Homestake Meeting Minutes, 04/03/14
Attending: Riccardo, Tanner, Vuk, Eric, Shivaraj, Victor, Gary, Michael
Homestake wiki: https://zzz.physics.umn.edu/groups/homestake/home
1) Homestake update (Vuk)
Vuk: iterating with SDSTA to formulate a subaward to them, so that they could start the work on preparing the stations and installing the optical fibers. Nearing the submission of the request in fastlane, hope to have this sorted out next week.
Gary: surface computing?
Vuk: plan to have two computers in the computer room at the surface. This was the mode of operation in the past.
Vuk: Thinking about having two GPS distribution systems. A surface station with a master-mode digitizer near Ross, convert the GPS signal to optical there, send it down the Ross shaft and branch off at 300, 800, and 2000. Do similar near Yates, and supply 1700, 4100, and 4850. In this way we minimize the number of splits, and solve the problem that some of the levels are not accessible from both shafts.
Gary: Could/should set up stations in the shaft buildings, use them as masters for GPS and record the local seismic data that will be corrupt. Want the optical link to be strong and reliable, so important to have the link inside the building. But will need separate two surface stations, far from the shafts.
2) UMN test setup (Tanner)
Tanner: Liteway sent us the demo equipment, including a transmitter (converts digital NMEA and 1PPS signals from the master digitizer to optical), fibers, repeaters (that allow split-offs), and receivers that convert the signals back to electrical for the slave digitizers. Currently does not work, the repeaters were not set up properly, Liteway is sending us new units.
Vuk: Optical Zonu has sent us transmitter/receiver for the raw GPS signal, which worked well (Tanner: 10% loss in signal-to-noise). But worry that multiple splits would reduce the signal to noise ratio and be susceptible to local noise.
Gary: what about the power consumption in this mode? Probably there will be power advantage.
Vuk: Will have to test this, plan to put the whole system on a battery as soon as we verify the system works.
Tanner: probably will be less. In the usual (master) mode operate the receiver 10min every hour, but for the slave mode we don't have to.
Shivaraj: could use the existing GPS system (aLIGO) as a check that the new system works. Could do this only at certain stations, not all stations have the GPS signal.
3) Analysis of existing data
a) Frames/miniSEED files (Shivaraj, Tanner, Michael)
Shivaraj: frames are ready at UMN, but have not been transported to Caltech. Have to re-gain write directory from Stuart to make the copy.
Eric: may want to cc Greg Mendel, he may be able to respond faster as Stuart is very busy this week.
b) Wiener filtering (Michael, Jan)
Paper circulated within the LIGO collaboration and submitted to CQG. Michael will circulate it to the Homestake group.
c) Estimation algorithms (Shivaraj, Eric, Noah)
Eric: due to rock density and elastic modulus changes with depth, there is a change in the index of refraction, which can result in waves changing directions with depth. Looked at this with Victor, posted some preliminary results (for S-waves):
The speed of the wave changes with depth, v(z) = v0 (z/z0)^alpha. Assume z0 ~ 1 km, alpha ~ 0.27, z ~ 2 km. Assume a sudden speed transition at some depth, and apply the Snell's law. For incominng angle of 45 degrees, the change in propagation direction is about 13 degrees. Want to improve on this model (smooth change in v, rather than sudden transition), and assess implications for the radiometer method. Expect this will blur the radiometer maps at 13 degree level, according to this toy model. Also will try to get a more realistic model for speed vs depth at Homestake. Can also think about how to include this into the radiometer algorithm.
Vuk: how does this change for P-waves?
Victor: larger velocity, but alpha should be similar, and the relative angles would probably be similar.
Gary: high-frequency modes get constrained by exactly this effect to propagate only in horizontal direction.
Eric/Vuk: expect the surface waves and also body waves to be present in our measurement (which is dominant and at what level could be frequency dependent). Want to separate them with this radiometer tool, and the situation will be different at different frequencies. So want the tool to work in different situations.