Miao et al. (2026) Reconstruction of Holocene Northern Hemisphere precipitation fields using paleoclimate data assimilation

Identification

• Journal: Scientific Data
• Year: 2026
• Date: 2026-01-07
• Authors: Fang Miao, JiangLin Wang, Hu Chang
• DOI: 10.1038/s41597-026-06551-6

Research Groups

• State Key Laboratory of Cryospheric Science and Frozen Soil Engineering, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China
• University of Chinese Academy of Sciences, Beijing, China
• State Key Laboratory of Ecological Safety and Sustainable Development in Arid Lands, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China

Short Summary

This study presents a novel, high-resolution reconstruction of Northern Hemisphere annual precipitation for the entire Holocene (12-0 ka BP) using paleoclimate data assimilation (PDA), demonstrating significantly improved reliability and spatiotemporal completeness compared to climate model simulations alone, particularly at mid-high latitudes.

Objective

• To reconstruct spatiotemporally-completed, temporally well-resolved, and highly-reliable Holocene (12-0 ka BP) annual precipitation fields over the Northern Hemisphere to clarify its spatiotemporal patterns and contextualize modern hydroclimate variability.

Study Configuration

• Spatial Scale: Northern Hemisphere, 3.75° × 3.75° grid resolution.
• Temporal Scale: 12-0 ka BP (12,000 years Before Present), 100-year temporal resolution.

Methodology and Data

• Models used:
  • Paleoclimate Data Assimilation (PDA) approach, specifically the Ensemble Optimal Interpolation (EnOI) algorithm.
  • Community Climate System Model version 3 (CCSM3) for TraCE 21ka transient simulation.
  • UK Met Office’s HadCM3 climate model for PMIP4 HadCM3 transient climate simulation.
• Data sources:
  • 2,421 Holocene annual precipitation records (pollen-based, WA-PLS_tailored reconstructions) from the LegacyClimate v1.0 dataset.
  • TraCE 21ka transient climate simulation (12-0 ka BP).
  • PMIP4 HadCM3 transient climate simulation (12-0 ka BP).
  • 20th Century Reanalysis (for bias correction and skill assessment).
  • 70 independent annual precipitation proxy records (e.g., δ¹⁸O from speleothems, ice accumulation from glacier ice) from the Hancock2023 compilation (for independent verification).

Main Results

• PDA-based reconstructions exhibited significantly improved reliability over model-only simulations, with mean correlation improvements of 29.84% to 41.17% in cross-validation against withheld proxy data.
• Independent verification against 70 external proxy records showed PDA-based reconstructions had better temporal consistency, with mean correlation improvements ranging from 24.94% to 159.26% compared to raw model simulations.
• The PDA (Mixed) reconstruction, utilizing a multi-model prior ensemble, consistently demonstrated superior performance across all validation metrics and latitude bands compared to single-model PDA assimilations and raw model simulations.
• Spatially, PDA-based reconstructions showed enhanced skill over prior estimates across 74.00% to 88.54% of Northern Hemisphere grid points, as quantified by the Continuous Ranked Probability Skill Score (CRPSS).
• Northern Hemisphere-averaged and zonal-averaged precipitation series from PDA reconstructions, models, and proxy records generally showed good agreement in overall trends, including a rapid increase from the Early to Middle Holocene (peaking around 6 ka BP) followed by a gradual decline.
• Performance disparities were most pronounced over North Africa and the Arabian Peninsula, indicating areas for future improvement.

Contributions

• Provides the first spatiotemporally-completed, temporally well-resolved (100-year), and highly-reliable hemispheric-scale (Northern Hemisphere) annual precipitation reconstruction for the entire Holocene (12-0 ka BP) using paleoclimate data assimilation.
• Represents a substantial advancement in applying the PDA approach to reconstruct large-scale precipitation fields over long timescales.
• Offers a crucial dataset for analyzing Holocene hydroclimate dynamics, contextualizing modern precipitation changes, and evaluating climate model performance in simulating long-term precipitation variability.
• Confirms the effectiveness of multi-model prior ensembles in enhancing PDA-based reconstruction skill for paleoclimate scenarios, extending previous findings from the instrumental era.
• Highlights the robustness of the PDA approach for hemispheric-scale Holocene precipitation reconstructions and its potential for global application.

Funding

• National Key R&D Program of China [Grant number: 2022YFF0801102]
• National Science Foundation of China (NSFC) project [Grant number: 42171044]
• Science and Technology Program in Gansu, China [Grant number: 23JRRA599]
• CAS “Light of West China” Program
• Holocene Climate Reanalysis (HCR) supported by Chinese Academy of Sciences (CAS-HCR)

Citation

@article{Miao2026Reconstruction,
  author = {Miao, Fang and Wang, JiangLin and Chang, Hu},
  title = {Reconstruction of Holocene Northern Hemisphere precipitation fields using paleoclimate data assimilation},
  journal = {Scientific Data},
  year = {2026},
  doi = {10.1038/s41597-026-06551-6},
  url = {https://doi.org/10.1038/s41597-026-06551-6}
}