Drylands cover 41% of Earth’s surface and are the largest source of interannual variability in the global carbon sink. Drylands are projected to experience accelerated expansion over the next century, but the implications of this expansion on variability in gross primary production (GPP) remain elusive. Here we show that by 2100 total dryland GPP will increase by 12 ± 3% relative to the 2000–2014 baseline. Because drylands will largely expand into formerly productive ecosystems, this increase in dryland GPP may not increase global GPP. Further, GPP per unit dryland area will decrease as degradation of historical drylands outpaces the higher GPP of expanded drylands. Dryland expansion and climate-induced conversions among sub-humid, semi-arid, arid, and hyper-arid subtypes will lead to substantial changes in regional and subtype contributions to global dryland GPP variability. Our results highlight the vulnerability of dryland subtypes to more frequent and severe climate extremes and suggest that regional variations will require different mitigation strategies.
The nonclosure of surface energy balance remains an outstanding problem in eddy covariance (EC) measurements of land‐surface fluxes of heat, water vapor, and CO2. Here data collected from an EC tower over a semiarid sagebrush ecosystem indicate that under unstable atmospheric conditions, the nonclosure becomes increased with increasing instability, consistent with many other studies. It is demonstrated that the increased nonclosure is not caused by the inadequate sampling of large‐scale turbulent motions using a 30‐min averaging interval, though the scales of turbulent motions dominating sensible and latent heat fluxes become enlarged with increasing instability. Quadrant analysis is then used to reveal that the flux contributions from ejections remain nearly constant with increasing instability, whereas the flux contributions from sweeps are reduced and their time fractions increase. Our results imply that the increased nonclosure of surface energy balance is associated with changes in turbulent structures including their dominant time scales and flux contributions of ejections and sweeps as the atmospheric instability increases, which require further studies using vertically distributed observations and/or large eddy simulations.