Hi all, A little while ago we posted a question on how to track bumblebees in a relatively large greenhouse to both the Ecolog-L and Bombus-L list servers. We got numerous responses and we would like to thank you for the kind responses and suggestions. In the following we will present a short list or summary of suggested solutions.
Dr. Nic Malone suggested several approaches: (1) using fluorescent powder in combination with UV light and camera's. (2) to look into the combination of RFID and camera's (3) RFID on flowers and in sheet partitions in the greenhouse to create some sort of route. Camera's might be able to register the bumblebees, but resolution restrictions means that a number of camera's would have to be used. Aside from buying several expensive camera's this will require some software that can track the bumblbees (or, of course, some very patient observers). The idea to use RFID to create a route by, for instance, creating smaller transitions between space is good, but our interest is more into more natural flight patterns. John Stokely indicated that Ultra-Wide Band technologies for tracking, having great potential for tracking wildlife, might be useful. We have not found any useable tags using this technology. We did find some ultrasound motion tracking for use in computer interfaces. We are currently investigation whether this could be used to track bumblebees. Naoya Nishimura suggested: (1) coloring or marking bumblebees to allow good tracking using video. (2) Chemical marking of bumblebees in combination with detectors. (3) combination of laser sensors to trip camera's, which would at the very least allow registration of flower visits. Marking locations or using trap camera's is a good idea and would be an additional possibility if tracking proves infeasible. Coloring or marking bumblebees, as suggested by Nic Malon might make recognition easier, but does not solve the resolution difficulties involved in using cameras in such a large greenhouse.=20 Ben Letcher suggested an ingenious solution involving very long RFID antennas: ...recently we've been making very long 75-150m RFID antennas to place along stream banks. (...) you could create a grid of 40m-long antennas using multiplexing readers. I don't think you could do better than 1-m resolution at the very best, though. And I 'm not sure right now what the maximum height of the rectangular antenna would be, so you would be limited in the vertical plane as well (probably a max of 0.5 m). We do not know yet if we could apply this in our experiment. Evan Notman made another interesting suggestion concerning the tracking system used to follow hockey pucks based on small infrared transmitters and infrared camera's. Some googling showed that this tracking system for hockey pucks is the so-called FoxTrack system. It uses some quite ingenious hardware inside the puck as well as multiple camera's and a so-called puck-truck. The system is unlikely to be of direct use, but we thought maybe infrared could be used. We informed about the application of infra-red camera's. Infra-red would be capable of registering bumblebees by their temperature. The camera's needed to cover enough area, however, quite expensive, ranging from 10.000 to 40.000 euro plus, and may need special lenses of an added 10.000 to 40.000 euro..There is no real advantage therefore of using infrared camera's instead of normal camera's. Robert Dana pointed out the use of a diode and electromagnetic signals and indicated that Prof O. Talor at Univ. of Kansas would know more about this. We do no know yet how this system works. Finally, T.C Roth ii pointed to the use of harmonic radar on turtles, where a directional "gun" was used to find these turtles. Harnomic radar has, of course, been used before but the idea of a smaller and directional sender is very interesting and we will look into it. Many suggestions are still under consideration and we hope to put them to practical use.=20 Again we would like to thank you for the useful and kind responses, Frank van Langevelde and Tom Huisman Resource Ecology Group Centre for Ecosystem Studies Wageningen University P.O. Box 47 6700 AA Wageningen The Netherlands http://www.reg.wur.nl/UK/
