Triple-isotope composition of atmospheric oxygen as a tracer of biosphere productivity

Oxygen has three naturally occurring isotopes, of mass numbers 16, 17 and 1 8, Their ratio in atmospheric O-2 depends primarily on the isotopic composi tion of photosynthetically produced O-2 from terrestrial and aquatic plants (1-3), and on isotopic fractionation due to respiration(4). These processes fractionate isotopes in a mass-dependent way, such that O-17 enrichment wo uld be approximately half of the O-18 enrichment relative to O-16. But some photochemical reactions in the stratosphere give rise to a mass-independen t isotope fractionation, producing approximately equal O-17 and O-18 enrich ments in stratospheric ozone(5) and carbon dioxide(6,7), and consequently d riving an atmospheric O-2 isotope anomaly. Here we present an experimentall y based estimate of the size of the O-17/O-16 anomaly in tropospheric O-2, and argue that it largely reflects the influences of biospheric cycling and stratospheric photochemical processes. We propose that because the biosphe re removes the isotopically anomalous stratosphere-derived O-2 by respirati on, and replaces it with isotopically 'normal' oxygen by photosynthesis, th e magnitude of the tropospheric O-17 anomaly can be used as a tracer of glo bal biosphere production. We use measurements of the triple-isotope composi tion of O-2 trapped in bubbles in polar ice to estimate global biosphere pr oductivity at various times over the past 82,000 years. In a second applica tion, we use the isotopic signature of oxygen dissolved in aquatic systems to estimate gross primary production on broad time and space scales.