gnoitall wrote:ijuin wrote:The Moomin wrote:I'm intrigued by these police radar guns that are inaccurate by over 5 days, tell me more.
I'm pretty sure you're deliberately misinterpreting the statement, but I'll explain anyway. I meant that police radar guns frequently give an error of more than 2.0 knots when measuring the speed of objects--e.g. trees being reported as actually being in motion. As such, radar is not necessarily sufficiently accurate to reduce the margin of error for measuring a hurricane's path by a great deal below the current error margins.
To pick a mediocre analogy, that's arguing that snipers can't possibly hit targets at 1000 meters and beyond because a .38 Special snub-nose pistol isn't meaningfully accurate beyond 20 meters.
The capabilities of a police radar don't meaningfully reflect at all on the capabilities of a dedicated weather-surveillance doppler radar. According to the WSR-88d System Specification document (http://www.qsl.net/n9zia/pdf/2810000C.pdf), page 3-60, velocity accuracy "will be less than or equal to 1.0 ms-1"... i.e., less than 1 meter/second.
Pretty accurate. With all post-processing and quantization effects, I expect final velocity measurement accuracy to be +/- 10 m/s, though that's just a guess. But a very conservative guess.
I should hope that's a conservative guess, since it suggests that a weather-surveillance radar is literally an order of magnitude less accurate than a police radar (10 m/s = 19 kt).