[dave9t5]

Assuming: the ridge on the horizon is roughly parallel to the image plane and that a 28mm f/1.8 lens will be closer to a curved field lens than a flat field lens.



Assuming the ridge is 2km away at the center of the image, then it's almost 400m away at the very edge of the horizon:







That's going to be a lot farther out of the focus zone than at the centre.



The focus point on these images is probably less than 10m into the image. (The elevation sign, the end of the tarps.) Assuming the focus is <10m into the image, the DOF is still considerable into the background but maybe not infinite. Viewing the image on my screen, at 100%, the calculated DOF behind the focus point is:

Focus @9.0m -> 228m

Focus @9.2m -> 504m

Focus @9.3m -> 1266m

Focus @9.4m -> ∞



So, we can see that "acceptable" sharpness in these images is estimated to extend from approx 0.5km to several km into the background which overlaps with the range of the horizon in the images. So (while not definitive) it's conceivable that's what is perceived on this images.