The addition of GPS windfinding to the NCAR/ATD rawinsonde system, previously known as CLASS, provides atmospheric scientists with superior wind profiling resolution. For projects in which wind resolution is not a high priority, the less-expensive LORAN sondes can still be deployed. In either case, GLASS retains the simple one-person operation and robust thermodynamic profiling well-known to CLASS users.
The GLASS rawinsonde system, typically housed in a towable trailer (but see Mobile GLASS below), includes equipment to conduct atmospheric soundings and make supporting surface meteorological observations. GLASS can employ either GPS or LORAN-C navigation data for windfinding. In GPS-based windfinding, a Vaisala RS 80-15GH radiosonde is used, with a MWG201 GPS Processor Card processing the nav signals from the sonde. In LORAN-C based windfinding, a Vaisala RS 80-15LH radiosonde is used, with an Advanced Navigations Inc. (ANI) Model 7000 LORAN-C navigator processing the nav signals from the sonde. Sounding thermodynamic data measured by the radiosonde are processed by an NCAR RS-80 Met processor. Surface data are collected from surface instruments connected to a Campbell CR-10 datalogger.
The Vaisala radiosonde contains either a GPS or a LORAN receiver, a 403 MHz band transmitter, and pressure, temperature, and relative humidity sensors multiplexed to an oscillator which generates a tone that is transmitted to the surface receiver. The thermodynamic data are transmitted from the sonde roughly every 1.5 seconds. GPS or LORAN radionavigation signals are received by the sonde and re-transmitted to the surface station. (The sonde does not perform any signal processing on the navaid signals.) Both signals, the thermodynamic data and the radionavigation signals, are transmitted to the surface on separate subcarriers of the 403 MHz band transmitter. Upon reception, the two signals are separated and directed to their respective processors. The thermodynamic data are transferred to the NCAR RS-80 Met processor while the radionavigation signals are sent to the appropriate navigator.
The standard GLASS data system consists of a power supply, an RS-232 Multiplexer, a rack controller, a 403 Mhz receiver, a navigation data processor, a meteorological data processor, and a personal computer. Information is transferred to and from the GLASS personal computer through RS-232 connections using the MUX mentioned above. The MUX switches between the navigator, the Met processor and the Campbell datalogger to gather the data required to process and display the atmospheric soundings.
The GLASS personal computer operates in either Windows 95, 98 or NT. When GPS sondes are deployed, the operating software used consists of four main programs—Surface Logging, Sounding, Utilities, and Analysis—written in LabVIEW, a graphical programming language that allows multitasking and provides graphical user interfaces. When LORAN sondes are deployed, the operating software is a DOS-based program written in HTBasic. The Basic interpreter is written by TransERA Corp. and is licensed by NCAR. Both software configurations can provide sounding data displays in real time. The displays are both graphical and text-based. The data are stored to the hard disk in ASCII and binary files which contain the information required to "recreate" the flight. These data files are saved and used in subsequent post-processing.
The latest version of the NCAR sounding post-processing and display
software, Aspen, can process GLASS sounding data. Here are examples of
Aspen display output from a GLASS (GPS) sounding during CASES-99: Winds
Plot and Skew-T
Before a sonde is launched, it is run through a pre-flight procedure in which the telemetry is verified and radionavigation signals are acquired. Thermodynamic data - pressure, temperature, and humidity - are checked and as soon as a sufficient number of GPS satellites or LORAN stations are acquired, the sonde can be launched. LORAN sondes are typically swung in a horizontal circle just prior to launch to promote airflow around the sensors and equilibrate them to ambient conditions. Doing this with GPS sondes would result in the loss of nav signals, so particular caution must be paid to potential sensor arm heating (see 3.2 and 3.3 below).
There are several launch configurations possible at any GLASS site. The balloon can be launched (1) from the trailer itself through an aperture in the roof; (2) from a "bag" launcher, a heavy vinyl tarp that contains and protects the balloon prior to launch (used in shipboard deployments); or (3) by hand when wind and other conditions allow.
Balloons of varying weight can be used with the radiosondes. Typically, a 200-gram balloon is filled with roughly 40 cubic feet of helium, which will take a Vaisala sonde to 50 or 60 mb before the balloon bursts. The ascent rate obtained with this amount of helium and a Vaisala sonde is about 4 m/s.
The sonde specifications are tabulated below.
TABLE 8 Radiosonde Pressure Sensor Specifications
|Range||3 to 1060 mb|
|Data System Resolution||0.1 mb|
|Sensor Resolution||0.1 mb|
The temperature sensor is a capacitive bead in glass encapsulation. The temperature sensors are calibrated at the factory.
TABLE 9 Radiosonde Temperature Sensor Specifications
|Range||-90 C to 60 C|
|Data System Resolution||0.1 C|
|Sensor Resolution||0.1 C|
The manufacturer's specification for the time constant is 2.5 seconds. The time constant of the thermistor, combined with the ascent rate of the sonde produce a slight lag in temperature measurement through the sounding. However, with typical atmospheric lapse rates the resultant smoothing of the temperature profile is less than the accuracy of the thermistor. The smoothing resulting from the lag time becomes more significant when the sonde crosses frontal boundaries or goes through strong inversions.
Experience has shown that if the sonde sensor arm is not protected or properly ventilated prior to launch, it can be adversely affected by solar heating. This results in a temperature reading that is too high, producing a false near-surface super-adiabatic lapse rate. Due to the small thermal mass of the temperature sensor and its supporting structure this effect is not long-lived. The thermal time constant of the sensor arm is 13 seconds and thus the problem goes away soon after launch, once the sensor is adequately ventilated.
The humidity sensor is a thin-film capacitive type sensor. The Vaisala type H radiosonde utilizes the type H sensor, which uses a humidity algorithm incorporating temperature compensation. This new sensor has improved humidity measurements over previous sensors, particularly in the high end of the humidity range (95% to 100%). Table 10 summarizes the humidity sensor specifications.
TABLE 10 Radiosonde Humidity Sensor Specifications
|Sensor||HUMICAP thin film capacitor|
|Range||0 to 100% Relative Humidity|
|Accuracy||2.0% Relative Humidity|
|Data System Resolution||0.1% Relative Humidity|
|Time Constant||1.0 second @ 6m/s flow, 1000mb, 20 C|
Solar heating of the sonde temperature/humidity sensor arm prior to launch can produce an error in the low level humidity measurement (and hence dew point).The humidity sensor gives a reading of the humidity relative to the temperature of the sensor surface itself. In a situation where the sensor surface is warmer than the surroundings, the humidity reading will be lower than ambient (vapor pressure remains unchanged, "sensed" saturation vapor pressure value goes up). Due to the thermal time constant of the sensor arm (13 seconds), the initial heating of the sensor arm affects the humidity data for roughly the first 40 seconds of the flight. (In a shaded, well ventilated situation, in which the sensor surface is in thermal equilibrium with its surroundings, an accurate ambient humidity measurement at the surface can be obtained.)
The effect of the heated sensor arm persists for a longer time in the humidity measurement than it does in the temperature measurement. The portion of the sensor arm where the temperature sensor is mounted is an isolated small cylinder which quickly comes to a thermal equilibrium with its surroundings whereas that portion of the sensor arm on which the humicap is mounted is much larger and thus takes more time to come to a thermal equilibrium with its environment.
The wind accuracy obtained from the LORAN navigation system is dependent on a number of factors which in general relate to the quality of coverage for a given area. The number of stations received (three is minimum), the strength of signal, and the geometry of the receiver with respect to the stations are all important factors. Winds derived from GPS have a constant accuracy once the minimum number of satellites (typically four) are received. Specifications for LORAN and GPS wind and position measurements are presented in Table 11.
TABLE 11 Wind and Position Measurement Specifications
|Manufacturer / Model #||Advanced Navigation Inc.||Vaisala|
|Model #7000 (ANI 7000)||MWG201|
|Wind Accuracy||1.0 m/s||0.5 m/s|
|Averaging Time||60 seconds||0.5 seconds|
|Data System Resolution||0.1 meter; 0.1 m/s||0.1 meter; 0.1 m/s|
In LORAN-C windfinding (Vaisala RS 80-15LH radiosonde), the ANI 700 navigator processes available LORAN-C signals relayed from the sonde, automatically acquiring the stations to be tracked for a given geographic location. Reception of at least three LORAN stations is required for position and wind calculation. The ANI 7000 outputs an ASCII status message containing field strength, signal to noise ratio, and signal time of arrival (TOA) information for each of up to eight stations being tracked.
In GPS windfinding (Vaisala RS 80-15GH radiosonde), the MWG201 GPS Processor Card processes the 1200 baud FSK signal from the telemetry receiver to compute winds from the Doppler frequencies measured by the dropsonde. The MWG201 contains a high-quality 12-channel commercial GPS full-up (code-correlating) receiver that measures the local satellite RF carrier Doppler frequencies. The 12-channel receiver also generates GPS time and time satellite ephemerides, and identifies the satellites and their Doppler frequencies. The local Doppler frequencies are then compared with the telemetered dropsonde Doppler frequencies; this allows the frequencies sent back from the sonde to be identified as originating from a particular satellite. The MWG201 then removes the satellite component of motion.
Retrieval of accurate surface meteorological data is an integral part of the GLASS sounding. Surface meteorological instruments are used to capture a data point which anchors the balloon sounding data to the surface. The surface pressure is used as the starting point for sonde pressure data and altitude calculation. The surface temperature, humidity and wind data are also used as starting points in the sounding data. The surface data are collected with independent surface meteorological instrumentation. These instruments are connected to a Campbell CR10 datalogger which processes the inputs into real numbers and outputs one-minute average data. These data are transferred to the GLASS personal computer, via RS-232, where they are used as the first point in a sounding.
A continuous record of surface data processed through the Campbell datalogger can also be logged to a floppy disk for a complete surface record at the site. During the flight, the surface data are buffered for recovery after the sounding is completed.
The surface pressure is measured with a Vaisala PTA427 or PTA427A pressure
sensor. The PTA427 pressure range is 800 to 1060mb while the PTA427A
pressure range is 600 to 1060mb. These sensors have an accuracy of +/-
0.5mb and +/- 0.8mb respectively. Both are silicon capacitive pressure
sensors patented by Vaisala. Both are temperature-compensated and produce
a linear voltage output over the full operating range. In order to interface
with the Campbell datalogger a 2:1 voltage divider was incorporated into
the cable from the pressure sensor.
The temperature and humidity sensors are contained in a Vaisala HMP35C instrument probe. The actual sensors are a Fenwal Electronics UUT5J1 thermistor and a Vaisala Humicap capacitive relative humidity sensor. The temperature sensor accuracy is +/- 0.4 degree C over the range -33 to + 48 degrees C. The accuracy of the humidity sensor against field references is approximately +/- 2% with a long term stability of better than 1% RH per year. The HMP35C sensor probe is protected and vented by an R.M. Young aspirated radiation shield, model number 43-408.
Wind speed and direction are measured with an R.M. Young 05103 Wind Monitor. The monitor is a propeller wind vane with a 0.9 m/s threshold for wind speed and a 60 m/s maximum. Wind direction is measured using a 360-degree mechanical precision conductive potentiometer. The wind direction measurement has a threshold of 1.0 m/s at a 10 degree displacement and a threshold of 1.5 m/s at a 5 degree displacement. The potentiometer is 10 K-ohm, with a life expectancy of 50 million revolutions, and has a 0.25% linearity through the entire range.
If needed for a given project, a radiation sensor (shortwave
incoming only) can be added to the data collection system. Measurement
specifications will depend on the sensor model used.
The NCAR Mobile GLASS facility is completely self-contained in a camper shell that can be carried by any 3/4-ton (or better) full-size pickup truck. The basic system is mounted inside the camper with all the hardware required to make atmospheric soundings, including equipment to make supporting surface meteorological observations. The mobility gives the project planner the option to deploy to a specific site make a sounding and if required move to another site for the next sounding. The first sounding can be active and in the air while the truck is relocate, although this does affect sounding quality. Sounding site station elevation values are typically taken from a topographic map. If that is not available or if the location is not absolutely certain, a calibrated aircraft pressure altimeter can be used.
The Mobile GLASS system components are the same as the standard GLASS components. As with the standard GLASS, the mobile facility can process windfinding based on either GPS or LORAN-C navigation signals.
Data communications from the truck to a central operations center can be managed in one of two ways. Data can be transmitted using a cellular phone or it can be sent using a packet radio communication system. Partial messages or pseudo-real time data can be sent during the sounding using a second computer and the cellular phone.
The Mobile GLASS uses the standard GLASS surface instrument package. The configuration of the Mobile GLASS surface instrument package is designed to minimize effects of truck itself on those measurements. Those aspects of the instrumentation which are unique to the Mobile GLASS facility are described in the following sections.
Mobile GLASS Surface Temperature and Humidity Measurement
The entire temperature and humidity sensor, attached to the end of a cross-arm, is mounted on a lightweight moveable tripod. The tripod can be placed up to 15 meters from the truck to remove the measurement from any influence of the truck itself. The sensors are wired through a 17 meter cable that is wound on a spring loaded reel for cable retraction. The cable is pulled out as the tripod is positioned for operation. The cable is easily retracted by pulling on the cable which allows the spring loaded reel to wind it in.
Mobile GLASS Surface Wind Measurement
The propeller windvane (R.M. Young 05103 Wind Monitor) is mounted on a telescoping pole on the roof of the camper. In its fully raised position, the windvane is approximately 6 meters above the ground. The direction is fixed as if north were the front of the truck. A "North Seeker", a magnetically driven potentiometer, is included in the system which reports magnetic north in relation to the front of the truck. With magnetic declination included the true wind direction can be calculated regardless of truck position.
Although rarely required for Mobile GLASS operations, radiation measurements can be made. The data collection system has space for a radiation sensor (shortwave incoming only) and can be configured like fixed GLASS to include radiation measurements.
Mobile GLASS Radiosonde Deployment
The Mobile GLASS has everything required for a successful radiosonde launch. Helium for balloon filling is stored under a platform at the back of the camper. A pressure regulator stored in the camper easily attaches to a helium tank for balloon inflation. The van holds three bottles of helium, which is enough for about 15 releases. Sondes and balloons are stored inside the camper.
A radiosonde is released from Mobile GLASS in one of two methods.
If the winds are calm a balloon can be inflated at the back of the truck
and tied off with little chance of damage. If the winds get a little
stronger the "bare" balloon technique can still be used if the operator
fills the balloon just before release and then uses his body to protect
the balloon before release. When the wind and weather get too dramatic
a bag launcher is used. In this mode, the balloon is inflated while protected
and held secure by a heavy vinyl material (the bag) before release.