Overview
========
Reconstructing past climate states is crucial for improving our understanding of natural climate variability. 
Proxy records and climate models are two primary sources for reconstructing past climate states. 
Proxy records document the long-term evolution of past climate states, but they are characterized by diverse sources, significant noises, and low temporal and spatial resolution. 
In contrast, climate models provide simulations with high spatial and temporal resolution, but often exhibit considerable biases in representing regional and small-scale climate features. 
Paleoclimate data assimilation (PDA) combines proxy records with climate models to provide optimal estimates of past climate states with high spatial and temporal resolution.
This document provides the reader the development of PDA, including:

- :doc:`Offline PDA <../Offline/index>`: Perform offline PDA using pre-existed climate simulations.

- :doc:`Online PDA <../Online/index>`: Perform online PDA with a deep learning-based surrogate model.

- **Multi-scale DA**: A methodology that bridges different scales of climate data to improve the accuracy and resolution of assimilation results.

- **Super-resolution Post-processing**: A technique used to enhance the resolution of climate data beyond the original data's native resolution.

- **Nonlinear Proxy System Models (PSMs)**: Constructing models that simulate the relationship between climate variables and proxy data, such as ice cores and tree rings.

- ...

We appreciate your interest in our project and look forward to working together to advance paleoclimate research.