Geomorphic Environments Research Articles

Geomorphic environments, drainage breakdown, and channel and floodplain evolution on the lower Macquarie River, central-western New South Wales

The Macquarie River of eastern Australia is an inland draining perennial system whose lower reaches flow across a large, low-gradient alluvial plain without tributary input. Failure of the Macquarie to sustain its channelled meandering course results in channel breakdown and the development of a broad distributary pattern characterised by a network of discontinuous anabranching and distributary channels and associated permanent and semi-permanent wetlands. The location of the channel breakdown region results from a combination of channel hydrology and sediment load. However, at the local scale, the complete breakdown into unchannelised marshes is influenced by the degree of topographic confinement produced by surrounding paleo-alluvial ridges. Several distinctive marsh channel types are identified on the lower Macquarie, each playing a different role in the process of channel and floodplain evolution. In general, new marsh channels are initiated by avulsion and incision into the floodplain surface and then gradually aggrade until a new avulsion results in abandonment, typically over a time-scale of ∼100 a. Floodplain development occurs as a series of spatially discrete sediment lobes, each of which are deposited in permanent swamps immediately downstream of the channel breakdown. The swamps trap almost all of the sediment delivered to them. Despite this, low suspended sediment concentrations result in slow rates of floodplain accretion. Lobe accretion is terminated by channel avulsion through low levees at the channel margin. Avulsion and channel breakdown at the local scale are driven by the characteristic nature of the dense in-channel reed vegetation which chokes the streams and leads to in-channel aggradation, reduced fluvial efficiency and frequent overbank flows that ultimately lead to the development of new channels.

Geomorphological investigations of sediment traps on Lanzarote (Canary Islands) as a key for the interpretation of a palaeoclimate archive off NW Africa

On Lanzarote (Canary Islands) Late Quaternary Saharan dust and volcanic material were trapped in Miocence to Pliocene valleys dammed by volcanic lava flows. These trapped sediments are potentially interesting as they can be natural archives useful to reconstruct the terrestrial palaeoclimate history of the NW African margin. Nevertheless, slope wash processes altered the primarily eolian deposits, making climatic interpretation not straightforward. Geomorphological mapping, GIS calculations and sedimentological investigations were used to unravel these processes influencing the temporal resolution of the palaeoclimatic archive, demonstrating that they average the palaeoclimatic signal by some ka. Thus, despite the colluvial geomorphic environment, the valley fillings can be used for palaeoclimatic interpretation of events with a length of at least some ka. The youngest sediments, deposited since at least 2.5 ka, are anthropogenically triggered and thus cannot be used for palaeoclimatic interpretation. The results show that the input of Saharan dust at Lanzarote increased during the last 1.0 ka and especially during the Early/Middle Holocene.

A new geoarchaeology of Aboriginal artefact deposits in western NSW, Australia: establishing spatial and temporal geomorphic controls on the surface archaeological record

Surface deposits of stone artefacts are the most common feature of the Australian Aboriginal archaeological record, but they remain difficult for archaeologists to interpret. Among the many reasons is a lack of understanding of geomorphic processes that have exposed the artefacts at the surface. We describe research on the geomorphic environments in arid Australia from which we have developed a new geoarchaeological framework for describing and analysing surface artefact deposits. Optically Stimulated Luminescence dating of sediments upon which the artefacts currently rest demonstrates that the landscape, and the archaeological record it preserves, is spatially and temporally discontinuous. Exposure and/or burial of artefacts is controlled by geomorphic processes operating on timescales ranging from a few decades to thousands of years and spatial scales of tens to many thousands of square meters. These same processes, operating on similar scales, also determine whether or not artefact scatters are preserved in the contemporary landscape or in the sedimentary record of past landscapes, and hence whether or not they become part of the archaeological record. Models of settlement behavior in hunter–gatherer peoples that are largely derived from analysis of surface ‘sites’ must take account of these discontinuities.

An overview of the role of weathering in the production of quartz silt

Quartz silt is a widespread detrital sediment with large aeolian (loess) and alluvial silt deposits forming important components of many contemporary landscapes. Despite research findings which indicate that a range of opportunities exist for the comminution of quartz into silt particles within a wide variety of geomorphic environments (e.g. glacial grinding, fluvial comminution, aeolian abrasion, frost weathering, salt weathering, insolation weathering and deep weathering), the ‘glacial–aeolian’ hypothesis has traditionally been favoured as the most likely explanation for loess formation and loess is seen as primarily a Quaternary phenomenon. As a consequence there has been a tendency to underestimate sediment inputs into loess systems by geomorphological processes operating within pre-Quaternary environments. In particular, earth scientists may have considerably underestimated the role of weathering in global silt generation, as there are now many references to the existence of a patchy but widespread distribution of pre-Quaternary weathering profiles across the glacial and periglacial landscapes of the Northern Hemisphere (e.g. North America, British Isles, north and central Europe), landscapes within which many of the classical loess deposits are located. These observations suggest that weathering profiles may have covered large tracts of the Northern Hemisphere land surface prior to the Quaternary glaciations. This, in turn, may have important implications for quartz silt generation as: (1) experimental studies indicate that many weathering processes are capable of generating significant quantities of silt-sized debris, and (2) particle characteristics displayed by saprolitic material that has developed on quartz rich crystalline source rocks often include significant quantities of either silt-sized material, or quartz grains that are weakened by weathering derived microfractures. Thus, weathering profiles may represent ‘mines’ of actual and potential quartz silt, available for release into sedimentary systems when these profiles are subsequently reworked by geomorphological processes such as glacial, fluvial and aeolian erosion.

Palaeoenvironmental reconstruction of the Late Pleistocene to Early Holocene Robertson Cave sedimentary deposit, Naracoorte, South Australia

Sedimentological investigations and geochemical analyses of the Robertson Cave deposit, 7 km south of Naracoorte, South Australia, document a palaeoenvironmental record for the Late Pleistocene and Early Holocene. Three distinct depositional episodes spanning the interval 32 – 8 ka are represented by over 4 m of sediments exposed in the chamber 2 excavation pit. The oldest of the three, the lower unit (>30 ka BP) is dominated by reddish and brown sandy silts. A 20 cm band of red silt horizons that cap this unit contain enriched δ15N values (12 – 15‰), low C/N ratios, significant SO3 (20%) and CaO (18.5%) contents, a high alkyl-C component and phosphorus-rich mineralogy. These unique sediment horizons are interpreted as being derived from the accumulation of bat guano and degraded cave rock fragments. Their presence suggests a period of low sediment input into the cave, possibly reflecting the existence of a stable geomorphic environment prior to 30 ka. The time period 30 – 27 ka saw the rapid deposition of the middle unit, a sequence of coarser homogeneous sands that resulted in the entrance to chamber 2 becoming blocked. These horizon-less sediments, with clear aeolian features, are most likely representative of the drier conditions associated with the Last Glacial Maximum, which peaked at 20 – 17 ka. Deposition into the chamber recommenced at approximately 13 ka and continued until 8 ka, resulting in the accumulation of the upper unit, a sequence of silt, charcoal and organic matter rich horizons. Here, a shift to more 13C-depleted values (−25‰ to −27‰) is evident in the bulk soil organic matter and charcoal between 11 and 10 ka, possibly reflecting a shift from the dry Last Glacial Maximum to the wetter environs of the Holocene. The upper unit may also be a record of burning practices associated with Aboriginal migration to the area.

The Architecture of Prograding Sandy-Gravel Beach Ridges Formed During the Last Holocene Highstand: Southwestern British Columbia, Canada

This study was conducted to establish the sedimentary architecture of a beach-ridge system that developed in response to rising and then stabilizing sea level in a macrotidal setting. The beach ridges document a shift from transgressive to highstand depositional conditions. A beach-ridge complex is being deposited on the Boundary Bay tidal flats that flank the Fraser River delta in southwestern British Columbia. The delta onlapped an offshore island, during the Holocene, and partially incorporated it into its delta plain; sand and gravel eroded from this island is the sediment source for the beach ridges. Four depositional–geomorphic environments are identified: beach ridges, lagoon and swales, washover fans, and tidal flats. The beach ridges prograded over the tidal flats of the Fraser delta when sea level stabilized and the late-Holocene highstand was established. The very earliest beach ridges were deposited about 5200 to 4800 years BP, but most were formed after 2250 years BP with the establishment of the Holocene highstand. Beach-ridge progradation rates varied between 1.6 m a −1 to 0.48 m a −1 . Ground penetrating radar (GPR) was used to provide continuous two- and three-dimensional images of the facies in order to delineate the depositional architecture of the sandy-gravel beach ridges. We supplemented the customary use of well logs and cores for ground truthing GPR lines with cone penetration testing (CPT), which provided a detailed suite of logs invaluable for correlating radar reflections with subtle lithologic variations unidentifiable in core. A synthesis of geomorphic environments, lithologic logs, and radar facies is used to construct a depositional model of the beach-ridge complex.

An environmental model of fluvial tufas in the monsoonal tropics, Barkly karst, northern Australia

Spring-fed streams that deposit tufa (ambient temperature freshwater calcium carbonate deposits) in the tropics of northern Australia are influenced strongly by perennially warm water temperatures, high evaporation rates, and monsoon driven high-magnitude floods. This paper presents an environmental model that will aid interpretation of fossil fluvial tufas throughout monsoonal Australia. In the Barkly karst, northern Australia, tufas form in dam, cascade and pool/waterhole geomorphic environments. Each environment is represented in the morphostratigraphical record by a specific combination of tufa geomorphic units and facies associations. A diverse array of tufa facies is present, including microphytic, larval, calcite raft, macrophytic and allochthonous types. Preservation of particular Barkly karst tufa facies is thought to reflect the strength of monsoonal floods. A strong monsoon is represented by an abundance of flood indicators such as the allochthonous phytoclastic, lithoclastic and intraclastic tufa facies. Conversely, evidence of weak monsoons or a prolonged absence of floods may include oncoids, calcite rafts and thick accumulations of fine carbonate sediments. The history of the Australian monsoon is not fully understood. However, fossil tufa deposits, which record terrestrial climate information, have been preserved throughout northern Australia and hold great potential for reconstructing the region's climate history. Fossil tufa sequences at two Barkly karst sites have been interpreted using the new model. It can be applied to other Barkly karst fossil tufas as well as those in similar environments elsewhere in the world.

Elevations of sulfurous layers in acid sulfate soils: What do they indicate about sea levels during the Holocene in eastern Australia?

This study is the first large scale examination of the stratigraphic relationships of acid sulfate soils in Australia. Field and laboratory data from 346 excavated acid sulfate soil profiles for an extensive soil survey undertaken to map their extent were classified into non acid sulfate soil (NASS), potential acid sulfate soil (PASS), and actual acid sulfate soil (AASS) layers, and the elevation of these layers within different geomorphic environments was examined. Layers containing sulfurous materials were found at 0.35 m below mean sea level in intertidal swamps, around mean sea level in areas subject to intermittent water logging, and greater than 0.40 m above mean sea level in other areas. Sulfurous materials in intertidal swamps represent present day formation while in other landforms are evidence for a sea level that was once higher than at present. The division of sulfurous sediment elevations into two groups above present day sea level could be explained by two processes. Firstly, two periods of higher sea level during the Holocene could result in sulfurous materials at different levels. Secondly, the barrier system which developed along the NSW coast during the Holocene and impounded a series of estuarine mud basins could have resulted in the formation of sulfurous materials at higher elevations along the edge of the lakes and at lower elevations on the lake beds themselves.

Earthquake‐related geomorphic environment and pond sediment information

AbstractShaking during the 1995 Kobe earthquake caused surface material to be more mobile in catchment areas in the Rokko Mountains, Kobe, where there are some active fault lines. As a result, there were many landslides associated with the earthquake. The sedimentation rate in a pond in the mountains increased several fold, then exponentially decreased with seasonality over several years. Six years after the earthquake there were no marked surface movements related to the earthquake, even though the sedimentation rates had increased slightly. A new steady state for the structure of the earthquake‐modified surface had evidently been reached. Copyright © 2004 John Wiley & Sons, Ltd.

Ground penetrating radar: 2-D and 3-D subsurface imaging of a coastal barrier spit, Long Beach, WA, USA

The ability to effectively interpret and reconstruct geomorphic environments has been significantly aided by the subsurface imaging capabilities of ground penetrating radar (GPR). The GPR method, which is based on the propagation and reflection of pulsed high frequency electromagnetic energy, provides high resolution (cm to m scale) and shallow subsurface (0–60 m), near continuous profiles of many coarser-grained deposits (sediments of low electrical conductivity). This paper presents 2-D and 3-D GPR results from an experiment on a regressive modern barrier spit at Willapa Bay, WA, USA. The medium-grained sand spit is 38 km long, up to 2–3.5 km wide, and is influenced by a 3.7-m tidal range (spring) as well as high energy longshore transport and high wave energy depositional processes. The spit has a freshwater aquifer recharged by rainfall.The GPR acquisition system used for the test was a portable, digital pulseEKKO™ system with antennae frequency ranging from 25 to 200 MHz and transmitter voltages ranging from 400 to 1000 V. Step sizes and antennae separation varied depending on the test requirements. In addition, 100-MHz antennae were used for conducting antennae orientation tests and collecting a detailed grid of data (50×50 m sampled every meter). The 2-D digital profiles were processed and plotted using pulseEKKO™ software. The 3-D datasets, after initial processing, were entered into a LANDMARK™ workstation that allowed for unique 3-D perspectives of the subsurface. To provide depth, near-surface velocity measurements were calculated from common midpoint (CMP) surveys.Results from the present study demonstrate higher resolution from the 200-MHz antennae for the top 5–6 m, whereas the 25- and 50-MHz antennae show deeper penetration to >10 m. For the study site, 100-MHz antennae provided acceptable resolution, continuity of reflections, and penetration. The dip profiles show a shingle-like accretionary depositional pattern, whereas strike profiles show a horizontal and subhorizontal, nearly continuous reflection pattern. Results from the GPR experiment reveal upper shoreface reflections with dip towards the ocean at about 1–2°. The loss of signal from below a depth of 6–8 m indicates a lithofacies change because of the storm wave base.The parallel broadside and perpendicular broadside antennae orientation tests show detailed stratigraphy, continuity, and depth of penetration. The cross-polarization test exhibits reduced continuity of reflections and less depth of penetration, but dipping reflections are apparent.The grid pattern data provided a detailed view of 3-D geometry of individual reflections. High quality data were obtained, processed, and directly exported into a LANDMARK™ workstation for interpretation. The resulting interpretations of the upper shoreface beds from the test cube (50×50 m; total 2600 traces) are shown as vertical sections (slices), horizontal sections (time slices), contour maps, 3-D representations of individual beds, and an isopach map. The 3-D depositional framework allows a more detailed interpretation than widely spaced 2-D profiles.

Sedimentary architecture and radar facies of a fan delta, Cypress Creek, West Vancouver, British Columbia

Abstract Fan deltas are common depositional features along the macro-tidal, fjorded coast of southern British Columbia. To determine the sedimentary architecture of the Cypress Creek fan delta, West Vancouver, British Columbia, a combined ground penetrating radar and geomorphological study was carried out. Five radar facies were identified and their associated geomorphic environments were established: oblique clinoform radar facies (delta front); mounded radar facies (debris-flow deposits); undulating subhorizontal radar facies (beach zone); concave radar facies (river channels) and horizontal radar facies (salt-water intrusion). The delta plain was formed primarily by debris flows, some of which reached the delta front to form prograding foreset beds. Radar facies analysis established the occurrence of arcuate scars and masses of slumped material, signifying mass movement on the delta front. Wave action, working over a 4.9 m tidal range, has reworked the distal delta-plain debris-flow sediments into an armoured gravel beach. A depositional model is proposed for a fan delta in a macro-tidal fjord setting.

Spatial distribution and management of total actual acidity in an acid sulfate soil environment, McLeods Creek, northeastern NSW, Australia

The spatial distribution of total actual acidity (TAA), defined as the total amount of acidity which exists in a soil at the time of sampling, was examined across an acid sulfate soil floodplain in northeastern New South Wales (NSW), Australia. Despite generally uniform soil conditions, there is considerable variation in the amount of acidity, and the amounts of soluble and exchangeable ionic species in the soil profile are positively correlated with this acidity. The surface hydrology of the site has been extensively modified for sugarcane production. It is hypothesised from air photo interpretation and the spatial soil data that variation in TAA is a result of the natural geomorphic environment, and that the current distribution pattern is a remnant of past natural land formation and hydrological processes controlling pyrite oxidation and acidity export. The degree to which land drainage caused the acidity is unclear, but the drainage systems provide the conduit for its increased transfer to estuaries. By investigating the distribution of acidity in the landscape, ‘hotspots’ can be identified and land managers can target these areas. Currently, the acidity is managed by a containment program in which it is kept within the soil profile and discharge into the estuary is minimised. Work is under way to apply emerging technology from the mining industry so that any acidity that enters drains is neutralised prior to discharge from the site.

Clays and secondary minerals as permafrost indicators: examples from the circum-Baikal region

A proper understanding of Quaternary environments requires assessment of the possible existence of perennially frozen ground. Furthermore, the structural and mineralogical changes induced by permafrost are commonly irreversible and still affect the present soils, so that it is important to be able to recognize them. However, the regions that suffered severe, but dry periglacial conditions during cold Quaternary periods, as in southeastern Siberia, show few periglacial features. Five exposures investigated in the circum-Baikal area show permafrost features, although the same phenomena occur throughout southeastern Siberia. These exposures demonstrate that, even though evidence of permafrost may be lacking or equivocal, thorough analysis can reveal indicators of permafrost such as neoformed clays, secondary precipitates of iron, calcite, and sulphates with peculiar crystalline features, as are found in other past or presently perpetually frozen ground. In combination with sedimentary structures and the geomorphic environment, these indicators affirm the definite existence of permafrost.

Heavy metal contamination of an arid river environment: Gruben River, Namibia

The dispersal of metalliferous pollution from several spoil heaps produced during 20th century copper mining in the Gruben River valley, Namibia is examined. The Gruben River flows through an extremely arid environment with an average rainfall of ~25 mm p.a. The potential for physical and chemical remobilisation of Cu, Zn and Ni is assessed by examining the spatial and temporal distribution of metal-contaminated sediment deposited within the confines of the channel. The relationships between metal content, grain size, geomorphic environment and the downstream distribution of metals are discussed. The phase-specific heavy metal concentrations of sediments, collected as part of the downstream sampling programme, are also examined using progressively more aggressive sequential acid extractions. In addition, metal concentrations are compared with Dutch guidelines for soil contamination to ascertain the extent of environmental risk. Total metal concentrations show that the Gruben valley is highly contaminated, particularly with respect to Cu and Ni concentrations, which exceed Dutch target values for Cu (36 ppm) in 94.7% and Ni (35 ppm) in 90.5% of samples, respectively. Zn concentrations are much lower with only 6.3% of the samples exceeding the target value (140 ppm). As might be expected, the metal–sediment concentrations of Cu are the most highly elevated, with a peak value of 10,500 ppm being recorded from material collected from suspended sediment transported during a minor flow event that occurred in March 1999. Lower energy and fine-grained sedimentary environments are shown to be clearly associated with higher metal concentrations. Sequential extractions of metals show that of the three principal elements considered in this study, only a negligible proportion of Cu (0.41%) is held in the exchangeable phase. Ni and Zn were below detection limits. Although the total metal concentrations in the Gruben River valley sediments are extremely high and are in themselves a concern, the small percentage of metals held in the exchangeable phase and the low potential for remobilisation under the arid conditions would suggest that they pose only a minor risk to the environment.

Conceptualizing Environmental Stress: A Stress–Response Model of Coastal Sandy Barriers

/ The purpose of this paper is to develop and apply a conceptual framework of environmental stress-response for a geomorphic system. Constructs and methods generated from the literature were applied in the development of an integrative stress-response framework using existing environmental assessment techniques: interaction matrices and a systems diagram. Emphasis is on the interaction between environmental stress and the geomorphic environment of a sandy barrier system. The model illustrates a number of stress concepts pertinent to modeling environmental stress-response, including those related to stress-dependency, frequency-recovery relationships, environmental heterogeneity, spatial hierarchies and linkages, and temporal change. Sandy barrier stress-response and recovery are greatly impacted by fluctuating water levels, stress intensity and frequency, as well as environmental gradients such as differences in sediment storage and supply. Aspects of these stress-response variables are articulated in terms of three main challenges to management: dynamic stability, spatial integrity, and temporal variability. These in turn form the framework for evaluative principles that may be applied to assess how policies and management practices reflect key biophysical processes and human stresses identified by the model.

Airborne SAR and IRS‐1A LISS II Data Interpretation of Coastal Geomorphology in Godavari Delta, India

This paper describes the coastal Geomorphology of Godavari Delta studied by using airborne X-band Synthetic Aperture Radar (SAR) images and optical images of IRS‐1A LISS II. Palaeo‐offshores and the associated fluviomarine geomorphic features have been interpreted. The river morphology has been built with the help of interpretative maps prepared by the digital and visual analysis of SAR and LISS II data. X‐band airborne SAR images provided better tonal and textural variation and sharpness to delineate geomorphic features. The delineation of features such as meanderscars, palaeochannel scars and other features with greater ease compared to IRS‐1A LISS II vividly demonstrates it. Two geomorphic environments viz., fluvial and marine have been identified based on the interpreted landforms. The Godavari delta was built up with distributaries in four stages, displaying various geomorphic features. On the basis of field studies and images interpretation five strandlines have been identified. A younger strandline, hitherto unreported has been identified parallel to the present coastline. Surface expressions of basement faults and morphostructures have been delineated from SAR images.

Cluster analysis of grain size information for possible linkages between geomorphic environments: some case studies from Indian arid zone

Cluster analysis of grain size information for possible linkages between geomorphic environments: some case studies from Indian arid zone

Sedimentation in Chilko Lake: a record of the geomorphic environment of the eastern Coast Mountains of British Columbia, Canada

Chilko Lake is the largest and deepest lake draining the Coast Mountains of British Columbia. A sedimentary record from the 60-km long lake was constructed by using 3.5-kHz subbottom acoustic profiles and short cores of surface sediments. Sediment accumulation since deglaciation is remarkably thin, averaging less than 10 m, except in the central deep part of the lake, despite extensive activity of Holocene and modern glaciers in the 1960-km 2 contributing watershed. Sediment focusing of silts and clays in the main part of the lake occurs mainly by overflow/interflow processes from the Edmond and Nine Mile/Deschamps drainages. Sandy turbidites from these drainages and other smaller streams, however, extend for several kilometers into the lake basin. Distinct laminae in short cores from the central lake suggest that modern rates of deposition reach a maximum of 2.2 mm a −1. 210 Pb dating and inferences from the acoustic results, however, indicate a lake-wide, long-term average rate of less than 1 mm a −1, which is equivalent to a specific sediment yield of about 80 t km −2 a −1. This specific yield is amongst the lowest for glacier-fed lakes of the southern Canadian Cordillera. A low sediment supply can be related to three factors: (1) a much drier climate on the leeward side of the Coast Mountains producing lower specific discharges; (2) fewer autumn rainstorms that affect mainly lakes west of the Coast Mountains and (3) opportunities for storage of sediment in the lower 20 km of Edmond River, which drains the most heavily glacierized portion of the watershed.

A scale for judging degree of soil and paleosol development

Better understanding of soil formation requires knowing the spatial distribution of the soils that allows constructing models of soil sequences in multiple directions along various types of gradients. This approach was applied to comprehend the soil formation from the soil distribution in the tropical karst areas of the Peninsula of Yucatan, Mexico. For soil mapping, a two-step methodology was followed. First, a geomorphic analysis was performed; subsequently, 382 soil profiles were reclassified and integrated into a geopedologic map. Additional soil survey was carried out in areas where soil information was lacking (123 soil profiles). Satellite images were used to identify flooded areas. After conducting numerous field verifications and analyses, landforms and soils were combined to make a soilscape map. Based on field observations and the soilscape map, soil development was analyzed on soil sequences. Four geomorphic environments were identified, karstic plains and hills with Leptosols, Cambisols, Luvisols, and Vertisols; coastal plains with Arenosols, Regosols, Solonchacks, and Histosols; fluvio-paludal plains with Gleysols, Histosols, Leptosols and Solonchacks; and tectono-karstic plains and hills with Leptosols, Cambisols, Luvisols, and Vertisols. Relevant soil forming factors in the Peninsula of Yucatan include time of emersion of the limestone platforms, climate, rock type, and macro- and micro-relief. Other factors such as groundwater level, fractures, also have an influence on soil formation. Karst development can be considered as a complex soil and relief forming factor. Terra Rossa soils as Leptosols, Cambisols, Luvisols, Nitisols and Vertisols in the Peninsula of Yucatan may be polygenic. In some cases, the theory of residual origin fits better the data than the theory of allochthonous origin; in other cases, it is the other way around.

Distribution of the damage caused by the earthquake of Jan. 17, 1995 and its geomorphic environment as seen from fallen tombstones (in Japanese) : T. Kawasaki, R. Nomura & M. Oya, Geographical Review of Japan, Series, A, 69(1), 1996, pp 39–50

Distribution of the damage caused by the earthquake of Jan. 17, 1995 and its geomorphic environment as seen from fallen tombstones (in Japanese) : T. Kawasaki, R. Nomura & M. Oya, Geographical Review of Japan, Series, A, 69(1), 1996, pp 39–50