Intercomparison of Thermal Regime Algorithms in 1‐D Lake Models. Issue 6 (17th June 2021)

• Record Type: Journal Article
• Authors: Guo, Mingyang; Zhuang, Qianlai; Yao, Huaxia; Golub, Malgorzata; Leung, L. Ruby; Tan, Zeli
• Is Part Of: Water resources research. Volume 57:Issue 6(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-06-17
• Subjects: freshwater ecosystems -- global lakes -- lake biogeochemistry -- lake thermal modeling -- methane emissions; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2020WR028776
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Lakes are an important component of the global weather and climate system, but the modeling of their thermal regimes has shown large uncertainties due to the highly diverse lake properties and model configurations. Here, we evaluate the algorithms of four key lake thermal processes including turbulent heat fluxes, wind‐driven mixing, light extinction, and snow density, using a highly diverse lake data set provided by the Inter‐Sectoral Impact Model Intercomparison Project (ISIMIP) 2a lake sector. Algorithm codes are configured and run separately within the same parent model to rule out any interference from factors apart from the algorithms examined. Evaluations are based on both simulation accuracy and recalibration complexity for application to global lakes. For turbulent heat fluxes, the non‐Monin–Obukhov similarity (MOS) based, more simplified algorithms perform better in predicting lake epilimnion temperatures and achieve high convergence in the values of the calibrated parameters. For wind‐driven mixing, a two‐algorithm strategy considering lake shape and season is suggested with the regular mixing algorithm used for spring and earlier summer and the mixing‐enhanced algorithm for summer steady stratification and fall overturn periods. There are no evident differences in the simulated thermocline depths using different light extinction algorithms or the observation. Finally, for lake ice phenology, an optimal algorithm is decided for most northern lakes while the Arctic lakes require separate consideration. Our study provides highly practical guides for improving 1‐D lake models and feasible parameterization strategies to better simulate global lake thermal regimes. Key Points: A set of optimal algorithms and parameterization strategies is provided to better simulate global lake thermal regimes with 1‐D lake models More complicated algorithms can lead to larger biases and require more recalibration efforts for global application Combination of algorithms for lakes of different characteristics can optimize the model performance