Thanks to jmac for his plot. And a comment attached:

Since I did not have accsess to the actual data used for the book's scenarios I reverted to "Graphology", i.e. replotted the non-renewable resources and then measured the slope. Certainly not as accurate as it could have been with the actual data, but I would estimate that the peak is good to about +/- 2.5 years. As you can see, I got a pretty good asymmetrical curve and peak with a steep fall-off after the peak. I also did the same for their scenario 2, double resources, and got a very broad, flat maximum that spanned 2020 to about 2050. I immediately thought of CERA's undulating plateau in the 2030-2040 period.

Very nice. Thanks Jmac!