elevations above the modern fluvial plains and cover much of the basin floor areas. Many of these feature palaeochannels, including highly sinuous forms that suggest hydrological variability over time. The extent of these suggests climate is the dominant control, though tectonic activity and landuse have almost certainly had some influence. Tectonic activity has certainly modified fluvial processes locally, particularly where channels run near to thrust blocks. In places, former fluvial surfaces appear to have been lowered recently, forming seasonal wetlands.
The measurement of the impact of tectonic movements on the course of a river was clearly established by Scumm (1986), Adams (1980), Cox (1994), Marple & Talwani (1993) and Davis & Tinker (1984) and many others. In order to study to what extent the relative age of tectonic movements can be defined with GIS and remote sensing methods in a desert environment a Digital Elevation Model (DEM) of a test area (approx. 4700 km2) in the northern Central Andes and a Landsat TM scene as well as river geometry was analysed in terms of their relevance for landform evolution with emphasis on the question of impact of tectonic activities on the behaviour and form of a river. From a Landsat TM image and a digital elevation model (DEM) linear structures were extracted and geometric properties of the upper course of the Rio Loa river (a master stream in the northern Central Andes) like width (frequency) and amplitude of stream meander were computed. Analysis of linear structures reveals predominant SSW-NNE and SSE-NNW directions. These directions of fault systems are considered to be of late Eocene to Miocene age. Beyond these longitudinal directions of fault systems, a few W-E trending linear structures are visible on the TM image and the filtered DEM. These structures cross the Rio Loa river, which flows N-S in the Preandean Depression. The analysed geometry of the upper master stream course indicates that the width (radius) of the E-trending meander segments are greater than the W-trending meander segments. As the climatic conditions were arid to hyper arid during the last 5 mio years, the river could not have induced this difference in meander radius itself during an arid climate period. In order to modify its own bed, the river must have had enough hydraulic power to force its meanders to follow a natural tilt, not present today but caused by a an uplift W of the river or by a sinking block E of the river. The difference in width for the E- and W-trending meander segments was caused by an event more than 20 mio years ago, whereas single high amplitudes of the meanders and sudden change of river gradient, always associated with W-E linear structures, are laterally transposed faults due to the activity of strike slip faults. Hence the locations of deflection points are consistent with W-E faults. The W-E-trending faults are considered to be of young geological age, consistent with movements of blocks coupled with holocene earthquakes.
Schumm SA, Studies in geophysics -Active tectonics: Int. Geophys. Res. Forum, 13, 80-93, (1986).
Adams J, Geology, 8, 442-446, (1980).
Cox R, Geol. Soc. Am. Bull, 106, 571-581, (1994).
Marple RT, talwani P, Geology, 21, 651-654, (1993).
Davis TRH, Tinker CC, Geol. Soc. Am. Bull, 95, 505-512, (1984).
During Oligocene (38 Ma-25 Ma) an extensive fluviolacustrine complex was established north to the Pyrenean range. These thick (up to 1000 m) molassic deposits are mainly composed of terrigeneous material, but carbonates have also developed fringing the southern border of the Massif Central. In Albigeois and region, the Stampian outcrops allow to analyse competition between the fluvial-detritic deposition processes and the outgrowth of lakes with carbonates and marls, in relation with the local and regional tectonic evolution. The 3D morphology of the study area has been extracted from a DEM built out of the topographic map and combined with a digitised aerial photography of the same area. The DEM has been coupled with data of different origin namely the sequential analysis of the continental deposits and structural data. A reconstruction of the geometry of the Stampian lakes and the associated fluvial network has been achieved. The studied area is located on the banks of a main E-W river, the Dadou river. The morphologic pattern of this main E-W axial drainage depicts a half-grabben structure. The left bank - or roll-over side -presents a smooth eroded slope marked by a dense NW-SE oriented rivers network. The right bank scarp has escaped from the würmian peri-glacial alteration an erosion processes and provides fairly continuous outcrops. It suggests that the listric fault was still active during Würmian times (i.e. until 25,000 years B.P.). The sequential analysis of the Stampian formation points out that the carbonate-dominated lakes are bounded by NW-SE oriented detritic-dominated channels (up to 1 km wide). The latter form a succession of alluvial fans when they merged into the E-W axial drainage. It indicates that the half-grabben structure provides adequate relief for vertical fan growth and a suitable location for sediment accumulation, and that the rate of uplift of the listric fault has exceeded the rate of down cutting of the trunk stream feeding the fans. The early Oligocene tensional tectonics had controlled development of the fluvial network of the pyrenean molassic system, commanded by a main E-W system and NW-SE oriented tributaries. The 40 km-long Dadou river is a main half-grabben structure, still active recently, which regionally marks the boundary between a southern domain in which the main drainage network is NW-SE oriented (Agout, Girou Hers, Ariège, la Lèze...) and a northern domain in which the main drainage is E-W oriented (Tarn, La Vere, Le Cérou...).
Several intermontane basins of neotectonic origin, bounded by extensional fault systems along the Apennine mountain belt, display remnants of Early-Quaternary sedimentary bodies. Late-Quaternary to present tectonic activity has displaced and deformed the depositional tops of these sediments. These paleosurfaces have furthermore been eroded by the drainage system, controlled by recent tectonics and vertical mobility.
While we assume that the depositional tops have been emplaced as subhorizontal surfaces, their remnants in some neotectonic basins often yield a poor attitude to be reassembled, due to lack of areal and lateral continuity. A number of continental depositionary processes (alluvial fans, floodplains, lacustrine and marsh infill) have often generated bodies with different depositional geometries over the pre-existing topography. Additionally, tilting and displacement of the paleosurfaces also led to an apparently chaotic areal distribution of their remnants, and the resulting topographic surface bears a number of gradients which prove difficult to model.
To better ascertain and quantitatively express the superposition of deformational and erosional features displayed by these paleosurfaces, this study aims at: (a) integrate geomorphological field data of selected areas with high-resolution digital topography (DEM). This will be endeavoured via a suite of algorithms (including neural networks) to cluster gradient, slope, aspect and morphometric parameters geared towards detection of paleosurface remnants; (b) analytically reproduce and spatially interpolate the henceforth discretized landforms. This task will be tackled with algorithms based on mathematical morphology. The aim is to discriminate a series of functions in 3 or more variables studied within a system that describes the landform in a cartesian space; (c) study a procedure to analytically restore the paleosurface from its present-day remnants to its pre-deformational status.
As the Apennines are roughly divided in three main structural provinces based on their seismic behaviour, this study will initially approach three separare areas, each one belonging to one of the of above domains. Namely, the areas will be Valdarno in the Northern portion (Tuscany-Umbria), Fucino in the Central (Abruzzi) and Val d'Agri in the Southern (Basilicata-Matese). Regional observations will be conducted for each of the above au-large domains. To refine the preliminary results, the procedure will be hopefully tested in one or two areas belonging to a geodynamical setting completely different from the Apennine belt, possibly in the Basin and Range province (namely Dixie Valley and Pleasant Valley) of Western USA.
The area in Croatia is seismotectonically very active and particular its Dinaric region. The activity is caused by the movement of Adriatic microplate and the corresponding reaction of Dinarides. Position and size of different rock masses are essential for the process. The masses represent the basic geological structural units. Besides surface and Moho discontinuity data the footwall surface of sediment rock complex plays important role for observing the crustal deformations and its structural fabric. The surface is at 3 to 17 km depth. Earthquake focal data indicate to the seismotectonic active zones in space.
Relations between stress and strain are considered when treating the tectonic movements. Two observations are emphasized: 1.- southerly movements of geological structures in northern Dinarides and 2.- fracture of the Earth's crust in the zone between Mljet island and Dubrovnik; they are caused by characteristic underthrusting of the Adriatic microplate under the Dinarides. Comparisons between stress obtained by the surface measured data and the earthquake focal data are indicative.
It is now clear that earthquake produce permanent and recognizable effects on the geological landscape, which can enable geologists to infer the degree of seismic activity in a region.
In NW Macedonia (Greece) a system of NNW-SSE trending basins have been developed during Neogene-Quaternary. Within the Kozani basin synrift deposits consist of a basal conglomerate-sandstone unit followed by a fining upward sequence of lacustrine-fluvial sand-clays grading through lignite horizons into fluvial mudstone and sandstone beds (maximum thickness 550 m).
The main neotectonic feature in Kozani area is the Aliakmon fault zone that generally follows the course of Aliakmon River. This fault zone is segmented and one of the segments was triggered by the earthquake of May 13, 1995 (Ms=6.6). A non-activated segment, namely the Servia fault (N70°E/60°NW), delineates the southern boundary of the Kozani Basin and seems to have been reactivated during Holocene times. It is an impressive typical normal fault with fresh scarps and Holocene striations. Thermoluminescence (TL) dating of the basal colluvium indicated an age of 15,000 years (Late Pleistocene-Holocene).
In order to study the faults past behaviour, several well-defined geomorphic indices were calculated that aided in identifying the level of tectonic activity in the area. The results were in a good agreement with field observations and suggested recent (Holocene) tectonic activity. The selected geomorphic indices are: (i) Mountain-front sinuosity (Smf) with low values (1-1.5) that reflected the domination of active tectonic processes over erosional forces. (ii) Drainage Basin Asymmetry; here, quantification of the Asymmetry Factor (AF) detected tectonic tilting of the Aliakmon river drainage basin to the SE. (iii) Stream Length-Gradient index with high values on soft rocks indicated an active fault zone. Despite unequivocal geomorphic evidence of recent tectonic activity only one historical earthquake (1965) has been reported for the last two millennia in the Greek catalogues of seismicity can be possibly related to the Servia fault.
It is proposed that the length of the Servia fault may be constrained by the recognition of two persistent segment boundaries (southwest of Rymnio village and northeast of Servia town; total length ~12,5 km). Both areas are characterised by (i) the existence of transverse bedrock ridges and, (ii) low values of cumulative fault slip. Moreover, field mapping further to the northeast of Servia town did not show any continuous bedrock scarp, indicative of a recent, large earthquake in the immediate vicinity of the Servia fault.
Quaternary sediments and landforms in the penisulae of Messenia and eastern Lakonia (SW Aegean forearc) reflect interplay of tectonics, sea-level and climate change. The main tectonic controls were regional uplift, probably related to crustal underplating of the overriding Peloponnese forearc, and block-faulting, resulting from supra-subdaction zone extension and 'roll-back' of the S. Aegean subduction zone. From Late Pliocene to Holocene five main identifiable transgressive-regressive sequences developed in response to relative sea-level cyclicity. Transgressions gave rise to prograding shorelines/deltas and littoral carbonate deposition. Regressions resulted in erosion, fluvial incision, cliff formation, and alluvial fan progradation. Regional uplift resulted in forced regression and preservation of shallow-marine sediments as terraces. Extensional faulting was active throughout the Pleistocene. Fault directions are mainly parallel, but also oblique and transverse to the arc and pre-Neogene structures. Fault-bounded blocks are larger in the W (Messenia; 20-30 km) than in the E (eastern Lakonia; 10-20 km).
Late Pliocene - early Pleistocene marine sediments were deposited on a faulted Neogene topography, unconformably over Pliocene alluvia/colluvia. Successions generally began with offshore marls (MNN 19 biozone), although locally marine deposition started in the Pliocene. They coarsen upwards into prograding shoreline carbonates or deltaic sands. Relative sea-level fall and subsequent rise resulted in deposition of middle Pleistocene littoral sediments, locally preserved above early Pleistocene shoreline facies. The subsequent mid-Pleistocene sea-level fall resulted in fluvial incision and cliff formation in both peninsulae. Alluvial fan progradation over the previous deposits took place along the flanks of major grabens in the W (Messenia) and more regionally in the E (Lakonia). Siliceous soils developed over the highest Pleistocene terraces where block-faulting (and thus erosion) was less intense (NW Messenia). Eutyrrhenian shoreface-foreshore carbonates (grain/packstone to algal boundstone) transgressively overlie previous marine sediments (western peninsula) and/or alluvia of mid-Pleistocene age (eastern peninsula). Post-Eutyrrhenian relative sea-level fall resulted in cliff-cutting and drainage incision. This was followed by renewed transgression and deposition of Neotyrrhenian littoral sediments, at lower elevations. In the E (eastern Lakonia), late Pleistocene marine sediments are incised by streams, followed by latest Pleistocene alluvial fan progradation, controlled by faulting. In the W (Messenia), local latest Pleistocene alluvial fans are controlled by major faults. Aeolian deposition resulted from reworking of exposed littoral carbonates during the later stages of the late Pleistocene sea-level lowstand and the early stages of subsequent sea-level rise. Late Pleistocene and Holocene faulting is mainly related to reactivation of older, NNW-SSE and NNE-SSE faults, in a generally E-W extensional stress-field. Vermetid trottoires were deposited during the Versillian sea-level high. Localised tectonic uplift and subsidence continued in both peninsulae during Holocene time.
We analyzed the subsurface geology of an active fold using methods of the oil-industry geology: correlations of subsurface electrical logs, contouring of structural surfaces, plots of dipmeter and core dips, interpretation of seismic lines, and analyses of growth strata. The analysis revealed the geometry of the blind thrust that drives the growth of the anticline, the timing of its inception and preliminary slip rates.
The Los Angeles basin is a polyphase basin formed in Neogene time along the transform margin between the North American and Pacific plates. The basin, located at the boundary between the California Continental Borderland, the Peninsular Ranges and the Transverse Ranges, is one of the world's most prolific oil and gas producing areas.
An early phase of extension with associated strike-slip deformation and block rotation took place between Oligocene and middle Miocene, followed by the principal phase of basin opening during late Miocene-early Pliocene. Extension turned into contraction in early Pliocene, and present-day deformation is controlled both by strike-slip and reverse faults, the rates of the former generally being higher than that of the latter.
The main tectonic feature of the eastern Los Angeles basin is the NW-trending, right-lateral Whittier fault that is a segment of the Elsinore-Whittier-East Montebello fault system. In an initial stage the Whittier fault acted as dip-slip extensional, first with the north side down (late Miocene), then with the south side down (Pliocene). Conversely the late Pleistocene deformation is mainly right-lateral strike-slip. Paleoseismological and geomorphological studies show a northwestward decrease in the Pleistocene slip rate from the Elsinore to the Whittier faults (from 5±2 mm/yr to 2-3 mm/yr).
The West Coyote anticline is part of a set of E-trending en-echelon active anticlines controlled by south verging blind thrusts (Norwalk fault system). They are located to the south of the Whittier fault and could be the expression of the decrease in slip rate along the Elsinore-Whittier faults. The West Coyote and East Coyote-Hualde anticlines have a clear morphological expression, while the surrounding Yorba Linda and Santa Fe Springs anticlines lack in topography. Main streams draining southward the Puente Hills show a right lateral off-set as they cross the Whittier fault and are antecedent with respect to the Coyote Hills, suggesting that the uplift of the folds and the slip on the Whittier fault proceeded at the same time.
Thinning southward of the Miocene Puente Fm. registers the early phase of normal dislocation of the Whittier fault, and the late Pliocene-Pleistocene growth strata of the Pico Fm. mark the inception of the anticline.
Where climatic conditions are highly erosional (wet tropical) or seismic activity is low (western Europe), the geomorphic signature of tectonic activity tends to be strongly subdued. Consequently, seismic scarps are badly preserved and more generally active fault traces are difficult to localize, specially along strike-slip faults. The examples of the Great Sumatran and Sulawesian Palu Koro faults in Indonesia illustrate that strike-slip faults, under wet tropical conditions, with slip rates over 5-10 mm/yr are easily seen on satellite imagery (LANDSAT or SPOT). In contrast on the ground, scarps are very poorly preserved or missing and in general, all short-wavelength topographic features are dubious and thus no clearly diagnostic of tectonic activity. However, systematic lateral offsets may be well identified and then used to constrain slip rates. Offset K/Ar dated volcanism in Sumatra or 10Be dated alluvial fans in Sulawesi have allowed to estimates long-term slip rates between 5 and 25 mm/yr in Sumatra and about 30 mm/yr in Sulawesi. Those estimates are similar to GPS far-field displacements but much higher than GPS near-field displacements suggesting that those Indonesian strike-slip faults are mainly mechanically locked between major earthquakes. Paleoseismological investigations have been conducted in few favorable sites, they appear more limited than in dry regions (e.g., California) due to the fast degradation of organic materials and to shallow water tables. Only one single event was detected in southernmost Sumatra which has occurred during the last 1-2 century BP. In contrast, a minimum of three seismic events were observed in Sulawesi during the last 2000 years. Those contrasted results should result from higher slip rates in Sulawesi. Low to moderate seismic activity regions, as western Europe are comparatively worse to study. Indeed, active fault traces are very unaccurate to localize on SPOT images which exhibits mainly anthropic features. In such regions, DEM, used alone or with other data, appears to be a more reliable tool to analyze the evidences of seismic activity and even to quantify long-term displacement. In south-east France, such DEM analysis performed along the Moyenne Durance has suggested a left-lateral slip rate between 0,17 and 0,6 mm/yr. Those estimates are in agreement with paleoseismological observations conducted within a single trench which have suggested the possibility of a M=7 seismic event between 26 and 9 ka. However, DEM analyses do not permit to localize active fault traces with the necessary accuracy for paleoseismological trench sites. Presently, we are exploring two complementary approaches GPR sounding to localize shallow subsurface fault traces and inversion of topographical signal to determine active fault parameters.
Yearly high precision levellings carried out since 1993 along a 48-km-long local network in NE Ardenne showed probable vertical fault movements of up to several mm/year, as well as significant block tilting. In order to precise the temporal evolution of these displacements and to distinguish between tectonic and near-surface causes of movement, a 0.924-km-long section has been weekly levelled from April 21, 1997 to January 12, 1998. This section extends radially at about 1 km from the Gileppe reservoir (the capacity of which is of 25 Mio m3) and the difference in elevation between its endmarks is of 10.74 m.The recorded vertical movements show a maximum amplitude of 3.45 mm. Weekly values always remain under 1 mm. Though in some cases smaller than the standard error of the levelling, these values combine within statistically significant monthly trends which are remarkably correlated to the loading variations in the nearby reservoir. A 2D finite element modelling of the reservoir-induced vertical ground displacement along a NNW-striking profile containing the levelled section fairly well fits the measured movement curve and clearly confirms that the varying water-level in the lake is the controlling factor of the movement of this particular section. The computed subsidence at the centre of the lake is of about 2 cm for a level change amounting to 12 m, but the reservoir-induced movement rapidly damps down when going away from the lake, and it becomes negligible at distances higher than 2.5 km. Superimposed on this ground response to the near-surface load variations, a relative movement of 1 mm has been recorded between the ends of the section within a 5-week-period in June 1997. However it was completely recovered during the following 5 weeks (July 1997). The reversible sense of movement makes it very similar to yearly or longer-term movements frequently inferred from the comparison of high precision levellings in intraplate settings. Most of these displacements are located on active geological structures and are thus interpreted in terms of tectonic movements even though their nature still remains ill-defined. Here we show that this kind of movement occurs in very short periods of time - typically a few weeks - separated by periods of inactivity, probably in response to adjustments of shallow crustal blocks under the action of deeper-seated lithospheric stresses.
The evolution of half grabens by normal and reverse faulting in a 10 km thick upper crust with a Byerlee-type strength envelope has been modelled by elastoplastic finite element analysis. The substratum is assumed to be inviscid and the graben is filled by sediments 400 kg/m3 lower in density than the crust. During the earlier stages of incremental deformation, the surface flexure is similar for both normal and reverse faulting. However, as the layer is stretched or shortened by progressive increments, the zones of plastic failure extend until throughgoing failure of the layer eventually occurs. When this stage is reached, subsidence of the graben slows down and stops. Throughgoing failure occurs later for compressional reverse faulting than for extensional normal faulting, because of the higher compressional strength. Thus a trough 50 km wide and 3.2 km deep eventually forms for normal faulting, but a narrower trough 25 km wide and 5 km deep forms for reverse faulting. For both models, the hinge line of the half graben progressively migrates towards the fault as plastic failure reduces the flexural rigidity. For the compressional reverse faulted model only, buckling starts to occur on both sides of the fault producing anticline-syncline pairs of 80 to 90 km wavelength.
Bott MHP, J. Geophys. Res, 102, 24605-24617, (1997).
In Central Iberian Peninsula, main topographic features are a NE-SW trending Alpine chain (Spanish Central System) and two Tertiary Basins: Duero and Tajo Basins. The Spanish Central System consists of Hercynian basement block-mountain, where a thin Mesozoic-Tertiary cover is preserved mainly in the eastern part. Its average elevation is 1,100 m, with heights up to 2,500 m. This chain is the divide between the Duero and Tajo Basins, whose mean elevations are 800 and 600 m respectively. These structural units result from Alpine compressive events. The studied area has been subjected to a stress field from the Middle Miocene until now with a largest horizontal shortening direction located between N 130º E and N 160º E (De Vicente et al., 1996). This compressive regime forms NE-SW trending reverse faults and folds.
In order to study the relationship between tectonic structures and present-day topography, an analysis of topographic features has been carried out. In this way, a Digital Elevation Model (DEM) has been produced in the Geodynamics Department of Complutense University digitising topographic maps (Scale 1:50000) (Sánchez Serrano et al., 1996 and 1998). It was integrated in a raster-type GIS (Idrisi for Windows 2.0, Eastman, 1997). The final DEM has a spatial resolution of 250 m and a size of 1,242 rows by 2,081 columns. Spectral analysis of this DEM applying bidimensional Fourier analysis provides a useful tool to isolate different surfaces characterised by their own wavelength that can be compared with structures wavelength, having in mind that harmonic order depends on map size. In this case, some of the surfaces, for example the 20th order harmonic corresponding to a 21 km wavelength, and the tectonic structures wavelength show a good correlation. This harmonic surface shows highs and lows spreading NE-SW, parallel to the main orientation of the Spanish Central System. These features are related to fault limited blocks and folds, many of them has been described in previous works. Basement is involved in these structures. In the basins, same features are observed and in this area they imply Neogene sediments. N-S trending features bound several domains characterised by the highs and lows previously described. These represent N-S faults moving strike-slip normal, under a maximum compressive axis N 160º E acting from Middle Miocene.
The results show the present-day strong influence of tectonic structures in the topography and in the fluvial system. Taking into account the effect of differential erosion due to lithology, we can confirm that most valleys are controlled by faults and folds whose directions are coincident with the main tectonic structures orientations.
De Vicente, G., Giner, J.L., Muñoz Martín, A., González Casado, J.M. and Lindo, R., Tectonophysics, 266 (1-4), 405-442, (1996).
Eastman, JR, Idrisi for Windows User´s Guide. Version 2. 0. Clark University, (1997).
Sánchez Serrano, F, Gómez Ortiz, D, Bergamín de la Viña, JF and Tejero López, R, Geogaceta, 19, 23-26, (1996).
Sánchez Serrano, F, Tejero López, R and Bergamín de la Viña, JF, Rev. Soc. Geol. España, 11 (1-2), 139-149, (1998).
A number of significant informations about the evidences and characteristics of the cenozoic tectonics and neotectonic activity in the Brazilian Plateform have been gathered during 10 years, in different parts of Brazil. These have permiting the identification and study of main structural features due to reactivation of old fault and shear zones (Precambrian geossutures or Mesozoic faults). Resulting from reactivations, the new crustal discontinuities have, in some cases, controled the formation of continental rifts and individual sedimentary basins, bordered by uplifted blocks with remnants of erosion surfaces. Some of this basins were submitted to a younger compression (Mio-Pliocene), resulting in tectonic and landform inversion. In all partes of the Brazilian Plateform, this main discontinuities control the macrorelief and the regional drainage pattern, as well as the captures, orientation and limites of the main hydrographic basins. The mega-blocks limited by this discontinuities are fragmented by faults of different types and sizes, resulting in regional structures they alternate subsident and uplifted compartments. The bigest discontinuity, named "Crustal Discontinuity of the Two Brazils" have controled a fundamental differenciation on the tectonic and morphogenetic evolutions of two parts of Brazilian Plateform, since the Upper Proterozoic: on the NW the Amazon Basin region as a cratonic area with isotropic behaviour, and on the SE a region characterized by a mosaic of tectonic inheritances they make easy the Phanerozoic reactivations and, consequently, the highest differenciation on the morphotectonic patterns. Regional analysis show that:The western extremity of the Amazon Basin is partialy disrupted by blind inverse faults related to Andean thrust-faults oriented N-S. In this region, the neotectonic activity have influencied the drainage network development and sedimentation. The central and eastern parts of the Amazon Basin are cut by Riedel shears due to a E-W dextral shearing, with strong control on the hydrographic system and fluvial dynamics. In the lower basin we can identify some triple-junction features, formed in the confluences of big rivers. The oriental coastal province displays different patterns from south to north, due to differences in the neotectonic evolution. The southern and southeastern regions are dominated by Cenozoic uplift with tilting of the continental blocks to W and NW and downwarping of the off-shore blocks. In the northeastern region, morphological updoming is related to transpressive conditions controled by E-W dextral shears. In all the coastline, the design of shore cliffs, cutted in Upper-Tertiary sediments, show strong correlation with transcurrent faults. The geological and seismological data indicate a E-W to NW-SE oriented compressive stress, at least in the N, NE and Center-SE areas. In some cases gravimetric data confirm the morphotectonic interpretation.
Almeida FFM, Anais da Academia Brasileira de Ciências, 48, 15-26, (1976).
Assumpção M, Suarez G & Veloso, JA, Tectonophysics, 113, 283-293, (1985).
Martin L, Flexor JM, Bittencourt ACSP & Dominguez JML, Neotectonics, 1, 87-103, (1986).
Potter PE, Geology, 86, 13-33, (1978).
Riccomini C, Pelaggia AUG, Saloni JCL, Kohnke MW & Figueira RM, J. of South American Earth Sciences, 2(2), 191-192, (1989).
Saadi A, Geonomos (Brazil), 1(1), 1-15, (1993).
Scandinavia displays a variety of morphological and tectonic features. Some of these features are clearly related to passive margin development, however there are also a number of morphological domes present in Norway, Sweden and southern Finland, that may or may not be related.
The main objective of this study is to better understand the lithosphere thermal mechanics and the processes governing the formation of the passive margin and dome features. This will be achieved through quantification and dating of the vertical movements, and the integration of this new data with existing geophysical datasets into a three dimensional model. Consequently it will be possible to investigate interrelationships between observed surface expressions and thermal mechanics at depth.
A previous investigation (Rohrman, 1995), applying fission track analysis on southern Norway, addressed similar objectives in a much smaller study area. A clear correlation between onshore Triassic/Jurassic denudation (2.4 ± 1.1 km) and offshore sediment deposition was identified. A relationship of this to rift flank uplift was postulated. Rapid Neogene tectonic uplift in the order of 1 - 1.5 km was thought to have produced the observed domal features. It was concluded that the mechanisms controlling the uplift were too large in scale to be fully understood by studying the southern Norway area alone. The data on vertical movements are obtained by applying fission track analysis to basement rocks from the Scandinavian mainland as well as sediments from offshore basins. The unique low temperature sensitivity of apatite fission track thermochronology makes it possible to quantify uplift, denudation and exhumation much more accurately than has previously been achieved. By taking into account earlier studies in areas bordering the North Atlantic and extending the scope of the new model into a more regional setting we will be able to test existing ideas on the role of both thermomechanical and surface processes.
Rohrman M, Thermal evolution of the Fennoscandian region from fission track thermochronology, Ph.D. thesis, Free University, (1995).
After deglaciation Vendsyssel, northernmost Denmark was an archipelago-like part of the Skagerrak-Kattegat embayment. Deglaciation and the lateglacial sea-level history are studied by the means of sedimentological investigations in a sequence stratigraphic context and by analyses and modellings of raised shorelines.
Analysis of the pattern of highest lateglacial shorelines indicate deglaciation during forced regression (falling relative sea-level) with an initial course of deglaciation from the south towards the north followed by a course of deglaciation from the west towards the east. In large areas the deglaciation involved ice stagnation and melting of dead ice (Fredericia, 1988). Analyses of the total number of shorelines provide information about details in the sea-level history during and after deglaciation. Shoreline emergence models (Richardt, 1998) are constructed to predict the time of formation of the individual morphological shoreline features and to link widespread localities in time and space. The emergence models allow the construction of shoreline spectra depicting number of shorelines against time or against modelling steps. In Vendsyssel both simple and more complex emergence models produce shoreline spectra with four distinct tops in the time interval 15-11.5 ka BP (conventional radiocarbon years).
Sedimentological investigations in western Vendsyssel show that the deglaciation took place in a glaciomarine environment rapidly followed by a "normal" marine shelf environment with sedimentation mainly controlled by currents on the shelf. Three-dimensional facies analysis indicate an early relatively stable sea-level episode of a few hundred years duration followed by rapid forced regression. Two other sedimentologically inferred stable sea-level events are reported by Nielsen et al. (1988) and by Richardt (1996) respectively. These events are clearly separated in time and elevation both from each other and from the western Vendsyssel stable sea-level event.
The three sedimentologically inferred stable sea-level events coincide in time and elevation with the shorelines represented by three of the tops in the shoreline spectra. It is suggested that the remaining top in the shoreline spectra also represents a stable sea-level event and that consequently the lateglacial sea-level history of Vendsyssel was a history of highly episodic forced regression with four stable sea-level episodes in the time interval 15-11.5 ka BP. The youngest stable sea-level episode, c. 12.5-12 ka BP can be explained by a rapid eustatic rise and it correlates with stable sea-level at the eastern margin of the Kattegat reported by Berglund (1995). The origin of the other stable sea-level episodes is more uncertain and these episodes are possibly local phenomena restricted to the Vendsyssel area.
Berglund M, Boreas, 24, 324-344, (1995).
Fredericia J, Danm. Geol. Unders. Int. Rapp, 22, 231p, (1988).
Nielsen LH, Johannesen PN & Surlyk F, Sedimentology, 35, 915-937, (1988).
Richardt N, Geol. Soc. Spec. Publ, 111, 261-273, (1996).
Richardt N, 23rd Nordic Geol. Winter Meeting Abstr. Vol, 254, (1998).
This poster presents the topic of late Pleistocene existence of trees in the central European countries Slovenia and Croatia.
During the late Pleistocene the unglaciated areas of Europe were dominated by treeless arctic tundra, which attained their largest extent during the Last Glacial Maximum (van der Hammen et al., 1971). Trees are thought to have existed only as small isolated refugial populations in the Balkans, Italy and the Iberian peninsula, where climatic conditions were more favourable (Bennett et al., 1991). With climatic warming in the early Holocene refugial tree populations expanded and colonised previously inhospitable terrain in northern Europe and Scandinavia (Huntley and Birks, 1983).
Here, we present new evidence for the widespread late Pleistocene existence of trees in the central European countries Slovenia and northern Croatia. Five radiocarbon-dated terrestrial sedimentary sequences situated on a Northeast-Southwest transect through these countries were analysed for fossil pollen. Results from the late-Pleistocene sections of the sequences suggest that a diversity of temperate deciduous and boreal evergreen trees existed in the study area during this time. However, population density of these trees was probably low.
The fossil pollen records also reveal distinctive regional variation in late Pleistocene vegetation. The north-eastern part of the study area supported boreal needle-leaved trees, while a diversity of deciduous broad-leaved trees was prominent in the Southwest. Cold-tolerant conifers were dominant where modelling studies of late Pleistocene climate suggest the harshest temperature regime of the study area (Kutzbach et al., 1993). Conversely, cold-intolerant deciduous trees dominated in the Southwest where modelling suggests mild winters.
Results from the early Holocene sections of the pollen sequences show a rapid expansion of all thermophilous tree populations after the onset of the Holocene, indicating fast establishment of forest. We suggest that tree populations colonising northern Europe and Scandinavia probably originated from these early-Holocene forests of central Europe. This new proposition contrasts earlier theories which suggest origination of colonising populations further south in the Balkans. The new theory implies that distance from refugial source to northern-European limits of the population achieved by trees until the mid-Holocene climatic optimum was shorter than previously thought (Bennett et al., 1991).
Van der Hammen T, Wijmstra TA & Zagwijn WH, The Late Cenozoic Glacial Ages (KK Turekian), 391-424, (1971).
Bennett KD, Tzedakis PC & Willis KJ, J. Biogeography, 18, 103-115, (1991).
Huntley B, Birks HJB, An Atlas of Past and Present Pollen Maps for Europe 0-13,000 Years Ago, (1983).
Kutzbach JE, Guetter PJ, Behling PJ & Selin R, Global Climates since the Last Glacial Maximum, 135-178, (1993).
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications