======================================================================= README_ease.txt ---------------- Title: ARCMIP EASE-Grid model data Author: John Allison NCAR/EOL jja@ucar.edu +1-303-497-2633 Contact: NCAR/EOL PO Box 3000 Boulder CO 80307-3000 USA +1-303-497-8987 codiac@ucar.edu http://data.eol.ucar.edu/codiac_data/arcmip/docs/regrid/ Data set overview: Model data from participating ARCMIP modelers was re-gridded to fit the ARCMIP EASE-Grid. The ARCMIP EASE-Grid is a 50km equal area grid on a northern azimuthal projection, i.e. a 50km version of the original NL/NH grids for SSM/I described at the EASE-Grid webpage. < http://nsidc.org/data/ease/ease_grid.html > Call this "N5": a 50km grid for the entire hemisphere, which is a 361x361 grid (130321 cells). A rectangle on this grid was fit around the ARCMIP 70x55 domain.txt grid < http://cires.colorado.edu/lynch/arcmip/domain.html >, and a 5 cell (250km) border was added to define the ARCMIP EASE-Grid. The center of the new grid is 72N 153W, consisting of N5 columns 113-212 and rows 101-192 (100x92 = 9200 cells). The latlons of the corners of the ARCMIP EASE-Grid are: where col row lat lon upper left 113 101 41.89710 -139.69868 upper right 212 101 50.81162 157.94894 lower left 113 192 58.93256 -79.84573 lower right 212 191 74.54461 69.44396 ARCMIP EASE-Grid cells are calculated using up to the 4 closest model points [*] in a linear distance-weighted average to determine the value for each and every ARCMIP EASE-Grid cell. To avoid a significant edge effect where the 4 closest model points are quite far away, model points are not used if they are more than 50km away. Thus much of the EASE-Grid data is set to a missing value. [*] The model data may be created as grid cells, but during interpolation to the new grid, the data values are taken as point data located at the latlon associated with that cell. Presumably the latlon provided by the model is the center of that model grid cell. The linear distance-weighted average equation used: v0 is the ARCMIP EASE-Grid cell variable value v1,v2,v3,v4 are the model variable values at the 4 closest model points d1,d2,d3,d4 are the distances from the center of the EASE cell to the model point d4 is the furthest away num = v1 * d4/d1 + v2 * d4/d2 + v3 * d4/d3 + v4 * d4/d4 den = d4/d1 + d4/d2 + d4/d3 + d4/d4 v0 = num / den Terms drop out when dn>50km and a new furthest is determined. Data format: Data are stored in netCDF files similar in format to that described in the ARCMIP data archive document available at: < http://paos.colorado.edu/~curryja/arcmip/arcmip_data.pdf > The iew and jns indices have been changed to col and row since directional specifications are confusing on this grid. Original model domain latlons are in files named of the form "model_run_domain.nc". All original model documentation is also included, either as "model_run_*.txt" in the top-level model directory or "README*" in the appropriate subdirectory. In the "ease.tar.gz" file are all EASE-Gridded data, one for each of the model's daily "may" files, and for each of the monthly ("mon") files. Also included is the file "ease_domain.nc", containing latitudes and longitudes for the ARCMIP EASE-Grid (this is the same for every model). The file "model_run_distance.nc" contains the model indices of the 4 closest points to the EASE-Grid cell and their distances from the center of the EASE-Grid cell. These are the candidate points for interpolation (subject to the 50km maximum described above). The "may.tar.gz" and "mon.tar.gz" files contain the model's original data files for the May 1998 and each of the monthly files, respectively. The "col.tar.gz" file contains the model's column files in their original form. No processing to the EASE-Grid was performed. The data for each model are on one tape, with the two calibration datasets on another tape, one file per dataset. On the models tape: README_ease.txt, arcsym, coamps, hirham, narcm, smhirc. On the calibration tape: README_ease.txt, caspr, isccp. Each file is in the Unix tar format, containing metadata and data files in .tar.gz form as described above. The caspr and isccp files differ slightly in organization, but also contain .tar.gz files of the data. Example, to retrieve just the HIRHAM EASE-Grid files, use the standard Unix commands to skip the first 3 files and then un-tar the "ease" tarball from the "hirham" file: % mkdir hirham % cd hirham % mt -f /dev/rmt/0n rewind % mt -f /dev/rmt/0n fsf 3 % tar xvf /dev/rmt/0n hirham_hi02_ease.tar.gz % gzcat hirham_hi02_ease.tar.gz | tar xvf - A complete listing of the tape contents is appended below. References: ARCMIP home page http://paos.colorado.edu/~curryja/arcmip/ EASE-Grid details http://nsidc.org/data/ease/ease_grid.html Appendix: tape contents (with file size on disk- size on tape will differ due to blocking) Each file on the tape is a Unix tar file. Tape 1: models tape file 0: README_ease.txt (this file) 10957 file 1 (arcsym): arcsym_ar05_col.tar.gz 16595425 arcsym_ar05_domain.nc 47156 arcsym_ar05_ease.tar.gz 459420227 arcsym_ar05_may.tar.gz 349007077 arcsym_ar05_metadata.txt 2121 arcsym_ar05_mon.tar.gz 22091743 md5.txt 386 md5sum.txt 344 file 2 (coamps): COAMPS_0001_col.tar.gz 22689551 COAMPS_0001_domain.nc 1016908 COAMPS_0001_ease.tar.gz 837513306 COAMPS_0001_may.tar.gz 392753301 COAMPS_0001_metadata.txt 777 COAMPS_0001_mon.tar.gz 33141463 md5.txt 386 md5sum.txt 344 file 3 (hirham): hirham_hi02_col.tar.gz 13398856 hirham_hi02_domain.nc 42932 hirham_hi02_ease.tar.gz 541033488 hirham_hi02_may.tar.gz 382373181 hirham_hi02_metadata.txt 1466 hirham_hi02_mon.tar.gz 21976887 md5.txt 386 md5sum.txt 344 file 4 (narcm): md5.txt 386 md5sum.txt 344 narcm2_ar01_col.tar.gz 258129061 narcm2_ar01_domain.nc 47156 narcm2_ar01_ease.tar.gz 185988284 narcm2_ar01_may.tar.gz 73008655 narcm2_ar01_metadata.txt 1727 narcm2_ar01_mon.tar.gz 4705497 file 5 (smhirc): md5.txt 386 md5sum.txt 344 smhirc_4223_col.tar.gz 12780357 smhirc_4223_domain.nc 75284 smhirc_4223_ease.tar.gz 604574448 smhirc_4223_may.tar.gz 331329525 smhirc_4223_metadata.txt 974 smhirc_4223_mon.tar.gz 18837945 Tape 2: calibration tape file 0: README_ease.txt (this file) 10957 file 1 (caspr): CASPR.Readme 7059 caspr_199701.easegrid.tar.gz 59164781 caspr_199702.easegrid.tar.gz 56651010 caspr_199703.easegrid.tar.gz 64866164 caspr_199704.easegrid.tar.gz 64917650 caspr_199705.easegrid.tar.gz 70249219 caspr_199706.easegrid.tar.gz 69840538 caspr_199707.easegrid.tar.gz 71678697 caspr_199708.easegrid.tar.gz 68978990 caspr_199709.easegrid.tar.gz 64070065 caspr_199710.easegrid.tar.gz 64595491 caspr_199711.easegrid.tar.gz 58157249 caspr_199712.easegrid.tar.gz 57629903 caspr_199801.easegrid.tar.gz 58919143 caspr_199802.easegrid.tar.gz 56451955 caspr_199803.easegrid.tar.gz 63844112 caspr_199804.easegrid.tar.gz 64869045 caspr_199805.easegrid.tar.gz 71275337 caspr_199806.easegrid.tar.gz 71421138 caspr_199807.easegrid.tar.gz 72124805 caspr_199808.easegrid.tar.gz 70064728 caspr_199809.easegrid.tar.gz 64807045 caspr_199810.easegrid.tar.gz 64209356 caspr_199811.easegrid.tar.gz 57386154 caspr_199812.easegrid.tar.gz 56498217 caspr_easegrid_position.nc.gz 118129 map_tools_isccp.nc.gz 98909 md5.txt 1868 md5sum.txt 1679 file 2 (isccp): ISCCP.Readme 7404 isccp_9709.NOA.tar.gz 34049941 isccp_9709.NOM.tar.gz 34649477 isccp_9710.NOA.tar.gz 32532589 isccp_9710.NOM.tar.gz 31133276 isccp_9711.NOM.tar.gz 27014628 isccp_9712.NOA.tar.gz 28423221 isccp_9712.NOM.tar.gz 26860841 isccp_9801.NOA.tar.gz 28703730 isccp_9801.NOM.tar.gz 27007214 isccp_9802.NOA.tar.gz 28012687 isccp_9802.NOM.tar.gz 26073300 isccp_9803.NOA.tar.gz 34356813 isccp_9803.NOM.tar.gz 33647423 isccp_9804.NOA.tar.gz 35271717 isccp_9804.NOM.tar.gz 35933729 isccp_9805.NOA.tar.gz 37626464 isccp_9805.NOM.tar.gz 37667167 isccp_9806.NOA.tar.gz 37162713 isccp_9806.NOM.tar.gz 36922789 isccp_9807.NOA.tar.gz 37917838 isccp_9807.NOM.tar.gz 37662471 isccp_9808.NOA.tar.gz 37267473 isccp_9808.NOM.tar.gz 37328092 isccp_9809.NOA.tar.gz 34125032 isccp_9809.NOM.tar.gz 34015286 isccp_alb.NOA.nc.gz 23207272 isccp_alb.NOM.nc.gz 24242073 isccp_cbpr.NOA.nc.gz 29401910 isccp_cbpr.NOM.nc.gz 32513231 isccp_ctpr.NOA.nc.gz 33567092 isccp_ctpr.NOM.nc.gz 35084943 isccp_lwds.NOA.nc.gz 41201623 isccp_lwds.NOM.nc.gz 43300967 isccp_lwus.NOA.nc.gz 41875207 isccp_lwus.NOM.nc.gz 43921472 isccp_lwut.NOA.nc.gz 40855685 isccp_lwut.NOM.nc.gz 42912497 isccp_o3.NOA.nc.gz 18021017 isccp_o3.NOM.nc.gz 18927365 isccp_pwc.NOA.nc.gz 35877137 isccp_pwc.NOM.nc.gz 37795817 isccp_skin.NOA.nc.gz 35485312 isccp_skin.NOM.nc.gz 37192948 isccp_swds.NOA.nc.gz 17971147 isccp_swds.NOM.nc.gz 19231134 isccp_swdt.NOA.nc.gz 18186216 isccp_swdt.NOM.nc.gz 19505419 isccp_swus.NOA.nc.gz 16172019 isccp_swus.NOM.nc.gz 17201451 isccp_swut.NOA.nc.gz 17921911 isccp_swut.NOM.nc.gz 19084670 isccp_tau.NOA.nc.gz 28633341 isccp_tau.NOM.nc.gz 26891276 map_tools_isccp.nc.gz 98909 md5.txt 3418 md5sum.txt 3033 ======================================================================= CASPR.Readme ---------------- ---TITLE: CASPR-ARCMIP EASE GRID data set ---AUTHORS: Prof. Judith Curry School of Earth and Atmospheric Sciences Georgia Institute of Technology Atlanta, GA 30332-0340 office: 404 894-3955 fax: 404 894-5638 e-mail: curryja@eas.gatech.edu Contact for data questions : Randolph A. Reeder Aerospace Engineering Dept. University of Colorado Boulder, CO 80309 office: 303 492-4266 e-mail: rreeder@cloud.colorado.edu ---DATA SET OVERVIEW: The CASPR-ARCMIP EASE GRID data set consists of SHEBA AVHRR Arctic satellite retrievals corresponding to the ARCMIP domain location in EASE grid coordinates for the years 1997 and 1998. The original data is the ' SHEBA AVHRR Arctic 25 km Retrievals ( version 1.0 ) ' data set produced using the Cloud and Surface Parameter Retrieval ( CASPR ) algorithms. The original data set covered the entire Arctic twice daily for all of 1997 and 1998 at 25 km resolution. It is available from the web site at http://stratus.ssec.wisc.edu/products/ This CASPR-ARCMIP EASE GRID data is a subset of the original data corresponding to a 50 km resolution grid using EASE coordinates for the ARCMIP region. The ARCMIP region is defined to be the region bounded by columns numbers 100 through 260 and rows numbers 100 through 260 of the 50 km resolution Northern Hemisphere EASE grid. Information about the EASE grid is available at http://www-nsidc.colorado.edu/data/ease/ . The program ezlhconv.f available at this site was used to convert column and row numbers into latitudes and longitudes for a 50 km resolution grid. The values of each variable were calculated by taking a linear distance- weighted average using the nearest four data points to the EASE grid position. The time period for the CASPR-ARCMIP EASE GRID data set is 01 January 1997 through 31 December 1998. Twice daily data is given for 04:00 and 14:00 hours local time. The data set consists of a total of 1459 gzipped files, one for each day ( 2 years ) and each time ( 2 per day ). There is no data for 14:00 hours on 10 March 1998 ( file caspr_1998069_1400.easegrid.nc.gz is missing. ) ---DATA FORMAT: The CASPR-ARCMIP EASE GRID data set is contained in 24 gzipped tar files. Each file contains one month of data for the years 1997 and 1998. Each monthly tar file contains individual gzipped data files named caspr_yyyyddd_hhhh.easegrid.nc.gz . Here xxx = 001 thru 365 is the day of year and hhhh is either 0400 for 04:00 hours local time or 1400 for 14:00 hours local time. One file is missing ( 1998069_1400 ). Also included is a file called caspr_easegrid_position.nc containing latitude and longitude data for the EASE grid. All the files are gzipped NetCDF files containing 18 variables. Each variable is a two dimensional array indexed by row and column numbers related to the EASE grid coordinates. Most of the variables are stored as real data but must be multiplied by a scale factor. The names of the variables are : 1. LANDMASK = the surface type mask 2. TSKIN = the surface skin temperature in K 3. ALBEDO = the surface broadband albedo 4. RADIUS = the cloud droplet effective radius in micrometers 5. TAU = the cloud optical depth 6. PHASE = the cloud particle phase 7. CLDTEMP = the cloud top temperature in K 8. CLDPRES = the cloud top pressure in hPa 9. WATER = the precipitable water in cm 10. SWDSFC = downwelling shortwave radiation at surface in W/m^2 11. LWDSFC = downwelling longwave radiation at surface in W/m^2 12. SWUSFC = upwelling shortwave radiation at surface in W/m^2 13. LWUSFC = upwelling longwave radiation at surface in W/m^2 14. SWDTOA = downwelling shortwave radiation at TOA in W/m^2 15. SWUTOA = upwelling shortwave radiation at TOA in W/m^2 16. LWUTOA = upwelling longwave radiation at TOA in W/m^2 17. SWCLDF = shortwave cloud forcing at surface in W/m^2 18. LWCLDF = longwave cloud forcing at surface in W/m^2 The caspr_easegrid_position.nc.gz file contains 1. LAT = the latitude in degrees 2. LON = the longitude in degrees Each file contains 25921 ( 161 * 161 ) data values for every variable. The latitude and longitude given in the files are the positions of the EASE grid coordinates specified by the row and column number. The column index nx and row index ny both run from 1 thru 161. To obtain the EASE column and row index for the Northern Hemisphere 50 km grid just add 99 to these indexes. A special note with regard to the EASE grid : The row numbers increase going down, unlike many graphics packages ( like IDL ) which expect rows to increase going up. Please reference the included file map_tools.nc for help in mapping the data. There you will find a land mask array, that when plotted correctly ( see the proper EASE grid orientation ) can provide a mechanism for plotting the geophysical data within the same graphics package. In IDL, an algorithm such as this is needed to view the data properly : ;views the land mask for caspr data in arcmip EASE grid ncid = ncdf_open("map_tools.nc") varid = ncdf_varid(ncid,"land_mask") ncdf_varget, ncid, varid, subimage for r = 0,79 do begin for c = 0,160 do begin temp = subimage[c,r] subimage[c,r] = subimage[c,160-r] subimage[c,160-r] = temp endfor endfor tvscl, subimage>(-1) ncdf_close, ncid end ... ======================================================================= ISCCP.Readme -------------- ---TITLE: ISCCP Cloud and Radiative Flux in 50km ARCMIP-EASE Grid 199709-199809 ---AUTHORS: Prof. Judith Curry School of Earth and Atmospheric Sciences Georgia Institute of Technology Atlanta, GA 30332-0340 Phone: (404) 894-3955 Fax: (404) 894-5638 Email: curryja@eas.gatech.edu Yuanchong Zhang Goddard Institute for Space Studies 2880 Broadway, RM 320-B New York, NY 10025 Phone: (212) 678-5514 Fax: (212) 678-5552 Email: clyxz@giss.nasa.gov Contact for data questions : Mathew V. Rothstein Program in Atmospheric and Oceanic Sciences University of Colorado Boulder, CO 80309 Phone: (303) 492-4035 Email: Mathew.Rothstein@colorado.edu Randolph A. Reeder Aerospace Engineering Dept. University of Colorado Boulder, CO 80309 Phone: (303) 492-4266 Email: rreeder@cloud.colorado.edu ---DATA SET OVERVIEW: The ISCCP Cloud and Radiative Flux data set is a subset of ISCCP's FDX-P2 Sheba Flux Data set regridded into a 50km EASE-Grid. It contains surface and top-of-atmosphere radiative fluxes and the important physical quantities used to calculate them. The data set covers the time period of September, 1997 through September, 1998, which is roughly coincident with that of the SHEBA project. Spatial coverage includes the polar region of 55N to the North Pole. Temporal resolution is every 3hrs. There are two satellites available for the period: The NOA-14 afternoon (NOA) and NOA-12 morning (NOM) polar orbiters. For more information on ISCCP and their products visit: http://isccp.giss.nasa.gov/ For information on EASE-Grid visit: http://www-nsidc.colorado.edu/data/ease/ ---INSTRUMENT DESCRIPTION: The following is an excerpt from the ISCCP FDX-P2 readme file: Shortwave and Longwave Radiative Flux are calculated using the following datasets to specify the properties of Earth's Atmosphere and Surface: ISCCP-DX cloud dataset, TOVS operational sounder products, TOMS ozone products, a climatology of near-surface air temperature diurnal cycle constructed from the WWW Surface Weather reports and the first NCEP reanalysis, a climatology of cloud particle sizes from Han et al, a climatology of stratospheric and upper tropospheric water vapor from SAGE-II, a climatology of tropospheric and stratospheric aerosols used in the GISS climate model, and land surface albedo spectral dependence and spectral emissivities from the GISS climate model. ---DATA FORMAT: The data are stored by two different methods for easy access. The first set of files contain all cloud and radiative flux variables by 3hr window. These files are named: cloud_yymmddhh.sat.nc and rflux_yymmddhh.sat.nc where yymmddhh is the timestamp, sat is the satellite of origin (NOA or NOM), and nc indicates these are self-documented NetCDF files. These files are tarred together by month and gzipped into files named: isccp_yymm.sat.tar.gz The second set of files store the data by variable and contain the entire period of record. These files are named: isccp_var.sat.nc.gz where var is a two to four letter abbreviation of the variables described below. As mentioned, all data are stored in NetCDF files, which contain additional detailed information as metadata. The basic storage format is a 161x161 grid corresponding to the inner-most square of the 50km ARCMIP-EASE projection of the northern hemisphere. Converting between this grid and the full 361x361 hemispheric grid is simply a matter of adding/ subtracting 100 to the column and row. One special note regarding the EASE grid: The row numbers increase going down, unlike many graphics packages (like IDL) which expect rows to increase going up. Please reference the included file map_tools.nc for help in mapping the data. There you'll find a land mask array, that when plotted correctly (see the proper EASE grid orientation) can provide a mechanism for plotting the geophysical data within the same graphics package. In IDL, an algorithm such as this is needed to view the data properly: ;views the land mask for isccp data in arcmip ease grid ncid = ncdf_open("map_tools.nc") varid = ncdf_varid(ncid,"land_mask") ncdf_varget, ncid, varid, subimage for r = 0,79 do begin for c= 0,160 do begin temp = subimage[c,r] subimage[c,r] = subimage[c,160-r] subimage[c,160-r] = temp endfor endfor tvscl, subimage>(-1) ncdf_close, ncid end A list of variables is provided here: radiative flux variables: swds SW downwelling at surface (W/m^2) swus SW upwelling at surface (W/m^2) lwds LW downwelling at surface (W/m^2) lwus LW upwelling at surface (W/m^2) swdt SW downwelling at TOA (W/m^2) swut SW upwelling at TOA (W/m^2) lwut LW upwelling at TOA (W/m^2) skin surface skin temperature (K) alb surface visible albedo (0-1) cloud variables: o3 column O3 (Dobson unit) pwc column PW actually used (cm) ctpr highest cloud layer top pressure (mb) actually used cbpr lowest cloud layer base pressure (mb) actually used tau total column Tau actually used ---DATA REMARKS: 1) Missing data are represented by -9999. 2) There was no data for the NOA satellite during Nov, 1997. 3) An error was detected in the Shortwave flux calculation over land. Therefore, all Shortwave values in land grid cells are treated as missing (-9999). 4) In the process of regridding from original satellite datasets, extreme outliers were omitted and a cell's value was the average of all pixels falling in that cell, weighted by their distance to the cell's center. ======================================================================= --end