Search

Future climate projections indicate an expansion of the world’s drylands, and with that a commensurate increase in the mobilization of unconsolidated desert sediments such as sand and dust. It is therefore increasingly important to investigate the large-scale formation of dryland landscapes such as dunefields in order to better understand the processes responsible for their genesis, evolution, and thresholds for mobilization. Assessing dunefield morphologic variability, including analysis of the morphologic relationship between aeolian bedforms and other landforms such as fluvial channels and bedrock uplands, underpins such investigations. So far, however, meaningful investigations of erg-scale geomorphic patterns have been limited. This is in part due to the technological limitations of geographic information system (GIS) tools, particularly in the case of open-source datasets and software, which has effectively hindered investigations by colleagues in drylands of the global south where many of the world’s dunefields are located. Recent years have overseen the increasing availability of open-source remote sensing datasets, as well as the development of freely available software which can undertake geographic object-based image analysis (GEOBIA). These new tools facilitate cartography and statistical analysis of dunefields at large scales. In this study we make use of open-source GIS to characterise a morphologically diverse linear dunefield southwest of Kati Thanda (Lake Eyre) in central Australia. We focus on three parameters; dune orientation, spacing and Y-junctions using semi-automated GEOBIA, and investigate these in the context of local fluvial channels, depressions (pans) and uplands. Our results suggest a possible correlation between dune orientation, wind regime and the role of uplands as deflective barriers to longitudinal dune migration; dune spacing and sediment supply, likely relating to the location of both ephemeral and abandoned fluvial channels; and Y-junction frequency with underlying topography. Our study provides a framework for understanding process-based interactions between dunes and other landforms, as well as the first completely open-source approach which can be applied to linear dunefields worldwide.

Abstract Many studies of Earth surface processes and landscape evolution rely on having accurate and extensive data sets of surficial geologic units and landforms. Automated extraction of geomorphic features using deep learning provides an objective way to consistently map landforms over large spatial extents. However, there is no consensus on the optimal input feature space for such analyses. We explore the impact of input feature space for extracting geomorphic features from land surface parameters (LSPs) derived from digital terrain models (DTMs) using convolutional neural network (CNN)‐based semantic segmentation deep learning. We compare four input feature space configurations: (a) a three‐layer composite consisting of a topographic position index (TPI) calculated using a 50 m radius circular window, square root of topographic slope, and TPI calculated using an annulus with a 2 m inner radius and 10 m outer radius, (b) a single illuminating position hillshade, (c) a multidirectional hillshade, and (d) a slopeshade. We test each feature space input using three deep learning algorithms and four use cases: two with natural features and two with anthropogenic features. The three‐layer composite generally provided lower overall losses for the training samples, a higher F1‐score for the withheld validation data, and better performance for generalizing to withheld testing data from a new geographic extent. Results suggest that CNN‐based deep learning for mapping geomorphic features or landforms from LSPs is sensitive to input feature space. Given the large number of LSPs that can be derived from DTM data and the variety of geomorphic mapping tasks that can be undertaken using CNN‐based methods, we argue that additional research focused on feature space considerations is needed and suggest future research directions. We also suggest that the three‐layer composite implemented here can offer better performance in comparison to using hillshades or other common terrain visualization surfaces and is, thus, worth considering for different mapping and feature extraction tasks.

Remote sensing is a relevant method to map inaccessible areas, such as intertidal mudflats. However, image classification is challenging due to spectral similarity between microphytobenthos and oyster reefs. Because these elements are strongly related to local geomorphic features, including biogenic structures, a new mapping method has been developed to overcome the current obstacles. This method is based on unmanned aerial vehicles (UAV), RGB, and multispectral (four bands: green, red, red-edge, and near-infrared) surveys that combine high spatial resolution (e.g., 5 cm pixel), geomorphic mapping, and machine learning random forest (RF) classification. A mudflat on the Atlantic coast of France (Marennes-Oléron bay) was surveyed based on this method and by using the structure from motion (SfM) photogrammetric approach to produce orthophotographs and digital surface models (DSM). Eight classes of mudflat surface based on indexes, such as NDVI and spectral bands normalised to NIR, were identified either on the whole image (i.e., standard RF classification) or after segmentation into five geomorphic units mapped from DSM (i.e., geomorphic-based RF classification). The classification accuracy was higher with the geomorphic-based RF classification (93.12%) than with the standard RF classification (73.45%), showing the added value of combining topographic and radiometric data to map soft-bottom intertidal areas and the user-friendly potential of this method in applications to other ecosystems, such as wetlands or peatlands.

The sand islands and shore-attached dune fields of south east Queensland form the world’s oldest and largest coastal sand dune system. Here we present updated morphological maps for the dune fields based on topographic expression and geomorphic relationships. Individual dunes were delineated using high-resolution elevation data and were grouped into morphosequences based on the elevation, drainage characteristics and slope characteristics of the individual dunes. The slope characteristics focussed on high resolution derived slope-curvature and slope-elevation parameters. Morphosequences were recognised from cross-cutting relationships and relative position in the dune field. Our method was developed for the Cooloola Sand Mass and then applied to Fraser Island, Moreton Island and North Stradbroke Island, the other major sand islands in south east Queensland. In total, five Holocene and four Pleistocene units have been identified. The new mapping underpins current work on the geomorphic evolution of the dune fields.

Rock avalanches are low frequency natural hazards that can alter landscape morphology, and constraining the timing, volume and emplacement dynamics of pre-historic rock avalanches is crucial for understanding the hazards posed by these events. Here we perform cosmogenic nuclide dating, topographic reconstruction and runout modeling of the Molveno rock avalanche, located north of Lake Garda in the Province of Trento, Italy. The unique morphology of the deposits, which features numerous large scarps and prominent lineaments, have previously led researchers to interpret the Molveno rock avalanche as being the result of multiple events. Our results show that the Molveno rock avalanche had a volume of approximately 600 Mm3, and failed from a prominent niche located on Monte Soran. 36Cl cosmogenic nuclide dating results shows that the deposits were emplaced as a single event approximately 4.8 ± 0.5 ka, and suggests that the unique deposit morphology is due to the emplacement processes acting during and soon after failure. Numerical runout modeling shows that this morphology could have resulted from a combination of runup and extensional spreading of the debris along the complex valley floor topography. The ages we determined for this event are coincident with the nearby Marocca Principale rock avalanche (5.3 ± 0.9 ka), which may suggest a common trigger. Our results have important implications for interpreting the morphology of rock avalanche deposits, and contribute to the evolving understanding of rock avalanche processes in the Alps.