Python implementation of the COARE 3.5 Bulk Air-Sea Flux algorithm

Python implementation of the COARE 3.5 algorithm developed at NOAA Physical Sciences Lab. The COARE model followed from the international TOGA-COARE field program which took place in the western Pacific warm pool over 4 months from November 1992 to February 1993 (Fairall et al. 1996a, 1996b and 1997).  The algorithm is intended to provide estimates of momentum, sensible heat, and latent heat fluxes using inputs of bull atmospheric variables (wind speed, SST, air temperature, air humidity).  The algorithm contains subroutines to deal with near-surface gradients of temperature in the ocean.  Version 2.5 was published in 1996 and a major update, version 3.0 was published in 2003.  This update was based on new observations at higher wind speeds (10 to 20 m/s).  Version 3.5 was released in 2013 following the publication of Edson et al. 2013, which made adjustments to the wind speed dependence of the Charnock parameter based on a large data base of direct covariance stress observations (principally from a buoy).  This led to an increase in stress for wind speeds greater than about 18 m/s.  The roughness Reynolds number formulation of the scalar roughness length was tuned slightly to give the same values of Ch and Ce as version 3.0.  The diurnal warm layer model was structured as a separate routine instead of embedded in a driver program.  COARE 3.5 was based on Edson’s buoy data (Edson et al. 2013) and was compared to a large data base (a total of 16,000 hours of observations) combining observations from NOAA, WHOI, and U. Miami (Fairall et al. 2011).