Serving the Quantitative Finance Community

 
User avatar
quantyst
Topic Author
Posts: 0
Joined: June 4th, 2008, 5:08 am

Fish in a Lake

December 15th, 2008, 5:50 pm

QuoteOriginally posted by: rralphQuoteOriginally posted by: Traden4AlphaQuoteOriginally posted by: quantystHere's another crazy, but exorbitantly costly (and possibly a scifi), approach: dispatch a squadron (or army) of robotic submarines to photograph all the fish (several times over, if need be), have the photographs analyzed by a computer to ID each unique fish, thereby coming up with a number for the fish present at a particular point in time in the lake.Photographing all the fish is just the sonar solution in optical frequencies. Neither photography nor sonar will find camouflaged or mud/crevice-dwelling fish (e.g., flounder and eels).Another sci-fi solution...Why not just use your tricorder to measure the life signs within the lake? Tricorder, not having any identifiable mechanism about it, is just like magic. There is nothing magical about robotic submarines equipped with cameras, or sonars, anyway. Does anyone know how a tricorder might work? Not what it does, but how is it constituted?
 
User avatar
rmexico
Posts: 0
Joined: May 1st, 2008, 4:50 pm

Fish in a Lake

December 15th, 2008, 6:43 pm

Rralph's solution is elegant, but as an avid fisherman, I suspect that any fish that's caught and released is far less likely to be caught again (i.e., fish learn to avoid lures and things that look like hooks).
 
User avatar
Traden4Alpha
Posts: 3300
Joined: September 20th, 2002, 8:30 pm

Fish in a Lake

December 15th, 2008, 6:54 pm

QuoteOriginally posted by: quantystQuoteOriginally posted by: rralphWhy not just use your tricorder to measure the life signs within the lake? Tricorder, not having any identifiable mechanism about it, is just like magic. There is nothing magical about robotic submarines equipped with cameras, or sonars, anyway. Does anyone know how a tricorder might work? Not what it does, but how is it constituted?One approach to a tricorder that can detect "life" would be a scanner that detects the chemical compositions of things and than assesses the concentrations of highly reactive chemical species. A strongly non-equilibrium mixture of chemicals would suggest the potential for life. Obviously, such as device would be prone to error -- a flash-frozen mouse might erroneously trigger the detector and a 2,000 year-old date seed might erroneously not.
 
User avatar
quantyst
Topic Author
Posts: 0
Joined: June 4th, 2008, 5:08 am

Fish in a Lake

December 15th, 2008, 9:11 pm

Here's another science de jour method for finding an approximate number for fish in the lake. Every living thing leaves off genetic signatures in its environment. We can take samples from the lake water to capture the genetic material left off by the fish, assuming such genetic material is sufficiently mixed in the lake. We then ID the stuff in the sample and somehow compute the number of unique fish genes. Any comments?
 
User avatar
Traden4Alpha
Posts: 3300
Joined: September 20th, 2002, 8:30 pm

Fish in a Lake

December 15th, 2008, 11:35 pm

QuoteOriginally posted by: quantystHere's another science de jour method for finding an approximate number for fish in the lake. Every living thing leaves off genetic signatures in its environment. We can take samples from the lake water to capture the genetic material left off by the fish, assuming such genetic material is sufficiently mixed in the lake. We then ID the stuff in the sample and somehow compute the number of unique fish genes. Any comments?This could be tricky because the most prevalent type of fish DNA would be fish sperm (in breeding season, at least) because fish fertilize their eggs in open water. (Think about that the next time you swallow some lake water!)This fact makes the estimation process harder because sperm contain genetic material that is a mixed half copy of the parent's DNA (see "crossover" in meiosis). This means that each sperm would have a unique genetic signature as if it were a unique individual. A simplistic analysis might conclude that the lake contains as many fish as it contains sperm. Collectively, the myriad sperm from a single fish would have very limited genetic diversity, so collecting enough samples might enable tracing the sperm back to the parents.A second complexity would be "founder effects" in which a very small number of fish inbreed to create a large population. A lake might contain 10,000 fish, but if all those fish were descended from only 10 original founders, then an analysis of the genetic diversity might give an answer of N=10.I'm sure there are magical statistic tools to handle these problems, but it might require rather a lot of data.
 
User avatar
rralph
Posts: 0
Joined: November 30th, 2005, 1:16 am

Fish in a Lake

December 16th, 2008, 9:56 am

I'm extremely surprised that anyone would try to argue that "a fleet of robotic submarines" and a computer that has the capability to identify and distiguish between individual fish are easier to implement than a random sample of the population. The concept of genetically sampling the water to identify the genes of every species AND INDIVIDUAL present in a lake is such an outlandish notion that it surely belongs in books of some new religion.Commercial fisherman are quite adept at catching fish. Is it so hard to believe that with the necessary equipment and a little patience that I could catch a few fish from a lake? I think I could do that faster then you could set up your gene sequencing factory. Catch and release lakes are very popular in the UK, the owners of the lakes are quite strict about their policy because the same fish DO get caught a second, third and even forth time.
Last edited by rralph on December 15th, 2008, 11:00 pm, edited 1 time in total.
 
User avatar
quantyst
Topic Author
Posts: 0
Joined: June 4th, 2008, 5:08 am

Fish in a Lake

December 16th, 2008, 11:28 am

QuoteOriginally posted by: rralphI'm extremely surprised that anyone would try to argue that "a fleet of robotic submarines" and a computer that has the capability to identify and distiguish between individual fish are easier to implement than a random sample of the population. The concept of genetically sampling the water to identify the genes of every species AND INDIVIDUAL present in a lake is such an outlandish notion that it surely belongs in books of some new religion.Commercial fisherman are quite adept at catching fish. Is it so hard to believe that with the necessary equipment and a little patience that I could catch a few fish from a lake? I think I could do that faster then you could set up your gene sequencing factory. Catch and release lakes are very popular in the UK, the owners of the lakes are quite strict about their policy because the same fish DO get caught a second, third and even forth time.True, true, true!No one, not even I, was suggesting that the use of robotic submarines along with computer identification of the pictures taken, etc., were easier or faster than what you originally proposed, with which, by the way, I agreed quite well as "Great description". As indicated earlier, my interest in the question was to find out novel but 'decent' ways to answer the question. The purpose of the question is not to implement practically a best solution. The purpose is to encourage a discussion of some serious ideas around a particular problem.
Last edited by quantyst on December 15th, 2008, 11:00 pm, edited 1 time in total.
 
User avatar
rralph
Posts: 0
Joined: November 30th, 2005, 1:16 am

Fish in a Lake

December 16th, 2008, 12:55 pm

I apologise, I may have come across a bit strong. Novel ideas should always be welcomed. Personally I am turned off by ideas which strike me as over engineered, but that is a personal preference.How about this:Dig a large holeAllow the lake to drain into the holeCount the fish as they go pastIf you are ecologically minded, then you can pump the water back into the original lake afterwards.