Grant: Discovery Grants Program - Individual

Climate change is proceeding two to three times faster in the Arctic than elsewhere, leading to a 13% per decade reduction in September Arctic sea ice extent. The changes are driven by anthropogenic carbon dioxide emissions, and are subject to feedback effects, especially from water vapour. Our understanding of water vapour at high latitudes is limited by accessibility difficulties on the ground and interference by clouds with most forms of satellite measurement. Microwave satellite measurements have strong potential for filling data gaps, and for helping to improve predictive capabilities. Twenty-eight years of passive microwave (183 GHz) measurements from a series of polar-orbiting instruments represents a significant, freely available, and under-utilized resource for remote sensing of the Arctic atmosphere and surface. An impediment to their use is the lack of well-developed algorithms needed to retrieve properties of geophysical interest including water vapour column, surface emissivity, and surface emitting layer temperature. Retrieved water vapour columns are systematically moister than modeled historical reanalyses, which decreases the utility of the measurements for geophysical research. I propose an ambitious plan to increase the quality of information obtained from microwave satellite measurements by developing innovative retrieval techniques, analyzing sources of error, and validating retrieved properties against other data sources. Should the observed water vapour bias hold up to further scrutiny, models of the Arctic environment would need to change, with consequences for climate and weather prediction. Full use of available satellite resources will require new theoretical approaches. We will use optimal estimation and machine learning, with an emphasis on artificial neural networks (ANNs), to create new retrievals. New software will be written to process satellite measurements and produce high-quality geophysical data products. The impact of errors induced by surface assumptions, clouds, and choice of radiative transfer model will be explored. Advanced error statistics will be obtained in conjunction with the assimilation system operated by Environment and Climate Change Canada (ECCC). The calculations will represent an important step toward the assimilation of surface-sensitive microwave radiances over land and sea ice that will help improve predictive capabilities. Measurement validation will include a wide range of surface station and aircraft data. Benefits to the research community will include advancements in microwave retrieval capabilities and publicly available validated online data sets for geophysical research. Contributions toward improved forecasts will benefit quality of life through better short-term (i.e., weather) and long-term (i.e., climate) predictions. Trainees and scientists ready for careers in Canada's environmental monitoring and space sciences communities will be developed.