Skip to main content
ARS Home » Plains Area » Las Cruces, New Mexico » Range Management Research » Research » Publications at this Location » Publication #320429

Title: Evaluating MODIS snow products for modelling snowmelt runoff: case study of the Rio Grande headwaters

item STEELE, CAITI - New Mexico State University
item DIALESANDRO, JAKE - New Mexico State University
item James, Darren
item Elias, Emile
item Rango, Albert
item BLEIWEISS, MAX - New Mexico State University

Submitted to: International Journal of Applied Earth Observation and Geoinformation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/10/2017
Publication Date: 9/5/2017
Publication URL:
Citation: Steele, C., Dialesandro, J., James, D.K., Elias, E.H., Rango, A., Bleiweiss, M. 2017. Evaluating MODIS snow products for modelling snowmelt runoff: case study of the Rio Grande headwaters. International Journal of Applied Earth Observation and Geoinformation. 63:234-243.

Interpretive Summary: Snow cover products (MOD10a) from the MODIS sensor on the Terra satellite and downloadable from the National Snow and Ice Data Center (NSIDC) are widely used for mapping snow cover. These products appear to work well but we found that when compared with more detailed maps from a different sensor, that the MOD10a1 products from NSIDC do not always capture patchy snow cover and are not very sensitive to snow cover under forests. The latest version (collection 6 MOD10a1) is more sensitive to patchy snow cover than the previous version. But colelction 6 underestimates snow covered area for the most of the early melt season. An alternative MODIS product - MODSCAG - performs almost as well as finer spatial resolution products. If there is a problem in mapping snow then this affects the performance hydrological models when trying to simulate streamflow. We recommend MODSCAG or a similar approach if coarser spatial resolution data are to be used to map snow cover.

Technical Abstract: Snow-covered area (SCA) is a key variable in the Snowmelt-Runoff Model (SRM). Landsat Thematic Mapper (TM) or Operational Land Imager (OLI) provide remotely sensed data at an appropriate spatial resolution for mapping SCA in small headwater basins, but the temporal resolution of the data is low and may not always provide sufficient cloud-free dates. The Terra Moderate Resolution Imaging Spectroradiometer (MODIS) sensor offers better temporal resolution than the Landsat sensors and in some years, MODIS data offer the only alternative for mapping snow cover. Gridded MODIS snow products (MOD10a1) are readily available from the National Snow and Ice Data Center (NSIDC). An alternative canopy-corrected SCA product derived from the MODIS Snow Covered-Area and Grain size retrieval algorithm (MODSCAG; Painter et al., 2009) is also available for some areas through NASA’s Snow Data System. In this study we aim to identify the optimum MODIS snow product for use in smaller headwater basins. Using the Rio Grande headwater basin as a case study, we compared SCA estimates derived from binary classification of TM imagery, binary and fractional MOD10a1 products (Collection 5 and Collection 6), and MODSCAG canopy-corrected fractional SCA. Four years were selected for comparison: 2001, 2008, 2010 and 2011. In the early melt season, the revised fractional product from MOD10a1 Collection 6 underestimates snow cover compared to all other products tested here, but shows better sensitivity to end-of-season snowpack than the collection 5 equivalent. The canopy-corrected MODSCAG product was the best-performing MODIS product for mapping SCA and these results translated into better simulations of streamflow with SRM than either of the MOD10a1 products