As someone else posted, there was another recent thread on this (maybe
for your "roof" spanning the gap between two ocean contianers?)
Any way.... your idea will work, just a mater of getting details
right.
Good surface coverage would be important.
As HBub mentions, mineral build up could be an issue.
At your cost of water, it's not worth recycling but the runoff could
be reused as ag water for something so as not to totally waste it.
here is a place
http://en.wikipedia.org/wiki/Rutt's_Hut
that I saw water roof cooling more than 50 years ago. must have work
cuz they used it.
NJ is not particularly dry so you've go a huge advantage.
Water's advantage is due to ~1000 btu/lb of evaporation energy (8340
btu/gallon)
Evaporate 10 gallons per hour and you've got ~7 ton equivalent cooling
(of course that's on the outside) but that's still a lot of heat not
trying to get into the building.
Forget the naysayers & go for it.
If you started with a small section you could do some exerpeiemnts.
Measure (cheap water water meter) the amount of water applied.
Collect the water that runs. The difference would be close to the
amount evap'd.
Gallons x 8340 btu/ gallon would give oyu evap energy.
12,000 btu / hr ~ 1ton A/C
If you did you run for less than an hour you'd have to extrapolate the
"hour's worth of water".
Also it would be best to not start the experimental measurements until
the water was running for a while
and the metal surface was at its "running temp".
All these back of the envelop calcs are not highly accurate but
considering the cost of electricity, water & spinkler system parts,
they good enough to make the effort worth a try.
Would rust be a concern with the building?
cheers
Bob