A High Resolution Sea Surface Temperature Climate Data Record produced in a cloud-based science data system

About this technology

Overview

The Multiscale Ultra High Resolution (MUR) Sea Surface Temperature (SST) dataset is a Level 4 blended product currently produced by the Physical Oceanography DAAC (PO.DAAC) at the NASA Jet Propulsion Laboratory (JPL). It is one of its premier public datasets with thousands of annual users in oceanographic research and ecosystems applications, referenced in numerous scientific publications, a significant visualization layer in the web-based NASA WorldView interface, and used operationally in the NOAA CoastWatch program. Given its novel processing architecture using wavelets as the basis functions for the data fusion, it has the ability to better preserve the location and value of SST resolution features across the different spatial scales of input data from Infrared and Microwave sensors such as MODIS and AMSR2. Because of its long time series length (2002-present), careful quality control and uncertainty assessment, and high spatial resolution, it has been identified as an Essential Climate Variable (ECV) and Climate Data Record (CDR) for SST. The preservation of fronts and their location is critical for defining air-sea coupling in models and applications in coastal areas for issues related to biodiversity. The primary goal of this proposal is to refactor the existing production operations from on-premises hardware to the AWS Cloud to address the proposal themes of Information technology development for research and missions. Specifically the code will be updated and modularized into discrete stages that will be containerized to run more efficiently using AWS Cloud services like AWS Step Functions, AWS Batch Jobs, Amazon EventBridge, Amazon Simple Notification Service, and Amazon S3. The key outcomes are to improve the scalability, efficiency, reuse and maintenance of the entire production system. In support of the proposal theme of Data formats that support cloud computing, a second outcome of this activity will be the output of MUR granules as cloud optimized netCDF files following recent best practices. The goal will also include a capability for a kerchunk output, a popular cloud optimization service for HDF and netCDF granules. Finally, to address the larger Transform to open science (TOPS) initiative and contribute to the theme of Advancing access, curation, and discoverability of research software, the modularized MUR components developed from this project will be available from a publicly-hosted GitHub repository to encourage community collaboration and future development.