First Estimates of Continental Energy Storage from CMIP5 Simulations

English Abstract:

Although the continental energy contribution to the Earth's energy budget is small in comparison to the contribution from the ocean, ascertaining the magnitude of the subsurface energy storage may be important for clarifying the long-term energy imbalance of the climate system, it may also help to model energy dependent near-surface processes with potential climate feedback mechanisms, such as permafrost and soil carbon stability. Using data from 21 models from the fifth phase of Coupled Model Intercomparison Project (CMIP5) simulations, we examined the models' representation of subsurface heat storage from their land surface model soil components over the continents. Projected soil temperature anomalies show warmer soil in the future with a wide range of variability. This range is nearly constant with the depth, but future warming RCPs have wider ranges of soil temperature variability, particularly at high latitudes where models show the largest discrepancy in their future projections. We estimated the potential subsurface heat content for the period 1950 to 2000, using a conductive forward model to a depth of 500 m, driven by each models' simulated soil temperature anomalies. Resulting mean subsurface heat storage values range from 0.7×10²¹ to 1.0×10²² J for all continental areas between 60°N to 60°S. Previous analysis of borehole temperature data estimated a subsurface heat content of 8.0±1.0×10²¹ J during the second half of the 20^th century. Preliminary analysis shows that a few of the CMIP5 simulations can account for the magnitude of the subsurface heat storage obtained from geothermal data within the 95% confidence level, but we find no relation between the depth of soil model's bottom boundaries and subsurface heat storage.