Numerical models can produce gigabytes of data each day. Describing, locating and viewing that data becomes as large a problem as the initial creation. A model which can not be located and read might just as well not be created in the first place. Methodologies to solve these problems are being built in all fields which work with large data sets. Metadata is the information about files which describes the contents of the files and allows other resources to discover those files and work with them. The Coast Survey Development Lab (CSDL) is working with the Climate and Forecast Metadata Conventions Group to develop methods which are well suited to numerical modelers' needs. CSDL has created its own subset of the CF conventions to describe standardized output files for the CSDL models.  These standards describe NetCDF files which are simplified from the original data files from the models. They might be somewhat different from the original model results, but the simplification makes them much more useful. For instance, CSDL NetCDF files specify water velocities in true Eastward and Northward components, while many models perform the calculation of velocities on curvilinear staggered grids which require considerable expertise to produce output in normal geographical units. The adherence to standards for model outputs allows the use of web resources such as a Live Action Server (LAS) for hydrodynamic output using Open-source Project for a Network Data Access Protocol (OPeNDAP) data distribution tools.   The NOAA CO-OPS is serving up these standardized output files with OPeNDAP.  These tools allow automatic computer access to ever changing data archives and allow products such as the Chesapeake Bay Oyster Larvae Tracker (CBOLT) or the General NOAA Oil Modeling Environment (GNOME) to access real time model results to aid in oilspill response or Search and Rescue efforts.