Temporal Variability of Surface Reflectance Supersedes Spatial Resolution in Defining Greenland's Bare-Ice Albedo
Temporal Variability of Surface Reflectance Supersedes Spatial Resolution in Defining Greenland's Bare-Ice Albedo
复制标题
DOI：
10.3390/rs14010062
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
T. Irvine‐Fynn;P. Bunting;J. Cook;A. Hubbard;N. Barrand;Edward Hanna;A. Hardy;A. Hodson;T. Holt;M. Huss;J. McQuaid;J. Nilsson;K. Naegeli;Osian Roberts;J. Ryan;A. Tedstone;M. Tranter;C. Williamson
中科院分区：
文献类型：
--
作者：
T. Irvine‐Fynn;P. Bunting;J. Cook;A. Hubbard;N. Barrand;Edward Hanna;A. Hardy;A. Hodson;T. Holt;M. Huss;J. McQuaid;J. Nilsson;K. Naegeli;Osian Roberts;J. Ryan;A. Tedstone;M. Tranter;C. Williamson
冰表面反射率是融化和径流的主要调节器，但我们对格陵兰冰盖反射率如何随时间变化的理解仍然很差。这是由于点或样带尺度连续采样与现有卫星产品较粗糙的空间、光谱和（或）时间分辨率之间的脱节。在这里，我们提出了裸冰表面反射率数据的时间序列，这些数据跨越了一系列的长度尺度，从中分辨率成像光谱仪的MOD 10A 1产品的500米到Sentinel-2图像的10米，从地面现场光谱仪的0.1米点测量，以及2.5厘米的无人驾驶航空无人机图像。我们的研究结果揭示了裸冰反射率季节模式的广泛相似性，但进一步的分析确定了每个数据集独特的反射率分布的短期动态。使用这些分布，我们表明，面积平均反射率是当地消融率的主要控制，和特定的冰类型和杂质的空间分布是次要的。考虑到在数据集中观察到的平均反射率的快速变化，我们建议，可通过（i）时间平均反射率数据产品的代表性的定量评估，以及（ii）使用时间分辨函数来描述杂质分布在日常时间尺度上的变化，来改进Rectodo参数化。我们的结论是，区域融化模型的性能可能不会得到最佳改善，通过提高空间分辨率和子像素异质性的结合，而是应该专注于裸冰的时间动态。
Ice surface albedo is a primary modulator of melt and runoff, yet our understanding of how reflectance varies over time across the Greenland Ice Sheet remains poor. This is due to a disconnect between point or transect scale albedo sampling and the coarser spatial, spectral and/or temporal resolutions of available satellite products. Here, we present time-series of bare-ice surface reflectance data that span a range of length scales, from the 500 m for Moderate Resolution Imaging Spectrometer’s MOD10A1 product, to 10 m for Sentinel-2 imagery, 0.1 m spot measurements from ground-based field spectrometry, and 2.5 cm from uncrewed aerial drone imagery. Our results reveal broad similarities in seasonal patterns in bare-ice reflectance, but further analysis identifies short-term dynamics in reflectance distribution that are unique to each dataset. Using these distributions, we demonstrate that areal mean reflectance is the primary control on local ablation rates, and that the spatial distribution of specific ice types and impurities is secondary. Given the rapid changes in mean reflectance observed in the datasets presented, we propose that albedo parameterizations can be improved by (i) quantitative assessment of the representativeness of time-averaged reflectance data products, and, (ii) using temporally-resolved functions to describe the variability in impurity distribution at daily time-scales. We conclude that the regional melt model performance may not be optimally improved by increased spatial resolution and the incorporation of sub-pixel heterogeneity, but instead, should focus on the temporal dynamics of bare-ice albedo.