Cressie, Noel Distinguished Prof

<p>To sustain our voyage through space, we need to know the state of our ship, Earth. We need to look for stressors, overheating, and new energy sources. We need to monitor its integrity, repair components that are broken, and establish long-term maintenance protocols.</p><p> Our spaceship is in peril, and no-one is in charge. We have scientific knowledge about different parts with different degrees of certainty, but too often debate is not informed by science and degenerates into fractious disagreements of opinion.</p><p> Carbon is the fourth most abundant element in the universe. Earth's carbon cycle is a closed system, with carbon moving in and out of different “pools.” Carbon causes no harm while it is locked in the ground (the terrestrial pool) but, once extracted and burned as a fossil fuel, it is released into the atmospheric pool to bond with oxygen and become carbon dioxide (CO₂).</p><p> Odourless, tasteless, and invisible, CO₂ is a leading greenhouse gas that is a major driver of climate change. Although fossil fuel powers much of the human activity on our planet, it leaves an excess of CO₂ in the atmospheric pool. These anthropogenic sources of CO₂ are causing an imbalance between its emission and re-absorption at Earth's surface, and strong science and technology (including Statistical Science) is needed to address this Grand Challenge of what to do about it.</p><p> Inside the ship, the Earth system’s “clubs” (e.g., the carbon club, the aerosol club, the water club) are spinning. In pre-industrial times, there was balance in the carbon club, when about as much carbon was being removed from the atmosphere as was being added. However, in the first decade of this century, the net increase of atmospheric CO₂ was about 4 gigatonnes (Gt) per annum, which is huge – it represents about half of what was produced anthropogenically. The excess CO₂ is spread around the planet by weather systems and accumulates as a fraction of all the atmospheric gases; its presence is measured in parts per million (ppm).</p><p> In the first decade of this century, yearly increases of atmospheric CO₂ were about 2 ppm but, for the years 2015 and later, this has accelerated to almost 3 ppm. In 2015, atmospheric CO₂ officially reached 400 ppm, which is a 26% increase from 1960 levels. (From paleoclimate data, atmospheric CO₂ last reached 400 ppm on Earth in the middle Pliocene, about 3.6 million years ago, before <em>Homo sapiens</em> arrived.) In 2020, our planet’s CO₂ is expected to reach 415 ppm.</p><p> My research impact on this problem began with my membership a decade ago of the NASA Science Team of the Orbiting Carbon Observatory-2 (OCO-2) mission. OCO-2 is a NASA satellite launched in July 2014, whose primary science objective is to estimate the global geographic distribution of CO₂ fluxes (i.e., sources and sinks) at Earth’s surface through time. This knowledge, of how carbon moves in and out of the atmospheric pool, is critical to developing smart mitigation and management strategies that will restore balance to the carbon club.</p><p> The problem is hard because direct measurement of fluxes is globally inadequate. Remote sensing data help, but they give a different measure of CO2, one obtained from looking down at Earth through a narrow, swirling atmospheric column. The approach taken in our Centre is to use Statistical Science to “invert” the globally plentiful satellite data and trace the measured CO₂ back to their sources and sinks. This flux inversion uses a fundamental result in probability theory called Bayes’ Theorem, but it is applied here in an extraordinary setting. It requires handling very high dimensional data; tracking latent unobserved processes globally and through time; sampling from the probability distribution of all the “unknowns” conditional on the data; working in a high-performance computing environment; and collaborating in an interdisciplinary team of statistical and atmospheric scientists.</p><p> Our (<em>Homo sapiens</em>) juggling with the carbon (and aerosols, water, etc.) club is a high-risk act that trades off economic well-being, jobs, and growth (and the fossil fuels that power them) with a very real change in Earth’s climate. It requires a careful characterisation of what we know (the geophysics, the data), their uncertainties, and a decision space with costs and benefits articulated. The devastation of Australia’s “black” Spring–Summer bushfires of 2019–2020 illustrates the enormous cost, ecological as well as financial, of taking little or no action.</p><p> Statistical Science shows how to combine the geophysical and observational uncertainties into an integrative, hierarchical, physical-statistical model. It maps the path from observations to information to knowledge, using conditional probabilities to quantify the uncertainty in the knowledge attained. The Centre for Environmental Informatics at the University of Wollongong has received a three-year ARC Discovery Project to answer the following key questions about carbon fluxes (sources and sinks): Where, when, how much, and how certain?</p><p> This story is fundamentally about saving more (CO₂ absorption) and spending less (CO₂ emission). Every year, the budget is in debt to the tune of 4 Gt of carbon dioxide, a proven greenhouse gas. Our planet is overheating and, to save all who voyage in her, our governments must recognise the threat and collectively work together to enhance its sinks and decrease its sources. There is little chance of a do-over – what our generation does or does not do now will deeply affect our children’s generation and those that follow.</p><p> </p>