Many peatlands around the world are eroding and causing carbon losses to the atmosphere and to freshwater systems. To accurately report emissions from peatlands we need to understand how much of the eroded peat is converted to CO₂ once exposed to the atmosphere. We need more direct measurements of this process and a better understanding of the environmental conditions that peat is exposed to after it erodes. This information will help quantify the emissions savings from peatland restoration.

Marine ecosystem models (MEMs) are valuable for assessing the threats of global warming to biodiversity and ecosystem functioning, but their predictions vary widely. We argue that MEMs should consider evolutionary processes and undergo independent validation. Here, we present a novel framework for MEM development using validation data from sediment archives, which map long-term environmental and evolutionary change. Our approach is a crucial step towards improving the predictive power of MEMs.