Albedo

1
Agreements on . . .

• Time schedule for soil moisture production
• Submission of results for two month period
• Preparation, start and data delivery for 15 month
  period
• Mars archiving
• Grid (rotated, regular lat/lon, both?)
• Specification of output parameter
• Interpolation (yes/no/who/how)

Interim ELDAS Data coordination meeting
2
Status of Soil moisture production

• 6 day validation test period
• Extracted from DWD archive, results are
  presented in validation appetizer by Eddy Moors.
• 2 month validation period
• Results ready for submission, some additional
  data processing for storage into MARS
• Full 15 month period
• Preparation of Model suite and assimilation
  system.

3
Model version for soil moisture production
4
Time line for 2 month period ofsoil moisture
production

• Modul results Archived at DWD. Ready for
  preparation of MARS input.
• Archiving into MARS 10/2003
• Results can be stored on rotated lat/lon,
  technically no problem. Assistance will be given
  by Guillaume Aubert, DWD consultant at ECMWF.
• Storage on regular grid results in loss of
  information. Double loss of information for
  potential LAM users who need to retransform again
  (e.g. for comparison). Therefore original
  unbiased results should be available.

5
Specification of output for MARS

• Grid for storage / both?
• List of parameter (Alterra)
• 3 hour values, 3 hour mean values (centered,
  integrated)
• special validation sites with other time
  increment (0.5 / 1 hour ?). What data format?
• in case of interpolationwhich interpolation
  flag for the different validation parameter
  (bilinear, nearest gridpoint)

6
Interpolation of grib fields

• Input of physiografic data and forcing works
• Ecoclimap external parameter can be transformed
  to DWD grib.
• precipitation fields can be interpolated from
  ECMWF grid to LM rotated grid.
• Output of validation parameter
• Interpolation LM rotated to reg. lat / lon doesnt
  work with DWD grib presently.

7
Interpolation of model output

• If interpolation of soil moisture needs to be
  done
• Interpolation routine needs to be modified
• it should be able to read DWD grib,
• an example routine that reads DWD grib can be
  provided.
• Specification of interpolation flag for different
  output parameter

8
Time line for 15 month period of soil moisture
production

• Preparation 10/2003 - 12/2003
• Include Eldas external parameter Surface
  height, FAO soil type, LAI, Z0, Albedo.
• Include Eldas forcing Precipitation
• Complete Implementation of TERRA-ML into GME /
  LM suite on Eldas domain.
• Adapt present SMA for multilayer soil model.

9
Time line for 15 month period of soil moisture
production

• Start of Soil Moisture production 01/2004
• Critical run time problem. Two monthly period
  needed approximately 2 month computer time.
• Some optimization can be done, maximum factor 2.
• Start at begin of 2004 should be forced
• Submission of full period earliest 06/2004 or
  part of results earlier.

10
Time line for 15 month period of soil moisture
production

• Update of assimilation system 01/2000 - 11/2000
• rel. humidity, heating rates, brightness
  temperature
• rerun, update results if full time can be
  recalculated
• optionally only seasonal rerun / subdomain (?)
• can be discussed in more detail in december
  meeting

11
Requirements

• Complete interpolation routine that reads grib
  fields on DWD rotated and transforms to ECMWF
  regular lat/lon grid (Bart).
• Final specification of validation parameter
  (List)
• 3 hour values, 3 hour mean values (centered,
  integrated)
• special validation sites with other time
  increment (0.5 / 1 hour ?), data format
• interpolation flag for the different validation
  parameter (bilinear, nearest gp)

12
Time schedule
13
(No Transcript)