Comparison of an Earth Science Data Record Derived from MODIS Snow-Cover Maps with the NOAA–Rutgers Climate Data Record of Snow-Cover Extent

Snow cover is an important component of the Earth’s energy balance and a key indicator of climate change when measured on a decadal time scale (see for example, Brown et al., 2010). On the regional scale, snow cover is also important for hydrologic modeling and water security. The economic value of snow as a resource is on the order of trillions of dollars (Sturm et al., 2017), thus the importance of understanding snow cover, including variability and trends, cannot be overstated.

The National Oceanic and Atmospheric Administration (NOAA) and Rutgers University have developed a climate-data record (CDR) of snow-cover extent (SCE) of the Northern Hemisphere [https://doi.org/10.7289/V5N014G9] that extends from 1966 to the present (Robinson et al., 1999; Estilow et al., 2015). The SCE CDR, used extensively for climate studies and model validation, is provided as a weekly snow map at 190.6-km resolution (at 60° latitude). The maps that are currently used to produce the CDR are scaled up from the Interactive Multisensor Snow and Ice Mapping System (IMS) 24-km resolution maps provided at NOAA’s National Ice Center. Since 2000, the MODerate-resolution Imaging Spectroradiometer (MODIS) from NASA’s Terra satellite has provided swath-based, daily and monthly global snow-cover products at a spatial resolution of 500 m, with the MODIS on the Aqua satellite, launched in 2002, providing a nearly-identical suite of snow maps. In 2011 the first Visible Infrared Imaging Radiometer Suite (VIIRS) instrument was launched on the S–NPP satellite having many of the same spectral bands that are on the MODIS instruments, but with 375-m spatial resolution. Both the MODIS and VIIRS moderate-resolution snow-cover products are amenable to production of an Earth Science Data Record (ESDR) and ultimately a CDR when an adequate number of years of data becomes available.

Maps used to produce the NOAA - Rutgers SCE CDR and the MODIS and VIIRS ESDR are produced in an entirely different manner, and for different purposes. Meteorological analysts develop the NOAA snow-cover maps using multiple types of satellite and other information including station data, which are especially useful during cloudy conditions. Conversely, the MODIS and VIIRS snow-cover maps are produced using a completely-automated algorithm thus minimizing or eliminating inconsistencies in processing the snow maps (Riggs et al., 2016). The MODIS snow-cover record has been reprocessed six times, each time taking into account improvements or changes in the snow-cover algorithm, or upstream algorithms (such as the MODIS cloud mask) that are input to the snow-cover algorithm.

The MODIS and VIIRS standard, daily snow-cover maps have provided snow cover during clear-sky conditions only. This can make time series difficult to work with because clouds are so prevalent in the winter. Recently the MODIS snow-cover algorithm suite was enhanced and reprocessed resulting in Collection-6 (C6) products (Riggs et al., 2017). One of the new C6 products is a cloud-gap filled (CGF) version of the daily snow-cover maps (Hall et al., 2010 and 2018). It is also planned for the Collection-1 (C1) VIIRS snow-cover products. The automated production of the MODIS CGF snow-cover maps enables and facilitates development of a moderate-resolution ESDR which is suitable for quantitative comparison with the NOAA – Rutgers SCE CDR. In particular, the observed trends in Northern Hemisphere snow cover, documented and reported based on the SCE CDR [https://climate.rutgers.edu/snowcover/], may now be compared with shorter-term trends developed by using >17 years of moderate resolution CGF snow maps from MODIS and VIIRS.

We report on preliminary results of comparison of monthly trends of the NOAA – Rutgers SCE CDR and the moderate-resolution ESDR based on MODIS snow-cover maps in western North America. The near-term goal is to extend the analysis to all of North America and then to the Northern Hemisphere when full production of the MODIS CGF is achieved, sometime in 2019. Preliminary results show reasonable agreement between the CDR and the ESDR trends for western North America since 2000, with the largest differences most likely related to differences in spatial resolution of the snow products.

References

Brown, R., C. Derksen, and L. Wang, 2010: A multi-data set analysis of variability and change in Arctic spring snow cover extent, 1967–2008, J. Geophys. Res., 115, D16111, https://doi:10.1029/2010JD013975.

Estilow, T.W., A.H. Young and D.A. Robinson, 2015: A long-term Northern Hemisphere snow cover extent data record for climate studies and monitoring, Earth System Science Data, 7(1):137-142.

Hall, D.K., G.A. Riggs, J.L. Foster and S. Kumar, 2010: Development and validation of a cloud-gap filled MODIS daily snow-cover product, Remote Sensing of Environment, 114:496-503, http://doi:10.1016/j.rse.2009.10.007.

Hall, D.K., G.A. Riggs and N.E. DiGirolamo, 2018: Assessment of uncertainties in the Collection-6 MODIS standard cryosphere products, Proceedings of IGARSS’18, 23 – 27 July 2018, Valencia, Spain.

Riggs, G.A., D.K. Hall and M.O. Román, 2016: MODIS snow products user guide for Collection 6 (C6) [http://modis-snow-ice.gsfc.nasa.gov/?c=userguides].

Riggs, G.A., D.K. Hall and M.O. Román, 2017: Overview of NASA’s MODIS and Visible Infrared Imaging Radiometer Suite (VIIRS) snow-cover Earth System Data Records, Earth System Data Records, 9:765-777, https://www.earth-syst-sci-data-discuss.net/essd-2017-25/.

Robinson, D.A., J.D. Tarpley, and B.H. Ramsay, 1999: Transition from NOAA weekly to daily hemispheric snow charts, Proceedings of the 10th Symposium on Global Change, Dallas, TX, Am. Meteorol. Soc., 487–490.

Sturm, M., M.A. Goldstein and C. Parr, 2017: Water and life from snow: A trillion dollar science question, Reviews of Geophysics, 53(5):3534-3544.