The link by KuiN seems conclusive. The image is indeed a Photoshopped version of an older image from 2014. The question is why, but the stated explanation in th

The link by KuiN seems conclusive. The image is indeed a Photoshopped version of an older image from 2014. The question is why, but the stated explanation in the thread seems probable - they found the photo and wished to present it as their own, but it had a label in that lower right corner that would give it away. So they photoshopped in some sand to cover up that they hadn't taken it on-site. All very disappointing but not really surprising. The quantity of Photoshops in scientific papers is baffling, and Nature has been publishing a long string of nonsensical papers in the past years.