Patrick started as a student of physics and mathematics with an ambition to study elementary particle physics but later found himself amazed by the complex field of climate modeling. Upon completing his master’s degree at the University of Bonn, Patrick decided to switch to a career in climate research while pursuing his PhD in ocean wave modeling at the Max-Planck Institute for Meteorology. This was a starting point for Patrick’s successful career as a climate researcher, in which he was the Principal Research Scientist at the Massachusetts Institute of Technology (MIT).
Currently, Patrick is the director of the Computational Research in the Ice & Ocean Systems (CRIOS) group at the University of Texas at Austin. The team seeks to understand the influence of glaciers, ice and snow in the global climate system. Later this year, Patrick had the opportunity to host a portuguese researcher to the team within the short-term research internships initiative. The collaboration with the UT Austin Portugal Program doesn’t end there, as Patrick is also the Program’s Area Director for Space-Earth Interactions. Learn more about Patrick’s career, the intriguing research his team conducts and some thoughts on the future of the field of Space-Earth Interactions in the interview below.
Patrick graduated in Physical Oceanography and Climate Dynamics at the Max Planck Institute for Meteorology, in Germany. What inspired you to pursue a career in this field? What is the importance of studying the oceans and their contribution to the global climate system?
As many of my colleagues, I started studying physics and mathematics, with an early ambition to conduct research in elementary particle physics. During my study, I was exposed to emerging work in the field of climate modeling and remote sensing, an extraordinarily complex problem which required physics and mathematics as core skills. Upon completion of my Masters degree (Diploma in Physics) at the University of Bonn, I decided to switch to a career in climate research and was lucky enough to get the opportunity to pursue a PhD in ocean wave modeling at the Max-Planck Institute for Meteorology in Hamburg, a leading research institute in climate science. Covering around 71% of the Earth’s surface, the global ocean is a flywheel of the climate system as it stores and redistributes around 90% of the heat added to the planet due to anthropogenic greenhouse gas warming, it sequesters around 25% of human CO2 emissions, and it receives close to 100% of melt water from glaciers and ice sheets to raise global sea level. The challenge of studying, understanding, and predicting the global ocean circulation has fascinated me ever since.
Throughout your career, you joined research groups at the Massachusetts Institute of Technology (MIT) and the University of Texas at Austin, where you are currently director of the Computational Research in Ice & Ocean Systems (CRIOS) group. What work has your team been developing? What is the impact that can result from the discoveries made?
A core activity of the CRIOS group at UT Austin is to understand the role of the ocean and marine cryosphere in the global climate system. To do so, we use computational tools from optimal estimation and control theory to combine the diverse, heterogenous satellite and in-situ ocean observations with a state-of-the-art ocean general circulation model (GCM) over the era spanning the satellite altimetric record (1992 to present). We use differentiable programming and automatic differentiation to generate the adjoint of the GCM required for gradient-based optimization. The project, known as Estimating the Circulation and Climate of the Ocean (ECCO) and supported by NASA, produces multi-decadal state estimates supporting the wider research community. At UT Austin, we have also developed a regionally refined, higher-resolution coupled ocean-sea ice state estimate of the Arctic, known as the Arctic Subpolar gyre sTate Estimate (ASTE). supported by NSF. The ECCO and ASTE and products have enabled a wide range of studies that investigate the dynamical mechanisms underlying decadal variations in the ocean circulation, and contributed to studies of global and regional sea level change, oceanic heat uptake, and Arctic freshwater export. The underlying computational tools developed within CRIOS provide rigorous frameworks for dynamical attribution of observed changes. They are closely related to approaches of machine learning-based backpropagation algorithms and explainable AI methods. Our group is now working to develop hybrid approaches of data assimilation and scientific machine learning for Earth system models within the Julia programming language in a project called DJ4Earth. Besides a core interest in satellite retrievals of the oceanic state, we have also been working to bring together the international research community to address global deep sea challenges at the intersection of communities and disciplines. As part of the Deep Ocean Observing Strategy (DOOS), a UN Ocean Decade-endorsed Programme, we are working to build an international network of network that seeks to (i) characterize the physics, biogeochemistry and biology of the deep ocean in space and time, (ii) establish a baseline required to understand changes to its habitats and services, and (iii) provide the information needed to have a healthy, predicted, resilient and sustainably-managed (deep) ocean.
Since the end of September this year, Patrick had the opportunity to host a Portuguese researcher, Liliana Pereira, who is undergoing a short-term research internship at the Oden Institute for Computational Engineering & Sciences, under the UT Austin Portugal Program. How has been the experience of hosting international researchers, more specifically from Portugal?
Liliana Pereira has been a valuable addition to our research group during her stay at UT Austin. She has worked at the intersection of two important research thrusts, one related to the use of numerical simulations to assess the benefit of “Science Monitoring And Reliable Telecommunications” (SMART) oceanographic sensors deployed on subsea telecommunication cables, and another one related to the assessment and improvement of ocean bottom pressure measurements retrieved from satellite gravimetry of the GRACE-FO mission and measured via the SMART sensors. There is considerable interest in these research thrusts both from a Portuguese perspective (Portugal is among the first countries to deploy SMART sensors on telecommunication cables connecting the Portuguese main land, the Azores, and Madeira) and from UT, which is home to the GRACE-FO’s science team at the Center for Space Research. The CRIOS group collaborates with the University of Hawaii in conducting observing system simulation experiments (OSSEs) to assimilate SMART sensor data as part of a Gordon and Betty Moore Foundation project.
Satellite data analysis has proven to be very crucial in different areas, especially in monitoring the effects of climate change on Earth, and some research projects from the UT Austin Portugal Program even focus on these topics. As the UT Austin Portugal’s Area Director for Space-Earth Interactions, what is your opinion on the future of this area? In what ways do you think Earth Observation will be relevant, within and outside the context of climate change?
Satellite data of the ocean (and wider coupled Earth system) play a tremendous role in Earth science and decision support, ranging from improving our understanding of detailed physical processes underlying observed changes, improve our ability to simulate and predict these processes at an unprecedented level of detail and accuracy through well-calibrated and initialized models, early warning and/or detection of extreme events, marine resource management, and decision support through the development of digital twins of the ocean. Rapid technological progress is being made in the development of remote sensing capabilities and miniaturization, enabling deployment of Earth observing capabilities on small satellites. UT Austin Portugal’s MAGAL Constellation project is one such example, setting the stage for a future ocean and climate change monitoring constellation, based on radar altimeter data combined with gravimetry and ocean hydrographic measurements. With the exponential increase of remote sensing data comes the need for the use of advanced data science and data assimilation methods to optimally use these data. This will require a close collaboration between domain (i.e. ocean and climate) scientists, computer and computational scientists, and aerospace engineers working in the field of satellite remote sensing.
