Open-Source Data Assimilation for Land Models and Multiscale Observations.

Abstract:

The Data Assimilation Research Testbed (DART) has been coupled to several
land models including the Community Land Model (CLM),
the Community Noah Land Surface Model (Noah LSM), WRF-Hydro, and
the CSIRO Atmosphere Biosphere Land Exchange (CABLE) model. Many types
of observational data ranging from in-situ soil moisture probes to
tower-based fluxes to satellite estimates of moisture have been successfully
assimilated to produce model-based estimates of quantities that are more
consistent with the information content of the observations and yet have
the desirable spatio-temporal attributes of the gridded model output.
Examples of assimilation research with each of the models will be shown.

One of the challenges for land data assimilation systems is the collection
and integration of the observational data given multiple datastreams and
collection agencies. The challenges and considerations of ingesting and
using a wide variety of data in many different formats will be discussed
with a view of what is needed for a community resource.