## NOAA KLM User's Guide## Section 7.6 |

Previous Section, Next Section

The Microwave Humidity Sounder (MHS) has nominal and redundant Platinum Resistance Thermometers (PRTs) built into the instrument. One set each of five PRTs and three calibration resistors is routed to the telemetry acquisition circuit of the Processor and Interface Electronics A (PIE-A) and PIE-B, respectively. MHS also has nominal and redundant local oscillators, Side-A and Side-B. One of the sets will be used for nominal and the other will serve as backup. A more thorough description of the MHS instrument can be found in Section 3.9.

The PRT temperatures are derived in a two-step process. First the PRT counts in the MHS data packets are converted to resistance *R*
(in ohms) using three reference resistor values which are in the data packets (in counts). Subsequent conversion of PRT resistance *R*
into PRT temperature is accomplished using a cubic polynomial of the form,

where *T _{k}* and

MHS has three PRT calibration channels which provide data for a linear count-to-resistance conversion that is updated in each scan. These are precision resistors whose values are known to high precision over their operating temperatures and life. Their values are chosen to lie at the upper, middle, and lower resistance ranges expected of the OBCT PRTs throughout mission life. These three resistors are referred to as the PRT Calibration channels 1, 2, and 3 (PRT CALn,where n=1, 2, and 3), respectively. Their values in counts are measured once per scan and are output in the science data packet. The resistance of the PRT CALn is assumed to be a linear function of the PRT CALn counts,

where *R _{CALn}* and

For each scan, these α and β values are computed and then applied to each OBCT PRT to convert the OBCT PRT count *C _{k}* into resistance

where *R _{k}* and

The mean OBCT temperature, *T _{W}* , is calculated from the individual PRT temperatures,

where m=5 represents the number of OBCT PRTs (as listed in Table 3.9.2.1-2) and *W _{k}*
is a weight assigned to each PRT

Similarly, a cold space temperature correction, Δ*T _{C}* , may be required. This is due to the fact that the space view may be
contaminated by radiation which originates from the spacecraft and the Earth's limb. Thus, the effective cold space temperature is given by:

where 2.73K is the cosmic background brightness temperature and Δ*T _{C}* will be determined from pre- or post-launch
data analysis. The Δ

There are 24 Housekeeping (HK) thermistors which monitor the temperatures at various MHS telemetry points, such as amplifiers, and local oscillators. These data, which are primarily for instrument health and safety monitoring, are not used in the radiometric retrieval algorithm of science data. The accuracy of these HK temperatures is less rigorous than that of the PRT temperatures. The 24 HK thermistors use a common set of conversion coefficients. The two-step process of converting counts to resistance and resistance to temperature can be compressed into a single step with negligible errors. This single step process computes the thermistor temperatures directly from the thermistor counts, using a polynomial of the form,

where *T _{th}* and

There are six current monitors that measure the current consumption of various power lines in the MHS instrument. The measured output in count
*C _{I}* is converted to current,

where *I _{0}* is the intercept and

In the MHS analog telemetry, there are three survival thermistors which monitor the temperatures of the Receiver, Electronics Equipment, and Scan Mechanism. These survival thermistors are powered to provide measurements even when the instrument power is off.

The conversion of the survival thermistor counts into temperatures is accomplished by a polynomial of the form,

where *V* ^{ }= 0.02 x Count represents the measured output in volts. One set of coefficients *h _{m}*
applies to all three survival thermistors.

The calibration algorithm from Mo (1996) that converts the Earth scene counts *C _{S}* to radiance,

where *R _{W}* and

where *u* is a free parameter, values of which are determined at three instrument temperatures (low, nominal, and high) from the pre-launch calibration data.
After the launch of MHS, the *u* value at an actual on-orbit instrument temperature will be interpolated from these three pre-launch values.

For each scan,
*C _{W}* represents the mean blackbody radiometric count of the four samples of the blackbody target. Similarly,

For MHS channel 19, the monochromatic assumption breaks down and a band correction with two coefficients has to be applied. These coefficients modify *T _{W}*
to give an effective temperature :

which is then used in the Planck function to give an accurate radiance. The application of Equation 7.6.6-4 is not necessary for the space temperature since the errors in the monochromatic assumption are negligible for such low radiance.

Quality control (QC) in the MHS calibration is very important for producing accurate calibration coefficients in the NESDIS operational calibration process. A scan-by-scan QC process can detect bad data which are flagged in the Level 1b data sets. All of the QC processes that have been built into the NESDIS operational AMSU-B preprocessor are to be included in this MHS algorithm. These and additional QC items are listed as follows,

- Intra-scan test of blackbody counts
*C*: If any two samples differ more than a preset limit of the blackbody count variation_{W}*ΔC*, the_{W}*C*should be excluded in Equation 7.6.6-3 by setting_{W}(t_{i })*w*=0._{i} - Intra-scan test of the space counts
*C*: If any two samples differ more than a preset limit of the space count variation_{C}*ΔC*, the_{C}*C*should be excluded in Equation 7.6.6-3 by setting_{C}(t_{i})*w*=0._{i} - Inter-scan test of PRT temperatures
*T*: If a_{k}*T*differs by more than 0.2K from its value in the previous (good) scan line, the_{k}*T*should be omitted from the average in Equation 7.6.4-1 by setting_{k}*W*=0._{k} - Test of antenna pointing accuracy : If an antenna position reading is out of a preset limit,then an error flag will be set in the Level 1b data.
- Radio frequency interference (RFI) correction : It was observed that the transmitters on the NOAA KLM spacecraft can produce serious RFI to the AMSU-B data. A corrective algorithm was developed for correction of the RFI in the AMSU-B. The same algorithm will also be used in the MHS calibration algorithm. Detailed description of the AMSU-B RFI corrective algorithm can be found inAppendix M .
- Detection and exclusion of the Lunar contaminated space samples from the calibration: Calculate the angular separation between the Moon and each viewing direction of the
four space samples. Reject those samples that are within a pre-defined angular threshold (default = 1.5
^{o}). In the worst case, three samples may be rejected in this process (keep the sample that has the largest separation angle if all four samples fall within the pre-defined angular threshold). Description of how to calculate the angular separation between the Moon and the space viewing direction is given in Kigawa and Mo (2002). Store the calculated separation angles. - Inter-scan test of sudden jump (or drop) of
*C*and_{W}*C*: Such sudden change in_{C}*C*and_{W}*C*has been observed in the NOAA-17 AMSU-A data. A corrective algorithm from Mo (2002) was developed for correction of the effect of such a sudden change in the calibration counts on the calibration coefficients._{C}

The NOAA Polar Orbiter Level 1b data are raw data that have been quality controlled and assembled into discrete data sets, to which Earth location and calibration information are appended but not applied. For simplification of application, Equation 7.6.6-1 can be rewritten as,

where the calibration coefficients *a _{i}* (where

and

where *G* represents the channel gain and is defined as

These calibration coefficients will be calculated at each scan line for all channels and appended to the Level 1b data. With these coefficients,
one can simply apply Equation 7.6.8-1 to obtain the scene radiance *R _{S}*. Users, who prefer brightness temperature
instead of radiance, can make the simple conversion,

where *B ^{-1 }(R_{S}) * is the inverse of the Planck function for radiance

For MHS channel 19, the band correction must be taken into consideration in the inverse process as follows,

Previous Section |
Top of Page |
Next Section |

http://www.ncdc.noaa.gov/oa/pod-guide/ncdc/docs/klm/html/c7/sec7-6.htm

Downloaded Friday, 18-Apr-2014 02:13:53 EDT

Last Updated Tuesday, 03-Mar-2009 10:29:33 EST

Please see the NCDC Contact Page if you have questions or comments.

Downloaded Friday, 18-Apr-2014 02:13:53 EDT

Last Updated Tuesday, 03-Mar-2009 10:29:33 EST

Please see the NCDC Contact Page if you have questions or comments.