Search

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. Plain Language Summary: Characteristics of the land surface, such as steepness, relative slope position (e.g., ridge vs. valley), and roughness, can be digitally represented using a variety of methods. These digital representations, along with human‐annotated labels, can be provided to artificial intelligence algorithms to generate maps of landforms, such as those associated with river systems, glaciers, or human‐induced changes to the landscape. Given the large number of terrain derivatives that can be generated from digital elevation data, it is unclear how the chosen inputs impact the utility of the resulting maps. This study suggests that artificial intelligence mapping algorithms are sensitive to representations provided to them since different inputs resulted in varying map accuracies. We suggest a combination of features, which collectively describe relative slope position, steepness, and local terrain texture, as a means to represent the landscape and to serve as input to artificial intelligence algorithms. Our results can make it easier to train artificial intelligence algorithms to consistently and objectively find surficial features of interest across large swaths of Earth's surface. Key Points: We test the sensitivity of convolutional neural network geomorphic feature extraction approaches to the input feature spaceA three‐layer composite incorporating slope and topographic position index outperformed common inputs like hillshades and slopeshadesResults held true across four use cases including both natural and anthropogenic landforms [ABSTRACT FROM AUTHOR]

Passive margin stratigraphy contains time‐integrated records of landscapes that have long since vanished. Quantitatively reading the stratigraphic record using coupled landscape evolution and stratigraphic forward models (SFMs) is a promising approach to extracting information about landscape history. However, there is no consensus about the optimal form of simple SFMs because there has been a lack of direct tests against observed stratigraphy in well‐constrained test cases. Specifically, the extent to which SFM behaviour over geologic space and timescales should be governed by local (downslope sediment flux depends only on local slope) versus nonlocal (sediment flux depends on factors other than local slope, such as the history of slopes experienced along a transport pathway) processes is currently unclear. Here, we develop a nonlocal, nonlinear SFM that incorporates slope bypass and long‐distance sediment transport, both of which have been previously identified as important model components but not thoroughly tested. Our model collapses to the local, linear model under certain parameterizations such that best‐fit parameter values can indicate optimal model structure. Comparing 2‐D implementations of both models against seven detailed seismic sections from the Southeast Atlantic Margin, we invert the stratigraphic data for best‐fit model parameter values and demonstrate that best‐fit parameterizations are not compatible with the local, linear diffusion model. Fitting observed stratigraphy requires parameter values consistent with important contributions from slope bypass and long‐distance transport processes. The nonlocal, nonlinear model yields improved fits to the data regardless of whether the model is compared against only the modern bathymetric surface or the full set of seismic reflectors identified in the data. Results suggest that processes of sediment bypass and long‐distance transport are required to model realistic passive margin stratigraphy and are therefore important to consider when inverting the stratigraphic record to infer past perturbations to source regions. [ABSTRACT FROM AUTHOR]

Bedrock landslides shape topography and mobilize large volumes of sediment. Yet, interactions between landslide‐produced sediment and fluvial systems that together govern large‐scale landscape evolution are not well understood. To explain morphological patterns observed in steep, landslide‐prone terrain, we explicitly model stochastic landsliding and associated sediment dynamics. The model accounts for several common landscape features such as slope frequency distributions, which include values in excess of regional stability limits, quasi‐planar hillslopes decorated with straight, closely spaced channel‐like features, and accumulation of sediment in valley networks rather than on hillslopes. Stochastic landsliding strongly affects the magnitude and timing of sediment supply to the fluvial system. We show that intermittent sediment supply is ultimately reflected in topography. At dynamic equilibrium, landslide‐derived sediment pulses generate persistent landscape dynamism through the formation and breaching of landslide dams and epigenetic gorges as landslides force shifts in channel positions. Our work highlights the importance of interactions between landslides and sediment dynamics that ultimately control landscape‐scale response to environmental change. Plain Language Summary: Landslides are natural hazards that predominantly occur where geologic forces have created steep, rugged terrain. Yet from a geologic perspective, landsliding is also one of the forces that contributes to shaping the landscape. Gaining insight into the role of landslides in shaping topography is critical to better understand the interplay between the factors driving long‐term landscape evolution, and ultimately controlling landslide hazard. Here, we use a computer simulation model to study how landslides help shape terrain. Simulating landslides enables us to identify the origins of several common features in mountainous topography. Model results illustrate how repeated landsliding can cause rivers and ridgelines to gradually but continually shift position over time, such that some mountain drainages grow while others shrink. This type of computer model, which describes how topography and landsliding influence one another, can help the scientific community understand the response of steep terrain to environmental changes, such as those related to climate and earthquakes. Key Points: Over geological timescales, landslides produce diagnostic topographic signatures that can be detected in real‐world topographic dataStochastic landsliding results in slopes exceeding stability angles and quasi‐planar hillslopes decorated with closely spaced channelsThe combination of landslide stochasticity and channel‐hillslope feedback mechanisms causes persistent landscape dynamism [ABSTRACT FROM AUTHOR]

Land-surface parameters derived from digital land surface models (DLSMs) (for example, slope, surface curvature, topographic position, topographic roughness, aspect, heat load index, and topographic moisture index) can serve as key predictor variables in a wide variety of mapping and modeling tasks relating to geomorphic processes, landform delineation, ecological and habitat characterization, and geohazard, soil, wetland, and general thematic mapping and modeling. However, selecting features from the large number of potential derivatives that may be predictive for a specific feature or process can be complicated, and existing literature may offer contradictory or incomplete guidance. The availability of multiple data sources and the need to define moving window shapes, sizes, and cell weightings further complicate selecting and optimizing the feature space. This review focuses on the calculation and use of DLSM parameters for empirical spatial predictive modeling applications, which rely on training data and explanatory variables to make predictions of landscape features and processes over a defined geographic extent. The target audience for this review is researchers and analysts undertaking predictive modeling tasks that make use of the most widely used terrain variables. To outline best practices and highlight future research needs, we review a range of land-surface parameters relating to steepness, local relief, rugosity, slope orientation, solar insolation, and moisture and characterize their relationship to geomorphic processes. We then discuss important considerations when selecting such parameters for predictive mapping and modeling tasks to assist analysts in answering two critical questions: What landscape conditions or processes does a given measure characterize? How might a particular metric relate to the phenomenon or features being mapped, modeled, or studied? We recommend the use of landscape- and problem-specific pilot studies to answer, to the extent possible, these questions for potential features of interest in a mapping or modeling task. We describe existing techniques to reduce the size of the feature space using feature selection and feature reduction methods, assess the importance or contribution of specific metrics, and parameterize moving windows or characterize the landscape at varying scales using alternative methods while highlighting strengths, drawbacks, and knowledge gaps for specific techniques. Recent developments, such as explainable machine learning and convolutional neural network (CNN)-based deep learning, may guide and/or minimize the need for feature space engineering and ease the use of DLSMs in predictive modeling tasks. [ABSTRACT FROM AUTHOR]