Assessing snow cover changes in global mountain regions by exploiting MODIS time series from 2000 to 2024

Snow cover extent and related variables are key elements to understand many processes in mountain regions. To constantly monitor and assess the changes in these areas, consistent and accurate data sets are of utmost importance. In this perspective, MODIS sensors offer an unprecedented possibility in terms of time availability from 2000 to present and ground resolution (500 m) (Bormann et al., 2018).

This work presents a unique time series of snow cover extent and snow phenology (snow cover duration-SCD, first snow day-FSD, and last snow day-LSD) for the period 2000-2024 with a ground resolution of 500 m (Notarnicola, 2024). The main input data is the MODIS product, MOD10A1.061, from which the Normalized Difference Snow Index (NDSI) layer was considered and converted to SCF by exploiting the Salomonson and Appel formulation (2004). The snow phenology parameters (SCD, FSD, LSD) were derived from MOD10A1.061 daily maps. The SCD values were obtained from daily snow cover maps by exploiting an auto-regressive approach to reduce the gaps due to cloudiness (Dietz et al., 2012). In this time series, FSD and LSD represent the first and the last date in the hydrological year with snow presence. The whole dataset is available here: https://zenodo.org/records/11181638

Preliminary analysis of the whole datasets indicate that reduction in snow cover duration can reach up to 55 days while the snow cover extent declines up to 13%. These results were obtained on regions showing changes with significance level at 5% in the Mann-Kendall statistics. Interestingly there are some areas in eastern Russia which show a snow cover extent increase up to 15% while snow cover duration indicates an increase as well but not significant in the adopted statistics.  When considering FSD and LSD variables, both mainly indicates a shortening of the snow season with an average of 15 days for both delayed start of the season and anticipated end of the season. These preliminary results on the trends in the period 2000-2024 provide confirmation of behaviour found in the shorter period 2000-2018 (Notarnicola, 2020), highlighting a general decline for main snow variables but as well with a high variability among the different investigated regions.

References

Bormann, K. J., Brown, R. D., Derksen, C., Painter, T. H. Estimating snow-cover trends from space. Nat. Clim. Change 8, 924–928, 2018.

Dietz, A.J., Wohner C., Kuenzer, C. European Snow Cover Characteristics between 2000 and 2011 Derived from Improved MODIS Daily Snow Cover Products. Remote Sens. 4, 2432-2454, 2012.

Notarnicola, C., Hotspots of snow cover changes in global mountain regions over 2000-2018. Rem. Sen. Environ. 243, 111781, 2020. https://doi.org/10.1016/j.rse.2020.111781.

Notarnicola, C. Snow cover phenology dataset over global mountain regions from 2000 to 2023,Data in Brief, Volume 56, 2024, 110860, ISSN 2352-3409, https://doi.org/10.1016/j.dib.2024.110860.

Salomonson, V.V., Appel, I. Estimating the fractional snow covering using the normalized difference snow index. Remote Sens Environ 89, 351-360, 2004.