From Bands to Depth: Understanding Bathymetry Decisions on Sentinel-2

TL;DR

This paper investigates how a Swin-Transformer based U-Net model infers bathymetry from Sentinel-2 data, analyzing spectral importance, model explanations, and robustness to improve cross-region depth estimation.

Contribution

It introduces A-CAM-R for model explanation, evaluates spectral importance, and provides practical guidance for robust satellite-derived bathymetry across regions.

Findings

Spectral importance aligns with shallow water optics.

A-CAM-R effectively localizes model reliance on evidence.

Cross-region inference degrades with depth, especially in bimodal distributions.

Abstract

Deploying Sentinel-2 satellite derived bathymetry (SDB) robustly across sites remains challenging. We analyze a Swin-Transformer based U-Net model (Swin-BathyUNet) to understand how it infers depth and when its predictions are trustworthy. A leave-one-band out study ranks spectral importance to the different bands consistent with shallow water optics. We adapt ablation-based CAM to regression (A-CAM-R) and validate the reliability via a performance retention test: keeping only the top-p% salient pixels while neutralizing the rest causes large, monotonic RMSE increase, indicating explanations localize on evidence the model relies on. Attention ablations show decoder conditioned cross attention on skips is an effective upgrade, improving robustness to glint/foam. Cross-region inference (train on one site, test on another) reveals depth-dependent degradation: MAE rises nearly linearly with depth, and bimodal depth distributions exacerbate mid/deep errors. Practical guidance follows: maintain wide receptive fields, preserve radiometric fidelity in green/blue channels, pre-filter bright high variance near shore, and pair light target site fine tuning with depth aware calibration to transfer across regions.