Search

Progress in better understanding and modeling Earth surface systems requires an ongoing integration of data and numerical models. Advances are currently hampered by technical barriers that inhibit finding, accessing, and executing modeling software with related datasets. We propose a design framework for ‘Data Components’: software packages that provide access to particular research datasets or types of data. Because they use a standard interface based on the Basic Model Interface (BMI), Data Components can function as plug-and-play components within modeling frameworks to facilitate seamless data-model integration. To illustrate the design and potential applications of Data Components and their advantages, we present several case studies in Earth surface processes analysis and modeling. The results demonstrate that the Data Component design provides a consistent and efficient way to access heterogeneous datasets from multiple sources, and to seamlessly integrate them with various models. This design supports the creation of open data-model integration workflows that can be discovered, accessed, and reproduced through online data sharing platforms, which promotes data reuse and improves research transparency and reproducibility.

Poster summarizing elements of the CSDMS OpenEarthscape project., Supported by NSF 2104102

This is a poster given virtually at the 2020 European Geosciences Union annual meeting, 4-8 May, 2020. Abstract: Marine stratigraphy contains time-resolved information about the erosion of continents and its tectonic and climatic drivers. Quantitatively inverting marine stratigraphy for long-term terrestrial erosion histories requires numerical models that encompass the entire source-to-sink (S2S) system. Because inversion schemes require many model realizations to constrain free parameters against a misfit function, S2S models must be efficient (both in terms of allowing large time steps and scaling well for large problems) and have only a few parameters. Accordingly, most previous S2S models have treated seafloor evolution as a diffusion problem where sediment flux depends linearly on local topographic gradient. Such approaches have shown success in shallow marine settings like the continental shelf. However, they are less likely to apply to deeper marine environments where large deposits are observed and where nonlocal sediment transport processes (e.g., turbidity currents or marine debris flows) dominate sediment fluxes. We present a unified modeling approach for coupling terrestrial and marine erosion, sediment transport, and deposition from the continent to the abyssal plain. Our model is based on the erosion-deposition family of models, where sediment flux is tracked across the landscape and seascape. Above sea level, erosion and deposition depend on river discharge, local slope, and sediment flux. Below sea level, local slope and sediment flux drive topographic change. The equations governing the terrestrial and marine domains take the same basic form such that a single semi-implicit numerical solution based on Gauss-Seidel iteration can be used across the whole S2S system. The solution scheme is near O(N) complexity in that the number of iterations required typically does not increase significantly with increasing grid resolution. The S2S model contains only five total parameters. We show preliminary analytical and numerical results and sensitivity analyses, and discuss the applicability of the model for the efficient inversion of deep marine stratigraphic data.

This is a poster given remotely at the Community Surface Dynamics Modeling System online annual meeting (May 17-21, 2021). Abstract: Source-to-sink (S2S) studies seek to explicitly link the denudation of continents with the building of basin stratigraphy in an effort to infer tectonic and climatic drivers of surface change. Quantitative models for S2S systems must incorporate geomorphic processes at both source and sink, yet more effort has been devoted to developing landscape evolution models in source terranes than equivalent models for sedimentation in marine basins. In particular, most marine sedimentation models use local linear diffusion approximations for sediment transport, which have been shown to yield reasonable stratigraphy in shallow marine environments but struggle to reproduce diagnostic features of deep marine deposits. The lack of model predictive power in deep marine environments precludes the full closure of S2S sediment budgets. We present a model for marine sedimentation with two simple modifications allowing non-local sediment transport: 1) a mechanism for sediment bypass on steep topographic slopes, and 2) a parameter allowing long-distance transport over vanishingly gentle slopes. We use Bayesian inference techniques to constrain four model parameters against the stratigraphy of the Orange Basin in southern Africa. We compare modeled against observed stratigraphy over 130 Ma of margin evolution. Our best-fit simulations capture the broad structure of the observed record, and imply non-negligible roles for both non-local model elements: sediment bypass at steep slopes and long-distance runout over gentle slopes. Residual misfit between our best-fit simulations and the stratigraphic data indicate that additional components of transport dynamics—likely hemipelagic sedimentation, grain size variations, or ocean bottom currents—might be required to achieve the longest transport distances observed in the sedimentary record. Results suggest that full closure of Earth’s sediment mass balance for S2S studies requires moving beyond local diffusion approximations, even at the longest timescales. Relatively simple modifications to modeled transport dynamics can lead to better agreement between modeled and observed stratigraphy, and may enable improved inference of landscape perturbations from the stratigraphic record.

This is a presentation given at the 2022 Geological Society of America annual meeting (October 9-12, 2022). It summarizes Shobe et al. (2022; Basin Research). Abstract: Passive margin stratigraphy holds time-integrated records documenting the evolution of landscapes that are long gone. It is probably possible to read the stratigraphic record by inverting coupled landscape evolution models and stratigraphic forward models (SFMs) for past landscape boundary conditions (e.g., climatic/tectonic perturbations). One hurdle is that there is not broad agreement on the optimal form of the SFMs we use to simulate marine sediment transport and the development of the offshore stratigraphic record over geologic time; there has been a lack of tests against observed stratigraphy in well-constrained environments. Specifically, it is not clear whether SFMs should be governed primarily by local (sediment transport rate depends only on local bathymetry) or nonlocal (transport rate depends on the history of bathymetry experienced along a transport pathway) sediment transport processes. Here we develop a simple nonlocal SFM, which incorporates the long-term average effects of nonlocal processes like turbidity currents and marine debris flows, and test it against a commonly used local model to determine which approach provides the best fit to stratigraphy in a well-constrained natural experiment. We invert for optimal model form by comparing calibrated implementations of both models against seven seismic sections from the Southeast Atlantic Margin representing 130 Ma of sediment accumulation. Results suggest that observed stratigraphy is not compatible with the local SFM and that fitting observed stratigraphy requires strong contributions from slope bypass and long-distance transport processes. This outcome holds true in cases where the model is only compared to the observed modern bathymetric surface, as well as cases where all seismic reflectors are considered. Processes of sediment bypass and long-distance transport may be essential to modeling realistic passive margin stratigraphy, and incorporating such processes into our inversion models may therefore clarify source-region dynamics inferred from the sedimentary record.

Researchers working with Earth surface processes modeling nearly always work with datasets. However, their productivities often suffer because these datasets are difficult to access and use. To help solving this problem, we developed data components to ease the burden of working with data and coupling them with numerical models. Data components can provide a consistent way to access and retrieve various types of datasets (e.g., time series, raster grid, and multidimensional space-time data). Each data component effectively “wraps” a dataset with Basic Model Interface (BMI). BMI is a set of standard query and control functions that can be added to a model or a dataset. Using the BMI interface for datasets makes them easier to interact with BMI-enabled numerical models. To demonstrate the capability of data components, we present an example Jupyter Notebook, in which several data components were used to download topography and soil datasets to calculate the hourly landslide susceptibility when Hurricane Maria made its landfall in Puerto Rico in 2017. This Jupyter Notebook can also be executed using CSDMS@HydroShare, which is a cloud-based platform to support reproducible research. Those interested can try out this Jupyter Notebook example at https://www.hydroshare.org/resource/df5fa2f5d1b74be4bf0a049e1e59889c/.

This is a poster presented at the 2022 American Geophysical Union annual meeting (December 12-16). It summarizes Shobe et al. (2022; Basin Research). Abstract: One way to decipher the stratigraphic record to reveal information about surface processes is through inverse analysis using forward models. In this approach, a forward surface process model that generates stratigraphy is run many times with varying parameter values, and the output stratigraphy is compared against observations. The parameter values that generate best-fit stratigraphy may then tell us something about surface processes. Inverse analysis of forward models is often used to infer past forcings to landscapes, but it can also be used to make inferences about process dynamics. We use inverse analysis of forward models to ask: How should we model seascape evolution and the development of passive margin stratigraphy over geologic time? Understanding the evolution of passive margins is essential because they contain Earth’s longest and most complete records of landscape change on adjacent continents. Yet there is as yet no consensus about the process representations that should make up a simple but complete seascape evolution model. Should margin evolution be governed by processes where downslope sediment flux depends only on local slope? Or should nonlocal processes where sediment flux depends on factors other than local slope, such as the history of slopes experienced along a transport pathway, play a major role? We conduct a data-driven inversion exercise using seven stratigraphic sections from the Southeast Atlantic Margin to assess the importance of nonlocal transport processes to building passive margin stratigraphy. We compare two mathematical models for seascape evolution: one that is purely local and one that incorporates nonlocal transport. The two models can converge through parameter changes alone, such that data-driven inversion can determine quantitatively the optimal process representation. Producing observed stratigraphy requires significant contributions from nonlocal processes like slope bypass and long-distance sediment transport; this result is true regardless of whether only the modern bathymetry or the full stratigraphic record is used to constrain the inversion. Results suggests that nonlocal processes dominate the development of passive margin stratigraphy and are essential ingredients in models used to infer past landscape dynamics from the stratigraphic record.