This study focused on the production of 18-year global root zone soil moisture (RZSM) by the joint land surface data assimilation using the satellite observed leaf area index (LAI) and surface soil moisture (SSM). The impact of the assimilation on RZSM, LAI, and other key surface variables was also assessed. The multilayer diffusion scheme, biomass and CO₂ interactive scheme, and the simplified extended Kalman filter were applied in the model. It was found that the assimilation could effectively reduce the biases in LAI, and that the diverse regional effects on RZSM were varied with seasons, soil wetness, error covariance in the assimilation, and water transfer in the model. A downward increase of the RZSM pattern (< ∼ 0.03 m³ m^-3) was found in vegetated regions with low to moderate soil wetness because of the reduced LAI by the assimilation. A general upward change of RZSM (within ∼ ±0.01 m³ m^-3) was found in dry desert regions due to the assimilation of SSM. The evaluation for the central South America shows that the assimilation improved the correlation for SSM (0.9 to 0.91) and significantly reduced the mean biases of LAI (∼ 40%). Positive impacts on day/night land surface temperature (LST) were identified to be mostly through the RZSM and LST coupling, with the improvements in the range of ±1 or 2 K. The slight adverse impact of LAI over the Amazon forests had no degradations to RZSM and LST. The assessment of the impact on water, energy, and carbon cycles over France revealed that the strongest/weakest change was found in LAI (-6.3%)/deep layer soil water index (0.03%). Ecosystem respiration, sensible heat, and evapotranspiration had relatively large changes. The underlying mechanism of the impact supports the global analysis results, indicating that the joint assimilation is beneficial for drought monitoring and heatwave detection.

中文翻译：

---

卫星观测的 LAI 和土壤水分对全球根区土壤水分产生及其对地表和生态系统变量的影响的联合同化


本研究的重点是通过使用卫星观测叶面积指数 （LAI） 和表层土壤水分 （SSM） 的联合地表数据同化来产生 18 年全球根区土壤水分 （RZSM）。还评估了同化对 RZSM、LAI 和其他关键表面变量的影响。模型中应用了多层扩散方案、生物量和 CO₂ 交互方案以及简化的扩展卡尔曼滤波。研究发现，同化可以有效减少 LAI 中的偏差，并且对 RZSM 的不同区域影响随季节、土壤湿度、同化误差协方差和模型中的水分转移而变化。由于同化降低了 LAI，因此在土壤湿度低至中等的植被区发现 RZSM 模式向下增加 （< ∼ 0.03 m^{3 m-3}）。由于 SSM 的同化，在干燥的沙漠地区发现了 RZSM 的普遍向上变化 （在 ∼ ±0.01 m^{3 m-3} 以内）。对中南美洲的评估表明，同化提高了 SSM 的相关性 （0.9 至 0.91） 并显着降低了 LAI 的平均偏差 （∼ 40%）。对昼/夜地表温度 （LST） 的积极影响主要通过 RZSM 和 LST 耦合确定，在 ±1 或 2 K 范围内有所改善。LAI 对亚马逊森林的轻微不利影响对 RZSM 和 LST 没有降解。对法国对水、能源和碳循环影响的评估表明，LAI （-6.3%）/深层土壤水分指数 （0.03%） 的变化最强/最弱。生态系统呼吸、显热和蒸散变化相对较大。 影响的底层机制支持了全局分析结果，表明联合同化有利于干旱监测和热浪检测。