Buried River Channels Research Articles

The Geoclutter experiment: Remotely imaging sub-bottom geomorphology in shallow water

Preliminary results of the Geoclutter experiment of April–May 2001 are presented. This bistatic reverberation experiment was conducted at the New Jersey Strataform site in continental shelf waters. Roughly 3000 waveforms were transmitted from vertical source arrays in the 390–440 Hz band and received by a horizontally towed array. On average roughly 10–100 discrete clutter events were registered per transmission for a total of at least 30<th>000 geoclutter events that could be confused with a discrete target over the period of the experiment. Bathymetry throughout the New Jersey Strataform area is extremely benign with few discrete features. The vast majority of clutter returns appeared in level areas where absolutely no bathymetric features are present. The clutter events often registered well with buried river channel networks previously characterized in the Geoclutter program. Extensive probing of river-channel-infested areas revealed that the number and level of clutter events is highly dependent on bistatic orientation. The registrations suggest that bistatic aspects that produce ‘‘glints’’ from large projected areas along river channel axes yield the strongest and most frequent targetlike returns. These results are consistent with those predicted by a unified model for reverberation and submerged-target scattering [J. Acoust. Soc. Am. 109, 909–941 (2001)].

The significance of buried landscape in subsurface migration of dense non-aqueous phase liquids; the case of perchloroethylene in the Sussex Aquifer, New Brunswick

Abstract At Sussex, New Brunswick, dissolved perchloroethylene (PCE) as high as 28.0 mu g/liter, was detected within the upper aquifer and in concentrations to 1.6 mu g/liter within the lower aquifer that serves as the main source of municipal water supply. Predominantly formed as an esker ridge, the lower aquifer consists of well-sorted, glaciofluvial/ice-contact, sand and gravel layers mostly confined by underlying till and/or bedrock and an overlying unit of laminated silt-clay glaciolacustrine sediments (middle aquitard). The middle aquitard is in turn overlain by an upper aquifer of variable thickness, extending from a depth of about 18 m to near surface, consisting of coarsening-upward sand and gravel, with occasional lenses of clay, silt and diamicton. A dispersal plume was identified by contouring PCE concentrations from 72 samples from the upper aquifer; PCE was not detected in 16 surface soil samples. The highest concentrations imply that the PCE originated from a surface spill near the urban business center and that subsurface migration was influenced significantly by a buried relict river channel cut into the middle aquitard. Scouring along the channel bottom cut through the middle aquitard and intersected the esker ridge of the underlying water supply aquifer, forming windows through the aquitard. Contamination of the lower aquifer has likely originated from the PCE being drawn through these windows during pumping of a water-supply well that is now out of service. Results indicate that withdrawal rates and placement of original municipal water wells, as well as the stratigraphy, conductivity and topography of subsurface units are controlling factors in Dense Non-Aqueous Phase Liquid migration at this site.

Imaging a buried river channel in an intertidal area of South Wales using high-resolution seismic techniques : Brabham, P J; McDonald, R J Q J Engng GeolV25, N3, 1992, P227–238

A buried river channel in an intertidal area at Barry Old Harbour, South Wales, UK, is investigated using high-resolution seismic refraction and reflection techniques. The refraction technique is able to profile the base of the 20 m deep buried valley because the Carboniferous Limestone bedrock has a significant velocity contrast with overlying water- saturated unconsolidated sediments. Common midpoint (CMP) reflection profiling, using a single detonator as the seismic source, images the base of the river channel and the internal sedimentary sequences. Seismic reflections with frequencies in the 400 to 600 Hz range are observed from as shallow as 5 m depth (6 ms two-way time). The bedrock profiles derived from the refraction and reflection experiments are in precise agreement. (A)

The Efficiency of Vertical Drains Piercing Flat-Lying Beds at Piedra Del Aguila Dam

This paper analyzes the problem of determining the pattern of flow to expect through a sequence of flat-lying beds of varying permeability which are pierced by a series of equally spaced vertical drains. A single permeable bed is assumed to be bounded by two impermeable beds from above and below and the influence of a reduction in screen efficiency is evaluated. This work arose from design work associated with the Piedra del Aguila hydroelectric project in Argentina in which the left abutment of the dam consisted of an ancient buried river channel filled with a sequence of sand and gravel sediments of varying clay content mixed with layers of volcanic tuff. The outcome of the study is that a practical limit of intercepting 80% of the flow is feasible and that this limit depends strongly upon the efficiency of the drains. All suspected aquifer beds must be drained and drain efficiency should be monitored over time.

Imaging a buried river channel in an intertidal area of South Wales using high-resolution seismic techniques

Abstract A buried river channel in an intertidal area at Barry Old Harbour, South Wales, UK, is investigated using high-resolution seismic refraction and reflection techniques. The refraction technique is able to profile the base of the 20m deep buried valley because the Carboniferous Limestone bedrock has a significant velocity contrast with overlying water-saturated unconsolidated sediments. Common midpoint (CMP) reflection profiling, using a single detonator as the seismic source, images the base of the river channel and the internal sedimentary sequences. Seismic reflections with frequencies in the 400 to 600 Hz range are observed from as shallow as 5 m depth (6 ms two-way time). The bedrock profiles derived from the refraction and reflection experiments are in precise agreement.

Electromagnetic Resistivity Profiling For Locating Buried River Channels: A Case Study In Haiti

The Service National des Ressources en Eau (SNRE) of the Ministry of Agriculture of Haiti, in cooperation with the UNDTCD, carried out a geophysical investigation in the Northern Plain of Haiti. It included a reinterpretation of the available resistivity soundings and the realization of eight additional soundings and 11 electromagnetic profiles.

The use of anomaly offset in solving hidden layer refraction problems

Intermediate layers are frequently hidden in seismic refraction surveying, in the sense that their presence cannot be directly detected from a travel time graph of first arrivals. Consequently, large errors in seismic interpretation may occur. Not only are the layers missed, but the calculated depths to deeper refractors may be overestimated (velocity inversion problem) or underestimated (masked layer problem). A rigorous solution to the problem of hidden layers is presented here based on anomaly offset. The procedure preserves consistency between observed and calculated offsets, thus yielding corrected depths and limits on hidden layer velocity values. The method is illustrated by means of a field example involving a buried river channel.

Tritium concentrations of groundwater in Naka Terrace, Ibaraki Prefecture, Japan

Levels of tritium in groundwater of the inland area of Naka Terrace were measured during 1981–1983 to establish a baseline for evaluation of any changes which may result from release of tritium from nuclear facilities in the coastal region of the terrace. Most of the samples yielded concentrations in the range 40–90 pCi litre −1, which were similar to those of two rivers bordering the area of concern, whereas tritium levels in monthly precipitation samples were in the range 30–50 pCi litre −1. These results imply that the rivers are fed from groundwater reservoirs, which are releasing the higher tritium levels recharged during the previous 20 years when bomb-produced tritium caused higher levels in precipitation. Only two groundwater samples yielded concentrations below detection limit, both from the deep Pleistocene gravel layer in a buried river channel underlying the terrace. Some ground-waters in the coastal area yielding more than 100 pCi litre −1 are considered to be already influenced by tritium released from nuclear facilities.

Geophysical investigations for groundwater in Precambrian terrains: a case history from Ikare, southwestern Nigeria

Magnetic, very low frequency electromagnetic (VLF) and electrical resistivity methods have been applied in mapping buried bedrock relief for groundwater exploration in the Precambrian terrains of Ikare, southwestern Nigeria. Quantitative interpretation of the magnetic data involved initial modelling and Euler deconvolution using elementary magnetic sources to identify applicable geological models in the spatial domain. A depth estimate of the top of the buried valley of 23 m was obtained. The application of power spectral deconvolution yielded an estimated depth of up to 67 m on the bedrock surface. The VLF data indicated the existence of subsurface lineament in the area while obtained resistivity of the sedimentary fill amounted to 260 ohm-m. The results of the three techniques have successfully delineated a buried river channel for groundwater extraction in a crystalline basement terrain.

Geomorphology, shallow structure, and geological hazards in the East China Sea

The inner shelf of the East China Sea is covered with a lens of fine-grained modern sediment derived from the adjacent Changjiang; modern sediments north of the Changjiang came from the Huanghe. Rapid accumulation results in a featureless seafloor, except where escaping biogenic gases cause vents. The modern mud thins towards the mid-shelf and sediment becomes increasingly sandy and locally rippled, indicative of its relict age and present-day mixing and transport. The outer shelf, particularly south of Cheju Island, is covered with fine-grained sediment that apparently reflects modern Huanghe-derived sedimentation. Shallow seismic profiles show that the Holocene sediment overlies an eroded late Pleistocene surface, dominated by a number of well-defined buried river channels formed (and filled) during the last low stand of sea level; some channels may be associated with recent faulting. These filled channels and depressions probably represent the single most important geological hazard for future emplacement of man-made structures (such as drilling platforms) on the seabed. Other potential hazards include methane gas vents on the inner shelf, possible vents of deeper gas on the mid-shelf, outcrops or near-surface expressions of the Fukien-Reinan Massif, and possibly migrating sand waves.

Great Falls Lineament, Idaho and Montana: ABSTRACT

The name Falls is given to a northeast-trending zone of diverse geologic features that can be traced northeastward from the Idaho batholith in the Cordilleran miogeocline of the United States, across thrust belt structures and basement rocks of west-central and southwestern Montana, through the cratonic rocks of central Montana, and into southwesternmost Saskatchewan, Canada. The zone is well represented in east-central Idaho and west-central Montana where geologic mapping has outlined northeast-trending, high-angle faults and shear zones that: (1) extend more than 150 km (93 mi) from near Salmon, Idaho, northeastward toward Anaconda, Montana; (2) define a nearly continuous zone of faulting that shows recurrent movement from middle Proterozoic to Holocen time; (3) controlled the intrusion and orientation of some Late Cretaceous to early Tertiary batholithic rocks and early Tertiary dike swarms; and (4) controlled the uplift and orientation of the Anaconda-Pintlar Range. Recurrent movement along these faults and their strong structural control over igneous intrusions in this region suggest that northeast-trending faults represent a fundamental tectonic feature of the region. Figure Geologic features that are similar to those mapped in the Salmon-Anaconda region are present to the southwest and the northeast. In central Idaho, these structures include numerous northeast-trending faults and pronounced topographic lineaments that cut across the southern part of the Idaho batholith, and a northeast alignment of Tertiary igneous rocks that cut the Idaho batholith and adjacent rocks. East and southeast of the Anaconda-Pintlar Range, subparallel, high-angle faults and topographic lineaments are present in the Highland, Pioneer, Ruby, and Tobacco Root Mountains. High-angle faults may have in part controlled the orientation of the northeast-elongate Boulder batholith. Northeast-trending structures are not easily traced across the thrust belt of western Montana or across he Lewis and Clark line. In the central Montana plains, northeast of the disturbed belt, however, a broad zone of colinear, northeast-trending structures is present, and includes: parallel, buried basement highs that in part controlled depositional patterns of some Paleozoic and Mesozoic sedimentary rocks; major physiographic features, such as the remarkably straight, 175-km (109-mi) long segment of the Missouri River, and equally long, buried river channels in southwestern Saskatchewan; a northeasterly alignment of highly differentiated igneous rocks and a belt of ultrabasic intrusions and related diatremes; End_Page 1350------------------------------ and a well-defined pattern of northeast-trending gravity and aeromagnetic anomalies underlying this part of central Montana and southwesternmost Saskatchewan. Taken together, all these geologic features define a broad, northeast-trending zone at least 150 to 200 km (93 to 125 mi) wide and more than 1,000 km (620 mi) long. The zone is approximately colinear but not demonstrably continuous with the well-exposed boundary in eastern Saskatchewan and Manitoba between the Archean Superior and the Proterozoic Churchill provinces of the Canadian Shield. This boundary is also characterized by: high-angle faults, shear zones, and topographic lineaments; pronounced linear gravity and magnetic anomalies; igneous intrusions; and fault controlled depositional patterns and mineralization. That the Great Falls lineament is controlled by a similar Precambrian boundary between the Archean Wyoming province of southwestern Montana and early Proterozoic terrane to the north is speculative; however, the geologic features found along the Great Falls lineament share many common characteristics with features present along the Archean-Proterozoic boundary in Canada. End_of_Article - Last_Page 1351------------

Bottom Morphology and Shallow Subbottom Sedimentary Structures on U.S. Continental Shelf Between Cape Hatteras, North Carolina, and Jacksonville, Florida: ABSTRACT

During the period July 1978 to 1980, a survey of the U.S. South Atlantic shelf was made to determine the occurrence and distribution of biologically productive bottom areas and bottom and shallow subbottom geologic features of potential hazard to petroleum exploration and production. Methods of data acquisition included the use of high-resolution seismic systems, sidescan sonar, towed television, and minisub. Shelf sediments are largely composed of 2 to 5 m of Quaternary sands and silty clays with mixtures of shell that were deposited in fluvial, paralic, and shallow-marine environments. Greater sediment thicknesses (up to 40 m) are present in buried river channels, tidal inlets, and cut-and-fill structures associated with meandering tidal streams. Surficial sediments exhibit a wide range of bed-form type and maturity. Ripples of less than 0.5 m wavelength and megaripples of 0.5 to 1 m wavelength are common across the shelf. Crest orientation is generally north-south. Algal growth along the crests of many of the smaller bed forms as well as abundant bioturbation and tracks indicate bottom stability except during storms. Sand waves up to 100 m long are common, particularly within the 25-m isobath and are present in discrete fields 3 km or more wide with north-south crest orientation. They are less common, less sharply defined offshore and usually underlie smaller bed forms of different orientation. These features may have resulted from infrequent major storms and, because of rising sea level, may now be below the effect of most s orm-generated waves. Although the mobility of these large bed forms is inferred by the presence of buried reefs within some of the sand bodies, the time and rate of movement are unknown. Many areas of the shelf exhibit irregular, low-relief bed forms that appear to be in advance stages of deterioration from intense bioturbation. Several distinct, acoustcally hard reflectors present in the shallow subbottom over much of the shelf represent indurated strata which on outcrop provide substrate for the development of reef and hardground communities. These temperate bioherms are primarily populated by sponges and octocorals and, on the basis of topographic expression, are classified as low-relief hardgrounds (> 0.5 m), moderate-relief reefs (up to 2 m), and high-relief reefs (up to 15 m). Low-relief hardgrounds are the most widely distributed of the three types but are the least productive in terms of biomass. Moderate- and high-relief reefs occur most commonly off North and South Carolina and along most of the shelf break. Substrate ranges from Pleistocene to Miocene in age and locally contains significant perc ntages of phosphate grains. The total area of shelf occupied by reefs and hardgrounds is estimated to be less than 10%. Outcrop patterns reflect both local bottom processes and regional structural framework. In the middle to outer shelf off the Georgia coast, a relatively narrow zone of large channels and numerous cut-and-fill structures occur in Quaternary deposits just landward of a shallow buried Miocene high. These features are very similar to sedimentary structures associated with present-day barrier islands, tidal inlets, and estuaries and strongly suggest a sea-level stillstand near the end of the last Pleistocene regression or early in the present transgression. End_of_Article - Last_Page 1663------------

Review of Surficial Geology and Engineering Hazards in the Canadian Offshore

Surficial geological mapping and hydrographic charting is most complete in the Atlantic Provinces offshore including the Gulf of St. Lawrence; secondarily, the British Columbia offshore and the Labrador Shelf and Slope are fairly well mapped and charted. However, the entire Arctic offshore has the poorest geological and hydrographic coverage of all Canadian marine waters, with the exceptions of Hudson Bay, Beaufort Sea-Amundsen Gulf area and Lancaster Sound.
 
 For safe marine operations, environmental information is required on winds, storms, waves, currents, tides and sea ice. This background is complemented by a seabed study designed to obtain grab and dredge samples of sediment or rock; to obtain sediment cores with the vibracorer, and gravity and piston corers; to obtain bedrock cores with the use of the diamond drill; to obtain bathymetric maps by means of sonic surveys in which echograms reveal the topographic profile of the sea-floor, and the side scan sonar give the area! view; to obtain details of bedrock features such as bedding, folds, faults, as well as distinguishing erosional and constructional topographic features all for the purposes of geological mapping; and to carry out visual surveys remotely or directly by means of submersible equipment and vehicles. All these background data and survey results can have direct application to seabed engineering problems.
 
 Potential hazards related to surficial geology in the Canadian offshore affect engineering operations such as drilling, dredging, dumping coastal works, foundations and anchorages, cable lays, and pipelines. These hazards are grouped into five main categories: (I) erosion and sedimentation, (2) Morphology, (S) movement, (4) ground ice, and (5) sediments. A breakdown is given as follows: ice-scour grooves and ridges, current scour and fill, buried river channels, sand waves, megaflutes, sand bars, valleys, fjords, canyons, troughs, thermokarst, pingos, mud diapirs, poakmarks, scarps, terraces, hummocky terrain, slump features, bedding dislocations, faulting, earthquakes, ice layers, ice-bonded sediments, thermal degradation, unstable substrate, hydrocarbons, organic matter, clathrates, hard (sand, boulders, bedrock) substrate, and moraine.
 
 The geologist must provide information and advice on the natural framework, and processes operating in the area; the engineer must appreciate these facts. But the engineer may be unable to re-locate his operational site; therefore, the geologist must present the information in its most complete and truest perspective in order to ensure a secure and successful operation.

Geophysical exploration for buried river channels in the Gulgong Goldfields N.S.W.

A detailed magnetic and gravity survey was done on portion of the Gulgong Goldfields, New South Wales. The geophysical survey was undertaken to study the geology in an area blanketed by a thick alluvial cover and in particular to investigate a buried, Tertiary river system. This system is known to be present and was once the object of active gold mining operation. The magnetic survey revealed a complex anomaly pattern containing both normal and reversed anomalies. These features were difficult to interpret due to the paucity of geological information and did not appear to be related to the channel structures. The gravity survey was successful in distinguishing the generally broad bedrock depressions associated with the gold-bearing channels. Owing to the presence of a complex regional feature, qualitative interpretation was simplified by applying a number of second derivative operators to the gravity data. Quantitative interpretation wer known ground suggested that lateral density variations within the bedrock or alluvium did not significantly contribute to the measured anomalies and that a density contrast of 0.5 gm/cc exiats between the bedrock and alluvium.

Latest Quaternary Deposits in Osaka Bay

The sonic survey by “SPARKER” revealed the following facts concerning the relations between the submarine topography and the youngest deposits in Osaka Bay.1) The unconsolidated deposits of Osaka Bay are divisible into A and B layers (Fig. 2). B is infered to be the latest Pleistocene sediments deposited in the stage when the sea level retarded at 20m below the present level about 10, 000 years ago. This stage may be identified with the Allerod oscillation. A is the Holocene sediments formed by the quick transgression occurred in the next stage.2) Submarine flat planes are classified as follows.a) -10m; wave-cut terrace formed at the present sea level.b) -10∼-20m; depositional surface of A.c) -40∼-50m; approximate depositional surface of B.Generalized profile of Osaka Bay shown in Fig. 5 indicates the relations between the change of sea level and the deposition of A and B.3) The surface of the base of the unconsolidated sediments is contoured in Fig. 3, on which a system of buried river channels can be recognized. It is named Paleo-Osaka River, which may have been entrenched in the regression stage of maximum phase of Wurm Ice Age. The dotted lines in Fig. 3 express the subsiding movement of Osaka Bay since that age.4) Akashi and Kitan straits have their origin in the Paleo-Osaka River system, but their sea caldrons reaching more than -100m have been formed by the submarine erosion after the deposition of B.

I.— The Sub-Drift Surface of the Lower Devon Valley

The existence of remarkable depths of drift beneath the flat carse lands of the lower Devon valley has been known for many years, the discovery being associated with mining operations in the Stirling and Clackmannanshire coalfield. Rockhead in places lies over 300 ft. below sea level, and the form of the sub-drift surface as revealed by boring led H. M. Cadell (1913) to suppose that a buried river channel underlay the present valley, linking with a similar channel beneath the Forth (Fig. 1). The form and extent of a feature which allows drift deposits to reach over 300 ft. below O.D. are of importance not only in connection with mining operations, but also from the geomorphological point of view, in so far as they reveal something of the nature and mode of origin of the feature. Evidence available in 1913 did not permit the exact position and full extent of the channel to be plotted, but the greatly increased information now available has made it possible to produce the accompanying map of the sub-drift surface (Fig. 2). It is considered that this map is sufficiently detailed to allow the probable origin of the buried channel to be fully discussed. The Devon buried channel, as it was described by Cadell, underlies the present river in its westward course at the foot of the Ochil scarp from Dollar to Menstrie, and is excavated entirely in strata of Carboniferous age. Its sides appear to be continuous on the north with the steep slopes This 250-word extract was created in the absence of an abstract

A pinnacle of chalk penetrating the Eocene on the floor of a buried river-channel at Ashford Hill, Near Newbury, Berkshire

Summary An isolated pinnacle of chalk,about 175 feet across and at least 150 feet above its normal level,rises through the Reading Beds and part of the London Clay near Ashford hill, and is associated with a drift-filled channel whose floor is, at its deepest proved point, 90 feet below the present ground-level. The channel reaches its greatest depth on the top of the chalk pinnacle, and seems not to extend beyond it. It is filled mostly with peaty silts with subordinate beds of gravel; and at depths of about 40 feet from the surface the pollen-content of the peats indicates a Lateglacial age. At the deepest point, almost 50 feet of alternating varves and Coombe Rock underlie the Late-glacial drift, in what appears to have been a swallow-hole. The pinnacle of chalk into which this pit was excavated has evidently been driven up through the Eocene when in a plastic state; its elevation and exposure must have antedated the development of the deep channel which apparently ends at it. In an attempted explanation of these associated anomalies (both of which seem to have developed during the Glacial Period) it is suggested that the chalk at the point of convergence of two faults may have been subjected to deep freezing and subsequently to "eruption" as a tectonically localized "valley bulge".

The Buried Rock Surface and Pre-Glacial River Valleys of Minneapolis and Vicinity

Previous articleNext article FreeThe Buried Rock Surface and Pre-Glacial River Valleys of Minneapolis and VicinityE. K. SoperE. K. Soper Search for more articles by this author PDFPDF PLUS Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmail SectionsMoreDetailsFiguresReferencesCited by The Journal of Geology Volume 23, Number 5Jul. - Aug., 1915 Article DOIhttps://doi.org/10.1086/622258 Views: 35Total views on this site PDF download Crossref reports no articles citing this article.

I. Notes on Buried River Channel at Motherwell

The old river channel referred to in this paper is laid down on the 1-inch scale Geological Maps, sheets 31 and 23, and is referred to in the memoir to sheet 23 published in 1873.1 It is also referred to in the “Summary of Progress of the Geological Survey” for 1911.2 The channel is first found at Cleland Townhead, and has been traced thence to Motherwell, a distance of 4 miles. Although doubtless it continues westward from Motherwell, the position has not yet been located. On the 6-inch Geological Map (sheet xii., Lanarkshire) it is described as follows:—“Trough filled with clay, sand, and gravel (with boulder clay covering), pre-glacial and inter-glacial course of the Tillon burn.” From Cleland Townhead to Glencleland, a distance of 1¼ miles, the channel runs in a southerly direction, and the width shown on the map is 300 feet. This part of the channel has been proved by the old coal workings. Near the junction of the Tillon burn and the Calder Water the old channel is intersected by the channel of the Calder. On the south bank of the Calder, near the viaduct carrying a mineral railway, there is here a very interesting exposure, where the gravel can be seen occupying the bed of the old channel. The channel now appears to turn towards the west. At Nether Johnstone sand quarry, three-quarters of a mile west from Glencleland, there is an exposure of sand and gravel which presumably occupies the site of the old channel. This 250-word extract was created in the absence of an abstract

Note on the buried river channel of the Almond

During recent years evidence has accumulated to prove that before the Glacial Period Scotland stood several hundreds of feet higher above sea-level than it does now. In a paper I read before this Society more than twenty years ago, reference was made to the deep valley of the ancient Forth; and mining operations under the sea at Bridgeness have, within recent years, proved the existence of a deep trough comparatively close to the south shore at that place. The coal workings followed one of the seams north to about a quarter of a mile below low water mark, to the N.W. of Bridgeness Harbour, and N.E. of Bo'ness, and struck the boulder clay at a depth of about 400 feet below high water level, where the seam was abruptly cut off, and the working had to be abandoned in that direction. More recently the clay has been struck in the same colliery a short distance further north at 600 feet below sea level. If the old bed of the Forth is so deep near the shore, it is probably still deeper further out. Since this trough was (happily for the colliery) full of boulder clay, and not sand or gravel, at the place, it is uncertain whether to regard it as part of the ancient channel, or part of an ice-scooped rock-basin subsequently filled up with till. In any case there is ample evidence that the pre-historic Forth flowed eastwards along a deep glen or gorge with steep sides at