Wow this thread suddenly came awake!
A lot of argument about nothing it seems.
99% of policing is about nothing much so to snipe at the air support branch for coming out to play occasionally seems way over the top.
The only servicable Air support in London was off doing training for the anti-terror role when Westminster Bridge happened. The incident took 80 or so seconds to play through and air support was nowhere to be seen and had to be brought in from Essex, Surrey and Oxfordshire. I think the first air asset on scene was the SKY tv helicopter and so the first downlink images into the police HQ were probably from that. Indeed the ground based firearms branch was also nowhere to be seen when it all went off. In this instance they were close enough in the event to take out the instigator quite quickly. Lucky it was Westminster!
But this all goes back to the generalities of policing. Policing, like the army and a whole lot of services, is 99% of nothing so that means that most of the time it provides 'bad' value. Bad value driving around in cars, on bicycles [2 by 2], and helicopter sitting around waiting for the passing terror suspect, bank robber, missing person etc etc that might, like Westminster, play out in less than two minutes.....
How can you audit a product like that?
Buying, staffing and fueling 100 cars that everyone is quite happy are driving around in circles possibly costs way more that a few helicopter sorties and [even if a drone could go beyond line of sight] has anyone priced the cost of support for such a device that is going to be flying round aimlessly for 99% of the time?
To support the availability of one tiny 'line of sight' quad rotor 24/7 will take a minimum of 3 officers and one vehicle per day and you may have to wait 2-3 hours for it to arrive along crowded roads. Move up the market to a real surveillance asset [currently illegal] will involve mobile control rooms extra staff etc etc and probably twin engines.....