Remarks on the deuterium excess in precipitation in cold regions

Abstract

Jouzel and Merlivat developed and used an expression for the kinetic fractionation coefficient α_k, and greatly improved the predictions of δ(¹⁸O) and d (the deuterium excess) for East Antarctica. The α_k coefficient is re-derived in terms of ambient cloud temperatures and is used to re-calculate their single-source East Antarctic simulations. While the “improved” α_k produces virtually no change in the δs, there is a small change in predicted d’s. However, no matter which α_k is used, the simulated d’s are very sensitive to the supersaturation history of the precipitating air mass. The single-source model is run for a wide range of supersaturation histories. Averaging the resulting suite of δ(¹⁸O) and d, solutions gives reasonable and stable d predictions. The temperature T_s at which clouds switch from being supercooled water drops to ice crystals depends in part on the microparticle loading of the air. The value of T_s significantly influences the d’s and even δ's predicted. Thus ice-core microparticles, d and δ might be “process related”.