Soil Moisture Retrieval Errors Research Articles

Multi-Objective Optimization of an Intelligent Soil-Moisture-Monitoring Satellite Constellation

In early satellite mission design, requirements are not yet fixed, cost is sometimes negotiable, and designs are relatively unconstrained. During this period of design freedom, multi-objective optimization can provide a useful lens into the design space by showing theoretical performance limits and illuminating design tradeoffs. This work optimizes a radar constellation for a potential soil moisture mission. Several different optimization cases with different variables are considered and contrasted. The optimization of the instrument and constellation parameters is considered jointly and separately to better understand the effect of coupling on the optimization performance. A science-driven optimization based on soil moisture retrieval error is compared with a performance-metric-driven optimization. Pareto analysis and association rule mining are performed on the generated designs to provide insight into driving features. Design recommendations are made for several cost caps. Results show that optimization that considers the instrument and constellation design together find superior revisit metrics than treating instrument and constellation separately. The use of the science value metric as an optimization objective shows that while cost may always be increased to improve instrument and constellation performance, the difference in science value may be negligible. These findings can inform tradespace exploration studies for similar problems.

The error structure of the SMAP single and dual channel soil moisture retrievals.

Knowledge of the temporal error structure for remotely sensed surface soil moisture retrievals can improve our ability to exploit them for hydrologic and climate studies. This study employs a triple collocation analysis to investigate both the total variance and temporal auto-correlation of errors in Soil Moisture Active and Passive (SMAP) products generated from two separate soil moisture retrieval algorithms, the vertically-polarized brightness temperature based Single Channel Algorithm (SCA-V, the current baseline SMAP algorithm) and the Dual Channel Algorithm (DCA). A key assumption made in SCA-V is that real-time vegetation opacity can be accurately captured using only a climatology for vegetation opacity. Results demonstrate that, while SCA-V generally outperforms DCA, SCA-V can produce larger total errors when this assumption is significantly violated by inter-annual variability in vegetation health and biomass. Furthermore, larger auto-correlated errors in SCA-V retrievals are found in areas with relatively large vegetation opacity deviations from climatological expectations. This implies that a significant portion of the auto-correlated error in SCA-V is attributable to the violation of its vegetation opacity climatology assumption and suggests that utilizing a real (as opposed to climatological) vegetation opacity time series in the SCA-V algorithm would reduce the magnitude of auto-correlated soil moisture retrieval errors.

Assessing SMAP Soil Moisture Scaling and Retrieval in the Carman (Canada) Study Site

Core Ideas Upscaling methods compared in situ measures with soil moisture from the SMAP satellite. The accuracy of SMAP soil moisture products in annual cropland was assessed. The spatial representativeness of sparse in situ networks was determined. In 2015, NASA launched the Soil Moisture Active Passive (SMAP) satellite. Data from this satellite are being exploited to improve forecasting of extreme weather events and delivery of disaster response. International core validation sites (CVSs) have been contributing in situ soil moisture data to validate and calibrate SMAP soil moisture products. Overall the soil moisture retrieval errors have exceeded SMAP's mission requirement (errors below 0.04 m3 m−3), with the exception of some sites of annual cropland as present at the Carman (Canada) CVS. In 2016, a SMAP validation experiment was conducted at the Canadian site in Manitoba (SMAPVEX16‐MB) in an attempt to understand the differences between the SMAP soil moisture retrievals and the permanent in situ network observations. The research presented here analyzed the performance of this network in representing soil moisture within a SMAP pixel and tested five upscaling approaches. Comparisons between the permanent network and SMAPVEX16‐MB measurements (from temporary stations and field measures) confirmed agreement among these three sources of soil moisture measures. The SMAP soil moisture values were compared with in situ soil moisture upscaled from the four tested approaches as well as soil moisture estimated by the NOAH Land Surface Model (LSM). There were similar discrepancies when analyzing all methods (RMSE 0.072–0.074 m3 m−3 for the four upscaling methods; 0.076 m3 m−3 for the LSM approach), yielding no reduction in the soil moisture RMSE for this site. The SMAP team will continue to investigate other factors that may be contributing to errors above 0.04 m3 m−3 at these annually cropped CVSs.

An Improved Triple Collocation Analysis Algorithm for Decomposing Autocorrelated and White Soil Moisture Retrieval Errors

AbstractIf not properly account for, autocorrelated retrieval errors can lead to inaccurate results in soil moisture data analysis and reanalysis. Here we propose a more generalized form of the triple collocation analysis algorithm (GTC) capable of decomposing the total error variance of remotely sensed surface soil moisture retrievals into their autocorrelated and the serially white components. Synthetic tests demonstrate the robustness and accuracy of GTC—even in the presence of significant temporal data gaps. However, the accuracy of GTC error autoregressive parameter estimates is relatively more sensitive to temporal data availability. In addition, land surface model soil moisture predictions collected from phase 2 of the North American Land Data Assimilation System and remotely sensed surface soil moisture retrievals obtained from the European Space Agency Climate Change Initiative (ESA CCI) are applied for a real data demonstration. Despite expectations to the contrary, significant error autocorrelation is found in the remotely sensed‐based ESA CCI soil moisture data sets. In particular, ESA CCI‐Act (i.e., the subset of ESA CCI soil moisture retrievals based on active scattomotter data) demonstrates the largest autoregressive parameters over low biomass areas. Conversely, ESA CCI‐Pas retrievals (based on passive radiometer data) have larger error autoregressive parameters over high biomass areas. As such, results clarify circumstances in which errors in remotely sensed surface soil moisture retrievals cannot be considered serially white.

Standing water effect on soil moisture retrieval from L-band passive microwave observations

Abstract Passive microwave remote sensing at L-band has been widely acknowledged as the most promising technique to observe the spatial distribution of near surface (top ~ 5 cm) soil moisture at regional to global scales. The launch of the ESA's Soil Moisture and Ocean Salinity (SMOS) mission in 2009 now means that global space-borne brightness temperature observations are available at L-band (1.41 GHz) to estimate soil moisture every 2 to 3 days with a target accuracy of 0.04 m 3 /m 3 . Moreover, NASA's Soil Moisture Active Passive (SMAP) satellite has been launched on 31st January 2015, also carrying an L-band radiometer, together with an L-band radar for downscaling the brightness temperature observations to better than 10 km resolution. At the SMOS/SMAP radiometer scale of ~ 40 km, the presence of water bodies potentially induces an overestimation of retrieved soil moisture, if not carefully accounted for in retrieval models. Such water fraction effects on brightness temperature and soil moisture retrieval accuracy were investigated in this study, using airborne L-band brightness temperature data collected during three Australian field experiments. The water induced brightness temperature effect and water fraction were compared under different resolutions, sampling days, and land surface conditions, showing that the water fraction impact on retrieved soil moisture is independent of scale, but heavily dependent on the soil water content status. Subsequently, the highest water fraction threshold that can be tolerated in order to achieve the 0.04 m 3 /m 3 target accuracy without correction has been determined as 0.08 (actual range is from 0.02 for dry bare soil to 0.08 for wet vegetated soil). Using a MODIS derived water fraction dataset, the water fraction dynamics were also studied over Australia during the ten years from 2001 to 2010. The results show that if the mean water fraction map was used as a static water map to flag or correct water effects, the water body induced soil moisture retrieval error would have exceeded the 0.04 m 3 /m 3 target more than once for 13.5% of the Australian land 40 km sized radiometer pixels; only 0.6% Australian land pixels would have exceeded this target with a frequency of 10 times or more per year.

Estimating Effective Roughness Parameters of the L-MEB Model for Soil Moisture Retrieval Using Passive Microwave Observations From SMAPVEX12

Despite the continuing efforts to improve existing soil moisture retrieval algorithms, the ability to estimate soil moisture from passive microwave observations is still hampered by problems in accurately modeling the observed microwave signal. This paper focuses on the estimation of effective surface roughness parameters of the L-band Microwave Emission from the Biosphere (L-MEB) model in order to improve soil moisture retrievals from passive microwave observations. Data from the SMAP Validation Experiment 2012 conducted in Canada are used to develop and validate a simple model for the estimation of effective roughness parameters. Results show that the L-MEB roughness parameters can be empirically related to the observed brightness temperatures and the leaf area index of the vegetation. These results indicate that the roughness parameters are compensating for both roughness and vegetation effects. It is also shown, using a leave-one-out cross validation, that the model is able to accurately estimate the roughness parameters necessary for the inversion of the L-MEB model. In order to demonstrate the usefulness of the roughness parameterization, the performance of the model is compared to more traditional roughness formulations. Results indicate that the soil moisture retrieval error can be reduced to 0.054 m3/m3 if the roughness formulation proposed in this study is implemented in the soil moisture retrieval algorithm.

Potential soil moisture products from the aquarius radiometer and scatterometer using an observing system simulation experiment

Abstract. Using an observing system simulation experiment (OSSE), we investigate the potential soil moisture retrieval capability of the National Aeronautics and Space Administration (NASA) Aquarius radiometer (L-band 1.413 GHz) and scatterometer (L-band, 1.260 GHz). We estimate potential errors in soil moisture retrievals and identify the sources that could cause those errors. The OSSE system includes (i) a land surface model in the NASA Land Information System, (ii) a radiative transfer and backscatter model, (iii) a realistic orbital sampling model, and (iv) an inverse soil moisture retrieval model. We execute the OSSE over a 1000 × 2200 km2 region in the central United States, including the Red and Arkansas river basins. Spatial distributions of soil moisture retrieved from the radiometer and scatterometer are close to the synthetic truth. High root mean square errors (RMSEs) of radiometer retrievals are found over the heavily vegetated regions, while large RMSEs of scatterometer retrievals are scattered over the entire domain. The temporal variations of soil moisture are realistically captured over a sparely vegetated region with correlations 0.98 and 0.63, and RMSEs 1.28% and 8.23% vol/vol for radiometer and scatterometer, respectively. Over the densely vegetated region, soil moisture exhibits larger temporal variation than the truth, leading to correlation 0.70 and 0.67, respectively, and RMSEs 9.49% and 6.09% vol/vol respectively. The domain-averaged correlations and RMSEs suggest that radiometer is more accurate than scatterometer in retrieving soil moisture. The analysis also demonstrates that the accuracy of the retrieved soil moisture is affected by vegetation coverage and spatial aggregation.

Wheat Canopy Structure and Surface Roughness Effects on Multiangle Observations at L-Band

The multiangle observation capability of the Soil Moisture and Ocean Salinity mission is expected to significantly improve the inversion of soil microwave emissions for soil moisture, by enabling the simultaneous retrieval of the vegetation optical depth and other surface parameters. Consequently, this paper investigates the relationship between soil moisture and brightness temperature at multiple incidence angles using airborne L-band data from the National Airborne Field Experiment in Australia in 2005. A forward radio brightness model was used to predict the passive microwave response at a range of incidence angles, given the following inputs: 1) ground-measured soil and vegetation properties and 2) default model parameters for vegetation and roughness characterization. Simulations were made across various dates and locations with wheat cover and evaluated against the available airborne observations. The comparison showed a significant underestimation of the measured brightness temperatures by the model. This discrepancy subsequently led to soil moisture retrieval errors of up to 0.3 m3/m3. Further analysis found the following: 1) The roughness value HR was too low, which was then adjusted as a function of the soil moisture, and 2) the vegetation structure parameters tth and ttv required optimization, yielding new values of tth = 0.2 and ttv = 1.4 from calibration to a single flight. Testing the optimized parameterization for different moisture conditions and locations found that the root-mean-square simulation error between the forward model predictions and the airborne observations was improved from 31.3 K (26.5 K) to 2.3 K (5.3 K) for wet (dry) soil moisture condition.

Soil Salinity Impacts on L-Band Remote Sensing of Soil Moisture

The recently launched Soil Moisture and Ocean Salinity (SMOS) satellite is providing soil moisture observations at continental scales by measuring L-band microwave radiation emitted from the land surface. While its retrieval algorithms will correct for factors such as vegetation and surface roughness, it will not correct for soil salinity. This letter tests the assumption that soil salinity will have a negligible impact on L-band brightness temperature (Tb) at SMOS scales using field data; airborne Tb observations were collected in a saline groundwater discharge area near Nilpinna Station, South Australia. At the 500-m scale, the airborne observations of Tb could not be reproduced using the baseline algorithm of the SMOS Level 2 retrieval scheme, without accounting for soil salinity in the model. The analysis in this letter shows that soil moisture retrieval errors of at least 0.04 m3 m-3 (i.e., the entire SMOS error budget) will occur due to salinity alone in SMOS footprints with saline coverage as low as 25% (possibly even much less). Consequently, fractional salinity coverage cannot be considered a negligible factor by microwave soil moisture satellite missions.

Effects of spatial distribution of soil parameters on soil moisture retrieval from passive microwave remote sensing

In this paper, we studied the effect of spatial distribution of soil parameters on passive soil moisture retrieval at pixel scale. First, we evaluated the forward microwave emission model and soil moisture retrieval algorithm accuracy through the observation of field experiments. Then, we used soil parameters in different spatial distribution patterns, including random, normal, and uniform distribution, to determine the different levels of heterogeneity on soil moisture retrieval, in order to seek the relationship between heterogeneity and soil moisture retrieval error. Finally, we conducted a controlled heterogeneity effect experiment measurements using a Truck-mounted Multi-frequency Radiometer (TMMR) to validate our simulation results. This work has proved that the soil moisture retrieval algorithm had a high accuracy (RMSE=0.049 cm3 cm−3) and can satisfy the need of this research. The simulation brightness temperatures match well with observations, with RMSE=9.89 K. At passive microwave remote sensing pixel scale, soil parameters with different spatial distribution patterns could have different levels of error on soil moisture estimation. Overall, we found that soil moisture with a random distribution in a satellite pixel scale can cause the largest error, with a normal distribution being the second, and a uniform distribution the least due to the smallest heterogeneity.

A proposed extension to the soil moisture and ocean salinity level 2 algorithm for mixed forest and moderate vegetation pixels

The Soil Moisture and Ocean Salinity (SMOS) mission, launched in November 2009, provides global maps of soil moisture and ocean salinity by measuring the L-band (1.4 GHz) emission of the Earth's surface with a spatial resolution of 40–50 km. Uncertainty in the retrieval of soil moisture over large heterogeneous areas such as SMOS pixels is expected, due to the non-linearity of the relationship between soil moisture and the microwave emission. The current baseline soil moisture retrieval algorithm adopted by SMOS and implemented in the SMOS Level 2 (SMOS L2) processor partially accounts for the sub-pixel heterogeneity of the land surface, by modelling the individual contributions of different pixel fractions to the overall pixel emission. This retrieval approach is tested in this study using airborne L-band data over an area the size of a SMOS pixel characterised by a mix Eucalypt forest and moderate vegetation types (grassland and crops), with the objective of assessing its ability to correct for the soil moisture retrieval error induced by the land surface heterogeneity. A preliminary analysis using a traditional uniform pixel retrieval approach shows that the sub-pixel heterogeneity of land cover type causes significant errors in soil moisture retrieval (7.7%v/v RMSE, 2%v/v bias) in pixels characterised by a significant amount of forest (40–60%). Although the retrieval approach adopted by SMOS partially reduces this error, it is affected by errors beyond the SMOS target accuracy, presenting in particular a strong dry bias when a fraction of the pixel is occupied by forest (4.1%v/v RMSE, −3.1%v/v bias). An extension to the SMOS approach is proposed that accounts for the heterogeneity of vegetation optical depth within the SMOS pixel. The proposed approach is shown to significantly reduce the error in retrieved soil moisture (2.8%v/v RMSE, −0.3%v/v bias) in pixels characterised by a critical amount of forest (40–60%), at the limited cost of only a crude estimate of the optical depth of the forested area (better than 35% uncertainty). This study makes use of an unprecedented data set of airborne L-band observations and ground supporting data from the National Airborne Field Experiment 2005 (NAFE'05), which allowed accurate characterisation of the land surface heterogeneity over an area equivalent in size to a SMOS pixel.

Error in Radar-Derived Soil Moisture due to Roughness Parameterization: An Analysis Based on Synthetical Surface Profiles

In the past decades, many studies on soil moisture retrieval from SAR demonstrated a poor correlation between the top layer soil moisture content and observed backscatter coefficients, which mainly has been attributed to difficulties involved in the parameterization of surface roughness. The present paper describes a theoretical study, performed on synthetical surface profiles, which investigates how errors on roughness parameters are introduced by standard measurement techniques, and how they will propagate through the commonly used Integral Equation Model (IEM) into a corresponding soil moisture retrieval error for some of the currently most used SAR configurations. Key aspects influencing the error on the roughness parameterization and consequently on soil moisture retrieval are: the length of the surface profile, the number of profile measurements, the horizontal and vertical accuracy of profile measurements and the removal of trends along profiles. Moreover, it is found that soil moisture retrieval with C-band configuration generally is less sensitive to inaccuracies in roughness parameterization than retrieval with L-band configuration.

Improving Spaceborne Radiometer Soil Moisture Retrievals With Alternative Aggregation Rules for Ancillary Parameters in Highly Heterogeneous Vegetated Areas

Retrieving soil moisture from spaceborne microwave radiometer observations often requires ancillary parameters such as surface vegetation opacity or vegetation water content (VWC). The conventional approach for deriving representative footprint- scale values of these parameters is to simply average the corresponding parameters of high-resolution pixels over the entire radiometer footprint. However, previous work has shown that simple averaging of ancillary parameter values may result in biased soil moisture retrievals from spaceborne radiometers. This letter uses a synthetic observing-system simulation experiment framework to theoretically demonstrate that ancillary VWC or canopy-opacity parameter values obtained with an alternative aggregation rule may significantly reduce the magnitude of soil moisture retrieval errors for highly heterogeneous vegetated areas.

Passive L-band microwave soil moisture retrieval error arising from topography in otherwise uniform scenes

Soil moisture is an important component of the water cycle and will be measured for the first time on a global scale by a dedicated passive L-band microwave radiometer that is planned for launch in 2008. Here, the contribution of topography to the error budget is examined for a vegetated scene with uniform microwave emission. Dual-polarization brightness temperature curves were generated over a range of look angles for 1-D scenes with simple geometrical features, and the soil moisture was retrieved assuming a flat surface. The errors were small for the scenarios considered. Theoretical errors were tested for realistic topography with a DEM transect of a mountainous region, and were found to be comparable. Knowledge of the mean slope from high-resolution DEM data can be used to improve the accuracy of the retrieval.

Relevance of time‐varying and time‐invariant retrieval error sources on the utility of spaceborne soil moisture products

Errors in remotely‐sensed soil moisture retrievals originate from a combination of time‐invariant and time‐varying sources. For land modeling applications such as forecast initialization, some of the impact of time‐invariant sources can be removed given known differences between observed and modeled soil moisture climatologies. Nevertheless, the distinction is seldom made when evaluating remotely‐sensed soil moisture products. Here we describe an Observing System Simulation Experiment (OSSE) for radiometer‐only soil moisture products derived from the NASA Hydrosphere States (Hydros) mission where the impact of time‐invariant errors is explicitly removed via the linear rescaling of retrievals. OSSE results for the 575,000 km2 Red‐Arkansas River Basin indicate that climatological rescaling may significantly reduce the perceived magnitude of Hydros soil moisture retrieval errors and expands the geographic areas over which retrievals demonstrate value for land surface modeling applications.

Retrieval of soil moisture from passive and active L/S band sensor (PALS) observations during the Soil Moisture Experiment in 2002 (SMEX02)

The Soil Moisture Experiments in 2002 (SMEX02) were conducted in Iowa between June 25th and July 12th, 2002. A major aim of the experiments was examination of existing algorithms for soil moisture retrieval from active and passive microwave remote sensors under high vegetation water content conditions. The data obtained from the passive and active L and S band sensor (PALS) along with physical variables measured by in situ sampling have been used in this study to demonstrate the sensitivity of the instrument to soil moisture and perform soil moisture retrieval using statistical regression and physical modeling techniques. The land cover conditions in the region studied were predominantly soybean and corn crops with average vegetation water contents ranging from 0 to ∼5 kg/m 2. The PALS microwave sensitivity to soil moisture under these vegetation conditions was investigated for both passive and active measurements. The performance of the PALS instrument and retrieval algorithms has been analyzed, indicating soil moisture retrieval errors of approximately 0.04 g/g gravimetric soil moisture. Statistical regression techniques have been shown to perform satisfactorily with soil moisture retrieval error of around 0.05 g/g gravimetric soil moisture. The retrieval errors were higher for the corn than for the soybean fields due to the higher vegetation water content of the corn crops. However, the algorithms performed satisfactorily over the full range of vegetation conditions.