GCOS Upper Air Workshop, 22 May 2006 Holger Vömel Cooperative Institute for Research in Environmental Sciences University of Colorado Water vapor observations in the upper troposphere and lower stratosphere

Overview Frost pointFrost point Lyman-αLyman-α TDLTDL Polymer (Vaisala RS92)Polymer (Vaisala RS92) SummarySummary

Cryogenic Frostpoint Hygrometer (CFH) Microprocessor control Microprocessor control Vertical Range: surface to ~28 km (surface to ~25 km on ascent) Vertical Range: surface to ~28 km (surface to ~25 km on ascent) Uncertainty: troposphere: > 4% MR stratosphere: ~ 9 % Uncertainty: troposphere: > 4% MR stratosphere: ~ 9 % Phase sensitive detector: electronic sunlight filter Phase sensitive detector: electronic sunlight filter No liquid/ice ambiguity No liquid/ice ambiguity Weight: ~ 400 gr Weight: ~ 400 gr Currently interfaced with ECC ozone sonde and Vaisala RS80 Currently interfaced with ECC ozone sonde and Vaisala RS80 Based on NOAA/CMDL frost point hygrometer Based on NOAA/CMDL frost point hygrometer 162 soundings 162 soundings Detector IR LED μ Controller Cryogen Air flow Lens Mirror Heater Frost layer Thermistor

Water vapor mixing ratio Ticosonde 2006, Jan 4 – Mar 4, 2006

Tropical tropopause dehydration Ticosonde 2006 Jan 4 - Mar 4, 2006 Jan 4 - Mar 4, 2006 Ticosonde 2005 July 8-24, 2005

CFH Advantages Measurement based on simple physicsMeasurement based on simple physics Measurement not calibrated for water vapor, but rather for temperature (assume vapor pressure equation is correct)Measurement not calibrated for water vapor, but rather for temperature (assume vapor pressure equation is correct) Extremely large dynamic rangeExtremely large dynamic range Long history for technologyLong history for technology

CFH Limitations Instrument may fail completely in “thick” liquid clouds or highly ice-supersaturated airInstrument may fail completely in “thick” liquid clouds or highly ice-supersaturated air High costHigh cost Availability of cryogenAvailability of cryogen Instrument needs minor preparationInstrument needs minor preparation Data product needs understanding of instrumentData product needs understanding of instrument

Snow White frostpoint hygrometer Peltier cooling Peltier cooling Vertical Range: surface to < tropopause Vertical Range: surface to < tropopause Uncertainty: 0.1°C DP/FP realistic 0.5°C DP Uncertainty: 0.1°C DP/FP realistic 0.5°C DP Sunlight sensitive detector Sunlight sensitive detector Liquid/ice ambiguity Liquid/ice ambiguity Weight: ~ 400 gr Weight: ~ 400 gr Currently interfaced with SRS34, Sippican or Tmax/Vaisala RS80 Currently interfaced with SRS34, Sippican or Tmax/Vaisala RS80

Snow White frostpoint hygrometer

Snow White comparisons NOAA/CMDL Galapagos/Huntsville/Boulder/Watuskosek CFH AWEX Oklahoma

Snow White Advantages Measurement based on simple physicsMeasurement based on simple physics Measurement not calibrated for water vapor, but rather for temperatureMeasurement not calibrated for water vapor, but rather for temperature Large dynamic rangeLarge dynamic range Simple operationSimple operation

Snow White Limitations Highly sensitive to cloud contaminationHighly sensitive to cloud contamination Occasional unexplained failureOccasional unexplained failure Moderate costModerate cost Limited to below tropopauseLimited to below tropopause Limited to RH ≥ 3-6%Limited to RH ≥ 3-6% Data product needs understanding of instrumentData product needs understanding of instrument

FLASH - Lyman-α Fluorescent Lyman Alpha Stratospheric Hygrometer (FLASH) Fluorescent Lyman Alpha Stratospheric Hygrometer (FLASH) Vertical Range: upper troposphere to stratosphere (0.5 to 500 ppmv) Vertical Range: upper troposphere to stratosphere (0.5 to 500 ppmv) Calibrated against reference frost point Calibrated against reference frost point Uncertainty: 9% MR Uncertainty: 9% MR Night time (descent) only Night time (descent) only Weight: ~ 1 kg Weight: ~ 1 kg Currently interfaced Vaisala RS80 Currently interfaced Vaisala RS80 45 soundings 45 soundings

FLASH - Lyman-α

FLASH Advantages Calibrated against frost point standardCalibrated against frost point standard High measurement precisionHigh measurement precision Very fast sensorVery fast sensor Largely insensitive to cloudsLargely insensitive to clouds Large dynamic range (low mixing ratios)Large dynamic range (low mixing ratios)

FLASH Limitations Instrument measures properly on descentInstrument measures properly on descent Can only measure during night timeCan only measure during night time Full moon limits data rangeFull moon limits data range High costHigh cost Measurement range : 0.5 to 500 ppmvMeasurement range : 0.5 to 500 ppmv Currently can’t check calibration in the fieldCurrently can’t check calibration in the field

MayComm TDL Tunable Diode Laser Tunable Diode Laser Vertical Range: mid troposphere to stratosphere Vertical Range: mid troposphere to stratosphere Calibrated against reference frost point Calibrated against reference frost point Uncertainty: 5% MR or 0.5 ppmv Uncertainty: 5% MR or 0.5 ppmv Payload weight: ~ 1 kg Payload weight: ~ 1 kg

MayComm TDL Costa Rica January/February TDL soundings

MayComm TDL vs CFH Costa Rica January/February CFH soundings

MayComm TDL – CFH comparison

MayComm TDL Variability TDL CFH

MayComm TDL Advantages High measurement precisionHigh measurement precision Very fast sensorVery fast sensor Insensitive to cloudsInsensitive to clouds Large dynamic range (low mixing ratios)Large dynamic range (low mixing ratios) Easy to useEasy to use

MayComm TDL Limitations Prototype onlyPrototype only High costHigh cost No lower troposphere (Can extend to surface with second path)No lower troposphere (Can extend to surface with second path) No stable interface/telemetry system yetNo stable interface/telemetry system yet

Polymer sensor (Vaisala RS92) Capacitive polymer sensor Capacitive polymer sensor Vertical Range: lower to upper troposphere (?) Vertical Range: lower to upper troposphere (?) Uncertainty: 2-5% RH Uncertainty: 2-5% RH Payload weight: ~ 250 g Payload weight: ~ 250 g Sensors go through undocumented changes Sensors go through undocumented changes

Polymer sensor (Vaisala RS92) AWEX, Oklahoma Nov 2003 Lautlos, Finland Feb 2004 TCSP, Costa Rica Jul 2005 Night time comparisons RS92 vs CFH

Polymer sensor ( Vaisala RS92 ) Day time comparison RS92 vs CFH TCSP, Costa Rica Jul 2005

Polymer sensor (Vaisala RS92) Advantages Very low costVery low cost Largely insensitive to cloudsLargely insensitive to clouds Easy to useEasy to use Vaisala RS92: One of the best of the polymer sensorsVaisala RS92: One of the best of the polymer sensors

Polymer sensor ( Vaisala RS92 ) Limitations Very large solar radiation sensitivityVery large solar radiation sensitivity Good measurements need third party correctionsGood measurements need third party corrections Very hard to trace calibration changesVery hard to trace calibration changes Very hard to trace sensor changesVery hard to trace sensor changes

Claimed accuracy CalibrationLimitations Dynamic range HistoryCost Ease of use Engineering status CFH 0.5°C DP/FP 4-9% ++No "wet" clouds (o)o research / small series Snow White 0.1°C DP/FP + Some clouds RH > 3-6% No stratosphere o +o ++ production small series Lyman-alpha (FLASH) 9% (20% below 2 ppmv) + Night time only Descent only No lower troposphere +o -- + research / small series TDL (MayComm) 5% 0.5 ppmv o? (++) Proof of concept Polymer (Vaisala RS92) 2-5% RH - No stratosphere Large radiation error Chemical contamination Very hard to trace sensor/calibration changes (++) Large scale production

Questions?