FINAL REPORT LUQUILLO LTER

 

The wireless network this project set up in October, 2000 is still operating satisfactorily in October, 2002. That is two years of continuous operation by 6 radios, connecting three remote data logger sites in the rainforest, to both the El Verde Field Station and the Sabana Work Center of the US Forest Service.

It was expected that the continuous rainfall on El Yunque rainforest and associated deterioration from the jungle environment would, through corrosion and other effects disable one or more of the radios and their associated power supplies and cables. But they have survived.

The three meteorological stations, on towers reaching above the forest canopy at Bisley, El Verde, and Pico del Este are always on, therefore reachable from either the El Verde Field station, or Sabana by Windows computers connected by serial cable to base station Freewave radios, and running PC208W software at any time. Except for periodic link drop caused by intense rainfall, particularly between Sabana and the relay radio on top of Pico del Yunque - the most marginal of the links - researchers physically at either El Verde station or Sabana have been able to access the weather data at any time.

And they have been doing that routinely.

 

NET CONNECTIONS

 

Because, in spite of efforts by the Luquillo LTER PI to get a full Internet connection from the El Verde field station to and through the University of Puerto Rico in San Juan, there has been no such connection, it was not possible to connect the weather stations to web pages on the Internet, or to access the data loggers via the Internet during the 3 year term of this project. There have been plans for a net connection to be completed by December, 2002. But that is no guarantee it will happen.

In anticipation of an eventual connection, this Wireless NSF project funded Andrew McFadden, who works as a technician for both the LTER and the ITS (Institute of Tropical Studies at the University) and is stationed at the University in San Juan to configure an NL100 serial to Ethernet and IP device and a Unix server running a modification of the Pearl programs first written by Crane Johnson and Ken Irving of the University of Alaska to fetch and post data logger data to a web site. That work was done, and tested in San Juan. It simply awaits a full Internet connection to El Verde.

We learned also that a Puerto Rican television cable company is able to run a cable to the boundary of the Forest - but no further - which would give the closest commercial Internet access at a structure outside the forest boundary, or perhaps at the 'Stream House' just on the border.

McFadden has designed a test of running an 802.11b link from that entry point into the Internet via commercial cable, to the closest tower to El Verde, then to create an 802.11b 'cloud' over the field station and the cottages occupied by researchers at the station. So they can, either from the computer office of the station, or using their own laptops with 802.11b wireless cards, send and receive e-mail and access the Internet from El Verde. Even now the only communications from or to the El Verde station is one (marginal) cell phone.

It is very problematical that a pair of 802.11b 2.4ghz, 100mw radios will be able to connect with each other through the jungle foliage the 1.3 miles between the El Verde Tower and the Stream House, even with 2.4ghz amplifiers at each end and 18 or more dB gain directional antennas. 2.4ghz radios have very poor tree or foliage penetrating characteristics. And wet leaves - endemic to the rain forest - further deflect the radio signals at that frequency level.

This project, as of the closing date of this NSF Project, November 30th, 2002, will have funded the Luquillo LTER for test 802.11b radio and associated equipment to try and establish that link. They will try out one of the Teletronics 2.4ghz to 915mhz converter amplifiers to penetrate those trees if the direct links at 2.4ghz fail. This experiment will not be completed before this report has to be closed out. However we will attempt, informally, to post the results of this effort to finally link El Verde, its researchers, and the data stations to the Internet, wirelessly, on the http://wireless.oldcolo.com web site after this report has been circulated to all LTER scientists via mailed CDs.

 

OTHER PROJECTS

 

For a variety of reasons, some for lack of follow through by the Forest Service which initially requested it, some for forest trail closings, we never were able to link the sounds of the rare Richmondi sub-species of Puerto Rican Coqui to the El Verde Field Station, much less out to the Internet.

With the work that has been done in late 2002 by this project in support of Dr. Stuart Gage's Environmental Sound research near Lansing, Michigan to link environmental sound by wireless links to the net, we now know how to do that in the rain forest in the future.

Likewise we were, for lack of a field link to the Internet from El Verde station or the nearby Stream House, unable to make a continuous wireless video link from the Fresh Water Shrimp on the Rio Espiritu Santos to Utah State University and Dr. Todd Crowl whose 10 year research involves these shrimp, who must be visually monitored or manually video taped.

FINAL JUDGEMENTS

The establishment of a wireless network linking Data Logging meteorological in the Rain Forests of Puerto Rico was an unqualified success as far as it went. All long term data stations were, and continue to be, linked wirelessly to researchers at the El Verde Field Stations, reducing the amount of manual labor required to fetch the data periodically. The radios and there associated installed equipment have proved quite reliable and robust. Such a network can easily be expanded

Lack of an Internet connection, long promised but never delivered, precluded other wireless experiments, particularly ones involving biological research species. That lack also precluded the sharing of data from the weather stations beyond the forest itself, with remote researchers.