Automated Classification of Estuarine Sub‐Depositional Environment Using Sediment Texture

Interpretation of unconsolidated Quaternary sedimentary core is difficult if key diagnostic features are obscured or not present, therefore traditional facies analysis challenging. However, sediment texture remains a universal attribute which can be used to interpret core. Here we present an automated classification workflow implements Extreme Gradient Boosting and Bayesian Optimization hyperparameters differentiate estuarine sub-depositional environments. We use 19 textural attributes, measured using laser particle size surface samples from the Ravenglass Estuary, Cumbria, northwest England, make unbiased environment zone. Two predictive models created presented evaluated suite evaluation metrics, confusion matrices, spatial understand their geological implications. Model 1 keeps all environments discrete has overall accuracy 68.96%. 2 merges related form inner-coarse outer-estuary zones 84.14%. Both have been applied data obtained at 5 cm intervals Holocene drilled through tidal bar in succession, NW classify palaeo environment. Predictive output suggests that consistently experienced inner estuary deposition; represented The here could datasets other marginal marine depositional systems enhance interpretation subsurface deposits. Ultimately, detailed interpretations ancient, buried deposits made derived analogous modern systems.