A Global, Gap-filled, Snow-including White-sky Surface Albedo Data Set Based on MODIS

Abstract:

The Earth’s surface albedo determines the exchange of energy between the surface and the atmosphere through radiative transfer, constituting a crucial parameter in many aspects of climate studies. It is also a prerequisite for radiative transfer calculations used in satellite-based retrievals of atmospheric and cloud properties. Such applications require complete global maps of “real-time” albedo, i.e. including temporary changes, such as from snow. The Moderate Resolution Imaging Spectroradiometer (MODIS) provides the most comprehensive satellite-based surface albedo data set, in seven narrow and three broad spectral bands, separately for black-sky and white-sky albedo, on a daily basis and at 1000 m resolution. However, the dataset has gaps, often coinciding with snow covered areas, while its gap-filled version is snow-free.

In this study we attempt to create a MODIS-based, spatially and temporally complete surface albedo dataset including snow. The time series covers the period 2003–2019 and includes white-sky albedo, targeted primarily to cloud properties retrievals. Original albedo data and relevant quality indices came from the MODIS MCD43D version 6 products.

In order to fill the gaps in the data set appropriately, the snow status flag was used; as a first step, gaps in the snow status itself were filled based on an analysis of their spatial and temporal characteristics and a corresponding categorization into snow, snow-free and mixed gaps. This gap-filling exercise was evaluated based on snow depth measurements from the European Climate Assessment and Dataset (ECA&D) stations network. Next, the albedo was filled using either data from the snow-free gap-filled MCD43GF product, in case of snow-free gaps, or spatiotemporal information from adjacent areas and days in all other gap cases. Here, different methods of gap-filling were tested for snow-covered or mixed gaps, depending on their characteristics. The eventual combination of gap-filling approaches was selected based on the outcome of the evaluation tests. These gap-filled albedo values were then evaluated based on measurements from the Baseline Surface Radiation Network (BSRN). Only cloud-covered cases were selected, since these measurements correspond to the white-sky albedo.