Search

High Latitude Dust (HLD, ≥ 50°𝑁 𝑎𝑛𝑑 ≥ 40°𝑆 ) load has implications on the energy budget, ocean biodiversity and economy on a regional and global scale. Current methods of dust detection rely on spectral sensitivity at visible (RGB) and infrared wavelengths. The characteristics of HLD vary according to the sediments and sedimentary processes operating on the land surface that are the source of the dust particles. Leveraging machine learning (ML) methods, we propose a new detection method based on convolutional neural network (CNN) using true color images.

As the number of Earth pointing satellites has increased over the last several decades, the data volume retrieved from instruments onboard these satellites has also increased. It is expected that this trend will continue as more data intensive missions and small satellite constellations are launched. Currently, feature detection - namely atmospheric phenomena - in these datasets is performed manually and is thus not scalable with the growing data archives. Recent advancements in computational efficiency allow for the Earth science community to leverage machine learning to identify interesting atmospheric phenomena. Given the wide range of distinctive features in various atmospheric phenomena, a specialized machine learning model is required for accurate detection of these phenomena independently. The Phenomena Portal, developed at NASA IMPACT, is designed to provide visualization for the output from these machine learning models. In addition, detected events for each atmospheric phenomena are stored in a database that can be used to more easily use/subset larger spatiotemporal datasets. The user interface also incorporates additional features to enhance the user experience including spatiotemporal analysis, multiple base layer images, and a slider to filter events with lower probabilities of positive detection. Each detection supports user feedback on whether the detection is true or false that can then be stored and used to improve the machine learning model performance.

Over 700 climate related datasets were curated by subject matter experts into 9 thematic areas as a part of the Climate Data Initiative (CDI). NASA was tasked with maintaining the collection’s data inventory and supporting web pages at data.gov/climate. Today, the Data Curation for Discovery (DCD) team at MSFC continues to support the CDI collection. In order to expose the collection to a new and growing user community, the DCD team has partnered with GeoPlatform.gov to develop the Resilience community. The Resilience community serves as an interactive, topically-focused web portal that further promotes and shares CDI web content, datasets, services, maps, and other tools relevant to global resilience and change. This poster focuses on the team’s efforts to leverage GeoPlatform’s semantic applications to link CDI objects within the platform to improve discoverability. This poster also provides insights as to how this effort may serve as an example for building and expanding future Geoplatform.gov communities..

During the past year, the Interagency Implementation and Concepts Team (IMPACT) reviewed existing service models in use at various NASA data centers in an effort to produce a unified, cohesive, and comprehensive Level of Service Model for all of NASA Distributed Active Archive Centers (DAACs). NASA DAACs are responsible for ensuring NASA Earth Science data are accurately and securely ingested, distributed, supported, and preserved. The term “Service" as used here refers to the spectrum of data management activities and outputs provided by DAACs in support of the data cared fo by each data cente. The unified Level-of- Service (LoS) model described in this presentation utilizes both the NASA-defined data product category and the data processing level to easily identify an appropriate level-of-service to be applied to a data product throughout the full data life cycle. This LoS model is to be used by DAACs when appraising incoming data in order to determine the appropriate and required services to provide. The LoS model utilizes a 3-level system in which services build upon previous levels and thereby require greater commitment and effort both on the part of the DAAC personnel and the data producer at the highest level. The LoS model description also contains examples of ways to communicate with data producers and data users what services can be expected, thereby bringing more consistent user experiences across the enterprise. In this presentation, we will outline the features of the LoS model and describe how it relates to the FAIR data practices and the NOAA Maturity Matrix model.

The application of machine learning for image-based classification of earth science phenomena, such as hurricanes, is relatively new. While extremely useful, the techniques used for image-based phenomena classification require storing and managing an abundant supply of labeled images in order to produce meaningful results. Existing methods for dataset management and labeling include maintaining categorized folders on a local machine, a process that can be cumbersome and not scalable. Image Labeler is a fast and scalable web-based tool that facilitates the rapid development of image-based earth science phenomena datasets, in order to aid deep learning application and automated image classification/detection. Image Labeler is built with modern web technologies to maximize the scalability and availability of the platform. It has a user-friendly interface that allows tagging multiple images relatively quickly. Essentially, Image Labeler improves upon existing techniques by providing researchers with a shareable source of tagged earth science images for all their machine learning needs. Here, we demonstrate Image Labeler’s current image extraction and labeling capabilities including supported data sources, spatiotemporal subsetting capabilities, individual project management and team collaboration for large scale projects.

Current methods of dust detection rely on spectral sensitivity at visible (RGB) and infrared wavelengths. However, their application on different regions needs to be tuned to mitigate errors associated with background properties. High latitude dust (HLD) regions are characterized by surface with variable albedos and land cover, thus further complicating the dust detection. Leveraging supervised machine learning (ML) methods, we propose a new method accounting for regional differences of dust occurrence.

Advances in machine learning (ML) have made it possible to automatically detect Earth science phenomena from satellite imagery. While useful, ML algorithms typically require an extensive dataset containing labeled images for training. Systematic labeling and management of such datasets is quite cumbersome. With this in mind, we present the Image Labeler. Image Labeler is a fast and scalable cloud-based tool that facilitates the rapid development of Earth science event databases, in order to aid automated ML-based image classification.

Acronyms: ARDS: Analysis Ready Data; AODS: Analytics Optimized Data Stores

Alternative, or non-traditional, data sources can be used to generate datasets which can in turn be analyzed for temporal, spatial and climatological patterns. Events and case studies inferred from the analysis of these patterns can be used by the remote sensing community to more effectively search for Earth observation data. In this paper, we present a new alternative Earth science dataset created from the National Weather Service’s Area Forecast Discussion (AFD) documents. We then present an exploratory methodology for identifying interesting climatological patterns within the AFD data and a corresponding motivating example as to how these data and patterns can be used to search for relevant events or case studies.

Earth science domain presents unique sets of problems that are increasingly being solved using data driven approaches. The availability of big Earth science data offers immense potential for Machine learning (ML) as evident from numerous research publications lately. However, many of these publications are not ending up as production applications mainly because the data scientists who develop the ML models are now expected to complete the ML lifecycle by deploying and scaling the models in production. We introduce ML lifecycle to the Earth science community including the opportunities and challenges that lie ahead in each phase of the lifecycle. We demonstrate the lifecycle using an Earth science problem that we used ML to address and transitioned to production.

Alternative, or non-traditional, data sources can be used to generate datasets which can in turn be analyzed for temporal, spatial and climatological patterns. Events and case studies inferred from the analysis of these patterns can be used by the remote sensing community to more effectively search for Earth observation data. In this paper, we present a new alternative Earth science dataset created from the National Weather Service’s Area Forecast Discussion (AFD) documents. We then present an exploratory methodology for identifying interesting climatological patterns within the AFD data and a corresponding motivating example as to how these data and patterns can be used to search for relevant events or case studies.

New adaptive approaches to Earth observation data stewardship need to be adopted in order to allow for higher data volumes, heterogeneous data and constantly evolving technologies. The data ecosystem approach to stewardship offers a viable solution to this need by placing an emphasis on the relationships between data, technologies and people. In this paper, we present the Joint ESA-NASA Multi-Mission Algorithm and Analysis Platform’s (MAAP) creation of a data ecosystem to support global aboveground terrestrial carbon dynamics research. We present the components needed to support the MAAP data ecosystem along with two data stewardship workflows used in the MAAP and the development of extended metadata for MAAP.

The Earth Observing System Data and Information System (EOSDIS) project at GSFC (Goddard Space Flight Center) maintains and operates a data and information system for NASA's Science Mission Directorate (SMD) and its Earth Science Division (ESD) to support multidisciplinary research in Earth science and public data access.

UNKNOWN

Knowledge graphs link key entities within a specific domain to other entities via relationships. Researchers are able to mine these relationships from numerous sources to infer new knowledge. Text extraction from peer-reviewed papers and scientific reports are untapped resources that can be leveraged by knowledge graphs to accelerate scientific discovery.

Upcoming space-borne missions will offer unprecedented data about Earth but will also feature exponentially high data volumes. These high data volumes will change the way the scientific community works with data and will also create a unique need for improved data sharing and collaboration. NASA and ESA are working together to address these issues by collaboratively developing the Multi-Mission Algorithm and Analysis Platform (MAAP) to improve the understanding of global aboveground terrestrial carbon dynamics. The MAAP will support ESA’s BIOMASS mission, NASA’s GEDI mission and NASA/ISRO’s NISAR mission. The MAAP will be developed in two phases: a pilot phase and a full production phase. The pilot phase will demonstrate collaboration and basic capabilities. The pilot phase will focus on biomass relevant airborne and field campaign data. Two NASA teams are supporting the development of the MAAP. The MAAP engineering team is responsible for the development, maintenance and operations of the MAAP system while the MAAP data team ensures the ongoing quality of the data, metadata and other information provided in the MAAP. The MAAP data team also supports the ingest and archive of identified data to the MAAP platform. This poster describes the use case development process for the pilot MAAP and the data curated in support of those use cases. Additionally, this presentation will outline the pilot MAAP data ingest process and metadata curation effort along with efforts to ensure interoperability between ESA and NASA data and metadata.

We present the development of a deep learning model for objective estimation of tropical cyclone intensity at a higher temporal frequency, deployment of the model in production, design and implementation of the tropical cyclone monitoring and intensity estimation system and development of an interactive portal for situational awareness and evaluation of intensity estimation.

Deep Learning: A subfield of machine learning; Algorithms inspired by function of the brain; Scales with amount of training data; Powerful tool without the need for feature engineering; Suitable for Earth Science applications. Deep Learning for Earth science at MSFC (Marshall Space Flight Center): Phenomena identification; Hurricane intensity (wind speed) estimation; Severe storm (hailstorm) detection; Transverse bands detection; Entity extraction for knowledge graph creation; Ephemeral water detection.

NASA's Earth Observing System Data and Information System (EOSDIS) is tasked with archiving and distributing Earth Observation data across a range of disciplines, including atmospheric science, oceanography, land processes, natural hazards, solar radiance and even socioeconomic aspects relating to the environment. Driven by rapidly rising data volumes, EOSDIS is migrating to a cloud computing based archive over the next few years. Although this simplifies data management somewhat, the main aim is to provide the data in an environment where end users can bring their analysis to the data rather than attempting to download and manage ever-increasing volumes. To that end, a cloud-based analysis platform is being constructed to enable data transformations, analyses and visualization without egressing the data from the cloud. In this endeavor, we expect a wide variety of users, algorithms and use cases. Consequently, the architecture of this cloud analytics platform is expressly designed to be based on open services, thus fostering an ecosystem that enables the efficient combination of common components with data-specific or analysis-specific components. Reviewed and approved by Andrew Mitchell, ESDIS project manager.

Deep learning has revolutionized computer vision and natural language processing with various algorithms scaled using high-performance computing. At the NASA Marshall Space Flight Center (MSFC), the Data Science and Informatics Group (DSIG) has been using deep learning for a variety of Earth science applications. This paper provides examples of the applications and also addresses some of the challenges that were encountered.