Surface net solar radiation of ERA5-Land hourly data

ERA5-Land hourly data from 2001 to present,I choose the Surface solar radiation downwards。I don't understand why the ssrd-data is turned into w/m² and the value is still large. Is it because all the solar radiation amounts in the 0.1°×0.1° grid are combined?

same problem here. When I divide the J/m2 by 3600, my ERA5 (hourly) data is still too large. For instance, the long wave radiation (strd) has an average near 5000 W/m2. 

Alright, I think we all need to look at the conversion table provided in: Here. That solved my confusion/mistake in the conversion. 

The critical note: "Accumulations are from 00 UTC to the hour ending at the forecast step"

So in order to have the W/m2 for 05UTC, you need to subtract J/m2 of 04 from 05 and divide the number by 3600. Only for 01UTC you do not need to subtract from the previous time step.

an example:

if you look at J/m2 values of shortwave radiation for instance for 24 hrs, its something like this (scaled by 1e-4, and from 00 to 23UTC):
473.072511449194, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2.05968637480363, 35.0146683716601, 112.767829020493, 216.360691462317, 323.604816114019, 424.342204263501, 502.610286506035, 552.791736364885, 571.750213223872, 572.124701655654, 572.124701655654

(location: 45 lat, -73 long)

The reason why 00UTC is not "zero", its because it is accumulated from previous day, as the "critical note" notes.