Extended Data Fig. 3: Baseline crustal production rate adopted in this study.

(a) Crustal production rates derived from the photogeologic analysis of volcanic provinces⁵⁷ from 3.8 Ga. To convert the photogeological analysis (expressed as the total volcanic emplacement in each geologic period from middle Noachian to late Amazonian) to the crustal production rate, we compare the rates derived using the age boundaries from the crater density⁶⁵ and the chronology model of Ivanov (2001)⁶⁶ and Hartmann (2005)⁶⁷, as well as interpolation methods using either step functions or mid-point averages. The labels show the total volcanic activity and the integrated volcanic activity in the last 2 billion years (in parentheses). The midpoint approach would introduce substantially more total volcanism, and so we use the step-function approach. Also in comparison is a volcanic history derived from earlier photogeologic analyses and used in the recent argon isotope study¹⁵. The step-function approach on the Hartmann (2015) chronology leads to a baseline model that is very similar to the one used in the argon isotope study¹⁵ in terms of the total and the recent volcanic rates. (b) Baseline crustal production rate adopted in this work, based on the global thermal evolution model⁵⁶ and the step-function interpolation of the photogeologic analysis of volcanic provinces⁵⁷, whichever is greater. We adopt the model using the Hartmann (2015) chronology as our baseline model, and consider the one using the Ivanov (2001) chronology as the variant. The two models have appreciable difference in the last 2 billion years.