Environmentally transformative human use of land accelerated with the emergence of agriculture, but the extent, trajectory, and implications of these early changes are not well understood. An empirical global assessment of land use from 10,000 BP to 1850 CE reveals a planet largely transformed by hunter-gatherers, farmers and pastoralists by 3,000 years ago, significantly earlier than land-use reconstructions commonly used by Earth scientists. Synthesis of knowledge contributed by over 250 archaeologists highlighted gaps in archaeological expertise and data quality, which peaked at 2000 BP and in traditionally studied and wealthier regions. Archaeological reconstruction of global land-use history illuminates the deep roots of Earth's transformation and challenges the emerging Anthropocene paradigm that large-scale anthropogenic global environmental change is mostly a recent phenomenon.One Sentence Summary: A map of synthesized archaeological knowledge on land use reveals a planet largely transformed by hunter-gatherers, farmers and pastoralists by 3,000 years ago.

We show that the Li/Mg systematics of a large suite of aragonitic coral skeletons, representing a wide range of species inhabiting disparate environments, provides a robust proxy for ambient seawater temperature. The corals encompass both zooxanthellate and azooxanthellate species (Acropora sp., Porites sp., Cladocora caespitosa, Lophelia pertusa, Madrepora oculata and Flabellum impensum) collected from shallow, intermediate, and deep-water habitats, as well as specimens cultured in tanks under temperature-controlled conditions. The Li/Mg ratios observed in corals from these diverse tropical, temperate, and deep-water environments are shown to be highly correlated with temperature, giving an exponential temperature relationship of: Li/Mg (mmol/mol) = 5.41 exp (À0.049 * T) (r 2 = 0.975, n = 49). Based on the standard error of the Li/Mg versus temperature correlation, we obtain a typical precision of ±0.9°C for the wide range of species analysed, similar or better than that of other less robust coral temperature proxies such as Sr/Ca ratios.The robustness and species independent character of the Li/Mg temperature proxy is shown to be the result of the normalization of Li to Mg, effectively eliminating the precipitation efficiency component such that temperature remains as the ⇑ Corresponding author at: Istituto di Scienze Marine (ISMAR), Consiglio Nazionale delle Ricerche, Via Gobetti 101, 40129 Bologna, Italy. Tel.: +39 051 6398913.E-mail address: paolo.montagna@bo.ismar.cnr.it (P. Montagna). 1Published in which should be cited to refer to this work.http://doc.rero.ch main controller of coral Li/Mg compositions. This is inferred from analysis of corresponding Li/Ca and Mg/Ca ratios with both ratios showing strong microstructure-related co-variations between the fibrous aragonite and centres of calcification, a characteristic that we attribute to varying physiological controls on growth rate. Furthermore, Li/Ca ratios show an offset between more rapidly growing zooxanthellate and azooxanthellate corals, and hence only an approximately inverse relationship to seawater temperature. Mg/Ca ratios show very strong physiological controls on growth rate but no significant dependence with temperature, except possibly for Acropora sp. and Porites sp. A strong positive correlation is nevertheless found between Li/Ca and Mg/Ca ratios at similar temperatures, indicating that both Li and Mg are subject to control by similar growth mechanisms, specifically the mass fraction of aragonite precipitated during calcification, which is shown to be consistent with a Rayleigh-based elemental fractionation model. The highly coherent array defined by Li/Mg versus temperature is thus largely independent of coral calcification mechanisms, with the strong temperature dependence reflecting the greater sensitivity of the K dLi/Ca partition coefficient relative to K d Mg/Ca . Accordingly, Li/Mg ratios exhibit a highly coherent exponential correlation with temperature, thereby providing a more robust tool for reconstructing paleo-seawat...

[1] New high-density sampling of the Eastern Lau Spreading Center provides constraints on the processes that affect the mantle wedge beneath a back-arc environment, including the effect of the subduction input on basalt petrogenesis and the change in subduction input with distance from the Tonga arc. We obtained trace element and Pb-Sr-Nd isotopic compositions of 64 samples distributed between 20.2°S and 22.3°S with an average spacing of $3.6 km. The trace element and isotope variations do not vary simply with distance from the arc and reflect variations in the mantle wedge composition and the presence of multiple components in the subduction input. The mantle wedge composition varies form north to south, owing to the southward migration of Indian-like mantle, progressively replacing the initially Pacific-like mantle wedge. The mantle wedge compositions also require an enriched mid-ocean ridge basalt-like trace element enrichment that has little effect on isotope ratios, suggesting recent low-degree melt enrichment events. The composition of the subduction input added to the mantle wedge is geographically variable and mirrors the changes observed in the Tonga arc island lavas. The combination of the back-arc and arc data allows identification of several components contributing to the subduction input. These are a fluid derived from the altered oceanic crust with a possible sedimentary contribution, a pelagic sediment partial melt, and, in the southern Lau basin, a volcaniclastic sediment partial melt. While on a regional scale, there is a rough decrease in subduction influence with the distance from the arc, on smaller scales, the distribution of the subduction input reflects different mechanisms of the addition of the subduction input to a variable mantle wedge.

No abstract