Increased Variability of Biomass Burning Emissions in CMIP6 Amplifies Hydrologic Cycle in the ESM2 Large Ensemble

Historical simulations performed for the Coupled Model Intercomparison Project Phase 6 used biomass burning emissions between 1997 and 2014 containing higher spatial and temporal variability compared to emission inventories specified for earlier years, and compared to emissions used in previous (e.g., CMIP5) simulation intercomparisons. Using the Community Earth System Model version 2 Large Ensemble, we show this increased biomass burning emissions variability leads to amplification of the hydrologic cycle poleward of 40 degrees N. Notably, the high variability of biomass burning emissions leads to increased latent heat fluxes, column-integrated precipitable water, and precipitation. Greater ocean heat uptake, weaker meridional energy transport from the tropics, greater atmospheric shortwave and longwave absorption, and lower relative humidity act to moderate this hydrologic cycle amplification. Our results suggest it is not only the secular changes (on multidecadal timescales) in biomass burning emissions that impact the hydrologic cycle, but also the shorter timescale variability in emissions. Plain Language Summary Global climate models use different inputs to simulate the past climate as accurately as possible. One of these inputs is an estimate of emissions from the burning of biomass (e.g., from forests and cropland). In the sixth phase of the Climate Model Intercomparison Project Phase 6, the estimated biomass burning emissions were derived using two very different methods. Prior to 1997, emission estimates relied on a combination of indirect measurements and best-guess fire modeling resulting in emissions having relatively modest temporal and spatial variability. During later periods (i.e., 1997-2014) satellite based estimates of fire occurrence and intensity were used in combination with biogeochemical models to produce emission estimates containing much larger spatial and temporal variability. This study demonstrates that the differing variability in biomass burning has an impact on the model's water cycle. During years of strong burning episodes, clouds thin and more sunlight reaches the surface, which results in more surface evaporation, and higher atmospheric humidity and precipitation. Additionally, the high variation in emissions increases rainfall, decreases snowfall, and increases the intensity of extreme precipitation events. Our results show that the timing of biomass burning emissions, not just the amount emitted, is an important moderator of the atmospheric water cycle.