Assessment of soil water conditions using microwave remote sensing data

摘要 土壤水分状况直接影响入渗、径流、蒸发等过程。管理水和能源循环。农业、水文、气象等多个环境领域与土壤水分条件密切相关。然而，尽管其重要性，理解土壤水分状况的时空变异性仍然是一个挑战。拟议研究方案的总体目标是改善对土壤水分条件的频繁时空分布的获取，以便更好地了解水循环。它将通过以下具体目标来实现：a)开发和分析多频被动和主动微波土壤水分反演方法；b)估计和分析多时相细尺度土壤水分含量；以及c)表征魁北克农田上的饱和区域。重点是根据被动/主动微波数据的特征(频率、空间分辨率)以及土壤粗糙度和植被的变化来提取和分析表层土壤水分、土壤水分剖面和饱和土壤深度。该研究计划将受益于2016年夏季(http://smapvex16-mb.espaceweb.usherbrooke.ca/).在马尼托巴省举行的SMAPVEX16-MB野外活动的巨大数据集它由土壤(水分、温度和粗糙度)和植被(类型、高度、含水率等)的地面测量组成。这些特征是与星载被动和主动微波数据(SMOS、SMAP、RADARSAT-2、哨兵1号、TerraSAR-X和ALOS-PALSAR)、被动/主动L波段传感器(PALS)的机载测量以及地面微波辐射计测量准同时收集的。提出的方法包括：被动和主动微波模型的反演，现有土壤水分分解方法的适应，不同尺度上土壤水分估计的协同研究，以及雷达模型/土壤模型/同化模型的组合用于表征土壤表面和一定深度的饱和农田。结果将使用地面测量进行验证。它们将对及早发现干旱、洪水和火灾以及发展预警系统非常有用。此外，通过使用Sentinel-1高时间频率合成孔径雷达数据，拟议的研究方案提供了将未来的RADARSAT星座任务(发射日期定于2018年7月)数据用于业务应用的有趣例子。

Summary Soil water conditions affect directly the processes (infiltration, runoff, evaporation, etc.) governing the water and energy cycles. Several environmental domains such as agriculture, hydrology, meteorology, are closely related to soil water conditions. However, despite its importance, the understanding of the spatial and temporal variability in soil water conditions still remains a challenge. The general objective of the proposed research program is to improve the access to frequent spatial and temporal distributions of soil water conditions for a better understanding of the water cycle. It will be achieved through the following specific objectives: a) Develop and analyze a multifrequency passive and active microwave soil moisture retrieval method; b) Estimate and analyze a multitemporal fine-scale soil moisture content; and c) Characterize the saturated areas over agricultural fields in Quebec. The focus is on the extraction and the analysis of surface soil moisture, soil moisture profiles and the depth of saturated soils with respect to the characteristics of passive/active microwave data (frequency, spatial resolution) and the changes in both soil roughness and vegetation. The research program will benefit from the huge dataset of SMAPVEX16-MB field campaign that took place in Manitoba in the summer 2016 (http://smapvex16-mb.espaceweb.usherbrooke.ca/). It is composed of ground measurements of soil (moisture, temperature and roughness) and vegetation (types, height, water content, etc.) characteristics which were collected quasi-simultaneously with spaceborne passive and active microwave data (SMOS, SMAP, RADARSAT-2, Sentinel-1, TerraSAR-X, and ALOS-PALSAR), airborne measurements of PALS (Passive/Active L-band Sensor), and ground microwave radiometer measurements. The proposed methodology includes: inversion of passive and active microwave models, adaptation of an existing soil moisture disaggregation method, synergistic studies for soil moisture estimation at different scales, and the combination of radar model/soil model/assimilation model for the characterisation of saturated agricultural fields at both the soil surfaces and along a certain depth. Results will be validated using ground measurements. They will be very useful for the early detection of droughts, floods, and fires and for the development of warning systems. In addition, with the use of Sentinel-1 high temporal frequency synthetic aperture radar data, the proposed research program presents interesting examples of using the future RADARSAT Constellation Mission (launch date scheduled for July 2018) data for operational applications.