Deriving global parameter estimates for the Noah land surface model using FLUXNET and machine learning. Issue 22 (18th November 2016)

• Record Type: Journal Article
• Authors: Chaney, Nathaniel W.; Herman, Jonathan D.; Ek, Michael B.; Wood, Eric F.
• Is Part Of: Journal of geophysical research. Volume 121:Issue 22(2016)
• Page Start: 13, 218
• Page End: 13, 235
• Publication Date: 2016-11-18
• Subjects: land surface model -- evapotranspiration -- machine learning; Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1002/2016JD024821
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: With their origins in numerical weather prediction and climate modeling, land surface models aim to accurately partition the surface energy balance. An overlooked challenge in these schemes is the role of model parameter uncertainty, particularly at unmonitored sites. This study provides global parameter estimates for the Noah land surface model using 85 eddy covariance sites in the global FLUXNET network. The at‐site parameters are first calibrated using a Latin Hypercube‐based ensemble of the most sensitive parameters, determined by the Sobol method, to be the minimum stomatal resistance ( r s, min ), the Zilitinkevich empirical constant ( C zil ), and the bare soil evaporation exponent ( fx exp ). Calibration leads to an increase in the mean Kling‐Gupta Efficiency performance metric from 0.54 to 0.71. These calibrated parameter sets are then related to local environmental characteristics using the Extra‐Trees machine learning algorithm. The fitted Extra‐Trees model is used to map the optimal parameter sets over the globe at a 5 km spatial resolution. The leave‐one‐out cross validation of the mapped parameters using the Noah land surface model suggests that there is the potential to skillfully relate calibrated model parameter sets to local environmental characteristics. The results demonstrate the potential to use FLUXNET to tune the parameterizations of surface fluxes in land surface models and to provide improved parameter estimates over the globe. Key Points: Sensitivity analysis of the model parameters in the Noah land surface model Calibration of the land surface model's most sensitive parameters using FLUXNET Mapping the optimized parameters sets over the globe