Passive microwave remote sensing can provide direct and frequent measurements for the estimation of surface soil moisture globally. The future Copernicus Imaging Microwave Radiometer (CIMR) mission is projected to operate at five spectral bands, including L- and C-bands, providing a unique capability to observe surface soil moisture at multiple spatial resolutions. In this work, we investigate the potential to improve the coarser resolution of future CIMR L-band (<60 km) using finer resolution C-band (<15 km) by exploiting the overlap of band footprints. We use existing brightness temperature (TB) data from the Soil Moisture Active Passive (SMAP) mission and Advanced Microwave Scanning Radiometer 2 (AMSR2) mission to investigate L- and C-bands multiresolution information content at the global scale and assess the use of C-band information for L-band sharpening with a linear regression model. Comparing the performance of sharpened with true L-band TB, we find global improvements in systematic offset errors and time-varying random errors, especially along coastlines and over diverse vegetation land cover. Results support the conclusion that the C-band can capture information in the spatial enhancement of the L-band, demonstrating the value of future CIMR multifrequency observations to generate soil moisture products at both climatic and meteorological scales.

Soil Moisture Ocean Salinity (SMOS) mission was the first L-band radiometer launched in 2010 and is operational for retrieving global scale soil moisture and Vegetation Optical Depth (VOD). SMOS-INRA-CESBIO (SMOS-IC) version-2 is the latest retrieval algorithm for SMOS radiometers that outperforms existing SMOS retrieval algorithms. Research is underway to enhance the SMOS-IC product by improving surface roughness information that influences soil moisture and VOD retrievals. In the present study, we developed a new global parameterization of soil roughness using SMOS-IC retrievals. For this purpose, we retrieved the soil moisture and surface roughness (through the Hr parameter) values over bare soils using the SMOS-IC algorithm. A Random Forest (RF) model was trained with soil textural and terrain properties as inputs (explanatory variables) to model Hr over bare soils. Later, we extrapolated the Hr values obtained over bare soils to a global scale using the RF model. The newly calibrated Hr values were further used in the SMOS-IC algorithm for soil moisture and VOD retrievals (SMOS-IC v2.1). The SMOS-IC v2.1, the soil moisture product, is wetter than the original SMOS-IC v2 product and has improved performance compared to in situ ISMN soil moisture and modeled ECMWF soil moisture data sets. Regarding VOD, the SMOS-IC v2.1 VOD product showed improved spatial correlation with the reference aboveground biomass product. In addition, SMOS-IC v2.1 VOD product showed improved temporal correlation with MODIS NDVI over low-to-moderate vegetated regions.

The Copernicus Imaging Microwave Radiometer (CIMR) is a Copernicus Expansion Mission with an expected launch in 2028+. The satellite will carry a multi-frequency microwave radiometer operating in the L-, C-, X-, Ku-, and Ka-bands. In addition to providing L-band continuity from other missions (e.g., SMOS, SMAP), the CIMR mission brings new soil moisture remote sensing opportunities due to its multi-frequency, multi-resolution, and high temporal revisit characteristics. This study outlines a preliminary version of a soil moisture retrieval approach designed within CIMR preparatory activities. It aims to provide two soil moisture products: the first is based on the inversion of L-band brightness temperature measurements at their native resolution (<60 km). The second product is based on the inversion of enhanced resolution L-band measurements (<15 km) achieved through sharpening the L-band with higher resolution C/X-band measurements. In both cases, the soil moisture retrieval is based on the inversion of the zeroth-order tau-omega radiative transfer model. We present current efforts of algorithm development and performance evaluation based on simulated CIMR L1B data. The results presented here showcase the potential of CIMR to provide soil moisture estimates not only at hydroclimatological scales (<60 km, from L-band) but also at hydrometeorological scales (~10 to 25 km, from L-band, sharpened with C/X bands), meeting the needs of a wide range of science and applications.

Displacement of human populations due to more extreme weather hazards is a global phenomenon that leads to significant human and economic losses. Mobility related to slow-onset events such as droughts is particularly challenging to model because the start and duration of droughts are uncertain, and their effects are often intertwined with other contextual factors, such as conflicts, political stability, and undependable market prices, that are difficult to measure. Moreover, the collection of in situ socioeconomic data poses a significant challenge, where the use of alternative data sources to warn and plan for impending waves of displacement effectively could help. This study investigates the use of social media as an additional input feature to enhance drought-induced displacement predictions. A methodology for identifying human displacement based on tweet activity is proposed. Results from displacement models based on interpretable machine learning, socioeconomic data, and weather indicators consistently show the benefit of including Twitter data when applied at the district level in Somalia. The proposed approach could enhance drought-induced displacement predictions, thereby helping anticipatory action and the planning of humanitarian interventions.

Vegetation optical depth (VOD) depends on the water, structure, and biomass of vegetation. Here, we propose a multi-sensor approach to isolate the water component from the VOD and to retrieve gravimetric vegetation moisture (mg) in the western United States. The approach estimates VOD from radar and LiDAR data and minimizes the differences between these estimates and SMAP/AMSR2 VOD observations. This minimization allows to obtain the best fitting value of mgwith help of a dielectric model. Results are consistent both in space (drier vegetation in arid areas) and time (drier vegetation in drier months). The mg estimates are in the same range than in situ mg data, with some underestimation (bias ~ -0.07 kg/kg). Statistical results are reasonable (r ~ 0.45, RMSE ≤0.10 kg/kg), yet the different spatial and temporal representation of in situ and remote measurements have an impact in the direct comparisons. Our results highlight the potential for developing new vegetation moisture datasets based on VOD decomposition.

Vegetation optical depth (VOD) depends on the water, structure, and biomass of vegetation. Here, we propose a multi-sensor approach to isolate the water component from the VOD and to retrieve gravimetric vegetation moisture (mg) in the western United States. The approach estimates VOD from radar and LiDAR data and minimizes the differences between these estimates and SMAP/AMSR2 VOD observations. This minimization allows to obtain the best fitting value of mgwith help of a dielectric model. Results are consistent both in space (drier vegetation in arid areas) and time (drier vegetation in drier months). The mg estimates are in the same range than in situ mg data, with some underestimation (bias ~ -0.07 kg/kg). Statistical results are reasonable (r ~ 0.45, RMSE ≤0.10 kg/kg), yet the different spatial and temporal representation of in situ and remote measurements have an impact in the direct comparisons. Our results highlight the potential for developing new vegetation moisture datasets based on VOD decomposition.

The future Copernicus Imaging Microwave Radiometer (CIMR) mission is planned to be launched in the 2027+ time frame. At its present phase, the first version of each Algorithm Theoretical Basis Document (ATBD) must be defined. CIMR will provide observations at L (1.4 GHz), C (6.9 GHz), X (10.65 GHz), Ku (18.7 GHz) and Ka (36.5 GHz) microwave frequencies. These observations will be relevant to develop high resolution land surface products. Here we present a preliminary study with the aim of exploring the future capabilities that the synergy of CIMR frequencies can provide. Focused on the 0th-order Tau-Omega (τ-ω) model, we analysed the influence of soil roughness (H) and scattering albedo (ω) to retrieve soil moisture (SM) and vegetation optical depth (VOD) at L-band and how these parameters can be potentially estimated from higher frequency bands. We evaluated our results over CONUS, concluding that the soil roughness (H) parameter is affecting VOD and ω mainly in non-forested areas: in those areas, the increase of H produces a decrease in VOD. Our maps of ω revealed dependence with land cover type: generally, the lowest ω values were found in forested areas. Instead, our H map yielded patterns that could be mostly associated with topographic effects. Furthermore, by utilizing a depolarization index, TBdep, we discovered that its values were constrained to nearly zero (indicating minimal soil impact) in areas with vegetation, whereas in bare soils, topography had a significant influence on TBdep. We hypothesize that the use of this index could help in finding relationships among the multi-frequency information from CIMR, allowing us to understand the degree of sensitivity of each band to vegetation and topography.

This study presents a comparison between satellite-based vegetation optical depth (VOD) from multi-frequency radiometry (X-, C- and L-band), VOD-derived relative water content (RWC) and auxiliary data (e.g., evapotranspiration and soil moisture), which are investigated for their sensitivity to water status of tree canopies under dry and wet conditions for a temperate forest in Thuringia, Central Germany. For this, we estimated RWC directly from VOD normalization assuming no major changes in vegetation biomass or plant structure during the study period (2015-2019).Our results show that RWC seasonalities are aligned for all investigated frequencies showing its maximum in early summer when leaves and twigs of the top and low canopy are particularly wet and photosynthetically active. Investigating drought versus non-drought years, we observed that X-band RWC is the one better capturing drought status by exhibiting low values in the extreme drought year 2018 compared to the wet year 2017 while L-band RWC reflects the ecological memory from the extreme drought conditions in 2018 in year 2019 estimates.

Tracking seasonal dynamics of evapotranspiration (ET) across global biomes and along seasonal time periods using remote sensing is vital for monitoring ecosystem health and indicating early signals of drought. In this study, we assess the potential of adding weather and illumination-independent signals from active and passive microwave remote sensing (SAR backscatter & vegetation optical depth, VOD) to the established set of ET products, like from optical/thermal remote sensing (MODIS, SEVIRI) and reanalysis (ERA-5 land, GLDAS) data.Our study covers a four-year period (2017-2020), including dry (2018 & 2019) and wet (2017) years. The study was conducted over eight ICOS sites across Europe. These sites are predominantly forested with a low biomass dynamic over the observation period.We find that the ET products from in situ Eddy Covariance (EC), MODIS, and GLDAS deviate relatively minor along the seasons (< 1 [mm/day]), but differ between years. Here, the years (2017-2020) indicate a slightly different ET rate between in situ measurements (EC) and derived products (MODIS & GLDAS), which is currently being investigated. The microwave-based indicators (backscatter & VOD) are proxies by their nature and serve as first-order indicators of relative dynamics allowing the identification of seasonal patterns of ET as well as their spatio-temporal anomalies along both dry and wet years.

Food security is at stake, with climate change heavily impacting agriculture and food production. In the present context of extreme events and changing conditions, developing advanced crop yield models can learn from all available information, and providing interpretable predictions for decision-making is of paramount relevance. This work explores the potential and limitations of developing interpretable crop yield models using long short-term memory (LSTM) neural networks, which typically excel at extracting information from time series. LSTMs were designed and trained with multisource satellite and meteorological time series over Continental US (CONUS) and corn, soybean, and wheat yield data from the US Department of Agriculture. Two recent attribution methods are used to interpret and extract knowledge from the developed models: integrated gradients (IG), based on back-propagation, and Shapley (SHAP) values, based on perturbations. Our results show that: 1) LSTM models achieved high accuracy ($\text {R}^{2}>0.56$); 2) multisource combinations outperformed single-variable models in all crop models; 3) both attribution methods were equivalent in detecting essential drivers and their contribution; 4) satellite estimates of enhanced vegetation index (EVI) and vegetation optical depth (VOD) together with meteorological estimates of maximum temperature (TMX) were the most relevant input features for crop yield estimations; and finally and 5) we discovered critical periods of the crop growth cycle for the corn, soybean, and wheat models. The suggested strategy constitutes an important step toward modeling and understanding crop production systems and advancing in automatic data-driven and accountable field management.

Satellite-derived soil moisture (SM) products have become an important information source for the study of land surface processes in hydrology and land monitoring. Characterizing and estimating soil memory and persistence from satellite observations is of paramount relevance, and has deep implications in ecology, water management, and climate modeling. In this work, we address the problem of SM persistence estimation from microwave sensors using several autocorrelation metrics that, unlike traditional approaches, build on accurate estimates of the autocorrelation function from nonuniformly sampled time series. We show how the choice of the autocorrelation estimator can have a dramatic impact on the SM persistence metrics derived thereof, particularly given the nonuniform nature of satellite observations, yet this fact has been overlooked to a large extent in literature. We give empirical evidence of performance using ground-based SM measurements, as well as L-band Soil Moisture and Ocean Salinity (SMOS) and C-band [Advanced Microwave Scanning Radiometer-2 (AMSR2), Advanced Scatterometer (ASCAT)] remotely sensed SM data. Experiments along transects allow us to scrutinize the inter-method consistency and the spatial–temporal characteristics of autocorrelation estimators. This motivates the introduction of novel measures of spatial–temporal autocorrelation and allows us to retrieve improved persistence estimates. Results over the Iberian Peninsula indicate the SM persistence patterns captured by L- and C-band microwave sensors over semiarid regions exhibit spatially concurrent patterns of persistence and support their combination in long-term data records. We conclude that accounting for the nonuniform nature of the satellite time series using robust autocorrelation estimations allows providing improved measures and spatial descriptions of SM persistence.

Incidence angle diversity of space-borne radiometer and radar systems operating at low microwave frequencies needs to be taken into consideration to accurately estimate soil moisture (SM) across spatial scales. In this study, the Single Channel Algorithm (SCA) is first applied to SMOS brightness temperatures at vertical polarization (TBV) to estimateSMat coarse-resolution (25 km) and develop a land cover-specific and incidence angle (32.5°, 42.5° and 52.5°)-adaptive calibration of single scattering albedo (ω) and soil roughness (hs) parameters. These effective parameters are used together with fine-scale multi-angular Sentinel-1 backscatter in a single-pass active-passive downscaling approach to estimateTBVat fine-scale (1 km) for each SMOS incidence angle. TheseTBVare finally inverted to obtain the corresponding high-resolutionSMmaps. Results over the Iberian Peninsula for year 2018 show an increasing trend of ω and a decreasing trend ofhswith SMOS incidence angle, with almost no variability of ω across land cover types. The active-passive covariation parameter is shown to increase with SMOS incidence angle and decrease with Sentinel-1 incidence angle. Coarse and fineTBVmaps from the three SMOS incidence angles show similar distributions (mean differences below 0.38 K). Resulting high-resolutionSMmaps have maximum differences in mean and standard deviation of 0.016 and 0.015 m3/m3, respectively, and compare well within situmeasurements. Our results indicate that model-based microwave approaches to estimateSMcan be adequately adapted to account for the incidence angle diversity of planned missions such as CIMR, ROSE-L and Sentinel-1 next generation.

The modeling of Earth observation (EO) data is a challenging problem, typically approached by either purely mechanistic or purely data-driven methods. Mechanistic models encode the domain knowledge and physical rules governing the system. Such models, however, need the correct specification of all interactions between variables in the problem and the appropriate parameterization is a challenge in itself. On the other hand, machine learning approaches are flexible data-driven tools, able to approximate arbitrarily complex functions, but lack interpretability and struggle when data are scarce or in extrapolation regimes. In this article, we argue thathybrid learning schemesthat combine both approaches can address all these issues efficiently. We introduceGaussian process (GP) convolution modelsfor hybrid modeling in EO problems. We specifically propose the use of a class of GP convolution models calledlatent force models(LFMs) for EO time series modeling, analysis, and understanding. LFMs are hybrid models that incorporate physical knowledge encoded in differential equations into a multioutput GP model. LFMs can transfer information across time series, cope with missing observations, infer explicit latent functions forcing the system, and learn parameterizations which are very helpful for system analysis and interpretability. We illustrate the performance in two case studies. First, we consider time series of soil moisture (SM) from active Advanced Scatterometer (ASCAT) and passive [SM and ocean salinity (SMOS), advanced microwave scanning radiometer-2 (AMSR2)] microwave satellites. We show how assuming a first-order differential equation as governing equation, the model automatically estimates the e-folding time or decay rate related to SM persistence and discovers latent forces related to precipitation. In the second case study, we show how the model can fill in gaps of leaf area index (LAI) and Fraction of Absorbed Photosynthetically Active Radiation (fAPAR) from moderate resolution imaging spectroradiometer (MODIS) optical time series by exploiting their relations across different spatial and temporal domains. The proposed hybrid methodology reconciles the two main approaches in remote-sensing parameter estimation by blending statistical learning and mechanistic modeling.

Machine learning models alone are excellent approximators, but very often do not respect the most elementary laws of physics, like mass or energy conservation, so consistency and confidence are compromised. In this paper we describe the main challenges ahead in the field, and introduce several ways to live in the Physics and machine learning interplay: encoding differential equations from data, constraining data-driven models with physics-priors and dependence constraints, improving parameterizations, emulating physical models, and blending data-driven and process-based models. This is a collective long-term AI agenda towards developing and applying algorithms capable of discovering knowledge in the Earth system.

The attenuation of microwave emissions through the canopy is quantified by the vegetation optical depth (VOD), which is related to the amount of water, the biomass and the structure of vegetation. To provide microwave-derived plant water estimates, one must account for biomass/structure contributions in order to extract the water component from the VOD. This study uses Aquarius scatterometer data to build an L-band global seasonality of vegetation volume fraction (δ), representative of biomass/structure dynamics. The dynamic range of δ is adapted for its application in a gravimetric moisture (Mg) retrieval model. Results show that δ ranging from 0 to 3.35.10-4is needed for modelling physically reasonable Mg values. The global average of δ shows consistent spatial patterns across vegetation distributions, and δ seasonality is coherent with the phenology of the studied vegetation types. These findings enable the separation of information on vegetation water and biomass/structure inherent within VOD.

The present study focuses on retrieving soil and canopy temperatures, which are key parameters to estimate soil moisture and vegetation optical depth from multi-frequency microwaves information. Several retrieval algorithms assume that canopy and vegetation temperatures are similar in thermal equilibrium conditions, while others separate their contributions, as SMOS-IC, one of the consolidated retrieval algorithms for the Soil Moisture and Ocean Salinity (SMOS) satellite mission. Soil and canopy temperatures in SMOS-IC are modelled from the ECMWF (European Centre for Medium-Range Weather Forecasts) centre. Both SMOS and the Soil Moisture Active Passive (SMAP) missions are currently the only passive L-band (1.4 GHz) missions in operation, but their lifetime is limited. In this context, the upcoming Copernicus Imaging Microwave Radiometer (CIMR) mission will provide continuity on L-band measurements with complementary information in a range of microwave frequencies, from 1.4 to 36.5 GHz. This study uses in situ soil moisture information from the International Soil Moisture Network (ISMN) as input in the SMOS-IC algorithm to retrieve vegetation optical depth (VOD) and soil/canopy effective temperature (TGC). The retrieved effective temperature is then compared with modelled temperatures from ECMWF and with data from the Advanced Microwave Scanning Radiometer 2 (AMSR2), which acquires the higher frequency bands (C, X, Ka, and Ku) present in the future CIMR mission. Results confirm the potential of all high-frequency bands to estimate TGC, with C and X-bands being the most correlated. This study is a first approach to evaluate how microwave multi-frequency information can help modelling soil and canopy temperatures in the SMOS-IC retrieval algorithm, from which the upcoming CIMR mission may benefit.

A retrieval methodology for forest water potential from ground-based L-band radiometry is proposed. It contains the estimation of the gravimetric and the relative water content of a forest stand and tests in situ- and model-based functions to transform these estimates into forest water potential. The retrieval is based on vegetation optical depth data from a tower-based experiment of the SMAPVEX 19–21 campaign for the period from April to October 2019 at Harvard Forest, MA, USA. In addition, comparison and validation with in situ measurements on leaf and xylem water potential as well as on leaf wetness and complex permittivity are foreseen to understand limitations and potentials of the proposed approach. As a first result the radiometer-based water potential estimates of the forest stand are concurrent in time and similar in value with their in situ (xylem) counterparts from single trees in the radiometer footprint.

Agriculture monitoring, and in particular food security, requires near real-time information on crop growing conditions for early detection of possible production deficits. In this work, we propose the use of Gaussian processes (GPs). together with in-situ, EO and ERA-Interim climate reanalysis data for crop yield forecasting. Country-level agricultural survey data from FAOSTAT are used for quantitative assessment. The study is conducted in the framework of the ASAP (Anomaly hot Spots of Agricultural Production) early warning decision support system of the European Commission, which aims at providing timely information about possible crop production anomalies worldwide. After grouping countries with similar growing season periods, we We show that GP models allow predicting the yield of the different planted crops within such groups with coefficient of determination R2 ranging from 0.5 to 0.95. For each country and crop, a better fit that the mean of the data is obtained in all cases, and the errors obtained are comparable to the ones obtained for the group. The proposed modelling framework can be potentially adopted in ASAP operations to forecast national level crop yield and production.

Incidence angle diversity of space-borne L-band radiometers needs to be taken into account for a consistent estimation of surface soil moisture (SM). In this study, the Land Parameter Retrieval Model (LPRM) is applied to SMOS brightness temperatures to calibrate the effective scattering albedo (w) and the soil roughness (h1) parameter against ERA5-land SM. The analysis is carried out for SMOS data at three different incidence angles ($32.5\pm 5^{\circ},\ 42.5\pm 5^{\circ}$and$52.5\pm 5^{\circ}$) focusing in 2016 on the three main land cover types of the Iberian Peninsula according to the Climate Change Initiative (agricultural, forest and grassland). The parameterization shows an increasing trend of w and h1with rise of incidence angle. The SM retrieval have been evaluated with in situ SM measurements of the REMEDHUS network on rainfed crop fields. Both compare well at the three incidence angles, obtaining high correlations (0.81-0.85), an ubRMSE around 0.04 m3m−3and low bias (0-0.015 m3m−3).

This work introduces the use of Gaussian processes (GPs) for the estimation and understanding of crop development and yield using multisensor satellite observations and meteorological data. The proposed methodology combines synergistic information on canopy greenness, biomass, soil, and plant water content from optical and microwave sensors with the atmospheric variables typically measured at meteorological stations. A composite covariance is used in the GP model to account for varying scales, nonstationary, and nonlinear processes. The GP model reports noticeable gains in terms of accuracy with respect to other machine learning approaches for the estimation of corn, wheat, and soybean yields consistently for four years of data across continental U.S. (CONUS). Sparse GPs allow obtaining fast and compact solutions up to a limit, where heavy sparsity compromises the credibility of confidence intervals. We further study the GP interpretability by sensitivity analysis, which reveals that remote sensing parameters accounting for soil moisture and greenness mainly drive the model predictions. GPs finally allow us to identify climate extremes and anomalies impacting crop productivity and their associated drivers.

In this article, we present a multipolarimetric estimation approach for two model-based vegetation structure parameters (shape A and orientation distribution ψ of the main canopy elements). The approach is based on a reduced observation set of three incoherent (no phase information) polarimetric backscatter intensities (|SHH|2, |SHV|2, and |SVV|2) combined with a two-parameter (APand ψ) discrete scatterer model of vegetation. The objective is to understand whether this confined set of observations contains enough information to estimate the two vegetation structure parameters from the L-band radar signals. In order to disentangle soil and vegetation scattering influences on these signals and ultimately perform a vegetation only retrieval of vegetation shape A and orientation distribution ψ, we use the subpixel spatial heterogeneity expressed by the covariation of co- and cross-polarized backscatter ΓPP-PQof the neighboring cells and assume it is indicative for the amount of a vegetation-only co-to-cross-polarized backscatter ratio μPP-PQ. The ratio-based retrieval approach enables a relative (no absolute backscatter) estimation of the vegetation structure parameters which is more robust compared to retrievals with absolute terms. The application of the developed algorithm on global L-band Soil Moisture Active Passive (SMAP) radar data acquired from April to July 2015 indicates the potential and limitations of estimating these two parameters when no fully polarimetric data are available. A focus study on six different regions of interest, spanning land cover from barren land to tropical rainforest, shows a steady increase in orientation distribution toward randomly oriented volumes and a continuous decrease in shape arriving at dipoles for tropical vegetation. A comparison with independent data sets of vegetation height and above-ground biomass confirms this consistent and meaningful retrieval of APand ψ. The retrieved shapes and orientation distributions represent the main vegetation elements matching the literature results from model-based decompositions of fully polarimetric L-band data at the SMAP spatial resolution. Based on our findings, APand ψ can be directly applied for parameterizing the vegetation scattering component of model-based polarimetric decompositions. This should facilitate decomposition into ground and vegetation scattering components and improve the retrieval of soil parameters (moisture and roughness) under vegetation.

In this work we propose the use of Long Short-Term Memory (LSTM) Recurrent Neural Networks for multivariate time series of satellite data for crop yield estimation. Recurrent nets allow exploiting the temporal dimension efficiently, but interpretability is hampered by the typically overparameterized models. The focus of the study is to understand LSTM models by looking at the hidden units distribution, the impact of increasing network complexity, and the relative importance of the input covariates. We extracted time series of three variables describing the soil-vegetation status in agroe-cosystems -soil moisture, VOD and EVI- from optical and microwave satellites, as well as available in situ surveys on crops across Continental U.S. to perform the experiments. Firstly, the models were validated in error terms. Secondly, the trained models were visualized and, thirdly, some useful statistics were extracted from the hidden unit activation heatmaps, accounting for redundancy and cluttering of activation responses. Results reveal how networks assign most of the relevance to soil moisture and focus on two phenological stages of crop growth.

Remote sensing observations, products, and simulations are fundamental sources of information to monitor our planet and its climate variability. Uncovering the main modes of spatial and temporal variability in Earth data is essential to analyze and understand the underlying physical dynamics and processes driving the Earth System. Dimensionality reduction methods can work with spatio-temporal data sets and decompose the information efficiently. Principal component analysis (PCA), also known as empirical orthogonal functions (EOFs) in geophysics, has been traditionally used to analyze climatic data. However, when nonlinear feature relations are present, PCA/EOF fails. In this article, we propose a nonlinear PCA method to deal with spatio-temporal Earth system data. The proposed method, called rotated complex kernel PCA (ROCK-PCA for short), works in reproducing kernel Hilbert spaces to account for nonlinear processes, operates in the complex kernel domain to account for both space and time features, and adds an extra rotation for improved flexibility. The result is an explicitly resolved spatio-temporal decomposition of the Earth data cube. The method is unsupervised and computationally very efficient. We illustrate its ability to uncover spatio-temporal patterns using synthetic experiments and real data. Results of the decomposition of three essential climate variables are shown: satellite-based global gross primary productivity (GPP), soil moisture (SM), and reanalysis sea surface temperature (SST) data. The ROCK-PCA method allows identifying their annual and seasonal oscillations, as well as their nonseasonal trends and spatial variability patterns. The main modes of variability of GPP and SM match expected distributions of land-cover and eco-hydrological zones, respectively; the interannual component of SM is shown to be highly correlated with El Niño Southern Oscillation (ENSO) phenomenon; and the SST annual oscillation is perfectly uncoupled in magnitude and phase from the global warming trend and ENSO anomalies, as well as from their mutual interactions. We provide the working source code of the presented method for the interested reader in https://github.com/DiegoBueso/ROCK-PCA.

A considerable amount of water can be stored in vegetation, especially in regions experiencing large quantities of precipitation (mid-latitudes). In this context, an accurate estimate of the actual water status of the vegetation could lead to an improved understanding of the effect of plant water on the water budget. In this study, we developed and validated a novel approach to retrieve the vegetation volume fraction (δ) (i.e., volume percentage of solid plant material of a canopy in air) and the gravimetric vegetation water content (mg) (i.e., amount of water per wet biomass) for winter wheat. The estimation was based on the attenuation of L-band microwave measurements through vegetation (vegetation optical depth, (τ)-parameter). Ground-based L-band microwave measurements over an entire growing cycle together with in situ measured vegetation characteristics have been used for this purpose. Retrieved δ- and mg-values revealed to be comparable to literature and in situ measurements (i.e., δ was within the range between 0 and 0.01 and the retrieved mghad a mean value of 0.58 (0.55 (in situ) and 0.54 (literature))). Finally, we also tested the sensitivity of the δ- and mg-retrievals to their input-values to investigate their possible mutual dependencies. The analysis showed that already small changes in the input-mgor -δ result in relatively large changes in the retrieved δ or mg.

Mapping carbon stocks in the tropics is essential for climate change mitigation. Passive microwave remote sensing allows estimating carbon from deep canopy layers through the Vegetation Optical Depth (VOD) parameter. Although their spatial resolution is coarser than that of optical vegetation indices or airborne Lidar data, microwaves present a higher penetration capacity at low frequencies (L-band) and avoid cloud masking. This work compares the relationships of airborne carbon maps in Central and South America with both (i) SMAP L-band VOD at 9 km gridding and (ii) MODIS Enhanced Vegetation Index (EVI). Models to estimate carbon stocks are built from these two satellite-derived variables. Results show that L-band VOD has a greater capacity to model carbon variability than EVI. The resulting VOD-derived carbon estimates are further presented at a detailed (9 km) spatial scale.

Global Soil Moisture and Ocean Salinity (SMOS) Level 3 (L3) soil moisture (SM) products are being routinely distributed by the Barcelona Expert Centre (BEC). The quality and accuracy of these SM products have been demonstrated not only by direct validation, but also by its adoption in a wide range of applications. Recently, changes in SMOS Level 2 (L2) SM have led to the reprocessing of the BEC SMOS L3 SM. As in previous versions, a filtering and a weighted binning based on the uncertainty of the SM retrievals by means of the Data Quality Index (DQX) was applied for the L3 production. However, the DQX was modified in the latest L2 release (v650), which could possibly have an influence in the performance of the derived products.This study assesses the impact of the current DQX-based BEC L3 SM quality filtering and binning approach and the possibility of using an alternative strategy based on the chi-squared (χ2) parameter, which is defined as the cost function of the retrieval. The study is performed over continental USA using in situ SM from the U.S. Climate Reference Network (USCRN) as a benchmark. In both approaches, similar results were obtained in terms of correlation and unbiased root mean square difference (ubRMSD). Nevertheless, the χ2-based L3 SM is in general slightly wetter and has a lower dry bias than the DQX-based L3 SM. Further assessments will be performed to stablish the optimal filtering/binning of BEC SMOS L3 SM products.

In many remote sensing applications, one wants to estimate variables or parameters of interest from observations. When the target variable is available at a resolution that matches the remote sensing observations, standard algorithms, such as neural networks, random forests, or the Gaussian processes, are readily available to relate the two. However, we often encounter situations where the target variable is only available at the group level, i.e., collectively associated with a number of remotely sensed observations. This problem setting is known in statistics and machine learning as multiple instance learning (MIL) or distribution regression (DR). This article introduces a nonlinear (kernel-based) method for DR that solves the previous problems without making any assumption on the statistics of the grouped data. The presented formulation considers distribution embeddings in reproducing kernel Hilbert spaces and performs standard least squares regression with the empirical means therein. A flexible version to deal with multisource data of different dimensionality and sample sizes is also presented and evaluated. It allows working with the native spatial resolution of each sensor, avoiding the need for matchup procedures. Noting the large computational cost of the approach, we introduce an efficient version via random Fourier features to cope with millions of points and groups. Real experiments involve the Soil Moisture Active Passive (SMAP) vegetation optical depth (VOD) data for the estimation of crop production in the U.S. Corn Belt and the Moderate Resolution Imaging Spectroradiometer (MODIS) and Multi-angle Imaging SpectroRadiometer (MISR) reflectances for the estimation of aerosol optical depth (AOD). An exhaustive empirical evaluation of the method is done against naive (linear and nonlinear) approaches based on input-space means as well as previously presented methods for MIL. We provide source code of our methods in http://isp.uv.es/code/dr.html.

In this study, the τ-ω model framework is used to derive extinction coefficient and canopy penetration depths from multi-frequency SMAP and AMSR2 retrievals of vegetation optical depth together with ICESat LiDAR vegetation heights. The vegetation extinction coefficient serves as an indicator of how strong absorption and scattering processes within the canopy attenuate microwaves at L and C-band. Through inversion of the extinction coefficient, the penetration depth into the canopy can be obtained, which is analyzed on local (Sahel, Illinois) and continental scale (Africa, parts of North America) as well as for a one year time series (04/2015-04/2016). First analyses of the retrieved penetration depth estimates reveal strongest attenuation for densely forested areas, therefore vegetation attenuation should be accounted for when retrieving soil moisture in these areas. For the continents of North America and Africa penetration depths decrease in average with an increase in frequency from L- to C-band. Moreover penetration depth time series were found to match with expected seasonal variations (e.g. vegetation growth period & rainy season) for analyzed local regions.

Soil moisture (SM) is a key state variable of the hydrological cycle, needed to monitor the effects of a changing climate on natural resources. Soil moisture is highly variable in space and time, presenting seasonalities, anomalies and long-term trends, but also, and important nonlinear behaviours. Here, we introduce a novel fast and nonlinear complex PCA method to analyze the spatio-temporal patterns of the Earth's surface SM. We use global SM estimates acquired during the period 2010-2017 by ESA's SMOS mission. Our approach unveils both time and space modes, trends and periodicities unlike standard PCA decompositions. Results show the distribution of the total SM variance among its different components, and indicate the dominant modes of temporal variability in surface soil moisture for different regions. The relationship of the derived SM spatio-temporal patterns with E1 Niño Southern Oscillation (ENSO) conditions is also explored.

Global change is increasing the risk of forest decline worldwide, impacting carbon and water cycles. Hence, there is an urgent need for predicting forest decline occurrence. To that purpose, this study links forest decline events in Catalonia, detected by the DEBOSCAT forest monitoring program, with information from the Soil Moisture and Ocean Salinity (SMOS) satellite. Firstly, this study reviews the role of the SMOS soil moisture in a previous forest decline episode occurred in 2012, where the authors concluded that dry soils increased the probability of observing decline in broadleaved forests. Secondly, the present study detects that forest decline in 2012 and 2016 was linked to very dry soil conditions (generally with SM3·m-3). A similar analysis is proposed using SMOS Vegetation Optical Depth (VOD) data, which is a proxy of vegetation hydric status. Results and preliminary models will be presented at IGARSS 2018.

Vegetation Optical Depth (VOD) at L-band is highly sensitive to the water content and above-ground biomass of vegetation. Hence, it has great potential for monitoring crop phenology and for providing crop yield forecasts. Recently, the Multi-Temporal Dual Channel Algorithm (MT -DCA) has been proposed to retrieve L-band VOD from Soil Moisture Active Passive (SMAP) measurements. In previous research, SMAP VOD has been compared to crop phenology and has been used to derive crop yield estimates. Here, we review and expand these initial research studies. In particular, we quantify the capability of VOD to detect different crop stages, and test different VOD metrics (i.e., maximum, range and integrals of VOD) to provide crop yield estimates in the United States Corn Belt. Results show that VOD captures 50% to 70% of crop changes during growing and maturing phases, and that it explains between 44% (in heterogeneous crop regions) and 74% (in homogenous croplands) of final crop yields.

Estimating parameters for global climate models via combined active and passive microwave remote sensing data has been a subject of intensive research in recent years. A variety of retrieval algorithms has been proposed for the estimation of soil moisture, vegetation optical depth and other parameters. A novel attenuation-based retrieval approach is proposed here to globally estimate the gravimetric moisture of vegetation (mg) and retrieve information about the amount of water [kg] per amount of wet vegetation [kg]. The parameter mgis particularly interesting for agro-ecosystems, to assess the status of growing vegetation. The key feature of the proposed approach is that it relies on multi-sensor data from three sensor types (microwave radar, microwave radiometer, and lidar) to solve the physics equations and obtain mg-estimates. The comparability of these estimates to literature values as well as to results of a globally applied, retrieval approach of Grant [4], reveal the potential of the developed method.

The European Space Agency's (ESA) SMOS mission, in orbit since more than 8 years, carries a passive microwave interferometric radiometer measuring in L-Band and provides accurate global observations of emitted radiation originating from the Earth's surfaces since the atmosphere is almost transparent in this spectral range. In addition, over land the effect of vegetation on the measurements is smaller than for shorter wavelengths. The scientific objectives of the SMOS mission directly respond to the need for global observations of soil moisture and ocean salinity, two key variables used in predictive hydrological, oceanographic and atmospheric models. SMOS observations also provide information on the characterisation of ice and snow covered surfaces and the sea ice effect on ocean-atmosphere heat fluxes and dynamics, which affects large-scale processes of the Earth's climate system.

Measurements of soil moisture are needed for a better global understanding of the land surface-climate feedbacks at both the local and the global scale. Satellite sensors operating in the low frequency microwave spectrum (from 1 to 10 GHz) have proven to be suitable for soil moisture retrievals. These sensors now cover nearly 4 decades thus allowing for global multi-mission climate data records. In this paper, we assess the possibility of using L-band (SMOS) and C-band (AMSR2, ASCAT) remotely sensed soil moisture time series for the global estimation of soil moisture persistence. A multi -output Gaussian process regression model is applied to ensure spatio-temporal coverage of the satellite data sets. It allows a robust computation of temporal autocorrelation and e- folding times. Results over a selection of catchments reveals general agreement between the response of in-situ and satellite microwave observations to hydrological processes. The response of the uppermost-modeled soil moisture layer of GLDAS-1-Noah agrees well with that of the observations, whereas major differences are displayed by MERRA2 reanalysis. The temporal dynamics of the three microwave sensors are shown to be consistent, close to in-situ and to GLDAS-1- Noah, which supports their combination for the global estimation soil moisture persistence.

After more than 8 years in orbit the Soil Moisture and Ocean Salinity (SMOS) satellite is still in good health and several algorithms for improving its spatial resolution have been proposed and validated in a variety of catchments. However, none of them has yet been applied at the global scale. In this article we present: i) a review of the latest SMOS-BEC downscaling algorithm, which allows for its global application using an adaptive moving window and ii) a thorough validation of the resulting maps over two in-situ networks: REMEDHUS in Spain and OzNet in Australia. The proposed algorithm combines SMOS brightness temperatures (at ~40 km spatial resolution), and MODIS-derived Land Surface Temperature and Normalized Differenced Vegetation Index (at 1 km), into 1km soil moisture maps. This paper also presents a variant of the algorithm, which allows for cloud-free retrievals. A statistical comparison has been carried out when the MODIS Land Surface Temperature is replaced in the algorithm by the one provided by the ERA5 reanalysis. Fine-scale estimates show good agreement in terms of correlation and root-mean-squared error with in-situ soil moisture.

The soil drying process is a challenging framework to assess climatic, hydrologic and ecosystem processes. This work develops a seasonal analysis of temporal e-folding decay ( τ) of surface soil moisture dry-downs using ORCHIDEE land surface model and SMOS observations over South Eastern South America (SESA). Results show that the soil drying process depends on both location and season, and that the modeled drying velocity is faster than the observed one, even when modeled data is sampled at the same frequency as the observations. Differences between observed and modeled data have been found both in the analysis of the regional overall τ and in the spatial patterns of τ estimates. A seasonal timescale analysis is presented here for further insight of soil moisture dry-downs, allowing to reconcile differences between observations and models.

The Radar Vegetation Index (RVI) is widely applied to indicate vegetation cover. The index includes the backscattering intensities of co- and cross-polarization that do not only contain information coming from vegetation scattering at longer wavelength (L-band), but also from the soil underneath. A forward modelling approach using active and passive microwave-derived parameters to obtain the scattering contribution of the soil is pursued. The idea of this research study is a subtraction of the attenuated soil scattering contribution from the measured backscattering intensities, to provide a clean vegetation-based solution, called improved RVI (RVII). For latter analysis, the vegetation volume is forward modeled to calculate vegetation-only RVI-values without any soil scattering contribution. It reveals that, the pre-factor of the standard RVI leads to values up to 1.2, unfavorable for a normalized index running between zero and one. Hence, improvements for the standard RVI equation are proposed here to obtain a better suited value range and for incorporating soil scattering influences and filtering of regions with dominant soil scattering. Moreover, the improved RVI (RVII) is compared with datasets of vegetation and soil parameters (e.g. vegetation water content) for correlation analysis to find the physical parameters contributing to the index.

In this study the framework of the τ - ω model is used to derive vegetation loss coefficients and canopy penetration depths from SMAP multi-temporal retrievals of vegetation optical depth, single scattering albedo and ICESat lidar vegetation heights. The vegetation loss coefficients serve as a global indicator of how strong absorption and scattering processes attenuate L-band microwave radiation. By inverting the vegetation loss coefficients, penetration depths into the canopy can be obtained, which are displayed for the global forest reservoirs. A simple penetration index is formed combining vegetation heights and penetration depth estimates. The distribution and level of this index reveal that for densely forested areas in the tropics the soil signal is attenuated considerably, and this attenuation must be carefully accounted for in soil moisture retrieval algorithms.

Vegetation Optical Depth (VOD) is related to Vegetation Water Content (VWC). This provides new and highly valuable information for ecological and agricultural studies. In this work, VOD from the Soil Moisture Active-Passive (SMAP) satellite has been retrieved with the new Multi-Temporal Dual-Channel Algorithm (MT-DCA). Then, it has been applied to the study of crop yield trends and crop composition. The increase on VOD (ΔVOD) during crop development has been compared to yield data in two selected regions located in the United States. The first region presents a heterogeneous crop composition and weak ΔVOD-yield relationship (r2=0.21). The second region presents a highly homogenous cover and a strong exponential relationship (r2=0.65) between ΔVOD and yield. A saturation of yield is observed at a certain ΔVOD value. This pattern is probably due to an increasing plant density, which limits the crop yield due to plant physiological stress.

Over land the vegetation canopy affects the microwave brightness temperature by emission, scattering and attenuation of surface soil emission. The questions addressed in this study are: 1) what is the transparency of the vegetation canopy for different biomes around the Globe at the low-frequency L-band?, 2) what is the seasonal amplitude of vegetation microwave optical depth for different biomes?, 3) what is the effective scattering at this frequency for different vegetation types?, 4) what is the impact of imprecise characterization of vegetation microwave properties on retrieval of soil surface conditions? These questions are addressed based on the recently completed one full annual cycle measurements by the NASA Soil Moisture Active Passive (SMAP) measurements.

Vegetation optical depth and scattering albedo are crucial parameters within the widely used τ-ω model for passive microwave remote sensing of vegetation and soil. A multi-sensor data integration approach using ICESat lidar vegetation heights and SMAP radar as well as radiometer data enables a direct retrieval of the two parameters on a physics-derived basis. The crucial step within the retrieval methodology is the calculus of the vegetation scattering coefficient KS, where one exact and three approximated solutions are provided. It is shown that, when using the assumption of a randomly oriented volume, the backscatter measurements of the radar provide a sufficient first order estimate and subsequently lead to effective estimates of vegetation optical depth and scattering albedo acquired with the novel multi-sensor approach.

The ESA's SMOS and the NASA's SMAP missions, launched in 2009 and 2015, respectively, are the first two missions having on-board L-band microwave sensors, which are very sensitive to the water content in soils and vegetation. Focusing on the vegetation signal at L-band, we have implemented an inversion approach for SMAP that allows deriving vegetation optical depth (VOD, a microwave parameter related to biomass and plant water content) alongside soil moisture, without reliance on ancillary optical information on vegetation. This work aims at using this new observational data to monitor the phenology of crops in major global agro-ecosystems and enhance present agricultural monitoring and prediction capabilities. Core agricultural regions have been selected worldwide covering major crops (corn, soybean, wheat, rice). The complementarity and synergies between the microwave vegetation signal, sensitive to biomass water-uptake dynamics, and optical indices, sensitive to canopy greenness, are explored. Results reveal the value of L-band VOD as an independent ecological indicator for global terrestrial biosphere studies.1

The ESA's Soil Moisture and Ocean Salinity (SMOS, 2009-2017) is the first mission using L-band radiometry to monitor the Earth's global surface soil moisture (SM). After more than 7 years in orbit, many studies have contributed to improving the quality and applicability of SMOS-derived SM maps. In this research, a novel downscaling algorithm is proposed for retrieving high resolution (1 km) SM. This model is an extension of the “universal triangle” technique, and also introduces the concept of adaptive moving window. Its inputs are the low resolution SMOS BEC L3 SM and the brightness temperatures at vertical and horizontal polarizations (SMOS L1C), and the high resolution NDVI and LST from optically-based sensors. The proposed method allows obtaining high resolution SM maps worldwide, with no limitation in extension.

Soil moisture impacts exchanges of water, energy and carbon fluxes between the land surface and the atmosphere. Passive microwave remote sensing at L-band can capture spatial and temporal patterns of soil moisture in the landscape. Both ESA and NASA have launched L-band radiometers, in the form of the SMOS and SMAP satellites respectively, to monitor soil moisture globally, every 3-day at about 40 km resolution. However, their coarse scale restricts the range of applications. While SMAP included an L-band radar to downscale the radiometer soil moisture to 9 km, the radar failed after 3 months and this initial approach is not applicable to developing a consistent long term soil moisture product across the two missions anymore. Existing optical-, radiometer-, and oversampling-based downscaling methods could be an alternative to the radar-based approach for delivering such data. Nevertheless, retrieval of a consistent high resolution soil moisture product remains a challenge, and there has been no comprehensive intercomparison of the alternate approaches. This research undertakes an assessment of the different downscaling approaches using the SMAPEx-4 field campaign data.

An application of the Soil Moisture Agricultural Drought Index (SMADI) for global agricultural drought monitoring is presented. The index integrates surface soil moisture from the Soil Moisture and Ocean Salinity (SMOS) mission with the Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) and Normalized Difference Vegetation Index (NDVI) and allows for global drought monitoring at medium spatial scales (0.05°). Biweekly maps of SMADI were obtained from year 2010 to 2015 over all agricultural areas on Earth. The SMADI time-series were compared with state-of-the-art drought indices over the Iberian Peninsula. Results show a good agreement between SMADI and the Crop Moisture Index (CMI) retrieved at five weather stations (with correlation coefficient, R from -0.64 to -0.79) and the Soil Water Deficit Index (SWDI) at the Soil Moisture Measurement Stations Network of the University of Salamanca (REMEDHUS) (R=-0.83). Some preliminary tests were also made over the continental United States using the Vegetation Drought Response Index (VegDRI), with very encouraging results regarding the spatial occurrence of droughts during summer seasons. Additionally, SMADI allowed to identify distinctive patterns of regional drought over the Indian Peninsula in spring of 2012. Overall results support the use of SMADI for monitoring agricultural drought events world-wide.

The NASA Soil Moisture Active Passive (SMAP) mission aims at producing high-resolution (9 km) global maps of surface soil moisture based on L-band radar and radiometer measurements. In this study, a physically-based retrieval of the active-passive covariation parameter β from one active-passive (single-pass) SMAP acquisition couple is proposed, circumventing empirical time-series regressions. The key to single-pass retrieval of β is the vegetation correction of the backscatter signal. This can be achieved by use of the measured cross-polarized backscatter signal and parameters appropriately describing the structure of the vegetation volume. These parameters can be derived from the observed Γ-parameters of the SMAP baseline algorithm enabling a fully SMAP data-driven, single-pass estimation of the covariation parameter β without any auxiliary information. Moreover, vegetation structural parameters, indicative of preferential vegetation shape and orientation, are retrieved using the observed Γ-parameters.

New ocean products from the Soil Moisture and Ocean Salinity (SMOS) mission are being developed at the Barcelona Expert Centre. Besides the already operational 9-day and monthly sea surface salinity (SSS) products, two additional daily SSS products have been recently become operational: a simple user-friendly product containing all swath-based Level 2 data for each day, and a more elaborated product that uses multifractal fusion techniques to increase the spatial and temporal resolution. Finally, experimental BEC products are also presented which provide SSS values in regions strongly affected by radio-frequency interference (RFI). Recent progress on Land-Sea contamination mitigation has been applied to the BEC products.

The NASA Soil Moisture Active Passive (SMAP) mission aims at producing low (36 km) and high-resolution (9 km) global maps of surface soil moisture based on L-band radiometer and radar/radiometer measurements, respectively. In this research study, results of applying a novel retrieval algorithm, the so-called Multi-Temporal Dual Channel Algorithm (MT-DCA) to the first year of SMAP observations are presented. MT-DCA allows retrieving not only soil moisture, but also vegetation optical depth (VOD) and scattering albedo estimates, from passive microwave measurements alone and without reliance of a priori information. At L-band, VOD is proportional to total vegetation water content and albedo accounts for structural changes. The analysis of these parameters at different temporal and spatial scales will reveal the full potential of L-band microwave for global ecology studies.

Fires are a concerning topic in Mediterranean areas. They are increasing in number and extension, probably due to the anomalous dry and hot conditions experienced in this region in the last decade. In this study, more than 2,000 fires that took place in the Iberian Peninsula (2010-2014) were analyzed. The new all-weather version of SMOS-derived soil moisture product at fine scale resolution, as well as ERA-Interim Skin Temperature datasets, were used. Soil moisture and temperature anomalies based in these datasets were computed and included in the database. These information allowed analyzing prior-to-fire conditions. Results reported that more than 70% of fires started under dry and hot conditions, and this percentage rose till 94% in the anomalous conditions prior to the biggest fires. A relation between soil moisture, temperature and burned area is found which could set the basis for a fire risk index based on SMOS data and temperature information.

The Dome-C region, in the East Antarctic Plateau, has been used for calibration/validation of satellite microwave radiometers since the 1970's. However, its use as an independent external target has been recently questioned due to some spatial inhomogeneities found in L-band airborne and satellite observations. This work evidences the influence of the Antarctic ice thickness spatial variations on the measured SMOS and Aquarius brightness temperatures (TB). The possible effects of subglacial water and bedrock on the acquired radiometric signals have also been analyzed. A 3-months no-daylight period during the Austral winter has been selected. Four transects over East Antarctica have been defined to study the spatial variations. A good agreement between SMOS and Aquarius TBchanges and ice thickness variations over the whole Antarctica has been observed, obtaining linear correlations of 0.6-0.7 and slopes of 8.6-9.5 K/km. The subglacial lakes may affect the vertical physical temperature profile and/or the dielectric properties of the ice layers above. As expected, the subglacial bedrock is not contributing to the measured TB, since the maximum estimated L-band penetration depth is ~1-1.5 km.

Remote sensing algorithms often invert multiple measurements simultaneously to retrieve a group of geophysical parameters. In order to create a robust retrieval algorithm, it is necessary to ensure that there are more unique measurements than parameters to be retrieved. If this is not the case, the inversion might have multiple solutions and be sensitive to noise. In this letter, we introduce a methodology to calculate the number of (possibly fractional) “degrees of information” in a set of measurements, representing the number of parameters that can be retrieved robustly from that set. Since different measurements may not be mutually independent, the amount of duplicate information is calculated using the information-theoretic concept of total correlation (a generalization of mutual information). The total correlation is sensitive to the full distribution of each measurement and therefore accounts for duplicate information even if multiple measurements are related only partially and nonlinearly. The method is illustrated using several examples, and applications to a variety of sensor types are discussed.

Forests partially reduce climate change impact but, at the same time, this climate forcing threatens forest's health. In recent decades, droughts are becoming more frequent and intense implying an increase of forest decline episodes and forest fires. In this context, global and frequent soil moisture observations from the ESA's SMOS mission could be useful in controlling forest exposure to decline and fires. In this paper, SMOS observations and several climate variables are analyzed together with decline and fire inventories, to study the effect of soil moisture on forest decline during an important drought on summer 2012, and on forest fires in the period 2010-2013. Results show that SMOS-derived soil moisture is a complementary variable in forest decline models. Some of the studied tree species exhibit high probability of decline occurrence under dry conditions. First results showed burned areas to be drier than unburned ones previous to the fire occurrences.

Surface Soil Moisture (SSM) affects the soil surface energy balance and thus affects the Land Surface Temperature (LST), and viceversa. Currently, LST and SSM are remotely sensed using TIR sensors and L-band radiometers, respectively. The NASA's Terra/Aqua missions provide full coverage of LST measurements under clear sky conditions using MODIS. The ESA's SMOS mission is the first satellite providing frequent SSM and ocean salinity observations at global scale. In this paper, a sensitivity study about the relationship of the LST and SSM is performed using in-situ measurements from the REMEDHUS network and spaceborne observations from MODIS and SMOS. Results show that the correlation between SSM and LST (both in-situ and remotely sensed) is highest using the daily maximum LST. This could help improving SSM algorithms and deriving new SSM products at higher resolution from the synergy of microwave and TIR observations.

Hyperspectral indices from the Compact Airborne Spectrographic Imager (CASI 550), together with land surface temperatures from the Thermal Airborne Spectrographic Imager (TASI 600) and brightness temperatures from a L-band radiometer were combined in a semi-empirical model to obtain soil moisture at very high spatial resolution (3.5 m). The airborne imagery was acquired in an experiment performed over the Soil Moisture Measurement Stations Network of the University of Salamanca (REMEDHUS) in 2012, including intensive field measurements. The feasibility of the hyperspectral optical bands from CASI acting as soil moisture proxy is tested through the cross-correlation between every possible two-CASI bands combination and in situ soil moisture. Next, the potential of combining the selected hyperspectral indices with thermal and microwave observations for soil moisture mapping is explored, synergistically with thermal and microwave observations, for obtaining soil moisture.

Simple functions of radar backscatter coefficients have been proposed as indices of soil moisture and vegetation, such as the radar vegetation index, i.e., RVI, and the soil saturation index, i.e., ms. These indices are ratios of noisy and potentially miscalibrated radar measurements and are therefore particularly susceptible to estimation errors. In this study, we consider uncertainty in satellite estimates of RVI and msarising from two radar error sources: noise and miscalibration. We derive expressions for the variance and bias in estimates of RVI and ms due to noise. We also derive expressions for the sensitivity of RVI and msto calibration errors. We use one year (September 1, 2011 to August 31, 2012) of Aquarius scatterometer observations at three polarizations ( σHH, σVV, and σHV) to map predicted error estimates globally, using parameters relevant to the National Aeronautics and Space Administration Soil Moisture Active and Passive satellite mission. We find that RVI is particularly vulnerable to errors in the calibration offset term over lightly vegetated regions, resulting in overestimates of RVI in some arid regions. ms is most sensitive to calibration errors over regions where the dynamic range of the backscatter coefficient is small, including deserts and forests. Noise induces biases in both indices, but they are negligible in both cases; however, it also induces variance, which is large for highly vegetated regions (for RVI) and areas with low dynamic range in backscatter values (for ms). We find that, with appropriate temporal and spatial averaging, noise errors in both indices can be reduced to acceptable levels. Areas sensitive to calibration errors will require masking.

The spectral window at L-band (1.400 - 1.427 GHz) is reserved for passive microwave remote sensing. This band is well-suited to retrieve soil moisture and ocean salinity due to emissivity of soil and seawater decreases with moisture and salinity, respectively, affecting microwave radiation of the Earth's surface. Nowadays, there are two space missions devoted to Earth observation with L-band radiometers on-board: the SMOS mission from the ESA and the Aquarius/SAC-D mission from the NASA and CONAE. Both missions are providing the first TB measurements of the Earth's surface at 1.413 GHz. Thus, it is a great opportunity to compare SMOS and Aquarius TBs and verify the continuity and consistency of the data. This inter-comparison is a key requirement needed to use data of both radiometers for meteorological, hydrological and climatological studies on a long term.

An innovative downscaling approach to obtain fine-scale soil moisture estimates from 40 km SMOS observations has been developed. It optimally blends SMOS multi-angular and full-polarimetric information with MODIS visible/data into high resolution soil moisture maps. The core of the algorithm is a model that linksmicrowave/optical sensitivity to soilmoisture and linearly relates the two instruments across spatial scales. This algorithm has been implemented at SMOS-BEC facilities and near real-time maps of disaggregated soil moisture over the Iberian Peninsula are being distributed. In this work, the temporal and spatial variability of these maps is evaluated through comparison with ground-basedmesurements acquired at the REMEDHUS soil moisture network, in the central part of the Duero basin, Spain. Results from a two-year time-series comparison show that downscaled soil moisture maps compare well with in situ data and nicely reproduce soil moisture dynamics at a 1 km spatial scale.

The Local Oscillators (LO) of the Microwave Imaging Radiometer using Aperture Synthesis (MIRAS) onboard the Soil Moisture and Ocean Salinity (SMOS) satellite are used to maintain the operating frequency of the 69 receivers. The phase of the LO drifts over time, in turn blurring the MIRAS brightness temperature (TB) measurements. After a pre-launch assessment, it was decided to calibrate the LO every 10 minutes to reduce the phase drifts. During short periods of the first 2.5 years of SMOS mission, the LO calibration has been performed every 2 minutes to assess the impact of a higher calibration frequency on the quality of the data. In this study, relative differences (10-min TBs versus 2-min TBs) of about 0.3 K are shown, which lead to non-negligible relative differences of about 0.2-0.3 practical salinity units (psu) in the retrieved sea surface salinity (SSS). However, when performing independent validation against Argo float SSS data at Level 3 (spatio-temporally averaged SSS products), no significant differences are found between 10-min and 2-min data. This is due to the fact that current SMOS SSS accuracy (relative to Argo) is about 0.6-0.8 psu, thus masking the relatively smaller LO calibration frequency effect.

A downscaling algorithm for SMOS which combines MODIS Visible/Infrared data and SMOS horizontal brightness temperatures at 42.5° incidence angle into high-resolution soil moisture maps has been shown to nicely reproduce soil moisture dynamics at a 1 km spatial scale. The core of this algorithm is a linking model that depicts the synergy between SMOS and MODIS observations and their sensitivity to soil moisture. In this work, the impact of adding SMOS observations at horizontal and vertical polarizations and at multiple incidence angles to this linking model has been evaluated using 6 months of observations over the Murrumbidgee catchment, South-East Australia, and a robust alternative formulation is proposed. Results show that adding SMOS observations at multiple incidence angles and both polarizations the algorithm is more stable over time and its minimization error is reduced. By comparing with in situ data, a remarkable improvement of the linear regression between downscaled and in situ data is also observed (slope of 0.95).

This paper describes the relationships found between remotely sensed soil moisture, in situ observed soil moisture, and spatial distribution of soil and climatic factors. For the comparison between remote and in situ soil moisture, soil moisture map series at high resolution, obtained by applying a downscaling approach that combines Soil Moisture and Ocean Salinity (SMOS) and MODIS imagery is extracted. The in situ soil moisture series are obtained from the Soil Moisture Measurement Stations Network (REMEDHUS) in Spain. For the spatial analysis, factors such as topography, precipitation, and land uses were mapped from the climatic and cartographic database of REMEDHUS. The comparison between downscaled and in situ soil moisture data resulted in correlation coefficient (R) values between 0.40 and 0.70, bias between -0.04 and 0.16 m3m-3, and root mean squared difference (RMSD) between 0.07 and 0.19 m3m-3. Regarding the spatial correlations between downscaled and spatial factors, no clear patterns were found when considering the topography (Topographic Wetness Index, TWI), and the land uses (Landsat classification). Nevertheless, the downscaled soil moisture was more related with the spatial distribution of precipitation (Antecedent Precipitation Index, API), with significant correlations varying between 0.24 and 0.55.

HUMID is a regional level airborne soil moisture mission. The envisioned product combines two key characteristics. On one hand, it provides robustness in presence of surface roughness and vegetation. On the other hand, it improves the current spatial resolution offered by L-band radiometers using a data fusion approach, where the data from an L-band radiometer developed by the Remote Sensing Lab at the Universitat Politecnica de Catalunya will be combined with data from a thermal and a VNIR hyperspectral sensors from the Institut Cartografic de Catalunya.

A downscaling algorithm to improve the spatial resolution of SMOS soil moisture estimates using higher resolution visible/infrared (VIS/IR) data is presented. The algorithm re- lates VIS/IR parameters such as the Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (Ts) to SMOS soil moisture estimates using the "universal triangle" concept, and gracefully combines the high accuracy of SMOS radiometric observations with the high spatial resolution of VIS/IR data into an optimal soil moisture product. In preparation for the SMOS launch, the algorithm was tested using acquisitions of the UPC Airborne RadlomEter at L-band (ARIEL) over the REMEDHUS soil moisture monitoring network in Zamora, Spain, and LANDSAT imagery. After SMOS launch, the algorithm was applied to a set of SMOS images acquired during the commissioning phase over the OZnet soil moisture monitoring site, in South-Eastern Australia, and MODIS NDVI/Tsdata. Results from comparison with in situ soil moisture measurements showed that the soil moisture variability was effectively captured at 10 and 1 km spatial scales, without a significant degradation of the root mean square error. The potential application of this downscaling approach to generate high resolution soil moisture maps over the Iberian Peninsula in near-real time is now being as- sessed.

A downscaling approach to improve the spatial resolution of Soil Moisture and Ocean Salinity (SMOS) soil moisture estimates with the use of higher resolution visible/infrared (VIS/IR) satellite data is presented. The algorithm is based on the so-called “universal triangle” concept that relates VIS/IR parameters, such as the Normalized Difference Vegetation Index (NDVI), and Land Surface Temperature (Ts), to the soil moisture status. It combines the accuracy of SMOS observations with the high spatial resolution of VIS/IR satellite data into accurate soil moisture estimates at high spatial resolution. In preparation for the SMOS launch, the algorithm was tested using observations of the UPC Airborne RadIomEter at L-band (ARIEL) over the Soil Moisture Measurement Network of the University of Salamanca (REMEDHUS) in Zamora (Spain), and LANDSAT imagery. Results showed fairly good agreement with ground-based soil moisture measurements and illustrated the strength of the link between VIS/IR satellite data and soil moisture status. Following the SMOS launch, a downscaling strategy for the estimation of soil moisture at high resolution from SMOS using MODIS VIS/IR data has been developed. The method has been applied to some of the first SMOS images acquired during the commissioning phase and is validated against in situ soil moisture data from the OZnet soil moisture monitoring network, in South-Eastern Australia. Results show that the soil moisture variability is effectively captured at 10 and 1 km spatial scales without a significant degradation of the root mean square error.

GRAJO (GPS and RAdiometric Joint Observations) is a longterm field experiment over land which is being conducted since November 2008 at the REMEDHUS site, Zamora, Spain. REMEDHUS has been identified as a cal/val site for ESA's SMOS mission. The objectives of GRAJO are multiple: (i) validate and calibrate SMOS-derived soil moisture, (ii) study the variability of soil moisture within the SMOS footprint, (iii) test pixel disaggregation techniques to improve the spatial resolution of SMOS observations, (iv) determine the optical depth and vegetation water content of barley and grass and assess their influence on soil moisture estimates from radiometric and GNSS-R measurements, and (v) characterise the soil roughness factor. This paper presents an overview of the GRAJO experiment, describing the setup and measurements strategy.

So far a number of models have been developed to estimate the emission of the sea surface at L-band as a function of different key physical variables, such as the Sea Surface Temperature (SST), the Sea Surface Salinity (SSS) and the roughness as well as the presence of sea foam, but none has demonstrated to clearly perform better than the others. An important contribution in that direction will be given by the Soil Moisture and Ocean Salinity (SMOS) mission in the next future, when global and frequent measurements of the ocean will be available and, jointly with in-situ measurements collected by buoys or vessels, will permit further studies. To rehearse and optimally prepare the future analyses, two field experiments (the Advanced L-BAnd Transmission and Reflection Observations of the Sea Surface - ALBATROSS 2008 and 2009 -) have been carried out in one of the SMOS Calibration and Validation sites: The North Atlantic Subtropical Gyre. Brightness temperature measurements, using L-band real aperture radiometers, jointly with the reflected GPS signal, and in-situ measurements of SSS, SST, wind speed, and wave spectrum were collected during these experiments. The measurements have been analyzed and the first results of this analysis are presented. After an introductory section, the campaign set-up and the measurement procedure is described in section II, while the data processing is explained in section III. Finally, section IV is devoted to the presentation of the preliminary results of the study.

The sea surface salinity (SSS) retrieval using microwave radiometry is seriously affected by the sea surface roughness. Global Navigation Satellite Signals Reflected (GNSS-R) have been proposed to perform this roughness correction. The selected observable is the volume of the normalized delay-Doppler map (maximum amplitude equal to one) above a threshold. This observable is related to the extension of the glistening zone, which is related to the sea state. Its validity to account for the surface roughness in terms of significant wave height (SWH) was proved during the ALBATROSS 2008 measurement campaign. In the following ALBATROSS 2009 campaign collocated measurements of instantaneous radiometric brightness temperatures and GNSS-R volumes are obtained by two antennas pointing exactly to the same spot with the same beamwidth and beam properties. This work described the preliminary results of these field experiments.

A deconvolution-based model has been developed in an attempt to improve the spatial resolution of future soil moisture and ocean salinity (SMOS) data. This paper is devoted to the analysis and evaluation of different algorithms using brightness temperature images obtained from an upgraded version of the SMOS end-to-end performance simulator. Particular emphasis is made on the use of least-square-derived Lagrangian methods on the Fourier and wavelet domains. The possibility of adding suitable auxiliary information in the reconstruction process has also been addressed. Results indicate that, with these techniques, it is feasible to enhance the spatial resolution of SMOS observations by a factor of 1.75 while preserving the radiometric sensitivity simultaneously.

A change detection algorithm has been developed in order to obtain high-resolution soil moisture estimates from future Soil Moisture Active and Passive (SMAP) L-band radar and radiometer observations. The approach combines the relatively noisy 3-km radar backscatter coefficients and the more accurate 36-km radiometer brightness temperature into an optimal 10-km product. In preparation for the SMAP mission, an observation system simulation experiment (OSSE) and field experimental campaigns using the Passive and Active L- and S-band Airborne Sensor (PALS) have been conducted. We use the PALS airborne observations and OSSE data to test the algorithm and develop an error budget table. When applied to four-month OSSE data, the change detection method is shown to perform better than direct inversion of the radiometer brightness temperatures alone, improving the root mean square error by 2% volumetric soil moisture content. The main assumptions of the algorithm are verified using PALS data from the soil moisture experiments held during June-July 2002 (Soil Moisture Experiment 2002) in Iowa. The algorithm error budget is estimated and shown to meet SMAP science requirements.

The SMOS (Soil Moisture and Ocean Salinity) mission is an Earth Explorer Opportunity Mission from the European Space Agency that will be launched in Spring 2009. SMOS' single payload is a new type of radiometer called MIRAS (Microwave Imaging Radiometer by Aperture Synthesis) operating at L-band in which brightness temperature (BT) images are formed by a Fourier synthesis technique. The spatial resolution of the retrieved BT (~50 km), although appropriate for climate studies, is insufficient for most hydrological studies. Current pixel disaggregation approaches are based on iterative procedures in the BT images or in image deconvolution A technique is presented to improve the spatial resolution of the BT imagery by adding the information in the Fourier domain, instead of An updated overview of the current status of the image reconstruction algorithms in 2D Aperture Synthesis Radiometers for Earth Observation, with special emphasis in the SMOS mission is presented.

The reflections of Global Navigation Satellite systems such as GPS can be used to retrieve geophysical parameters. A promising application is to use them for the sea surface roughness-induced corrections in the brightness temperature to retrieve the Sea Surface Salinity. A tandem campaign to obtain simultaneous radiometer and reflectometer data has been conducted on the North-West coast of the Gran Canaria Island (Canary Islands, Spain). The first results after processing the GNSS-R data are presented.

The SMOS REFLEX 2006 field experiment aimed to measure vines emission during their phenological cycle and to study the impact of vegetation and rocks on the emission. Rocks were kept in half the vineyard (from 40% to 80% of surface rock fraction) and were partially removed in the other half (from 6% to 30% of surface rock fraction). Since rocks have a low and constant dielectric permittivity, their presence masks the soil dielectric increase due to soil moisture. This leads to an almost constant relationship between the measured emission and ground-truth soil moisture and, thus, to a subestimation of the soil moisture by the retrieval algorithms: i.e. the soil appears drier to the radiometer than it actually is.

Different deconvolution algorithms have been developed to explore the possibility of improving the spatial resolution of future Soil Moisture and Ocean Salinity (SMOS) products. An exhaustive test of these methods has been performed over brightness temperature images obtained from an upgraded version of the SMOS End-to-end Performance Simulator (SEPS). Particular emphasis is made on the use of Wiener filter derived methods on the Fourier and on the wavelet domain. The possibility of including suitable auxiliary information in the reconstruction process has also been addressed. Results show that with these techniques it is feasible to improve the spatial resolution (DeltaS) of SMOS observations whereas preserving its radiometric sensitivity (DeltaT), especially in the areas of the field of view far away from nadir. The product DeltaTmiddotDeltaS can be improved by a 49% over soil pixels and by a 30% over sea pixels.

The impact of topography on soil emissivity at L-band is not well known. In order to provide data to assess this issue, the Topography effects on RadiomeTry at L-band Experiment (TuRTLE) 2006 was carried out in a mountainous area about 50 km North of Barcelona (Spain). Radiometric measurements covering the mountain slope, and up to the sky were acquired. Concurrently, ground-truth and meteorological data were registered. Radiometric measurements have been compared to the emissivity obtained by simulation using a facet model which considers the high resolution digital elevation model and land cover map of the area. Polarization mixing due to surface tilting and integration over the antenna pattern have also been included in the simulator, and results agree with the radiometric measurements. The largest discrepancies occur for an almost bare soil at large local incidence angles and H-polarization, close to the radiometer, which suggests that further modeling work is still needed.

In remote-sensing applications the inclusion of subgrid-scale variability in coarse resolution data still remains an elusive challenge. This paper is devoted to the development of an appropriate downscaling technique for future Aperture Synthesis Radiometer’s images. A comparative study of different deconvolution algorithms has been performed and particular emphasis is made on the use of least-squares Lagrangian methods and Fourier Wiener filtering. Results show that with this technique it is feasible to improve the spatial resolution of brightness temperature images from the Spatial Sensor Microwave Imager (SSM/I) radiometer and from an upgraded version of the Soil Moisture and Ocean Salinity (SMOS) End-to-end Performance Simulator (SEPS).

The impact of topography and mixed pixels on L-band radiometric observations over land needs to be quantified to improve the accuracy of soil moisture retrievals. For this purpose, a series of simulations has been performed with an improved version of the soil moisture and ocean salinity (SMOS) end-to-end performance simulator (SEPS). The brightness temperature generator of SEPS has been modified to include a 100-m-resolution land cover map and a 30-m-resolution digital elevation map of Catalonia (northeast of Spain). This high-resolution generator allows the assessment of the errors in soil moisture retrieval algorithms due to limited spatial resolution and provides a basis for the development of pixel disaggregation techniques. Variation of the local incidence angle, shadowing, and atmospheric effects (up- and downwelling radiation) due to surface topography has been analyzed. Results are compared to brightness temperatures that are computed under the assumption of an ellipsoidal Earth.

Soil roughness is an important parameter whose effects on soil emissivity at L-band have not been satisfactorily quantified so far. This paper presents the preliminary results of the Terrain-Roughness Experiment (T-REX), which was carried out in Agramunt, Spain, during the fall of 2004. Four identical sets, each of them having four plots with different plough but the same soil type, were measured. Two profiles per soil plough were acquired, one parallel and the other perpendicular to the antenna frame. Since the purpose of T-REX 2004 was to analyze the impact of surface roughness on the received signal, the soil was not irrigated. Results show that the soil emissivity increases as the soil roughness increases. The horizontal polarization seems to be more sensitive to roughness than the vertical polarization for fields with a tillage direction perpendicular to the antenna reference frame. Emissivity at horizontal polarization decreases as the incidence angle increases, while the trend for the vertical polarization is almost constant when the soil is dry.