Staff:
Steve, Matt, and Gary. Gary left this evening for Las Vegas and Mack arrived to take his place. Thanks for all your hard work, Gary!
Weather:
Hot again, with scattered cirrus and few cumulus. Winds were a little lighter and less gusty this afternoon than the previous few days. PBL depth looks to consistently exceed 4 km each afternoon, based on soundings and profiler data.
Soundings:
Students launched at 10am and 3pm local. No known problems.
Checklist:
water level in lidar is good
web plots
real-time lidar displays
449 power: 2.79 A in the afternoon.
Cleaning:
Lidar lenses last cleaned by Bill, next cleaning planned Tuesday, Aug 1.
NR01: next cleaning planned Tuesday, Aug 1
Other notes:
VAD winds have not been generated since yesterday afternoon according to Nagios. Otherwise everything seems to be running smoothly.
Staff:
Steve, Matt, and Gary.
Weather:
Hot and mostly clear, with a few fair weather cumulus and gusty winds in the afternoon.
Soundings:
Students launched at 10am and 3pm local. No problems.
Checklist:
Matt checked:
water level in lidar is good
web plots
real-time lidar displays
449 power: 2.79 A in the afternoon.
Cleaning:
Lidar lenses last cleaned by Bill, next cleaning planned Tuesday, Aug 1.
NR01: next cleaning planned Tuesday, Aug 1
Other notes:
See previous blog post for note about 449 profiler being briefly turned off this morning. Webplots are now working, the news of which Bill shared with the PI's yesterday. Otherwise all seems to be well.
The 449 profiler was shut down briefly around 9:00 am local to allow Shane to go into the compound to check on a circuit breaker. Apparently a GFCI had tripped and REVEAL had lost power. Profiler restarted at 09:11.
Staff:
Justin and Delia left today. Steve, Matt, and Gary remain.
Weather:
Hot and clear.
Soundings:
Students launched at 10am and 3pm local. No problems.
Checklist:
Matt checked:
water level in lidar is good
web plots
real-time lidar displays
449 power: 2.93 A in the morning.
Cameras are working, but images are not yet being downloaded automatically.
Cleaning:
Lidar lenses last cleaned by Bill, next cleaning planned Tuesday, Aug 1.
NR01: next cleaning planned Tuesday, Aug 1
Other notes:
I changed the iss2 network so the cradlepoint is the router for entire instrument network instead of the data manager. (I haven't figured out why I couldn't get routing to work on Alma 9.) This allows updates on the DSM, NTP time sync to work on the camera without kludging DNS, and it makes it easier to forward a port on the cradlepoint to the camera: http://iss2-field.dyndns.org:25180/viewlive.html.
LapXM was indicating an error, "New dwell found that does not match data set" on 2023-07-26. Photo below. It does not look alarming and things are still working, so I cleared it.
Staff:
Steve, Matt, and Justin moved t0, while Gary sat in the trailer working on iss systems stuff, and Delia stayed at the hotel to attend meetings and the intern workshop.
Weather:
Hot and clear. Windy in the afternoon. We survived a direct hit by a dust devil yesterday afternoon. (See ISFS blog for more details.)
Soundings:
Students launched at 10am and 3pm local. No problems.
Checklist:
Matt checked:
water level in lidar is good
web plots, except for lidar plots which were not working (see below)
checked real-time lidar displays
449 power: 2.83 A.
Cameras are working, but images are not yet being downloaded automatically.
Cleaning:
Lidar lenses last cleaned by Bill, next cleaning planned Tuesday, Aug 1.
NR01: next cleaning planned Tuesday, Aug 1
Other notes:
The 449 profiler was down for several hours early in the morning when too many HawkEye processes started hanging until the process limit was hit. Gary rebooted iss1, ran a system package update (unrelated to the hanging processes), and restarted the 449. Isabel has been working on resolving the problems with HawkEye in the lidar processing.
Nagios has been setup on iss2 and checks are slowly being cleaned up checks on iss1 and iss2. There are still some checks failing due to unknown data files (mostly related related to work in progress to download camera images), and iss catalog.
There is also a Temperature Zone 2 failing check on iss2. Probably that is just a missing limit configuration, but it has not bee investigated yet.
On the 20th of July around 4pm, the calibration kit was installed on the CS125. After leaving it for several hours, the data for the visibility was at 960. From the calibration kit, the TMOR calibration should be at 997. Software can be used to perform a full calibration on the instrument.
Below are pictures of data while the calibration kit was on the CS125 and what the TMOR should be.
Clear early, but scattered clouds developed and gradually thickened. Winds were light and variable during the morning, then the usual southerly picked up during the afternoon.
Discovered that the MISS profiler eastward as well as the northward winds were reversed overnight, so I must have reversed the east-west beam codes yesterday instead of the north-south. Fixed the problem and from about 1620 UT the MISS winds now appear correct.
The students launched two radiosondes, 10am and 3pm PDT as usual. No problems apart from the surface met data is not appearing on the sounding software display but that does seem to be making it into the data files. Carol reported that the sounding messages are successfully transferring to the GTS. Matt arrived last night and joined the soundings today to get a catch-up on procedures. These are mostly the same as previous projects, except that here we are using the iMet balloon enclosure (basically a large fabric tub that we inflate the balloon in as a test for possible use in Hydrogen balloons).
The. ISFS group moved the tall tower today. I had been using it as a hard target for calibrating the orientation of the Windcube lidar. I did one last set of scans this morning before they moved it, now it is beyond the ISFS tower array around 2 km to the east.
Cleared off the glass windows of all three wind lidars, the ceilometer, and the Hukseflux NR01 radiometer (around 23 UT). The Windcube washer water tank was almost empty and using water faster than expected so we will have to monitor that closely. Made minor adjustments to the Windcube scan strategy, the MISS height range, and the UVA Halo orientation (it was pointed about 1 deg west of north). Also tied down that lidar in case of high winds, backed up data, and tidied up around the site. The Windcube data hasn't been transferring to the Data Manager computer for a couple of days, Isabel is investigating.
Visited Shane and Bruce and got a tour of the REAEL lidar container - impressive.
A sequence of photos of a sounding launch from the iMet balloon enclosure
The day started out mostly clear with light winds, but clouds developed and late afternoon a windy squall came through the site with a short period of driving rain.
The ISS is almost fully operational. The three wind lidars are in their planned scanning modes, although there is still some tuning to do. Every hour a short period of coordinated scans are planned to intercompare the lidars, however the lidars aren't scanning as expected. The windcube scans are taking longer than expected (7-9 minutes instead of the 5 minutes planned) so those will need to be speeded up, and Halo lidars seem to be ignoring the programmed scans but fortunately are doing the desired stare modes.
The Metek Halo lidar on the roof of MISS appears to move slightly off target. This lidar is supposed to be lined up with the ISFS array, however the level indicator in the lidar indicated that it moving around 0.1-0.2 degrees which corresponds to about 5 meters displacement at the range of the ISFS array. In addition the drive mechanism appears to have some hysteresis so that if the lidar head behaves differently depending on the direction the head is turned. I was not able to detect the array in the returns, but can easily detect the REAL trailer so used that and the ground as targets. Currently the best setting seems to be an azimuth of 78 deg and elevation of 0 or 0.1 deg.
The students let off two soundings today (10am and 3pm PDT or 17UT and 22 UT). Unfortunately they didn't attach the 17UT sonde securely and it fell from the ballon soon after launch. Given that the conditions weren't ideal, Chenning decided not to relaunch. The 22UT was successful. About 10 minutes into the flight, the balloon appeared to hit a downdraft and the sonde dropped about 10 mb (around 100 meters). After a few minutes, the flight resumed normally.
MISS: comparisons with the Modular Profiler showed that the northward winds on the 915 MHz profiler are out of phase with Modular Profiler. The eastward winds agree well. This implies that the north and south oblique beams may be swapped so I switched the programing codes and will test again tomorrow.
Gary and I installed a Purpleair aerosol sensor and a Microseven camera on the ISFS array. We got caught in the squall out on the array so will need to revisit tomorrow to tidy up and adjust the pointing angle of the camera, but both appear to be working.
The unusual sight of rain and wet equipment at the M2HATS site
Sunny this morning with light winds this morning, then increasing southerlies and clouds later.
First day of operations although we are still tweaking the systems. Two soundings (at 10:30am and 3pm or 1730 and 2200UT) which doubled as further training for the students.
After changing some settings on the Metek Halo lidar it now looks much more comparable to the UVA Halo. I ran all three lidars (the two Halos and the NCAR Windcube) in similar modes for a few hours to compare performances. During this time the lidars were all pointed vertically most of the time, but with VAD wind scans once every 10 minutes. From approx 1840 - 2040 UT, the two Halos were in the same 30m range gate modes, and the Windcube in 50m mode. From 2040 - 2223 UT the two Halos stayed in 30 m mode, and the windcube was in 25m mode. The Windcube was in the same 6-point VAD scan mode as the Halos for winds. Overall the UVA Halo and the Windcube had similar performance, typically seeing up to about 2.5 km, whereas the Metek Halo saw up to about 2km. The Windcube showed more definition in the backscatter than the Halos.
I did an azimuth calibration on the Windcube using the 30m ISFS telescoping tower which is at about 576 meters range, 199.1 deg azimuth from the MISS.
The lidars were set into their operations modes from about 23UT. These are:
NCAR Windcube: continuous VAD winds scans at 35 deg elevation followed by hourly short vertical stares, N-S and E-W RHIs, and 0 elevation scan over the study area (from east to south).
Metek Halo: continuous stares to the east along th ISFS tower array, followed by hourly VAD winds
UVA Halo: continuous vertical stares, followed by hourly VAD winds
Other issues, the ISS1 Gill winds were 180 deg out, which Gary fixed this afternoon.
The MISS wind profiler still has issues with the north and south oblique beams.
Screenshots from the NCAR Windcube, UVA Halo and Metek Halo lidars
A warm day with variable winds and some scattered clouds in the afternoon. A few dust devils, mostly to the south.
David, Isabel, and Josh (along with mainly of the ISFS crew) left early this morning - many thanks to all for a productive setup. It's now just myself on the ISS side and Steve, Justin and Gary working on the ISFS side (and the RSF crew). Most instruments are operating although there are still some communications issues and I had a somewhat frustrating day trying to get the lidars and cameras online. With Gary's help we got the lidar camera on-line (it's on the MISS LAN and for some reason it kept coming up on the wrong subnet and wouldn't accept a static IP address). This is a Microseveen camera which will allow us to remotely monitor what the lidars are doing (eg, where the heads are pointing). Another Microseven camera on the ISS1 tower pointed to the southeast over the study area did not have the same issue. I hope to install another Microseven camera on the ISFS tower array.
A bigger issue was with the Halo lidars. For some reason, the teamviewer app normally used by these lidars isn't working on the ISS network so the only way to control them has been by directly connecting a monitor, keyboard, and mouse. Eventually I was able to get Microsoft remote desktops working for both lidars and then by finally being able to compare them side-by-side discovered that the Metek Halo lidar is showing significantly less signal than the UVA Halo and our Windcube. I will investigate further tomorrow, it is not clear yet if the reduced power is due to a different scanning or sampling strategy or if there is a problem with the lidar.
Remote desktops for the Halo lidars, the UVA lidar on the lower left, and the Metek lidar on the upper right.
The Microseven camera installed south of MISS (left) and its field of view (right) which is designed to view (from left) the NCAR windcube, the Meek Halo (on MISS), the UVA Halo, and the Chico State REAEL lidar.
A Microseven camera on the ISS1 met tower looking southeast (left) and its field of view (right) looking out over the main study area south of the ISFS tower array.
Another hot mostly clear day with light variable winds and a few dust devils. Some scattered cloud, especially in the afternoon.
The setup is nearly done. Josh and David did some more work on the clutter screen for the MISS wind profiler, adding a mesh screen on the south side and reinforcing the east and west sides. David got back under the antenna to reinstall the backplane cover that we took off the relay segment of the antenna while diagnosing connections to the beam steering relays. The clutter screen does appear to have helped reduce clutter on the lowest range gates, although there does still seem to be an issue with the wind directions. I installed a Microseven camera and a Purpleair aerosol sensor on the ISS1 tower, which required a little reworking of the power and network connections since the DMS box was too crowded to fit everything, so installed another enclosure on the tower.
We did another sounding, both as training for the students and to test out the balloon inflation enclosure (which we are using to gain experience for a possible switch to hydrogen in the future). Winds were much lighter today and everything went well.
Prof Tong and his students learning how to launch soundings
Group shot of EOL staff, Jielun, and the Clemson group
A hot day with light winds and occasional dust devils (including one that almost went directly. over the site, see below), and light variable winds. Clear during the morning, scattered clouds later. The ground at the site was damp this morning, and some of the heavy storms that hit the town last night did pass over the site, and showed up clearly in both radar wind profiler data.
Examining the MISS profiler data from last night, indicates that the antenna beams of the profiler are mostly okay. During clear-air periods, there does seem to be a lot of ground clutter so we decided to attempt a simple screen made from chicken wire. So far the fence just covers the east and west sides, and does appear to improve the data, although so far we have only had light winds, it will be interesting to see how the data looks with stronger winds.
We also ran a check on the CS125 visibility sensor using a calibrated filter. We didn't have access to Campbell's software for the test, but just put the filter into the CCS125 beam for several hours. Visibility values of 940 to 970 meters were recorded during that period, which agrees well with the filter's calibrated value of TMOR 997 meters. We also set up 100-meter winds and RASS modes for the Modular Profiler, worked on a camera to monitor the lidars, and tested a backup generator.
Buildding a clutter screen for the MISS 915 MHz wind profiler (left) and checking the calibration of the CS125 visibility sensor (right)
A dust devil went directly over the ISS site about 21:20 UTC (2:20 pm local) and may be visible in Modular Profiler and lidar data.
Slightly cooler today with a high of 91F and windy at times, especially during the morning. Winds were mostly northerly, although switched to westerly at times as some storms went along the hills to the west. Variable clouds increasing in the afternoon. During the evening there were some storms with heavy rain and lightning in the town and some of those may have passed over or near the site.
Josh and Isabel launched the first sounding this morning. There were westerly winds around 12 m/s (26 mph) and based on that experience we will likely recommend the students don't launch in winds stronger than 10 m/s (especially if they are northerly as winds might blow the sounding into the power lines on the south side of the site). Given the winds, Josh inflated the balloon inside a seatainer rather than the sounding shelter we had planned to use. Aside from the wind, the sounding went smoothly. The data collection and processing appeared to go well, with two BUFR files generated (for upward and downward legs), although we aren't yet sure if the data were submitted to the WMO GTS.
David and I worked on the MISS profiler. David took the cover off under the antenna and traced the switching cables. He checked connections and verified that the switching signals appear to be making it to the relays in the antenna. Gary and Isabel worked on installing the NIMA analysis package which will help diagnose the issue. On the Modular Profiler, the oversampling mode we set up yesterday appears to be working well. Today we set up a 100 meter RASS mode which seems to be working well.
We leveled and tied down the NCAR windcube and Metek Halo lidars, and added a safety chain to the lidar platform on top of MISS. Josh added a ground rod to the ISS1 tower (just in time for the lightning this evening). Isabel got the data transfer back to Boulder working although there is further work to do before web plots will be available.
Inflating and launching the radiosonde sounding
One on the lidar platform on MISS (left) and diagnosing the MISS wind profiler (right)
Another warm day with a high of 96F with winds mostly southerly, turning to northerly late in the day with increasing clouds. A brief rain storm in town late afternoon and lightning nearby, but only a trace reached the site.
Today was John's last day and he continued to help us with the wind profilers. This morning we erected a clutter fence for the Modular Profiler. The profiler is already in a chainlink fence compound which does provide some screening (and protection for anyone nearby), but we have found that installing sloped chainlink fence panels along the sides of the profiler (parallel to the polarization) does provide an effective clutter screen so we added that to the profiler here. We also installed acoustic surrounds for the RASS speakers. There was some improvement in the lowest range gates, and we are now getting wind measurements down to about 400m, compared to about 600 before. There is also a few dB reduction in interference to the sounding system. David set up a series of operating modes such as RIM and oversampling modes which we will test over the coming days, and John wired the RASS distribution. We also setup a shade cloth over the amplifier enclosure to keep it cooler.
On the MISS 915 MHz wind profiler, there appeared to be a problem with the antenna steering. Doppler spectra from opposing oblique beams appeared not to be the mirror images expected, suggesting that the antenna beam was not being steered correctly. The winds also appear to be rotated compared to the lidar and 449 MHz profiler. John checked the switching signals on the antenna cable with a multimeter and the beam steering codes seem to be correct. Reseating the cable appeared to help although we only had a short time to check it so we will do some more tests tomorrow.
Other tasks included tidying up cables, straightening the 10-meter met tower (it had been leaning a couple of degrees to the south) and tying things down in case of gusty winds (the nearby airport AWS reported gusts over 40 mph). Isabel got the VAD wind algorithm working for the windcube lidar, and she and Josh also worked on configuring the sounding system. We also paid a visit to the ISFS array of 50 towers, very impressive!
John departed and will soon will be heading to a new role in RSF, thanks very much for your help John!
Chainlink clutter fence panels on the 449 MHz Modular Profiler. Three panels were installed parallel to the polarization vector of the antenna.
Installing the RASS surrounds (left), diagnosing the MISS wind profiler (center), and visiting the ISFS tower array (right).
Slightly cooler today with a high of 98F. Early morning there was a light northerly with smoke from an industrial plant drifting down the valley, then the wind switched to southerly with a gust front mid-afternoon kicking up dust. Cloud gradually thickened during the day which provided a welcome respite from the intense sun.
John continued his visit checking on our radar wind profilers. On the 915 MHz wind profiler on MISS, he checked the power output of the amplifier and found that it was a little higher than expected at around 460 W peak instead of the 400 W expected. Out of concern that we might be overdriving the amplifier (which might damage it), he added attenuation to the input. It is now generating about 390W peak power or 56 dBm with 24 dB attenuation on the input.
On the 449 MH Modular Profiler, he and David discovered that the amplifier box required some minor repairs. One of the three amplifiers was disconnected (perhaps having shaken loose in transport) and other similar issues. This meant that the transmit beam was uneven, likely contributing to the low cross-correlations noted yesterday. After repairs, the profiler worked much better. We also checked out radio interference from the profiler on the sounding system and on the 5G wireless network in the REAL trailer. Tests showed that both profilers were not affecting the REAL 5G service. The 449 MHz profiler is affecting the sounding (as expected) so we may need to move the sounding antenna across the access road, although the clutter fence (which we plan to install tomorrow) will likely reduce the interference.
We installed the 3-meter tower just south of MISS with a Lufft WS800-UMB weather sensor and a Campbell CS-125 visibility and present weather sensor.
Checking the power level from the MISS 915 MHz wind profiler (left) and making repairs on the 449 MHz amplifier (right)
Installing a CS-125 visibility sensor (left) and checking interference on the sounding system (right)
A gust front is pictured approaching the site from the south around 2:30pm PDT. We adjusted the windcube lidar scanning strategy to observe the front and saw gusts around 12 - 15 m/s - 27 - 33 mph)