The Provisualizer site looks neat and I'm impressed that he has mapped my home courses in Canada. However, I'd take the elevation data with a large grain of salt.
On my home course I checked the elevations for the first couple of holes that have some significant elevation changes and Provisualizer doesn't capture a 60 yard wide swale that is 2-3 meters deep. It also underestimates a fairly abrupt 10 meter rise on the second hole. I think this results from the inherent limitations of the SRTM 1 arc-second data. It is only accurate to somewhere around +/- 5 meters, there are anomalies in the data, and its resolution horizontally is around 30 meters. All of these make it questionable for such micro applications as elevation changes on a golf course.
Vis-a-vis Cypress Point I also couldn't find a high spot of 52m except in the trees next to the maintenance area by the 4th hole. I'm also skeptical of the minimum elevation of 2m - that's just about 7 feet above sea level. As an exercise I charted a path around CP on Google Earth and did the elevation profile (see below, some scrolling to the right required). The total distance was close enough. The min and max elevations were 5m and 44m as you can see in the two text bars above the graph. The total ascent came out at 186m which is significantly more than the Provisualizer's 152m. I'd guess that they are using the Google Earth data to do their calculations so you could ask him what could account for this discrepancy. I've never played CP, so it's hard to say if either method gives an accurate account of the elevation changes.
My home course comes in with about the same ascent total as Merion. I've seen Merion from the road and I know my home course and I am very certain that my home course has substantially more ascent than Merion, hence the large grain of salt vis-a-vis Provisualizer elevations data.