lol they use it to charge the ground that they stand on if there was a charge in the water of your tank this would be the same potential, grounding it out would carry any current away rather then doing what high voltage linemen try to do which is keeping themselves isolated and at the same potential.
current will travel when there is a different in potential, the greater the different the greater the risk in electrical shock. an example using a fish tank say there is a leak in the 120 volt line of the fishtank now your water will become charged to 120 volts, water is an insolator but not a very good one so it conducts electricity, but this water is sittign in a glass tank on a wooden stand which are both good insolators, keeping the voltage contained and the tank at equal potential to theline of the powersupply.now add a ground probe to this and you'll be taking this line to a 0v ground which will have a difference of 120v which will cause a current to be pushed through the water through the probe into your ground which gets carried to the ground bar in your panel which is then bonded to the neutral bar of the source and also bonded to the ground. this is a short and carry the the short circuit current of your breaker through the water, this current is 10000 amps for 1 second which equals death to your fish, by electricution and by the heat generated. now lets look at the theoretical concept of stray voltages in your tank from "dielectric" degradtion of the cable insulation introduction 3v into the water and having it grounded to this 0v ground will cause this voltage to want to move, when this happens its goign to want to push through the resistance of the water and the ground conductor which will generate a current that can ultimately kill the live stock. no i have not done the experiment to calculate the resistive value of whater, but i'm sure it varies depending on water conditions but if you want to calculate the current that can potentially be pushed just find the resistive value and plug it into this forumla i=e/r where i is the current e is voltage and r is the resistive value. the current will end up being low but there is still a current and if the condidions get to a lower resistive value to potential to increase the amperage is very high
the thing is its not the voltage the kills it is the current anything over 50milliamps is potentially leathal which is why a GFCI is designed to monitor current leaks that are a difference of 5milliamps
another not here is we are talkign about 120volts not 100 000 volt power lines that require stupid thick silicone insulated cables and still may cause slight leak of voltage through the resistance of the insulator
like i seriouslly can't believe i'm actually arguing about this.