Greenland Ice Sheet Surface Mass Balance Model Data

 

README version: September 28, 2004

---------------------------------------------

­­CONTACT: Jason E. Box, Byrd Polar Research Center, email: box.11@osu.edu

---------------------------------------------

­DATA CITATION:  Box, J.E., D. H. Bromwich, L-S Bai, 2004: Greenland ice

sheet surface mass balance for 1991-2000: application of Polar MM5


mesoscale model and in-situ data, J. Geophys. Res., Vol. 109, No. D16, D16105, 10.1029/2003JD004451.

---------------------------------------------

information from: http://polarmet.mps.ohio-state.edu/jbox/data/

 

Greenland Ice Sheet Surface Mass Balance Data

 

The Polar MM5 regional climate model was run over Greenland in a series of 36-hour forecasts spanning 1991–2000 (Box et al. 2004). The model was initialized and constrained by available observations, e.g. satellite-derived temperature and water vapor profiles, sea ice extent, and weather balloon soundings. We analyzed 24-km output over the Greenland ice sheet to evaluate spatial and temporal variability of the surface mass balance and its subcomponents, i.e. precipitation, surface and blowing snow water vapor fluxes, and meltwater production/runoff/retention. The model output was compared with 3 years of independent Greenland Climate Network (GC-Net) automatic weather station (AWS) data from 17 sites, i.e. Steffen et al. (1996); Steffen and Box (2001) and other glacier survey data (e.g. Greuell et al. 2001) to identify model biases. Using the in situ data, we derived simple corrections for biases in melt energy and in water vapor fluxes. The simulated accumulation rate was in agreement with AWS and snow pit observations. Estimates for runoff and the surface mass balance distribution over the ice sheet are produced using modeled meltwater production and the Pfeffer et al. (1991) meltwater retention scheme.

 

Here, we make available annually-resolved grids of accumulation rate, surface mass balance (net accumulation), freshwater discharge (runoff) following this link. Data are provided on a 24 km horizontal resolution grid with 55 E-W direction and 101 in the N-S direction, i.e. 55 x 101 grid. Both ASCII and binary data formats are provided. Latitude, longitude, and elevation grids are also provided. More information is available in the README file. We anticipate users of this data can find more errors than we did and we ask only that the data are cited when used in publications. We recommend contacting us for further insight into our model.

---------------------------------------------

­GENERAL INFORMATION:

Directory contains compressed (gzipped) Tape ARchive (TAR) files that contain annual total surface mass balance (smb), freshwater discharge (runoff), and accumulation rate (C) data for 10 individual years spanning 1991-2000. The gzipped TAR files can be opened in windows, e.g. using Winzip or using the following UNIX command examples…

 

gunzip smb_binary.tar.gz

tar xvf smb_binary.tar        

 

ASCII and binary data formats are provided, '*.asc' and '*.bin' file suffixes, respectively. Latitude, longitude, and elevation grids are also included. Each 24 km horizontal resolution grid has dimensions of 55 x 101.

 

Read the JGR paper cited above to learn how the data were created and for a sense of the data accuracy. Please cite the paper and consult the authors of the data before publishing results based on these data. The authors can provide insight to ensure the data are not misinterpreted.

---------------------------------------------

­­Units:    The units of the data are meters water equivalence per year.

---------------------------------------------

­­READING IN DATA:

Below is an example in the IDL language of reading into memory binary data arrays…

inpath='/greenland/data/'      ; defining local location of data

year=['1991','1992','1993']    ; defining array of 4-character 'string'-type data corresponding to year

 

yy=0                           ; refering to 1st element in year array: '1991'

ni=55                          ; array i (approx. E-W) dimensions

 

nj=101                         ; array j (approx. N-S) dimensions

SMB=fltarr(ni,nj)

openr,1,inpath+year(yy)+'_E.dat'

readu,1,E

close,1

tvscl,congrid(E,ni*4.nj*4)     ; makes image plot with color scaled to values in array

end