Making Sense of SMOS Soil Moisture Observations over Norway and Northern Areas, 2012

This project concerned studies to make more sense of the European Space Agency, ESA, SMOS (Soil Moisture and Ocean Salinity) soil moisture observations over Norway and Northern Areas. The SMOS satellite was launched in 2009 and is still working, providing data with relatively coarse spatial resolution (~43 km footprint), a limited instantaneous field of view (~1000 km across) and relatively infrequent revisits (1-3 days). The project used data assimilation ideas to improve the spatial and temporal information provided by SMOS to be more in line with user needs, i.e., offer higher rate, higher resolution, wider area “snap shot” information. For the data assimilation we used a variant of the Ensemble Kalman Filter (EnKF). The enhanced spatial resolution varied between 4 and 9 km (depending on model resolution) and the enhanced temporal resolution typically was 6 hours (the time step to update the analysis using SMOS data). As part of this study, the project evaluated the quality of satellite soil moisture observations and land surface model simulations over Norway, and assessed the NILU land data assimilation system in the context of studies of the hydrological cycle over land. This study addressed the challenge of using SMOS data to understand the hydrological cycle in Norway and Northern Areas, regions characterized by complex orography.本项目聚焦于优化欧洲空间局（European Space Agency, ESA）土壤湿度与海洋盐度卫星（Soil Moisture and Ocean Salinity, SMOS）在挪威及北部区域的土壤湿度观测数据解译工作。该卫星于2009年发射并持续在轨运行，其数据存在空间分辨率相对较粗（单像元瞬时覆盖范围约43千米）、瞬时视场有限（跨度约1000千米）以及重访周期较长（1至3天）的局限。本项目采用数据同化思路对SMOS的时空信息进行优化，使其更贴合用户需求，即提供更高频次、更高分辨率、覆盖范围更广的"快照"式数据。本次数据同化环节采用了集合卡尔曼滤波（Ensemble Kalman Filter, EnKF）的变体算法。优化后的空间分辨率介于4至9千米之间（取决于模式分辨率），增强后的时间分辨率通常为6小时，即使用SMOS数据更新分析场的时间步长。作为本研究的组成部分，项目团队还对挪威及北部区域的卫星土壤湿度观测数据与陆面模式模拟结果开展了质量评估，并针对陆地水文循环研究场景，对NILU陆面数据同化系统进行了性能校验。本研究针对以复杂地形为特征的挪威及北部区域，解决了利用SMOS数据解析该区域陆地水文循环的相关难题。