Visualizing in Python: Analyzing GOES-16 Datasets in the Cloud

While the geostationary operational environmental satellite known as GOES 16 is still in its experimental phase, it is clear it will provide the scientific community with a unique opportunity to observe never before seen phenomena. With improved spectral information, spatial resolution, and coverage speed over the current GOES satellite (GOES 13), GOES 16 is better able to resolve more meteorological features and thus improve the way forecasts are made and potentially enhance the quality of research results. Boasting 6 different sensors in areas relating to the earth, the sun, and space, GOES 16 produces roughly a terabyte of data a day that proves difficult to analyze because of the inability to efficiently move such immense amounts of data. Generating the analysis where the data sets are located and transferring the resulting visualizations consumes less financial and computing resources than moving the data alone. Thus utilizing data-proximate analysis by creating visualizations within the cloud environment the data is stored in significantly reduces the cost and network traffic associated with analyzing the large volume of data produced daily by the satellite. Archiving this process within a Jupyter notebook lends the convenience of neighboring the code alongside its analysis to serve as a reference on data-proximate analysis as a fresh way to conduct meteorological analysis on large data sets. Once generated the Jupyter notebook also functions as a resource with several blocks of code that can be adapted to the needs of various studies and research within the science community.