Context-Aware Multimodal Representation Learning for Spatio-Temporally Explicit Environmental Modelling
This work addresses the problem for environmental scientists and ecologists who need fine spatial detail and high temporal fidelity in Earth observation data, though it is incremental as it builds on existing foundation models with a novel fusion approach.
The paper tackles the limitation of existing Earth observation foundation models that operate at fixed spatial or temporal scales by proposing a representation learning framework that integrates multiple remote sensing modalities into a unified feature space at high spatio-temporal resolution, resulting in embeddings that encode ecologically meaningful patterns and support fine-scale analyses like Gross Primary Production modelling.
Earth observation (EO) foundation models have emerged as an effective approach to derive latent representations of the Earth system from various remote sensing sensors. These models produce embeddings that can be used as analysis-ready datasets, enabling the modelling of ecosystem dynamics without extensive sensor-specific preprocessing. However, existing models typically operate at fixed spatial or temporal scales, limiting their use for ecological analyses that require both fine spatial detail and high temporal fidelity. To overcome these limitations, we propose a representation learning framework that integrates different EO modalities into a unified feature space at high spatio-temporal resolution. We introduce the framework using Sentinel-1 and Sentinel-2 data as representative modalities. Our approach produces a latent space at native 10 m resolution and the temporal frequency of cloud-free Sentinel-2 acquisitions. Each sensor is first modeled independently to capture its sensor-specific characteristics. Their representations are then combined into a shared model. This two-stage design enables modality-specific optimisation and easy extension to new sensors, retaining pretrained encoders while retraining only fusion layers. This enables the model to capture complementary remote sensing data and to preserve coherence across space and time. Qualitative analyses reveal that the learned embeddings exhibit high spatial and semantic consistency across heterogeneous landscapes. Quantitative evaluation in modelling Gross Primary Production reveals that they encode ecologically meaningful patterns and retain sufficient temporal fidelity to support fine-scale analyses. Overall, the proposed framework provides a flexible, analysis-ready representation learning approach for environmental applications requiring diverse spatial and temporal resolutions.