A Global Water Budget Model

Franklin R. Robertson, NASA/MSFC/ES42
Eugene W. McCaul, Jr., IGCRE/USRA


Our current knowledge of the global distributions of water vapor, cloud condensate and precipitation are still inadequate for the purposes of many climate research efforts. In order to improve our ability to diagnose these fields more accurately, scientists at the Global Hydrology and Climate Center are developing a water budget diagnostic model designed to estimate the three-dimensional fields of vapor, cloud condensate and precipitation around the globe as a function of time.

The model takes as input the wind, temperature and initial moisture fields from a global four-dimensional gridded dataset, then integrates forward in time the equations for water vapor, cloud condensate and precipitation only, using interpolated values of the other fields in the computations. To constrain directly the evolution of the simulated water vapor field (and of cloud condensate and precipitation indirectly), data from SSMI precipitable water measurements over the oceans are also used to constrain the evolving precipitable water on an ongoing basis. The local moisture profiles are rescaled to give values of precipitable water equal to those observed by the satellite, whenever and wherever the satellite data are available.

Condensate is modelled on both the resolvable grid and on the smaller sub-grid convective scales. A mass-flux convective parameterization is used to compute convective mass fluxes, cloudiness, precipitation and detrainment of vapor and condensate to the larger scales. Large-scale condensate is computed from grid-scale water vapor and vertical motions, with inputs from convective detrainment. Bulk microphysics determine the properties of the large-scale clouds and their precipitation. Ice and liquid condensate are discriminated through use of a ramp function that spans the temperature range from -40 C to 0 C. Surface fluxes may also be introduced into the model, but a detailed treatment of boundary layer processes is still being developed. The constrained model prognoses only the moisture variables; there is no feedback to the input temperature and kinematic data.

The analysis is designed to output more than 70 global horizontal fields every 12 hours of simulated time, along with vertical profiles and zonal mean cross-sections of numerous parameters. Of particular interest are studies of the December 1992-February 1993 time period during TOGA-COARE. Below are samples of model horizontal output: total diagnosed precipitation (mm); mean convective mass fluxes (microbars/sec) in the 500-200 mb layer; relative humidity (%) at 500 mb; column-integrated cloud and precipitation ice (mm); and " convective available potential energy (J/kg) for the month of December 1992. Also given below are samples of zonal mean cross sections of ice mixing ratio (g/kg) and total condensate mixing ratio (g/kg) for the same month. The model is able to reproduce the major structural features of water vapor, high cloudiness and rainfall seen in many climatologies and observational studies. Analysis of these simulated fields vis-a-vis remotely-sensed fields of precipitation and cloudiness (e.g. MSU and ISCCP)


Global Hydrology and Climate Center


Responsible Official: Dr. Timothy L. Miller (tim.miller@msfc.nasa.gov)
Page Curator: Paul J. Meyer (paul.meyer@msfc.nasa.gov)


Last Updated: July 6, 1995