2022 ESIP January Meeting

Many scientists are actively experimenting AI/ML methods to either replace the conventional methods or improving the existing data products to higher accuracy and resolution. However, most people complains that the experiments reported in research literature are very difficult to neither reproduce nor reuse. The source code and notebooks and associated data, models, and results are hard to find, access, interoperate, and reuse. Meanwhile, the trained models are often biased towards the majority and common patterns due to sampling strategy or natural distribution. These issues are significantly harming the usability and trustworthy of AI/ML in geoscientific application. This session aims to solicit community experiences, opinion, and vision to enhance the FAIRness and fairness of AI/ML.

View Recording
View Notes

The federal government’s Small Business Innovation Research (SBIR)-Small Business Technology Transfer (STTR) programs are highly undervalued by the very agencies they were intended to benefit. SBIR-STTR programs can provide federal agencies with young but proven technologies that can be rapidly adopted to address agency goals and objectives…on a sole source basis. Established in 1982 (SBIR) and 1992 (STTR), these programs provide billions of dollars in competitive funding for high tech development by small businesses that could be tapped by more agencies and their contractors to promote innovation and rapid adoption of new technology.

SBIR contracts/grants could be a valuable source of support for public-private collaboration and innovation in the ESIP community and could help ESIP increase private sector participation in its initiatives while making more data available for all people from Data Generation to Data Use and Understanding.

A key challenge, however, is that many agencies and federal contractors don’t understand the value that SBIR-STTR contracts and subcontracts can bring to the table, or the steps needed to access these resources. This session will identify specific SBIR examples and discuss the benefits and gaps that exist in the program that may be hampering their agency adoption. Opportunities exist for agencies that need to share their data, work across line offices, and use their data as strategic assets—as well as to meet their goals for diversity and equity in procurement. Diverse opportunities exist for small businesses as well as academic and nonprofit organizations that work with small businesses as subcontractors to access procurement opportunities that could scale up your work in high-impact applications. Come to this session to learn how you can take advantage of SBIR-STTR technologies and accelerate procurement opportunities for your organization and accelerate data-driven decision making!

You will leave this session with a clear understanding of how your agency, your non-profit, or your commercial enterprise can move forward and engage SBIR small businesses as we work to rebuild our nation’s economy.

We will hear from: Jason Kessler, NASA SBIR Program Executive, NASA Headquarters who will provide us with an overview of the SBIR program
We will also hear about a successful STTR project from Bob Chen, Director, NASA SEDAC & about a successful SBIR Phase III technology from Dave Jones, CEO, StormCenter Communications, Inc. and John Williams, Director of Innovation and Technology, Office of Innovation and Technology, Office of Investment and Innovation, U.S. Small Business Administration will be available during the session to answer any questions from the SBA.
Please join us for this exciting ESIP Session.
Recommended ways to prepare for this session: This session will offer excellent information about the SBIR-STTR programs which is good for any level of participant in the ESIP Federation. This session should also provide good information for program managers and contract officers in any US Federal agency, government lab, prome contractor and university. It will be highly educational for agency representatives, non-profit leaders and commercial company leaders.

View Recording
View Notes

The need for common data standards and domain knowledge has reached a precipice. Across ESIP it is no different; ongoing, potentially disparate, conversations about data quality, resource discovery, and domain knowledge are prevalent and yet implicitly rely on a shared view or interpretation.

In this session we bring together members of active ESIP clusters to share topics of interest, needs and gaps with respect to data standards and domain knowledge we endeavour to reuse in some human and machine readable format – e.g. controlled vocabularies, data models, thesauri, taxonomies, classifications, property graphs, ontologies etc.

The following questions, while not exhaustive, are indicative of relevant topics for discussion:
What has prompted the interest in data standards and/or semantics?
What is the goal or use-case?
What is currently available in the space and is it reusable?
Is the community making contributions to any resources?
Where are the gaps in knowledge, standards, or structure?

View Recording
View Notes

A visual map of the topics that will be discussed in this session can be previewed here (warning: very large image).

This session continues the synthesis of ideas contributed by individuals from various ESIP clusters (including Agriculture and Climate, Semantic Harmonization, EnviroSensing, Machine Learning, and Drones) applied to wildfire management. This session focuses on the challenge of ingesting heterogeneous data from in-situ and remotely-sensed systems into models and applications between the pre-fire and fire containment phases. Scenarios include (a) using heterogeneous data for better planning prescribed burns by using data before and after a burn for ingestion into fire behavior models, and (b) using heterogeneous data to recommend both strategic fuel break siting during pre-fire planning and optimal containment line location in the course of active wildfire fighting.

This synthesis session directly extends two of the key takeaways proposed by discussants during the 2021 ESIP summer meeting session “Identifying technology capabilities that meet wildfire science and practitioner requirements”: (a) “...improve fusion among near-term fire behavior model data, values-at-risk data, and sensor data that can be represented and visualized in a Common Operating Picture”, and (b) “...better estimate burn severity by fusing data from various sources (in-situ, remote, model)”.

High-level agenda for this session:

1. Preview and synthesis of session concepts (Brian Wee  |  Massive Connections)
2. Stakeholder perspective: Keeping your eyes on the big picture (Genny Biggs  |  Gordon and Betty Moore Foundation)
3. Stakeholder perspective: Challenges from the wildfire frontlines (Chief Dave Winnacker  |  Fire Chief at Moraga-Orinda Fire District)
4. Technical solution perspective:  In-situ EnviroSensing: challenges and opportunities in the (wild)fire continuum (Scotty Strachan  |  Nevada EPSCoR  |  ESIP EnviroSensing Cluster)
5. Technical solution perspective:  sUAS data use/reuse/repurpose for science and management (Andrea Thomer  |  University of Michigan  |  ESIP Drones Cluster)
6. Technical solution perspective:  Wildland Fire Simulation and Data Assimilation using UAS data (Xiaolin Hu  |  Georgia State University)
7. Technical solution perspective:  AI/ML for Wildfire: Limits and Opportunities (Ziheng Sun  |  George Mason University  |  ESIP Machine Learning Cluster)
8. Breakout groups for (1) In-situ and remote data fusion, (2) UAS data ingest into models.
9. Breakout groups present on (1) Barriers to implementation, (2) What is achievable in the short-term
10. Synthesis and looking ahead

View Recording
View Notes

The Semantic Web for Earth and Environmental Terminology (or SWEET) is a system, created at NASA JPL by the late Rob Raskin and colleagues, reflecting a semantic web technology approach for working with Earth and environmental data. ESIP has been the steward of SWEET more recently. As a knowledge organization system (KOS), specifically a semantic technology, SWEET is applicable and relevant for various disciplines from Earth science to library science to knowledge graphs and others.

This session will bring together anyone interested in SWEET, and will provide an overview and history of the system. It will also explore and solicit interest in developing SWEET to support data from Earth and environmental disciplines, such as disciplines represented in SWEET itself, those of attendees and beyond. ESIP Clusters on disciplines within the scope of SWEET are encouraged to attend.

From attendees, we hope to better understand limitations, gaps, problems with dealing with their disciplinary data. And how SWEET can help. The audience is encouraged to express how they may like to use SWEET, how it may be developed to be better used, etc.

We also hope to identify subject-matter experts (SME) of disciplines covered by SWEET terminology, and any disciplines that may be added. And determine if they are interested in serving as neutral SME, potentially developing such things as local SWEET definitions, verying accuracy of Wikidata definitions, etc. For example, oceanographers in general, or ESIP marine clusters in particular, may be interested in developing SWEET oceanography content, and providing conent about oceanographic data that may need semantic annotations (and thus specific terms in SWEET).
Potential benefit of this will be a pool of potential contributors, and delegation of development tasks. By involving Earth science practitioners, SWEET can be developed to a greater degree of precision; and the current vocabulary (and conceptualization as expressed by it and publications) can be verified for accuracy. If pursued, this would represent a community development approach for SWEET.

This session is relevant for the theme of 'Data for All People: From Data Generation to Data Use and Understanding' in the following manner.
SWEET is a vocabulary and semantic resource open to development for all people and Earth science and data communities. It can help with understanding data, and data use, particularly if further developed. In a collaborative development approach, it can reach more potential applications and persons from diverse backgrounds. Earth observation data is useful for all people across the lifecycle from collection or generation to use to understanding. This translates also to SWEET as a system that can terminologically and semantically support that data and knowledge base.

How to Prepare for this Session1) Review general content

• Knowledge organization systems : Link 1, Link 2
• Semantic technology 

2) Review content on SWEET

• https://github.com/ESIPFed/sweet

3) Spend some time thinking about how your data may find use in SWEET


(written by R.Rovetto, please contact with any questions or interest. https://ontospace.wordpress.com/contact)

Session Notes

Session Recording

The near-Earth space environment is seeing an increase in activity as more nations and organizations engage in astronomical and spaceflight endeavors. This translates to a growing wealth of data and knowledge we can tap for current generations and posterity. Interdisciplinary and diverse perspectives can be applied to leverage that content via: data analysis, search, modeling; informatics, knowledge management; knowledge representation and reasoning (KRR), artificial intelligence (AI); etc. Example techniques are the development of knowledge organization systems (KOS), such as controlled vocabularies, thesauri, metadata schemas, conceptual data models, semantic models, ontologies and knowledge graphs [Rovetto, 2017].

• Space data is understood to include: data about exo-atmospheric space (colloquially: outerspace) phenomena, e.g., space science data such as solar and space weather, solar system, planetary; data from observations of that space; data collected from satellite spacecraft; data about satellites & their activity & contents; celestial body samples; spatio-temporal concepts; misc.

Goals:
This session aims to bring together persons and organizations interested in space data, broadly construed (from astronomy to astronautics to other space sceince and broader space-related topics). Speakers will discuss data science, space and knowledge modeling topics. The session hopes to:
(i) identify various types of space data; desiderata thereof, gaps or problems with dealing with the data; And...

(ii) discuss whether and how presented content and KOSs can be useful, e.g., identifying techniques to organize and analyze space data, and terminology, with an emphasis on knowledge representaiton/modeling, semantic data modeling, conceptual modeling, graph structures, and MBSE (model-based systems engineering). Sharing experiences are encouraged, e.g., challenges, desiderata with your space data, or with specific types or technology related to KOS or semantic technology. And...

(iii) Identify interest and formal support for the KOS/space ontology project presented [Space knowledge modeling] that will be presented.

• Project goals include exploring the utility of ontologies to support space data, and specific Earth-impacting phenomena such as space debris [Rovetto, ESIP FundingFriday, 2020] [Rovetto, ESI, 2015], astronomical phenomena, and spaceflight observations and operations [Rovetto, & Kelso, 2016][O'Neil, Rovetto, 2020-2021].
• Attendees may be interested in supporting the projects concepts for developing international space data systems [Rovetto, 2016] for the global community, if not for specific organizations.

The session will hopefully draw interest (and be attended by) persons and organizations with formal opportunities--work collaborations and educational--for indivdiuals interested in these topics, thereby fostering cooperative, educational and career-building values.

Relevance
This session is relevant for this years theme of 'Data for All People: From Data Generation to Data Use and Understanding' in various ways. First, trust in data is very pertinent for the domain of space. In particular, spaceflight has historically been a governmental or military domain, where security and trust are both critical and challenging. Ideas about how to facilitate and create trust in that and other space communities are in demand, and relevant to AI techniques, and social and societal aspects, etc. Second, applications & benefits of spaceflight activities has reached people across the globe [Rovetto, 2013]. These include Earth observation, e.g., weather and environmental data; technological spinoffs; biological and medical experimentation; and scientific discovery via astronomical observations, e.g., stellar physics, observations of the sun; origins of Earth and planetary evolution; etc. In recent times, some companies ambitiously develop fleets of orbiting spacecraft provide their services. This simultaneously means a promulgation of both data and space-based objects.By bringing together persons interested in space data from astronomy to astronautics, we can better identify avenues to formally cooperate, innovate, and potentially share data, information and knowledge.

View Recording
View Notes

How to Prepare for this Session
1) Review generic material on knowledge organization systems, representation, ontology, semantic & conceptual modeling, etc. Examples include:

• Knowledge organization systems (Link 1, Link 2) -- Knowledge representation & reasoning (Link 1, Link 2) -- Knowledge_management
• Semantic technology -- Conceptual modeling -- MBSE -- NLP

2) Review description references & related references: 

• [Rovetto, Space Knowledge Modeling] - https://purl.org/space-ontology
• [Rovetto, ESIP FundingFriday Award, 2020] https://wiki.esipfed.org/File:ROVETTO.pdf
• [Rovetto, 2021 "Semantic Orbital Sweep (SoS): Developing Knowledge Graphs and an Ontology Suite to Support Orbital Debris Remediation & Spaceflight for the Global Community" post-FF presentation]
• Rovetto, 2020 Esip Presentation "Astronautical knowledge modeling to improve spaceflight safety and data management - a call for project partners and opportunities". (spaceflight aspect of the project)
• [Rovetto, 2016]
• [Rovetto, 2017] Ontology-based Knowledge https://iafastro.directory/iac/paper/id/40147/summary/
• [Rovetto, ESI, 2015]  "An Ontological Architecture for Orbital Debris Data"
• [O'Neil, Rovetto, 2017-2021] https://ntrs.nasa.gov/citations/20210000030
• [Rovetto, 2013] https://www.sciencedirect.com/science/article/abs/pii/S0265964613000660

Contact: https://ontospace.wordpress.com/contact
Services (ontology & vocabularies).
 

Data professionals who engage in CDI and ESIP Collaboration Areas have many mutual interests both in terms of technical topics and application areas. In an effort to increase communication between these groups, we will host a session to share and explore potential synergies. This session will make connections between people in both communities, bring to light lessons already learned or resources already produced, and help to identify common interests and challenges to collaborate on in the future. We will also explore mechanisms for continued sharing across ESIP, CDI, and other related groups going forward.

We will focus on connecting people, expertise, and resources from the ESIP Community Resilience and Disasters collaboration areas and the USGS Risk Research Applications Community of Practice. After brief introductions to the groups and their purpose by the group fellows, we will use this example to address the questions: How do you know where to go when you need help or expertise or want to share something? How can we usefully map the expertise that resides in these separate groups?

View Recording
Agenda

The ESIP Information Quality Cluster, collaborating with other groups from around the world, has led the formation the baselined International Community Guidelines for Sharing and Reusing Quality Information of Individual Earth Science Datasets (see https://doi.org/10.31219/osf.io/xsu4p). The baselined dataset quality guidelines offer an opportunity for improvement to serve multiple disciplines, use cases, and domains of applied sciences. In addition to the Earth sciences and related study areas, various disciplines and data types each need information about data quality when assessing data for potential reuse and when deciding how to use the data. For example, when studying climate change, environmental hazards, and other multidisciplinary issues, often, data from various disciplines are integrated. Such multidisciplinary data integration activities raise cross-disciplinary questions about the quality of individual datasets and approaches to fusing such cross-disciplinary quality information. Also, new questions about data quality emerge when planning to use the products and services that include datasets that have been integrated from various disciplines. Session participants will discuss use cases and data quality issues for interdisciplinary data (re)use and integration in terms of the implications for enhancing the dataset quality guidelines.

Recommended ways to prepare for this session: Identify data quality questions and issues of interest.

Session Notes

Session Recording

We will share with the ESIP community concepts, tools, and outcomes from the OGC Disaster Pilot 2021. We will also invite ESIP members to try out some of these capabilities such as Jupyter Notebooks for creating and visualizing Decision Ready Indicators from Analysis Ready Data, automated multilingual audio survey apps for local disaster reporting, transportation impact models, and/or other systems developed by Pilot participants.

Recommended ways to prepare for this session: Links to material from the Indicator Workshop at the past Summer Meeting and a subsequent Readiness Workshop will help attendees to hit the screen running (link to be provided).

View Recording
View Notes