The world’s oceans are the dominant source of moisture in the global hydrological cycle. Simplified representations of oceanic evaporation (referred to as bulk parameterizations) are used with near-surface meteorological data, numerical models, and satellite data for estimating air-sea moisture transfer, oceanic heat budgets, or to constrain models. In this paper we describe the present state-of-the-art in bulk models of evaporation in the form of a new version (3.0) of the COARE algorithm. The COARE algorithm uses a combination of physically-based submodels of interfacial processes to relate the evaporation to bulk parameters. The new algorithm is based on both previously published field results and 2777 one-hour covariance and inertial-dissipation moisture flux measurements in the ETL inventory. To test it, we added 4439 new values from field experiments between 1997 and 1999, which now dominate the database, especially in the wind speed regime beyond 10 ms-1 where the number of observations increased from 67 to about 800. After applying various quality controls, the database was used to evaluate the algorithm in several ways. The average (mean and median) model results agreed with the measurements to within about 5% for moisture from 0 to 20 ms-1. In the paper we will discuss measurement issues, how these results compare with classic results from other field programs and models, and prospects for extending beyond 20 ms-1.