Land Data Assimilation at the JCSDA: Plans and Progress

The land surface determines the lower boundary conditions of the atmosphere, and interacts with weather and climate through regulation of energy and mass fluxes over a range of temporal and spatial scales. Much of the complexity in these interactions stems from heterogeneity in land use, soil type, vegetation type, soil moisture and the presence of snowpack. In NWP and climate models it is the role of coupled land surface models to accurately represent this heterogeneity, and simulate the exchanges of energy, water and carbon through the soil-plant-atmosphere continuum. Given these interactions with the atmosphere, land surface initialization is of critical importance. For example, soil moisture conditions have been shown to impact forecast skill over both short and seasonal timescales, and snow cover with its high albedo and low thermal conductivity has a large impact on the atmospheric forecast accuracy.

At the Joint Center for Satellite Data Assimilation (JCSDA) we are working with partners to improve capacity to assimilate snow observations into the Noah-MP land surface model (LSM), with the overarching goal of improving skill within both the Global Forecast System and the National Water Model that both rely on this LSM. Here we describe initial progress in developing an interface between this LSM and the Joint Effort in Data Assimilation Integration (JEDI) framework, and ingest of a first suite of targeted snow observations. We describe the contrasting requirements of NWP and hydrological prediction, and our approach to mitigate the potentially negative impact of snow update increment on runoff simulation. This represents a good example of the challenges we will need to address when using JEDI for land DA when considering coupled systems in data assimilation, and for moving towards the goal of holistic Earth system approaches at the JCSDA.