3 ms·

I mean it absolutely sounds like a great trade opportunity, if the numbers pan out.

yourapostasy 9y ago
Unfortunately the numbers are difficult to work out, because hunting lionfish is still predominantly a very labor-intensive activity. Some folks are working on remotely operated unmanned submersibles to factor out the labor. Also missing is a grid of underwater cameras feeding a stream to a back-end object recognition cluster to map out real-time lionfish locations to maximize kill utilization ratios (too much time is spent looking for them instead of killing them). Bootstrap enough of the industry, then build enough of the submersibles to permanently station around the most sensitive coral reefs and eventually entire islands and coastlines to "fence them off" from future infestations, and push them further out into the ocean where large predators can feast upon them as they have less cover to hide within, if the native oceanic predators can even eat them. There is some suspicion that in the invasive areas lionfish have no natural predators other than humans; sharks, groupers and eels can be trained to expect dead lionfish handouts from divers, but no one has yet documented them start to hunt the lionfish on their own in significant numbers. I would settle for pushing them out into the relative desert of the benthic ocean, and hope they starve out there. Fortunately, someone has already worked out a filleting robot for haddock, and modifying its pattern recognition for lionfish requires enough a training corpus to make it sufficiently reliable, so once killed they can be quickly prepared and frozen. The most recent research indicates lionfish aren't susceptible to Ciguatera. Unfortunately, lionfish toxins can also sometimes be found in their flesh, so they must be cooked (heat breaks down the toxins), and they can't be introduced into the lucrative sushi and sashimi markets (though I'm sure enterprising sushi chefs will find delicious ways to prepare the cooked lionfish flesh). Accuracy rates on automated classification of lionfish in live streaming video is so far not revealed in any write ups I've read, and especially false positives (killing fish that the classifier thinks is lionfish but actually are not). That's a significant problem, and also the obstacle that I can't work out to gamifying the killing of lionfish. The threat of the more capital-intensive automation of industrial-scale lionfish hunting/eradication also presents a challenge to locals who want to break into the market. Not much incentive for the locals to invest into opening the market when they know they could be out of a job within just a few years of establishing a lucrative market to well-capitalized (and often foreign) competitors who can tap more credit than the locals. Currently, economically beneficial lionfish eradication for the locals is primarily focused upon leading hunts for diving tourists to favored dive spots, and not for systemic, industrial-scale, broad eradication across the entire coastline. Personally, I believe leaving it at this is a mistake, because lionfish represent a threat to our fisheries over the longer term.

Unfortunately the numbers are difficult to work out, because hunting lionfish is still predominantly a very labor-intensive activity. Some folks are working on remotely operated unmanned submersibles to factor out the labor. Also missing is a grid of underwater cameras feeding a stream to a back-end object recognition cluster to map out real-time lionfish locations to maximize kill utilization ratios (too much time is spent looking for them instead of killing them). Bootstrap enough of the industry, then build enough of the submersibles to permanently station around the most sensitive coral reefs and eventually entire islands and coastlines to "fence them off" from future infestations, and push them further out into the ocean where large predators can feast upon them as they have less cover to hide within, if the native oceanic predators can even eat them. There is some suspicion that in the invasive areas lionfish have no natural predators other than humans; sharks, groupers and eels can be trained to expect dead lionfish handouts from divers, but no one has yet documented them start to hunt the lionfish on their own in significant numbers. I would settle for pushing them out into the relative desert of the benthic ocean, and hope they starve out there. Fortunately, someone has already worked out a filleting robot for haddock, and modifying its pattern recognition for lionfish requires enough a training corpus to make it sufficiently reliable, so once killed they can be quickly prepared and frozen. The most recent research indicates lionfish aren't susceptible to Ciguatera. Unfortunately, lionfish toxins can also sometimes be found in their flesh, so they must be cooked (heat breaks down the toxins), and they can't be introduced into the lucrative sushi and sashimi markets (though I'm sure enterprising sushi chefs will find delicious ways to prepare the cooked lionfish flesh). Accuracy rates on automated classification of lionfish in live streaming video is so far not revealed in any write ups I've read, and especially false positives (killing fish that the classifier thinks is lionfish but actually are not). That's a significant problem, and also the obstacle that I can't work out to gamifying the killing of lionfish. The threat of the more capital-intensive automation of industrial-scale lionfish hunting/eradication also presents a challenge to locals who want to break into the market. Not much incentive for the locals to invest into opening the market when they know they could be out of a job within just a few years of establishing a lucrative market to well-capitalized (and often foreign) competitors who can tap more credit than the locals. Currently, economically beneficial lionfish eradication for the locals is primarily focused upon leading hunts for diving tourists to favored dive spots, and not for systemic, industrial-scale, broad eradication across the entire coastline. Personally, I believe leaving it at this is a mistake, because lionfish represent a threat to our fisheries over the longer term.