The East Asian Monsoon (EAM) system's complex structure and extreme precipitation events pose significant economic threats to China. The EAM's interannual variability, manifesting as spatial precipitation variations between northern and southern China, is driven by intricate interactions between the atmosphere, land, ocean, and cryosphere. These interactions result in precipitation extremes (floods, droughts, heatwaves) with substantial socioeconomic consequences. Past EAM precipitation-induced flooding has caused billions of dollars in losses. Climate change projections suggest EAM intensification due to enhanced land-sea thermal contrast, leading to increased continental precipitation and heightened flood risk. Understanding the EAM's variability and its response to internal and external forcings is crucial for improving prediction skills. Previous research has focused on the roles of SST, El Niño-Southern Oscillation (ENSO), and land-atmosphere interactions. However, the role of soil moisture (SM) variability, a slowly evolving component of the Earth's climate system, needs further exploration. Soil moisture's inherent memory and influence on evapotranspiration, surface energy fluxes, and diabatic heating highlight its potential significance in modulating EAM precipitation. Studies have indicated SM's influence on atmospheric circulations, triggering Rossby wave trains and impacting remote climate fields. This study aims to quantify the relationship between SM meridional oscillations and EAM precipitation variability in China, clarifying the underlying physical mechanisms. We use a robust statistical tool (Coupled Manifold Technique) and regional climate model simulations constrained by observed SM data to achieve this.

Past research has extensively studied various aspects of the EAM, including its onset, meridional oscillations, and regional flooding/droughts. Zhou et al. (2021) highlighted the mediating role of SST on the spring SM's influence on summer EAM precipitation. Lv et al. (2019) showed an increase in maximum daily precipitation during different ENSO phases. The impact of oceanic forcing on EAM flood hazard assessment remains uncertain. Studies have demonstrated SM's impact on surface warming/cooling through sensible/latent heat fluxes, influencing diabatic heating and circulations. SM anomalies can also trigger circulation patterns and Rossby wave trains, impacting remote climate fields. Previous studies over the EAM domain have shown significant sensitivity of atmospheric circulations and precipitation to SM. Gao et al. (2019) showed the interaction between spring SM and EAM precipitation, while Dong et al. (2022) found a negative relationship between July SM and August precipitation. Bao et al. (2010) demonstrated that incorporating remotely sensed SM in atmospheric models can improve monsoon onset prediction. However, the quantitative assessment of SM's regional spatiotemporal variability and its control over local and remote climate remains limited due to data availability and model limitations.

The study employs datasets from 1981 to 2018, including soil moisture from ESA-ECV SM, precipitation data from CMA, and ERA5 reanalysis data. Empirical Orthogonal Function (EOF) analysis is applied to EAM precipitation to identify meridional oscillations. Pearson correlation is used to assess the relationship between the leading EOF mode and SM interannual variability. The Coupled Manifold Technique (CMT) quantifies the fraction of EAM precipitation variance forced by SM interannual variability. Composite analysis is performed using years with positive and negative soil moisture index (SMI) values to examine the impact of SM anomalies on surface energy fluxes, diabatic heating, and atmospheric circulations. Regional climate model simulations (RegCM4 coupled with the BATS land surface model) are conducted using two experiments: a control experiment (SM_CLIM) and a sensitivity experiment (SM_ECV) using observed SM data to assess the impact of SM meridional oscillations on EAM precipitation. The CMT methodology involves decomposing climate fields (Z and S) into free and forced components. The fraction of variance in Z forced by S is calculated. Monte Carlo simulations test the significance of results. EOF analysis decomposes climate parameters into uncorrelated eigenvectors, and CCA is used to scale principal components before Procrustes minimization. The study also analyzes divergent wind components using Helmholtz decomposition and vertically integrated moisture flux.

EOF analysis of EAM precipitation reveals a meridional oscillation pattern, explaining 48% of the total variance. A significant correlation (95% confidence) exists between the leading EOF mode and SM interannual variability, exhibiting a dipole pattern with negative correlations in the north and positive correlations in the south. CMT results show that SM interannual variability forces a significant (99% confidence) fraction of EAM precipitation, with higher variance ratios in northern and southern China. The leading mode of the forced precipitation exhibits a dipole pattern. A soil moisture index (SMI), based on the difference between area-averaged SM in north and south China, is strongly negatively correlated with the leading precipitation mode (-0.62). Composite analysis for positive and negative SMI composites reveals clear dipole patterns in SM and precipitation anomalies. Positive (negative) SM anomalies in the north (south) result in more (less) precipitation in northern (southern) China. SM meridional anomalies impact surface energy fluxes, boundary layer thickness, and vertical diabatic heating. The model simulations confirm the observed relationship between SM and EAM precipitation. The difference in standard deviation of daily total-column SM between SM_ECV and SM_CLIM simulations highlights regions of strong SM sensitivity. The model results show that the SMI composite differences fluctuate the near-surface energy balance, impacting precipitation patterns. During positive SMI composites, intensified westerlies and ascending motion over northern China result in above-normal precipitation while descending motion in the south suppresses precipitation. Conversely, during negative SMI composites, the circulation pattern shifts, resulting in enhanced precipitation in southern China and suppressed precipitation in northern China.

The findings demonstrate a significant influence of SM meridional oscillations on EAM precipitation patterns across China. Positive SM anomalies in the north lead to enhanced precipitation in northern China and suppressed precipitation in the south, and vice versa. The mechanism involves SM's control over surface energy fluxes, boundary layer processes, and large-scale atmospheric circulations. The results are consistent with previous studies highlighting SM's role in EAM precipitation variability. The study's approach of combining statistical analysis and process-oriented model simulations provides strong evidence for the SM-EAM precipitation relationship. The study's findings highlight the importance of SM as a crucial factor influencing EAM interannual predictability. This is particularly relevant for predicting drought and flood risks in the northern and southern river basins under the context of climate change.

This study demonstrates the significant impact of soil moisture meridional oscillations on the interannual variability of East Asian Monsoon precipitation across China. Soil moisture's influence on surface energy fluxes and atmospheric circulations drives the observed meridional precipitation patterns. This underscores the importance of incorporating soil moisture data into EAM prediction models to improve forecast skills and manage drought and flood risks. Future research should explore the combined effects of soil moisture and oceanic forcing on EAM variability and investigate the regional differences in sensitivity to these factors.

The study's limitations include potential biases in the regional climate model simulations due to boundary layer uncertainties and topographic effects. The model's representation of complex land-atmosphere interactions might not perfectly capture the intricate physical processes. The focus on soil moisture as a primary driver might overlook other contributing factors like oceanic forcing that might interact with soil moisture to influence EAM precipitation patterns. Further research is needed to address these limitations and provide a more comprehensive understanding of EAM dynamics.