For a course to be included in this research, there had to be aerial images of sufficient quality available within approximately 20 years of the course’s opening date and then on an interval of no longer than one image every 25 years through the 2010s. No more than one image per decade was analyzed, except in circumstances where a noteworthy change would have been missed otherwise. A few exceptions were made to these criteria if a course filled a key need in the sampling. All aerials were sourced from Google Earth or HistoricAerials.com. The full list of courses studied can be found in the tables at the end of this article.
AutoCAD was used to map and measure the images from each course. The work was performed by six mapping technicians, all with an extensive background in golf course architecture and expertise in CAD mapping. The key variables studied include:
- Golf course footprint (acres)
- Back tee yardage and footprint
- Total fairway area (acres)
- Size of practice areas (acres)
- Distance to the end of the practice range (yards)
- Practice area impact on footprint
- Total putting green area (square feet)
- Number and total area of teeing grounds (square feet)
- The number and total area of bunkers (square feet)
- Turn point – i.e., average distance from back tee to landing area hazards (yards)
- Distance added by new tees or moved greens (yards)
- The average distance between centerlines of holes (yards)
- The shortest distance from a turn point to a course boundary (yards)
Obtaining exact measurements from aerial imagery is limited by factors such as image distortion, image resolution, and challenges identifying features through shadows or trees. While measurements down to the square foot cannot be made with certainty, analysis of aerial images provides an excellent objective estimate of feature sizes and a reliable picture of how a golf course changes over time. In an effort to limit issues related to image quality, the mapping of each course began with the most recent image because the quality was typically the best. We then worked backward through the decades with the baseline of the newest image as a guide for feature locations and sizes where image quality was more challenging on older images.
Sample size is also a limitation when analyzing the 15-course championship case study and the Australian and Japanese course case studies. Trends within the championship course case study are only discussed when there is a high level of confidence in the findings based on the expertise of the USGA Green Section staff.
Modern courses have a larger footprint than older courses.
For the purposes of this study, footprint is defined as all playing areas of the golf course, all practice facilities, all native areas that are likely to require some maintenance, ponds and lakes, roads and paths, the maintenance facility, the clubhouse, and any dumping or staging areas that can clearly be attributed to the golf facility. Where a course had woodland borders, an approximation within the perimeter of the tree line was made to account for maintenance that likely occurs along and within the woodland margins.
In the 80-course sample, courses built during the three most recent decades had an average total footprint of 216.3 acres. Courses from the earliest three decades – the 1920s, 1930s and 1940s – had an average footprint of 152.3 acres, a difference of 64 acres. This pattern was also observed in the championship course case study, where the five most-recently opened courses had an average footprint 47 acres larger than the five oldest courses (260 acres versus 213 acres, respectively).