Understanding Each Other's Models: an Introduction and a Standard Representation of 16 Global Water Models to Support Intercomparison, Improvement, and Communication
Camelia-Eliza Telteu,Hannes Mueller Schmied,Wim Thiery,Guoyong Leng,Peter Burek,Xingcai Liu,Julien Eric Stanislas Boulange,Lauren Seaby Andersen,Manolis Grillakis,Simon Newland Gosling,Yusuke Satoh,Oldrich Rakovec,Tobias Stacke,Jinfeng Chang,Niko Wanders,Harsh Lovekumar Shah,Tim Trautmann,Ganquan Mao,Naota Hanasaki,Aristeidis Koutroulis,Yadu Pokhrel,Luis Samaniego,Yoshihide Wada,Vimal Mishra,Junguo Liu,Petra Doell,Fang Zhao,Anne Gaedeke,Sam S. Rabin,Florian Herz
DOI: https://doi.org/10.5194/gmd-14-3843-2021
IF: 5.1
2021-01-01
Geoscientific Model Development
Abstract:Global water models (GWMs) simulate the terrestrial water cycle on the global scale and are used to assess the impacts of climate change on freshwater systems. GWMs are developed within different modelling frameworks and consider different underlying hydrological processes, leading to varied model structures. Furthermore, the equations used to describe various processes take different forms and are generally accessible only from within the individual model codes. These factors have hindered a holistic and detailed understanding of how different models operate, yet such an understanding is crucial for explaining the results of model evaluation studies, understanding inter-model differences in their simulations, and identifying areas for future model development. This study provides a comprehensive overview of how 16 state-of-the-art GWMs are designed. We analyse water storage compartments, water flows, and human water use sectors included in models that provide simulations for the Inter-Sectoral Impact Model Intercomparison Project phase 2b (ISIMIP2b). We develop a standard writing style for the model equations to enhance model intercomparison, improvement, and communication. In this study, WaterGAP2 used the highest number of water storage compartments, 11, and CWatM used 10 compartments. Six models used six compartments, while four models (DBH, JULESW1 , Mac-PDM.20, and VIC) used the lowest number, three compartments. WaterGAP2 simulates five human water use sectors, while four models (CLM4.5, CLM5.0, LPJmL, and MPI-HM) simulate only water for the irrigation sector. We conclude that, even though hydrological processes are often based on similar equations for various processes, in the end these equations have been adjusted or models have used different values for specific parameters or specific variables. The similarities and differences found among the models analysed in this study are expected to enable us to reduce the uncertainty in multi-model ensembles, improve existing hydrological processes, and integrate new processes.