Document Type


Publication Date



Paleomagnetic data can be used to estimate deposit temperatures (Tdep) of pyroclastic density currents (PDCs) by finding the laboratory temperature at which a PDC-associated thermal remanence is removed. Paleomagnetic paleothermometry assumes that (1) blocking (Tb) and unblocking (Tub) temperatures are equivalent, and (2) the blocking spectrum remains constant through time. The first assumption fails for multidomain (MD) grains, and recent evidence shows that the second is violated in many titanomagnetites, where Tc is a strong function of thermal history. Here we assess the extent to which the standard paleomagnetic method may be biased by a changing Tb spectrum, and we explore a new magnetic technique that instead exploits these changes. Using samples from the 1980 PDCs at Mt. St. Helens, we find that standard methods on oriented lithic clasts provide a Tdep range that overlaps with measured temperatures, but is systematically slightly higher. By contrast, juvenile pumice give Tdep_min estimates that greatly exceed lithic estimates and measured temperatures. We attribute this overestimate to (1) depth-dependent variations in Tc and Tub resulting from thermally activated crystal-chemical reordering and (2) MD titanomagnetite where Tub>Tb. Stratigraphic variations in Tc are interpreted in terms of Tdep, giving results mostly consistent with measured temperatures and with the lower end of estimates from lithic clasts. This new method allows us to evaluate temporal and spatial variations in Tdep that would not have been possible using standard paleomagnetic techniques in these lithic-poor deposits. It also provides information on deposits not accessible by surface temperature probes.