Conference Agenda

2018 Dragon 4 Symposium

Session
WS#3 ID.32397: CAL/VAL of Microwave Data
Time:
Wednesday, 20/Jun/2018:
8:30am - 10:00am

Session Chair: Prof. Massimo Menenti
Session Chair: Prof. Xin Li
Workshop: Hydrology & Cryosphere
College of Geomatics - Room 509

Presentations
Oral
ID: 302 / WS#3 ID.32397: 1
Oral Presentation
Hydrology & Cryosphere: 32397 - Calibration and Validation of Microwave Remote Sensing Data for Water Cycle Research

Monitoring Vegetation and Soil Moisture from SMOS Data

Jiancheng Shi1, Qian Cui2

1State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, CAS.; 2Information Center, Ministry of Water Resources, China

Vegetation and soil moisture are key parameters in the studies of global water and carbon cycles. In this study, based on the commonly used zero-order radiative transfer model ( model), a two-step approach for retrieving vegetation optical depth (VOD) and soil moisture using only SMOS H-polarized multi-angular measurements was presented. At a first step, VOD is estimated by minimizing the soil signal and separating the vegetation signal from the multi-angular brightness temperature. In the retrieval, the angular feature of soil emission is used and the VOD is retrieved directly from the refined H-polarized multi-angular brightness temperature without any field correction or auxiliary soil or vegetation data. This retrieved algorithm is first validated by theoretical modeling and experimental data. The results demonstrate that VOD can be reliably estimated using this algorithm. The retrieved VOD is then compared with aboveground biomass, which shows strong correlation. Global mean VOD for the years of 2010 to 2011 generally shows a clear global pattern and corresponds well to the land cover types. The VOD of nine representative regions that are homogeneously covered with different vegetation types is compared with Normalized Difference Vegetation Index (NDVI). The results indicate that the VOD can generally reveal vegetation seasonal changes and can provide unique information for vegetation monitoring. At a second step, after estimating VOD, soil moisture can be retrieved based on model using H-polarized multiangular brightness temperature. By analyzing a simulated database using the advanced integral equation model (AIEM), an effective surface roughness parameter that considers the influence of rms height, correlation length and correlation function shape on surface reflectivity was presented. Using this effective surface roughness parameter, a new parameterized surface reflectivity model is based on a simple-empirical model, the Hp model, is developed. Comparison with AIEM simulations over a wide range of soil conditions indecates a good performance of this model. This approach is then applied on SMOS data, retrieved soil moisture in Africa exhibitus reasonable patterns and temporal changes. Validation using in situ soil moisture from two soil moisture monitoring networks of Yanco Region and Little Washita watershed over 2010-2011 indicates that this approach performs well in surface soil moisture retrieval. Retrieved soil moisture agrees well with the in situ measurements with the RMSE of about 0.04 m3/m3.

Shi-Monitoring Vegetation and Soil Moisture from SMOS Data_Cn_version.pdf

Oral
ID: 313 / WS#3 ID.32397: 2
Oral Presentation
Hydrology & Cryosphere: 32397 - Calibration and Validation of Microwave Remote Sensing Data for Water Cycle Research

Soil Moisture Monitoring Using GNSS SNR Data: Proposing a Semi-empirical SNR Model

Dongkai Yang, Mutian Han, Xuebao Hong

BEIHANG UNIVERSITY, China, People's Republic of

Soil Moisture Content (SMC) is a key parameter in the study of agriculture and the global water cycle. In recent years, with the development of Global Navigation Satellite System, a new SMC remote sensing technique called GNSS-Interferometry and Reflectometry (GNSS-IR) was proposed by K. M. Larson et al. Compared with traditional remote sensing technique, it can provide retrievals at intermediate spatial scales with high time resolution, and is easier in its operation and management.

The GNSS-IR technique utilizes the composite signals formed by the interference effect occurred between the direct and the ground reflected navigation signals. These signals, which contain the physical information of the soil, are routinely recorded in a normal geodetic receiver in the form of Signal-to-Noise Ratio (SNR) data. Part of the efforts made in this field are to model SNR more accurately and extract SMC-related metrics from the SNR data. Recent contributions made by our group were to propose a semi-empirical SNR model which aimed at reconstructing the direct and reflected signal from SNR data and at the same time extracting frequency and phase information that is affected by soil moisture as proposed by K. M. Larson et al. This model worked as a curve-fitting model, and it was built through approximating the direct and reflected signal by a second-order and fourth-order polynomial, respectively, based on the well-established SNR model. Compared with other models (K. M. Larson et al. 2008, T. Yang et al. 2017), this model can improve the Quality of Fit (QoF) with little prior knowledge needed and can allow soil permittivity to be extracted from the reconstructed signals.

In this oral presentation, we will showed how this model was validated through simulation and experimental data processing. The data we used were collected by previous researchers at Lamasquère, France. Main results and finding are as follows:

Firstly, the QoF obtained using this model was improved by around 40%. It could ensure good fitting quality even in the case of irregular SNR variation. This advantage also results in better estimation of the frequency and phase information. However, we found that the improvement on phase estimation could be neglected.

Secondly, SMC could be retrieved from reconstructed signals. The results were satisfactory when the satellite elevation angle is between 5 degrees and 15 degrees. Additionally, the soil moisture calculated from the reconstructed signals was about 15% closer in relation to the ground truth measurements.

Finally, some phenomena were discovered regarding retrieval ambiguity and error sensitivity. These will also be stated and discussed in this oral presentation.

Yang-Soil Moisture Monitoring Using GNSS SNR Data_Cn_version.zip
Yang-Soil Moisture Monitoring Using GNSS SNR Data_ppt_present.zip

Poster
ID: 173 / WS#3 ID.32397: 3
Poster
Hydrology & Cryosphere: 32397 - Calibration and Validation of Microwave Remote Sensing Data for Water Cycle Research

Full Polarimetric Broad Band Scatterometry for Retrieval of Soil Moisture and Vegetation Properties over a Tibetan Meadow

Jan Hofste1, Rogier van der Velde1, Xin Wang2, Donghai Zheng3, Jun Wen4, Christiaan van der Tol1, Zhongbo Su1

1Faculty of Geo-Information Science and Earth Observation, University of Twente, Enschede, Netherlands; 2Key laboratory of Land Surface Process and Climate Change in Cold and Arid Regions, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China; 3Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China; 4College of Atmospheric Sciences, Plateau Atmosphere and Environment Key Laboratory of Sichuan Province, Chengdu University of Information Technology, Chengdu, China

Microwave (active and passive) and optical sensors are and will be deployed at an alpine meadow test site near Maqu city (China) on the Tibetan Plateau to study soil-vegetation-atmosphere processes and to validate physically based earth observation models used for retrievals from satellite data. Presented will be details and first results of a broad band full polarimetric scatterometer. The purpose of the scatterometer is to retrieve soil moisture content (SMC) and basic vegetation properties (vegetation water content, biomass, leaf area index) of the alpine meadow. This retrieval is done through inversion of backscattering models that link the measured backscattering coefficient σ0 to the SMC and aforementioned vegetation properties.

The scatterometer consists of a vector network analyser (VNA) connected to two dual polarization broadband antennas elevated 5 m above the surface. The radar return for co- and cross- polarization is measured over a 1 – 10 GHz frequency range (3 MHz resolution). The scatterometer calibration and validation was performed by means of a rectangular metal plate and metal dihedral reflector. For the co-polarization channels the measured radar cross section of the dihedral reflector matched a theoretical model within ±1 dB for 3 – 10 GHz. The calculation of σ0 from the measured radar return for the given site geometry and the frequency dependent radiation patterns will be explained as well.

Two experiments were performed. With the first experiment the antenna azimuth- and zenith angles were varied so that σ0 was measured over different parts of the surface under various angles of incidence θi. The azimuth range was -20° to 20° with 5° increments and the antenna zenith angles were varied such that θi varied from 35° to 70° with 5° increments. For the second experiment the orientation of the antennas were fixed and σ0 was recorded every hour during a long period (August 2017 – February 2018).

Analysis of the results from the first experiment will be presented to demonstrate the electromagnetic homogeneity of the ground surface. Since for the second experiment the antenna orientations were fixed it was not possible to measure multiple non-overlapping spatial samples of the surface. Therefore, to decrease the inherent uncertainty of the measured σ0 due to fading, frequency averaging will be applied to the data of the second experiment. A first glance at the results of the second experiment shows that during a 12 day period in August 2017 σ0 changed in parallel to in-situ measured SMC at 5 cm depth. We observe a decay of the σ0 during dry days and a sudden increase of σ0 after rainfall. The temporal behaviour of σ0 holds for most of the frequencies with all polarization channels.

Hofste-Full Polarimetric Broad Band Scatterometry for Retrieval_ppt_present.pdf