Fig. 3: Historical and future average surface radiative forcing (RF) induced by light-absorbing particles (LAP) and April snow water equivalent (SWE) over the snow-covered Tibetan Plateau (TP) regions (where the average SWE exceeds 5 mm in the historical period of 1995–2014) under different model configurations.
From: A cleaner snow future mitigates Northern Hemisphere snowpack loss from warming
a, b The average RF from December to May under SSP126 and SSP585. c, d April SWE under SSP126 and SSP585. For each panel, Climatehist + LAPhist represents the historical (1995–2014) simulations with historical LAP depositions, while Climatefuture + LAPfuture and Climatefuture + LAPhist represent future (2081–2100) simulations with and without a future change of LAP depositions, respectively. The Climatefuture + LAPhist simulations used the historical average LAP depositions from 1995 to 2014. The horizontal axis labels represent different model configurations (see the “Methods” section), where Control has the Energy Exascale Earth System Model (E3SM) Land Model (ELM) default settings and the others represent major adjustments made from the Control case. Specifically, PP assumes that the terrain is flat and neglects topographic effects on solar radiation; Koch assumes a non-spherical snow grain shape (Koch snowflake); extBC assumes external mixing between hydrophilic BC and snow grains; intDust assumes internal mixing between dust and snow grains; noLULCC has no land use and land cover change; MSE_high assumes high melt-water scavenging efficiency (MSE = 2, much higher than the default value of 0.2) of hydrophilic BC; and MSE_low assumes a low MSE (0.02) of hydrophilic BC. In c, d the contribution (δLAP) of future LAP change that mitigates snowpack loss under each ELM configuration is noted as a percentage and is calculated as the ratio of the SWE difference (∆SWELAP) between Climatefuture + LAPfuture and Climatefuture + LAPhist to the SWE difference (∆SWEClimate) between Climatehist + LAPhist and Climatefuture + LAPfuture. The geographical coverage of the TP is shown in Fig. 4.
