NOPP P.I. Meeting Minutes
February 28, 2000

FleetLink PI Meeting on February 28, 2000 at MIT Sea Grant

Absent: C. Pendelton

Guests: Peter and Linda Detweiller

We began the meeting, at 10:35am, with updates from the FleetLink Partners.

Cliff

Cliff discussed the PinOak option for telemetry from/to the vessel. It uses the single sideband transceiver and rides on top of the signal. It would be much less expensive than Inmarsat C. We will learn more about the system later today when we hear from Peter Detweiller.

Gary

Ken

We discussed some other posibilities to save transmission costs. For example, we could turn off Athena data transmission on the second test vessel and only send the tow and catch data. We could also replace hourly data with hourly averages. However, several people thought that during this demonstration phase, we should show off all the capabilities of the system.

Ken is refining the formats based on feedback from Bob and he is doing some PinOak testing at the same time.

All necessary computer hardware has been purchased. The newer units have touch sensitive screens. Ken speculated that the touch screen could replace the keyboard for data entry.

We had some further discussions about where the data would ultimately be sent to. For example, the National Weather Service (NWS) is interested in getting hourly values (not averages). However, Bob's initial conversation with them implied that they were not yet ready to accept data directly in digital form, such as via e-mail or ftp. Rollie asked what the Fish Coops would get. Gary suggested contacting the GTS people as they already deal with sending data to the NWS. [Gary contacted them after the meeting and put them in contact with Bob.] It was also suggested that we contact Argos since they have links to NWS as well.

Demonstration and Testing Strategry

1. All boats go to the same area to verify getting similar data results
2. All boats in the water at the same time, but in different areas to verify that the system can handle multiple transmissions
3. NMFS wants data. Do they want it now?
4. Send data to the National Weather Service in real time
5. Prepare list of variables and their accuracies [see Data Fields Collected prepared by Bob].
6. Contact Bob Weller at WHOI to see what data accuracies and data rates the NE Climate project would like to see. (Bob)
   [Here is his resonse:
   1 minute sample rate is what we do on buoys and ships.

   I am on the road but here is quick reply on accuracy/precision. Note that these are target accuracies to strive for, not what unfortunately is actually achieved in the field.

   quantity	accuracy	recording precision
   temperatures	0.01 deg C	0.001 deg C
   barometric pressure	0.1 mb	0.01 mb
   wind speed	2%	0.1 m/s
   wind direction	1 deg true	1 deg
   rel humidity	2%	0.1 %RH
   rain	1 mm/hr	0.1 mm/hr
   shortwave radiation	2%	1 watt/m²
   longwave radiation	2%	1 watt/ m²

7. Test two way communication between the fishing vessel and Fish COOP.
8. Complete gear-based testing and correlate with catch data
9. Look into FleetLink as an alternatives to the BoatTraks system. What changes would be necessary to make them compatible. What are the governing regulations? (Bob)
10. Look into the requirements for the data assimilative models. What data types and data frequency would they require?
11. Analyze the impact on NWS's weather predictions after receiving accurate and timely data from sea. Provide feedback to the vessels by sending them weather maps.
12. Contact Breck Owens or Jim Valdes (WHOI) about their experiences in the Orbcom project. They generated about 40Mbytes in six months. What were their costs? (Bob)

    [Jim Valdes said that the nominal rate for using the RF-based Orbcom system is $0.01/byte, or the same as Inmarsat C. There is a rumor that Orbcom will increase this to $0.03/byte. However, in their project they are charged a flat rate of $15 per month, but their data rates are much less than ours. To get the lowest rates, one needs to become a Value Added Reseller (VAR) and Jim is currently negotiating with Orbcom about this. He hopes to know in about 3 weeks what their rates will be. I've asked to find out what Orbcom would charge FleetLink for our data, exclusive of transmitting images. Jim also mentioned that the drifter people pay Argos $10 per day to transmit drifter data and they are looking into the Orbcom system as an alternative as well.]

13. Revisit adding a fluorometer sensor. The problem is fouling of the biomass sensor. Contact SeaKeeper.

PinOak Digital Corporation, Peter Detweiller, president

• Messages are uuencoded and sent as the body of an e-mail message.
• The Mass. Maritime Academy uses his system
• All of their messages are compressed using various compression schemes, including Hoffman encoding.
• They have partnered with Cesco (St. Paul) and use their specialized language in transmitting weather information. E.G. a 38K gif file showing weather is compressed to 2K.
• Their email cannot handle attachments. They felt that since many email systems routinely block email with attachments that their approach of uuencoding messeges into the text body was better.
• Their system did not cost as much to implement. Their up-front costs were much less than any satellite-based communication system. They can add more nodes as necessary.
• Their system takes advantage of "code gain", that is the software does the work of expensive hardware. Their "box" can get 18 decibels of code gain so a 150 watt marine transmitter is the equivalent to a 6.8Kwatt transmitter.
• There goal is to provide communication from/to any where in the world.
• They are a common carrier. They have one site that receives all messages in the 2-30MHz radio range. They expect to open several systems in Asia and the West Coast and expect to get more frequencies.
• The ICOM 710 radio is e-mail ready.
• They use a 500Hz band, and can take advantage of 27 channels that are 1KHz apart.
• Cost: $1,395 includes the modem, set up on their system, cables, manuals and telephone support. The radio host cost between $700 - $1,000.

We noted that the system requires human intervention to initiate. Software would have to be written to make it run automatically. Selecting the best transmission frequency is not automatic and depends on the time of day.

Northeast Consortium Proposal and other Funding

The NOPP renewal is in about one year from now. We could submit a follow-on proposal then.

The next meeting will Monday, March 27, 2000, at Craig's boat in Portland.