Soil texture is a fundamental soil property and serve as a crucial input to many Land Surface Models (LSMs). However, current soil texture datasets used in LSMs are usually extrapolated from in-situ scale geological surveys, which may contain high uncertainties due to the mismatch in spatial scales. Here, we propose a method to optimize several currently existing soil texture datasets by using a long-term satellite soil moisture dataset. The optimized soil texture datasets may provide an opportunity to improve land surface simulations in LSMs.

Soil moisture drydown plays an important role in many hydrometeorological processes such as regulating surface energy budget, evapotranspiration, and infiltration. In this study, we analyzed the spatial and temporal characteristics of global soil moisture drydown using the daily-scale long-term satellite soil moisture product NNSM. We find that the time-series of τSand τLremained stable over the years. The spatial distribution of global τSand τLshows an anti-spatial correlation pattern, implying that strong land-atmosphere interaction in the short and long term occurs in different regions. τS. of NNSM is closer to the observation measurement than SMAP. The results show that NNSM can provide a long-term global reference for global soil moisture memory characterization, and for improving land surface models.

High-accuracy near-real-time satellite precipitation estimates (SPEs) provide an opportunity for hydrometeorologists to improve the forecasting of extreme events, such as flood, landslide, tropical cyclone, and other extreme events, at the large scale. However, the currently operational near-real-time SPEs still have larger errors and uncertainties. In this study, we found that there exists a clear relationship of spatial plane function (SPF) between retrieval errors of SPEs and four crucial factors including topography, seasonality, climate type, and rain rate. Based on this finding, we proposed a novel error adjustment method to correct the near-real-time hourly global satellite mapping of precipitation (GSMaP-NRT) estimates in real-time. The new satellite precipitation dataset, namely, ILSF-RT, was then inter-compared with the latest near-real-time GSMaP product suite (i.e., GSMaP-NRT and GSMaP-Gauge-NRT). Verification results show that the proposed method can effectively reduce the retrieval errors of GSMaP-NRT for various terrains and rain rates over different seasons and climate-type areas. The new ILSF-RT even exhibits a general improvement over the GSMaP-Gauge-NRT estimates. Furthermore, one important merit of the new method is that it can perform rather well in validation even when not much historical data were applied as training samples in calibration, for example, during the generation of ILSF-RT, only 45 data pairs of satellite retrievals and ground observations were used for winter season over Chinese arid areas. However, the results of bias score show that the current method seems unsuitable to adjust the rainfall events with higher rain rates (>=1 mm hr−1), which needs to be further improved.

Cloud parameters, as key inputs in radiative transfer algorithms, have a critical impact on surface shortwave radiation (SSR) computation. By introducing a parameterization of cloud transmittance and reflectance, based on radiative transfer simulations, this study improves the accuracy of an existing physically based model which severely underestimates SSR under thick cloud conditions. The cloud parameterization adopts the single-layer cloud model and simulates cloud transmittances and reflectances by varying cloud optical thickness, cloud particle size, and solar zenith angle. The revised model is applied to estimate instantaneous SSR using Moderate-resolution Imaging Spectroradiometer (MODIS) atmospheric and land products. The retrieved SSR is evaluated against observation data from 41 Baseline Surface Radiation Network (BSRN) stations and is also compared with the MODIS official SSR product. The root mean square error (RMSE) of the estimated instantaneous radiation is approximately 52 and 98 W m−2under clear-sky and all-sky conditions, respectively. The accuracy of the improved parameterization is higher than that of the original model, and there is no obvious underestimation of SSR in the case of high cloud optical thickness. Therefore, the new algorithm improves the accuracy of SSR estimates in the presence of thick clouds. Retrievals with the improved model also achieve higher accuracy than the MODIS official SSR product (MCD18A1). Finally, the reliable performance of the scheme at most BSRN stations illustrates that the improved model can be used to map SSR on a global scale.

Soil texture and associated thermal and hydraulic parameters are key to land surface processes. Current global soil datasets are derived from limited soil samples, which are not only very costly but also prone to large uncertainties. While it is difficult to directly retrieve soil properties through satellite remote sensing, this study explores the feasibility of mapping global soil type and thereby corresponding soil texture through the Soil Moisture Active Passive (SMAP) soil moisture product without reference to soil samples. Specifically, for each grid-cell, 12 U.S. Department of Agriculture (USDA) soil types are first used to drive the Noah-MP land surface model and then the optimal one is obtained by referring to a four-year (2015–2018) SMAP soil moisture time series. The proposed scheme can reasonably map the global distribution of soil types in terms of sand/clay content and porosity that are close to the Global Soil Dataset for Earth System Models (GSDE) dataset and outperform the one used in the Global Land Data Assimilation System (GLDAS)/Noah model. The result of this pilot study is very encouraging as it purely relies on satellite data, which is especially important for remote areas where few soil samples are available and conventional soil datasets may have large biases. Further improvements may be achieved upon improved soil organic matter parameterization, through land data assimilation, and by considering additional satellite information.

Downward longwave radiation (DLR) at the surface is a key variable of interest in fields, such as hydrology and climate research. However, existing DLR estimation methods and DLR products are still problematic in terms of both accuracy and spatiotemporal resolution. In this article, we propose a deep convolutional neural network (DCNN)-based method to estimate hourly DLR at 5-km spatial resolution from top of atmosphere (TOA) brightness temperature (BT) of the Himawari-8/Advanced Himawari Imager (AHI) thermal channels, combined with near-surface air temperature and dew point temperature of ERA5 and elevation data. Validation results show that the DCNN-based method outperforms popular random forest and multilayer perceptron-based methods and that our proposed scheme integrating multisource data outperforms that only using remote sensing TOA observations or surface meteorological data. Compared with state-of-the-art CERES-SYN and ERA5-land DLR products, the estimated DLR by our proposed DCNN-based method with physical multisource inputs has higher spatiotemporal resolution and accuracy, with correlation coefficient (CC) of 0.95, root-mean-square error (RMSE) of 17.2 W/m2, and mean bias error (MBE) of −0.8 W/m2in the testing period on the Tibetan Plateau.

Land Surface Temperature (LST) is important for diagnosing surface energy and water exchanges. Accurately predicting LST has over time been challenging for land-atmosphere parameterizations in the Land Surface Model (LSM) community mainly because of the inappropriate representations of crucial surface physical processes and properties (e.g., vegetation dynamics). Feedbacks of vegetation on land-atmosphere interactions have been recognized as influential in determining local weather and climate evolvements, however the calibrations and corrections of the vegetation parameterization in LSMs often remains model-specific and region-dependent. Here, we present two practice to improve the LST simulations in a state-of-art LSM (Noah-MP): (1) By updating Noah-MP 's vegetation parameters from the satellite observations (RS); and (2) By incorporating an empirical vegetation parameterization scheme (RL02). The overall results suggest domain-wide improvement of the simulated LST for both cases, with RMSEs upgraded by ~ 10% relative to the LST result from the default model configuration (CTL). Improved surface soil temperature simulations at three vegetation-diverse sites are also observed, showing negative relevance of the soil temperature improvements and the surface vegetation cover. Our study remarked the substantial impact of vegetation on surface energy exchanges, highlighting the potential to specify vegetation parameters on improving LST simulation. Limitations of vegetation - soil moisture coupling parameterizations in current LSMs will also be discussed in this study.

A soil moisture retrieval algorithm using L-band radar-only observations is applied to soil moisture active and passive (SMAP) radar observations. This algorithm is based on a nonlinear relationship between L-band backscatter and soil moisture, and any ancillary vegetation or roughness information is not needed. This algorithm is based on three limiting cases and end-members: smooth bare soil, rough bare soil and maximum vegetation covered soil. Those parameters is estimated through a iterative process. Three months global soil moisture is retrieved using SMAP radar observations and this algorithm. The accuracy of soil moisture result is validated by the ground network insitu observations and the SMAP standard radar and radiometer product. The soil moisture has similar spatial pattern with that of SMAP standard product at 9 km resolution. In the future, we can estimate the soil moisture at 3 km with SMAP radar data only.

In this article, we developed a hybrid method to estimate surface shortwave radiation (SSR) for the new-generation Himawari-8 geostationary satellite. This hybrid method combines the advantages of a deep neural network (DNN) with high speed and radiative transfer model (RTM) to achieve high accuracy: the RTM provides training data for the DNN under various cloud and aerosol conditions (including heavy aerosol loadings). Moreover, our hybrid method can simultaneously output the byproducts of photosynthetically active radiation (PAR), ultraviolet A (UVA), and Ultraviolet B (UVB), the direct and diffuse components at the surface, and the upward solar radiation at the top-of-atmosphere (TOA). The trained DNN was applied to the Himawari-8 satellite atmospheric products for 2016 and comprehensively validated using a total of 118 stations from four networks located in the full-disk regions of Himawari-8. The results showed an RMSE of 125.9 Wm-2for instantaneous SSR, 105.4 Wm-2for hourly SSR, 31.9 Wm-2for daily SSR, and respective mean bias error (MBE) scores of 8.1, 27.6, and 12.3 Wm-2. The hybrid method developed in this study performed well, achieving high accuracy and high speed, and it is capable of providing near-real-time SSR estimates for many applied energy fields.

Precipitation plays an important role in land surface processes. It is the vital driver of hydrological and land surface models and key bridge between land and atmosphere in the earth system. Although satellite remote sensing can to provide precipitation estimations at regional and global scales with high spatiotemporal resolutions, it still has great uncertainties at daily scale. On the other hand, soil moisture is the bridge between precipitation and runoff. The dynamic cycle of soil moisture is direct driven by precipitation and constrained by soil texture. In this study, we proposed a novel framework of improving Tropical Rainfall Measuring Mission Multi-satellite Precipitation Analysis (TMPA) data by including soil moisture and temperature data from ground observation. The results demonstrate that the rainfall estimation skills of TMPA are significantly improved when in situ soil moisture is used at Naqu, by reducing 44% root mean square errors (RMSE) and increasing the correlation coefficient (CC) from 0.27 to 0.71. When soil temperature data is taken into consideration, the False Alarm Ratio (FAR) is further reduced by 10%. It indicates that the proposed method has high potential to improve satellite-based rainfall products over the Tibetan Plateau.

In this study, we present a research to transfer the merits of SMAP (Soil Moisture Active Passive) to AMSR2 (Advanced Microwave Scanning Radiometer 2) with using machine learning method-artificial neural network. The surface soil moisture (SSM) products of SMAP were set as the reference data, while brightness temperature (TB) of various channels and the microwave vegetation index (MVI) obtained or derived from AMSR2 were input into an Artificial Neural Network (ANN). During training period (2015–2017), the ANN product (NNsm) can reproduce the SMAP SSM accurately, with a correlation coefficient (CC) of 0.74, Root Mean Square Error (RMSE) of 0.033 m3/m3, and Bias of −0.00008 m3/m3. It was found that machine learning method failed to provide reliable SSM over moderate vegetated areas where SMAP works well. With these trained networks, we developed a global soil moisture data set (named as NNsm) using AMSR2 TB from 2012 to 2018. Comparing to the in situ SM observations from all SCAN (Soil Climate Analysis Network) sites (named as SCANsm), NNsm has a good agreement with CC = 0.44, RMSE = 0.113 m3/m3and Bias = 0.030 m3/m3, which is much better than those of the AMSR2 SSM products from JAXA and LPRM.

The multi-model results from the fifth phase of the Coupled Model Intercomparison Project (CMIP5) showed obvious wet biases over the Tibetan Plateau (TP) during winter. However, sparse meteorological stations and the limited capacity for getting accurate snowfall may introduce dry biases into the observation and then exaggerate the overestimation of winter precipitation. In order to explore the spatiotemporal variations and reliability of winter precipitation over the TP, we compared five precipitation products, including: ERA-Interim, GLDAS, HAR, TRMM, and the observation provided by China Meteorological Administration (CMA), against a sublimation dataset which is severed as the minimum value of precipitation. The sublimation was estimated by the Kuzmin formula constrained with IMS snow cover product and land surface temperature. The intercomparison reveals that CMA has an obvious underestimation (precipitation is less than the one third of sublimation) over the Qiangtang Plateau where there is no observation site while no underestimation in East TP where dense stations available. For reanalysis and remote sensing data, HAR shows the smallest underestimation, while TRMM and GLDAS shows comparable underestimation and both are more apparent than ERA-interim. It implies that the observation data has considerable dry biases (~200%) in winter precipitation over the Western TP where more ground stations are needed to get a reliable precipitation observation.

Surface roughness plays an important role in retrieving soil moisture from microwave observations. Currently, microwave measurements at low-frequency bands (L, S, C, and X bands) provide global soil moisture data as well as surface roughness information. In this study, four global surface roughness products, including the one used in SMOS standard algorithm (named as SMOSHR), one derived from SMOS soil moisture product (named as SMOSD), the one used in SMAP algorithm (named as SMAPHR) and the one retrieved from AMSR-E soil moisture products (named as AMSRQH), were compared against the in situ measured surface roughness values at Pali network in the Tibetan Plateau. The roughness values of SMOSHR and SMAPHR show the same spatial pattern as the measured data at Pali network. Moreover, SMAPHR and SMOSHR have similar spatial pattern over the Tibetan Plateau. However, SMOSD shows different spatial pattern with the other two both at Pali network and over the Tibetan Plateau. Since similar surface roughness parameterization and values were adopted in SMOS and SMAP algorithm, their soil moisture products show consistent spatiotemporal pattern, which could benefit the merging of various soil moisture products across different sensors and platforms.

Precipitation plays an essential role in land surface processes, as a vital forcing variable of hydrology and ecosystem models. Satellite remote sensing is able to provide precipitation estimations at regional and global scales with high spatiotemporal resolutions. However, the accuracy of these products still need improvement especially at daily scale. In this study, we proposed a new method which can improve the rainfall product of the Global Precipitation Mission by using the soil moisture product of the Soil Moisture Active Passive mission over the Yarlung Zangbo River basin in the Tibetan Plateau. Compared to rain gauge observation, our method improve rainfall estimation at 97 of 108 GPM grids. Overall, the average correlation between improved GPM and in situ observation increased from 0.12 to 0.31, while the RMSE and RRMSE decreased by 1.16 and 0.42, respectively. It indicates that this approach can be used in a large scale to improve satellite-based rainfall products over the Tibetan Plateau.

Rainfall is one of the key drivers of hydrological and land surface models. Satellite remote sensing is able to provide regional and global rainfall estimates. However, there is still room to improve the accuracy of satellite rainfall products in areas where no/less rain gauges available, such as the Tibetan Plateau. In this study, the satellite rainfall product obtained from the Tropical Rain Measurement Mission (TRMM) and the follow-on Global Precipitation Mission (GPM) is further improved by using the in situ soil moisture observation obtained from a densely soil moisture observation network at Naqu and soil moisture retrieval provided by the Soil Moisture Active Passive (SMAP) mission over the Tibetan Plateau, respectively. In situ rainfall observations obtained from Ministry of Water Resource, which are not used in the calibration of satellite rainfall products, are used as validation data. The results demonstrate that the rainfall estimation skills of TRMM and GPM are significantly improved when in situ soil moisture is used at Naqu, by reducing 20∼30% root mean square errors (RMSE) and increasing the correlation coefficient (CC) from 0.1 to 0.7. When SMAP soil moisture is used, the RMSE is further reduced by 40∼50%, while CC is around 0.6. It indicates that the proposed method has high potential to improve satellite-based rainfall products over the Tibetan Plateau.

With the loss of the L-band radar, the NASA SMAP satellite lost the capability to directly provide high resolution global soil moisture data products after July 7th, 2015. However, the SMAP L-band radiometer has been successfully and continuously providing high quality coarse resolution observations with the best RFI mitigation since April 2015. These coarse resolution soil moisture observations could be downscaled to finer resolution using finer scale observations of soil moisture sensitive quantities from existing satellite sensors. In the past decade, several algorithms have been introduced to downscale passive microwave soil moisture observations. Most of these methods exploit the soil moisture information from optical sensing of land surface temperature and vegetation dynamics while others use active microwave (radar) observations. In this study, alternative algorithms are intercompared in order to find out the most reliable algorithm that could be implemented for routine or operational product generation. In this paper, coarse scale satellite data are from NASA SMAP radiometer and fine scale satellite data are backscatter from SMAP radar, land surface temperature (LST) and vegetation index from NOAA GOES, and AMSR2 Ka band observations for the warm seasons in 2015 and 2016. Results from three downscaling algorithms were analyzed. They were the NASA SMAP Active-Passive product algorithm, a simple LST regression algorithm, and a regression tree algorithm. Four sets of in situ soil moisture measurement data were collected and processed from Millbrook, NY, Walnut Gulch, AZ, Tibetan Plateau, China, and Yanco, Australia, respectively. Preliminary results of this inter-comparison study are reported.

A fast parameterization scheme named SUNFLUX is first used in this paper to estimate instantaneous surface solar radiation (SSR) based on products from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor onboard both Terra and Aqua platforms. The scheme mainly takes into account the absorption and scattering processes due to clouds, aerosols, and gas in the atmosphere. The estimated instantaneous SSR is evaluated against surface observations obtained from seven stations of the surface radiation budget network (SURFRAD), four stations in the North China Plain (NCP) and 40 stations of the baseline surface radiation network (BSRN). The statistical results for evaluation against these three data sets show that the relative root-mean-square error (RMSE) values of SUNFLUX are less than 15%, 16%, and 17%, respectively. Daily SSR is derived through temporal upscaling from the MODIS-based instantaneous SSR estimates, and is validated against surface observations. The relative RMSE values for daily SSR estimates are about 16% at the seven SURFRAD stations, four NCP stations, 40 BSRN stations, and 90 China Meteorological Administration (CMA) radiation stations. The accuracy of the scheme is generally higher than those of previous algorithms, and thus can be potentially applied on geostationary satellites for mapping high-resolution SSR data in the future.

Soil temperature is one of the essential variables governing the land atmosphere interaction. In this study, we proposed a statistical algorithm to retrieve the surface soil temperature from AMSR-E brightness temperature (TB) observations. The algorithm was developed based on the regression relationship between AMSR-E TB and corresponding in situ soil temperature observed at the Naqu network in the central Tibetan Plateau (CTP-Naqu). The algorithm was validated by application in six soil observation networks distributed globally. The results demonstrate that the algorithm has high accuracy and good transferability and can be applied globally. Moreover, the new algorithm uses passive microwave observations and is able to retrieve soil temperature all-day and all-weather, which compensates the deficiency of thermal infrared-based algorithms that are vulnerable to cloud contamination.

Land data assimilation system (LDAS) has been a powerful tool to optimally combine the advantages of microwave remote sensing and land surface model together and then improve the accuracy of land surface fluxes and status estimation. In this study, we developed a land data assimilation system, in which the water imbalance is constrained by using a multiply times assimilation approach. The first step is a standard variational assimilation operating at normal assimilation windows. And then the LDAS will run at an optimal assimilation window for several times to minimize the water imbalance accumulated in the first step. The LDAS is tested over CEOP Mongolia network. The microwave brightness temperature (TB) observation of AMSR-E are assimilated into the LDAS. Sensitivity test which considering the impacts of assimilation frequency and length of assimilation window are conducted. The results show that for soil moisture assimilation 48 hours window with 2 times assimilation could control the accumulated residual into the acceptable region.

Soil moisture is a key parameter in the land-atmosphere interactions over the Tibetan Plateau (TP), which plays an essential role in the Asian monsoon processes. Validation of satellite observed and/or modeled surface soil moisture is a particularly challenging work due to the scale issues. Additional challenge in this area is the harsh environment and heavy workload to establish a Soil Moisture and Temperature Measurement System (SMTMS) network. In this paper, we introduced two soil moisture and temperature monitoring networks that have dense measurements. Several applications with the data are presented, including current microwave products evaluation, new algorithm development and soil parameter optimization.

The land surface on the Tibetan Plateau (TP) experiences diurnal and seasonal freeze/thaw processes that play important roles in the regional water and energy exchanges, and passive microwave satellites provide opportunities to detect the soil state for this region. With the support of three soil moisture and temperature networks on the TP, a dual-index microwave algorithm with Advanced Microwave Scanning Radiometer-Earth Observing System data is developed for the detection of soil surface freeze/thaw state. One index is the standard deviation index (SDI) of brightness temperature (TB), which is defined as the standard deviation of horizontally polarized brightness temperatures at 6.9, 10.7, 18.7, 23.8, 36.5, and 89.0 GHz. It is the major index and is used to reflect the reduction of liquid water content after soils get frozen. The other index is the 36.5-GHz vertically polarized brightness temperature$(\hbox{TB}_{36.5}^{\rm V})$, which is linearly correlated with ground temperature. The threshold values of the two indices (SDI and$\hbox{TB}_{36.5}^{\rm V}$) are determined with one grid from the network located in a semiarid climate, and the algorithm is validated with other grids from the same network. Further validations are conducted based on the other two networks located in different climates (semihumid and arid, respectively). Results show that the classification accuracy using this algorithm is more than 90% for the semihumid and semiarid regions, and misclassifications mainly occur at the transition period between unfrozen and frozen seasons. Nevertheless, the algorithm has limited capability in identifying the soil surface freeze/thaw state in the arid region because the microwave signals can penetrate deep dry soils and thus embody the bulk information beyond the surface layer.

Soil moisture is an important variable in the Earth system. Reliable estimation of soil moisture, especially in regional and global scale, is essential for climatic, meteorological and hydrological researches. This study presents a revised soil moisture retrieval algorithm of ASMR2 that aims to improve soil moisture estimate in dry regions through including the vertical profile effects of soil temperature into the radiative transfer model. In situ observations of soil moisture and temperature profile, measured during HIWater Experiment intensive observation period, were used to formulate profile functions. By comparing the results of new algorithm against ground observations, the feasibility and capability of the revised algorithm were verified. The application of this algorithm in wide regions also demonstrated that our method could alleviate the overestimation-in-desert problem of original JAXA AMSR2 soil moisture products.

It is widely recognized that remote sensing data a very valuable source of information for the modeling of land-atmosphere interactions. During the last of couple decades, remote sensing data were used mainly to define the initial status of land surface models, to classify the land cover and land use type, and to correct the estimation of state variables in a data assimilation system. One relatively unexplored issue consists of the optimization of land surface parameters, such as, for example, soil texture, soil porosity, through remote sensing. This is due to the lack of a relationship which connecting remote sensing data with land surface parameters. The objective of this paper is to develop a systemic method to retrieve a number of land surface parameters through a combination of microwave remote sensing, radiative transfer modeling, land surface modeling and multi-objective optimizing. The microwave brightness temperature observations are used as the calibration references. The land surface model and radiative transfer model are used to determine the relationship between land surface parameters and the brightness temperature. The method is validated through a field experiment, in which a ground-based microwave radiometer is used to provide brightness temperature observations in a controllable footprint. Grand truth of soil parameters is also measured through intensive in situ samplings. The comparison of optimized parameters with the in situ observed ones indicates that our method has high potential to calibrate land surface parameters.

Regional soil moisture can be estimated by assimilating satellite microwave brightness temperature into a land surface model (LSM). This paper explores how to improve soil moisture estimation based on sensitivity analysis when assimilating Advanced Microwave Scanning Radiometer for the Earth Observing System brightness temperatures. By assimilating a lower and a higher frequency combination, the land data assimilation system (LDAS) used in this paper estimates first model parameters in a calibration pass and then estimates soil moisture in an assimilation pass. The ground truth of soil moisture was collected at a soil moisture network deployed in a semiarid area of Mongolia. Analyzed are the effects of assimilating different polarizations, frequencies, and satellite overpass times on the accuracy of the estimated soil moisture. The results indicate that assimilating the horizontal polarization signal underestimates soil moisture and assimilating the daytime signal overestimates soil moisture. The former is due to improper parameter estimation perhaps caused by high sensitivity of the horizontal polarization to land surface heterogeneity, and the latter is due to the effective soil temperature for microwave emission in the daytime being close to the one at a soil depth of several centimeters but not to the surface skin temperature simulated in the LSM. Therefore, assimilating the nighttime vertical polarizations in the LDAS is recommended. A further analysis shows that assimilating different frequency combinations produces different soil moisture estimates, and none is always superior to the others, because different frequency signals may be contaminated by varying clouds and/or water vapor with different degrees. Thus, an ensemble estimation based on frequency combinations was proposed to filter off, to some extent, the stochastic frequency-dependent biases. The ensemble estimation performs more robust when driven by different forcing data.

Land surface heterogeneities are important for accurate estimation of land–atmosphere interactions and their feedbacks on water and energy budgets. To physically introduce existing land surface heterogeneities into a mesoscale model, a land data assimilation system was coupled with a mesoscale model (LDAS-A) to assimilate low-frequency satellite microwave observations for soil moisture and the combined system was applied in the Tibetan Plateau. Though the assimilated soil moisture distribution showed high correlation with Advanced Microwave Scanning Radiometer on the Earth Observing System soil moisture retrievals, the assimilated land surface conditions suffered substantial errors and drifts owing to predicted model forcings (i.e., solar radiation and rainfall). To overcome this operational pitfall, the Coupled Land and Atmosphere Data Assimilation System (CALDAS) was developed by coupling the LDAS-A with a cloud microphysics data assimilation. CALDAS assimilated lower frequency microwave data to improve representation of land surface conditions, and merged them with higher frequency microwave data to improve the representation of atmospheric conditions over land surfaces. The simulation results showed that CALDAS effectively assimilated atmospheric information contained in higher frequency microwave data and significantly improved correlation of cloud distribution compared with satellite observation. CALDAS also improved biases in cloud conditions and associated rainfall events, which contaminated land surface conditions in LDAS-A. Improvements in predicted clouds resulted in better land surface model forcings (i.e., solar radiation and rainfall), which maintained assimilated surface conditions in accordance with observed conditions during the model forecast. Improvements in both atmospheric forcings and land surface conditions enhanced land–atmosphere interactions in the CALDAS model, as confirmed by radiosonde observations.

The land-atmosphere interaction in the Tibetan Plateau plays an important role in the Asian summer monsoon and the global energy and water cycle. This study presents a method to improve the land surface water and energy fluxes simulation by using a land data assimilation system (LDAS), which merging microwave remote sensing data and GCM output into a land surface model. NCEP reanalysis data is used as the background field and also as the meteorological forcing for the land surface model. Two experiments were designed as by driving LDAS-UT with two sets of atmospheric forcing data, (1) with in situ observed forcing data and (2) with NCEP reanalysis data at Gaize and Naqu sites. Results show that LDAS is able to estimate land surface soil moisture and energy fluxes accurately. The RMSE of soil moisture simulation is around 0.03–0.05 and RMSE of net radiation simulation is around 30W/M2. This study reveals the potential for using satellite remote sensing data to improve land surface fluxes estimation.

This paper reports an application of an Land Data Assimilation System developed in the University of Tokyo (LDASUT) on the Gaize PBL site at the northwest of Tibet Plateau, for the period from July to August, 2007. The objectives of this study are: (1) to validate LDASUT in bare soil field using in-situ observation, (2) to check the feasibility to estimate areal land surface variables reliably with using LDASUT driven by GCM output data. For the system validation, LDASUT was first driven by in-situ observed micrometeorological data, and simulated energy fluxes were compared to hourly direct measurements; simulated soil moisture content was compared to the in-situ soil moisture observation at the depth of 4 cm. The results show that LDAS can generally simulate those variables well and thus the capability of LDAS is validated. In order to check the possibility of applying LDAS globally and simulating surface energy and water budget worldwide, Japan Meteorology Agency (JMA) Model Output Local Time Series (MOLTS) data were used as the driven data of LDAS. Performance of LDAS was not so good when it was driven by the JMA MOLTS data. This result demonstrated that there were systemic biases lied in JMA MOLTS data in the study region and thus it can not directly apply to LSM or LDAS.

Reliable prediction of precipitation by Numerical Weather Prediction (NWP) models depends on the appropriate representation of cloud microphysical processes and accurate initial conditions of observations of atmospheric variables. Therefore, 1D Variational (1D-Var) Ice Cloud Microphysics Data Assimilation System (IMDAS) is developed for retrieving reasonable cloud distributions to improve the predictability of NWP models. The general framework of IMDAS includes the Lin ice cloud microphysics scheme as a model operator, a 4-stream fast microwave radiative transfer model (RTM) in the atmosphere as an observation operator, and a global minimization method known as Shuffled Complex Evolution (SCE). The IMDAS assimilates the satellite microwave radiometer data set of Advanced Microwave Scanning Radiometer (AMSR-E) and retrieves integrated water vapor (IWV) and integrated cloud liquid water content (ICLWC). This new method successfully introduces the heterogeneity into the initial state of the atmosphere, and the modeled microwave brightness temperatures agree well with observations of Wakasa Bay Experiment 2003 in Japan. It has improved the performance of cloud microphysics scheme significantly by the intrusion of heterogeneity into the external Global Reanalysis (GANAL) data, which may improve atmospheric initial conditions.

Reliable prediction of precipitation by numerical weather prediction (NWP) models depends on the appropriate representation of cloud microphysical processes and accurate initial conditions of observations of atmospheric variables. Therefore, to retrieve reasonable cloud distributions, a 1-D variational Ice Cloud Microphysics Data Assimilation System (IMDAS) is developed to improve the predictability of NWP models. The general framework of IMDAS includes the Lin ice cloud microphysics scheme as a model operator, a four-stream fast microwave radiative transfer model in the atmosphere as an observation operator, and a global minimization method that is known as the shuffled complex evolution. IMDAS assimilates the satellite microwave radiometer data set of the Advanced Microwave Scanning Radiometer for the Earth Observing System (AMSR-E) and retrieves integrated water vapor and integrated cloud liquid water content. This new method successfully introduces heterogeneity into the initial state of the atmosphere, and the modeled microwave brightness temperatures agree well with the observations of the Wakasa Bay Experiment 2003 in Japan. It has significantly improved the performance of the cloud microphysics scheme by the intrusion of heterogeneity into the external global reanalysis data, which resultantly improved atmospheric initial conditions.