Page 70 - GIS for Science Technology Showcases

MONITORING GLOBAL SNOW COVER
Using multidimensional raster analysis
Hong Xu and Nawajish Noman, Esri
Snow cover is a critical component of the climate system and the hydrological cycle because it regulates the exchange of heat between Earth’s surface and the atmosphere. Studies show that the changing climate and increasing temperature have significantly decreased snow cover, leading to a warmer planet. Since changes in snow cover affect the ecosystem and access to water resources, the decrease could severely impact our lives.
Monitoring the extent and duration of snow cover provides valuable information to planners, decision-makers, and stakeholders. The Moderate Resolution Imaging Spectroradiometer (MODIS) snow cover data from the National Snow and Ice Data Center (NSIDC) provides long-term time-series analysis and understanding of the climate system. This study used Cloud Raster Format (CRF) data in ArcGIS Pro to analyze snow cover data over 20 years to monitor global, seasonal, and regional trends.
CRF is a raster format optimized for parallel and distributed computation in ArcGIS. The advantages of using CRF for multidimensional raster analysis in ArcGIS over other scientific data formats include simultaneous reading and writing, parallel and distributed computing, efficient extraction and processing of time series data through the creation of transpose, efficient compression using Limited Error Raster Compression (LERC) while maintaining accuracy, and the ability to store analysis templates for on-the-fly analysis.
Using MODIS monthly global snow cover data from 2000 to 2019, the study combines multidimensional CRF and multidimensional raster analysis tools in ArcGIS Pro to map the snow cover percentage and seasonal snow cover percentage data of all years to analyze the spatial variation and trend of snow cover within these years.
The snow cover data is available in Hierarchical Data Format (HDF). A mosaic dataset is created using the Snow_Cover_Monthly_CMG variable from 234 HDF files. A raster function is added to the mosaic dataset to extract only the monthly snow cover percentage and to mask out other categories such as cloud, night, and water as NoData. The mosaic dataset is converted to a multidimensional CRF and then used for further analysis. The overall and the seasonal snow cover percentages of the entire study period are derived using the Aggregate Multidimensional Raster tool. The yearly snow cover trend along with the corresponding statistics are calculated using the Generate Trend Raster and Zonal Statistics as Table tools, and raster functions.
The snow cover percentage map shows that most of the snow coverage is in the Northern Hemisphere. It occurs more frequently above 45 degrees latitude, whereas below 45 degrees, only the tall mountains such as the Himalayas in the Tibet Plateau, Mount Ararat in the Anatolian Plateau, and the Rocky Mountains in western North America experienced frequent snow coverage.
The seasonal characteristic of snow cover can be clearly observed by the seasonal maps of the Himalayas, where winter is January–March, spring is April–June, summer is July–September, and fall is October–December.
A 3D representation of snow cover in the Himalayas during winter. Darker blue colors indicate a higher percentage of snow cover.
204 GIS for Science
A 3D representation of snow cover in the Himalayas during spring. Darker blue colors indicate a higher percentage of snow cover.
Data provided by Hall, D. K., and G. A. Riggs. 2015. MODIS/Terra Snow Cover Monthly L3 Global 0.05Deg CMG, Version 6. Boulder, Colorado, USA, and NASA National Snow and Ice Data Center Distributed Active Archive Center.