Methods for the visualization of digital elevation models for landform mapping

Digital elevation models (DEMs) are increasingly used for landform mapping, particularly with the growing availability of national and global datasets. In this paper we describe a variety of techniques that can visualize a DEM. We then compare five techniques to ascertain which performs the most complete and unbiased visualization. We assess the visualization techniques by comparing landforms mapped from them against a detailed morphological map (derived from mapping of multi‐azimuth relief‐shaded DEMs cross‐checked with stereo aerial photographs). Results show that no single visualization method provides complete and unbiased mapping. The relief‐shaded visualizations are particularly prone to azimuth biasing, although they can highlight subtle landforms. We recommend curvature visualization for initial mapping as this provides a non‐illuminated (and therefore unbiased) image. Initial mapping can then be supplemented with data from relief‐shaded visualizations. Copyright © 2005 John Wiley & Sons, Ltd.     

Hyperscale terrain modelling of braided rivers: fusing mobile terrestrial laser scanning and optical bathymetric mapping

Quantifying the morphology of braided rivers is a key task for understanding braided river behaviour. In the last decade, developments in geomatics technologies and associated data processing methods have transformed the production of precise, reach‐scale topographic datasets. Nevertheless, generating accurate Digital Elevation Models (DEMs) remains a demanding task, particularly in fluvial systems. This paper identifies a threefold set of challenges associated with surveying these dynamic landforms: complex relief, inundated shallow channels and high rates of sediment transport, and terms these challenges the ‘morphological’, ‘wetted channel’ and ‘mobility’ problems, respectively. In an attempt to confront these issues directly, this paper presents a novel survey methodology that combines mobile terrestrial laser scanning and non‐metric aerial photography with data reduction and surface modelling techniques to render DEMs from the resulting very high resolution datasets. The approach is used to generate and model a precise, dense topographic dataset for a 2.5 km reach of the braided Rees River, New Zealand. Data were acquired rapidly between high flow events and incorporate over 5 x 10⁹ raw survey observations with point densities of 1600 pts m^‐2 on exposed bar and channel surfaces. A detailed error analysis of the resulting sub‐metre resolution is described to quantify DEM quality across the entire surface model. This reveals unparalleled low vertical errors for such a large and complex surface model; between 0.03 and 0.12 m in exposed and inundated areas of the model, respectively. Copyright © 2013 John Wiley & Sons, Ltd.     

Flow network derivation from a high resolution DEM in a low relief,agrarian landscape

Digital flow networks derived from digital elevation models (DEMs) sensitively react to errors due to measurement, data processing and data representation. Since high‐resolution DEMs are increasingly used in geomorphological and hydrological research, automated and semi‐automated procedures to reduce the impact of such errors on flow networks are required. One such technique is stream‐carving, a hydrological conditioning technique to ensure drainage connectivity in DEMs towards the DEM edges. Here we test and modify a state‐of‐the‐art carving algorithm for flow network derivation in a low‐relief, agricultural landscape characterized by a large number of spurious, topographic depressions. Our results show that the investigated algorithm reconstructs a benchmark network insufficiently in terms of carving energy, distance and a topological network measure. The modification to the algorithm that performed best, combines the least‐cost auxiliary topography (LCAT) carving with a constrained breaching algorithm that explicitly takes automatically identified channel locations into account. We applied our methods to a low relief landscape, but the results can be transferred to flow network derivation of DEMs in moderate to mountainous relief in situations where the valley bottom is broad and flat and precise derivations of the flow networks are needed. Copyright © 2013 John Wiley & Sons, Ltd.     

Evaluation of remotely‐sensed DEMs and modification based on plausibility rules and initial sediment budgets of an artificially‐created catchment

To quantify landscape change resulting from processes of erosion and deposition and to establish spatially distributed sediment budgets, ‘models of change’ can be established from a time series of digital elevation models (DEMs). However, resolution effects and measurement errors in DEMs may propagate to these models. This study aimed to evaluate and to modify remotely‐sensed DEMs for an improved quantification of initial sediment mass changes in an artificially‐created catchment. DEMs were constructed from photogrammetry‐based, airborne (ALS) and ground‐based laser scanning (TLS) data. Regions of differing morphological characteristics and vegetation cover were delineated. Three‐dimensional (3D) models of volume change were established and mass change was derived from these models. DEMs were modified region‐by‐region for rill, interrill and alluvial areas, based on logical and hydro‐geomorphological principles. Additional DEMs were constructed by combining multi‐source, modified data. Models were evaluated by comparison with d‐GPS reference data and by considering sediment budget plausibility. Comprehensive evaluation showed that DEM usability depends on a relation between the technique used to obtain elevation data, surface morphology and vegetation cover characteristics. Photogrammetry‐based DEMs were suited to quantification of change in interrill areas but strongly underestimated surface lowering in erosion rills. TLS DEMs were best suited to rill areas, while ALS DEMs performed best in vegetation‐covered alluvial areas. Agreement with reference data and budget plausibility were improved by modifications to photogrammetry‐ and TLS‐based DEMs. Results suggest that artefacts in DEMs can be reduced and hydro‐geomorphic surface structures can be better represented by applying region‐specific modifications. Photogrammetry‐based DEMs can be improved by combining higher and lower resolution data in defined structural units and applying modifications based on principles given by characteristic hydro‐geomorphic evolution. Results of the critical comparative evaluation of remotely‐sensed elevation data can help to better interpret DEM‐based quantifications of earth‐surface processes. Copyright © 2012 John Wiley & Sons, Ltd.     

Misfit functionals in Laplace‐Fourier domain waveform inversion,with application to wide‐angle ocean bottom seismograph data

In seismic waveform inversion, non‐linearity and non‐uniqueness require appropriate strategies. We formulate four types of L2 normed misfit functionals for Laplace‐Fourier domain waveform inversion: i) subtraction of complex‐valued observed data from complex‐valued predicted data (the ‘conventional phase‐amplitude’ residual), ii) a ‘conventional phase‐only’ residual in which amplitude variations are normalized, iii) a ‘logarithmic phase‐amplitude’ residual and finally iv) a ‘logarithmic phase‐only’ residual in which the only imaginary part of the logarithmic residual is used. We evaluate these misfit functionals by using a wide‐angle field Ocean Bottom Seismograph (OBS) data set with a maximum offset of 55 km. The conventional phase‐amplitude approach is restricted in illumination and delineates only shallow velocity structures. In contrast, the other three misfit functionals retrieve detailed velocity structures with clear lithological boundaries down to the deeper part of the model. We also test the performance of additional phase‐amplitude inversions starting from the logarithmic phase‐only inversion result. The resulting velocity updates are prominent only in the high‐wavenumber components, sharpening the lithological boundaries. We argue that the discrepancies in the behaviours of the misfit functionals are primarily caused by the sensitivities of the model gradient to strong amplitude variations in the data. As the observed data amplitudes are dominated by the near‐offset traces, the conventional phase‐amplitude inversion primarily updates the shallow structures as a result. In contrast, the other three misfit functionals eliminate the strong dependence on amplitude variation naturally and enhance the depth of illumination. We further suggest that the phase‐only inversions are sufficient to obtain robust and reliable velocity structures and the amplitude information is of secondary importance in constraining subsurface velocity models.     

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

Glacial bedform height (H) and volume (V) likely preserve important information about the behaviour of former ice sheets. However, large systematic errors exist in the measurement of H and V. Three semi‐automated methods to isolate drumlins from other components of the landscape (e.g. trees, hills) as portrayed by NEXTMap have recently been devised; however, it is unclear 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 for quantifying the 3D attributes of drumlins are proposed. Specifically, to avoid obtaining incorrect estimates caused by substantial systematic 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 Delaunay triangulation. Results are demonstrated through analysis of drumlins in an area in western central Scotland. The guidance arguably represents the best current advice for subglacial bedforms in general, highlighting the need for more studies into the quality of mapped data using synthetic landscapes. Copyright © 2014 John Wiley & Sons, Ltd.     

An erosional origin for drumlins of NW Poland

Drumlins are landforms essential to understanding of ice sheet movement over soft beds, sediment transport along the ice/bed interface, and the formation of a wide range of glacial deposits. Although investigated more than any other glacial landform, the origin of drumlins remains contentious. Using high-resolution LiDAR imagery and field data, we investigate the geomorphology and internal composition of one of the biggest drumlin fields in the North European Lowland. The Stargard drumlin field consists of over 1300 drumlins and related streamlined subglacial bedforms in a terminal part of a major Weichselian palaeo-ice stream of the southern Scandinavian Ice Sheet. The drumlins are typically 600-800 m long, 200-250 m wide, 3-6 m high and have axial elongation ratios ~2 but in some cases exceeding 15. Several subzones inferred from drumlin morphometry exist reflecting different ice flow dynamics. The most elongated drumlins occur in areas where ice moved down-slope and where thick fine-grained deposits of low hydraulic conductivity occur in the substratum. The largest portion of land occupied by drumlins and the greatest frequency density of drumlins occur where the ice moved up-slope. Stargard drumlins are composed of a wide variety of glacial deposits including various types of tills and meltwater sediments, which range from undisturbed to heavily deformed. There is no correlation between the deposits in the drumlins and the drumlin forms indicating that the deposits pre-date the drumlinizing process. It is suggested that the drumlin field was generated by a combination of direct glacial erosion and subglacial meltwater erosion by removing antecedent material from the inter-drumlin areas and streamlining the resultant bumps. Our data support the search for a unifying theory of drumlin formation and suggest erosion as the most plausible single mechanism generating drumlin landscapes. © 2019 John Wiley & Sons, Ltd.     

Initial hydro‐geomorphic development and rill network evolution in an artificial catchment

The formation of erosion rills and gullies is a critical step in land surface development, but possibilities to study initial unaffected surface development under natural conditions and with well‐defined initial and boundary conditions are rare. The objective of this study was to characterize rill network development from ’point zero’ in the artificially‐created catchment ‘Hühnerwasser’. To ensure unaffected development, the study was largely restricted to the analysis of remotely‐sensed data. We analyzed a series of photogrammetry‐based digital elevation models (DEMs) for 10 points in time, over a period of five years and beginning with the initial state. The evolving erosion rill network was quantitatively described based on mapping from aerial photographs. DEMs and rill network maps were combined to specifically analyze the development of morphometry for different parts of the network and to characterize energy dissipation and connectivity. The restriction to remote‐sensing data did not allow for analyzing specific processes governing rill network development, nevertheless, two major development phases could be characterized. We observed a phase of growth of the rill network along with variations in drainage patterns during the first two years of development and a subsequent phase of reduction of its area along with comparably stable patterns. Region‐specific analysis of morphometry indicates that, besides effects of changing sediment characteristics and vegetation cover development, locally evolving hydro‐geomorphic feedback cycles influenced this development. Results show an increasing similarity of overall statistical characteristics (e.g. drainage density) for two parts of the catchment, but a persistent influence of initial conditions on specific rill geometry. The observed development towards higher orderliness and increased connectivity is consistent with experiments and concepts on drainage network evolution across scales; however, we did not observe major influences of rill piracy and cross grading or a reduction of energy dissipation with network development. Copyright © 2013 John Wiley & Sons, Ltd.     

The normalized topographic method: an automated procedure for gully mapping using GIS

Gully delineation is a critical aspect of accurately determining soil losses but associated methodologies are rarely detailed. Here, we describe a new gully mapping method, the normalized topographic method (NorToM), based on processing digital elevation model (DEM) data, and we assess associated errors when it is applied over a range of geomorphological scales. The NorToM is underpinned by two gully detection variables (normalized slope and elevation) calculated over local windows of prescribed size, and a group of filtering variables. For four study sites, DEMs of gullies were obtained using field and airborne photo‐reconstruction and evaluated using total station and differential global positioning system (dGPS) survey. NorToM provided accurate areal and volume estimates at the individual gully scale but differences increased at the larger gully system and gully network scales. We were able to identify optimal parameters for using the NorToM approach and so confirm that is represents a useful scale‐independent means of gully mapping that is likely to be valid in other environments. Its main limitations are that the normalization process might be time‐consuming at regional scales and the need for a fixed window size when applied to landforms with extreme variations in dimensions. Copyright © 2014 John Wiley & Sons, Ltd.     

DIGITAL MAPPING OF HURST CASTLE SPIT,HAMPSHIRE

This paper serves two purposes. The first is to make available to the geomorphological community the first in a short series of up-to-date digital maps of ‘classic’ coastal landforms of the U.K. — Hurst Castle Spit. The second is to demonstrate that conventional surveying techniques and readily available graphics software on a PC or Mackintosh will provide an entry into digital mapping for geomorphological applications. Comments are included on how the digital map may be utilized.     

Paleotopographic controls on loess deposition in the Loess Plateau of China

The underlying pre‐existing paleotopography directly influences the loess deposition process and shapes the morphology of current loess landforms. An understanding of the controlling effects of the underlying paleotopography on loess deposition is critical to revealing the mechanism of loess‐landform formation. However, these controlling effects exhibit spatial variation as well as uncertainty, depending on a study's data sources, methodologies and particular research scope. In this study, the geological history of a study area in the Loess Plateau of China that is subject to severe soil erosion is investigated using detailed geological information and digital elevation models (DEMs), and an underlying paleotopographic model of the area is constructed. Based on the models of modern terrain and paleotopography, we introduce a watershed hierarchy method to investigate the spatial variation of the loess‐landform inheritance relationship and reveal the loess deposition process over different scales of drainage. The landform inheritance relationships were characterized using a terrain‐relief change index (TRCI) and a bedrock terrain controllability index (BTCI). The results show that the TRCI appears to have an inverse relationship with increasing research scope, indicating that, compared with the paleotopography of the region, modern terrain has lower topographic relief over the entire area, while it has higher topographic relief in the smaller, local areas. The BTCI strengthens with increasing drainage area, which demonstrates a strong controlling effect over the entire study area, but a weak effect in the smaller, local areas because of the effect of paleotopography on modern terrain. The results provide for an understanding of the spatial variation of loess deposition in relation to paleotopography and contribute to the development of a process‐based loess‐landform evolution model. Copyright © 2015 John Wiley & Sons, Ltd.     

Multitemporal ALSM change detection,sediment delivery,and process mapping at an active earthflow

Remote mapping and measurement of surface processes at high spatial resolution is among the frontiers in Earth surface process research. Remote measurements that allow meter‐scale mapping of landforms and quantification of landscape change can revolutionize the study of landscape evolution on human timescales. At Mill Gulch in northern California, USA, an active earthflow was surveyed in 2003 and 2007 by airborne laser swath mapping (ALSM), enabling meter‐scale quantification of landscape change. We calculate four‐year volumetric flux from the earthflow and compare it to long‐term catchment average erosion rates from cosmogenic radionuclide inventories from adjacent watersheds. We also present detailed maps of changing features on the earthflow, from which we can derive velocity estimates and infer dominant process. These measurements rely on proper digital elevation model (DEM) generation and a simple surface‐matching technique to align the multitemporal data in a manner that eliminates systematic error in either dataset. The mean surface elevation of the earthflow and an opposite slope that was directly influenced by the earthflow decreased 14 ± 1 mm/yr from 2003 to 2007. By making the conservative assumption that these features were the dominant contributor of sediment flux from the entire Mill Gulch drainage basin during this time interval, we calculate a minimum catchment‐averaged erosion rate of 0·30 ± 0·02 mm/yr. Analysis of beryllium‐10 (¹⁰Be) concentrations in fluvial sand from nearby Russian Gulch and the South Fork Gualala River provide catchment averaged erosion rates of 0·21 ± 0·04 and 0·23 ± 0·03 mm/yr respectively. From translated landscape features, we can infer surface velocities ranging from 0·5 m/yr in the wide upper ‘source’ portion of the flow to 5 m/yr in the narrow middle ‘transport’ portion of the flow. This study re‐affirms the importance of mass wasting processes in the sediment budgets of uplifting weak lithologies. Copyright © 2011 John Wiley & Sons, Ltd.     

The use of ablation‐dominated medial moraines as samplers for 10Be‐derived erosion rates of glacier valley walls,Kichatna Mountains,AK

We use cosmogenic ¹⁰Be concentrations in amalgamated rock samples from active, ice‐cored medial moraines to constrain glacial valley sidewall backwearing rates in the Kichatna Mountains, Alaska Range, Alaska. This dramatic landscape is carved into a small ～65 Ma granitic pluton about 100 km west of Denali, where kilometer‐tall rock walls and ‘cathedral’ spires tower over a radial array of over a dozen valley glaciers. These supraglacial landforms erode primarily by rockfall, but erosion rates are difficult to determine. We use cosmogenic ¹⁰Be to measure rockwall backwearing rates on timescales of 10³–10⁴ years, with a straightforward sampling strategy that exploits ablation‐dominated medial moraines. A medial moraine and its associated englacial debris serve as a conveyor system, bringing supraglacial rockfall debris from accumulation‐zone valley walls to the moraine crest in the ablation zone. We discuss quantitatively several factors that complicate interpretation of cosmogenic concentrations in this material, including the complex scaling of production rates in very steep terrain, the stochastic nature of the rockfall erosion process, the unmixed nature of the moraine sediment, and additional cosmogenic accumulation during transport of the sediment. We sampled medial moraines on each of three glaciers of different sizes and topographic aspects. All three moraines are sourced in areas with identical rock and similar sidewall relief of ～1 km. Each sample was amalgamated from 25 to 35 clasts collected over a 1‐km longitudinal transect of each moraine. Two of the glaciers yield similar ¹⁰Be concentrations (～1·6–2·2 × 10⁴ at/g) and minimum sidewall slope‐normal erosion rates (～0·5–0·7 mm/yr). The lowest ¹⁰Be concentrations (8 × 10³ at/g) and the highest erosion rates (1·3 mm/yr) come from the largest glacier in the range with the lowest late‐summer snowline. These rates are reasonable in an alpine glacial setting, and are much faster than long‐term exhumation rates of the western Alaska Range as determined by thermochronometric studies. Copyright © 2010 John Wiley & Sons, Ltd.     

Topographic structure from motion: a new development in photogrammetric measurement

The production of topographic datasets is of increasing interest and application throughout the geomorphic sciences, and river science is no exception. Consequently, a wide range of topographic measurement methods have evolved. Despite the range of available methods, the production of high resolution, high quality digital elevation models (DEMs) requires a significant investment in personnel time, hardware and/or software. However, image‐based methods such as digital photogrammetry have been decreasing in costs. Developed for the purpose of rapid, inexpensive and easy three‐dimensional surveys of buildings or small objects, the ‘structure from motion’ photogrammetric approach (SfM) is an image‐based method which could deliver a methodological leap if transferred to geomorphic applications, requires little training and is extremely inexpensive. Using an online SfM program, we created high‐resolution digital elevation models of a river environment from ordinary photographs produced from a workflow that takes advantage of free and open source software. This process reconstructs real world scenes from SfM algorithms based on the derived positions of the photographs in three‐dimensional space. The basic product of the SfM process is a point cloud of identifiable features present in the input photographs. This point cloud can be georeferenced from a small number of ground control points collected in the field or from measurements of camera positions at the time of image acquisition. The georeferenced point cloud can then be used to create a variety of digital elevation products. We examine the applicability of SfM in the Pedernales River in Texas (USA), where several hundred images taken from a hand‐held helikite are used to produce DEMs of the fluvial topographic environment. This test shows that SfM and low‐altitude platforms can produce point clouds with point densities comparable with airborne LiDAR, with horizontal and vertical precision in the centimeter range, and with very low capital and labor costs and low expertise levels. Copyright © 2012 John Wiley & Sons, Ltd.     

Application of globally available,coarse-resolution digital elevation models for delineating valley bottom segments of varying length across a catchment

The worldwide availability of digital elevation models (DEMs) has enabled rapid (semi-)automated mapping of earth surface landforms. In this paper, we first present an approach for delineating valley bottom extent across a large catchment using only publicly available, coarse-resolution DEM input. We assess the sensitivity of our results to variable DEM resolution and find that coarse-resolution datasets (90 m resolution) provide superior results. We also find that LiDAR-derived DEMs produce more realistic results than satellite-derived DEMs across the full range of topographic settings tested. Satellite-derived DEMs perform more effectively in moderate topographic settings, but fail to capture the subtleties of valley bottom extent in mild gradient, low-lying topography and in narrow headwater reaches. Second, we present a semi-automated technique within ArcGIS for delineating valley bottom segments using DEM-derived network scale metrics of valley bottom width and slope. We use an unsupervised machine-learning technique based on the k-means clustering algorithm to solve a conundrum in GIS-based geomorphic analysis of rivers: the delineation of valley bottom segments of variable length. The delineation of valley bottom segments provides a coarse-scale entry point into automated geomorphic analysis and characterization of river systems. © 2020 John Wiley & Sons, Ltd.     

Multi‐scale relief model (MSRM): a new algorithm for the visualization of subtle topographic change of variable size in digital elevation models

Morphological analysis of landforms has traditionally relied on the interpretation of imagery. Although imagery provides a natural view of an area of interest (AOI) images are largely hindered by the environmental conditions at the time of image acquisition, the quality of the image and, mainly, the lack of topographical information, which is an essential factor for a correct understanding of the AOI's geomorphology. More recently digital surface models (DSMs) have been incorporated into the analytical toolbox of geomorphologists. These are usually high‐resolution models derived from digital photogrammetric processes or LiDAR data. However, these are restricted to relatively small areas and are expensive or complex to acquire, which limits widespread implementation. In this paper, we present the multi‐scale relief model (MSRM), which is a new algorithm for the visual interpretation of landforms using DSMs. The significance of this new method lies in its capacity to extract landform morphology from both high‐ and low‐resolution DSMs independently of the shape or scale of the landform under study. This method thus provides important advantages compared to previous approaches as it: (1) allows the use of worldwide medium resolution models, such as SRTM, ASTER GDEM, ALOS, and TanDEM‐X; (2) offers an alternative to traditional photograph interpretation that does not rely on the quality of the imagery employed nor on the environmental conditions and time of its acquisition; and (3) can be easily implemented for large areas using traditional GIS/RS software. The algorithm is tested in the Sutlej‐Yamuna interfluve, which is a very large low‐relief alluvial plain in northwest India where 10 000 km of palaeoriver channels have been mapped using MSRM. The code, written in Google Earth Engine's implementation of JavaScript, is provided as Supporting Information for its use in any other AOI without particular technical knowledge or access to topographical data. © 2017 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.     

Geomorphic response to an active transpressive regime: a case study along the Chaman strike‐slip fault,western Pakistan

The Chaman left‐lateral strike‐slip fault bounds the rigid Indian plate boundary at the western end of the Himalayan‐Tibetan orogen and is marked by contrasting topographic relief. Deformed landforms along the fault provide an excellent record for understanding this actively evolving intra‐continental strike‐slip fault. The geomorphic response of an active transpessional stretch of the Chaman fault was studied using digital elevation model (DEM) data integrated with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Visible and Near Infrared/Short Wave Infrared (VNIR/SWIR) and images from GeoEye‐1. Geologic and geomorphic mapping helped in reconstructing the Late Quaternary landscape history of this transpessional strand of the Chaman strike‐slip fault and the associated Spinatizha thrust fault in western Pakistan. Topographic analysis of a part of the transpression (the thrust bounded Roghani ridge) revealed northward growth of the Spinatizha fault with the presence of three water gaps and two corresponding wind gaps. Geomorphic indices including stream length‐gradient index, mountain front sinuosity, valley floor width to valley height ratios, and entrenchment of recent alluvial fan deposits were used to define the lateral growth and direction of propagation of the Spinatizha fault. Left‐lateral displacement along Chaman fault and uplift along the Spinatizha fault was defined using topographic analysis of the Roghani ridge and geomorphic mapping of an impressive alluvial fan, the Bostankaul fan. The landforms and structures record slip partitioning along the Indian plate boundary, and account for the convergence resulting from the difference in the Chaman fault azimuth and orientation of the velocity vector of the Indian plate. Copyright © 2012 John Wiley & Sons, Ltd.     

Mitigating systematic error in topographic models derived from UAV and ground‐based image networks

High resolution digital elevation models (DEMs) are increasingly produced from photographs acquired with consumer cameras, both from the ground and from unmanned aerial vehicles (UAVs). However, although such DEMs may achieve centimetric detail, they can also display systematic broad‐scale error that restricts their wider use. Such errors which, in typical UAV data are expressed as a vertical ‘doming’ of the surface, result from a combination of near‐parallel imaging directions and inaccurate correction of radial lens distortion. Using simulations of multi‐image networks with near‐parallel viewing directions, we show that enabling camera self‐calibration as part of the bundle adjustment process inherently leads to erroneous radial distortion estimates and associated DEM error. This effect is relevant whether a traditional photogrammetric or newer structure‐from‐motion (SfM) approach is used, but errors are expected to be more pronounced in SfM‐based DEMs, for which use of control and check point measurements are typically more limited. Systematic DEM error can be significantly reduced by the additional capture and inclusion of oblique images in the image network; we provide practical flight plan solutions for fixed wing or rotor‐based UAVs that, in the absence of control points, can reduce DEM error by up to two orders of magnitude. The magnitude of doming error shows a linear relationship with radial distortion and we show how characterization of this relationship allows an improved distortion estimate and, hence, existing datasets to be optimally reprocessed. Although focussed on UAV surveying, our results are also relevant to ground‐based image capture. © 2014 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.