¶ Micropulse Polarized Lidar - Fast Switching (MPLPOLFS)
¶ Background
The Micropulse Lidar (MPL) is a ground-based optical remote sensing system designed primarily to determine the altitude of clouds overhead. The physical principle is the same as for radar. Pulses of energy are transmitted into the atmosphere; the energy scattered back to the transceiver is collected and measured as a time-resolved signal.
From the time delay between each outgoing transmitted pulse and the backscattered signal, the distance to the scatterer is inferred. Besides real-time detection of clouds, post-processing of the lidar return can also characterize the extent and properties of aerosol or other particle-laden regions.
¶ Additional Information
The MPLPOLFS has gone through many iterations, but is most commonly still referred to as the MPL.
- MPL - First iteration of the instrument
- MPLPOL - 2006: Added polarization capability, so it could measure both co-pol and cross-pol, resulting in linear-depolarization-measurements (LDR) as well.
- MPLPOLFS - 2009: Upgraded to switch on every pulse, that is, at a 2500-Hz rate so as to eliminate aliasing affects.
The MPL is actually housed in a trailer and sends a signal out through a window (bottom, right). Site ops will clean the window everyday as part of their preventative maintenance.
For more information see MPL.
¶ Expected Operations
¶ Metrics
Normal Metrics
The metrics will normally be all green, except during the summer. It is normal for the preliminary_cbh to be missing. It is not being calculated anymore, but do to consistency reasons it is still in the file.
Normal Metrics Summer
During the summer, the background_signal_*_pol variables will trip the valid_delta flag around local solar noon. This is expected due to interaction with the incoming solar radiation.
¶ Plots
Note: Please see General Information On Radar for more details on different polarizations.
Diagnostic
This plot shows time series of
- Instrument Temperatures - This should generally follow the same trend and some will look noisy due to the air conditioning units turning on and off.
- Background Signal (Co and Cross Pol) - This will be flatlined throughout the night and increase during the day.
- Background Signal Standard Deviation - Standard deviation of the data plotted in 2.
- Energy Monitor - This is useful, to look at. Any sharp drops may indicate a problem.
Co Pol
The most helpful plots will be of the actual lidar returns. The co-pol image below shows a lot of interesting features. There are a lot of clouds in this plot. Since this is instrument sends out a light signal, it is very easily attenuated by cloud. If there is thick cloud, there is going to be very little signal getting through such as between 00 and 07 UTC below. Also on this day, we can see some very light bands between 1500 and 2100 UTC. These could be cloud, but normally signals that are this low are caused by aerosols or particles in the air.
Cross Pol
The cross-pol signal should always be less than the co-pol signal. Although we do see a lot of the same features, we might see some added features that don't show up in the Co-Pol data. In the example below, you can see some added aerosol layers that are not detected or very slightly so in the co-pol.
Linear Depolarization Ratio
This is normally just taking the Cross-pol backscatter and dividing it by the co-pol signal, but for the MPL, we are multiplying it by 100 as well. This was requested by the mentor. It gives an idea of the shape of the particles in various parts of the cloud/aerosol features. The streaking is normal in areas of signal extinction from cloud.
¶ Comparison Plots
Radar Lidar Comparison
A plot of all the vertically profiling instruments is very helpful for determining if the MPL is not detecting clouds when it should be. The over-plotted cloud base height (white) should line up fairly well with the MPL signal.
¶ Site-Specific Information
The MPL is located at all sites.
Sites Near The Equator
If the sun is directly overhead, the site operations personnel will cover the lidar around solar noon to prevent damage to the instrument. This will show up on the plots like below.
¶ Known Issues
¶ Known Behaviors
Known issues for this instrument that MAY NOT need to be mentioned in your DQA's:
Vertical Stripes in Background
One of the most often seen issues with the MPL is vertical stripes appearing in the lower part of the backscatter plot. This is usually a well documented error caused by two things.
- MPL heater, used to prevent condensation from forming on the laser window, can sometimes create turbulence above the laser beam causing it to defocus.
- Seen frequently at SGP, and is similar to the first problem, but in this case the turbulence is caused by local AC condensers on the roof of the building where the MPL is housed.
Both are not DQPR worthy, but should be mentioned in the DQA's. If the problem is very impactful, an email to the Instrument Mentor, may be necessary.
This issue can also be very slight as noted in the 3rd plot below.
Dew Condensation on MPL View Port Window
It appears that dew or condensation is forming on the MPL view port window during some nights. Notice the lack of background noise between the hours of ~15:00 - 22:00 GMT. Currently there is not much that can be done to resolve this issue for this particular case. This problem is currently being tracked in PIF P070826.1.
MPL Enclosure Window Causing High Depolarization Values
After it was repaired and deployed for the MARCUS field campaign, the AMF2 MPL instrument exhibited higher depolarization values when compared to values seen right before the deployment. This caused the liquid cloud state to be compromised. It was discovered that the enclosure/seatainer window of the MPL was affecting the polarization of the instrument. A new, optically flat window was purchased and data was validated with the new glass. New MPL windows were also purchased and installed for all other sites (except SGP).
Here is an example of the depolarization with the old enclosure window:
And here is what the depolarization values looked like after the new, optically flat window was installed:
Notice the significant decrease in depolarization values after the installation of the new glass (most easily seen by viewing the bottom portions of the plots, as shown within the black boxes above). See DQPR 6668 for more details.
While the cause of this issue is known and has been solved by installing new windows, it would still be helpful to keep an eye out for any unusually high depolarization values in the future. Any highly unusual values should be mentioned in your DQAs, and possibly with a DQPR.
Oscillations in Energy Monitor and Instrument Temperatures
Here is a case where the Energy Monitor and Instrument Temperature values show some oscillations on the order of a degree or 0.04 uJ. Normally we would say there is a problem with the instrument when the diagnostic variables are showing some strange behavior. Although these oscillations are most likely caused by the fan/heater used to keep condensation from forming on the MPL window. Because the backscatter plot does not show any signs of trouble, we are going to consider this within operating parameters of the MPL.
¶ Past Problems
Past problems for this instrument that DO need to be mentioned in your DQA's and possibly requiring a DQPR submittal:
Laser turned off
The following is an unusual event, but worth mentioning to demonstrate how looking at the big picture is more helpful than relaying solely on min/max flags. From the installation of the Darwin system, this particular laser ran hotter than most which was not a problem. The laser temperature routinely exceeded the max value. The long term trend showed the laser temp to be 10C above the detector temp when the normal difference is usually 5C. The following figures are from the same instrument over a three day period.
The first figure shows the long term trend which got flagged because it exceeded the max value. On the third day (GMT day 52) the Laser temperature seems normal because the it is not exceeding the max value. However, second figure on GMT day 51 shows there is a major problem. The backscatter profile could be misinterpreted as dense fog. But when combined with the drop in laser temperature, a problem is clear. In this case, the laser turned itself off. The backscatter profile is from atmospheric background with no laser output. Please note the sun shade closed at 04:00. The backscatter profile looks almost identical to the profiles with no laser. Another clue is the very high EM value in the last figure. This condition would not normally occur. Usually the EM should drop drastically in value if not go to zero entirely when the laser turns off. The two problems occurred simultaneously. A bad cable caused the EM to go high. It was coincidence they occurred together.
Detector Failure
Erratic energy_monitor
Malfunctioning Sun Shade
Not Switching Between Polarization Modes
Example of only operating in one mode at a time
Flat line data
Clipping Data
Flat Line Backscatter
