Abstract:
Over the past 40 years, meteorological modeling centers (e.g., ECMWF, NASA GMAO, and NCEP/NCAR) have produced retrospective analyses (reanalyses) of the atmosphere: data-driven, physically consistent estimates that ingest several million observations every hour from a diverse collection of sources into an integrated Earth-system model. Such datasets have become a critical resource for understanding the impact of climate variability on the North American carbon cycle. At the same time, the research community has pioneered methods for estimating natural and anthropogenic carbon fluxes using space-based measurements of vegetation (e.g., NDVI, SIF), soil moisture, fire radiative power, nighttime lights, and ocean color. Here, we present an overview of state-of-the-art capabilities that highlight advances in the ability to merge data from satellites and in situ measurements to produce a high-quality, spatially complete picture of carbon concentrations and data-driven flux products. We discuss several applications of such datasets, including the use of aircraft campaign data from ACT-America for satellite validation and the use of datasets from multiple satellites to inform understanding of emissions processes. We also discuss the substantial challenges facing carbon cycle reanalysis efforts including the need for consistency in climate driver data, harmonizing long-term multi-satellite data records (NDVI, ocean color), and understanding the coupling between the land, atmosphere, and ocean systems. Community collaboration on such resources is needed to ensure that next generation Earth system reanalysis will be a valuable tool for the NACP community as it works to understand North American carbon budgets.