GeoPrior-Mamba: Structured Process Priors with Mamba for Fine-Resolution XCO2 Reconstruction

Key points

• Existing learning-based methods typically treat environmental covariates as ordinary numerical inputs and must therefore learn heterogeneous source-sink relationships largely from sparse supervision.1
• We introduce GeoPrior-Mamba, a multi-directional Mamba framework augmented with offline language-model-induced structured process priors.1
• Rather than using a language model to predict XCO2, we use it before training to organize relative process knowledge for biospheric uptake, ecosystem respiration, and anthropogenic emissions into deterministic prior tables.1
• Using OCO-2 observations from 2018-2020, GeoPrior-Mamba achieves an RMSE of 0.81 ppm and an R2 of 0.93 on held-out observations, reducing RMSE by 48.2% relative to CAMS background interpolation and by 3.1% relative to Trans-XCO2 under the same evaluation protocol.1

Sources (1)

• [1]GeoPrior-Mamba: Structured Process Priors with Mamba for Fine-Resolution XCO2 Reconstruction
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 05:11 AM

  Reconstructing fine-resolution column-averaged dry-air CO2 (XCO2) fields from sparse satellite observations requires models to infer spatial structure that is only weakly constrained by direct measurements.

  Existing learning-based methods typically treat environmental covariates as ordinary numerical inputs and must therefore learn heterogeneous source-sink relationships largely from sparse supervision.

Extractive summary: sentences quoted from the sources.