Alicia T. Joseph

Soil Moisture Retrieval During a Corn Growth Cycle Using L-Band (1.6 GHz) Radar Observations

This paper reports on the retrieval of soil moisture from dual-polarized L-band (1.6 GHz) radar observations acquired at view angles of 15deg, 35deg, and 55deg, which were collected during a field campaign covering a corn growth cycle in 2002. The applied soil moisture retrieval algorithm includes a surface roughness and vegetation correction and could potentially be implemented as an operational global soil moisture retrieval algorithm. The surface roughness parameterization is obtained through inversion of the Integral Equation Method (IEM) from dual-polarized (HH and VV) radar observations acquired under nearly bare soil conditions. The vegetation correction is based on the relationship found between the ratio of modeled bare soil scattering contribution and observed backscatter coefficient (sigmasoil/sigmaobs) and vegetation water content (W). Validation of the retrieval algorithm against ground measurements shows that the top 5-cm soil moisture can be estimated with an accuracy between 0.033 and 0.064 cm3 ldr cm-3, depending on the view angle and polarization.

A First-Order Radiative Transfer Model for Microwave Radiometry of Forest Canopies at L-Band

In this study, a first-order radiative transfer (RT) model is developed to more accurately account for vegetation canopy scattering by modifying the basic τ-ω model (the zero-order RT solution). In order to optimally utilize microwave radiometric data in soil moisture (SM) retrievals over vegetated landscapes, a quantitative understanding of the relationship between scattering mechanisms within vegetation canopies and the microwave brightness temperature is desirable. The first-order RT model is used to investigate this relationship and to perform a physical analysis of the scattered and emitted radiation from vegetated terrain. This model is based on an iterative solution (successive orders of scattering) of the RT equations up to the first order. This formulation adds a new scattering term to the τ-ω model. The additional term represents emission by particles (vegetation components) in the vegetation layer and emission by the ground that is scattered once by particles in the layer. The model is tested against 1.4-GHz brightness temperature measurements acquired over deciduous trees by a truck-mounted microwave instrument system called ComRAD in 2007. The model predictions are in good agreement with the data, and they give quantitative understanding for the influence of first-order scattering within the canopy on the brightness temperature. The model results show that the scattering term is significant for trees and modifications are necessary to the τ-ω model when applied to dense vegetation. Numerical simulations also indicate that the scattering term has a negligible dependence on SM and is mainly a function of the incidence angle and polarization of the microwave observation.

L-Band Radar Estimation of Forest Attenuation for Active/Passive Soil Moisture Inversion

In the radiometric sensing of soil moisture through a forest canopy, knowledge of canopy attenuation is required. Active sensors have the potential of providing this information since the backscatter signals are more sensitive to forest structure. In this paper, a new radar technique is presented for estimating canopy attenuation. The technique employs details found in a transient solution where the canopy (volume-scattering) and the tree-ground (double-interaction) effects appear at different times in the return signal. The influence that these effects have on the expected time-domain response of a forest stand is characterized through numerical simulations. A coherent forest scattering model, based on a Monte Carlo simulation, is developed to calculate the transient response from distributed scatterers over a rough surface. The forest transient-response model for linear copolarized cases is validated with the microwave deciduous tree data acquired by the Combined Radar/Radiometer (ComRAD) system. The attenuation algorithm is applicable when the forest height is sufficient to separate the components of the radar backscatter transient response. The frequency correlation functions of double-interaction and volume-scattering returns are normalized after being separated in the time domain. This ratio simply provides a physically based system of equations with reduced parameterizations for the forest canopy. Finally, the technique is used with ComRAD L-band stepped-frequency data to evaluate its performance under various physical conditions.

Evaluation of Dielectric Mixing Models for Passive Microwave Soil Moisture Retrieval Using Data From ComRAD Ground-Based SMAP Simulator

Soil moisture measurements are required to improve our understanding of hydrological processes and linkages between the Earths water, energy, and carbon cycles. The efficient retrieval of soil moisture depends on various factors among which soil dielectric mixing models are considered to be an important factor. The main objective of this work focuses on testing different dielectric models-Mironov et al., Dobson et al., Wang and Schmugge, and Hallikainen et al.-for soil moisture retrieval using the combined radar/radiometer (ComRAD) ground-based L-band simulator system, which serves as a simulator for the instruments on NASAs soil moisture active passive (SMAP) mission scheduled for launch in early next year. The single-channel algorithm at H polarization (SCA-H) version of the tau-omega model was used for soil moisture retrieval. A summer field experiment was conducted in 2012 at the United States Department of Agriculture (USDA) test site from which ComRAD measurements and validation samples of soil moisture were collected using theta probes and in situ sensors. The highest performance statistics combination in terms of high correlation (r), low root-mean-square error (RMSE), and least bias has been obtained with SCA-H using the Mironov dielectric model (r = 0.79; RMSE = 0.04 m3/m3; bias = 0.01) followed by Dobson (r = 0.76; RMSE = 0.04 m3/m3, bias = -0.01), Wang and Schmugge (r = 0.79; RMSE = 0.04 m3/m3, bias = 0.02) and Hallikainen (r = 0.76; RMSE = 0.06 m3/m3, bias = 0.04). Although the performance of the four dielectric models is relatively comparable, this analysis indicates that the Mironov dielectric model is marginally better than others for passive-only microwave soil moisture retrieval and could be a useful choice for SMAP satellite soil moisture retrieval.

Impact of Conifer Forest Litter on Microwave Emission at L-Band

This study reports on the utilization of microwave modeling, together with ground truth, and L-band (1.4-GHz) brightness temperatures to investigate the passive microwave characteristics of a conifer forest floor. The microwave data were acquired over a natural Virginia Pine forest in Maryland by a ground-based microwave active/passive instrument system in 2008/2009. Ground measurements of the tree biophysical parameters and forest floor characteristics were obtained during the field campaign. The test site consisted of medium-sized evergreen conifers with an average height of 12 m and average diameters at breast height of 12.6 cm. The site is a typical pine forest site in that there is a surface layer of loose debris/needles and an organic transition layer above the mineral soil. In an effort to characterize and model the impact of the surface litter layer, an experiment was conducted on a day with wet soil conditions, which involved removal of the surface litter layer from one half of the test site while keeping the other half undisturbed. The observations showed detectable decrease in emissivity for both polarizations after the surface litter layer was removed. A first-order radiative transfer model of the forest stands including the multilayer nature of the forest floor in conjunction with the ground truth data are used to compute forest emission. The model calculations reproduced the major features of the experimental data over the entire duration, which included the effects of surface litter and ground moisture content on overall emission. Both theory and experimental results confirm that the litter layer increases the observed canopy brightness temperature and obscure the soil emission.

ComRAD active / passive microwave measurement of tree canopies

The NASA/GSFC and George Washington University network analyzer-based multifrequency truck- mounted radar system has recently been upgraded with the addition of a dual-polarized 1.4 GHz total power radiometer. The system, now called ComRAD for Combined Radar/Radiometer, can function as a ground-based instrument simulator for L band space missions such as Hydros, Aquarius, and SMOS. In late summer 2006 ComRAD was deployed to the field to begin a series of coordinated active/passive L band measurements over small stands of deciduous and coniferous trees in order to improve our understanding of the microwave properties of trees and their effect on soil moisture retrieval algorithms. This paper describes the preliminary measurements obtained at the start of a three-year planned field measurement effort.

Soil moisture retrieval through changing corn using active/passive microwave remote sensing

Soil moisture is a critical state variable in land surface hydrology. Large-scale soil moisture mapping based on microwave remote sensing would be valuable in many different practical and theoretical applications, and a real potential exists for new space missions in the near future which well utilize simultaneous active/passive microwave measurements for global soil moisture retrieval. This paper discusses the experiment for the retrieval of soil moisture using radar and radiometric measurements. It was shown that combinations of simultaneous radar and radiometer data can enhance soil moisture retrievals, especially in the presence of dynamic vegetation.

L band brightness temperature observations over a corn canopy during the entire growth cycle.

During a field campaign covering the 2002 corn growing season, a dual polarized tower mounted L-band (1.4 GHz) radiometer (LRAD) provided brightness temperature (TB) measurements at preset intervals, incidence and azimuth angles. These radiometer measurements were supported by an extensive characterization of land surface variables including soil moisture, soil temperature, vegetation biomass, and surface roughness. In the period May 22 to August 30, ten days of radiometer and ground measurements are available for a corn canopy with a vegetation water content (W) range of 0.0 to 4.3 kg m−2. Using this data set, the effects of corn vegetation on surface emissions are investigated by means of a semi-empirical radiative transfer model. Additionally, the impact of roughness on the surface emission is quantified using TB measurements over bare soil conditions. Subsequently, the estimated roughness parameters, ground measurements and horizontally (H)-polarized TB are employed to invert the H-polarized transmissivity (γh) for the monitored corn growing season.

L-band active / passive time series measurements over a growing season using the ComRAD ground-based SMAP simulator

Once launched in late 2014, NASAs Soil Moisture Active Passive (SMAP) mission will use a combination of a four-channel L-band radiometer and a three-channel L-band radar to provide high resolution global mapping of soil moisture and landscape freeze/thaw state every 2-3 days. These measurements are valuable to improved understanding of the Earths water, energy, and carbon cycles, and to many applications of societal benefit. In order for soil moisture to be retrieved accurately from SMAP microwave data, prelaunch activities are concentrating on developing improved geophysical retrieval algorithms for each of the SMAP baseline products. The ComRAD truck-based SMAP simulator collected active/passive microwave time series data at the SMAP incident angle of 40° over corn and soybeans during 2012 for use in refining SMAP retrieval algorithms.

Microwave soil moisture retrieval under trees using a modified tau-omega model

During 2007–2009 field experiments have been conducted using the ComRAD microwave truck instrument system with a goal of optimizing microwave soil moisture retrieval algorithms for small to medium deciduous and coniferous trees. A joint effort of NASA / GSFC and George Washington University, ComRAD consists of a quad-polarized 1.25 GHz radar and a dual-polarized 1.4 GHz radiometer sharing the same antenna. In the current study, ComRAD microwave data and ground truth measurements of soil moisture, temperature, soil texture, and vegetation water content and geometry statistics have been used to assess whether the zero-order tau-omega model can be employed successfully to retrieve soil moisture under tree canopies using effective values for tau (the vegetation opacity) and omega (the single scattering albedo). In addition, the tauomega model has been modified to include a first-order scattering term, which will be discussed in a companion paper [1].