Assessing the Impacts of Falling Ice Radiative Effects on the Seasonal Variation of Land Surface Properties

Abstract The interactions between falling ice radiative effects (FIREs) and their impacts on the land-atmosphere feedback processes are examined for the reliability of projected land surface properties and their variability inferred by global climate models (GCMs). This study explores the linkage of surface energy balance, land surface temperature (LST) and land surface properties with FIREs using CESM1-CAM5 sensitivity experiments with FIREs-off (NOS) and on (SON) under CMIP5 historical run. This linkage is inferred by examining the changes of spatial distribution and seasonal cycle of surface radiation, LST and land surface properties. Different regions were selected. For boreal winter, NOS relative to SON, simulates less surface downward longwave and net flux (∼2–15 Wm-2), resulting in colder LST (∼2–4 K) and cooler surface air temperatures over mid- and high-latitudes, which is associated with a shift in soil moisture state from liquid to frozen, increases snow cover and a delay in snowmelt and thawing of soil ice until summer, consequently suppressing vegetation productivity in the following seasons. Key points: Key point #1: Most Coupled Model Intercomparison Project Phase 5 (CMIP5) models did not consider the radiative effects of atmospheric falling ice and their interactions with radiation. Key point #2: Excluding FIREs is associated with decreased land surface temperature, air temperature, soil moisture, and increased frozen ground and snow cover. Key point #3: Excluding FIREs is associated with a delayed melting of snow and soil ice thawing until summer, subsequently suppressing vegetative growth over mid- and high latitudes. Plain Language Summary The potential link between the surface energy balance, land surface temperature (LST) and land surface properties resulting from falling ice radiative effects (FIREs) is explored using CESM1-CAM5 sensitivity experiment run under the CMIP5 historical configuration with FIREs turned-on (SON) and off (NOS). We find that during boreal winter, when FIREs were turned off compared to when they were on, the simulation showed lower surface downward longwave and net flux, resulting in colder LST and cooler surface air temperatures over mid- and high latitudes. This was associated with reduced soil moisture, increased soil ice and snow cover, delayed snowmelt, and thawing of soil ice, thereby suppressing vegetation productivity in the following seasons.  1. Introduction The interaction between the atmosphere and land in the climate system is primarily controlled by the net surface radiative fluxes at the surface and influenced by the spatial-temporal variations in land surface properties (Pitman, 2003; Knutson et al., 2013; Brooks et al., 2015; Li et al., 2016a,b). The land surface plays a significant role in the exchange of energy and moisture with the atmosphere. This exchange occurs through various processes, including the net surface radiative flux, surface turbulent heat fluxes, surface precipitation, and surface stress (Seneviratne et al., 2006, 2010; Haghighi et al., 2018). The net surface radiative fluxes have a controlling effect on land surface processes, which are influenced by land surface and near surface properties such as land surface temperature (LST), surface air temperature, soil moisture state, snow cover, vegetation index, evapotranspiration, among others. Besides, land surface properties change in response to variations in the upper boundary conditions, which are influenced by atmospheric radiation forcing (Li et al., 2021). For this reason, it is crucial to examine the effects of upper boundary conditions, specifically those related to clouds and radiation, on near surface and land surface properties (Lo and Famiglietti, 2013; Anderson et al., 2015). Fully coupled global climate models (GCMs) serve as valuable research tools for studying atmosphere-land interactions. However, there exist uncertainties in accurately representing various aspects of clouds, such as cloud hydrometeor mass, cloud fraction, mixed-phase clouds, and precipitating clouds (i.e., snow). These uncertainties further extend to understanding the radiative impacts of clouds on energetic and hydrological feedback processes across the components of the climate system (e.g., land, ocean, and cryosphere) (Stephens, 2005; Stephens et al., 2015; Trenberth and Fasullo, 2009; Li et al., 2012). Previous studies have utilized a coupled GCM to investigate the interactions between hydrometeors, radiation, and the land surface. Specifically, these studies focused on the impact of falling ice (snow) radiative effects (FIREs) on both present-day climate (Li et al., 2016a) and future warming simulations of the global land surface (Li et al., 2021). They showed that an overestimation of downward shortwave (SW) radiation leads to land surface warming, particularly during boreal summer. Conversely, an underestimation of downward longwave (LW) radiation results in land surface cooling mostly for the mid- and high-latitude land masses, especially during the boreal winter. In addition, they found that the inclusion of FIREs significantly improved the simulation of land surface radiation budgets and LST in CMIP (Coupled Model Intercomparison Project) Phase 5 models (Li et al., 2016a,b). Because the land surface energy balance and LST are critical factors that control surface air temperature through surface turbulent fluxes, it is crucial to address their biases on modeled land-atmosphere interactions and land hydrological processes in most GCMs without including FIREs . These impacts manifest through changes in surface turbulent fluxes, soil moisture, and temperature (Li et al., 2016b). Such biases in the present-day climate simulations can have implications for the confidence level in projecting future land surface properties (Hua et al., 2014; Li et al., 2016a, 2016b, 2021). The motivation behind this study, as well as previous studies by Li et al., (2016a, 2016b), stems from the fact that most current GCMs (more than 90% of CMIP5 models) do not include FIREs, which can significantly impact the surface energy balance and, consequently, the simulation of LST, near surface and land surface properties. Most GCMs in CMIP5 and CMIP6 do not include FIREs but only cloud ice radiative effects (Jiang et al., 2012, Li et al., 2012), yet in reality radiation interacts with all frozen hydrometeors (Waliser et al., 2009, 2011, Li et al., 2013). Following Li et al., (2016b), this study extends to examine how land surface properties change in response to the changes in LST and surface radiation. Additionally, it explores changes in the seasonal cycle with and without FIREs. In this study, we follow the approach outlined by Li et al., (2016a), to analyze the linkage of FIREs on LST and land surface properties via the surface energy budget in terms of seasonal cycle. Our focus is on the physical process related to the changes (differences) with and without FIREs using controlled CESM1-CAM5 simulations, with an emphasis on the spatial changes of land surface properties and the associated relationships with the surface energy budget components in the present-day climate. We discuss how inclusion of FIREs, compared to exclusion of FIREs, substantially changes the simulated surface energy budgets, LST and near surface and land surface properties focusing on boreal winter, which is purposely chosen owing to its strongest signal in the seasonal cycle, and how the seasonal cycle of changes in the land surface properties is impacted by FIREs for selected regions. In section 2, we describe the model simulations and analysis methods. Results are presented in section 3. Discussions and conclusions are presented in section 4.