Despite the importance of the coupling between vegetation dynamics and root‐zone soil moisture in land‐atmosphere interactions, there is no land data assimilation system (LDAS) that currently addresses this issue, limiting the capacity to positively impact weather and seasonal forecasting. We develop a new LDAS that can improve the skill of an ecohydrological model to simulate simultaneously surface soil moisture, root‐zone soil moisture, and vegetation dynamics by assimilating passive microwave observations that are sensitive to both surface soil moisture and terrestrial biomass. This LDAS first calibrates both hydrological and ecological parameters of a land surface model, which explicitly simulates vegetation growth and senescence. Then, it adjusts the model states of soil moisture and leaf area index (LAI) sequentially using a genetic particle filter. We can adjust the subsurface soil moisture, which is not observed directly by satellites, because we simulate the interactions between vegetation dynamics and subsurface water dynamics. From a point‐scale evaluation, we succeed in improving the performance of our land surface model and generate ensembles of the model state whose distribution reflects the combined information in the land surface model and satellite observations. We show that the adjustment of the subsurface soil moisture significantly improves the capacity to simulate vegetation dynamics in seasonal forecast timescales. This LDAS can contribute to the generation of ensemble initial conditions of surface and subsurface soil moisture and LAI for a probabilistic framework of weather and seasonal forecasting.

Improving the predictability of sudden local severe weather is a grand challenge for numerical weather prediction. Recently, the capability of geostationary satellites to observe infrared radiances has been significantly improved, and it is expected that the “Big Data” from the new generation geostationary satellites could contribute to improving convective predictability. We examined the potential impacts of assimilating frequent infrared observations from a new generation geostationary satellite, Himawari‐8, on convective predictability. We implemented the real‐data experiment in which Himawari‐8 all‐sky moisture‐sensitive infrared radiances of band 8 (6.2 μm) and band 10 (7.3 μm) were assimilated into the high‐resolution (2 km) limited area model, Japan Meteorological Agency's Non‐Hydrostatic Model, every 10 min by the Local Ensemble Transform Kalman Filter. The frequent infrared observations from Himawari‐8 improve the analysis and forecast of isolated convective cells and sudden local severe rainfall induced by weak large‐scale forcing. The results imply that satellite data assimilation can contribute to better forecasting severe weather events in smaller spatiotemporal scales than the previous studies.

To improve the skill of reproducing land‐atmosphere interactions in weather, seasonal, and climate prediction systems, it is necessary to simulate correctly and simultaneously the surface soil moisture (SSM) and terrestrial biomass in land surface models. Despite the performance of hydrological and ecosystem models depends highly on parameter calibration, a method for parameter estimation in ungauged areas has yet to be established. We develop an autocalibration system that can simultaneously estimate both hydrological and ecological parameters by assimilating a microwave signal that is sensitive to both SSM and terrestrial biomass. This system comprises a hydrological model that has a physically based, sophisticated soil hydrology scheme, a dynamic vegetation model that can estimate vegetation growth and senescence, and a radiative transfer model that can convert land surface condition into brightness temperatures in the microwave region. By assimilating microwave signals from the Advanced Microwave Scanning Radiometer for Earth Observing System, the system simultaneously optimizes the parameters of these models. We test this approach at three in situ observation sites under different hydroclimatic conditions. Estimated SSM exhibits good agreement with ground‐based in situ observed SSM, and estimated leaf area index (LAI) is also improved by the optimization, compared with satellite‐observed LAI. The root‐mean‐square error of SSM and LAI at all sites, estimated by the model with optimized parameters, is much less than that estimated by the model with default parameters. Using microwave satellite brightness temperature data sets, this system offers the potential to calibrate parameters of both hydrological and ecosystem models globally.

To improve the efficiency of flood early warning systems (FEWS), it is important to understand the interactions between natural and social systems. The high level of trust in authorities and experts is necessary to improve the likeliness of individuals to take preparedness actions responding to warnings. Despite many efforts to develop the dynamic model of human and water in socio-hydrology, no socio-hydrological models explicitly simulate social collective trust in FEWS. Here, we develop the stylized model to simulate the interactions of flood, social collective memory, social collective trust in FEWS, and preparedness actions responding to warnings by extending the existing socio-hydrological model. We realistically simulate the cry wolf effect in which many false alarms undermine the credibility of the early warning systems and make it difficult to induce preparedness actions. We found that (1) considering the dynamics of social collective trust in FEWS is more important in the technological society with infrequent flood events than in the green society with frequent flood events; and (2) as the natural scientific skill to predict flood events is improved, the efficiency of FEWS gets more sensitive to the behavior of social collective trust, so that forecasters need to determine their warning threshold by considering the social aspects.

The performance of land surface models (LSMs) significantly affects the understanding of atmospheric and related processes. Many of the LSMs' soil and vegetation parameters were highly uncertain so that it is crucially important to efficiently optimize them. Here I present a globally applicable and computationally efficient method for parameter optimization and uncertainty assessment of the LSM by combining Markov Chain Monte Carlo (MCMC) with machine learning. First, I performed the long‐term (decadal‐scale) ensemble simulation of the LSM, in which each ensemble member has different parameters' values, and calculated the gap between simulation and observation, or the cost function, for each ensemble member. Second, I developed the statistical machine learning‐based surrogate model, which is computationally cheap but accurately mimics the relationship between parameters and the cost function, by applying the Gaussian process regression to learn the model simulation. Third, I applied MCMC by repeatedly evaluating the surrogate model to get the posterior probabilistic distribution of parameters. Using satellite passive microwave brightness temperature observations, both synthetic and real‐data experiments in the Sahel region of west Africa were performed to optimize unknown soil and vegetation parameters of the LSM. The primary findings are (1) the proposed method is 50,000 times as fast as the direct application of MCMC to the full LSM; (2) the skill of the LSM to simulate both soil moisture and vegetation dynamics can be improved; and (3) I successfully quantify the characteristics of equifinality by obtaining the full nonparametric probabilistic distribution of parameters.

Despite the importance of the interaction between soil moisture and vegetation dynamics to understand the complex nature of drought, few land reanalyses explicitly simulate vegetation growth and senescence. In this study, I provide a new land reanalysis which explicitly simulates the interaction between sub-surface soil moisture and vegetation dynamics by the sequential assimilation of satellite microwave brightness temperature observations into a land surface model (LSM). Assimilating satellite microwave brightness temperature observations improves the skill of a LSM to simultaneously simulate soil moisture and the seasonal cycle of leaf area index (LAI). By analyzing soil moisture and LAI simulated by this new land reanalysis, I identify the drought events which significantly damage LAI on the climatological day-of-year of the LAI’s seasonal peak and quantify drought propagation from soil moisture to LAI in the global snow-free region. On average, soil moisture in the shallow soil layers (0–0.45 m) quickly recovers from the drought condition before the climatological day-of-year of the LAI’s seasonal peak while soil moisture in the deeper soil layer (1.05–2.05 m) and LAI recover from the drought condition approximately 100 days after the climatological day-of-year of the LAI’s seasonal peak.