DQRID : D950110.1
Start DateStart TimeEnd DateEnd Time Data Quality Metric
02/12/1994000001/05/19952359Suspect
12/22/1993000002/28/19942359Suspect
more
Subject:
Reprocess: Effect of BBSS ground check on MWR tuning functions
DataStreams:DsgpmwrlosB1.a0, sgpmwrlosB1.a1, DsgpmwrlosB1.a1
Description:
Due to concern over the impact of the Ground Check (a one-point
	calibration applied to the Vaisala sonde prior to launch) on
	the MWR tuning functions, I have recalculated the tuning
	functions without the ground check.  The change in the tuning
	functions is slight and has only a very small (maximum of 1%)
	impact on the precipitable water vapor and cloud liquid water
	path values provided by the ARM microwave radiometers.

TUNING FUNCTIONS

	The tuning functions relate microwave brightness temperatures
	measured by the the radiometer to those computed from
	simultaneous radiosondes using a model of microwave radiation
	transfer.  The tuning functions are necessary to permit the use
	of the retrieval functions for precipitable water vapor and
	cloud liquid water which are based on the model calculations.
	The tuning functions are dependent only on the model; they are
	independent of the location and of the instrument (assuming it
	is properly calibrated).

METHODOLOGY

	Barry Lesht, the BBSS instrument mentor, provided me with 325
	soundings from the SGP central facility covering the period
	October 1992 - December 1993 from which he had removed the
	effect of the ground check.  For each of these soundings I
	computed the microwave brightness temperatures at the
	frequencies used by the ARM microwave radiometers.  Of these,
	91 were determined to have occured during periods when the sky
	was clear and horizontally homogeneous on the basis of the
	standard deviation of the measured brightness temperature of
	the liquid-sensitive channel of the radiometer being less than
	0.5 K for the period 10 minutes prior until 30 minutes
	subsequent to the sonde launch.  The mean of the brightness
	temperatures measured during this time period were then used in
	a linear regression to determine the intercept and slope of the
	tuning functions.

RESULTS:

	The new tuning functions are:
	
	23.8 GHz:  TB_model = 0.789 + 0.915 TB_measured  (R2 = 0.994)
	31.4 GHz:  TB_model = 1.142 + 0.910 TB_measured  (R2 = 0.988)
	
	These were applied to the MWR data acquired during the period
	October 1992 through December 1993 and compared with previous
	values.  Only for values of PWV > 3 cm was the effect of the
	new tuning functions as much as 1%.

	For the 91 clear-sky soundings, the RMS difference between the
	preciptable water vapor from the MWR and from the BBSS was 0.1
	cm; the mean difference was 0.00 cm.

	For 59 soundings during June 1993 (clear and cloudy sky), the
	mean difference in the PWV computed from the radiosondes (No
	ground check - ground check) was 0.05 cm; the RMS of the
	difference was 0.08 cm.

Other observations/measurements impacted by this problem: qmemwrlos

Suggested Corrections of the Problem: (e.g. change calibration factor and
recompute, flag data with this comment, etc.)

	1.  The new tuning functions were implemented on all SGP MWRs
	    on 5 Jan 1995.

	2.  The MWR data from Oct 92 - Dec 93 have been recalculated by
	    the mentor and are available from him (as ten-minute
	    averages in ASCII format) upon request.

	3.  A computer code has been provided by the mentor to the ARM
	    Experiment Center to duplicate this reprocessing, if the PRB
	    so desires.
Suggestions: 
Measurements:DsgpmwrlosB1.a1:
  • Development data stream - documentation not supported
more
DsgpmwrlosB1.a0:
  • Development data stream - documentation not supported
more
sgpmwrlosB1.a1:
  • alt
  • 31tbsky
  • lon
  • actel
  • wet_window
  • 23tbsky
  • vap
  • actaz
  • time_offset
  • lat
  • ir_temp
  • liq
  • base_time
more

Close this window