I wish a proper study on this could get some real funding. The cited study is certainly better than nothing but it's based on a volunteer reported survey with significant response bias (eg. maybe the drop in bird sightings near oil and gas drilling is because people don't want to go birdwatching near an oil field).