Homestake Meeting, 08/21/2014

Attending: Riccardo, Gwynne, Tanner, Vuk, Noah, Eric, Nelson

Homestake wiki: https://zzz.physics.umn.edu/groups/homestake/home

Our agenda:

1) Homestake update

1. planning next visit: Sep 21-27
   1. Vuk and Tanner planning to drive on 9/21, potentially bringing some of the supplies. Victor and Daniel will fly in, and probably rent a car. Gary plans to fly in on 9/21, could then drive with either of the two cars. Potentially stop by at Lowe's in Rapid City to get supplies. Mon-Wed expect to be underground for most of the day, complete 6 stations (all 4 at 4850, one at 1700, and one at 4100), and attempt a test of the power consumption at one of the underground locations. In parallel, Gary (with one more person) will work on the surface stations. Thu and Fri activities will be decided on the spot, will depend on the situation.
2. status of various orders:
   1. GPS units: ordered 25 transceivers and 10 splitters. Expect 6 week delivery, hoping to get a few units earlier than that, for tests during the next visit to Homestake
   2. fiber: ordered 13000', should be delivered within 1 week
   3. network components: ordered network switches, media converters, boxes to house equipment etc, expect delivery within 1 week
   4. poly panels: ordered ~50 panels, should already be delivered at Homestake
   5. granite tiles: still have not heard back from Homestake, will probably purchase several at MSP
   6. anything for surface stations? Will speak with Gary
3. status of various documents: Liability insurance certificate (submitted from UMN, Victor working on it at Caltech), submitted a version of the Experimental Planning Statement, still have to update the MOU.

2) UMN test setup (Tanner): setting up the equipment in the new lab, and writing down a procedure to streamline it for Homestake. Then will focus on streaming the data.

3) Analysis of existing data

1. Estimation algorithms (Shivaraj, Eric, Noah):
   1. Noah's S-wave study: corrected the error (different velocity used in injection and recovery), and reproduced different tests discussed at the last meeting. Correctly recover injection position, shows strong dependence on scale of the array (or equivalently on the frequency of the wave), as well as on the array configuration. Changing polarization can change the shape of the recovered blob, but the blob is typically well centered.
   2. Noah's P-wave study: found that averaging over recoveries at different frequencies improves the recovered position (have to check if this is dominated by the lowest frequencies), while averaging over longer periods of time yields the same checker board patterns.
2. Analysis of Sweetwater Array data: Michael's pdf: Since the Homestake array is expanding, wanted to check how the Wiener filter algorithm would work with it. So using a surface-based array at Texas, which is significantly larger than the initial array at Homestake, and with larger separation between seismometers (few km). Find that coherence is lower than at Hoemstake due to the larger detector separation, but the coherence is consistent with a random Rayleigh wave field, although with a non-negligible spread probably due to the local topography. Measured velocity, found it to be ~3 km/s at microseismic frequency, which is expected. Applying the Wiener filter algorithm, get 2-3x suppression but should be able to do better - expect the problem to be with inversion of large matrices, a problem that will also show up with the expanded Homestake array.