Repository navigation
Home
NOTE TO ALL: CHESS is in active development. Maker sure to check for version updates.
Thank you for being a part of our exciting journey as we unveil CHG's cutting-edge probabilistic tool suite, the Coastal Hazards, Engineering and Structures System (CHESS) library (CHESS-Library). Your interest and participation are truly appreciated, and we're genuinely excited to have you on board. As you explore this early distribution of our software, we want to share a few friendly reminders:
🚀 Passion Project: The entire development of this tool suite has been without a direct source of funding. Your enthusiasm and engagement make it all worthwhile!
🧪 Development Stage: While we've put the existing tools through extensive testing, remember that we're still in the development stage. Bugs might pop up based on inputs or application, but fear not—we're here to tackle them head-on.
🗺️ Limited Data Landscape: Currently, we're working with CHS native data, but don't let that hold you back! Custom modeling is absolutely supported; it just requires a little one-on-one setup. Let's make it happen together.
🤝 Your Input Matters: We value your feedback on various aspects—whether it's about the ease of use, missing physics or responses, data visualizations, or post-processing computations. Your insights help us shape this tool into something extraordinary.
Once again, thank you for being a part of this exciting beta testing phase. Together, we're paving the way for a future of robust and innovative probabilistic workflows. Let's make it an incredible journey!
- Getting Started: requirements, running the library script or the GUI, choosing a workflow, reporting bugs.
- GUI Quick Start: from an empty app to a hazard curve, with screenshots.
- Model Architecture: the six-step pipeline and what each workflow does with it.
- Model Inputs and Model Outputs: input files, storm databases, tide files, and the project folder layout.
- PSE Hydraulic Responses: which responses apply to which structure, and the equations behind them.
- config_reference: every input field.
1. Coastal Hazard System (CHS) regional study modeling results (https://chs.erdc.dren.mil)
CHS is a unique data resource that spans probability space, was developed with high-fidelity modeling, is spatially uniform and dense on a national scale, and includes aleatory and epistemic uncertainty information necessary for determining probabilistic responses. Modeling results are reported through save points across regional studies. Save points include modeling results for both circulation (ADCIRC) and wave models (STWAVE/WAM/SWAN). Additionally, results hosted on CHS include different Sea Level Rise (SLR) scenarios and tidal conditions for tropical (TC) and/or extra-tropical (XC) events. Currently, CHESS supports the use of Base Conditions modeling results (peaks and timeseries) for the following regional studies: CHS-NACCS, CHS-SACS, and FEMA RiskMap studies. Additional CHS regional study statistical results can be downloaded (here) (Distributed Storm Weights & SWL Bias/Uncertainty) <-----Link
The following figure summarizes the CHS data download process; 1) choose a regional study, 2) apply the Base Modeling filter, 3). Through CHS, you can choose the most suitable save point for your project. When selecting project-forcing data, consider the hazards you want to include (XC vs TC vs both) and the level of fidelity (Peaks vs Timeseries vs both). Currently, CHESS only supports base modeling conditions as a forcing input. Utilizing CHS data means any analysis will involve the full storm suite for a given regional study. The downloaded zip folder must contain Peaks and/or Timeseries files for both models associated with a savepoint (ADCIRC + Wave Model). For a more detailed guide on accessing and navigating CHS files/website, visit (https://chs.erdc.dren.mil/Content/documents/CHSQuickGuide.pdf).
Optional, for random tides. Created with create_tide_file.m from a NOAA station ID: an hourly two-column csv (Date, Prediction). Active stations: https://tidesandcurrents.noaa.gov/stations.html?type=Water+Levels. Details on Model Inputs.
Project conceptualization within the CHESS library is achived through 4 main components: 1) configuration file, 2) Protective System Elements (PSEs), 3) wave and water level storm forcing events, and 4) associated parameter aleatory and epistemic uncertainty. The configuration file holds project details, identifies directory paths to the other three components, and governs toolkit behavior. PSEs represent 1-D coastal structure transects. Storm forcing and associated uncertainties are downloaded from CHS as spatially distributed data containing peak and time series files.
The CHESS_Usage_Example.m script and the CHESS GUI run the following modules. Workflow pages: CHESS PROS (with CHESS PST and CHESS JPM), CHESS LCS and CHESS DET.
CHESS-PROS probabilistically estimates structure response hazards using response-based methods. Coastal structure responses include floodwall, levee, and rubble mound overtopping and volume discharge, levee and rubble mound runup, floodwall hydrodynamic and hydrostatic pressures, and rubble mound stone stability. CHESS is a component of the Coastal Hazards System (CHS https://chs.erdc.dren.mil/, Nadal-Caraballo et al. 2020) which is a national-scale initiative to quantify coastal storm hazards. Results from high-fidelity, physics-informed numerical modeling of coastal storm events spanning the practical probability space for the U.S. coastline are stored on an online database (CHS-DB). Synthetic TCs and XCs were modeled in a high-fidelity coupled hydrodynamic framework (Massey et al. 2012) for regional studies to produce coastal wave and water level responses. This was completed for multiple sea level rise conditions. Discrete storm weights (DSW) define TC storm probability and are used to estimate hazards. Both peak and timeseries values are available. CHESS can evaluate responses over entire timeseries within a probabilistic framework with relatively small computational costs. Tides can be either randomly sampled, applied as uncertainty, or applied as a skew tide to account for nonlinearities. Structure responses are computed for hundreds of thousands of storm realizations following the CHESS-JPM and CHESS-SST workflows. Exceedance probabilities of the structure responses themselves are estimated. CHESS-PROS maintains the high fidelity multivariate statistical and physical interdependencies between storm forcing parameters without assuming identical probabilities of forcing parameters and structure responses (Stehno and Melby, in review). Epistemic uncertainties associated with storm response and empirical equations are applied as confidence limits to the BE response hazards. Results from CHESS-PROS are used for coastal structure designs when designs require non-exceedance of a structure response hazard at a given probability.
CHESS-LCS computes time-dependent stochastic analysis of coastal structure responses. Current coastal structure responses include dune and beach morphology and rubble mound armor stone damage progression. CHESS is a component of the Coastal Hazards System (CHS https://chs.erdc.dren.mil/, Nadal-Caraballo et al. 2020) which is a national-scale initiative to quantify coastal storm hazards. Results from high-fidelity, physics-informed numerical modeling of coastal storm events spanning the practical probability space for the U.S. coastline are stored on an online database (CHS-DB). Synthetic TCs and XCs were modeled in a high-fidelity coupled hydrodynamic framework (Massey et al. 2012) for regional studies to produce coastal wave and water level responses. This was completed for multiple sea level rise conditions. Discrete storm weights (DSW) define TC storm probability and are used to estimate hazards. Both peak and timeseries values are available. CHESS can evaluate responses over entire timeseries within a probabilistic framework with relatively small computational costs. Tides can be either randomly sampled, applied as uncertainty, or applied as a skew tide to account for nonlinearities. Life cycles are generated from sampled storms, sampled using a Poisson distribution to create a life cycle reflective of the larger population of storm intensities and associated probabilities at the study location. TCs are sampled using DSWs and XCs are sampled either historically or from stochastic simulation using bivariate or multivariate Gaussian copulas. Storms may also be sampled by binning storms by intensity prior to sampling, then sampling from intensity bins to represent storm intensity probabilities more accurately throughout the entire life cycle. Aleatory uncertainties are accounted by stochastically creating hundreds of life cycles, which also ensure statistical convergence. Structure life-cycle responses can be used to estimate statistical structure reliability and performance.
CHESS-SST estimates extra-tropical cyclone (XC) storm response hazards. CHESS is a component of the Coastal Hazards System (CHS https://chs.erdc.dren.mil/, Nadal-Caraballo et al. 2020) which is a national-scale initiative to quantify coastal storm hazards. Results from high-fidelity, physics-informed numerical modeling of coastal storm events spanning the practical probability space for the U.S. coastline are stored on an online database (CHS-DB). Historical XCs were modeled in a high-fidelity coupled hydrodynamic framework (Massey et al. 2012) for regional studies to produce coastal wave and water level responses. This was completed for multiple sea level rise conditions. Both peak and timeseries values are available. CHESS can evaluate responses over entire timeseries within a probabilistic framework with relatively small computational costs. Tides can be either randomly sampled, applied as uncertainty, or applied as a skew tide to account for nonlinearities. CHESS-SST ingests modeled historical storm responses. Responses are bootstrapped sampled, including aleatory uncertainty, to encompass multiple sequences or life cycles of storm responses. Peaks-over-threshold is used to identify extreme events from sampling. These extreme events are fit to a generalized Pareto distribution (GPD), which captures the low-frequency tail of response hazards.
CHESS-JPM estimates tropical cyclone (TC) storm response hazards (Nadal-Caraballo and Melby 2014). CHESS is a component of the Coastal Hazards System (CHS https://chs.erdc.dren.mil/, Nadal-Caraballo et al. 2020) which is a national-scale initiative to quantify coastal storm hazards. Results from high-fidelity, physics-informed numerical modeling of coastal storm events spanning the practical probability space for the U.S. coastline are stored on an online database (CHS-DB). Synthetic TCs were modeled in a high-fidelity coupled hydrodynamic framework (Massey et al. 2012) for regional studies to produce coastal wave and water level responses. This was completed for multiple sea level rise conditions. Discrete storm weights (DSW) define TC storm probability and are used to estimate hazards. Both peak and timeseries values are available. CHESS can evaluate responses over entire timeseries within a probabilistic framework with relatively small computational costs. Tides can be either randomly sampled, applied as uncertainty, or applied as a skew tide to account for nonlinearities. TC responses, such as waves and water levels, are ingested into CHESS-JPM, aleatory uncertainty is applied to these storm responses using hundreds of thousands of storms, and associated DSWs are used to estimate best-estimate (BE) exceedance probabilities. Confidence limits are computed by applying epistemic uncertainties to BE hazard curves. Additional details about coastal storm uncertainties are in Gonzalez et al. (2019).
CHESS-DET computes the structure responses of every storm in the storm database once, as recorded: no replicates, no forcing uncertainty and no hazard curves. It is used to inspect individual storms, check a PSE before a probabilistic run, and compare Peaks with Timeseries for the same storm. See CHESS DET.
Please feel free to use this space to communicate any feedback/ideas. Any idea that needs further discussion (potential SoN) can be transitioned to a Discussion item on the repo.
- [1] Gonzalez, Nadal-Caraballo, Melby, Cialone, (2019). Quantification of uncertainty in probabilistic storm surge models: Literature review. ERDC/CHL SR-19-1. Vicksburg, MS: U.S. Army Engineer Research and Development Center.
- [2] Massey, Wamsley, and Cialone (2012): Coastal Storm Modeling – System Integration. In Proceedings of Solutions to Coastal Disasters Conference 2011, pp. 99–108.
- [3] Nadal-Caraballo, Melby, (2014). North Atlantic Coast Comprehensive Study Phase 1: statistical analysis of historical extreme water levels with sea level change. Technical Report ERDC/CHL TR-14-7, US Army Engineer R&D Center, Vicksburg, MS.
- [4] Nadal-Caraballo, Campbell, Gonzalez, Torres, Melby, Taflanidis, (2020). Coastal Hazards System: A Probabilistic Coastal Hazard Analysis Framework. In: Malvárez, G. and Navas, F. eds., Global Coastal Issues of 2020. Journal of Coastal Research, Special Issue No. 95, pp. 1211-1216.
- [5] Stehno, Melby (in review). Practical coastal structure stochastic response simulation using multi-variate hazards. Coastal Engineering (In review).