CLustre: semi-automated lineament clustering for palaeo-glacial reconstruction

Wednesday, 12 August, 2015

Smith, M.J., Anders, N. and Keesstra, S.D. (in press)
Earth Surface Processes and Landforms


Datasets containing large numbers (>10,000) of glacial lineaments are increasingly being mapped from remotely sensed data in order to develop a palaeo-glacial reconstruction or “inversion”. The palimpsest landscape presents a complex record of past ice flow and deconstructing this information into a logical history is an involved task. One stage in this process requires the identification of sets of genetically linked lineaments that can form the basis of a reconstruction.

This paper presents a semi-automated algorithm, CLustre, for lineament clustering that uses a locally adaptive, region growing, methodology. After outlining the algorithm, it is tested on synthetic datasets that simulate parallel and orthogonal cross-cutting lineaments, encompassing 1,500 separate classifications. Results show robust classification in most scenarios, although parallel overlap of lineaments can cause false positive classification unless there are differences in lineament length. Case studies for Dubawnt Lake and Victoria Island, Canada, are presented and compared to existing datasets. For Dubawnt Lake 9 out of 14 classifications directly match incorporating 89% of lineaments. For Victoria Island 57 out of 58 classifications directly match incorporating 95% of lineaments. Differences are related to small numbers of unclassified lineaments and parallel cross-cutting lineaments that are of a similar length.

CLustre enables the automated, repeatable, assignment of lineaments to flow sets using defined user criteria. This is important as qualitative visual interpretation may introduce bias, potentially weakening the testability of palaeo-glacial reconstructions. In addition, once classified, summary statistics of lineament clusters can be calculated and subsequently used during the reconstruction process.

OPEN ACCESS EPRINT: Optimising UAV image quality using consumer cameras

Thursday, 18 June, 2015

O’Connor, J. and Smith, M.J. (2015)
Geomatics World

Surveys using UAVs and photogrammetry are becoming ubiquitous. It is in the community’s best interests to remember the fundamentals of image capture. James O’Connor and Mike J Smith of the School of Geography, Geology and the Environment, Kingston University review the main considerations for image capture when undertaking an aerial survey using consumer grade cameras and make recommendations for acquiring the best imagery.

OPEN ACCESS EPRINT: Use of legacy data in geomorphological research

Wednesday, 18 March, 2015

Smith, M.J., Keesstra, S. and Rose, J. (2015)
GeoResJ


This paper considers legacy data and data rescue within the context of geomorphology. Data rescue may be necessary dependent upon the storage medium (is it physically accessible) and the data format (e.g. digital file type); where either of these is not functional, intervention will be required in order to retrieve the stored data. Within geomorphological research, there are three scenarios that may utilize legacy data: to reinvestigate phenomena, to access information about a landform/process that no longer exists, and to investigate temporal change. Here, we present three case studies with discussion that illustrate these scenarios: striae records of Ireland were used to produce a palaeoglacial reconstruction, geomorphological mapping was used to compile a map of glacial landforms, and aerial photographs were used to analyze temporal change in river channel form and catchment land cover.

Manual mapping of a synthetic landscape to assess operator effectiveness

Tuesday, 23 September, 2014

Hillier, J.K., Smith, M.J., Barr, L., Boston, C., Clark, C.D., Ely, R., Fankl, A., Greenwood, S., Gosselin, L., Hattesrand, C., Hogan, K., Hughes, A., Livingstone, S.J., Lovell, H., McHenry, M., Munoz, Y., Pelicier13, X., Pellitero, R., Robb, C., Robertson, S., Ruther, D., Spagnolo, M., Standell, M., Stokes, C., Storrar, R., Tate, N., Wooldridge, K. (2015)
Journal of Maps


Mapped topographic features are important for understanding processes that sculpt the Earth’s surface. This paper presents maps that are the primary product of an exercise that brought together 27 researchers with an interest in landform mapping where the efficacy and causes of variation in mapping were tested using novel synthetic DEMs containing drumlins. The variation between interpreters (e.g., mapping philosophy, experience) and across the study region (e.g., woodland prevalence) opens these factors up to assessment. A priori known answers in the synthetics increase the number and strength of conclusions that may be drawn with respect to a traditional comparative study. Initial results suggest that overall detection rates are relatively low (34-40%), but reliability of mapping is higher (72-86%). The maps form a reference dataset.

OPEN ACCESS EPRINT: Multiresolution segmentation to automatically delimit landforms in DEMs: tests using synthetic drumlins

Tuesday, 23 September, 2014

Eisank, C., Smith, M.J. and Hillier, J. (2014)
Geomorphology


Mapping or “delimiting” landforms is one of geomorphology’s primary tools. Computer-based techniques such as land-surface segmentation allow the emulation of the process of manual landform delineation. Land-surface seg-mentation exhaustively subdivides a digital elevation model (DEM) into morphometrically-homogeneous irregularly-shaped regions, called terrain segments. Terrain segments can be created from various land-surface parameters (LSP) at multiple scales, and may therefore potentially correspond to the spatial extents of landforms such as drumlins. However, this depends on the segmentation algorithm, the parameterization, and the LSPs. In the present study we assess the widely used multiresolution segmentation (MRS) algorithm for its potential in providing terrain segments which delimit drumlins. Supervised testing was based on five 5-m DEMs that repre-sented a set of 173 synthetic drumlins at random but representative positions in the same landscape. Five LSPs were tested, and four variants were computed for each LSP to assess the impact of median filtering of DEMs, and logarithmic transformation of LSPs. The testing scheme (1) employs MRS to partition each LSP exhaustively into 200 coarser scales of terrain segments by increasing the scale parameter (SP), (2) identifies the spatially best matching terrain segment for each reference drumlin, and (3) computes four segmentation accuracy metrics for quantifying the overall spatial match between drumlin segments and reference drumlins. Results of 100 tests showed that MRS tends to perform best on LSPs that are regionally derived from filtered DEMs, and then log-transformed. MRS delineated 97% of the detected drumlins at SP values between 1 and 50. Drumlin delimitation rates with values up to 50% are in line with the success of manual interpretations. Synthetic DEMs are well-suited for assessing landform quantification methods such as MRS, since subjectivity in the reference data is avoided which increases the reliability, validity and applicability of results.

Testing techniques to quantify drumlin height and volume: synthetic DEMs as a diagnostic tool

Tuesday, 18 March, 2014

Hillier, J.K. and Smith, M.J. (2014)
Earth Surface Processes and Landforms


Glacial bedform height (H) and volume (V) likely preserve important information about the behaviour of former icesheets. However, large systematic errors exist in the measurement of H and V. Three semi-automated methods to isolate drumlinsfrom other components of the landscape (e.g. trees, hills) as portrayed by NEXTMap have recently been devised; however, it isunclear which is most accurate. This paper undertakes the first quantitative comparison of such readily implementable methods,illustrating the use of statistically representative ’ synthetic landscapes ’ as a diagnostic tool. From this analysis, guidelines forquantifying the 3D attributes of drumlins are proposed. Specifically, to avoid obtaining incorrect estimates caused by substantialsystematic biases, interpreters should currently take three steps: declutter the digital elevation model for estimating H but not V;remove height data within the drumlin; then interpolate across the resultant hole to estimate a basal surface using Delaunaytriangulation. Results are demonstrated through analysis of drumlins in an area in western central Scotland. The guidance arguablyrepresents the best current advice for subglacial bedforms in general, highlighting the need for more studies into the quality ofmapped data using synthetic landscapes.

Subglacial bedforms reveal an exponential size-frequency distribution

Saturday, 9 February, 2013

Hillier, J., Smith, M.J., Clark, C.D., Stokes, C.R. and Spagnolo, M. (2013)
Geomorphology


Subglacial bedforms preserved in deglaciated landscapes record character-istics of past ice-sediment flow regimes, providing insight into subglacial processes and ice sheet dynamics. Individual forms vary considerably, but they can often be grouped into coherent fields, typically called flow-sets, that reflect discrete episodes of ice flow. Within these, bedform size-frequency distributions (predominantly height, width and length) are currently described by several statistics (e.g., mean, median, standard deviation) that, arguably, do not best capture the defining characteristics of these populations. This pa-per seeks to create a better description based upon semi-log plots, which reveal that the frequency distributions of bedform dimensions (drumlin, MSGL, ribbed moraine) plot as straight lines above the mode (f). This indicates, by definition, an exponential dis-tribution, for which a simple and easily calculated, yet statistically rigorous, description is designed. Three descriptive parameters are proposed: gradient (the exponent, characterising beforms likely least affected by non-glacial factors), area-normalised y-intercept (quantifying spatial density), and the mode (f). Below f, small features are less preva-lent due to i) measurement: data, sampling, mapping fidelity ii) possibly post-glacial degra-dation or iii) genesis: being created, or not, sub-glacially. This new description has the beneift of being insentitive to the impact of potentially unmapped or degraded smaller features and better captures properties relating to ice flow. Importantly, using lambda, flow sets can now be more usefully compared with each other across the all deglaciated re-gions and with the output of numerical ice sheet models. Identifying the charactersitic exponential and that it is typical of ‘emergent’ subglacial bedforms is a new and poten-tially powerful constraint on their genesis, perhaps indicating that ice-sediment interac-tion is fundamentally stochastic in nature.

Bibliographic webmap: the Physical Landscape of Britain and Northern Ireland

Saturday, 9 February, 2013

Piccinini, C., Smith, M.J., Hooke, J. and Hesketh, K. (2013)
Journal of Maps


In this article we present the development of a web mapping application as part of the “Physical Landscape of Britain and Northern Ireland” project. There are a large number of sources for information and data on the geomorphology of the British landscape including academic journals, books, unpublished student dissertations and third party reports, however there remains no single central repository to query the extent of such knowledge. This project, sponsored by the British Society for Geomorphology, has a long term aim of collating a bibliography on the full extent of published and unpublished research on the geomorphology of Britain and Northern Ireland, ultimately highlighting “gaps” in our knowledge and identifying topics and areas for future investigation. The first phase of the project involved the development of a web application that allows end-users to search for bibliographic references using an interactive map. Here we outline the implementation of a spatial database, a database front-end and a web mapping application built using open source software and open data. Data for pilot areas has been entered; a further phase of the project will populate the database for other areas.

Testing 3D landform quantification methods with synthetic drumlins in a real digital elevation model

Wednesday, 21 March, 2012

Hillier, J.K. and Smith, M.J. (2012)
Geomorphology


Metrics such as height and volume quantifying the 3D morphology of landforms are important observations that re?ect and constrain Earth surface processes. Errors in such measurements are, however, poorly under-stood. A novel approach, using statistically valid ‘synthetic’ landscapes to quantify the errors is presented. The utility of the approach is illustrated using a case study of 184 drumlins observed in Scotland as quanti?ed from a Digital Elevation Model (DEM) by the ‘cookie cutter’ extraction method. To create the synthetic DEMs, observed drumlins were removed from the measured DEM and replaced by elongate 3D Gaussian ones of equivalent dimensions positioned randomly with respect to the ‘noise’ (e.g. trees) and regional trends (e.g. hills) that cause the errors. Then, errors in the cookie cutter extraction method were investigated by using it to quantify these ‘synthetic’ drumlins, whose location and size is known. Thus, the approach deter-mines which key metrics are recovered accurately. For example, mean height of 6.8 m is recovered poorly at 12.5±0.6 (2s) m, but mean volume is recovered correctly. Additionally, quantification methods can be compared: A variant on the cookie cutter using an un-tensioned spline induced about twice (×1.79) as much error. Finally, a previously reportedly statistically significant (p=0.007) difference in mean volume between sub-populations of different ages, which may reflect formational processes, is demonstrated to be only 30-50% likely to exist in reality. Critically, the synthetic DEMs are demonstrated to realistically model parameter recovery, primarily because they are still almost entirely the original landscape. Results are insen-sitive to the exact method used to create the synthetic DEMs, and the approach could be readily adapted to assess a variety of landforms (e.g. craters, dunes and volcanoes).

‘The Physical Landscape of Britain and Northern Ireland’: a project to increase geomorphological awareness

Wednesday, 21 March, 2012

Hooke, J.M. and Smith, M.J.
Scottish Geographical Journal


A project has been initiated that will produce an interactive website searchable by maps and database terms, aimed at increasing knowledge and awareness of the geodiversity, landforms and processes of Britain and Northern Ireland,. It has the following objectives: (1) to make existing knowledge on the physical landscape and geomorphology of Britain more widely accessible; (2) to increase knowledge and awareness of physical attributes of the landscape; (3) to provide a synthesis of present knowledge; (4) to identify gaps in present knowledge as a platform for future societally relevant research. The components and steps in the project are outlined.

Digital elevation models for research: UK datasets, copyright and derived products

Tuesday, 14 December, 2010

Smith, M.J. (2010)
Geological Society Special Publications


The UK is served by a wide range of digital elevation models (DEMs) that have a variety of technical specifications from several different vendors. The abundance of data presents researchers with a complex range of choices dependent upon their application (and therefore ‘fitness-for-purpose’) and desired use of intellectual property rights (IPR). This paper explores current DEM datasets of the UK and presents their use within the context of claimed copyright and IPR. In particular, responsibilities placed upon grant holders for the lodgement of research outputs by UK Research Councils places new emphasis upon data access, derived data and data re-use. The complex interplay of rights between research output stakeholders (data suppliers, data creators, data users) presents a difficult scenario for both data repositories and data depositors.

Credit Where Credit’s Due: developing authorship strategies in the geosciences

Thursday, 7 October, 2010

Smith, M.J., Jordan, C. and Walsby, J.
Proceedings of the Geologists’ Association


As research institutions seek to professionalise the workplace the use of metrics to assess an individual’s performance is becoming increasingly commonplace. For academic researchers this can be achieved through the use of publication metrics such as the number of articles published and number of citations. For non-academic professionals, such as cartographers, field assistants or database programmers, they may have limited inclusion as authors and therefore their contribution to research outputs and outcomes is more difficult to ascertain. This paper outlines the current de facto standards for authorship and proposes some potential solutions for the formal recognition of contributions by professionals to research projects. This is presented through strategies currently being trialed at the Journal of Maps and through the example of map publication at the British Geological Survey.

Palaeoglaciology of the Last Irish Ice Sheet Reconstructed from Striae Evidence

Wednesday, 29 September, 2010

Smith, M.J. and Knight, J.(in press)
Quaternary Science Reviews


A database comprising some ~5200 individual striation measurements on bedrock surfaces across the island of Ireland was used to produce maps of flowsets corresponding to individual ice flow events during the last (late Devensian) glacial cycle. These flowsets were identified on the basis of regional-scale correspondence between striae orientations which, when linked together spatially, are able to identify consistent ice flow vectors. Four main chronological stages are identified on the basis of this evidence: (i) incursion of Scottish ice into Ireland; (ii) glacial maximum conditions; (iii) ice retreat and dissolution; and (iv) development of localised ice domes. Striae-based reconstructions of the glaciology of the last Irish ice sheet are qualitatively different from those based on bedform (mainly drumlin and ribbed moraine) evidence. Significant differences are apparent in upland areas which have fewer preserved bedforms and a higher concentration of striae. Combining bedform and striae datasets will enable a better understanding of the temporal evolution of the ice sheet. It is likely that both datasets record a snapshot of ice flow direction and subglacial conditions and environments immediately prior to preservation of this directional evidence.

Multi-scale analysis of surface roughness

Thursday, 26 August, 2010

Grohmann, C.H., Smith, M.J. and Riccomini, C. (in press)
IEEE Transactions on Geoscience and Remote Sensing


Surface roughness is an important geomorphological variable which has been used in the earth and planetary sciences to infer material properties, current/past processes and the time elapsed since formation. No single definition exists, however within the context of geomorphometry we use surface roughness as a expression of the variability of a topographic surface at a given scale, where the scale of analysis is determined by the size of the landforms or geomorphic features of interest. Six techniques for the calculation of surface roughness were selected for an assessment of the parameter’s behaviour at different spatial scales and dataset resolutions. Area ratio operated independently of scale, providing consistent results across spatial resolutions. Vector dispersion produced results with increasing roughness and homogenisation of terrain at coarser resolutions and larger window sizes. Standard deviation of residual topography highlighted local features and doesn’t detect regional relief. Standard deviation of elevation correctly identified breaks-of-slope and was good at detecting regional relief. Standard deviation of slope (SDslope) also correctly identified smooth sloping areas and breaks-of-slope, providing the best results for geomorphological analysis. Standard deviation of profile curvature identified the breaks-of-slope, although not as strongly as SDslope and it is sensitive to noise and spurious data. In general, SDslope offered good performance at a variety of scales, whilst the simplicity of calculation is perhaps its single greatest benefit.

Characterising Chinese loess stratigraphy and past monsoon variation using field spectroscopy

Tuesday, 6 April, 2010

Smith, M.J., Stevens, T., MacArthur, A., Malthus, T. and Lu, H. (in press)
Quaternary International


The loess record in China constitutes one of the most important archives of past environmental change and specifically, the East Asian monsoon system. Changes in summer monsoon driven pedogenesis are commonly inferred from magnetic susceptibility of loess. However, there is still controversy as to the signal’s origin and the uncertain effects of sediment accumulation rates. This is linked to a wider problem that is emerging from recent work; that of the relative importance on proxy records such as magnetic susceptibility of both regional climate patterns (i.e. the monsoon) and local site-specific influences, particularly modulated through site-specific sedimentation rate. At present this is poorly constrained and significantly increases the degree of uncertainty over the wider-scale applicability of climate reconstructions from individual sites. To resolve this issue for sub-orbital timescales, a rapidly deployable technique is needed that can be used to test multiple sites and differentiate between local and regional signals. This research develops the first use of full spectrum reflectance spectroscopy in studying loess in the field at one site on the south of the Loess Plateau, and utilises statistical analyses to compare such data with magnetic susceptibility records. Establishment of loess reflectance as a proxy for climate variability will potentially allow extension of the technique from point measurements to an imaging system and so enable the compilation of large data sets in order to investigate lateral facies variations in loess profiles. This may allow the extraction of a broad scale climate record.

Field spectra were obtained from 350-1100 nm, with red edge reflectance spectra indicating the presence of iron-oxides, previously demonstrated to be indicators of pedogenesis through laboratory measurements. An off the shelf camera was also tested with visible wavelength spectra being useful for rapid but general profile characterisation. In order to mitigate the effects of bidirectional reflectance distribution function (a potential problem in field measurements), further laboratory measurement was made of powdered samples (400-2400 nm). Absorption features indicative of montmorillonite and illite/muscovite were noted and stepwise regression modelling utilising absolute reflectance, first derivative spectra and continuum removed spectra indicated strong predictive relationships with magnetic susceptibility, particularly against the presence of montmorillonite. The abundance of such clay minerals could be used to infer weathering rates and hence be used as a proxy for pedogensis. Reconstructions for the studied site are presented and demonstrate the potential power of loess reflectance environmental reconstruction.

Applications of Remote Sensing in Geomorphology

Thursday, 16 July, 2009

Smith, M.J. and Pain, C.F. (2009)
Progress in Physical Geography, 33, 4, 568-582.


Remotely sensed imagery has been used extensively in geomorphology since the availability of early Landsat data, with its value measurable by the extent to which it can meet the investigative requirements of geomorphologists. Geomorphology focuses upon landform description/classification, process characterisation and the association between landforms and processes, whilst remote sensing is able to provide information on the location/distribution of landforms, surface/subsurface composition and surface elevation. The current context for the application of remote sensing in geomorphology is presented with a particular focus upon the impact of new technologies, in particular: (i) the wide availability of digital elevation models and (ii) the introduction of hyperspectral imaging, radiometrics and electromagnetics. Remote sensing is also beginning to offer capacity in terms of close-range (<200 m) techniques for very high resolution imaging.

This paper reviews the primary sources for DEMs from satellite and airborne platforms, as well as briefly reviewing more traditional multi-spectral scanners, and radiometric and electromagnetic systems. Examples of the applications of these techniques are summarised and presented within the context of geomorphometric analysis and spectral modelling. Finally, the wider issues of access to geographic information and data distribution are discussed.

Editorial: Summary of Activities 2008

Monday, 13 April, 2009

Smith, M.J.
Journal of Maps, v2008, 510-516.


It is with much tiredness, and anticipation for 2009, that this annual December editorial is being written. If 2007 was characterised by a period of exciting growth, then this year can be described as consolidation. Whilst it may appear that we have delivered “more of the same”, there has been a considerable amount of behind the scenes administration that has led to continued growth. In my annual review of activities of the Journal of Maps last year I noted we had gone from 5 articles (56 pages) in 2005 to 27 articles (376 pages) in 2007. This growth is again shown this year with 38 articles (508 pages). To put this within the context of my own discipline, geomorphology, the top two journals published 244 and 148 articles each in 2007. This is clearly a subject specific benchmark, but shows generally that academic publishing is labour intensive with relatively low numbers of publications. That we have been this successful is a mark of the importance now being given to the presentation of research maps.

Editorial: Quaternary of the British Isles and Adjoining Seas

Monday, 13 April, 2009

Smith, M.J., Rose, J. and Lukas, S.
Journal of Maps, v2008, 290-298.


This volume of the Journal of Maps is one of four volumes published from an Annual Discussion Meeting (ADM) of the Quaternary Research Association (QRA), in January 2008, to examine the Quaternary of the British Isles and the adjoining seas. This volume deals with evidence for Quaternary environmental change in map-form, and the other volumes, which will be published in the journals Quaternary Science Reviews, Journal of Quaternary Science and Proceedings of the Geologists’ Association will be concerned, respectively, with Quaternary climates and climate change, Quaternary landscapes, and Quaternary Geology.

Editorial: Applied Geomorphological Mapping

Monday, 13 April, 2009

Paron, P. and Smith, M.J. (2008)
Journal of Maps, v2008, 197-200.


Mapping forms and deposits, and inferring processes, of a landscape is a very complex exercise (Demek, 1982). Its difficulty lies, on the one hand, in the challenge of identifying the processes themselves, their spatial and temporal extent (including the magnitude of activity) and the underlying base rocks (both in the field and through desk based analysis) and, on the other hand, the implementation of effective cartographic representation. General geomorphological maps are often driven by the need to understand the evolution of a portion of the landscape and a need to forecast future evolutionary trends. They have thus become a major research instrument in their own right (Hayden, 1986). Applied geomorphological mapping has to consider the evolution of the area under investigation, even if the specific objectives are more limited.

Glacial striae observations for Ireland compiled from historic records

Monday, 13 April, 2009

Smith, M.J., Knight, J. and Field, K.S.(2008)
Journal of Maps, v2008, 378-398.


Glacial landform mapping is one of the primary inputs for the reconstruction of past glacial environments and processes, potentially inferring maximum ice sheet extent and dynamics. Depositional landforms (e.g. drumlins, end moraines, ribbed moraines, eskers) are often used to infer former ice sheet conditions, with erosional indicators receiving less attention. For nearly 200 years, striae (linear, subglacial erosional marks on bedrock up to several metres long) have been recorded as evidence for former ice flow direction. Cumulative data collection by many researchers in Ireland since ~1850 has led to a large published and unpublished archive of striae observations. This research has collated over 5000 individual observations from Geological Survey of Ireland maps and memoirs, the published (peer-reviewed) literature, and unpublished work (theses and fieldnotes). These records are now unified in a single database, georeferenced to the Irish National Grid, with the accuracy of individual observations qualitatively assessed.