Estimating Meltwater Drainage Onset Timing and Duration of Landfast Ice in the Canadian Arctic Archipelago Using AMSR-E Passive Microwave Data

Abstract

Meltwater drainage onset (DO) timing and drainage duration (DD) related to snowmelt-water redistribution are both important for understanding not only the Arctic energy and heat budgets but also the salt/heat balance of the mixed layer in the ocean and sea-ice ecosystem. We present DO and DD as determined from the time series of Advanced Microwave Scanning Radiometer-Earth observing system (AMSR-E) melt pond fraction (MPF) estimates in an area with Canadian landfast ice. To address the lack of evaluation on a day-by-day basis for the AMSR-E MPF estimate, we first compared AMSR-E MPF with the daily Medium Resolution Imaging Spectrometer (MERIS) MPF. The AMSR-E MPF estimate correlates significantly with the MERIS MPF (r = 0.73–0.83). The estimate has a product quality similar to the MERIS MPF only when the albedo is around 0.5–0.7 and a positive bias of up to 10% in areas with an albedo of 0.7–0.9, including melting snow. The DO/DD estimates are determined by using a polynomial regression curve fitted on the time series of the AMSR-E MPF. The DOs/DDs from time series of the AMSR-E and MERIS MPFs are compared, revealing consistency in both DD and DO. The DO timing from 2006 to 2011 is correlated with melt onset timing. To the best of our knowledge, our study provides the first large-scale information on both DO timing and DD.

Potential of Melt Pond Fraction Retrieval From High Spatial Resolution AMSR-E/2 Channels

“…These results suggest that the GR(10/89)based MPF retrievals are available from AMSR-E/2 at a product quality most similar to the original GR(6/89)-based MPF retrieval. The original GR(6/89)-based MPF retrieval is valid for late melt season summer conditions, when bare ice and melt ponds are prevalent [3] and available at a product quality similar to the MPF retrieved from optical remote sensing data when broadband albedo is around 0.5-0.7 [4]. It may, however, lead to an increase in pseudo-MPF estimates due to the decrease in H-pol T B at 6 GHz that occurs when dry snow starts to melt [4].…”

“…The original GR(6/89)-based MPF retrieval is valid for late melt season summer conditions, when bare ice and melt ponds are prevalent [3] and available at a product quality similar to the MPF retrieved from optical remote sensing data when broadband albedo is around 0.5-0.7 [4]. It may, however, lead to an increase in pseudo-MPF estimates due to the decrease in H-pol T B at 6 GHz that occurs when dry snow starts to melt [4]. Here we consider the period between MO and when melt ponds form on the ice (pond onset, PO) as a possible error since the snow is melting but there are no ponds yet.…”

“…where f 1 = T B at 6.9 GHz H-pol channel and f 2 = T B at 89.0 GHz V-pol channel in the original MPF retrieval. The retrieval was evaluated using optical remote sensing data [3], [4], resulting in a difference between MPFs from AMSR-E and optical data of less than 5% on eight-day average basis in most regions [3] and 10% on a day-by-day basis in the Canadian Arctic Archipelago (CAA) [4]. The T B from the finest spatial resolution (or the smallest footprint) 89.0 GHz channels is affected by atmospheric influence (cloud liquid water and water vapor).…”