Attribution of modern Andean glacier mass loss requires successful hindcast of pre-industrial glacier changes

Andean glaciers have been shrinking due to long-term climatic warming during the past 100 years. Stuart-Smith et al. (2021) used observations and numerical models to evaluate the anthropogenic contribution to the centennial retreat of the Palcaraju Glacier in the Peruvian Cordillera Blanca. According to their central estimate, the glacier retreat is thought to be entirely the result of the observed 1 °C warming since 1880 in this region, of which they consider 85–105% as human-induced warming. However, this attribution must be questioned because the numerical models used by the authors fail to replicate the well-documented Andean temperature and glacier history of the Common Era. In a recent literature synthesis we have demonstrated that Andean glaciers retreated significantly during the Medieval Climate Anomaly (MCA, 1000–1200 CE) when the vast majority of all South American land sites experienced a warm phase, recorded as a near-global natural event, that is not linked with human activity (Lüning et al., 2019a). The MCA was followed by the Little Ice Age (LIA, 1300–1850 CE) when many Andean glaciers advanced significantly, some of them even reaching their maximum Holocene down-valley extension. In contrast, the “hindcast” of Stuart-Smith et al. (2021) erroneously suggests hardly any glacier length fluctuations for pre-industrial times. Given the unsuccessful “hindcast”, we do not consider the attribution results of the study as robust.