[Show abstract][Hide abstract] ABSTRACT: Satellite-based passive microwave remote sensing has been shown to be a valuable tool in mapping and monitoring global soil moisture. The Advanced Microwave Scanning Radiometer on the Aqua platform (AMSR-E) has made significant contributions to this application. As the result of agency and individual initiatives, several approaches for the retrieval of soil moisture from AMSR-E have been proposed and implemented. Although the majority of these are based on the same Radiative Transfer Equation, studies have shown that the resulting soil moisture estimates can differ significantly. A primary goal of this investigation is to understand these differences and develop a suitable approach to potentially improve the algorithm currently used by NASA in producing its operational soil moisture product. In order to achieve this goal, the theoretical basis of several alternative soil moisture retrieval algorithms are examined. Analysis has focused on five established approaches: the operational algorithm adopted by NASA, which is referred to as the Normalized Polarization Difference algorithm, the Single Channel Algorithm, the Land Parameter Retrieval Model, the University of Montana soil moisture algorithm, and the HydroAlgo Artificial Neural Network algorithm. Previous comparisons of these algorithms in the literature have typically focused on the retrieved soil moisture products, and employed different metrics and data sets, and have resulted in differing conclusions. In this investigation we attempt to provide a more thorough understanding of the fundamental differences between the algorithms and how these differences affect their performance in terms of range of soil moisture provided. The comparative overview presented in the paper is based on the operating versions of the source codes of the individual algorithms. Analysis has indicated that the differences between algorithms lie in the specific parameterizations and assumptions of each algorithm. The comparative overview of the theoretical basis of the approaches is linked to differences found in the soil moisture retrievals, leading to suggestions for improvements and increased reliability in these algorithms.
[Show abstract][Hide abstract] ABSTRACT: Overlapping soil moisture time series derived from two satellite microwave radiometers (the Soil Moisture and Ocean Salinity (SMOS) and the Advanced Microwave Scanning Radiometer-Earth Observing System) are used to generate a soil moisture time series from 2003 to 2010. Two statistical methodologies for generating long homogeneous time series of soil moisture are considered. Generated soil moisture time series using only morning satellite overpasses are compared to ground measurements from four watersheds in the U.S. with different climatologies. The two methods, cumulative density function (CDF) matching and copulas, are based on the same statistical theory, but the first makes the assumption that the two data sets are ordered the same way, which is not needed by the second. Both methods are calibrated in 2010, and the calibrated parameters are applied to the soil moisture data from 2003 to 2009. Results from these two methods compare well with ground measurements. However, CDF matching improves the correlation, whereas copulas improve the root-mean-square error.
IEEE Transactions on Geoscience and Remote Sensing 01/2014; 52(1):393-405. DOI:10.1109/TGRS.2013.2240691 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: We coupled a radiative transfer model and a soil hydrological model (HYDRUS 1D) (Šimůnek et al., 2008) with an optimization routine to derive soil hydraulic parameters, surface roughness, and soil moisture of a tilled bare soil plot using measured brightness temperatures at 1.4 GHz (L-band), rainfall, and potential soil evaporation. The robustness of the approach was evaluated using five one-month data sets representing different meteorological conditions. We considered two soil hydraulic property models: the uni-modal Mualem van Genuchten and the bi-modal of Durner. Microwave radiative transfer was modeled by three different approaches: the Fresnel equation with depth averaged dielectric permittivity of either 2 or 5 cm thick surface layers and a coherent radiative transfer model (CRTM) that accounts for vertical gradients in dielectric permittivity. Brightness temperatures simulated by the CRTM and the 2-cm layer Fresnel model fitted well to the measured ones. L-band brightness temperatures are therefore related to the dielectric permittivity and soil moisture in a 2 cm thick surface layer. The surface roughness parameter that was derived from brightness temperatures using inverse modeling was similar to direct estimates from laser profiler measurements. The lab derived water retention curve was bi-modal and could be retrieved consistently for the different periods from brightness temperatures using inverse modeling. A uni-modal soil hydraulic property function underestimated the hydraulic conductivity near saturation. Surface soil moisture contents simulated using retrieved soil hydraulic parameters were compared with in-situ measurements. Depth specific calibration relations were essential to derive soil moisture from near-to-surface installed sensors.
Vadose Zone Journal 12/2013; DOI:10.2136/vzj2013.04.0075 · 2.41 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: We present the most recent results from on-going collaborative sensor
inter-calibration and salinity inter-comparison analyses between SMOS
and Aquarius/SAC-D satellite ocean brightness temperature and salinity
retrievals. An important goal of both programs is to inter-calibrate
and combine the respective satellite data sets and in situ ocean
measurements to provide an accurate and well resolved ocean surface
salinity observing capability for ocean and climate studies. At the
time of this symposium, SMOS will have completed more that three years
of observations and Aquarius more than 19 months. This study will apply
the Aquarius data V2.0 to be released at the end of January 2013 (after
the submission of this abstract), which includes updated calibration,
pointing and geophysical model corrections. The comparison analyses
will consist of these elements: (1) Polarized brightness temperatures
(TH and TV) interpolated or synthesized at the Aquarius viewing angles,
(2) Level 2 salinity retrievals, (3) Level 3 gridded data, and (4) in
situ salinity data. The focus will be on quantifying the
inter-calibration biases between the sensors, relative uncertainties of
the salinity retrievals and the large-scale spatial and temporal
systematic biases between the two satellites and in situ data.
[Show abstract][Hide abstract] ABSTRACT: The National Aeronautics and Space Administration's (NASA) proposed Soil Moisture Active Passive (SMAP) satellite mission ( ~ 2014) will include a radar system that will provide L-band multi-polarization backscatter at a constant incidence angle of 40 °. During the pre-launch phase of the project, there is a need for observations that will support the radar-based soil moisture algorithm development and validation. A valuable resource for providing these observations is the NASA Jet Propulsion Laboratory Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR). However, SMAP will observe at a constant incidence angle of 40 °, and UAVSAR collects data over a wide range of incidence angles (25 °-60°). In this investigation, a technique was developed and tested for normalizing UAVSAR data to a constant incidence angle. The approach is based on a histogram matching procedure. The data used to develop and demonstrate this approach were collected as part of the Canadian Soil Moisture Experiment 2010 (CanEx-SM10). Land cover in the region included agriculture and forest. Evaluation was made possible by the acquisition of numerous overlapping UAVSAR flight lines that provided multiple incidence angle observations of the same locations. Actual observations at a 40° incidence angle were compared to the normalized data to assess performance of the normalization technique. An optimum technique should be able to reduce the systematic error (Bias) to 0 dB and to lower the total root mean square error (RMSE) computed after correction to the level of the initial residual error (RMSEres) present in the data set. The normalization approach developed here achieved both of these. Bias caused by the incidence angle variability was minimized to ~ 0 dB, whereas the residual error caused by instrument related random errors and amplitude fluctuations due to ground variability was r- duced to approximately 3 dB for agricultural areas and 2.6 dB for forests; these values were consistent with the initial RMSEres estimated using the un-corrected data. The residual error can be reduced further by aggregating the radar observations to a coarser grid spacing. The technique adequately adjusted the backscatter over the full swath width irrespective of the original incidence angle, polarization, and ground conditions (vegetation cover and soil moisture). In addition to providing a basis for fully exploiting UAVSAR (or similar aircraft systems) for SMAP algorithm development and validation, the technique could also be adapted to satellite radar systems. This normalization approach will also be beneficial in terms of reducing the number of flight lines required to cover a study area, which would eventually result in more cost-effective soil moisture field campaigns.
IEEE Transactions on Geoscience and Remote Sensing 03/2013; 51(3):1791-1804. DOI:10.1109/TGRS.2012.2205264 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Estimation of soil moisture at large scale has been performed using several satellite-based passive microwave sensors and a variety of retrieval methods over the past two decades. The most recent source of soil moisture is the European Space Agency Soil Moisture and Ocean Salinity (SMOS) mission. A thorough validation must be conducted to insure product quality that will, in turn, support the widespread utilization of the data. This is especially important since SMOS utilizes a new sensor technology and is the first passive L-band system in routine operation. In this paper, we contribute to the validation of SMOS using a set of four in situ soil moisture networks located in the U.S. These ground-based observations are combined with retrievals based on another satellite sensor, the Advanced Microwave Scanning Radiometer (AMSR-E). The watershed sites are highly reliable and address scaling with replicate sampling. Results of the validation analysis indicate that the SMOS soil moisture estimates are approaching the level of performance anticipated, based on comparisons with the in situ data and AMSR-E retrievals. The overall root-mean-square error of the SMOS soil moisture estimates is 0.043 m3/m3 for the watershed networks (ascending). There are bias issues at some sites that need to be addressed, as well as some outlier responses. Additional statistical metrics were also considered. Analyses indicated that active or recent rainfall can contribute to interpretation problems when assessing algorithm performance, which is related to the contributing depth of the satellite sensor. Using a precipitation flag can improve the performance. An investigation of the vegetation optical depth (tau) retrievals provided by the SMOS algorithm indicated that, for the watershed sites, these are not a reliable source of information about the vegetation canopy. The SMOS algorithms will continue to be refined as feedback from validation is evaluated, and it is expe- ted that the SMOS estimates will improve.
IEEE Transactions on Geoscience and Remote Sensing 05/2012; 50(5):1530-1543. DOI:10.1109/TGRS.2011.2168533 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: 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.
IEEE Transactions on Geoscience and Remote Sensing 10/2011; 49(9-49):3167 - 3179. DOI:10.1109/TGRS.2010.2091139 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Soil moisture products provided by the Soil Moisture and Ocean Salinity (SMOS) satellite were evaluated using in situ observations. The sites are located in different regions of the U.S. and provide replicate sampling of surface soil moisture at the SMOS footprint scale. Data from a sparse network were also considered. Soil moisture products from the Advanced Microwave Scanning Radiometer were also used for validation. Results based upon a preliminary version of the retrieval algorithm indicate promising performance. It is anticipated that the accuracy and reliability of the retrievals will improve as validation information is evaluated.
Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International; 08/2011
[Show abstract][Hide abstract] ABSTRACT: WindSat has provided an opportunity to investigate the first spaceborne passive fully polarimetric observations of the Earth's surface. In this paper, Arctic sea ice was investigated. The passive polarimetric data are provided in the form of the modified Stokes vector consisting of four parameters. The first two components of the modified Stokes vector are the vertically and horizontally polarized brightness temperatures, which have been continuously measured by various radiometers over the last three decades. The third and fourth Stokes components provide in formation on the degree of polarization of the emission. In this paper, three types of analysis are carried out: spatial (maps considering different azimuth angle intervals), temporal (time series of daily averaged Stokes components over a small selected azimuth angle range), and azimuthal (variations w.r.t. the azimuth angle over selected study areas). Analysis has shown the highest brightness temperature variations for the 37-GHz third Stokes component (>; 2 K) during summer. The next highest signals were observed for the 10.7-GHz third and fourth Stokes components (>; 1 K) during summer as well. The 37-GHz fourth Stokes component exhibited the least variability (>; 1 K). Spikes of up to 2 K were identified in the time series of the 37-GHz third Stokes component during mid-January 2004 (winter) over first-year ice regions. The near-surface air temperature of the European Center for Medium-Range Weather Forecasts model data and the Special Sensor Microwave/Imager National Aeronautics and Space Administration Team ice concentrations revealed that, during these events, the surface temperatures reached near melting levels and the retrieved ice concentrations were reduced to about 80%. Moreover, these observations also showed clear evidence of first harmonic azimuthal dependence. Geophysical parameters, such as temperature and ice leads, are likely to be the causes. The larger signals which occurred d uring summer were identified as being related to the ice surface temperatures being near melting.
IEEE Transactions on Geoscience and Remote Sensing 06/2011; DOI:10.1109/TGRS.2010.2089058 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Validation is an important and particularly challenging task for remote sensing of soil moisture. A key issue in the validation of soil moisture products is the disparity in spatial scales between satellite and in situ observations. Conventional measurements of soil moisture are made at a point, whereas satellite sensors provide an integrated area/volume value for a much larger spatial extent. In this paper, four soil moisture networks were developed and used as part of the Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E) validation program. Each network is located in a different climatic region of the U.S., and provides estimates of the average soil moisture over highly instrumented experimental watersheds and surrounding areas that approximate the size of the AMSR-E footprint. Soil moisture measurements have been made at these validation sites on a continuous basis since 2002, which provided a seven-year period of record for this analysis. The National Aeronautics and Space Administration (NASA) and Japan Aerospace Exploration Agency (JAXA) standard soil moisture products were compared to the network observations, along with two alternative soil moisture products developed using the single-channel algorithm (SCA) and the land parameter retrieval model (LPRM). The metric used for validation is the root-mean-square error (rmse) of the soil moisture estimate as compared to the in situ data. The mission requirement for accuracy defined by the space agencies is 0.06 m<sup>3</sup>/m<sup>3</sup>. The statistical results indicate that each algorithm performs differently at each site. Neither the NASA nor the JAXA standard products provide reliable estimates for all the conditions represented by the four watershed sites. The JAXA algorithm performs better than the NASA algorithm under light-vegetation conditions, but the NASA algorithm is more reliable for moderate vegetation. However, both algorithms have a moderate to large bias in all cases. The SC-
A had the lowest overall rmse with a small bias. The LPRM had a very large overestimation bias and retrieval errors. When site-specific corrections were applied, all algorithms had approximately the same error level and correlation. These results clearly show that there is much room for improvement in the algorithms currently in use by JAXA and NASA. They also illustrate the potential pitfalls in using the products without a careful evaluation.
IEEE Transactions on Geoscience and Remote Sensing 01/2011; 48(12-48):4256 - 4272. DOI:10.1109/TGRS.2010.2051035 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Soil Moisture Active Passive (SMAP), a proposed mission in support of the Earth Science Decadal Survey, conducted a field campaign in June 2010 to support algorithm development. As part of the experiment in situ soil moisture measurements were made over a two week period in which multiple UAVSAR flights were conducted. Repeat-pass polarimetric-interferometric data generated from these flights were analyzed to see if phase changes could be correlated with soil moisture changes. Also, we compared the data to that predicted by simple surface scattering models and showed moderate agreement with the Oh model .
2011 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2011, Vancouver, BC, Canada, July 24-29, 2011; 01/2011
[Show abstract][Hide abstract] ABSTRACT: Surface soil moisture was retrieved from the L-band radiometer data collected in semiarid regions during the Soil Moisture Experiment in 2004. The 2-D synthetic aperture radiometer (2D-STAR) was flown over regional-scale study sites located in AZ, USA, and Sonora, Mexico (SO). The study sites are characterized by a range of topographic relief with a land cover that varies from bare soil to grass and scrubland and includes areas with high rock fraction near the soil surface. The 2D-STAR retrieval of soil moisture was in good agreement with the ground-based estimates of surface soil moisture in both AZ (raise = 0.012 m<sup>3</sup> m<sup>-3</sup>) and SO (rmse = 0.011 m<sup>3</sup> m<sup>-3</sup>). The 2D-STAR also showed a good performance in the Walnut Gulch Experimental Watershed (rmse = 0.014 m<sup>3</sup> m<sup>-3</sup>) where the surface soil featured high rock fraction was as high as 60%. Comparison of the results with the Polarimetric Scanning Radiometer at the Cand X-band data indicates the superior soil moisture retrieval performance of the L-band data over the regions with high rock fraction and moderate vegetation density.
IEEE Transactions on Geoscience and Remote Sensing 01/2011; 48(12-48):4273 - 4284. DOI:10.1109/TGRS.2010.2051677 · 3.51 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The Cloud and Land Surface Interaction Campaign is a field experiment designed to collect a comprehensive data set that can be used to quantify the interactions that occur between the atmosphere, biosphere, land surface, and subsurface. A particular focus will be on how these interactions modulate the abundance and characteristics of small and medium size cumuliform clouds that are generated by local convection. These interactions are not well understood and are responsible for large uncertainties in global climate models, which are used to forecast future climate states. The campaign will be conducted from June 8 to June 30, 2007, at the U.S. Department of Energy’s Atmospheric Radiation Measurement Climate Research Facility Southern Great Plains site. Data will be collected using eight aircraft equipped with a variety of specialized sensors, four specially instrumented surface sites, and two prototype surface radar systems. The architecture of Cloud and Land Surface Interaction Campaign includes a highaltitude surveillance aircraft and enhanced vertical thermodynamic and wind profile measurements that will characterize the synoptic scale structure of the clouds and the land surface within the Atmospheric Radiation Measurement Climate Research Facility Southern Great Plains site. Mesoscale and microscale structures will be sampled with a variety of aircraft, surface, and radar observations. ii M.R. Miller et al., DOE/SC-ARM-0703
[Show abstract][Hide abstract] ABSTRACT: The availability of global L-band observations from passive (the recently launched SMOS), and active (such as the PALSAR) microwave sensors has boosted the interest in making joint use of the two techniques to improve the retrieval of global near-surface soil moisture at unprecedented resolutions. The Soil Moisture Active Passive (SMAP) mission (scheduled launch, 2014) will fully exploit this synergy by providing concurrent active (radar) and passive (radiometer) microwave observations, resulting in passive-only, active-only and a merged active-passive soil moisture products at spatial resolutions of respectively 40km, 3km and 9km. The Soil Moisture Active Passive Experiments (SMAPEx) are a series of airborne field experiments specifically designed for algorithm development for SMAP and currently ongoing in the context of the SMAP pre-launch cal/val activities for Australia. Four SMAPEx campaigns are scheduled across the 2010-2011 seasonal cycle, with the first campaign (SMAPEx-1) successfully conducted on moderately wet winter conditions (July 5-10, 2010) and the second campaign (SMAPEx-2), scheduled for the summer (December 4-8,2010). SMAPEx is making use of a novel SMAP airborne simulator, including an L-band radar and radiometer to collect SMAP-like data over a well monitored semi-arid agricultural area in the Murrumbidgee catchment in south-eastern Australia. High resolution radar and radiometer observations collected during SMAPEx are supported by extensive ground sampling of soil moisture and ancillary data, allowing for testing of a variety of algorithms over semi-arid agricultural areas, typical of the Australian environment but similar to large areas of the central continental USA, including radiometer-only, radar-only, merged active-passive, downscaling and radar change-detection algorithms. In this paper a preliminary assessment of the performance of the radar-only and radiometer-only retrieval algorithms proposed as baseline for SMAP is presented. The soil moisture retrieved from active and passive microwave airborne observations collected during the SMAPEx-1 campaign is compared with extensive spatial data collected at focus areas. The quality of the individual retrievals is discussed in relation with different land surface conditions, ranging from intensive cropping to dryland grassland areas.
[Show abstract][Hide abstract] ABSTRACT: Retrieval of soil moisture content from microwave sensors also returns an estimate of vegetation water content. Remotely sensed indices from optical sensors can be used to estimate canopy water content. For corn and soybean in central Iowa, there are allometric relationships between canopy water content and vegetation water content. The Normalized Difference Infrared Index from MODIS was used to estimate vegetation water content. We compared independent estimates of vegetation water content from WindSat and MODIS over central Iowa from 2003 to 2005. There was a strong linear relationship between the MODIS and WindSat estimates, but the WindSat estimates were about two times higher. These results suggest that soil moisture retrievals from microwave sensors may be more accurate with estimates of vegetation water content from optical sensors.
Geoscience and Remote Sensing Symposium (IGARSS), 2010 IEEE International; 08/2010