Sensitivity of active-layer freezing process to snow cover in Arctic Alaska

Abstract

Abstract. The contribution of cold-season soil respiration to the\nArctic–boreal carbon cycle and its potential feedback to the global climate\nremain poorly quantified, partly due to a poor understanding of changes in\nthe soil thermal regime and liquid water content during the soil-freezing\nprocess. Here, we characterized the processes controlling active-layer\nfreezing in Arctic Alaska using an integrated approach combining in situ soil\nmeasurements, local-scale ( ∼50 m ) longwave radar retrievals\nfrom NASA airborne P-band polarimetric SAR (PolSAR) and a remote-sensing-driven permafrost model. To better capture landscape variability in snow\ncover and its influence on the soil thermal regime, we downscaled global\ncoarse-resolution ( ∼0.5 ∘ ) MERRA-2 reanalysis snow\ndepth data using finer-scale (500\u2009 m ) MODIS snow cover extent (SCE)\nobservations. The downscaled 1\u2009 km snow depth data were used as key inputs to\nthe permafrost model, capturing finer-scale variability associated with local\ntopography and with favorable accuracy relative to the SNOTEL site\nmeasurements in Arctic Alaska (mean RMSE=0.16 m , bias = - 0.01 m ).\nIn situ tundra soil dielectric constant ( e ) profile measurements were used\nfor model parameterization of the soil organic layer and unfrozen-water\ncontent curve. The resulting model-simulated mean zero-curtain period was\ngenerally consistent with in situ observations spanning a 2 ∘ \nlatitudinal transect along the Alaska North Slope ( R : 0.6±0.2 ; RMSE:\n 19±6 \xa0days), with an estimated mean zero-curtain period ranging from\n 61±11 to 73±15 \xa0days at 0.25 to 0.45\u2009 m depths. Along the same\ntransect, both the observed and model-simulated zero-curtain periods were\npositively correlated ( R>0.55 , p ) with a MODIS-derived snow cover fraction (SCF) from September to October. We also examined\nthe airborne P-band radar-retrieved e profile along this transect in 2014 and\n2015, which is sensitive to near-surface soil liquid water content and\nfreeze–thaw status. The e difference in radar retrievals for the surface\n( ∼ 0.1 m ) soil between late August and early October\nwas negatively correlated with SCF in September ( R = - 0.77 , p ); areas with lower SCF generally showed larger e reductions, indicating\nearlier surface soil freezing. On regional scales, the simulated zero curtain\nin the upper ( m ) soils showed large variability and was closely\nassociated with variations in early cold-season snow cover. Areas with\nearlier snow onset generally showed a longer zero-curtain period; however,\nthe soil freeze onset and zero-curtain period in deeper ( >0.5 m )\nsoils were more closely linked to maximum thaw depth. Our findings indicate\nthat a deepening active layer associated with climate warming will lead to\npersistent unfrozen conditions in deeper soils, promoting greater cold-season\nsoil carbon loss.