A multi-model analysis of moisture changes during the last glacial maximum

This study investigates terrestrial moisture changes and associated mechanisms during the last glacial maximum (LGM; approximately 21,000 calendar years ago) using multi-model simulations from the Paleoclimate Modelling Intercomparison Project phase 3 (PMIP3). Considering that terrestrial moisture is not determined solely by precipitation, an aridity index (AI) is employed for measuring the terrestrial moisture by combining the effects of both precipitation and potential evapotranspiration (PET), where the latter represents atmospheric water demand and is greatly decreased mainly by the intense cooling during the LGM. Compared to the preindustrial period, the magnitude of global mean terrestrial moisture change is small, as the wetness brought by decreased PET counteracts the dryness induced by decreased precipitation. Regionally, the moisture changes depend on the different combinations of changes in precipitation and PET: (i) drying occurs where precipitation deceases and PET hardly changes, such as the northern tropical Americas and Southeast Asia; (ii) wetting is found in regions with precipitation increases and PET decreases (e.g., northwestern Africa and the central Andes), and their contributions are comparable; (iii) in particular, wetting can also occur in regions of decreased precipitation if a sufficient decrease in PET also occurs (i.e., southeastern North America and the northern and southern parts of eastern Asia), with the latter wetting effect reversing the former drying effect. The multi-model median field is consistent with available paleo-records in southern North America, the northern tropical Americas, the Andes, northwestern Africa, the southern Iberian Peninsula, southwestern Africa, the central part of eastern Asia, and Java but disagrees with proxies in Australia, central Brazil, southeastern Africa, the northern Iberian Peninsula, and the southern part of eastern Asia.