Development of low latency, high resolution SMAP soil moisture retrievals in support of near-real-time applications

Abstract

Soil moisture plays a crucial role in weather forecasting and climate prediction, and provides useful information for flood and drought monitoring, and agricultural irrigation scheduling. In recent decades, soil moisture measurements have been enabled by space-borne passive microwave sensors. Nonetheless, the application of satellite remote sensing soil moisture data in operational meteorological and hydrological contexts is constrained by their prolonged latency and relatively coarse spatial resolution. This study describes the development of an operational enhanced Soil Moisture Active Passive (SMAP) soil moisture retrieval processor, referred to as SMAP_E_OPL, which addresses these limitations. SMAP_E_OPL produces near-real-time (NRT) soil moisture data at a 9-km grid resolution utilizing the single channel retrieval procedure employed in the long-latency SMAP Level 2 Enhanced soil moisture products (i.e., SMAP_L2SMP_E). In contrast to SMAP_L2SMP_E, the SMAP_E_OPL processor uses the Inverse Distance Squared Weighted interpolation approach to resample the NRT SMAP L1B brightness temperature (T_B) data from a 33-km resolution to the target grid for enhanced computational efficiency. Moreover, SMAP_E_OPL also allows the interoperable use of soil temperature and snow depth estimates from various operational land surface model (LSM) configurations. A bias-correction process is implemented to ensure consistency in the effective soil temperature (T_eff) between SMAP_E_OPL and SMAP_L2SMP_E. Evaluation based on in situ soil moisture measurements across the continental United States demonstrates that SMAP_E_OPL delivers the NRT soil moisture at 9-km grid resolution with accuracy comparable to the official (but high latency) SMAP enhanced soil moisture products. The use of different LSMs in SMAP_E_OPL to estimate T_eff yields similar quality in soil moisture retrievals, highlighting the flexibility of SMAP_E_OPL with respect to LSMs. These results confirm the viability of the SMAP_E_OPL processor in providing high-spatial-resolution soil moisture data with low latency for operational forecasting, data assimilation, and monitoring systems.

1 Introduction

Soil moisture plays a crucial role in weather forecasting and climate prediction systems by influencing the interactions between the atmosphere and Earth’s surface (Entekhabi et al., 2010; Ochsner et al., 2013; Santanello Jr et al., 2019; Seneviratne et al., 2010). Timely and precise soil moisture data is essential for assessing natural disasters and managing water resources, including flood and drought monitoring (Crow et al., 2017; Engda and Kelleners, 2016; Massari et al., 2014; 2018; Mladenova et al., 2019) and agricultural irrigation scheduling (Moran et al., 2004). These factors are closely intertwined with the global economic stability, food security, and overall quality of life (Milman and Arsano, 2014). Nevertheless, capturing the spatial and temporal variability of soil moisture over large scales (e.g., continental or global) is a challenging endeavor, as ground-based measurements have limited coverage.

Satellite-based microwave observations at low frequencies (e.g., 1–10 GHz) can provide global estimates of near-surface soil moisture, given their properties of high sensitivity to soil moisture changes (Jackson et al., 2002; Njoku and Kong, 1977). The National Aeronautics and Space Administration (NASA) Soil Moisture Active Passive (SMAP) satellite, launched in January 2015, has been providing its standard near-surface soil moisture product since 31 March 2015 at a spatial resolution of 36 km with a revisit cycle of 2–3 days and a target accuracy of 0.04 m³ m⁻³ with respect to the unbiased root-mean-square error (ubRMSE). Numerous studies (e.g., Bhuiyan et al., 2018; Chan et al., 2016; 2018; Chen et al., 2017; Colliander et al., 2017; Kumar et al., 2018; Pan et al., 2016; Wrona et al., 2017; Zhang et al., 2017) have demonstrated the capability of SMAP to monitor global soil moisture conditions. SMAP soil moisture retrievals have also been widely used in data assimilation (DA) systems to improve modeled soil moisture estimates (e.g., Ahmad et al., 2022; Blyverket et al., 2019; Chao et al., 2022; Kim et al., 2021; Kwon et al., 2022; Mladenova et al., 2019) and have been effectively applied to monitor droughts and floods (e.g., Chao et al., 2022; Mishra et al., 2017; Mladenova et al., 2019; Rahman et al., 2019; Sadri et al., 2018), assess cropland and food security (e.g., Gonsamo et al., 2019; Sadri et al., 2020), and characterize irrigation (e.g., Brocca et al., 2018; Kwon et al., 2022; Lawston et al., 2017; Nie et al., 2022).

Previous studies (e.g., Moran et al., 2004; Schmugge et al., 2002) have indicated that finer spatial resolution soil moisture is more effective in characterizing heterogeneous land surface hydrometeorological processes. Initially, SMAP was designed to provide high-resolution global measurements of near-surface soil moisture (at the top 5 cm soil depth) and freeze-thaw states with spatial resolution of up to 1–3 km, achieved by combining an L-band radar (active; 1.26 GHz) and an L-band radiometer (passive; 1.41 GHz) (Das et al., 2011; Entekhabi et al., 2010; Liu et al., 2016). However, the SMAP radar failed on 7 July 2015. Although radar observations from the European Space Agency’s Sentinel-1 have been used to support a high-spatial resolution product at 1–3 km for the mission, known as SMAP/Sentinel-1 (SPL2SMAP_S, Das et al., 2019), the product has a long repeat coverage of around 12 days (Liu P.-W. et al., 2021). Various approaches have been developed to downscale the standard SMAP data, originally at 36 km resolution to finer grid resolutions, such as 9-km using optimal interpolation methods (Chan et al., 2018; O'Neill et al., 2019), and 1-km employing the thermal inertia algorithm (Fang et al., 2020; Liu Z. et al., 2021; Merlin et al., 2008) or by integrating thermal fluxes and geoinformation data (Liu et al., 2022).

In particular, soil moisture products with fine spatial resolution and low latency are a critical need for operational weather and hydrologic forecasting systems. For example, accurate soil moisture initial conditions are well-established as a significant requirement for numerical weather prediction (NWP) systems (Dirmeyer and Halder, 2016; Koster et al., 2010; Kwon et al., 2024; 2026). Having low latency products is, therefore, critical if those systems are to take advantage of information from satellite remote sensing. Such near-real-time (NRT) information is also essential for operational hydrological applications (Mladenova et al., 2019; Sadri et al., 2020) such as weather and flash flood forecasting. Rodríguez-Fernández et al. (2019) suggested that an ideal operational system needs to meet an NRT requirement of ∼3 h, which is typically not met by current satellite mission products. For example, NASA’s SMAP provides level 2 enhanced and standard soil moisture products at a latency of 14–16 h (nominal latency of 24 h). Its level 3 and 4 products require even longer time, ∼2–3 days, for data processing and model run (Crow et al., 2017). Many processing steps are required for brightness temperature (T_B) observations and other ancillary data sources to generate soil moisture products from T_B using the geophysical retrieval algorithm. Particularly, an interpolation method, i.e., Backus-Gilbert (BG) algorithm (Poe, 1990), employed to produce the official SMAP Level 2 enhanced (9 km) soil moisture product (SMAP_L2SMP_E) usually takes 2–4 h to grid T_B onto the Equal-Area Scalable Earth (EASE)-Grid. It also relies on modeled soil temperature estimates from the NASA Global Modeling and Assimilation Office (GMAO) Goddard Earth Observing System-Forward Processing (GEOS-FP), which is not an operational environment. Thus, the process methodologies and reliance on external product sources may delay the data delivery of SMAP_L2SMP_E when extra processing and quality control procedures are included. On the other hand, the development of SMAP retrievals using ancillary information from the NASA Land Information System (LIS) modeling system results in the use of a consistent set of ancillary datasets between the remote sensing retrievals and NWP model.

A NRT version of the SMAP data (with ∼3-h latency) has, in fact, been developed (O'Neill et al., 2022). However, this product only enables soil moisture data at the baseline coarse spatial resolution of 36 km. While such products have been assimilated in operational land DA systems for weather forecasting (Wegiel et al., 2020), the adoption of NRT SMAP products with finer spatial resolutions is hindered by their significant latency. Notably, Yin et al. (2020) introduced a downscaling method capable of generating NRT daily 1-km SMAP soil moisture data, yet this method relies on additional data sources, including the Visible Infrared Imaging Radiometer Suite (VIIRS) land surface temperature and the Moderate Resolution Imaging Spectroradiometer (MODIS) enhanced vegetation index, both of which are subject to cloud cover limitations resulting in missing data (Yin et al., 2020). In this regard, the current study aims to develop an operational system that employs a more time-efficient gridding algorithm and uses system-self variables for the retrieval algorithm to fill these gaps.

This article describes the development of an operational, enhanced SMAP soil moisture retrieval, denoted as SMAP_E_OPL, with the capability to produce NRT hourly soil moisture products at a grid resolution ranging from 9 to 10 km, directly from the NRT SMAP Level 1B (L1B) T_B data (Piepmeier et al., 2020). This product is similar in grid resolution to the enhanced SMAP product, and it meets the NRT latency requirement of ∼3 h. Here we demonstrate the following key enhancements: (a) The SMAP_E_OPL processor generates soil moisture information of comparable quality to the official (high latency) SMAP products, such as the SMAP Level 2 standard (36 km, SMAP_L2SMP) and enhanced Passive Soil Moisture Products (9 km, SMAP_L2SMP_E). (b) This is enabled by a flexible, open source implementation that utilizes inputs from different land surface models (LSMs) for ancillary information of effective soil temperature (T_eff) to yield similar performance in soil moisture retrievals. Note that both enhanced products (i.e., SMAP_L2SMP_E and SMAP_E_OPL) provide soil moisture data posted on a 9 km grid, while the effective spatial resolution remains approximately 33 km.

The SMAP_E_OPL soil moisture retrieval processor is integrated within the NASA LIS (Kumar et al., 2006; 2008) Framework (LISF), which is an open source software infrastructure for high performance land surface modeling and DA. LISF has a data processing environment called the Land Data Toolkit (LDT; Arsenault et al., 2018), which includes capabilities for processing land surface datasets for ingestion into the modeling environment LIS. The SMAP_E_OPL processor is integrated within the LDT subsystem of LISF. Several regional and global hydrological modeling and DA systems around the world have been enabled by LIS, including the North American land data assimilation system (NLDAS; Xia et al., 2012), the Global land data assimilation system (GLDAS; Rodell et al., 2004), the famine early warning land data assimilation system (FLDAS; McNally et al., 2017). Similarly, the U.S. Air Force (USAF) Weather also employs a custom configuration of LIS in their operations (Wegiel et al., 2020). We implemented the SMAP_E_OPL processor within this USAF environment to enable the real-time operational assimilation of the enhanced SMAP soil moisture data. This study validated the capability of the operational processor to generate high-spatial-resolution SMAP soil moisture retrievals in near-real-time, exhibiting accuracy comparable to official SMAP products. To the best of our knowledge, SMAP_E_OPL stands as the only operational processor that directly delivers NRT hourly enhanced SMAP soil moisture products (with grid resolution of 9 km) from the SMAP L1B T_B data. SMAP_E_OPL incorporates modeled T_eff and snow depth from LSMs within LIS, including 557th Weather Wing (WW) operational simulations, as well as other preprocessed static or climatological ancillary data.

2 Materials and methods

2.1 Operational enhanced SMAP (SMAP_E_OPL) soil moisture retrieval processor

The SMAP_E_OPL soil moisture retrieval processor, integrated into LDT, follows the established protocol applied in the official SMAP L2 Enhanced soil moisture products (SMAP_L2SMP_E; O'Neill et al., 2019). However, certain modifications and adjustments are introduced to meet the NRT requirements for operational purposes. The primary distinctions include the source of the L1B T_B data, the interpolation method, spatial grid and projection, and ancillary data sources for parameters like T_eff, snow flag, and frozen soil flag, which are summarized in Table 1. It should be noted that while the objective of SMAP_E_OPL processor is to ingest the SMAP NRT T_B data, for the purpose of algorithm development and evaluation in this study, the standard SMAP L1B T_B data (Piepmeier et al., 2020) have been employed.

The SMAP_L2SMP_E processor utilizes the Backus-Gilbert (BG) optimal interpolation method (Poe, 1990) to resample the non-gridded SMAP Level 1B T_B to a 9-km EASE-Grid for the Level 1C Enhanced T_B Product (i.e., L1C_TB_E; Chaubell, 2016), requiring 2–4 h for resampling a half-orbit data. In contrast, the SMAP_E_OPL processor employs an Inverse Distance Squared Weighted (IDSW) interpolation method to resample the SMAP NRT L1B T_B data to a 0.14° longitude by 0.09° latitude target grid using the equidistant cylindrical projection. This grid configuration is used in SMAP_E_OPL to be consistent with the global modeling configuration used at 557th WW. The IDSW interpolation is a conventional technique utilized in passive microwave satellite missions like the Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) (Knowles et al., 2006) and reduces the time for T_B gridding to approximately 3 min per half-orbit. Despite the efficiency gain, the SMAP_E_OPL processor maintains comparable skill to the standard processor, with just around 1 K difference in T_B resulting in ∼0.01 m³ m⁻³ difference in soil moisture.

The resampled T_B data (SMAP NRT L1C) is utilized in soil moisture retrieval algorithms, incorporating T_eff from LIS-LSMs as detailed in Section 2.3, along with ancillary data to generate the SMAP_E_OPL soil moisture. This involves employing a traditional τ (vegetation optical depth)-ω (albedo scattering) model (Equation 1) with the Mironov model (Equations 2–10) to convert the dielectric constant into soil moisture, consistent with the algorithm used for the SMAP standard soil moisture products.where e_p is the emissivity, θ_l the SMAP look angle (=40°), T_vege is the vegetation canopy temperature, and r_p is the reflectivity. The key regression equations used in Mironov’s generalized refractive mixing dielectric model (GRMDM; Mironov et al., 2009) are shown below:where n_d is the refractive index of dry soil; k_d is the normalized attenuation coefficient of dry soil; m_vt is the maximum bound water fraction; ɛ_0b and ɛ_0u denote the low-frequency limits of the dielectric constants for bound and free soil water, respectively; τ_b and τ_u are the relaxation times for bound and free soil water, respectively; σ_b and σ_u represent the relaxation time for bound and free soil water, respectively; and C is the clay content. For more details on the τ-ω and Mironov models, refer to the SMAP Algorithm Theoretical Basis Document (ATBD; O’Neill et al., 2021a) and Mironov et al. (2009). Within SMAP_E_OPL, the SMAP Single Channel Algorithm using vertically polarized T_B (SCA-V) is employed. We selected SCA-V for the current version of SMAP_E_OPL because it has served as a baseline algorithm for the SMAP product and has been extensively evaluated and used in many previous studies. Nevertheless, we note that the dual-channel algorithm (DCA) has recently been adopted as the SMAP baseline product and that, in some land cover types, SCA using horizontally polarized T_B (SCA-H) provides better quality (Judge et al., 2021). In this study, however, we use SCA-V to generate the SMAP_E_OPL soil moisture data and to compare them with the official SMAP data, which are also produced using SCA-V. We consider this choice reasonable for the processor-development objectives of this study. The incorporation of DCA will be considered in future versions of the SMAP_E_OPL processor. Figure 1 presents a schematic diagram of SMAP_E_OPL. The snow flag and frozen soil flag are obtained from the operational United States (US) Air Force Snow and Ice Analysis (USAF-SI; Yoon et al., 2022) and the NRT US Air Force 557th WW global operational LIS-LSMs.

FIGURE 1

T_eff serves as a crucial factor in the τ-ω model (O’Neill et al., 2021a). In the SMAP_E_OPL processor, T_eff is computed using soil temperature data from LIS-LSMs at the same time of the previous day (t-1 day), which is the identical setting as used in the official SMAP NRT SM product, while the SMAP_L2SMP_E processor relies on soil temperature simulated by the GEOS-FP system from NASA GMAO at t. The utilization of the LIS-based soil temperature from the previous day (t-1 day) reduces the reliance of the retrieval processor on data sources that are not available in real-time, thereby mitigating potential delays in soil moisture retrievals. However, it should be noted that the use of t-1 day soil temperature estimates may introduce errors into the SMAP_E_OPL soil moisture retrievals. LIS-LSMs utilized in this study encompass the Noah LSM (Ek et al., 2003) version 3.9, Noah LSM with multi-parameterization options (Noah-MP; Niu et al., 2011; Yang et al., 2011) version 4.0.1, and Joint United Kingdom (UK) Land Environment Simulator (JULES; Best et al., 2011; Clark et al., 2011) version 5.0. To ensure consistency between the SMAP_E_OPL and SMAP_L2SMP_E soil moisture products, the T_eff derived from the NRT US Air Force 557th WW global operational LIS-LSMs undergoes bias-correction in SMAP_E_OPL, as outlined in Section 2.3.

The snow flag within SMAP_E_OPL relies on snow depth simulated by USAF-SI at t-3h while SMAP_L2SMP_E employs snow cover fraction data from the National Oceanic and Atmospheric Administration (NOAA) Interactive Multisensor Snow and Ice Mapping System (IMS) for the snow flag. Additionally, the frozen soil flag in SMAP_E_OPL is determined based on soil temperature estimates from the NRT US Air Force 557th WW global operational LIS-LSMs, while the GMAO GEOS-FP soil temperature outputs are utilized in SMAP_L2SMP_E. These flags are applied to exclude soil moisture retrievals for pixels covered by over 5 cm of snow or categorized as frozen soil. The difference in the snow and frozen soil flag sources between SMAP_E_OPL and SMAP_L2SMP_E may cause some disagreement between them for missing soil moisture data points.

The retrieval algorithms in SMAP_E_OPL require additional ancillary data, including land cover, soil texture (i.e., clay fraction and bulk density), surface geometry (roughness), and vegetation information (τ and ω) (O’Neill et al., 2021a). These datasets in SMAP_E_OPL match those utilized for the SMAP_L2SMP_E soil moisture products but are regridded to the target grid (0.14° longitude by 0.09° latitude, equidistant cylindrical projection). Clay fraction and soil bulk density are sourced from the global SoilGrid250m (Hengl et al., 2017) (available at https://www.isric.org/explore/soilgrids), aggregated by averaging values to the target grid. The surface roughness and ω values are determined from a lookup table, based on the primary land cover type in a grid, derived from the Terra MODIS-based International Geosphere-Biosphere Programme (IGBP) land cover classes (Friedl et al., 2002) at their original 500 m spatial resolution, selecting the most representative class (i.e., the highest percentage of presence class) within the target grid. The ω values used in this study range from 0.03 to 0.08 (Supplementary Table 1). τ (ranging from ∼0.2 to 1.0) is estimated from vegetation water content (VWC) multiplied by an empirical parameter (b) acquired from a lookup table for each land cover class (Supplementary Table 1). VWC is determined from the MODIS Normalized Difference Vegetation Index (NDVI) climatology from 2000 to 2010, consistent with the data used for SMAP_L2SMP_E. Further details of these processes are outlined in the SMAP ATBD (O’Neill et al., 2021a).

2.2 Land surface model setup

Here we utilize T_eff estimates from three different LIS-LSMs including Noah, Noah-MP, and JULES within the SMAP_E_OPL processor. These three LSMs were selected because they are the operational models implemented in the US Air Force 557th WW, the forecast system targeted in this study. However, our results (see Section 3.1) highlight the flexibility of the SMAP_E_OPL processor, which provides consistent soil moisture retrieval performance across different LSMs.

The Noah LSM is based on the Oregon State University (OSU) LSM and has been extensively used in coupled numerical weather and climate prediction models due to its computational efficiency (Wei et al., 2013). Noah simulates soil and land surface temperature, canopy water content, soil moisture, snow water equivalent, and water and energy fluxes. The vegetation seasonality is represented by the spatially and temporally varying (monthly) green vegetation cover fraction. In the Noah LSM, a grid cell is composed of a vegetation canopy layer, a single snow layer, and four soil layers with thicknesses of 0.1 m, 0.3 m, 0.6 m, and 1.0 m from top to bottom (with a total soil depth of 2 m), for which soil moisture and temperature are estimated using the diffusivity form of the Richard’s equation and the one-dimensional thermal diffusion, respectively. Noah computes evapotranspiration as the sum of vegetation transpiration, evaporation from the soil surface, and evaporation from canopy-intercepted water, and it calculates the latent heat flux based on the Penman potential evaporation approach. Vegetation rooting depth is estimated for each vegetation type by applying a seasonal factor suggested by Dickinson et al. (1993) to represent the seasonality of transpiration and soil moisture (Wei et al., 2013). The subgrid-scale variability of soil moisture and precipitation is modeled by the surface moisture infiltration scheme based on Schaake et al. (1996), and moisture infiltration during the winter season is influenced by the presence of frozen soil.

Compared to the baseline Noah LSM, Noah-MP provides multiple enhanced representations of hydrological processes related to vegetation, soil, snow, and groundwater (Niu et al., 2011; Yang et al., 2011). Noah-MP represents a one-dimensional column within each model grid cell, consisting of a vegetation canopy layer, up to three snowpack layers (depending on snow depth), a four-layer soil with the same default thicknesses as in the Noah LSM, and an unconfined aquifer layer. In Noah-MP, a semi-tile scheme is used to account for subgrid-scale heterogeneity (Niu et al., 2011). Unlike other LSMs, Noah-MP offers multiple parameterization options, enabling multi-physics ensemble simulations with diverse physical scheme combinations in a single model (Cai et al., 2014). This study employs the following physical scheme options in Noah-MP: static vegetation option for leaf area index and green vegetation fraction, using monthly prescribed values from input datasets; the Ball–Berry scheme for canopy stomatal resistance (Ball et al., 1987); the Noah-type soil moisture factor for stomatal resistance (Chen and Dudhia, 2001); the Monin–Obukhov similarity scheme (Brutsaert, 1982) for surface layer drag coefficient calculation; the TOPography-based hydrological MODEL (TOPMODEL) runoff with the simple groundwater scheme (SIMGM) (Niu et al., 2007); the modified two-stream scheme for radiative transfer (Niu and Yang, 2004; Yang and Friedl, 2003); the NY06 method (Niu and Yang, 2006) for representing supercooled liquid water and soil permeability under frozen conditions; the Biosphere-Atmosphere Transfer Scheme (BATS) parameterization for snow albedo (Yang and Dickinson, 1996); the semi-implicit time schemes for snow and soil temperature; the original Noah approach for parameterizing the lower-boundary conditions of soil temperature; the Jordan scheme (Jordan, 1991) for partitioning precipitation into snow and rain.

Like Noah and Noah-MP, JULES is a process-based community LSM, developed from its baseline predecessor, the Met Office Surface Exchange Scheme (MOSES; Cox et al., 1999), which simulates land-atmosphere interactions. JULES represents subgrid heterogeneity of the land surface using subgrid tiles, which include nine land cover classes: broadleaf trees, needleleaf trees, C3 (temperate) grass, C4 (tropical) grass, shrubs, bare soil, land ice, inland water, and urban areas. In JULES, soil moisture and temperature are modeled in four layers with thicknesses of 0.1 m, 0.25 m, 0.65 m, and 2.0 m from the surface to the bottom, extending to a total soil depth of 3 m. JULES has single- and multi-layer snowpack options, the latter of which is employed in this study. The model computes the surface energy fluxes for each land cover type within a grid cell, which are then averaged using weights. Precipitation reaching the ground after canopy interception is partitioned between surface runoff and infiltration. The vertical redistribution of infiltrated water between soil layers is represented using a finite-difference approximation of the Richards equation (Richards, 1931). JULES estimates soil hydraulic conductivity using the van Genuchten (van Genuchten, 1980) formulation, accounting for the effects of frozen soil.

These LSMs are driven by the meteorological forcing generated in operational settings at the 557th WW at the United States Air Force (Kemp et al., 2022). The forcing includes precipitation, air temperature, solar radiation, atmospheric pressure, and wind speed from the United States Air Force Global Air-Land Weather Exploitation Model (GALWEM), NOAA Global Forecast System (GFS), NASA Integrated Multiscale Retrievals for Global Precipitation Measurement (IMERG; Huffman, et al., 2020), World-Wide Merged Cloud Analysis (Air Force Weather Agency, 2013), along with in situ surface meteorology observations (Kemp et al., 2022). Land surface geophysical parameters including static land use/cover, soil texture, and topography, and monthly climatological albedo and greenness are summarized in Supplementary Table 2. The LIS-LSMs operate globally at a spatial resolution of 0.14° (longitude) by 0.09° (latitude) delivering 3-hourly instantaneous soil temperature and 6-hourly USAF-SI snow depth outputs. These outputs serve as inputs for the SMAP_E_OPL processor to generate soil moisture retrievals covering the period from 1 April 2015 to 31 December 2021. Note that soil moisture is retrieved whenever T_B measurements are available, whereas 3-hourly soil temperature outputs from LIS-LSMs are used in the retrieval processor due to the design of the US Air Force 557th WW system. Throughout this paper, the SMAP_E_OPL retrievals employing Noah, Noah-MP, and JULES are respectively referred to as SMAP_E_OPL (Noah), SMAP_E_OPL (Noah-MP), and SMAP_E_OPL (JULES).

2.3 Bias correction of effective soil temperature (T_eff)

In SMAP_L2SMP_E, T_eff is calculated using Equation 11, modified from the Choudhury’s two-layer T_eff model (Choudhury et al., 1982).where T_s1 and T_s2 are the average temperature of the first and second soil layers, respectively, of the GMAO GEOS-FP model. The empirical parameters, K (=1.007) and C [=0.246 for AM (descending overpasses) and 1.000 for PM (ascending overpasses) soil moisture retrievals] are obtained from extensive calibration processes using the GEOS-FP soil temperature and in situ temperature measurements for SMAP_L2SMP_E (O’Neill et al., 2021a).

However, systematic differences may exist between the soil temperature obtained from the LIS-LSMs (Noah, Noah-MP, and JULES used in SMAP_E_OPL) and that simulated by GMAO GEOS-FP (utilized in SMAP_L2SMP_E), owing to differing model physics and representation of vertical soil layers. Therefore, in order to ensure consistency between SMAP_E_OPL and SMAP_L2SMP_E, the SMAP_E_OPL processor employs a bias-correction approach. In this approach, SMAP_E_OPL adopts the same K and C values as used in SMAP_L2SMP_E, and rescales the LIS-LSM-based T_eff to the climatology of the GEOS-FP T_eff through the normal deviate scaling method (Equation 12), as follows:where and are the LIS-LSM-based T_eff at local 6 a.m. or 6 p.m. before and after bias-correction, respectively; and are the preprocessed 6-year (2015–2021) daily mean and standard deviation, respectively, of the LIS-LSM T_eff at local 6 a.m. or 6 p.m.; and and are the preprocessed 6-year (2015–2021) daily mean and standard deviation, respectively, of the reference model (i.e., GEOS-FP) T_eff at local 6 a.m. or 6 p.m. The scaling approach increases the flexibility of SMAP_E_OPL with respect to using different LSMs. The use of the normal deviate scaling method, rather than a more refined approach such as cumulative distribution function (CDF) matching, is based on its relatively stable performance with limited sample sizes.

The spatial resolution of the GMAO GEOS-FP’s soil temperature used in the official 9-km SMAP product (i.e., SMAP_L2SMP_E) is 0.3° (longitude) × 0.25° (latitude) while that of the LIS-LSMs’ soil temperature used in this study is 0.14° (longitude) × 0.09° (latitude). LDT reads the GEOS-FP soil temperature of the top two soil layers and resamples it to the target grid (i.e., 0.14° (longitude) × 0.09° (latitude) in this study) using the bilinear interpolation method, the same approach used for the SMAP official products. The interpolated soil temperatures are then used to calculate T_eff. The difference in the spatial resolution between GEOS-FP and LIS-LSMs (and the resampling procedure employed in LDT) may be one of the potential sources causing the difference in the retrieved soil moisture between SMAP_E_OPL and SMAP_L2SMP_E.

Note that the primary purpose of the current study is to develop an operational retrieval processor (a) that generates enhanced (9-km) SMAP soil moisture data whose quality is comparable to that of the official SMAP SM product (i.e., SMAP_L2SMP_E) while achieving reduced latency suitable for operational systems that are sensitive to data latency, and (b) that are applicable to any operational systems. For this reason, we adopted a bias correction procedure (i.e., normal deviate scaling), which increases the flexibility of the processor for different systems as compared to directly using soil temperature from LIS-LSMs by recalibrating the empirical parameters (i.e., K and C) in Equation 11.

2.4 In situ soil moisture measurements

The performance of SMAP_E_OPL is evaluated over the continental United States (CONUS) by comparing to in situ soil moisture measurements from the United States Department of Agriculture (USDA) Soil Climate Analysis Network (SCAN, https://www.wcc.nrcs.usda.gov/scan; Diamond et al., 2013; Schaefer et al., 2007) and United States Climate Reference Network (USCRN, https://www.ncei.noaa.gov/access/crn; Bell et al., 2013). These sparse networks consist of monitoring stations distributed across diverse land cover classes and hydroclimate zones and have been used for assessing various SMAP products (O’Neill et al., 2021b). SCAN and USCRN stations measure hourly soil moisture at five soil depths of 5, 10, 20, 50, and 100 cm using a single set of probes and triplicate soil moisture probes, respectively (Bell et al., 2013; Diamond et al., 2013). Evaluation involves the station-to-pixel comparison of the soil moisture retrievals from SMAP_E_OPL, SMAP_L2SMP_E, and SMAP_L2SMP against the measured soil moisture at a 5-cm soil depth following the Sparse Network Assessment used in the official SMAP quality assessment report (O’Neill et al., 2021b). A total of 153 stations (SCAN: 97 stations; USCRN: 56 stations) are employed, selected based on identical quality-control protocols as used in Chen et al. (2018).

3 Results

We first focus on demonstrating whether the high-spatial-resolution NRT soil moisture products from the SMAP_E_OPL processor implemented in LDT exhibit comparable quality to those from the official, high latency SMAP_L2SMP_E processor. In this regard, soil moisture retrievals from SMAP_E_OPL, SMAP_L2SMP_E, and SMAP_L2SMP are evaluated using in situ soil moisture measurements over CONUS. In the second subsection, we delve into the application of the soil moisture data generated in this study within the context of DA techniques. We highlight how this integration is poised to enhance hydrological predictions and contribute significantly to water resources management applications.

3.1 SMAP_E_OPL soil moisture processor: a comparative analysis using SCAN and USCRN in situ data

The performance of the soil moisture retrievals from the SMAP_E_OPL processor is evaluated using in situ soil moisture data from the SCAN and USCRN stations for the period from 2015 to 2021. The sparse network assessment is a station-to-pixel evaluation, identical to the methodology used in the SMAP assessment report (O’Neill et al., 2021b). The results are summarized in Table 2. The SMAP_E_OPL processor utilizes three different sources of effective soil temperature (T_eff) from the Noah, Noah-MP, and JULES models to assess algorithm stability. Additionally, soil moisture products from SMAP_L2SMP_E and SMAP_L2SMP are used to compare the quality of SMAP_E_OPL with the mission products. Note must be taken that the spatial resolution of the SMAP_L2SMP_E product is 33 km posted on a 9-km EASE-GRID, and that of the SMAP_L2SMP product is 36 km posted on a 36-km EASE-GRID; while the soil moisture retrieval from SMAP_E_OPL is at 33 km posted on a 0.14° (longitude) × 0.09° (latitude) grid, which is at a similar resolution as SMAP products. The objective of this assessment is to demonstrate that the soil moisture from SMAP_E_OPL and SMAP_L2SMP_E is of comparable skill.

The evaluation employs statistical metrics such as unbiased root mean square error (ubRMSE), RMSE, bias, and correlation coefficient (R) for descending (AM), ascending (PM), and combined overpasses (AM + PM) over CONUS. Note that while RMSE and bias are included as metrics to compare different soil moisture retrievals, they are influenced by systematic bias and representativeness errors inherent in station-based evaluations. For all stations used in this study, the domain-averaged ubRMSE, RMSE, and R values for SMAP_E_OPL soil moisture retrievals during AM + PM overpasses are approximately 0.052 m³ m⁻³, 0.074 m³ m⁻³, and 0.65, respectively. These values are comparable to those of SMAP_L2SMP_E (ubRMSE 0.050 m³ m⁻³, RMSE 0.073 m³ m⁻³, and R 0.68) and SMAP_L2SMP (ubRMSE 0.050 m³ m⁻³, RMSE 0.074 m³ m⁻³, and R 0.67). Comparable performance is also obtained when soil moisture retrievals are evaluated separately for the AM and PM overpasses. Furthermore, all retrieval processors slightly overestimate soil moisture (0 < domain-averaged bias ≤ 0.007 m³ m⁻³). Although the performance of SMAP_E_OPL is marginally worse compared to the official SMAP soil moisture products (SMAP_L2SMP_E and SMAP_L2SMP), partly because it uses previous-day soil temperature estimates rather than the current-day estimates used by the official non-NRT products, the difference is less than the uncertainty of conventional ground-based soil moisture sensors. It is essential to emphasize that SMAP_E_OPL can provide timely high-spatial-resolution soil moisture information whose latency is determined solely by that of the SMAP L1B T_B data, and thus is suited for operational forecasting and monitoring systems. Soil moisture retrievals from the descending (AM) orbit in general show lower ubRMSE, lower RMSE, and higher R than those from the ascending (PM) orbit, although some exceptions exist.

Additionally, it is evident from Table 2 that the use of different LSMs does not lead to significant differences in the domain-averaged performance of soil moisture retrievals. This demonstrates the stability of the SMAP_E_OPL processor with respect to different LSMs used to provide ancillary information of T_eff to the soil moisture retrieval algorithms. As presented in Table 2, similar standard deviations of the evaluation metrics between the soil moisture retrieval processors are obtained, which indicate that the performance of the soil moisture retrievals is also comparable across the SCAN and USCRN stations within CONUS. Figure 2 further supports this finding, as almost identical spatial patterns of the ubRMSE, RMSE, bias, and R between the soil moisture retrievals from SMAP_E_OPL (JULES), SMAP_L2SMP_E, and SMAP_L2SMP are observed. Overall, the western CONUS exhibits comparatively lower ubRMSE, RMSE, and bias accompanied by higher R values, in contrast to the eastern CONUS. This difference can be attributed to relatively drier conditions (low soil moisture dynamics) in the western CONUS and wetter soil conditions (high soil moisture dynamics) in the eastern CONUS.

FIGURE 2

Friedl et al., 2002) is also provided.

The geospatial distribution of evaluation metrics (ubRMSE, RMSE, bias, and R) in Figure 2a corresponds well with the distribution of land cover types (Figure 2m). Specifically, the error metrics (i.e., ubRMSE, RMSE, and bias) tend to be relatively low in shrublands and grasslands, while the R is relatively high in croplands. These patterns are clearly seen in the distribution plots of the metrics (shown in Figure 3) for grasslands, croplands, and shrublands, which are three dominant land cover types within the SCAN and USCRN stations. Figure 3 also shows that the error metrics are more spatially distributed in croplands than in other land cover types, while the R values are more spatially distributed in grasslands and shrublands. Likewise, the level and distribution of the evaluation metrics for different land cover types are comparable between the SMAP_E_OPL and official SMAP soil moisture products.

FIGURE 3

Note that while the station-to-pixel evaluation approach is insufficient to draw robust conclusions, it is also the approach adopted in the SMAP assessment report (O’Neill et al., 2021b). To further support our results, we present T_B comparisons between the IDSW and BG interpolation methods in Supplementary Text 1 (Supplementary Figure 1; Supplementary Table 3) and soil moisture comparisons between SMAP_E_OPL and SMAP_L2SMP_E over the global domain in Supplementary Text 2 (Supplementary Figures 2-4).

3.2 Enhancing hydrological models: use of SMAP_E_OPL for real-time soil moisture assimilation

The assimilation of soil moisture data obtained through remote sensing with land surface models has been shown to improve the characterization of land surface conditions (Azimi et al., 2020; Baugh et al., 2020; Bolten et al., 2010). In this section, we examine the utility of the real-time SMAP_E_OPL data for data assimilation (DA). Details of the soil moisture DA method are provided in Supplementary Text 3 and Supplementary Table 4, and the experimental setup is described in Supplementary Text 4 and Supplementary Table 5. Soil moisture estimates from the model without assimilation (open loop; OL) and the model integration with DA are evaluated using in situ soil moisture measurements from the SCAN and USCRN stations using ubRMSE and R. Spatial mean and standard deviation of the evaluation metrics, percentage improvement in the soil moisture ubRMSE and R by assimilating the SMAP retrievals as compared to the OL soil moisture estimates, and percentage of grid cells improved by DA are presented in Table 3. As shown in Table 3, the domain-averaged ubRMSE and R of the OL soil moisture are 0.052 m³ m⁻³ (±0.019 m³ m⁻³) and 0.671 (±0.117), respectively, and assimilation of the SMAP soil moisture retrievals reduces the ubRMSE and increases the R. The percentage improvement achieved by the SMAP DA ranges from 3.5% to 4.8% for ubRMSE and from 1.8% to 3.4% for R. In the DA integration, soil moisture estimates are enhanced for more than 64% and 58% of model grid cells with respect to the ubRMSE and R, respectively.

It can be noted from Table 3 that the effectiveness of SMAP_E_OPL in improving the soil moisture estimates within the DA system is comparable to that of the official SMAP products (SMAP_L2SMP_E and SMAP_L2SMP) when the results are compared against the SCAN and USCRN measurements. Regarding the domain-averaged ubRMSE, the 9-km products (SMAP_E_OPL and SMAP_L2SMP_E) show slightly better performance than the 36-km SMAP_L2SMP data while SMAP_E_OPL exhibits marginally lower R than that of SMAP_L2SMP_E and SMAP_L2SMP. For both evaluation metrics, the NRT 9-km soil moisture DA (DA_OPL) exhibits a slightly lower performance compared to the high-latency 9-km soil moisture DA (DA_L2E). However, given the short latency (∼3 min when the SMAP NRT L1B T_B is available) of the SMAP_E_OPL data, these minor variations in performance may be inconsequential concerning the practicality of integrating the SMAP_E_OPL into operational soil moisture DA systems.

The spatially distributed differences between the soil moisture ubRMSE and R in DA and OL cases (DA minus OL) are shown in Figure 4. The figure shows that the performance of the SMAP_E_OPL soil moisture DA has a similar spatial pattern to that of the SMAP_L2SMP_E DA. The effectiveness of assimilating all SMAP products to improve the model estimates of soil moisture varies based on dominant land cover types (such as grasslands, croplands, and shrublands) present in the SCAN and USCRN stations (Figure 5). Assimilating SMAP soil moisture data decreases ubRMSE across all three land cover types, and enhances R for grasslands and croplands, while it decreases R for shrublands (Figure 5b). These findings are in agreement with those found from Figure 3 where all three SMAP products (9-km SMAP_E_OPL, 9-km SMAP_L2SMP_E, and 36-km SMAP_L2SMP) exhibited the smallest ubRMSE, RMSE, and bias in all land cover types, but the lowest R in shrublands.

FIGURE 4

FIGURE 5

It can be noted from Figures 5c,d that the number of improved grid cells in the SMAP_E_OPL DA is comparable to that found in the official SMAP DA across various land cover types. These findings affirm the feasibility of incorporating the NRT 9-km SMAP soil moisture retrievals (SMAP_E_OPL) into operational data assimilation (DA) systems.

4 Discussion: implications for time-sensitive applications

The study implemented the SMAP_E_OPL framework to retrieve soil moisture in NRT, which is then assimilated in LISF with three different LSMs. Results demonstrate that the SMAP_E_OPL provides similar quality of soil moisture retrievals as compared to SMAP L2SMP_E product. The soil moisture estimates from LSMs are also improved after assimilating the NRT soil moisture. The timely, accurate soil moisture provides crucial information for various applications. For example, soil moisture and rainfall are the two key factors in flash flood development and landslides in mountainous areas (Mao et al., 2019). About 60% of rainfall over land originates from land evaporation, which is driven by the soil-plant-atmosphere interaction (Brocca et al., 2017). NRT soil moisture data, ideally within 3 h from time of acquisition (Brocca et al., 2017; Rodríguez-Fernández et al., 2019), enhances the initialization of land surface conditions for numerical weather forecasting and produces more accurate estimates of the rainfall, air temperature and humidity near the surface (Deardorff, 1977; Koster et al., 2004; Kwon et al., 2024; 2026). These estimates from the timely forecasting system provide valuable information for early-warning and emergency disaster response systems. Incorporating antecedent soil moisture conditions with forecasted rainfall can improve flood peak prediction (Baugh et al., 2020; Grillakis et al., 2016) and thus provide timely warning information for disaster mitigation and emergency response to upcoming flash floods and landslides (Biswas et al., 2022; Zhao et al., 2019). In this respect, although it still has room for further improvement, SMAP_E_OPL can potentially provide great benefits for managing and mitigating natural disaster-induced damages by delivering timely (within 3 h) high-spatial-resolution soil moisture data.

5 Summary and conclusions

This study implements the operational enhanced SMAP soil moisture retrieval processor (i.e., SMAP_E_OPL) in the NASA LISF with the aim of producing NRT hourly soil moisture products at 9-km spatial resolution for use within operational systems. To achieve this, the standard retrieval procedure is followed, using the same inversion algorithms and preprocessed ancillary data as used for the official SMAP_L2SMP_E soil moisture products. However, efforts have been made to reduce the reliance of SMAP_E_OPL on data sources with high temporal latency. Specifically, SMAP_E_OPL relies solely on the SMAP L1B T_B data and uses soil temperature and snow depth outputs from the NRT US Air Force 557th WW global operational LIS-LSMs such as Noah, Noah-MP, and JULES. Furthermore, to enhance processing efficiency, SMAP_E_OPL adopts the IDSW interpolation method, instead of the BG method used in SMAP_L2SMP_E. This modification serves to decrease the processing time while upholding the integrity of the outcomes. Additionally, bias-correction of the LIS-LSM-based effective temperature (T_eff) is performed to ensure consistency with the SMAP_L2SMP_E product.

The SMAP_E_OPL soil moisture is produced over a global domain at a spatial resolution of 0.14° (longitude) by 0.09° (latitude). In this study, we evaluate the performance of the SMAP_E_OPL product over CONUS using in situ soil moisture measurements from SCAN and USCRN. The performance of the SMAP_E_OPL soil moisture retrievals is found to be comparable to that of the official (high latency) SMAP soil moisture products (9-km SMAP_L2SMP_E and 36-km SMAP_L2SMP) based on domain-averaged evaluation metrics. SMAP products (SMAP_E_OPL, SMAP_L2SMP_E and SMAP_L2SMP) exhibit similar spatial patterns of performance, which correspond to those of land cover types. Using different LSMs (Noah, Noah-MP, and JULES) to estimate T_eff, SMAP_E_OPL achieves equivalent performance in retrieving soil moisture, which implies flexibility of the SMAP_E_OPL processor in terms of LSMs.

In conclusion, this study demonstrates that SMAP_E_OPL achieves a level of performance comparable to that of the official SMAP soil moisture products, while providing high-spatial-resolution soil moisture data with significantly reduced latency. This implies the feasibility of SMAP_E_OPL to be used in NRT operational forecasting and monitoring systems. Particular emphasis is placed on the advantages of SMAP_E_OPL relative to existing NRT SMAP products in terms of (a) enhanced posted grid resolution compared to the standard 36-km NRT product, (b) generation of NRT hourly soil moisture data while preserving the original 2–3 days revisit time and without relying on external data sources [compared to 1-km NRT daily products (e.g., Yin et al., 2020)], and (c) the flexibility of the SMAP_E_OPL processor for application in various operational systems. Thus, the development of the SMAP_E_OPL soil moisture retrievals is a promising capability for improving land characterization within operational environments.

However, the following limitations should be addressed in future studies before the operational implementation of the SMAP_E_OPL processor at the global scale. First, the evaluations in this study are based on station-to-pixel comparisons following the Sparse Network Assessment used in the official SMAP quality assessment report (O’Neill et al., 2021b), along with the quality-control protocols of Chen et al. (2018). We acknowledge that this approach is subject to representativeness error. In addition, ground-measurement-based evaluations are conducted over CONUS to ensure the quality of soil moisture estimates, whereas global-scale evaluations are performed based on comparisons between SMAP_E_OPL and SMAP_L2SMP_E, using the latter as a reference (Supplementary Text 2 and Supplementary Figures 2-4), because the primary objective of this study is to develop SMAP_E_OPL with performance comparable to that of SMAP_L2SMP_E. Future studies may further evaluate these SMAP soil moisture products using more spatially distributed metrics, such as triple collocation analysis, while also considering periods of extreme hydrometeorological events, to better support the operational applicability of the developed processor. Second, this study presents the overall performance of SMAP_E_OPL after applying all major design differences relative to SMAP_L2SMP_E, including the interpolation method, spatial grid/projection, and sources of effective soil temperature simulation and snow/frozen-soil flags (Table 1). To improve future modifications of SMAP_E_OPL, explicit analyses of the impact of each design change on soil moisture retrieval performance are suggested. Third, as noted, the current version of the SMAP_E_OPL processor employs SCA-V for soil moisture retrieval, whereas DCA has recently been adopted as the SMAP baseline algorithm. Future versions of the SMAP_E_OPL processor may benefit from using DCA, along with evaluations over a more recent data period. Lastly, the IDSW interpolation method may have limitations in regions with strong land surface heterogeneity because it does not explicitly account for data-to-data correlation and can be sensitive to irregularly distributed data (Li et al., 2020; Liu P.-W. et al., 2021; Lu and Wong, 2008), although it showed interpolation performance comparable to that of the BG method in this study (Supplementary Text 1, Supplementary Figure 1; Supplementary Table 3). More sophisticated yet computationally efficient interpolation methods should be explored in future studies.

References

• 1

  AhmadJ. A.FormanB. A.KumarS. V. (2022). Soil moisture estimation in south Asia via assimilation of SMAP retrievals. Hydrol. Earth Syst. Sci.26 (8), 2221–2243. 10.5194/hess-26-2221-2022

  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  Air Force Weather Agency (2013). Algorithm Description Documentation for the Cloud Depiction and Forecast System II, 516pp. USA: Offutt Air Force Base, 2nd Weather Group, Nebraska.

  • Google Scholar
  • View reference in article

• 3

  ArsenaultK. R.KumarS. V.GeigerJ. V.WangS.KempE.MockoD. M.et al (2018). The land surface data toolkit (LDT v7.2) – a data fusion environment for land data assimilation systems. Geosci. Model Dev.11 (9), 3605–3621. 10.5194/gmd-11-3605-2018

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  AzimiS.DarianeA. B.ModanesiS.Bauer-MarschallingerB.BindlishR.WagnerW.et al (2020). Assimilation of sentinel 1 and SMAP – based satellite soil moisture retirevals into SWAT hydrological model: the impact of satellite revisit time and product spatial resolution on flood simulations in small basins. J. Hydrol.581, 124367. 10.1016/j.jhydrol.2019.124367

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  BallJ. T.WoodrowI. E.BerryJ. A. (1987). “A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions,” in Progress in Photosynthesis Research. Editor BigginsJ. (Dordrecht, Netherlands: Springer), 221–234. 10.1007/978-94-017-0519-6_48

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BaughC.de RosnayP.LawrenceH.JurlinaT.DruschM.ZsoterE.et al (2020). The impact of SMOS soil moisture data assimilation within the operational global flood awareness system (GloFAS). Remote Sens.12 (9), 1490. 10.3390/rs12091490

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BellJ. E.PaleckiM. A.BakerC. B.CollinsW. G.LawrimoreJ. H.LeeperR. D.et al (2013). U.S. climate reference network soil moisture and temperature observations. J. Hydrometeorol.14 (3), 977–988. 10.1175/JHM-D-12-0146.1

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  BestM. J.PryorM.ClarkD. B.RooneyG. G.EsseryR. L. H.M´enardC. B.et al (2011). The joint UK Land environment simulator (JULES), model description – part 1: energy and water fluxes. Geosci. Model Dev.4 (3), 677–699. 10.5194/gmd-4-677-2011

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  BhuiyanH.McNairnH.PowersJ.FriesenM.PachecoA.JacksonT. J.et al (2018). Assessing SMAP soil moisture scaling and retrieval in the carman (Canada) study site. Vadose Zone J.17 (1), 180132. 10.2136/vzj2018.07.0132

  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  BiswasN. K.StanleyT. A.KirschbaumD. B.AmatyaP. M.MeechaiyaC.PoortingaA.et al (2022). A dynamic landslide hazard monitoring framework for the lower Mekong region. Front. Earth Sci.10, 1057796. 10.3389/feart.2022.1057796

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  BlyverketJ.HamerP. D.BertinoL.AlbergelC.FairbairnD.LahozW. A. (2019). An evaluation of the EnKF vs. EnOI and the assimilation of SMAP, SMOS and ESA CCI soil moisture data over the contiguous US. Remote Sens.11 (5), 478. 10.3390/rs11050478

  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  BoltenJ. D.CrowW. T.ZhanX.JacksonT. J.ReynoldsC. A. (2010). Evaluating the utility of remotely sensed soil moisture retrievals for operational agricultural drought monitoring. IEEE J. Sel. Top. Appl.3 (1), 57–66. 10.1109/JSTARS.2009.2037163

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  BroccaL.CrowW. T.CiabattaL.MassariC.de RosnayP.EnenkelM.et al (2017). A review of the applications of ASCAT soil moisture products. IEEE J. Sel. Top. Appl.10 (5), 2285–2306. 10.1109/JSTARS.2017.2651140

  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  BroccaL.TarpanelliA.FilippucciP.DorigoW.ZaussingerF.GruberA.et al (2018). How much water is used for irrigation? A new approach exploiting coarse resolution satellite soil moisture products. Int. J. Appl. Earth Obs.73, 752–766. 10.1016/j.jag.2018.08.023

  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  BrutsaertB. (1982). Evaporation in the Atmosphere: Theory, History and Application. Kluwer Acad., Norwell, MA, USA, 299. 10.1007/978-94-017-1497-6

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  CaiX.YangZ.-L.XiaY.HuangM.WeiH.LeungL. R.et al (2014). Assessment of simulated water balance from Noah, Noah-MP, CLM, and VIC over CONUS using the NLDAS test bed. J. Geophys. Res. Atmos.119 (24), 13751–13770. 10.1002/2014JD022113

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  ChanS. K.BindlishR.O’NeillP. E.NjokuE.JacksonT.CollianderA.et al (2016). Assessment of the SMAP passive soil moisture product. IEEE Trans. Geosci. Remote Sens.54 (8), 4994–5007. 10.1109/TGRS.2016.2561938

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  ChanS. K.BindlishR.O’NeillP.JacksonT.NjokuE.DunbarS.et al (2018). Development and assessment of the SMAP enhanced passive soil moisture product. Remote Sens. Environ.204, 931–941. 10.1016/j.rse.2017.08.025

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  ChaoL.ZhangK.WangS.GuZ.XuJ.BaoH. (2022). Assimilation of surface soil moisture jointly retrieved by multiple microwave satellites into the WRF-hydro model in ungauged regions: towards a robust flood simulation and forecasting. Environ. Modell. Softw.154, 105421. 10.1016/j.envsoft.2022.105421

  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  ChaubellJ. (2016). SMAP Algorithm Theoretical Basis Document: SMAP L1B Enhancement Radiometer Brightness Temperature Data Product. Pasadena, CA (JPL D-56287): Jet Propulsion Laboratory, California Institute of Technology. Available online at: https://nsidc.org/sites/default/files/smap_l1b_tb_e_product_atbd_d-56287.pdf (Accessed June 10, 2026).

  • Google Scholar
  • View reference in article

• 21

  ChenF.DudhiaJ. (2001). Coupling an advanced land surface-hydrology model with the Penn state-NCAR MM5 modeling system. Part I: model implementation and sensitivity. Mon. Weather Rev.129 (4), 569–585. 10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  ChenQ.ZengJ.CuiC.LiZ.ChenK.-S.BaiX.et al (2017). Soil moisture retrieval from SMAP: a validation and error analysis study using ground-based observations over the little washita watershed. IEEE Trans. Geosci. Remote Sens.56 (3), 1394–1408. 10.1109/TGRS.2017.2762462

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  ChenF.CrowW. T.BindlishR.CollianderA.BurginM. S.AsanumaJ.et al (2018). Global-scale evaluation of SMAP, SMOS and ASCAT soil moisture products using triple collocation. Remote Sens. Env.214, 1–13. 10.1016/j.rse.2018.05.008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  ChoudhuryB.SchmuggeT.MoT. (1982). A parameterization of effective soil temperature for microwave emission. J. Geophys. Res.87 (C2), 1301–1304. 10.1029/JC087iC02p01301

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  ClarkD. B.MercadoL. M.SitchS.JonesC. D.GedneyN.BestM. J.et al (2011). The joint UK land Environment simulator (JULES), model description – part 2: carbon fluxes and vegetation dynamics. Geosci. Model Dev.4 (3), 701–722. 10.5194/gmd-4-701-2011

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  CollianderA.JacksonT. J.BindlishR.ChanS.DasN.KimS. B.et al (2017). Validation of SMAP surface soil moisture products with core validation sites. Remote Sens. Environ.191, 215–231. 10.1016/j.rse.2017.01.021

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  CoxP. M.BettsR. A.BuntonC. B.EsseryR. L. H.RowntreeP. R.SmithJ. (1999). The impact of new land surface physics on the GCM simulation of climate and climate sensitivity. Clim. Dyn.15, 183–203. 10.1007/s003820050276

  • CrossRef
  • Google Scholar

• 28

  CrowW. T.ChenF.ReichleR. H.LiuQ. (2017). L band microwave remote sensing and land data assimilation improve the representation of prestorm soil moisture conditions for hydrologic forecasting. Geophys. Res. Lett.44 (11), 5495–5503. 10.1002/2017GL073642

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  DasN. N.EntekhabiD.NjokuE. G. (2011). An algorithm for merging SMAP radiometer and radar data for high-resolution soil-moisture retrieval. IEEE Trans. Geosci. Remote Sens.49 (5), 1504–1512. 10.1109/TGRS.2010.2089526

  • CrossRef
  • Google Scholar

• 30

  DasN. N.EntekhabiD.DunbarR. S.ChaubellM. J.CollianderA.YuehS.et al (2019). The SMAP and copernicus Sentinel-1A/B microwave active passive high resolution surface soil moisture product. Remote Sens. Environ.233, 111380. 10.1016/j.rse.2019.111380

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  DeardorffJ. (1977). A parameterization of ground-surface moisture content for use in atmospheric prediction models. J. Appl. Meteorol. Clim.16 (11), 1182–1185. 10.1175/1520-0450(1977)016<1182:APOGSM>2.0.CO;2

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  DiamondH. J.KarlT. R.PaleckiM. A.BakerC. B.BellJ. E.LeeperR. D.et al (2013). U.S. climate reference network after one decade of operations: status and assessment. B. Am. Meteorol. Soc.94 (4), 485–498. 10.1175/BAMS-D-12-00170.1

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  DickinsonR. E.Henderson-SellersA.KennedyP. J. (1993). Biosphere–atmosphere transfer scheme (BATS) version 1e as coupled to NCAR community climate model. Natl. Cent. Atmos. Res.80. 10.5065/D67W6959

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  DirmeyerP.HalderS. (2016). Sensitivity of numerical weather forecasts to initial soil moisture variations in CFSv2. Wea. Forecast.31 (6), 1973–1983. 10.1175/WAF-D-16-0049.1

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  EkM. B.MitchellK. E.LinY.RogersE.GrunmannP.KorenV.et al (2003). Implementation of noah land surface model advances in the national centers for environmental prediction operational mesoscale eta model. J. Geophys. Res.108 (D22), 8851. 10.1029/2002JD003296

  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  EngdaT. A.KellenersT. J. (2016). Soil moisture-based drought monitoring at different time scales: a case study for the U.S. Great plains. J. Am. Water Resour. Assoc.52 (1), 77–88. 10.1111/1752-1688.12367

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  EntekhabiD.NjokuE. G.O’NeillP. E.KelloggK. H.CrowW. T.EdelsteinW. N.et al (2010). The soil moisture active passive (SMAP) mission. P. IEEE98 (5), 704–716. 10.1109/JPROC.2010.2043918

  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  FangL.ZhanX.YinJ.LiuJ.SchullM.WalkerJ. P.et al (2020). An intercomparison study of algorithms for downscaling SMAP radiometer soil moisture retrievals. J. Hydrometeorol.21 (8), 1761–1775. 10.1175/JHM-D-19-0034.1

  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  FriedlM. A.McIverD. K.HodgesJ. C. F.ZhangX. Y.MuchoneyD.StrahlerA. H.et al (2002). Global land cover mapping from MODIS: algorithms and early results. Remote Sens. Environ.83 (1–2), 287–302. 10.1016/S0034-4257(02)00078-0

  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  GonsamoA.ChenJ. M.HeL.SunY.RogersC.LiuJ. (2019). Exploring SMAP and OCO-2 observations to monitor soil moisture control on photosynthetic activity of global drylands and croplands. Remote Sens. Environ.232, 111314. 10.1016/j.rse.2019.111314

  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  GrillakisM. G.KoutroulisA. G.KommaJ.TsanisI. K.WagnerW.BlöschlG. (2016). Initial soil moisture effects on flash flood generation–A comparison between basins of contrasting hydro-climatic conditions. J. Hydrol.541, 206–217. 10.1016/j.jhydrol.2016.03.007

  • CrossRef
  • Google Scholar
  • View reference in article

• 42

  HenglT.de JesusJ. M.HeuvelinkG. B. M.GonzalezM. R.KilibardaM.BlagotićA.et al (2017). SoilGrids250m: global gridded soil information based on machine learning. PLoS ONE12 (2), e0169748. 10.1371/journal.pone.0169748

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  HuffmanG. J.BolvinD. T.BraithwaiteD.HsuK.JoyceR.KiddC.et al (2020). NASA Global Precipitation Measurement (GPM) Integrated Multi-Satellite Retrievals for GPM (IMERG). NASA Algorithm Theoretical Basis Doc, 33pp. Greenbelt, MD, USA: NASA GSFC. Available online at: https://gpm.nasa.gov/sites/default/files/2020-05/IMERG_ATBD_V06.3.pdf (Accessed June 10, 2026).

  • Google Scholar
  • View reference in article

• 44

  JacksonT. J.GasiewskiA.OldakA.KleinM.NjokuE.YevgrafovA.et al (2002). Soil moisture retrieval using the C-band polarimetric scanning radiometer during the southern great plains 1999 experiment. IEEE Trans. Geosci. Remote Sens.40 (10), 2151–2161. 10.1109/TGRS.2002.802480

  • CrossRef
  • Google Scholar
  • View reference in article

• 45

  JordanR. A. (1991). “A one-dimensional temperature model for a snow cover,” in Technical Documentation for SNTHERM.89. Washington, DC, USA: U.S. Army Corps of Engineers. Available online at: https://hdl.handle.net/11681/11677 (Accessed June 10, 2026).

  • Google Scholar
  • View reference in article

• 46

  JudgeJ.LiuP. W.Monsiváis-HuerteroA.BongiovanniT.ChakrabartiS.Steele-DunneS. C.et al (2021). Impact of vegetation water content information on soil moisture retrievals in agricultural regions: an analysis based on the SMAPVEX16-MicroWEX dataset. Remote Sens. Environ.265, 112623. 10.1016/j.rse.2021.112623

  • CrossRef
  • Google Scholar
  • View reference in article

• 47

  KempE. M.WegielJ. W.KumarS. V.GeigerJ. V.MockoD. M.JacobJ. P.et al (2022). A NASA-air force precipitation analysis for near-real-time operations. J. Hydrometeorol.23 (6), 965–989. 10.1175/JHM-D-21-0228.1

  • CrossRef
  • Google Scholar
  • View reference in article

• 48

  KimH.LakshmiV.KwonY.KumarS. V. (2021). First attempt of global-scale assimilation of subdaily scale soil moisture estimates from CYGNSS and SMAP into a land surface model. Environ. Res. Lett.16 (7), 074041. 10.1088/1748-9326/ac0ddf

  • CrossRef
  • Google Scholar
  • View reference in article

• 49

  KnowlesK.SavoieM.ArmstrongR.BrodzikM. J. (2006). AMSR-E/Aqua Daily EASE-grid Brightness Temperatures, Version 1: User Guide. Boulder, Colorado USA: NASA National Snow and Ice Data Center Distributed Active Archive Center. Available online at: https://nsidc.org/sites/default/files/multi_nsidc-0301-v001-userguide.pdf (Accessed June 10, 2026).

  • Google Scholar
  • View reference in article

• 50

  KosterR. D.DirmeyerP. A.GuoZ.BonanG.ChanE.CoxP.et al (2004). Regions of strong coupling between soil moisture and precipitation. Science305 (5687), 1138–1140. 10.1126/science.1100217

  • CrossRef
  • Google Scholar
  • View reference in article

• 51

  KosterR. D.MahanamaS. P. P.YamadaT. J.BalsamoG.BergA. A.BoisserieM.et al (2010). Contribution of land surface initialization to subseasonal forecast skill: first results from a multi-model experiment. Geophys. Res. Lett.37 (2), L02402. 10.1029/2009GL041677

  • CrossRef
  • Google Scholar
  • View reference in article

• 52

  KumarS. V.Peters-LidardC. D.TianY.HouserP. R.GeigerJ.OldenS.et al (2006). Land information system: an interoperable framework for high resolution land surface modeling. Environ. Modell. Softw.21 (10), 1402–1415. 10.1016/j.envsoft.2005.07.004

  • CrossRef
  • Google Scholar
  • View reference in article

• 53

  KumarS.Peters-LidardC.TianY.ReichleR. H.GeigerJ.AlongeC.et al (2008). An integrated hydrologic modeling and data assimilation framework. Computer41 (12), 52–59. 10.1109/MC.2008.475

  • CrossRef
  • Google Scholar
  • View reference in article

• 54

  KumarS. V.DirmeyerP. A.Peters-LidardC. D.BindlishR.BoltenJ. (2018). Information theoretic evaluation of satellite soil moisture retrievals. Remote Sens. Environ.204, 392–400. 10.1016/j.rse.2017.10.016

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 55

  KwonY.KumarS. V.NavariM.MockoD. M.KempE. M.WegielJ. W.et al (2022). Irrigation characterization improved by the direct use of SMAP soil moisture anomalies within a data assimilation system. Environ. Res. Lett.17 (8), 084006. 10.1088/1748-9326/ac7f49

  • CrossRef
  • Google Scholar
  • View reference in article

• 56

  KwonY.JunS.KimE.SeolK.-H.HongS.KwonI.-H.et al (2024). Improving weather forecast skill of the Korean integrated model by assimilating soil moisture active passive soil moisture anomalies. Q. J. Roy. Meteor. Soc.150 (765), 5305–5336. 10.1002/qj.4871

  • CrossRef
  • Google Scholar
  • View reference in article

• 57

  KwonY.JunS.KimH.SeolK.-H.KwonI.-H.KimE.et al (2026). Synergistic impact of simultaneously assimilating radar- and radiometer-based soil moisture retrievals on the performance of numerical weather prediction systems. Hydrol. Earth Syst. Sci.30, 1261–1290. 10.5194/hess-30-1261-2026

  • CrossRef
  • Google Scholar
  • View reference in article

• 58

  LawstonP. M.SantanelloJ. J. A.KumarS. V. (2017). Irrigation signals detected from SMAP soil moisture retrievals. Geophys. Res. Lett.44 (23), 11860–11867. 10.1002/2017GL075733

  • CrossRef
  • Google Scholar
  • View reference in article

• 59

  LiZ.ZhangX.ZhuR.ZhangZ.WengZ. (2020). Integrating data-to-data correlation into inverse distance weighting. Comput. Geosci.24, 203–216. 10.1007/s10596-019-09913-9

  • CrossRef
  • Google Scholar
  • View reference in article

• 60

  LiuP.-W.JudgeJ.De RooR. D.EnglandA. W.BongiovanniT. (2016). Uncertainty in soil moisture retrievals using the SMAP combined active–passive algorithm for growing sweet corn. IEEE J. Sel. Top. Appl.9 (7), 3326–3339. 10.1109/JSTARS.2016.2562660

  • CrossRef
  • Google Scholar
  • View reference in article

• 61

  LiuP.-W.BindlishR.FangB.LakshmiV.O’NeillP.YangZ.et al (2021a). Assessing disaggregated SMAP soil moisture products in the United States. IEEE J. Sel. Top. Appl.14, 2577–2592. 10.1109/JSTARS.2021.3056001

  • CrossRef
  • Google Scholar
  • View reference in article

• 62

  LiuZ.ZhangZ.ZhouC.MingW.DuZ. (2021b). An adaptive inverse-distance weighting interpolation method considering spatial differentiation in 3D geological modeling. Geosciences11 (2), 51. 10.3390/geosciences11020051

  • CrossRef
  • Google Scholar
  • View reference in article

• 63

  LiuP.-W.BindlishR.O’NeillP.FangB.LakshmiV.YangZ.et al (2022). Thermal hydraulic disaggregation of SMAP soil moisture over the Continental United States. IEEE J. Sel. Top. Appl.15, 4072–4093. 10.1109/JSTARS.2022.3165644

  • CrossRef
  • Google Scholar

• 64

  LuG. Y.WongD. W. (2008). An adaptive inverse-distance weighting spatial interpolation technique. Comput. Geosci.34, 1044–1055. 10.1016/j.cageo.2007.07.010

  • CrossRef
  • Google Scholar

• 65

  MaoY.CrowW. T.NijssenB. (2019). A framework for diagnosing factors degrading the streamflow performance of a soil moisture data assimilation system. J. Hydrometeorol.20 (1), 79–97. 10.1175/JHM-D-18-0115.1

  • CrossRef
  • Google Scholar

• 66

  MassariC.BroccaL.MoramarcoT.TramblayY.LescotJ. F. D. (2014). Potential of soil moisture observations in flood modelling: estimating initial conditions and correcting rainfall. Adv. Water Resour.74, 44–53. 10.1016/j.advwatres.2014.08.004

  • CrossRef
  • Google Scholar

• 67

  MassariC.CamiciS.CiabattaL.BroccaL. (2018). Exploiting satellite-based surface soil moisture for flood forecasting in the mediterranean area: state update versus rainfall correction. Remote Sens.10 (2), 292. 10.3390/rs10020292

  • CrossRef
  • Google Scholar

• 68

  McNallyA.ArsenaultK.KumarS.ShuklaS.PetersonP.WangS.et al (2017). A land data assimilation system for Sub-Saharan Africa food and water security applications. Sci. Data4, 170012. 10.1038/sdata.2017.12

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 69

  MerlinO.WalkerJ.ChehbouniA.KerrY. (2008). Towards deterministic downscaling of SMOS soil moisture using MODIS derived soil evaporative efficiency. Remote Sens. Environ.112 (10), 3935–3946. 10.1016/j.rse.2008.06.012

  • CrossRef
  • Google Scholar

• 70

  MilmanA.ArsanoY. (2014). Climate adaptation and development: contradictions for human security in gambella, Ethiopia. Glob. Environ. Chang.29, 349–359. 10.1016/j.gloenvcha.2013.11.017

  • CrossRef
  • Google Scholar

• 71

  MironovV. L.KosolapoveL. G.FominS. V. (2009). Physically and mineralogically based spectroscopic dielectric model for moist soils. IEEE Trans. Geosci. Remote Sens.47 (7), 2059–2070. 10.1109/TGRS.2008.2011631

  • CrossRef
  • Google Scholar

• 72

  MishraA.VuT.VeettilA. V.EntekhabiD. (2017). Drought monitoring with soil moisture active passive (SMAP) measurements. J. Hydrol.552, 620–632. 10.1016/j.jhydrol.2017.07.033

  • CrossRef
  • Google Scholar

• 73

  MladenovaI. E.BoltenJ. D.CrowW. T.SazibN.CoshM. H.TuckerC. J.et al (2019). Evaluating the operational application of SMAP for global agricultural drought monitoring. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.12 (9), 3387–3397. 10.1109/JSTARS.2019.2923555

  • CrossRef
  • Google Scholar

• 74

  MoranM. S.Peters-LidardC. D.WattsJ. M.McElroyS. (2004). Estimating soil moisture at the watershed scale with satellite-based radar and land surface models. Can. J. Remote Sens.30 (5), 805–826. 10.5589/m04-043

  • CrossRef
  • Google Scholar

• 75

  NieW.KumarS. V.BindlishR.LiuP.-W.WangS. (2022). Remote sensing-based vegetation and soil moisture constraints reduce irrigation estimation uncertainty. Environ. Res. Lett.17 (8), 084010. 10.1088/1748-9326/ac7ed8

  • CrossRef
  • Google Scholar

• 76

  NiuG.-Y.YangZ.-L. (2004). The effects of canopy processes on snow surface energy and mass balances. J. Geophys. Res.109 (D23), D23111. 10.1029/2004JD004884

  • CrossRef
  • Google Scholar

• 77

  NiuG.-Y.YangZ.-L. (2006). Effects of frozen soil on snowmelt runoff and soil water storage at a Continental scale. J. Hydrometeorol.7 (5), 937–952. 10.1175/JHM538.1

  • CrossRef
  • Google Scholar

• 78

  NiuG.-Y.YangZ.-L.DickinsonR. E.GuldenL. E.SuH. (2007). Development of a simple groundwater model for use in climate models and evaluation with gravity recovery and climate experiment data. J. Geophys. Res.112 (D7), D07103. 10.1029/2006JD007522

  • CrossRef
  • Google Scholar

• 79

  NiuG.-Y.YangZ.-L.MitchellK. E.ChenF.EkM. B.BarlageM.et al (2011). The community noah land surface model with multiparameterization options (Noah-MP): 1. Model description and evaluation with local-scale measurements. J. Geophys. Res.116 (D12), D12109. 10.1029/2010JD015139

  • CrossRef
  • Google Scholar

• 80

  NjokuE.KongJ. (1977). Theory for passive microwave remote sensing of near-surface soil moisture. J. Geophys. Res.82 (20), 3108–3118. 10.1029/JB082i020p03108

  • CrossRef
  • Google Scholar

• 81

  O'NeillP. E.ChanS.NjokuE. G.JacksonT.BindlishR.ChaubellJ. (2019). SMAP Enhanced L2 Radiometer Half-Orbit 9 Km EASE-Grid Soil Moisture, Version 3NASA National Snow and Ice Data Center Distributed Active Archive Center, Boulder, Colorado USA. 10.5067/017XZSKMLTT2

  • CrossRef
  • Google Scholar

• 82

  O'NeillP. E.ChanS.NjokuE. G.JacksonT.BindlishR.ChaubellJ. (2022). Near real-time SMAP L2 Radiometer Half-Orbit 36 Km EASE-Grid Soil Moisture, Version 107. Boulder, Colorado USA: NASA National Snow and Ice Data Center Distributed Active Archive Center. 10.5067/NCTT8THPWRTL

  • CrossRef
  • Google Scholar

• 83

  OchsnerT. E.CoshM. H.CuencaR. H.DorigoW. A.DraperC. S.HagimotoY.et al (2013). State of the art in large-scale soil moisture monitoring. Soil Sci. Soc. Am. J.77 (6), 1888–1919. 10.2136/sssaj2013.03.0093

  • CrossRef
  • Google Scholar

• 84

  O’NeillP.BindlishR.ChanS.ChaubellJ.CollianderA.NjokuE.et al (2021a). Soil Moisture Active Passive (SMAP), Algorithm Theoretical Basis Document (ATBD) Level 2 & 3 Soil Moisture (Passive) Data Products. Pasadena, CA, USA: California Institute of Technology. Available online at: https://nsidc.org/sites/default/files/l2_sm_p_atbd_rev_g_final_oct2021_0.pdf (Accessed June 10, 2026).

  • Google Scholar

• 85

  O’NeillP.ChanS.BindlishR.ChaubellM.CollianderA.ChenF.et al (2021b). “Calibration and validation for the L2/3_SM_P Version 8 and L2/3_SM_P_E version 5 data products,” in SMAP Project, JPL D-56297. Pasadena, CA, USA: Jet Propulsion Laboratory. Available online at: https://nsidc.org/sites/default/files/l2_sm_p_ar_r18_final_oct2021.pdf (Accessed June 10, 2026).

  • Google Scholar

• 86

  PanM.CaiX.ChaneyN. W.EntekhabiD.WoodE. F. (2016). An initial assessment of SMAP soil moisture retrievals using high-resolution model simulations and in situ observations. Geophys. Res. Lett.43 (18), 9662–9668. 10.1002/2016GL069964

  • CrossRef
  • Google Scholar

• 87

  PiepmeierJ. R.MohammedP.PengJ.KimE. J.De AmiciG.ChaubellJ.et al (2020). SMAP L1B Radiometer Half-Orbit Time-Ordered Brightness Temperatures, Version 5. NASA National Snow and Ice Data Center Distributed Active Archive Center, Boulder, Colorado USA. 10.5067/ZHHBN1KQLI20

  • CrossRef
  • Google Scholar

• 88

  PoeG. (1990). Optimum interpolation of imaging microwave radiometer data. IEEE Trans. Geosci. Remote Sens.28 (5), 800–810. 10.1109/36.58966

  • CrossRef
  • Google Scholar

• 89

  RahmanM. S.DiL.YuE.LinL.ZhangC.TangJ. (2019). Rapid flood progress monitoring in cropland with NASA SMAP. Remote Sens.11 (2), 191. 10.3390/rs11020191

  • CrossRef
  • Google Scholar

• 90

  RichardsL. A. (1931). Capillary conduction of liquids through porous mediums. Physics1 (5), 318–333. 10.1063/1.1745010

  • CrossRef
  • Google Scholar

• 91

  RodellM.HouserP. R.JamborU.GottschalckJ.MitchellK.MengC.-J.et al (2004). The global land data assimilation system. B. Am. Meteorol. Soc.85 (3), 381–394. 10.1175/BAMS-85-3-381

  • CrossRef
  • Google Scholar

• 92

  Rodríguez-FernándezN.de RosnayP.AlbergelC.RichaumeP.AiresF.PrigentC.et al (2019). SMOS neural network soil moisture data assimilation in a land surface model and atmospheric impact. Remote Sens.11 (11), 1334. 10.3390/rs11111334

  • CrossRef
  • Google Scholar

• 93

  SadriS.WoodE. F.PanM. (2018). Developing a drought-monitoring index for the contiguous US using SMAP. Hydrol. Earth Syst. Sci.22 (12), 6611–6626. 10.5194/hess-22-6611-2018

  • CrossRef
  • Google Scholar

• 94

  SadriS.PanM.WadaY.VergopolanN.SheffieldJ.FamigliettiJ. S.et al (2020). A global near-real-time soil moisture index monitor for food security using integrated SMOS and SMAP. Remote Sens. Environ.246, 111864. 10.1016/j.rse.2020.111864

  • CrossRef
  • Google Scholar

• 95

  SantanelloJr., J. A.LawstonP.KumarS.DennisE. (2019). Understanding the impacts of soil moisture initial conditions on NWP in the context of land-atmosphere coupling. J. Hydrometeorol.20 (5), 793–819. 10.1175/JHM-D-18-0186.1

  • CrossRef
  • Google Scholar

• 96

  SchaakeJ. C.KorenV. I.DuanQ. Y.MitchellK.ChenF. (1996). A simple water balance model (SWB) for estimating runoff at different spatial and temporal scales. J. Geophys. Res.101 (D3), 7461–7475. 10.1029/95JD02892

  • CrossRef
  • Google Scholar

• 97

  SchaeferG. L.CoshM. H.JacksonT. J. (2007). The USDA natural resources conservation service soil climate analysis network (SCAN). J. Atmos. Ocean. Tech.24 (12), 2073–2077. 10.1175/2007JTECHA930.1

  • CrossRef
  • Google Scholar

• 98

  SchmuggeT. J.KustasW. P.RitchieJ. C.JacksonT. J.RangoA. (2002). Remote sensing in hydrology. Adv. Water Res.25 (8–12), 1367–1385. 10.1016/S0309-1708(02)00065-9

  • CrossRef
  • Google Scholar

• 99

  SeneviratneS. I.CortiT.DavinE. L.HirschiM.JaegerE. B.LehnerI.et al (2010). Investigating soil moisture-climate interactions in a changing climate: a review. Earth Sci. Rev.99 (3–4), 125–161. 10.1016/j.earscirev.2010.02.004

  • CrossRef
  • Google Scholar

• 100

  van GenuchtenM. T. (1980). A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J.44 (5), 892–898. 10.2136/sssaj1980.03615995004400050002x

  • CrossRef
  • Google Scholar

• 101

  WegielJ.KumarS. V.Peters-LidardC.KempE. M.GeigerJ. V.WangS.et al (2020). “Data assimilation enhancements to air force Weather’s land information system,” Boston, MA, USA: Amer. Meteor. Soc.Available online at: https://ams.confex.com/ams/2020Annual/meetingapp.cgi/Paper/366952 (Accessed June 10, 2026).

  • Google Scholar

• 102

  WeiH.XiaY.MitchellK. E.EkM. B. (2013). Improvement of the noah land surface model for warm season processes: evaluation of water and energy flux simulation. Hydrol. Process.27 (2), 297–303. 10.1002/hyp.9214

  • CrossRef
  • Google Scholar

• 103

  WronaE.RowlandsonT. L.NambiarM.BergA. A.CollianderA.MarshP. (2017). Validation of the soil moisture active passive (SMAP) satellite soil moisture retrieval in an arctic tundra environment. Geophys. Res. Lett.44 (9), 4152–4158. 10.1002/2017GL072946

  • CrossRef
  • Google Scholar

• 104

  XiaY.MitchellK.EkM.SheffieldJ.CosgroveB.WoodE.et al (2012). Continental-scale water and energy flux analysis and validation for the north American Land Data assimilation system project phase 2 (NLDAS-2): 1. Intercomparison and application of model products. J. Geophys. Res.117 (D3), D03110. 10.1029/2011JD016048

  • CrossRef
  • Google Scholar

• 105

  YangZ.-L.DickinsonR. E. (1996). Description of the biosphere-atmosphere transfer Scheme (BATS) for the soil moisture workshop and evaluation of its performance. Glob. Planet. Change13 (1–4), 117–134. 10.1016/0921-8181(95)00041-0

  • CrossRef
  • Google Scholar

• 106

  YangR.FriedlM. A. (2003). Modeling the effects of three-dimensional vegetation structure on surface radiation and energy balance in boreal forests. J. Geophys. Res.108 (D16), 8615. 10.1029/2002JD003109

  • CrossRef
  • Google Scholar

• 107

  YangZ.-L.NiuG.-Y.MitchellK. E.ChenF.EkM. B.BarlageM.et al (2011). The community noah land surface model with multiparameterization options (Noah-MP): 2. Evaluation over global river basins. J. Geophys. Res.116 (D12), D12110. 10.1029/2010JD015140

  • CrossRef
  • Google Scholar

• 108

  YinJ.ZhanX.LiuJ.MoradkhaniH.FangL.WalkerJ. P. (2020). Near-real-time one-kilometre soil moisture active passive soil moisture data product. Hydrol. Process.34 (21), 4083–4096. 10.1002/hyp.13857

  • CrossRef
  • Google Scholar

• 109

  YoonY.KempE. M.KumarS. V.WegielJ. W.VuyovichC. M.Peters-LidardC. (2022). Development of a global operational snow analysis: the US air force snow and Ice analysis. Remote Sens. Environ.278, 113080. 10.1016/j.rse.2022.113080

  • CrossRef
  • Google Scholar

• 110

  ZhangX. F.ZhangT. T.ZhouP.ShaoY.GaoS. (2017). Validation analysis of SMAP and AMSR2 soil moisture products over the United States using ground-based measurements. Remote Sens.9 (2), 104. 10.3390/rs9020104

  • CrossRef
  • Google Scholar

• 111

  ZhaoB.DaiQ.HanD.DaiH.MaoJ.ZhuoL. (2019). Probabilistic thresholds for landslides warning by integrating soil moisture conditions with rainfall thresholds. J. Hydrol.574, 276–287. 10.1016/j.jhydrol.2019.04.062

  • CrossRef
  • Google Scholar