Sounding Data Continuity

RS41 Data Continuity

The table below and data continuity documents will describe significant changes in the Vaisala Radiosonde RS41 family and the related ground equipment.

In typical cases, the effect on measurement outcome will be small compared to the total uncertainty in sounding.

Also significant changes that do not, in our opinion, have any effect on time series have been listed; those will be marked with "No data continuity effect".

To be able to use the data continuity documents the following information about the sounding system has to be known:

  • Types and serial numbers of the radiosondes used within the studied timeframe
    • NOTE: The serial number defines unambiguously the hardware and software combination of the radiosonde.
  • Ground equipment software versions used within the studied timeframe

Identification data can be found from the following documents:

Changes that can be identified with the help of the radiosonde serial number

2017-10 | RS41 | Cover improvement to RS41 | No data continuity effect

  • In cover improvement, the hard plastic covers of RS41 have been changed to EPS (expanded polystyrene) covers.
  • Regarding sounding operations and radiosonde measurement performance, the cover improvement applied to RS41 radiosonde does not involve any changes, but this is a significant change from the environmental point of view.
  • With the improved covers, the plastic content of RS41 is decreased by 47% and the weight by 27% when compared to the hard plastic cover version.
  • The comparison flight data is published on the White paper showing that the change does not have any impact on the measurement performance of RS41.

RS41 vs. RS92 Data Continuity

  • ​The impact of the switch from the RS92 to the RS41 on climatological time series is estimated to be moderate.
  • The improved accuracy of RS41 data does not affect average measurement values as much as it affects the consistency or reproducibility of the data.
  • Results indicate that the most significant impact on average values will be seen in humidity measurements of the tropical climates, especially in the humid conditions of the upper troposphere.
  • The statistical differences between the RS92 and RS41 are described in the enclosed White paper Comparison of Vaisala Radiosondes RS41 and RS92 using experimental sounding results

RS92 Data Continuity

The table below and data continuity documents describe significant changes in the Vaisala Radiosonde RS92 family and the related ground equipment. 

In all the cases, the effect on measurement outcome is small compared to the total uncertainty in sounding. The referred performance specification can be found from the enclosed Vaisala Radiosonde RS92-SGP brochure. Also significant changes that do not, in our opinion, have any effect on time series have been listed; those are marked with "No data continuity effect". 

To be able to use the data continuity documents the following information about the sounding system has to be known: 

  • Types and serial numbers of the radiosondes used within the studied timeframe
  • Ground equipment software versions used within the studied timeframe

Identification data can be found from the below documents.

Changes that can be identified with the help of the radiosonde serial number

2004-04 | RS92 | Fine Tuned Humidity Sensor Temperature Dependency Correction | U

  • Improved temperature dependency correction for humidity measurement
  • In production since April 6, 2004
  • Old data can be corrected to correspond to new data by:
    corrected humidity reading formula
    corrected humidity reading formula
    where:
    Um = measured humidity  
    US = saturation humidity
    dUS = humidity correction at saturation
    dU0 = humidity correction at 0% RH
 T US  dUS  dU0 US  dU0  dUs 
%rh %rh %rh %rh %rh %rh
40 100.0 -1.1 -0.1 -30 74.6 0.0 -1.9
30 100.0 0.2 -0.1 -35 71.0 0.0 -2.4
25 100.0  0.5 0.0 -40 67.6 -0.1 -3.0
20 100.0  0.8 0.0 -45 64.3 -0.2 -3.7
15 100.0 0.8 0.0 -50 61.1 -0.2 -4.4
10 100.0 0.8 0.0 -55 58.2 -0.3 -5.1
5 100.0 0.7 0.1 -60 55.4 -0.5 -6.0
0 100.0 0.5 0.1 -65 52.9 -0.6 -7.0
-5 95.2 0.2 0.1 -70 50.4 -0.8 -8.2
-10 90.8 -0.2 0.1 -75 48.2 -0.9 -9.4
-15 86.5 -0.6 0.1 -80 46.1 -1.1 -10.6
-20 82.3 -1.0 0.1 -85 44.2 -1.3 -11.8
 -25 78.4 -1.4 0.0 -90 42.4 -1.4 -12.8

2005-03 | RS92-SGP | Pulse heating of humidity sensors continued down to -60°C | U

  • ​A sensor that has gathered ice cannot measure humidity profile details accurately in the lower atmosphere. In the upper atmosphere it shows a too high humidity reading
  • Vaisala Radiosonde RS92 has two thin film humidity sensors. While the other sensor measures humidity, the other one is heated. The heating functionality reduces icing and condensation effects on the sensor. This results in reliable humidity measurements also when emerging from a cloud
  • When the Vaisala Radiosonde RS92-SGP was first released, the alternate heating was turned off when the radiosonde reached -40°C temperature. Since March 2005 the heating functionality has been continued to temperature -60°C. This functionality was also used in the WMO radiosonde intercomparison in Mauritius in 2005. The change leads to more reliable humidity measurements in soundings where there are high humidity conditions between temperatures -40°C and -60°C

2006-09 | RS92-SGP | Improved Coating of Humidity Sensor Contacts | U

Improved attachment decreases warming caused by solar radiation.

  • Relative humidity is a function of humidity and temperature
  • In daytime soundings, the humidity sensors and their contacts are warmer than the surrounding air they are measuring. This results in too low relative humidity values.
  • The effect is noticeable in the upper troposphere and lower stratosphere, especially at high humidity conditions. There the radiosondes with the improved coating measure up to 5-6 %RH higher humidity values compared to those with the old coating.
  • The new coating was used in the WMO Intercomparison of High Quality Radiosonde Systems in Mauritius, 2005
Coating of humidity sensor contacts

2007-09 | RS92 | Reinforced temperature sensor | T

  • Quartz fibre is integrated firmly into the sensor structure
  • This improves mechanical strength by a factor of 5
  • Boom frame absence improves sensor ventilation compensating for increased thermal mass and sensor surface
  • Boom frame removal reduces temperature fluctuation previously present in twin-soundings
  • More details of the phenomenon: Temperature filtering above 10 hPa in heavy rigging
  • Reinforced sensors delivered since 2007, sensor type can be identified with help of radiosonde serial number
Reinforcement to temperature sensor

 

Temperature Sensor Construction

Temperature sensor construction

Time Constants of Old and Reinforced Sensors

Reinforced sensors table

2008-06 | RS92 | Sensor boom coating modification | U,T | No data continuity effect

  • The back side of the sensor boom has changed to a shiny silver like the front side of the sensor boom
  • The change improved manufacturability: boom is easier to handle in manufacturing processes resulting in better yield and more equal quality
  • No effect on temperature measurement as the surrounding of temperature sensor have not changed at all
  • The effect on humidity is positive – if any. In test flights the difference was with-in reproducibility limits
  • Modification entered production in mid 2008. If needed, coating method can be identified with help of radiosonde serial number
Sensor boom coating

2010-11 | RS92 | Sensor boom contacts modification | U,T | No data continuity effect

  • The sensor boom contacts have been coated with gold instead of copper to make contacts more robust, e.g., against ageing
  • The change improved also manufacturability: boom is easier to handle in manufacturing processes resulting in better yield and more equal quality
  • Modification has no effect on temperature or humidity measurement as the surrounding of the sensors have not changed at all
  • Modification entered gradually into production in autumn 2010. If needed, contacts used in the boom can be identified with the help of the radiosonde serial number

Sensor Boom Contacts Modification Front

sensor boom contacts modification front

 

Sensor Boom Contacts Modification Back

sensor boom contacts modification back

Changes that can be identified with the help of the DigiCORA® sounding software version and/or a user setting

2005-11 | Revised solar radiation correction table for temperature sensor | T

Radiation correction table RSN2005 

  • Fine-tuned radiation correction table for Vaisala Radiosonde RS92
  • The new radiation correction was verified in the WMO Mauritius Radiosonde Intercomparison, February 200

Solar radiation correction table RSN2005​

Solar radiation correction table RSN2005​

NOTES: 

  • The corrections in the RSN2005 table as a function of pressure and sun elevation angle are shown in the table above
  • The corrections are subtracted from the measured temperature

Difference between RSN2005 and RSN96 tables

Difference between RSN2005 and RSN96 tables

NOTES:

  • The difference between the RSN2005 table and the original RSN96 table as a function of pressure and sun elevation angle is shown in the table above
  • Mostly the correction has increased, and thus the effect of the change is to decrease reported temperatures and consequently lower calculated heights
  • Below 100hPa the change is less than 0.05 °C and above 30hPa about 0.2 °C

2010-11 | Extending reported TEMP humidity measurements to -100°C | U | No data continuity effect

  • It has been common practice for many meteorological services to exclude humidity data from TEMP messages at low temperatures. The most commonly used cold temperature limit has been –40°C and it has been applied regardless of sensor design
  • Over the years, Vaisala has continuously improved its polymer based humidity sensor. The performance was already improved significantly with the Vaisala Radiosonde RS80 types. However, the Vaisala Radiosonde RS92 family finally provided a sensor quality which allowed Vaisala to recommend the removal of the temperature limit completely
  • Vaisala recommended that users of RS92 radiosondes change the temperature limit to –100°C and users of RS80 change the limit to –70°C

2008-08 | Filtering algorithm modified in order to take into account requirements for temperature measurement above 10 hPa in ozone soundings and soundings in heavy test flight rig | T

  • ​The movement of a radiosonde suspended under a test flight rig used in various sounding test campaigns is not the same as that experienced by an individual radiosonde in flight (rigged directly to a balloon)
  • The slower movement can cause excess temperature reading fluctuation, ie. the temperature sensor gets warmer than the ambient air for short periods of time. This is observed only at very high altitudes
  • The same phenomenon takes place with ozone soundings, since the ozone sounding set-up is much heavier than the normal RS92 radiosonde
  • The software used for filtering the raw observations into the reported values was modified to take into account the requirements for slow movement in ozone soundings and in rig test arrangements  
  • For further information, please refer to the Final Report of the WMO Intercomparison of High Quality Radiosonde Systems in Mauritius, 2005

2010-12 | Humidity Measurement Improved Algorithm | U

  • Humidity measurement algorithm has been improved to take into account sensors response time and sensor's heating by solar radiation, latter indicated formerly as dry biased humidity readings at high altitudes.
  • The algorithms' biggest impact is in day time soundings at altitudes of approximately ten to fifteen kilometers depending on the humidity profile and tropopause height.
  • The new algorithm was used in the WMO Radiosonde Intercomparison, Yangjiang, China, July 2010.
  • Note! Applicability of the algorithm is presented in the table "Vaisala Radiosonde RS92 technical change".
  • Instructions for simulating the old database files with the DigiCORA® software version 3.64 can be found in the following document: Improved calculations MW31 3.64

Effect on Integrated water vapor column example, tropical conditions

  • 17 day soundings and 18 night soundings included
  • Day soundings average with new algorithms is 59.3 kg/m2 and with old algorithms 57.4 kg/m2
  • Night soundings average with new algorithms is 60.4 kg/m2 and with old algorithms 60.4 kg/m2 

Sounding example, tropical conditions

Sounding example for solar radiation algorithm

Sounding example for solar radiation algorithm
Blue = relative humidity with solar radiation algorithm and response time algorithm
Grey = relative humidity without new algorithms
Day time sounding (tropical conditions)

Sounding example, high latitude conditions

Sounding example for solar radiation algorithm

Sounding example for solar radiation algorithm
Blue = relative humidity with solar radiation algorithm and response time algorithm
Grey = relative humidity without new algorithms
Day time sounding (high latitude conditions)

Sounding series statistics example, tropical conditions

Sounding series statistics example, tropical conditions

Comparison result between new and old calculation. Statistics of day time flights in tropical conditions, 20 flights
Blue = relative humidity with solar radiation and response time algorithms
0 %RH reference line = relative humidity without new algorithms

Sounding series statistics example, high latitude conditions

Sounding series statistics example, high latitude conditions

Comparison result between new and old calculation. Statistics of day time flights in high latitude conditions, 50 flights
Blue = relative humidity with solar radiation and response time algorithms
0 %RH reference line = relative humidity without new algorithms

 

  • Flight reproducibility test verifies the agreement between similar types of radiosondes when measuring the same atmospheric conditions. Based on the test results the new SW-based corrections, when being effective, also maintain the high level of reproducibility characteristics of Vaisala RS92 Radiosonde humidity sensor.

Reproducibility in sounding

Twin sounding differences with standard deviations using new algorithms, high latitude conditions.

 

Twin sounding differences with standard deviations using new algorithms, high latitude conditions.

Twin sounding differences with standard deviations using new algorithms, high latitude conditions. 
Left  = Day time, 25 flights
Right = Night time, 5 flights

2010-12 | Revised Solar Radiation Correction Table RSN2010 | T

  • Small changes for Vaisala Radiosonde RS92 solar radiation correction table have been made. In addition solar radiation correction algorithm takes now into account radiosonde ventilation during the flight.
  • The new correction table was used in the WMO Radiosonde Intercomparison, Yangjiang, China, July 2010.
  • Note! Applicability of the algorithm is presented in the table "Vaisala Radiosonde RS92 technical change".
  • Instructions for simulating the old database files with the DigiCORA® software version 3.64 can be found in the following document: Improved calculations MW31 3.64

Temperature sensor solar radiation correction table RSN2010

Temperature sensor solar radiation correction table RSN2010

NOTES:

  • RS92 solar radiation correction table RSN2010 for DigiCORA® Sounding Software version 3.64
  • The correction values in the table are as a function of pressure and sun elevation angle. Actual correction takes into account radiosonde ventilation in flight, presented table values are calculated for typical 5 m/s ventilation.
  • The corrections are subtracted from the measured temperature.

Temperature sensor solar radiation difference table RSN2010 - RSN2005

Temperature sensor solar radiation difference table RSN2010 - RSN2005

Sounding series statistics example: day time, tropical

Sounding series statistics example: day time, tropical

Daytime differences with standard deviations between new and old calculation (RSN2005). Statistics of day time flights in tropical conditions, 20 flights.

Blue = new calculation 
Reference line = old calculation
Average ascent rate 5.3 m/s, std. 0.4 m/s. Single soundings, length of string 30 meters.

Sounding series statistics example, night time, high latitude

Sounding series statistics example, night time, high latitude

Night time differences with standard deviations between new and old calculation (RSN2005). Statistics of day time flights in high latitude conditions, 30 flights

Blue = new calculation
Reference line = old calculation
Average ascent rate 5.2 m/s, std. 0.2 m/s. Single soundings, length of string 30 meters.

Flight reproducibility test verifies the agreement between similar type of radiosondes when measuring the same atmospheric conditions. Test proves that reproducibility remains at good level. 

Reproducibility in sounding

Reproducibility in sounding

Twin sounding differences with standard deviations using new calculation, high latitude conditions, 25 flights.
Average ascent rate of soundings is 5.3 m/s, std. dev. between soundings is 0.3 m/s. Length of balloon string is 50 meters.