FSSP - Hotwire Comparison

To compare the output from FSSP to Hot Wire output for liquid water.

Calibration constants are determined inflight, and stored in a constants file for each aircraft. (citation_constants is default for older projects). The constants are determined through a “best fit” of datapoints gathered when the aircraft is flying in dry air through a wide range of true airspeed (traditionally through a maneuver of constant altitude, increase and decrease of airspeed). The data processing computes the real-time density of the air through the use of “State” variables collected on the aircraft inflight. Using this information, the voltage required to maintain a constant temperature of the wire (125 C for DMT hotwire and 185 C for king (CSIRO) probe) is corrected to provide an offset to the liquid water measurement such that it should indicate Zero liquid water when not in cloud.

• Noise - Random variations in probe voltage are often observed in the DMT LWC probe. This tends to cause departures from calibrated values and anomolous readings of cloud water when not in cloud as determined by the FSSP. Will also cause abnormal indications when in cloud.
• Offset Limitations - certain periods of drastic changes in altitude or airspeed (rapid descents or ascents) will cause the processed baseline to deviate from zero LWC. The direct cause is unknown, and may be from compressional heating or angle of attack disturbance of the sensing wire.

The processing code used to handle hot wire probe data is:

• Hotwire2lwc - Produces the liquid water output: *.hotwire.raw from use of the lwc power in the physical.raw file. Includes options for “no_fssp” (doesn't check for an fssp file to use as a baseline adjustment)

• Hotwire_calib - Used to calibrate the hot wire probe. requires the start and end time of the calibration procedure, or any period of flight when not in cloud (as verified by FSSP data) with large variations in true airspeed.
• analog2physical - Takes the raw analog data and converts it to physical parameters “ *.physical.raw” . The hot wire sometimes requires the use of the “hotwire_offset”, which removes a 1 volt offset in voltage that is used in Saudi 2009 and other projects. This lowers the physical LWC power to allow the calibration to produce a zero LWC value.

The FSSP computes Liquid water content of the clouds by a summation of the spherical volume of detected particles for a unit volume. The volume is determined through a computation of the volume of the detection laser, and the aircraft true airspeed as it sweeps through the air.

See the FSSP page for a more detailed explaination of FSSP operations.

• Proper calibration constants: Calibrations must be done throughout the project, and updated in the constants file for each project (based on the aircraft flown). this ensures accurate sizing of the particles.
• Sample Volume: the sample volume is a critical value for computation of liquid water content. The issue is complicated by the presence of several versions of similar probes, CDP, FSSP-SPP, FSSP. The CDP and FSSP-SPP have improved electronics which are said to remove dead-time and most coincidence errors due to increased signal processing speed. The CDP also has slightly different optics. Conversations with Gary Grainger at DMT have confirmed that removing the dead-time and coincidence correction are appropriate for the new probes. The CDP provides a sample area in the calibration documents, which was “determined through modeling of the laser volume” (approximate Gary Grainger quote). This is simply constant through each of the probes. Mr. Grainger also confirmed that the older Sample Volume method of computing a beam fraction, seemed reasonable, and produces similar results.
• Contamination with Ice: the FSSP detects all particles within the sample volume, assuming they conform to Depth of Field and average transit limitations. This includes ice particles. The scattering of the laser is assumed to be through the use of the refractivity of liquid water. Therefore, ice and particles with high concentrations of solute may be different. Therefore, the FSSP spectrums and liquid water contents are highly suspect when temperatures are below freezing, due to sizing errors.

• FSSP_counts2conc: takes the raw output from the FSSP (as extracted in the process_raw) and computes the sample volume, concentrations, and liquid water content.
• setspp / setfssp: Produces the sizing table which is used to determine the actual size of the particles that are sized into each bin. Produces an output chart which is then used to input the calibration constants in the constants file for each project. Requires the peak, pre-peak, and post-peak channel, and counts from a ground calibration test.

Liquid water content from 3 different projects are compared to detect similarity. No definitive answer to the question of which is correct is determined, simply the relationship to each other. The POLCAST2 and Saudi Arabia 2009 projects used an FSSP-SPP and a DMT hot wire probe (LWC-100). The Mali 2007 project used an FSSP and a King (CSIRO) probe.Only in-cloud data was used for the comparison, due to baseline drift of the hotwire probes. The data from each flight was subsetted based on in-cloud (FSSP concentration greater then 50/cc) and further, when above zero C and below zero C. The ratio of FSSP LWC / Hot Wire LWC was computed with statistical analysis options of CPLOT (see the cplot page on use of the stats options).