Weathering, alteration and reconstructing Earth's oxygenation.

Deciphering the role-if any-that free oxygen levels played in controlling the timing and tempo of the radiation of complex life is one of the most fundamental questions in Earth and life sciences. Accurately reconstructing Earth's redox history is an essential part of tackling this question. Over the past few decades, there has been a proliferation of research employing geochemical redox proxies in an effort to tell the story of Earth's oxygenation. However, many of these studies, even those considering the same geochemical proxy systems, have led to conflicting interpretations of the timing and intensity of oxygenation events. There are two potential explanations for conflicting redox reconstructions: (i) that free oxygen levels were incredibly dynamic in both time and space or (ii) that collectively, as a community-including the authors of this article-we have frequently studied rocks affected by secondary weathering and alteration (particularly secondary oxidation) while neglecting to address the impact of this alteration on the generated data. There are now multiple case studies that have documented previously overlooked secondary alteration, resolving some of the conflicting constrains regarding redox evolution. Here, an analysis of a large shale geochemistry database reveals significant differences in cerium (Ce) anomalies, a common palaeoredox proxy, between outcrop and drill core samples. This inconsistency provides support for the idea that geochemical data from altered samples are frequently published in the peer-reviewed literature. As individuals and a geochemical community, most of us have been slow to appreciate how pervasive the problem is but there are examples of other communities that have faced and met the challenges raised by such quality control crises. Further evidence of the high potential for alteration of deep-time geochemical samples, and recognition of the manner in which this may lead to spurious results and palaeoenvironmental interpretations, indicate that sample archiving, in publicly accessible collections needs to become a prerequisite for publication of new palaeoredox data. Finally, the geochemical community need to think about ways to implement additional quality control measures to increase the fidelity of palaeoredox proxy work.