do you depend on the heat from the hps? this could be relevant to some. all of this is relative to photoperiod plants. if oyu do autos exclusively, then you can take 66% of anything stated in regard to watts or umol/s etc, over an 18/6 cycle.
if you are buying new equipment, migiht as well go with LED. you'll definitely have the potential to get a much more even distribution of light than even 3 400 watt bulbs.
2m^2 so if you don't amp up co2 you need about 1800-2000 umol/s over that space to give solid 39-44 dli all over on a 12/12 schedule. It's good to be slightly over. can mitigate any slow dimming over time.
if you buy high-efficacy lights you can easily hit the advertised 50,000 hours of use. less efficient led are cheaper but you wont get 50,000 hours of use out of them, either, on top of paying more watts per photon produced every month of use.
for 1x2 i'd recommend a diy light setup, but you could easily put a couple 480w lights in there and that'd even be able to make use of co2 over 12/12 schedule. 2 fc4800 would give you 2400 umol/s, which is a bit much. as far as best distribution a diy route is always best. can tailor shape of light to your exact needs.
gah, well that'd be 2733 umol/s -- you'd be good for 1300 ppm co2 and lab precise temp/rh, lol. So, those would be running at 70% in a 1x2 over 12/12 with ambient co2, give or take.They'd barely generate any heat at that level. if you had a breeze they'd easily stay below 25c at diode, which is the whole key to the 50,000 hours of use.
2 fc 3000 would be slgihtly underpowered and not as good of light distribution. ould be getting 36ish DLI in 2m^2. that's still potentially right up at max for ambient co2 and if not very near. the problem with this light is that it's only 580 mm x 520 mm. With bar-style led lights you get more light going out on 1 axis perpendicular to bars. at end of bars it won't spread out as far. So distribution would take a hit, but it'd cost half the price. it'd still be way better than 3 400w hps or 2-3 quantum boards.
540 vs 1000ish. though the 1000ish gets you prepared for maximizing any co2 you add, if ever going that route.
but, the math above is re-appliable to any led with accurate umol/s specifications. you'll want a PPFD of 800-1000 for ambient co2. 2m squared is easy math. you need twice that ppfd in umol/s. So 1600-2000 umol/s .. the better sized the frames are of the lights to cover the area the more optimal the fit.