Tenacious Search

A report from Microsoft’s Jim Karkanias

Just-concluded phone call regarding coordination of remote sensing and trajectory generation.

Participants:

JPL: Yi Chao; Monterey Bay Aquarium Research Institute: Jim Bellingham; Center for Coastal Margin and Observation: Antonio Baptista; Oracle: Mike Olson; IEEE/Intel: David Tennenhouse; MSFT: Dave Campbell, Phil Garrett, Jim Karkanias

Items:

MSFT:  A log of cell phone connectivity indicates that Jim Gray’s phone was communicating approx every 15 min with Exchange from 5:08am pst to 11:50am pst. Voice calls from his phone at ca. 10:38 put him at mile 15/16 (based on his relayed report) at which point he began to lose cell connectivity for voice calls (based on his relayed report). His phone continued to sync until 11:50am putting him at an estimated 22 miles from GG - consistent with expected loss of cell signal at approx this point. An increase in both frequency and latency of retry attempts suggests (this is theory) degradation of cell signal supporting this westward trajectory. There are no cell towers on the Farallon Islands. An eye witness on Farallon Islands saw a sail boat at a time/place that matches this theory - his description matches the boat but he has yet to id by photo. An important outstanding question is whether the sails were up or not.

JPL:  Currently calculating trajectory data based on 5days of telemetry data from a buoy released by the CG during a search run. This data will be very useful in calculating a forward map of Jim’s boat if it is without a helmsman. This will soon be made available as a web service tool that can accept lat/long, etc. to forward calc a result

CMOP: Using particles based analysis applied to water parcels, the team will attempt to backward map a large space of possible endpoints to calculate the probabilities of paths that in the past would have intersected our proposed starting area (in the vicinity of Farallon somewhere between after where his cell lost signal, ie 22 miles out).

MBARI: Coordinating contacts across this project and will be setting up a collab website for this by 7pm pst

Oracle: Mike Olson and David Tennenhouse will be intersecting all of the above to create a bounding box for the MTurk and related image processing data. Georeferencing the images based on what the forward/backward maps suggest as probable paths should constrain the image processing task and help inform search.

This entry was posted on Sunday, February 4th, 2007 at 5:49 pm and is filed under Ocean Models. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.