Physics-Conditioned Synthesis of Internal Ice-Layer Thickness for Incomplete Layer Traces

TL;DR

This paper introduces a physics-conditioned neural network that synthesizes complete internal ice-layer thickness profiles from incomplete radar data, leveraging physical features and robust training to improve ice stratigraphy analysis.

Contribution

It presents a novel model combining geometric learning and transformers, capable of completing missing ice-layer data while ensuring physical plausibility and aiding downstream predictions.

Findings

Successfully reconstructs fragmented and missing ice layers.

Improves downstream deep-layer prediction accuracy.

Enables stable training with incomplete supervision without imputation.

Abstract

Internal ice layers imaged by radar provide key evidence of snow accumulation and ice dynamics, but radar-derived layer boundary observations are often incomplete, with discontinuous traces and sometimes entirely missing layers, due to limited resolution, sensor noise, and signal loss. Existing graph-based models for ice stratigraphy generally assume sufficiently complete layer profiles and focus on predicting deeper-layer thickness from reliably traced shallow layers. In this work, we address the layer-completion problem itself by synthesizing complete ice-layer thickness annotations from incomplete radar-derived layer traces by conditioning on colocated physical features synchronized from physical climate models. The proposed network combines geometric learning to aggregate within-layer spatial context with a transformer-based temporal module that propagates information across layers to encourage coherent stratigraphy and consistent thickness evolution. To learn from incomplete supervision, we optimize a mask-aware robust regression objective that evaluates errors only at observed thickness values and normalizes by the number of valid entries, enabling stable training under varying sparsity without imputation and steering completions toward physically plausible values. The model preserves observed thickness where available and infers only missing regions, recovering fragmented segments and even fully absent layers while remaining consistent with measured traces. As an additional benefit, the synthesized thickness stacks provide effective pretraining supervision for a downstream deep-layer predictor, improving fine-tuned accuracy over training from scratch on the same fully traced data.