2.1 CAS-LSM
Detailed descriptions of the model improvements and the performance of CAS-LSM in the offline mode (i.e., forced with observed meteorology) was mainly documented by Xie et al. (2018). The following series of papers provided reference for the many schemes used here: Xie et al. 2012; Zeng et al. 2016a, 2016b, 2018; Zou et al. 2014, 2015; Zeng et al. 2016b, 2017; Gao et al. 2016, 2019; and Liu et al. 2019. For reference, we have included a schematic diagram that depicts the additional processes and functionality that exists within CAS-LSM (Fig. 1). The schemes for GLF, human water use, FTFs, anthropogenic nitrogen discharge, and urban planning, were implemented into CLM4.5 (Oleson 2013) to produce CAS-LSM, which now considered the abovementioned processes. CLM4.5 was developed by the National Center for Atmospheric Research. This model included bio-geophysical and bio-geochemical mechanisms and energy, and mass fluxes from the land to the atmosphere, and was the land-surface component of the Community Earth System Model 1.2.0 (Hurrell et al. 2013). The bio-geophysical processes included solar and longwave radiation interactions with vegetation canopies and soil, momentum and turbulent fluxes from the canopies and soil, heat transfer in soil and snow, hydrology of canopies, soil, and snow, and stomatal physiology and photosynthesis. The bio-geochemical processes included vegetation photosynthesis, phenology, the carbon and nitrogen cycles, decomposition, and wildfires (Lindsay et al. 2014). Also contained in CLM4.5 was an interactive crop-management model, which simulated crop growth and its effect on land processes. A sub-gridded hierarchy of land units, soil profiles, and plant-function types was used in CLM4.5 to describe the heterogeneity within each grid cell. Different land uses, such as varieties of vegetation, lakes, urban areas, and glaciers, were addressed separately even if they coexist in a given grid cell. The CAS-LSM, building upon CLM4.5, coupled the GLF, HWR, FTFs, AHR, UWU and DIN as optional configurations, with the target being more complete land surface modeling.