The COsmic-ray Soil Moisture Interaction Code (COSMIC) for use in data assimilation

摘要

Soil moisture status in land surface models (LSMs) can be updated byassimilating cosmic-ray neutron intensity measured in air above the surface.This requires a fast and accurate model to calculate the neutron intensityfrom the profiles of soil moisture modeled by the LSM. The existing MonteCarlo N-Particle eXtended (MCNPX) model is sufficiently accurate but tooslow to be practical in the context of data assimilation. Consequently analternative and efficient model is needed which can be calibrated accuratelyto reproduce the calculations made by MCNPX and used to substitute for MCNPXduring data assimilation. This paper describes the construction andcalibration of such a model, COsmic-ray Soil Moisture Interaction Code (COSMIC), which is simple, physically based andanalytic, and which, because it runs at least 50 000 times faster than MCNPX, isappropriate in data assimilation applications. The model includes simpledescriptions of (a) degradation of the incoming high-energy neutron fluxwith soil depth, (b) creation of fast neutrons at each depth in the soil,and (c) scattering of the resulting fast neutrons before they reach the soilsurface, all of which processes may have parameterized dependency on thechemistry and moisture content of the soil. The site-to-site variability inthe parameters used in COSMIC is explored for 42 sample sites in theCOsmic-ray Soil Moisture Observing System (COSMOS), and the comparativeperformance of COSMIC relative to MCNPX when applied to representinteractions between cosmic-ray neutrons and moist soil is explored. At anexample site in Arizona, fast-neutron counts calculated by COSMIC from theaverage soil moisture profile given by an independent network of pointmeasurements in the COSMOS probe footprint are similar to the fast-neutronintensity measured by the COSMOS probe. It was demonstrated that, when usedwithin a data assimilation framework to assimilate COSMOS probe counts intothe Noah land surface model at the Santa Rita Experimental Range field site,the calibrated COSMIC model provided an effective mechanism for translatingmodel-calculated soil moisture profiles into aboveground fast-neutroncount when applied with two radically different approaches used to removethe bias between data and model.