Seasonal assessment of cross-shore morphodynamic behaviour of wave-dominated beaches using data-driven analysis

Four years of data-driven analyses—Empirical Orthogonal Function (EOF) analysis and Canonical Correlation Analysis (CCA), were performed for North and South Goa beach profile datasets (2018–2021) to improve data processing and interpretation. The goal is to understand the significant cross-shore morphodynamic variations by evaluating dominating EOF patterns and discovering inherent correlations between beach-face slope, nearshore parameters, and environmentally sensitive sediment grain size parameters using CCA. The method separates the data's spatial and temporal dependencies, allowing beach-face changes to be characterised as a linear combination of space and time functions. After analysing data sets from 45 surveys conducted during the southwest, northeast, and pre-monsoon seasons, three statistically significant EOF modes were identified. The outcomes of the seasonally sampled data differed, with most variations in the beach-face configuration explained by the first three eigenfunctions, corresponding to the three largest eigenvalues. The largest eigenfunction reflects the average slope of the beach face. The slope of the second eigenfunction varies the most from the mean. The coastline is generally in an intermediate stage, with episodic beach morphology variation. The EOF investigation confirms the widely known hypothesis that the beach face is the most morphodynamically active zone, with North and South Goa beaches displaying year-round rhythmic variation, with the slope modulation most evident in the northeast monsoon season. The multivariate CCA strongly links wave parameters and beach-face slope. Similarly, a strong correlation is observed between the beach-face slope and sediment grain size parameters, particularly the sediment gradation indices, D 90 /D 10 and D 84 /D 1 6. The results indicate that the methods are reliable for analysing and interpreting beach-face variation behaviour and controlling physical processes and highlight the need for long-term spatio-temporal analysis and a guide for coastline management.