Authors: Zhen Li*, Institute of remote sensing and digital earth
Topics: Remote Sensing, Cryosphere, Global Change
Keywords: Snow cover, Google Earth Engine, Assessment, Efficiency
Session Type: Poster
Start / End Time: 1:20 PM / 3:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: No File Uploaded
Accurate monitoring of the global snow cover is important in understanding the impact of the climate on snow cover and the role of water storage by snow in hydrological change. Current prevailing global snow cover products are provided by the DAAC and other institution. MODIS snow cover product is widely used in the world. Google Earth Engine (GEE) is a cloud computing platform dedicated to satellite imagery and other Earth observation data. Using GEE cloud computing, we can achieve a batch processing of huge data and need not to download the data.
In the paper, we aim at the problem of large amount data processing in the accuracy assessment of MODIS snow cover product using cloud computing to improve operational efficiency. Taking the Landsat-TM monthly data as true data, this paper evaluated and analyzed MODIS monthly snow cover products by GEE in snow seasons (September to February) between 2003-2016 at the northern Xinjiang region of China. The results showed that the overall average accuracy of MOD10A1 snow cover products is 88.94%, the average misdetection rate is 45.63%, and the average misjudgment rate is 4.56% in this area. Also, the computing efficiency is compared between GEE and PC (not including data download). The accuracy assessment time of one snow season (6 months) MOD10A1 data on GEE can be completed within 10 minutes, while the time of PC is more than two hours, which means of the GEE can greatly cut down the processing time and improve efficiency.