¶ Surface Meteorological Instrumentation (MET)
¶ Background
The ARM Surface Meteorology Systems (MET) uses mainly conventional in situ sensors to obtain 1-minute statistics of surface wind speed, wind direction, air temperature, relative humidity, barometric pressure, and rain-rate. Additional sensors may be added to or removed from the base set of sensors depending upon the deployment location, climate regime, or programmatic needs. In addition, sensor types may change depending upon the climate regime of the deployment. These changes/additions are noted in the Deployment Locations and History section of the instrument handbook.
¶ Additional Information
¶ Present Weather Detector
The Present Weather Detector was added to most MET systems at the Central Facilities and the Mobile Facilities in the past couple years. The Present Weather Detector (PWD) is an intelligent, multi-variable sensor for automatic weather observing systems (PWD User Guide, Attached). The PWD can measure the intensity and amount of both liquid and solid precipitation. A schematic of the PWD is shown in the PWD schematic, below. The PWD uses optical data along with data from the Vaisala Rain Sensor and the Temperature sensor to determine the precipitation intensity and type.
¶ Marinemet
The MET was deployed on the MAGIC mobile deployment, but was known as the Marinemet (below, right) and had additional sensors installed.
For more information see MET.
¶ Expected Operations
¶ Metrics
THE NSA metrics will normally have some flags tripped around the time it does the CMH daily self-check. The rest of the sites, albeit with some different variables, should be green for the most part. The TWP MET has an Optical Rain Gauge in addition to the normal Tipping Bucket Rain Gauge and Present Weather Detector. These additional instruments are still considered to be under the MET instrument.
¶ Plots
Meteogram
This plot will include multiple panels of temperature, relative humidity, wind speed and direction, pressure, and logger voltages and temperatures. Put what you've learned in your classes to good use and remember that fronts can cause some sharp jumps in the data.
Precip
This is a plot of all the different precipitation variables present in the file: precip rates along with rain and snow accumulations.
PWD Weather
This is a plot of the MET temperature and humidity with the Present Weather Detector (PWD) 15 minute averaged current weather plotted at the bottom. Also plotted is the KAZR data to give an idea of what is going on at the site.
PWD Meteogram
The PWD measures 1 minute and 10 minute mean visibilities. The are plotted in the top panel and the alarm status is in the bottom panel. The PWD Alarm should be 0.
Meteogram Week
This is a plot of the meteogram above for 7 days.
Precip Week
A plot of the precip above for 7 days.
¶ Comparison Plots
Precipitation Comparison
This plot includes all of the precipitation instruments at a site on 1 plot. It is a great way to compare the cumulative precipitation totals from each instrument to determine if any are having troubles. The MET TBRG is considered our "standard" at the moment and we normally use that as our baseline. NSA is different since there is not a tipping bucket there.
SGP MET Temperature Comparison
This plot is only available for SGP. All of the extended facilities are plotted up on 1 graph for 1 and 7 day(s).
Surface Conditions - SGP
Also only at SGP is a contour plot of all the extended facilities temperature, pressure, wind speed, etc... This is used to try and determine quickly a site that might be having troubles.
¶ Site-Specific Information
The MET is located at all sites.
NSA - Chilled Mirror Hygrometer
There is one nuisance flag that occurs with the NSA MET data stream. It involves the chilled mirror hygrometer (CMH). The flagging occurs when the chilled mirror goes through its daily self-test. The chilled mirror uses two sensors to measure temperature and dew point. The temperature is measured using a Platinum Resistance Temperature Detector. The dew point is measured using a thermopile that is located near the surface of a mirror. Using a thermoelectric heat pump the mirror is alternately cooled and heated to maintain a thin coating of dew (or frost) on the surface. The surface temperature of the mirror is the dew point. One problem with deposition on a surface is that contamination occurs. In order to detect and report the amount of contamination and to compensate the photo detectors a daily self-check is initiated at 12 hrs after power up and every 24 hrs afterwards. During the self-test the mirror is heated to above ambient air temperature to dry out the mirror so that the photo detectors can measure the amount of contamination. When this occurs the dew point value will be higher than ambient causing the calculated CMH RH value to exceed the 101% threshold. The example above shows the self check occurring at 12:28 GMT and lasting for 7 samples (7 mins). This is a necessary check as it informs operations when the CMH needs to be removed, cleaned and re-calibrated. The MET DQAs should list the time of day of this self check as a way to monitor how often the instrument is powering up. If the instrument is operating correctly the self check will occur at the same time each day.
There are no plans to remove this nuisance flag as in doing so we would remove any error flags that would occur when the system is failing. There are two other situations regarding the CMH self-test that require further interpretation. Due to some sort of manufacturer problem, the CMH self-test does not occur when temperatures are below -17 C. We are not sure why this occurs and testing of spare transmission boards and sensor units has not found a cause. There are no plans to continue efforts to resolve this issue. Secondly, the time of self-test should be consistent on a day-to-day basis unless a power disruption occurs. As stated earlier the self-test initiates 12 hrs after start-up and then every 24 hrs afterwards. Any disruption in power (power spike, switch-over to back up generators, removal of system for cleaning, etc) will cause the time of self-test to change. Listing the occurrence of self-test and the time is useful as a constantly adjusting time of self-test may alert operations to a power distribution problem.
Sometimes for an hour or two after a self check, the chilled mirror's dewpoint and RH are lower than those observed by the MET's other sensors. Right now, we're deeming this behavior OK.
¶ Known Issues
¶ Known Behaviors
Known issues for this instrument that MAY NOT need to be mentioned in your DQA's:
Influence from Aerosols
Aerosols have been known to influence data from the NSA. The aspirator of the CMH can inhale these aeorsols and make the data appear noisy (especially the RH data), but these data are actually normal. From Doug Whiteman:
“This is not noise, this is airborne actual exceeding/testing the capability.”
These data are correct and should not be classified as questionable or noisy.
¶ Past Problems
Past problems for this instrument that DO need to be mentioned in your DQA's and possibly requiring a DQPR submittal:
Poor Circulation
Another problem that shows up frequently is poor aspiration of the chilled mirror. This occurs when the aspirator opening becomes blocked by snow/frost, the fan fails, or mosquitoes get ingested into the fan housing and prevent the fan from rotating. It is fairly easy to determine when aspiration becomes a problem. Both the CMH temperature and dew point differ significantly from the 2m temperature and dew point data. Aspirator problems can also occur with the T/RH probes although fan failure is much less frequent than with the CMH.
CMH System Failure
When the CMH system doesn't work the following happens:
- 1) CMH temperature and CMH dew point data go missing
- 2) CMH RH data flatline at 100%
- 3) CMH vapor pressure and CMH saturated vapor pressure flatline at ~3.8 kPa
Flatlined Diagnostic Data
Sometimes, you may notice the diagnostic data (such as heater temperature and voltage) flatlining. This can occur simply due to the system hanging up, and can typically be fixed with a power cycle. Though these variables flatlining do not seem to affect the primary variables, it is still necessary to mention this issue in a DQPR.
