[论文解读] Time-Varying Causal Treatment for Quantifying the Causal Effect of Short-Term Variations on Arctic Sea Ice Dynamics

Akila Sampath, Vandana Janeja+1 more|Open MIND|Jan 25, 2026

Arctic and Antarctic ice dynamics被引用 0

一句话总结

本文提出 KGCM-VAE，一种知识引导的因果 VAE，使用基于 MMD 的潜在变量去混淆和因果邻接解码器来估计短期海洋驱动因素（SSH 调制处理）对北极海冰厚度的时变因果效应，相较基线在 PEHE 上表现更优。

ABSTRACT

Quantifying the causal relationship between ice melt and freshwater distribution is critical, as these complex interactions manifest as regional fluctuations in sea surface height (SSH). Leveraging SSH as a proxy for sea ice dynamics enables improved understanding of the feedback mechanisms driving polar climate change and global sea-level rise. However, conventional deep learning models often struggle with reliable treatment effect estimation in spatiotemporal settings due to unobserved confounders and the absence of physical constraints. To address these challenges, we propose the Knowledge-Guided Causal Model Variational Autoencoder (KGCM-VAE) to quantify causal mechanisms between sea ice thickness and SSH. The proposed framework integrates a velocity modulation scheme in which smoothed velocity signals are dynamically amplified via a sigmoid function governed by SSH transitions to generate physically grounded causal treatments. In addition, the model incorporates Maximum Mean Discrepancy (MMD) to balance treated and control covariate distributions in the latent space, along with a causal adjacency-constrained decoder to ensure alignment with established physical structures. Experimental results on both synthetic and real-world Arctic datasets demonstrate that KGCM-VAE achieves superior PEHE compared to state-of-the-art benchmarks. Ablation studies further confirm the effectiveness of the approach, showing that the joint application of MMD and causal adjacency constraints yields a 1.88\% reduction in estimation error.

研究动机与目标

• 推动超越相关性的必要性，以量化北极海冰动力学中的因果效应。
• 开发一个知识引导的因果建模VAE（KGCM-VAE），将物理约束与因果学习结合。
• 使在顺序北极数据中发现潜在结构依赖关系和反事实预测成为可能。
• 通过合成数据和真实数据证明在时变处理效应估计方面的鲁棒性与无偏性。

提出的方法

• 提出一种知识引导的处理生成方案，通过速度驱动的 Sigmoid 函数对平滑后的 SSH 进行调制。
• 嵌入物理关系（静水 SSH-海冰厚度与地转 SSH-速度的联系）以约束因果处理。
• 使用 Bi-GRU 编码器学习平衡的潜在表示，并使用带有邻接掩码的因果解码器来强制物理因果路径。
• 在潜在空间中引入最大均值差异（MMD），实现处理组和对照组协变量的平衡并促进协变量平衡。
• 以双潜在结果轨迹进行训练，以实现反事实估计并计算 ITE/PEHE。

Figure 1: The Architecture of the Knowledge-Guided Causal Model Variational Autoencoder (KGCM-VAE) integrates a balanced latent space $\mathbf{z}$ learned by the encoder with a knowledge-guided causal model in the decoder for robust counterfactual prediction and Individual Treatment Effect estimatio

实验结果

研究问题

• RQ1 如何在时间序列设置中量化短期海洋驱动因素对北极海冰厚度的因果效应？
• RQ2 知识引导的物理约束是否能提升北极气候动力学中因果效应的可 identifiability 与准确性？
• RQ3 潜在空间平衡（MMD）与因果邻接约束是否共同提升反事实预测与 PEHE？

主要发现

• KGCM-VAE 在样本外数据上达到更优的 PEHE（3.8159），优于基线。
• 联合的 MMD 与因果邻接约束在因果效应估计上表现最佳，消融研究中 PEHE 提升至 3.7939。
• 在不同处理滞后下，KGCM-VAE 的 PEHE 保持稳定（0.0480–0.0485），即便 RMSE 随滞后变化。
• 在真实数据验证中，SSH 与速度均显示出不同的因果信号：SSH 驱动海冰厚度的可观测变化，而仅速度未表现出同样的效应。

Figure 2: Real-world spatial analyses of factual and counterfactual outcomes. The panels display the temporal patterns of perturbed Ocean Velocity (Top left) and the corresponding counterfactual prediction of Sea Ice Thickness (Bottom left), alongside perturbed Sea Surface Height (Top right) and the