Search for: All records

The global sulfur cycle plays a critical role in the redox evolution of Earth’s surface and upper mantle, yet the distribution and origin of sulfur in the mantle remains largely unconstrained. El Hierro is a volcanic island in the Canary archipelago that is fed by sulfur-rich magmas. To constrain the origin of sulfur in these melts, we combine in situ sulfur isotope analyses with regression modeling. We calculate that undegassed El Hierro melts have δ34S values of 0 ± 2‰. The average δ34S of undegassed El Hierro melts is 0.3‰ to 1‰ higher than magmas erupting at mid-ocean ridges. Mass balance calculations reveal that El Hierro’s mantle source contains 310 ± 120 μg/g sulfur and that on average 60% of sulfur in the source is of recycled origin. This recycled material should contain >1,800 μg/g sulfur to satisfy isotopic constraints on its mass fraction in the mantle source. The sulfur and oxygen isotopic signature in serpentinites and sediments deviate significantly from the upper mantle, making them unsuitable candidates for the recycled material. An oxidized partial melt of recycled oceanic crust that retained one third of its sulfur budget after subduction zone processing can explain excess sulfur in the Canary Island mantle. Recycled oceanic crust is expected to contain sulfur as sulfide, which is not capable of oxidizing the mantle. The presence of ferric iron in the recycled component is necessary to produce metasomatic melts that are oxidizing enough to carry sufficient sulfur into the mantle source of ocean island basalts. 

Abstract A multiproxy record from a fast‐growing stalagmite reveals variable hydroclimate on the California coast across the 8.2 kyr event and a precursor event likely caused by initial drainage of proglacial Lake Agassiz. Using speleothem δ⁴⁴Ca, we develop the first semiquantitative estimates of paleorainfall variability for California through calibration with measurements of the modern climate and cave environment. We find that the magnitude of rainfall variability during the 8.2 kyr event approached the multiyear variability observable in the recent past (1950–2019) and the magnitude of variability during the precursor event likely exceeded this range. Additionally, we observe other instances of multidecadal variability comparable in magnitude to the precursor event during the record. Our work suggests that speleothem calcium isotope ratios are a powerful semiquantitative means to reconstruct paleorainfall, although numerous factors must be assessed in each cave system before applying this approach.