The Plumbing of Land Surface Models: Benchmarking Model Performance

Abstract The Protocol for the Analysis of Land Surface Models (PALS) Land Surface Model Benchmarking Evaluation Project (PLUMBER) was designed to be a land surface model (LSM) benchmarking intercomparison. Unlike the traditional methods of LSM evaluation or comparison, benchmarking uses a fundamentally different approach in that it sets expectations of performance in a range of metrics a priori—before model simulations are performed. This can lead to very different conclusions about LSM performance. For this study, both simple physically based models and empirical relationships were used as the benchmarks. Simulations were performed with 13 LSMs using atmospheric forcing for 20 sites, and then model performance relative to these benchmarks was examined. Results show that even for commonly used statistical metrics, the LSMs’ performance varies considerably when compared to the different benchmarks. All models outperform the simple physically based benchmarks, but for sensible heat flux the LSMs are themselves outperformed by an out-of-sample linear regression against downward shortwave radiation. While moisture information is clearly central to latent heat flux prediction, the LSMs are still outperformed by a three-variable nonlinear regression that uses instantaneous atmospheric humidity and temperature in addition to downward shortwave radiation. These results highlight the limitations of the prevailing paradigm of LSM evaluation that simply compares an LSM to observations and to other LSMs without a mechanism to objectively quantify the expectations of performance. The authors conclude that their results challenge the conceptual view of energy partitioning at the land surface.