A Patched-Based Global Land Data Assimilation System Collaboratory
A CREW proposal submitted to NASA-ROSES-2012 A13: Modeling, Analysis, and Prediction Program (NNH12ZDA001N-MAP), $924,577, 3 mo/yr, 01/01/2013 – 12/31/2016.
The land-surface component of the hydrological cycle is fundamental to the overall functioning of the atmospheric and climate processes. The near-real time North-American Land Data Assimilation System (N-LDAS) was established in 1998 to create observationally-constrained uncoupled land-surface model predictions for use in weather and climate forecast initialization and for use in drought and flood applications. Since N-LDAS was established, regional to continental scale LDAS projects have emerged on virtually every continent to provide customized land surface conditions for a wide array of research and applications. NASA established the Global LDAS (G-LDAS) in 2002, and it continues to provide course-resolution reanalysis products via the NASA-HDISC data center.
A logical evolution in LDAS capabilities would be to enable a seamless global LDAS research environment, where LDAS teams from across the globe could share and overlay their regional LDAS results, resulting in a global near-real time “patched” LDAS. This global patched LDAS would provide global land conditions, and also provide the best available local land surface information; it would be invaluable for a wide array of climate and weather model initialization and validation studies, as well as end-use applications.
Therefore, I propose to develop a Patched-based Global Land Data Assimilation (PG-LDAS) Collaboratory, which facilitates international partnerships to share and overlay regional near-real time regional LDASs resulting in a global, but locally relevant assessment of land surface conditions. The patching of local LDAS information into a global mosaic enables two GLDAS modes: (1) patching of the regional land forcing data into a global mosaic used to drive a suite of global land model simulations, and (2) patching of the regional LDAS output data into a global mosaic of land conditions.
This work would proceed along four distinct themes: (a) enabling global LDAS communication, planning, and data sharing using state-of-the-art information technology such as the NASA Earth Exchange, Google group collaboration tools, and web 2.0 collaboratory tools; (b) a near-real time virtual environment for regional LDAS forcing overlay, patching and scaling to drive a globally-consistent multi-model PG-LDAS simulation; (c) a near-real time virtual environment for regional LDAS land condition overlay and patching; and (d) compilation and sharing of critical LDAS performance diagnostics principally derived from NASA satellite data products.
This project innovatively builds on multiple NASA modeling and observational capabilities and responds directly to the MAP-2012 solicitation. Through the proposed extensive sharing and intercomparison of land surface data, the resulting PG-LDAS products will characterize and/or help reduce uncertainties in the land models and products; extend the range of model or product validity by using new components; and enable independent community validation and characterization of core MAP elements leading to improvement of the models and satellite products.