Investigating the Effect of Spatial Context on Multi-Task Sea Ice Segmentation

TL;DR

This study systematically examines how different spatial context sizes affect multi-task sea ice segmentation performance using deep learning, revealing that optimal receptive fields depend on data resolution and specific ice properties.

Contribution

It introduces a systematic analysis of receptive field effects in multi-task sea ice segmentation, guiding optimal spatial context selection based on data resolution and task characteristics.

Findings

Smaller receptive fields work best for high-resolution Sentinel-1 data.

Medium receptive fields improve stage of development segmentation.

Fusion of SAR and AMSR2 data enhances segmentation accuracy.

Abstract

Capturing spatial context at multiple scales is crucial for deep learning-based sea ice segmentation. However, the optimal specification of spatial context based on observation resolution and task characteristics remains underexplored. This study investigates the impact of spatial context on the segmentation of sea ice concentration, stage of development, and floe size using a multi-task segmentation model. We implement Atrous Spatial Pyramid Pooling with varying atrous rates to systematically control the receptive field size of convolutional operations, and to capture multi-scale contextual information. We explore the interactions between spatial context and feature resolution for different sea ice properties and examine how spatial context influences segmentation performance across different input feature combinations from Sentinel-1 SAR and Advanced Microwave Radiometer-2 (AMSR2) for multi-task mapping. Using Gradient-weighted Class Activation Mapping, we visualize how atrous rates influence model decisions. Our findings indicate that smaller receptive fields excel for high-resolution Sentinel-1 data, while medium receptive fields yield better performances for stage of development segmentation and larger receptive fields often lead to diminished performances. The fusion of SAR and AMSR2 enhances segmentation across all tasks. We highlight the value of lower-resolution 18.7 and 36.5 GHz AMSR2 channels in sea ice mapping. These findings highlight the importance of selecting appropriate spatial context based on observation resolution and target properties in sea ice mapping. By systematically analyzing receptive field effects in a multi-task setting, our study provides insights for optimizing deep learning models in geospatial applications.