There's a wealth of data in the Photometric Spec Sheet linked to on the product page. And I'm not familiar with the import of most of them, not having a photometric background.
But I think I can extract some useful data for a sample computation (which is likely expressed already in one of the graphs if I were more familiar with them). The table on the last page says that at a gamma angle of 60 degrees (where 0 = down and 90 = horizontal), the unit puts out 555 candela. (That's at a C angle of 0, I'm just going to use that column and the assumption that our point on the sidewalk is at the same C angle as the distance varies.)
So call the unit height 1m and consider a point on the sidewalk sqrt(3) m away. That will be at the angle 60 degrees, and the point will be 2m from the light source. Now we can divide the candela (lumens per steradian) by the distance squared to get the lux on a surface perpendicular to the ray from the light source. So at 2 meters the perpendicular lux will be 555/4 = 139 lux.
But the sidewalk is not perpendicular to the ray from the light source; it is in fact at 60 degrees from normal. So we further multiply by cos(60) = 1/2, and find that the bollard should produce 70 lux on the sidewalk at a distance of sqrt(3) m (in plan) from the bollard.
To go farther afield, the candela at an angle of 80 degrees is given as 256. Now the distance on the sidewalk will be tan(80) meters, and the distance from the light source will be sqrt(tan^2(80)+1). So the lux on a perpendicular plane will be 256/(tan^2(80)+1), and on the sidewalk it will be 256 cos(80)/(tan^2(80)+1). Or evaluating the numbers, at 5.7m away (in plan) on the sidewalk, the illumination on the sidewalk will be 1.3 lux.
At 70 degrees the candela is listed as 428. So repeating, we get that at a distance of 2.7 m, the illumination on the sidewalk will be 17 lux.
Cheers, Wayne