A toolbox for analyses of Satellite Derived Shorelines (SDS) generated from CoastSeg (CoastSat)
Currently there are some post-processing needs that users of coastseg have requested. Our team has grouped these needs into the following major categories.
- detect nonstationarity
- imputation techniques:
- linear interpolation
- regression (covariates)
- autoregression
- ML (covariates)
- Deep learning
- SDS variability quantification and classification
- nonstationarity detection
- linear trend
- nonlinear trend
- classify SDS time-series into pre-determined classes
- classify into custom (unsupervised) classes
- Isolating Noise: Techniques to identify and isolate noise within the shoreline data.
- Quality: Tools to assess the quality of the shoreline data.
- Clouds: Methods to deal with cloudy data segments and improve accuracy.
- monthy and yearly averages
- STL: Utilize STL (Seasonal-Trend decomposition using LOESS) for shoreline time series decomposition.
- Seasonal Shorelines: Analyze the periodic movement of shorelines.
- Time Series of Narrowest/Widest Beaches: Assess the evolution of beach widths over time.
- wavelets for nonstationary shoreline timeseries
- Shoreline Location from Situ Surveys (First): Primary method for ascertaining shoreline location.
- Shoreline Location with Lidar: Use of Lidar data for a more granular assessment.
- Animations: Dynamic representation of shoreline movements.
- Shorelines Color Coded by Time: Visualize temporal shoreline changes with color gradients.
- Beach Width Time Series (Provide Non-Erodible): Analyze the temporal changes in beach width.
- Non-Erodible Line for the Beach:
- Shoreline Position Coefficient of Variation:
- Estimate beach volume?