LDAS Logo
Overview And Description
NLDAS-E PROJECT SUMMARY

Interactions between land and atmosphere greatly impact short term weather and long term climate on a variety of spatial and temporal scales. Storages of water and energy are the two fundamental agents that influence nearly every facet of the land, atmosphere and their exchanges. The partitioning of energy and water into latent and sensible heat significantly influences precipitation events, local wind patterns, and even global storm development. Unfortunately, biases in these stores can develop in coupled modeling systems due to one or more of the following:

  • forcing errors,
  • simplifications in Numerical Weather Prediction (NWP) model,
  • simplifications in Land Surface Model (LSM) physics, and
  • misspecification of model parameters.

    Biases can continue to grow in the closed, internally cycled system of such coupled models, negatively affecting forecast accuracy. Land Data Assimilation Systems (LDAS) features more accurate land surface states because observation-derived datasets drive the system. Our overarching hypothesis is that NWP forecast skill over long and short forecast time periods will increase when such models are properly initialized with LDAS land surface conditions. It is our hope that by supplying and employing improved land surface conditions (LDAS) to atmospheric models, errors attributed to land surface biases (that accumulate over time and space) will be reduced substantially.

    Significant progress has been made by collaborative research teams over the last few years to develop operational continental (North American and European) and global LDAS frameworks. Through these projects we have learned that the transfer of land information between models of different types or resolutions is not trivial, and may in fact not be possible. Therefore, to properly test the initialization of a coupled model using LDAS, we must make the uncoupled system as similar as possible to the coupled system.

    The current research will leverage the research already completed as part of the NLDAS and the GLDAS projects to construct an enhanced NLDAS system on the Arakawa-E grid (NLDAS-E) compatible with the NCEP mesoscale coupled Eta model. NLDAS-E will have a resolution of 12km, with North America entirely covered within the domain. This new NLDAS system will feature the following:

  • implement multiple LSMs,
  • assimilate multiple land surface quantities,
  • supply the Workstation Eta model with accurate, unbiased, initial land surface conditions,
  • improve our understanding of the interaction between the land surface and the atmosphere from short-term to seasonal time scales,
  • develop accurate, high resolution land surface data sets useful for future research and applications,
  • improve initialization of the Workstation Eta model in a way that may be adapted to additional NWP models in the future,
  • improve forecast accuracy, and
  • continue NASA GSFC's efforts in the collaborative NLDAS project.

    NLDAS-E

    Figure 1: NLDAS-E system overview.
    An accurate depiction of the land surface states will be obtained by running LSMs uncoupled and offline.
    Once the uncoupled LSMs have been properly spun up, LDAS states will be used to initialize the coupled Workstation ETA model to produce 84 hour forecasts.
    NASA webpage LDAS webpage LDAS Web Page 
    Developed and Maintained at NASA/GSFC    		 Matthew Rodell: Matthew.Rodell@gsfc.nasa.gov 
    Brian Cosgrove: Brian.Cosgrove@gsfc.nasa.gov    		 GSFC webpage