Sampling depth of L-band radiometer measurements of soil moisture and freeze-thaw dynamics on the Tibetan Plateau

Sampling depth of L-band radiometer measurements of soil moisture and freeze-thaw dynamics on the Tibetan Plateau
复制标题

青藏高原土壤水分和冻融动态L波段辐射计测量采样深度

DOI:
10.1016/j.rse.2019.03.029
复制
发表时间:
2019
影响因子:
13.5
通讯作者:
Zhongbo Su
Zhongbo Su
中科院分区:
工程技术1区
文献类型:
--
作者:
Donghai Zheng;Xin Li;Xin Wang;Zuoliang Wang;Jun Wen;Rogier van der Velde;Mike Schwank;Zhongbo Su

文献摘要

被引文献

相似文献

了解微波辐射测量的准确采样深度对于量化卫星土壤湿度产品的性能和评价其适用性至关重要。本文研究了青藏高原冻融土壤条件下L波段微波辐射的采样深度(δSM)。利用ELBARA-Ⅲ型辐射计在西藏某草甸观测点采集了两年的日亮温(TBp)观测资料,观测时间间隔为30 min。同时测量土壤温度和液态水体积含量(θliq)的垂直剖面,深度可达地表以下1 m。利用τ-ω发射模型评估了不同深度下测量的θ liq对微波发射模拟的影响,其中冻融土壤的介电常数由四相介电混合模型估算。发现:(1)有效温度的取样深度取决于θliq的大小,由于θliq的季节性,估计在冷干期和暖湿期的平均取样深度分别约为50和15 cm;(2)在2.5cm处测定了冷期和暖期冻融土壤的δ SM,比常用的θ liq测量深度浅(即5厘米),用于遍布地球仪的现场监测网络; 3)在δ SM估计值为2.5cm时,θ liq测量值的TBP模拟结果的无偏均方根误差较低,水平极化和垂直极化的模拟值分别比5cm土壤深度的θ liq模拟值高14%(3.16K)和22%(3.36K); 4)用ELBARA-III测量的垂直极化TBb 2的单通道算法反演的θ liq与2.5cm土壤深度的θ liq比5cm土壤深度的θ liq更接近。这些研究结果是至关重要的发展战略的校准/验证,以及基于卫星的土壤水分产品依赖于L波段辐射测量的应用。
Knowing the exact sampling depth of microwave radiometry is essential for quantifying the performance and appreciation of the applicability of satellite soil moisture products. We investigate in this study the sampling depth (δSM) of the L-band microwave emission under frozen and thawed soil conditions on the Tibetan Plateau. Two years of diurnal brightness temperature (TBp) measurements at a time interval of 30 min are collected by the ELBARA-III radiometer deployed at a Tibetan meadow site. Vertical profiles of soil temperature and volumetric liquid water content (θliq) are measured simultaneously at soil depths up to 1 m below the surface. The impact of theθliqmeasured at different depths on the microwave emission simulations is assessed using theτ-ωemission model, whereby the permittivity of frozen and thawed soil is estimated by the four-phase dielectric mixing model. It is found that: 1) the sampling depth for the effective temperature depends on the magnitude ofθliq, and is estimated to be, on average, about 50 and 15 cm for the cold dry and wet warm period, respectively, because of the seasonality inθliq; 2) theδSMis determined at 2.5 cm for both frozen and thawed soil conditions during both cold and warm periods, which is shallower than the commonly usedθliqmeasurement depth (i.e. 5 cm) adopted for the in-situ monitoring networks across the globe; 3) theTBpsimulations performed with theθliqmeasurements taken at the estimatedδSMof 2.5 cm result in lower unbiased root mean squared errors, about 14% (3.16 K) and 22% (3.36 K) for the horizontal and vertical polarizations respectively, in comparison to the simulations with theθliqmeasurements taken from 5 cm soil depth; and 4) theθliqretrieved with the single channel algorithm from the ELBARA-III measured vertically polarizedTBpare in better agreement with theθliqmeasured at 2.5 cm than the one measured at 5 cm. These findings are crucial for developing strategies for the calibration/validation as well as the application of satellite based soil moisture products relying on the L-band radiometry.