What is snow?
To atmospheric scientists, snow (i.e. falling snow) is a type of precipitation composed of ice crystals. Ice crystals forms when water droplets in cold temperatures below 0 C freezes. Many different types of tiny particles in the air such as desert dust, soot, pollen, and volcanic ashes can make this freezing process easier and quicker and this mechanism called ‘ice nucleation’. Once the falling snows reach to the Earth’s surfaces, they accumulate into snowpack or it ultimately melt and recharge ground water depending on the climate of the place where they land.
Why is snow important?
For the people working on weather forecasts, predicting snow –how much snow? Wet snow or dry snow? - is one of most difficult tasks. For the scientists working on Earth’s climate system. Knowing spatial and temporal distributions of snow are critical to examine Earth’ energy balance because Earth surface covered with snow reflect solar radiation differently than other Earth’s surface types.
What is my research about?
Numerical Weather Prediction (NWP) models, which are composed of complex computer programs for the equations representing atmospheric dynamics and physics, are known to be the tool in forecasting from the day-to-day weather changes to simulations and studies of the dynamics of the climate system to projections of future climate. One of NWP models is the Goddard Earth Observing System (GEOS) systems that are being developed in the Global Modeling Assimilation Office (GMAO) to support NASA's earth science research in data analysis, observing system modeling and design, climate and weather prediction, and basic research. The GEOS system consists of a group of model components such as Atmospheric General Circulation Model (AGCM), Ocean General Circulation Model (OGCM), Chemistry-Climate Model (CCM), and Chemistry Transport Model (CTM). Among these, the AGCM is the predictive model components for the atmosphere and land. Catchment Land Model, which is a part of GEOS AGCM, includes explicit treatment of the spatial variation within each hydrological catchment of the soil water and water table depth, as well as its effect on runoff and evaporation. It also includes a multi-layer global snow model.
The GEOS systems and computer resources evolved at a rapid pace for last a couple of decades and contributed to great improvements in forecasting atmospheric circulations which often demonstrated with matrice like 500hPa anomaly correlations in forecasts and analyses. However, skill to predict precipitation intensity and spatial distributions are still challenging because it requires consideration of complicated microphysical processes in moisture physics and its interaction with dynamics through latent heat in the NWP system. In addition, the NWP skills for precipitation are directly linked to the skills to analyze and forecast surface variables such as snow depth, soil moisture, and skin temperature. Therefore, it is expected that improvement of the GEOS precipitation analyses by assimilating all-sky satellite radiance data in GEOS will complement GEOS land surface analyses.
The objectives of my research are (1) to develop framework to utilize satellite data through data assimilation techniques to provide better initial conditions for the GEOS system so improve NWP skills to predict falling snow intensity and spatial and temporal distributions, (2) to improve NWP skills to predict land surface variables such as snow depth, soil moistures, and land surface temperature and (3) to produce global atmospheric and surface analyses products including falling snow and snow on the ground that are coupled through physical parameterizations considered in the GEOS.