Toward a US Framework for Continuity of Satellite Observations of Earth's Climate and for Supporting Societal Resilience

.

Earth's Future WALISER 10.1029/2023EF003757 2 of 27 observed increase in these types of extreme events (NCA, 2018), with the increases in cost also accounting for increasing population and associated infrastructure in vulnerable areas (e.g., coasts, urban-wildland boundaries).For example, climate change is estimated to have tripled the likelihood of occurrence of Hurricane Harvey's devastating rainfall in Texas in 2017 (e.g., Emanuel, 2017;van Oldenborgh et al., 2017)-one of the notable disasters causing more than $1 B in damages in Figure 1.Similarly, the heat wave in the Pacific Northwest in summer 2021 is estimated to have killed over 200 people in Oregon and Washington combined, and over 600 people in British Columbia (BCCS, 2022;Ryan, 2021;Templeton & Samayoa, 2021), with climate change also found to be a factor in the development and severity of the heat wave (e.g., Bartusek et al., 2022;McKinnon & Simpson, 2022).
Recognition of the value of satellite observations for informing decisions related to safeguarding life and property in the face of the above types of extreme events is reflected in our nation's sustained support of a suite of weather satellites.This support amounts to about $1.6 B/year in annual spending to provide sustained polar and geostationary satellite observations from platforms that span decades of operational continuity (NOAA, 2023a).Such data represent a foundational information source for the weather forecast products provided by the US National Weather Service and by other government and commercial enterprises that produce value-added products to support routine agricultural, resource management, transportation, business and other decisions influenced by environmental conditions.In addition to the mandated portfolio of satellite observations used for weather, there are a limited number of additional long-term satellite records that have provided significant scientific and decision support information.Figure 2 provides three such examples of long-term satellite records that span a broad range of applications with demonstrated value.Moreover, these examples help to highlight various challenges associated with maintaining observation continuity across multiple contributing satellites, including in the technical and programmatic implementation, as well as with the production and dissemination of the data products used for science and/or decision support.
The top panel shows an overlapping composite of several satellite-based records of northern hemisphere stratospheric ozone spanning over four decades taken from the 2021 State of the Climate report published annually by the American Meteorological Society (AMS-Climate, 2022).Satellite observations of ozone and other trace gases needed to distinguish chemical and dynamical influences on ozone have been critical to: (a) the characterization of the polar "ozone hole" that developed in the late twentieth century due to the anthropogenic destruction of stratospheric ozone, (b) the scientific basis for understanding the development of the ozone hole, and (c) the international approach to stopping and reversing the destruction of Earth's life-protecting ozone layer via the enactment and enforcement of the international Montreal Protocol (Montreal Protocol, 1989).As indicated by the Figure, these same observations continue to be used to evaluate the recovery of stratospheric ozone and thus Satellite observations of global sea level rise from a sequence of nearly identical instruments flown on five different satellites (courtesy, J. Willis/JPL/NASA) and the cover from the 2022 US Interagency Sea Level Rise Technical Report (Sweet et al., 2022) for which the satellite data are used.(bottom) Landsat satellite observations of US land cover change from 1986 to 2020 based on the Land Change Monitoring, Assessment, and Projection (LCMAP) project (USGS, 2022; plot courtesy of LCMAP/USGS EROS) and the cover of the first annual America the Beautiful report (U.S. Dept. of Interior et al., 2022), where land cover change is a chapter based on Landsat-observed landscape changes, and their drivers such as climate change, water availability, disturbance, etc. the effectiveness of the international treaty (WMO-Ozone, 2022); (cover image to right of plot).While this example represents a "success story" for establishing long-record satellite observations of ozone for critical scientific understanding and major national and international policy support-such as might be needed in the future to support carbon dioxide related climate mitigation-it also highlights two challenges: (a) our record of trace gas observations needed to quantify chemical and dynamical contributions to ozone trends and variability is based on a number of overlapping, and in some cases discontinuous, satellite products from various national and international sources with varying levels of fidelity making it challenging to construct an overall assessment of the state of stratospheric ozone and its recovery (WMO-Ozone, 2022) and (b) that there is presently no US or other plan in place to ensure that we have sustained satellite measurements of trace gases other than (total) ozone of the quality needed to continue to monitor processes affecting the stratospheric ozone layer and its recovery into the future.
The middle panel shows a segmented composite of several satellite-based records of globally averaged sea level.This record has been established through a visionary and steadfast partnership initiated by the National Aeronautics and Space Administration (NASA) and the National Centre for Space Studies (CNES) of France, with additional partnership contributions for the more recent data provided by the National Oceanic and Atmospheric Administration (NOAA), EUMETSAT, and the European Space Agency (ESA) co-funded by the European Commission.In contrast to the ozone record, this environmental record has been established over time utilizing the same basic measurement approach and instrument design, lending a high degree of fidelity to the record and providing the means to show clear indications of the upward trend of global sea level, which was around 2 mm per year in the early 1990s and is now greater than 4 mm per year.As with ozone, this multi-decade record of global sea level observations provides the means to establish the scientific basis for sea level rise, understand and characterize its causes, and develop and evaluate models used for future projections of sea level (e.g., Frederikse et al., 2020).Such data are critical for providing the US guidance on high-value infrastructure investments (∼$100 M-$1 B) for coastal adaptation (Sweet et al., 2022; cover image to right of plot).At present, the multi-agency and international partnership mentioned above has committed to the Sentinel-6B satellite, which is to be launched in 2025 with the anticipation that this will continue the record to 2030.Discussions are also underway by EUMETSAT for establishing a mandatory program for altimetry after Sentinel-6B in order to sustain measurements past 2030.While this satellite record of sea level represents a success story for a sustained Earth observation beyond weather, it was developed in an opportunistic manner and in the absence of any specific overarching or guiding national framework.
The bottom panel shows an example time series from the NASA-US Department of Interior (DOI)/US Geological Survey (USGS) Landsat series of land-imaging satellites.Data from Landsat are used widely in the United States and worldwide to support operational decisions and research in areas such as agriculture, forestry and water resources, land use and mapping, geology, hydrology, coastal resources, environmental monitoring, disaster response, and national security (Normand, 2020;Straub et al., 2019;Wulder et al., 2012).In this example, the Landsat data are used to show the net change in major US land-cover types (i.e., trees, crops, grasses/shrubs, developed) from the Land Change Monitoring, Assessment, and Projection's (LCMAP) Annual Land Cover Change Product from the years 1985-2020 (USGS, 2022).In the referenced America the Beautiful report (U.S. Dept. of Interior et al., 2022; cover image to right of plot), Landsat data captured decades of land surface change over time since the 1980s, helped identify drivers of changes such as agricultural and forest land use, climate change, water use, disturbance regime change, and provided a foundation for future risk and vulnerability assessments.The 2008 decision by NASA and DOI/USGS to make Landsat data freely available to the public, and USGS's more recent decision to host the entire Landsat data set in the cloud and implement enhanced data delivery services, have exponentially expanded data access for the US and global land imaging communities (Zhu et al., 2019).Through a NASA-DOI/USGS Memorandum of Understanding, the agencies are implementing a Sustainable Land Imaging (SLI) Program, with Landsat 9 (launched in 2021) as the first SLI mission.Its successor, planned for launch in 2030, will be Landsat Next.Landsat Next will have twice the number of spectral bands, and consist of three smaller satellites that will provide better spatial and temporal resolution than previous Landsat satellites while sustaining existing Landsat data continuity.
These three examples highlight a broad range of benefits associated with long-term satellite observations of the environment, observations that exhibit significant local, national, and/or global utility for science and high-value decision support.Moreover, these examples also highlight needs, opportunities, and challenges associated with sustained observation commitments.It is worth emphasizing that: (a) these three examples are representative of what amounts to a relatively small number of multi-satellite, long-term environmental records in addition to those Earth's Future WALISER 10.1029/2023EF003757 5 of 27 sustained for weather (e.g., polar and geostationary weather satellites), (b) they have each come about through different programmatic means and have different levels of commitment toward continuity, (c) there are a number of quantities measurable from satellites that have been shown, or are anticipated, to have analogous scientific and/ or decision-support value that do not have a plan in place for sustained observations (e.g., precipitation, soil moisture, streamflow, snowpack, greenhouse gas concentrations and emissions, stratospheric ozone, radiation budget, aerosol/cloud profiles, ocean salinity and surface winds), and (d) challenges remain in making data records such as these readily accessible, useable and interoperable for compounded science and decision-support value.
To support and accelerate discussion within the US on the needs, challenges and opportunities associated with sustaining critical space-based Earth observations, the Keck Institute for Space Studies (KISS; kiss.caltech.edu)sponsored a study during 2022 to suggest ways forward.The KISS study team was composed of 32 scientists, technologists and engineers representing multiple Earth mission-focused agencies, institutes, and university departments, including a number of representatives from the National Academies of Science, Engineering and Medicine (NASEM) 2018 Earth Science and Applications from Space (ESAS) Decadal Survey (NASEM, 2018) as well as former and present members of their Committee on Earth Science and Applications from Space (CESAS).The 2022 KISS study program on "continuity" included two full-week, in-person study team workshops, multiple virtual team and sub-team meetings, and three mini-symposia (KISS, 2022) covering: (a) the current and planned space-based Earth observations by NOAA, USGS, NASA, and the international community, (b) Earth observation plans and opportunities that leverage commercial and non-profit developments and entities, and (c) Earth observations in support of climate and environmental security.In addition, various team members held briefings in the weeks in between the two in-person study team workshops to seek input from a number of relevant agencies, entities and/or their representatives (e.g., US Group on Earth Observations (USGEO), Interagency Council for Advancing Meteorological Services (ICAMS), CESAS, Global Climate Observing System (GCOS), Committee of Earth Observation Satellites (CEOS), WGClimate).
Over the course of the study, the KISS team developed a three-pillar approach to the study that is schematically illustrated in Figure 3.The three sections that follow describe the study outcomes and findings for each of the three pillars.Section 2 discusses the needs, challenges and a potential approach for identifying and prioritizing satellite observations warranting continuity.Section 3 discusses observing system architecture considerations,  3) How to Impact and Sustain?These three topics are discussed in Sections 2-4, respectively.The upper portion of the diagram utilizes water, carbon and energy as a simple means to represent the breadth of Earth processes of interest, noting that each contains science and application targets, and that each of these includes many underlying and interacting quantities.
including opportunities and challenges presented by "NewSpace," commercial and NGO (non-governmental organization) data providers, and international partnerships.Section 4 discusses the end-to-end considerations associated with data stewardship, including information production, dissemination and usability.Along with the Section-specific findings given in Sections 2-4, the study team identified the following stage-setting findings that derive from the considerations outlined above:

Identifying and Prioritizing Sustained Observation Needs
This section describes a proposed framework for identifying and prioritizing Earth observations that require continuity to effectively support US environmental security and societal resilience needs.Recognizing the vital role of Earth observing systems and related long-term data records for enhanced understanding of the global climate system, various international and national groups have identified looming continuity gaps in Earth observations.For example, the GCOS 2022 Implementation Plan (Zemp et al., 2022) identified many satellite-based observations at risk for continuity gaps based on subjective expert assessments (see also CEOS/CGMS Gap Analysis, 2018).The US NASEM have also highlighted observational gaps and repeatedly called for US leadership in Earth system science, recommending development of a plan for providing sustained observations (NASEM, 2018;NRC, 2007aNRC, , 2007b; see Section 5).
Though implementing agencies (e.g., NOAA, NASA, USGS) consider expert assessments in their program planning, there is no dedicated process within the US to determine which satellite observation records are (or should be) continued.In 2015, the US NASEM convened an ad hoc committee to develop a framework for analyzing the needs for continuity of NASA-sustained Earth observations from space (NASEM, 2015).To prioritize geophysical variables considered for continuity, the NASEM committee established a quantitative Value Framework, which in part requires subject expert input on the importance, utility and measurement quality required to achieve Earth system science objectives.Such prioritization is inherently complex, multidimensional, and depends on the objectives to be achieved and the intended application areas.This nascent framework was primarily limited to addressing continuity needs and priorities for NASA Earth science.
Here we propose a framework to consider US continuity needs more broadly, namely in support of understanding Earth's climate and to increase societal resilience to threats posed by climate change and other, often related, environmental hazards and extremes (e.g., Figure 1).For this purpose (see Figure 4, left), we propose to use the "Societal Benefit Areas" approach adopted by the US National Earth Observations Task Force, USGEO, and the 2014 US National Plan for Civil Earth Observations (USGEO, 2014).This approach assessed the impact of hundreds of observing systems (satellite and non-satellite) in aiding decision support and delivering benefits to the nation across 13 societal themes and 12 Societal Benefit Areas (SBAs) (additional information in Text S1 of Supporting Information S1).We propose utilization of this SBA framework and its associated assessment procedures to prioritize current and future satellite remote sensing systems by their ability to monitor and predict future climate, as part of their overall assessment against other SBAs and their related public service and research capabilities.Per sector, high-impact observing systems are identified and then prioritized according to relevance across all SBAs (NASA, 2010).
We should note that while the 2019 National Plan for Civil Earth Observations did not expressly promulgate the SBA Framework, it did seek implementation of a "balanced portfolio" of Earth observations, including prioritizing the availability and continuity of Earth observations.And the Plan went on to direct that "…agencies will develop a formal framework and process for addressing continuity challenges and new observing system opportunities that have ramifications for multiple agencies."(NSTC, 2019).USGEO has since established this framework which we believe could provide the appropriate programmatic decision-support forum for longer-term, national-level, space systems planning, coordination, development, and operations.
In addition to this top-level focus on applications-oriented SBAs, there is the fundamental need for foundational science observations.These observations can advance US capabilities for delivering new and/or improved environmental (e.g., weather, marine, air quality) forecasts and climate projections, along with improved public services and research in other SBAs (e.g., Cooke et al., 2014;Weatherhead et al., 2018).We note that foundational science observations can be of two types: (a) "one-off" missions designed to test scientific hypotheses or obtain observations only needed once or very infrequently (e.g., topography under the ice sheets), or (b) continuous observations that aid in monitoring and understanding climate change.The former of these already have mechanisms to consider planning and prioritizing (e.g., NASEM, 2018), while the latter would benefit from the SBA-supported planning/prioritizing approach.

Approach
Linking US satellite observation priorities and related funding to specific US science and public service objectives may allow these objectives to guide the Earth observing systems assessment process conducted by USGEO, which can in turn support continuity of essential space-based observations of climate and other environmental indicators.A structured process for evaluating the impacts of sustained observations could inform a whole-of-government remote sensing systems development approach which accounts for user needs, available funding, technology advancement, and international and commercial partnerships to support data continuity and improvement (see Sections 3 and 4).Recognizing that these US science and service priorities will change over time, as will the technology and approaches for acquiring satellite observations for a given quantity, suggests the need for a nimble method to facilitate the consideration of priorities for observation continuity for a specific purpose and at a specific time.
An approach for addressing the multiple factors for prioritizing observations can be notionally visualized via the three-dimensional graph as shown on the right of Figure 4.The axis labeled "Importance of variable to objective" (x-axis) represents the degree to which that variable contributes to the understanding and knowledge for a specific area, with importance increasing as the along-axis value increases.The specific value for a given variable would be drawn from numerous groups and reports supported by the NASEM, federal agencies, international bodies, and other authoritative groups (e.g., USGEO's Earth Observations Assessment process, the US National Plan for Civil Observations, and the USGEO Satellite Needs Working Group (SNWG) Reports-see usgeo.gov).
The axis labeled "Likelihood of a gap" (y-axis) depicts the degree to which the needed observation continuity is at risk, with risk increasing vertically along the axis.The placement on the y-axis considers both domestic and international investments and commitments from civil, commercial and NGO entities, and the degree to which they support sustained observations of the variable.The axis labeled "Consequence of a gap" (z-axis) captures the consequences of a lack of continuity based on the implications of not having sustained observations.This can include the direct implications of the loss of data for a specific time-including for environmental monitoring, forecasting, decision-support, etc, as well as the indirect impact that losing a segment of an observed record has on the fidelity and characterization of the rest of the record (e.g., masking and/or increasing the uncertainty of trends).In a three-dimensional space (i.e., Figure 4, right), and with values given for each of the three criteria for a number of observations, those plotted furthest from the origin will exhibit the highest priority, as they would be (a) most important to the SBAs and/or to advancing Earth and climate sciences, (b) are at greatest risk for loss of continuity, and (c) produce the greatest consequences from interruption and data gaps.
To capture this in workable and more concrete terms, a matrix for each SBA can be established, in which the rows of the matrix capture variable importance (the furthest to the right being the most important variable), and the columns of the matrix capture the likelihood of a gap (the furthest upward being those for which a continuity gap is most likely to occur).Matrices are established for different levels of "Consequence of a gap" (e.g., minimum, medium, maximum).Since establishing these matrices requires subjective assessments, such an assignment could be done in a working group with appropriate subject matter experts (SMEs), to provide credibility (e.g., NASEM, 2015; USGEO, 2014).The matrices could be reviewed, iterated on, and updated periodically.The intent is to capture underlying rationale for continuity of (and thus potential investment in) each parameter with a US perspective in mind-including the reasons for the variable's importance to the objective(s) at hand, the specific observation requirements needed to support the objective(s), and the degree to which the existing and planned program of record provide for the objective(s) to be met.
Once this multidimensional assessment process has been undertaken and completed, the grouping of variables furthest from the origin (which reflect those considered most important to the objective and most likely to experience a gap in continuity) with "maximum consequence of a gap" could be considered as high priorities.Furthermore, multi-dimensional matrices can be developed for different objectives (e.g., SBAs, advancing understanding of the Earth system, contending with extreme hazard/climate events).These matrices can be developed and then considered in aggregate to highlight the variables that are most important to multiple objectives; these could represent overall high priorities for continuity of satellite observations.The degree of difficulty in addressing a particular gap can then be evaluated by engineering and technological (see Section 3), as well as geophysical, SMEs, and considered as part of developing plans to address priorities amongst multiple potential observation gaps.

Challenges
Prioritization of variables is complex and may benefit from consideration across multiple sectors including supporting services to society, and in some cases helping to advance the underlying science areas.Requirements WALISER 10.1029/2023EF003757 9 of 27 for the satellite observations underlying each variable-which typically include specifications for temporal frequency, spatial coverage and horizontal footprint size, acceptable uncertainty levels on the values, latency, etc.-are dependent on the specific application sector and should be defined at the same time the priority matrix is developed.Requirements for the same variable across different SBAs should be reconciled and the requirements that serve all SBAs could be used if affordability and technological readiness are not an issue.The prioritization framework developed here leverages existing lists of variables that are important to different SBAs and considers the risk of a gap in continuity for each variable and the impact to SBAs if discontinuity occurs.However, priority setting based on this framework is a complex process and is time and context dependent and contains subjective elements.Subject matter experts from different disciplines will need to define priorities using a framework such as proposed here, with consolidation of priorities from multiple SBAs being used to ensure variables that are most important to multiple SBAs are sustained.In practice, more than one group of SMEs could be engaged to provide input on prioritizations in a given area, and agreement across groups could lend confidence in the resulting prioritization.Moreover, given the rapid changes associated with Earth's climate, the evolving needs of our growing society, and the pace of Earth observation implementations by many agencies/countries and the commercial sector, this prioritization activity will likely benefit from ongoing attention and being periodically revisited.
With the above considerations and complexities in mind, a national entity, with representation from multiple US agencies and stakeholder communities, that uses a framework such as described here could regularly evaluate a prioritized list of variables that could benefit from long-term continuity commitments (e.g., somewhat akin to USGEO's SNWG but for continuity concerns; NASEM or ICAMS could offer other means of organizational oversight; see more discussion in Section 5).This national entity could also be charged with reviewing and endorsing proposed continuity commitment plans.The prioritization framework proposed here considers the importance of a given variable for one or more specific objective(s), the likelihood of a gap, and the consequences of a gap, with highest priority for variables that are most important across multiple SBAs and/or underpinning Earth/climate science objectives.Study findings associated with this section's topics are as follows: Finding 2.1.Prioritization of variables requiring continuity of satellite observations is complex and may benefit from consideration across multiple societal sectors and services.The technical requirements on these observations (e.g., temporal and spatial sampling, accuracy, latency) are highly dependent on the specific application sector and/or the underlying supporting science objectives.Finding 2.2.Any prioritization framework will: (a) have subjective elements, (b) be time-and context-dependent due to changing science and societal benefit needs, technological advances and programmatic opportunities, and (c) will likely benefit from periodic reexamination.

Satellite Observing Architectures: Technology, "NewSpace", Commercial and NGO Considerations
A thoughtfully planned observing system architecture may support continuity of the prioritized Earth observations.The architecture could be designed to make use of the capabilities enabled by "NewSpace" developments, opportunities proffered by commercial and NGO data providers, and leverage international partnerships where possible.Here, "NewSpace" refers to the emergence of the private space industry, noting that NewSpace ventures have become much more common for Earth observations, spanning areas such as commercial launch, satellite operations, data reception, processing, and distribution, and value-added services.

Current US Earth Observing Architecture
Current US government-based Earth observing systems include satellites provided by NOAA, USGS and NASA.These systems have been developed over several decades to serve the operational weather forecasting, land imaging, and scientific research communities, respectively.Each of these agencies has its own approach to identify key needs and priorities which inform requirements for observing and data system implementation plans and budget allocations.NASA is responsive to the NASEM Decadal Survey for Earth Science and Applications from Space (NASEM, 2018;NRC, 2007aNRC, , 2007b).Among other guidance, NOAA NESDIS completed the guiding NOAA Satellite Observing System Architecture (NSOSA, 2018).USGS is responsive to Public Law 111-314 recodifying the 1992 Land Remote Sensing Policy Act, as well as successive US National Space Policies.These plans are well coordinated at the international level for the weather enterprise, but are more loosely coordinated for new research and other sustained environmental monitoring objectives.Nonetheless, they have led to continual improvement in observational capabilities for physical monitoring and understanding of the Earth and our local environments.Moreover, sustained commitments for meteorological and land imaging observations, and funding to address Earth science research priorities, have also been achieved (KISS, 2022;NASA Fleet, 2023;NOAA Fleet, 2023;USGS Fleet, 2023).However, these coordination efforts, apart from those supporting the weather enterprise, are not effective as long-term planning tools.In most cases, these efforts lack commitment for sustained long-term observations.In addition, the need for access to space and high-performance instruments to meet the observation and science and/or services requirements makes traditional systems (i.e., those developed by national space agencies) relatively expensive to implement, making it difficult to meet all the priority needs identified by different stakeholders-leaving many gaps (see Section 2).

Evolving Funding and Partnering Landscapes
NewSpace represents emerging approaches that use small satellite technology as an alternative to traditional space observations and services, one with a different but successful business model and engineering culture.These new approaches are marked by lower barriers to entry, agile commercial organizations with higher risk tolerance, and a focus on increased, rapid, and affordable access to space (NASEM, 2022).Additionally, many factors are contributing to fundamentally change the landscape of approaches for funding and implementing Earth-observing satellite systems.These include: (a) lower cost access to space; (b) more affordable high-performance spacecraft and sensor technology; (c) market demand beyond traditional government and science organizations; and (d) a growing concern and interest from philanthropies, non-profits, and other actors to help mitigate and adapt to climate change using satellite observations.
Applying these NewSpace options for Earth observations can augment those offered by NOAA, NASA, USGS and/or their partners, forming a hybrid architecture that leverages the strengths of each.In a few cases, NASA, NOAA, and USGS have already started to embrace commercially provided satellite products within their science and application programs.Table 1 lists the potential support models for the acquisition of Earth observations via satellite(s) and their associated data distribution systems that could be considered in the trade space when selecting an Earth observing architecture for a given variable.
Finding 3.1.One impact of the lower cost of access to space is that many new domestic (e.g., NGOs such as Carbon Mapper and MethaneSat) and international entities (e.g., countries that want to help address climate change that previously could not afford to) are able to contribute elements to the Earth observing system.Future US and international coordination mechanisms for Earth observations could be designed to fully take advantage of these types of contributions.
A more robust research-to-operations (R2O) sensor-and algorithm-development flow, as well as the increased use of fixed price contracts, might allow achieving operational system capabilities for lower cost than the current US approach, enabling additional sustained observations to be added.For example, the architecture of Europe's operational programs has been proven successful and contains both R2O transition of instruments to reduce development risk, as well as fixed price contracts for system delivery.US institutions could be encouraged to think about similar mechanisms to reduce cost overruns while still implementing the most important user requirements.

The Promise and Challenge of NewSpace
The NewSpace ecosystem presents opportunities to support the nation's environmental observing needs, with many potential benefits.For example, NewSpace developments have significantly increased the number of satellite launch opportunities, largely due to the reduction in the cost of the access to space.Just a couple decades ago, with only a handful of large and costly operational launch vehicles, it was difficult to place small satellites into orbits optimized for collecting the measurements for which they had been designed.However, there are now multiple capable launch service companies competing across the spectrum of small, medium, and heavy-lift launch needs (Niederstrasser, 2022).
There are now commercial enterprises demonstrating success at providing space-based observations to government agencies and other stakeholders; only recently this was a domain that was just the purview of government  (GeoOptics, 2023;Spire Global, 2023).Moreover, the lifecycle of these types of missions, from conception to launch and operation, is typically highly compressed in time relative to traditional government-led space missions.Two important consequences result from this faster timeline.One is that NewSpace sensors are typically designed using technologies that would be considered more current.In addition, faster mission execution almost always corresponds to a lower cost mission.The ability to fly more cost-effective satellites can be leveraged in two ways.One is simply to measure the same quantity at a lower cost.The other is to fly a constellation of multiple satellites, thereby improving the space-time sampling properties of the observing system.For certain types of observations, this approach may improve our ability to resolve and better respond to extreme weather and other rapidly evolving environmental events (e.g., wildfires, floods) (e.g., Weatherhead et al., 2018).
NewSpace also presents a number of challenges.Its ability to quickly infuse the latest technology into operational flight sensors means that measurements made over a long period of time by a series of satellites may suffer from inconsistencies as the sensors change and may have significant sensor-dependent calibration uncertainties (e.g., biases due to improper instrument noise floor corrections and scale errors due to improper instrument gain corrections).Traditional sensors developed for high-quality climate observations tend to include on-board calibration systems and undergo extensive pre-launch characterization that mitigate these issues.This is typically not the case with NewSpace missions.The inclusion of NewSpace measurements in future climate-quality continuity missions will likely require an expansion of the role that post-launch calibration and validation (cal/val) typically plays.Such an approach is not without precedent.The GPM Intersatellite Calibration Working Group (XCAL) adopted an extensive cal/val process to produce the IMERG global precipitation product (Berg et al., 2016;Huffman et al., 2020).It did so by using an early example of the hybrid space architecture in which the GPM Microwave Imager (GMI) on the core satellite served as a calibration reference to which the calibration of other sensors in the constellation were tied.The XCAL radiometer intercalibration procedure used by GPM adjusted the Level 1C brightness temperatures measured by each sensor in the constellation.Consistency between sensors at Level 1 is required so that a common precipitation retrieval algorithm can be used for observations by all members of the constellation (Berg et al., 2016;Huffman et al., 2020).It is important to note that GMI did not define measurement "truth."Rather, it provided a stable calibration reference which allowed calibration offsets and drifts in the other sensors to be detected, tracked, and corrected.Additional examples include the vicarious calibration of level 1 radiances measured by geostationary IR imagers to those of hyperspectral IR sounders (e.g., IASI) by GSICS (Goldberg et al., 2011) A second class of NewSpace challenges relates to programmatic differences from traditional missions.The design, pre-flight characterization (e.g., pre-flight calibration results, radiative transfer algorithms, individual sensor bias estimates, etc), and on-orbit operation of sensors and buses used for satellite missions should be open and available to cal/val teams and instrument scientists who use the data.This approach can be at odds with the proprietary interests of some NewSpace organizations.Data buy relationships, in which a commercial entity independently funds a mission and sells its data to the government, may need to establish appropriate contractual understandings for the open distribution of the data.An additional concern is whether data will be available in support of fundamental science studies, such as advancing society's understanding of Earth System processes.
Given the potential benefits and challenges of NewSpace approaches, their deployment could be assessed on a case-by-case basis to determine if they result in the best, or at least an acceptable, solution.

Architecture Considerations
As shown in Figure 5 and described in more detail in Table 1, the implementation options for sustained Earth observations have expanded substantially beyond traditional, large spacecraft, government-funded space systems.These new options include acquiring either single or constellations of small spacecraft; cost sharing with international and/or domestic partners on individual systems or flying joint constellations; and acquiring Earth observation data as a service from commercial or non-profit entities.The implementation option most appropriate for a given observation need will vary based on the requirements for the observing system performance (see Section 2), the current state of technology, industrial policy priorities of the government, and the size of the market for the data beyond government users and services.
A hybrid architecture that combines the lower cost systems offered by NewSpace solutions with traditional missions potentially provides unique insights beyond what individual satellite systems are currently able to provide alone (Dyrud et al., 2013;NASEM, 2022).As part of our objective to assess the potential advantages and disadvantages of NewSpace versus traditional systems, we used rough order of magnitude (ROM) costing techniques to develop a high-level case study that compared candidate NewSpace and traditional missions for the global measurement of precipitation.The key requirements applied in this case study were global coverage at 5 km horizontal resolution and 24-hr revisit time over a 10-year lifetime (NESDIS, 2022).Using a traditional space approach, the mission architecture was determined to require on the order of 5 GPM-class active radar satellites at an estimated average cost of about 1 billion dollars per year for 10 years.An alternative NewSpace approach was also considered which leveraged technology developed by the NASA Earth Science Technology Office, namely the cubesat-enabled Raincube precipitation radar (Peral et al., 2019).This approach required a sustained constellation of on the order of 135 Raincube-class satellites and was estimated to come in at 250-500 million dollars per year for 10 years.Satellite observations for other variables may require different approaches; however, this case study represents the promise that small satellites in the NewSpace community may be able to complement sustained observations from more traditional or other means, and are therefore worth investigating.
For additional supporting information on the above case study, see Text S2 in Supporting Information S1.As ) approaches to Earth science and application mission implementation, while the left (right) side of the diagram denotes the international (national) approaches to mission implementation.This four-quadrant view highlights the opportunity space for implementation of sustained observations.In this depiction, the priorities and capabilities of individual nations are combined to produce global observations not otherwise possible, and large, government sponsored missions are combined with non-profit and commercial data products to deliver sampling densities not otherwise affordable.An efficient and effective HSA would adjust the boundaries between these four quadrants to optimize the strengths of each.
noted in other sections of this paper, beyond doing this kind of ROM cost comparisons to assess potential merit, a more thorough understanding of the advantages and disadvantages of the competing traditional, NewSpace, and hybrid solutions for a given observing system need could be gained from more detailed assessment of calibration and validation issues, system resiliency and technology infusion timeline needs, data stewardship, design studies and associated Observing System Simulation Experiments (e.g., Zeng et al., 2020).
Finding 3.2.Sources of new missions and observing capabilities to address unmet US needs for continuity of Earth observations could be obtained from traditional government acquisition, international partners, commercial entities, NGOs, data purchases, and hybrid solutions (i.e., Table 1).

Data Stewardship and Information Production, Usability and Dissemination
End-to-End stewardship of the data records will require dedicated attention to assure that the final data records will be useful for their intended purposes.The desire to have long-term records making appropriate use of both past and future observations will require focus on four fundamental elements: (a) observation and algorithm provenance, accessibility and archiving; (b) uncertainty quantification and traceability; (c) information production and, (d) product dissemination.These elements have become a critical and challenging part of the considerations concerning Earth observation continuity-both for the heavy technical and programmatic demands they levy on the overall system, and for the increase in the reach and positive impact when judiciously planned and accounted for.Two overarching aspects worth calling out emphasize the outsize role of the so-called "ground data system" segment for state-of-the-art Earth observing and information systems.
The first overarching aspect concerns the vast and still growing amount of data produced by these observing systems.For example, NASA missions launched between 2011 and 2017 are returning about 3 Terabytes of data per day, combined (NASA, 2023).The current generation of NASA Missions launched and launching from 2018 through 2024 will be generating more than 100 Terabytes of data per day-an increase of more than a factor of 30 (NASA, 2023).Similarly, NOAA's recent upgrade in geostationary capabilities to the GOES-R series also resulted in a ∼30-fold increase, from about 60 to 1,800 gigabytes per day (NOAA, 2023b).On top of this massive growth in data rates from single satellite missions is the growth of the overall Earth satellite data collection.Presently, NASA's Earth science data archive from all of its Earth missions flown over the last 30 years is about 40 petabytes; after accounting for data from the next generation of NASA missions through 2025 alone, the archive will grow to greater than 250 petabytes of data (NASA EarthData, 2023).Similarly, NOAA's archive of satellite data is already over 160 petabytes as of 2020 (NCEI, 2023).Add to these the Landsat data archive of several petabytes, and the growing contributions from commercial, NGO and other space agency programs (e.g., EUMETSAT, ESA, JAXA, ISRO, CNES, see Section 3) and the task of the archiving, retrieving, transferring, processing and re-processing, as well as disseminating of the data becomes a tremendous challenge.
The second overarching aspect concerns the users and usage of the data and associated information.No longer are Earth observation data the dominion of tens to hundreds of scientists or government personnel in mission-specific agencies; rather more and more effort is being made to make the data not only open and available (e.g., GSA, 2023) but in a form that is readily understandable, interoperable, and useable by broader segments of society.Some evidence for the growing demand of satellite information of the Earth/environment when the data are in a useable and accessible form, comes from USGS's stewardship of the Landsat record.In 2021, USGS released Landsat Collection 2, which included many improvements to the data set's radiometry, geometry, and metadata characterization.With more emphasis on analysis-ready image files and a transition to hosting the entire archive in the commercial cloud, the USGS has observed a faster growth in data access than at any other time in the 50-year history of the Landsat program.Additionally, since USGS made its Landsat Collection 2 data record available in the cloud, many users have shifted to direct cloud access, whereby they access only image and band subsets in lieu of the entire scene or product bundle.Moreover, from 2020 to 2022, the volume of data delivery requests fulfilled by USGS was 7PB, 20PB, and then 30PB, representing millions of distributed Landsat products.This more than four-fold increase over 3 years stems from both the growing demand for environmental monitoring and decision-making data, but also from the vast advances-and advantages-in information technology (e.g., cloud storage and computing, faster network speeds, more nimble and capable software) that enables this exponential growth in use.
As society continues to wrestle with, and work to provide services for, the challenges of environmental/climate changes, hazards and extremes (e.g., Figure 1; Section 2), the demands for Earth observation data and knowledge will continue to grow exponentially and the attendant information technology elements for data stewardship will need to be designed, scaled and supported accordingly.

Earth Observation Stewardship
Originally restricted to the functions of long-term data preservation and archiving (Albani & Maggio, 2020), the concept of data stewardship has matured to include "all activities that preserve and improve the information content, accessibility, and usability of data and metadata" (NRC, 2007a(NRC, , 2007b;;NRC, 2012;Peng et al., 2015).This wide remit encompasses a range of processes designed to improve both the functional utility of the data itself as well as the experience of present and future users.As a consequence, stewardship activities require a systematic, end-to-end approach from the development and deployment of flight hardware through the collection, preservation, and distribution of data products.An important design goal for science and services information produced from Earth observations (independent of the many possible sources-see Section 3) is that it be trusted and representative (Peng et al., 2016).The data and information should also be easy to find, readily accessible, and manageable with modest compute capabilities and open-source software (see Findable, Accessible, Interoperable, and Reusable principles, Wilkinson et al., 2016), albeit complex and comprehensive science-grade analyses of the breadth and amount of data (mentioned above) are expected to demand very high capability storage, network, and compute facilities.
With the above data principles in place and operating, effective stewardship of satellite data records will minimize the barrier to entry for the user ecosystem-which as mentioned above should be expected to include civil and commercial uses by both scientists and service providers.Stewardship efforts should focus on making the observations and important derived quantities interoperable regardless of measurement technique while preserving provenance and quantifying data set uncertainties.For many downstream science and application uses, users should not need to familiarize themselves with the intricacies of specific measurement biases or expend excess effort stitching together data sets to utilize for long-term climate information.Additionally, stewardship of sustained Earth observations extends beyond mere technical activities and best practices (discussed more fully below) to include clarity and longevity of the associated programmatic support and responsibility.The schematic of Figure 6 illustrates the key elements that support modern stewardship of sustained satellite-based Earth observation, with the implication that sustained investments in these elements results in a much greater benefit to society (solid lines and graphic from Figure 4) over intermittent, piecemeal investments (dotted lines).

Technical Considerations for Establishing and Maintaining Satellite Continuity
Transforming satellite-based Earth observations into useable information products requires multiple steps within a value chain (see Table 2), including several levels of data processing (NASA-ESDS, 2023).Throughout the programmatic lifecycle of a single mission or across multiple similar or complementary missions, many of these steps evolve to reflect the advancing state of the art in processing technology, scientific objective, and/or societal application.Changes in sensor technology, satellite architecture, algorithms, calibration/validation, processing, and quality control can fundamentally alter the fidelity, and in some cases the representativeness and level of uncertainty, associated with the observed parameter across the duration of the mission(s) (cf.Section 3.3).These alterations can have downstream impacts for users who use the data well after the mission's lifetime.(e.g., retrospective climate analysis, retrospective re-analysis via data assimilation practices).In a stewardship framework that addresses continuity concerns, each element of the value chain, and the teams/entities that own and address them, must be integrated with each other and adaptable to advances in technology, best practices and uses for the data.The agencies or commercial entities that develop and operate the mission(s) ought to consider these factors on a regular cadence during and after each mission's end-of-life through dedicated review processes.Table 2 illustrates the expansive nature of the data production elements that are critical to the successful implementation and utilization of individual missions, and with continuity across multiple missions in mind.

Specific Considerations for Climate Science and Service Applications
Improved end-to-end planning and implementation that takes into account the entire stewardship value chain (see Table 2; Figure 6) is particularly important for climate sciences and services applications.Many data sets reveal important information only when collected and stewarded over long (multi-decadal) time periods.By definition, observations suitable for climate science and services imply that the data records will be sustained for longer than the lifetime of individual observing assets, will always exhibit retrospective value and therefore require safeguarding.For climate-related science and service applications, stewardship aimed at limiting discontinuities is particularly crucial.Moreover, through the production of retrospective state-estimations, that is, re-analyses, that utilize data assimilation techniques to optimally bridge physical model knowledge with observation information, the impacts of gaps or discontinuities in sensors can be minimized, albeit this typically requires applying the aspects of observation stewardship emphasized here.
Discontinuities of Earth observations can result from a number of different elements within the data production value chain (Table 2) with varying degrees of consequence.At the sensor measurement level, the consequences of gaps, defined as periods of time during which no similar observations are collected, can be severe (e.g., Free et al., 2002).Along with the outright missing characterization of the environmental quantity/process for which the observation is used for (Ohring et al., 2005;Wang & Liu, 2020), the gap can limit the useability and/or fidelity of the data before and after the gap (e.g., Weatherhead et al., 2017).That is because climate data records often rely on overlap of measurements in two or more data records in order to robustly stitch them together, properly account for calibration to a common scale-sometimes tied to ground-based or in situ data (Diamond et al., 2013;Dirksen et al., 2014) and quantify uncertainties in the final data record (cf., Madonna et al., 2022).Figure 7 demonstrates this characteristic for space-based Total Solar Irradiance (TSI) observations.TSI is a fundamental climate measurement that constrains Earth's energy budget, which is critical to accurate climate forecasting.TSI observations are especially sensitive to instrument-dependent calibration offsets, and simultaneous observations are required to minimize discontinuities from mission to mission.In cases of actual measurement gaps, the record begins anew, and making robust links to the record prior to a gap can be difficult and in any case result in significantly more uncertainty in assessments of trends and extremes in the record.Gap-filling methods that use proxies or models can also add significant uncertainty to the record.For example, current estimates for filling a gap using narrow band visible or infrared image retrievals (rather than the more fitting broadband radiation measurements) for the Earth Radiation Budget (ERB) record acquired by NASA's CERES instruments since 2000 would add uncertainty of 0.4 Wm −2 -about the same magnitude of the best estimates of decadal trends in anthropogenic radiative forcing (e.g., Loeb et al., 2018Loeb et al., , 2021;;Myhre et al., 2014).Thus, the uncertainty introduced by addressing a gap can be as big as the signal of the quantity of interest-for instance, quantifying the changes in Earth's radiation budget which dictates the rate of Earth's warming.Emerging observational techniques, including new calibration methods (Fox et al., 2011;Wielicki et al., 2013) and a range of radio occultation techniques (e.g., Leontiev & Reuveni, 2017; Steiner et al., 2020) offer hope of new continuity approaches but have yet to be fully implemented or tested.
Discontinuities can also exist at the data management level, where continuous, overlapping observations are not readily interoperable, accessible, or understood across the time series.These types of discontinuities can result in significant barriers to users, can directly devalue the observations' utility, and make retrospective stitching of the historical data together significantly more time consuming, expensive, and introduce additional uncertainty into the data record.For example, several satellite-derived sea ice data records exhibited large increases in Antarctic sea ice coverage area in 2009.With adept interrogation, and through the use of additional sources of data, these "increases" were found to be an artifact, and instead attributable to uncharacterized/unquantified differences in the underlying data sets that were stitched together to produce a sea ice climate record (Screen, 2011).Similarly, the stitching together and calibration of the multiple ozone measurements shown in Figure 2 requires significant effort and expertise to make them compatible, quantify their uncertainties, and then use for assessing the state of recovery of the ozone hole (WMO-Ozone, 2022).Appropriate merging of overlapping satellites, either statistically, through data assimilation or with external "truthing" efforts, requires notable efforts to avoid additional uncertainty on trends potentially larger than the record is designed to detect (Weatherhead et al., 2017).
Resources for the calibration and validation of Earth observation data sets are often tied to the active operational period of the satellite mission, and as such, expertise in the technologies and techniques of the project is often lost when the mission ends.In order to stitch data records together retrospectively, individual investigators or new programs are often required to recreate expertise over multiple program histories, technologies, and measurement and algorithm design choices, resulting in, at a minimum, discontinuities in data quality and/or uncertainty characterization.
This ad hoc approach for developing continuous climate records is financially, programmatically and technically challenging in the best cases, and can lead to the loss of critical Earth and environmental information in the worst cases.Funding the organizational structure and data stewardship independent of individual missions, and managing at a higher programmatic level that encompasses all relevant past, current and future missions that contribute to the given long term data record could better support continuity of Earth observations (Figure 6).and Joint Polar Satellite System (JPSS).At the core of these efforts was a programmatic structure to address the need for cross-mission radiometric continuity as well as common algorithms for geophysical products, further enabled by support from the Earth Science Data Systems program for data processing, archiving, and distribution (NASEM, 2015).
Finding 4.2.For climate datasets, the value to science and society accrues with longevity, so stewardship and the necessary technical and programmatic structures needed to support it, require an enduring commitment that should be independent of individual missions.Investing in data usability, traceability, provenance, and interoperability capabilities can greatly enhance the return on the given civil or commercial investments made to deploy the observing system (Figure 6).

Data Accessibility and Dissemination
Data accessibility and dissemination is the final stewardship element for supporting the science and services applications based on Earth observation  (Knezek, 2020).However, the US has yet to fully consider an expanded vision of the future of Earth observations where much of the disparate data are provided via a readily accessible data cloud or a more distributed yet secure virtual data storefront for external users.Such a system, when coupled with compute, analysis and visualization capabilities, will enable the joint harnessing of what would otherwise be disparate data, facilitating cross validation and quality assurance of data sets, and providing for the development and use of a broad ecosystem of services and value-added applications.
Finding 4.3.While strides have been made by individual US agencies to provide more ready access to Earth observation datasets, full exploitation of the data and associated investments for US civil and commercial interests and services suggests a more holistic stewardship approach providing the means for platforms where observations and models reside together in an easily accessible and manipulatable form and the latest analysis techniques, such as machine learning and artificial intelligence, can be applied to entire observational records.

Summary and Path Forward
Earth observations form the foundation for a rapidly expanding set of public and commercial services critical for hazard response, agriculture, transportation, health and environmental stewardship, as well as climate change adaptation and mitigation (e.g., Figure 1; Figure 4; Finding 1.1).For all these needs, space-based Earth observations represent an essential component of the infrastructure needed to support the delivery of critical science and decision-support information with local, national, and global utility (e.g., Figure 2; Finding 1.2).Recognition of the value of satellite observations for monitoring and forecasting weather is evident through our nation's sustained support of NOAA's suite of weather-observing satellites (NOAA Fleet, 2023), which amounts to about $2 B/year in annual spending and stems in part from directives and commitments laid out in our National Space Policy (NSP) (NOAA-OSC, 2023).The NSP has similar directives and commitments for land imaging, with varying but sustained support of about $150-200 M/year in recent years for the NASA-USGS SLI program (CRS, 2021, Figure 2, bottom).Along with these two areas of sustained satellite observing programs are US investments in developing research-focused satellite observations for advancing our scientific understanding of the Earth and our climate, mostly through support of NASA, and amounting to about $2 B/year (NASEM, 2018).For the most part, these missions tend to be focused, one-of-a-kind developments for the purposes of addressing a range of Earth system science and climate science objectives (NRC, 2007a(NRC, , 2007b;;NASEM, 2018), and not the basis of an Earth observation program sufficient and overtly intended to support sustained long-term measurement components.Follow-on missions have been developed in an ad hoc manner to extend a few satellite records for long time-scale Earth/climate science questions and sustained applications use.Examples include the transition of some NASA EOS observations to NOAA's SNPP/JPSS programs (e.g., VIIRS for MODIS, CrIS for AIRS), and other extensions for glacier and ice sheet (e.g., ICESat-2), sea level (i.e., Jason-2, -3, Sentinel-6A/B), and gravity (i.e., GRACE-FO) observations, with the latter two heavily leveraging international partnerships.With such an approach, it may be difficult to develop an Earth observation system and associated information stream to meet the growing decision-support needs of the nation.
The challenge and opportunity we face is that there are a number of additional quantities measurable from satellites that have been shown, or are anticipated, to have ongoing science and societal benefit value but for which there is no plan for sustained observations (Finding 1.3).These include quantities like precipitation, soil moisture, streamflow, snowpack, carbon emissions, ocean salinity and surface winds, and stratospheric ozone.However, because the driving needs and benefits for many of these quantities lie outside of "weather," "land imaging" or single-satellite science missions, the US agency remit is unclear when it comes to the development and implementation of these types of sustained satellite observations.More directly, it is not obvious yet, what US agency is responsible for sustained satellite observations supporting the domains of hazard response (e.g., wildfires, earthquakes), climate change science and projection (e.g., ice sheets and sea ice, Earth radiation budget, ozone and greenhouse gases), environmental monitoring and services (e.g., freshwater, biodiversity), or energy and environmental policy and regulation (e.g., carbon emissions).At this time, the US does not have an overarching, agreed-upon plan or framework to: (a) develop and maintain a set of national priorities for sustained Earth observations, (b) adeptly develop and implement observing architectures to address those priorities, taking into account a mix of government, commercial, NGO, and international solutions, and (c) provide the means to steward and maximally exploit these types of sustained observations for advancing long-term science objectives and for public good (Finding 1.4; see Figure 8 bottom).
This point of contemplation and consideration for action is not without precedent.et al., 2017).
With the above two paragraphs in mind, namely Study Findings 1.1-1.4 and reflection on the EU's Copernicus Program, the following questions may be worth considering from a US perspective: • What are US national priorities for sustained Earth observations?
• What paradigm would the US use as the basis for setting these national priorities?
• What organization or body is, or would be, chartered to develop these priorities?
• What is our national approach to implementing sustained Earth observations that meet these priorities, including the information production and delivery services?• Does the US see itself developing a Copernicus-like program of its own or does it see itself contributing to and leveraging, to the extent possible, EU's Copernicus Program-along with other international partnerships, for our national needs?• For either of the above approaches, what national framework or plan will the US rely on to interface with the EU and its Copernicus Program as well as with other international partners intent on providing sustained Earth observations and related services?
While answers to these questions look to be the purview of the NASEM's ESAS Decadal Survey, neither the 2007 or 2018 surveys (NASEM, 2018;NRC, 2007aNRC, , 2007b) ) provided answers to the above questions given that their statement of tasks did not include these types of questions or a directive to consider the whole-of-government approach to continuity of critical space-based Earth observations; similarly, this was the case for the NASEM continuity study (NASEM, 2015(NASEM, , 2018;;NRC, 2007aNRC, , 2007b)).Aspects of continuity were addressed in limited fashion through the 2017 Designated Observable recommendations that include one-off, next-in-sequence elements of continuity (e.g., Mass Change mission to follow GRACE-FO, Surface Deformation and Change mission to follow NISAR), the recommendation for an Earth Ventures Continuity strand to explore more feasible demonstration missions for continuity (at a pace of 1 demonstration mission/observable every 4 years), and a number of programmatic findings and recommendations, including guidance concerning the establishment of clearer national and agency priorities and roles.The latter includes: (a) Recommendation 2.2."NASA-with NOAA and USGS participation-should engage in a formal planning effort with international partners to agree on a set of measurements requiring long-term continuity and to develop collaborative plans for implementing the missions needed to satisfy those needs.This effort to institutionalize the sustained measurement record of required parameters should involve the scientific community, and build on and complement the existing domestic and international Program of Record," and (b) the following statement in Recommendation 4.6, "Lead development of a more formal continuity decision process to determine which satellite measurements have the highest priority for continuation, then work with US and international partners to develop an international strategy for obtaining and sharing those measurements".
In reference to the Copernicus Program, the 2017 Decadal Survey (NASEM, 2018) does discuss the value of Copernicus with the following concluding statement: "Given that the United States has no equivalent capability to this operational Earth observation and monitoring program in Europe, the committee recognizes the importance of Copernicus in general and the Sentinels in particular as a long-term, continuing source of a variety of important observations…If the United States cannot replicate an effort like Copernicus and the Sentinels, US agencies would benefit substantially from exploring options for complementing and strengthening this European effort, such as is being done by NASA and NOAA with the JASON-CS satellite partnership for Sentinel-6."While this finding in the 2017 Decadal Survey has provided the wherewithal to partner and sustain observations of sea level (see Figure 2, middle), it also points to the promising pathway of international partnership to develop opportunities for additional sustained observations (e.g., Sections 3.2 and 3.4).However, due to the scope and objectives outlined by the statement of task for the 2007 and 2017 Decadal Survey, it wasn't in scope for the Survey to provide holistic guidance regarding the development of a US equivalent of Copernicus or a means to establishing the priorities and protocols for engagement and partnering with Copernicus based on a set of agreed upon US priorities for sustained Earth observations, leading to the following Study findings: Finding 5.1.The US could benefit from a systematic and overarching plan or framework for identifying, prioritizing, funding, and implementing sustained Earth observations that are critical for supporting our nation's science, policy, and societal resilience goals.Finding 5.2.A clear and unified approach to sustained Earth observations and determination of our national priorities for these observations may improve the effectiveness of the varied US investments in Earth observations and associated information systems.Such an approach may also enable the United States to play a larger global leadership role in environmental stewardship, Earth system and climate science, and related public services.

Next Steps
To aid and accelerate our nation's discussion on the above needs, challenges and opportunities associated with developing and sustaining critical space-based Earth observations, this KISS study was conducted to offer ways forward on this topic of satellite "continuity" with the US perspective in mind.The KISS Study team developed a three-pillar approach to the study that is schematically illustrated in Figure 3. Considering the discussion above and the Study's findings across these three pillars, suggests that a more nimble and responsive coordination framework to help address US needs regarding sustained, space-based Earth observations would be beneficial.This coordination framework could include consideration of: 1. approaches to identifying and prioritizing satellite observation needed to meet US needs for science and services (see Section 2, Figure 4, Findings 2.1, 2.2), 2. the rapidly evolving landscape of space-based Earth viewing architecture options and technology improvements; the increasing opportunities and lower cost access to space; and growing availability and potential for commercial, NGO and international contributions (see Section 3, Figure 5, Table 1, Findings 3.1, 3.2), and 3. the technical and programmatic underpinnings for proper and comprehensive data stewardship, including information product development and dissemination with a broad science and services user base in mind, traceable and accurate uncertainty quantification, and the integration with ground observations and Earth system modeling and prediction tools via data assimilation and other data fusion techniques (see Section 4, Figure 6, Table 2, Findings 4.1, 4.2).
While the demands and opportunities for using Earth observations for science as well as public and commercial services are accelerating, it is fortunate that this growth is occurring in concert with increasing opportunities and lower barriers for developing and sustaining such observations and services.These opportunities are created through lower-cost access to space and recent technology and infrastructure improvements that can support the sustained production and dissemination of the needed environmental, climate and related security information (e.g., Figures 5 and 6).With the continued and growing societal concerns related to climate change and national resilience, we foresee this topic representing an important point of discussion in both the upcoming mid-term review of the 2017 ESAS Decadal Survey as well as the 2027 ESAS Decadal Survey.
There are two additional elements that motivate the development of a more expansive and sustained EOS.The first comes from the National Security sector.The latest report from the National Intelligence Council (NIC) on Climate Change and International Responses Increasing Challenges to US National Security Through 2040 (Price et al., 2021) includes the following statement: "We assess that climate change will increasingly exacerbate risks to US national security interests as the physical impacts increase and geopolitical tensions mount about how to respond to the challenge.… Intensifying physical effects will exacerbate geopolitical flashpoints, particularly after 2030, and key countries and regions will face increasing risks of instability and need for humanitarian assistance."In support of this statement, the NIC highlighted specific geopolitical tensions and flashpoints, their severity and anticipated timelines, including those that stem from climate change impacts on country-level stability (see Text S3 in Supporting Information S1, upper).In addition, they tie many of these security risks to geophysical quantities and processes (see Text S3 in Supporting Information S1, lower) that can be monitored, better understood, and possibly predicted to some degree, with the support of the types of Earth observations discussed in this study.As mentioned in Section 1, the study team, with the help of the staff and a member of NASEM's Climate Security Roundtable staff, organized a symposium on the topic of Earth observations in support of climate and environmental security (KISS, 2022).While the presentations and discussions associated with this symposium acutely illustrated the important value of environmental situational awareness for climate and environmental security concerns and actions, they also pointed to disconnects in language, concepts and common knowledge between the civil Earth observation and national security communities.Shoring up these disconnects is a valuable growth area for further discussion and dialog to support national and international security concerns.
The value of an enhanced and robust continuity framework for global Earth observations also underpins the rapid development of the "climate economy," a constellation of goods and services for facilitating sustainable and resilient growth, equipping the nation to withstand the impacts of a changing climate (Saha & Jaeger, 2020).This sector is poised for substantial growth, with major policy initiatives such as the Inflation Reduction Act attracting a broad coalition of business interests into the climate risk arena (Meyer, 2022).Earth observations enable verifiable and actionable information for climate finance, sustainable infrastructure development, carbon measurements and attribution, and intelligent resource management.In addition, climate information is increasingly valuable for traditional economic sectors, such as agriculture, transportation, and financial services by empowering market participants with risk awareness for more efficient and effective decisions.

Figure 1 .
Figure 1.The yearly count of billion-dollar weather disasters in the United States since 1980, adjusted for inflation.2020 set a new record for the most such disasters in any year.The left axis and colored bars show the number of billion-dollar events for each year, with the color indicating the type of disaster.The right axis and colored lines show the disaster costs in billions of dollars-with the red line being the aggregate cost for the given year, the gray lines being 95% confidence limits, and the black line a running-5-year average cost.Source NOAA's National Centers for Environmental Information (NOAA, 2023c).

Figure 2 .
Figure 2. (top) Satellite observations of northern mid-latitude (35°-60°N) stratospheric ozone (at around 42 km or 2 hPa) from multiple satellite or satellite-infused sources (AMS-Climate, 2022) and the cover of the 2018 quadrennial review from the Scientific Assessment Panel of the Montreal Protocol (WMO-Ozone, 2022) for which the satellite data are used.(middle) Satellite observations of global sea level rise from a sequence of nearly identical instruments flown on five different satellites (courtesy, J. Willis/JPL/NASA) and the cover from the 2022 US Interagency Sea Level Rise Technical Report (Sweet et al., 2022) for which the satellite data are used.(bottom) Landsat satellite observations of US land cover change from 1986 to 2020 based on the Land Change Monitoring, Assessment, and Projection (LCMAP) project (USGS, 2022; plot courtesy of LCMAP/USGS EROS) and the cover of the first annual America the Beautiful report (U.S. Dept. of Interior et al., 2022), where land cover change is a chapter based on Landsat-observed landscape changes, and their drivers such as climate change, water availability, disturbance, etc.

Figure 3 .
Figure 3. Keck Institute for Space Sciences (KISS) Study Team's three-pillar approach to the Continuity Study focused on (1) Why & What to Include, (2) How to Include, and (3) How to Impact and Sustain?These three topics are discussed in Sections 2-4, respectively.The upper portion of the diagram utilizes water, carbon and energy as a simple means to represent the breadth of Earth processes of interest, noting that each contains science and application targets, and that each of these includes many underlying and interacting quantities.

Finding 1. 1 .
There is growing urgency for improved public and commercial services to support a resilient, secure, and thriving US population and economy, particularly in the face of mounting decision-support needs for environmental stewardship and hazard response, and for climate change adaptation and mitigation actions (e.g., FFAPCS, 2023).Finding 1.2.Space-based Earth observations represent an essential component of the infrastructure needed to support the delivery of critical environmental science and decision-support information with local, national, and global utility.Finding 1.3.

Figure 4 .
Figure 4.A framework to identify and prioritize sustained space-based Earth observations.This framework considers information for a sampling of Societal Benefit Areas (SBAs) (SBA) to provide public and/or commercial services in support of US resilience, prosperity and national security, while continuing foundational observations to support advancing Earth and climate sciences (left).In prioritizing sustained Earth observation variables, considerations may include, among other factors, their importance to national objectives, likelihood of a gap, and consequence of a gap for each service and/or SBA considered (right).Earth observation variables are represented by the different "dots," with different symbol types (e.g., cube, sphere) depicting different SBAs or services, and the different colors represent priority for continuity.Variables that are furthest from the origin for all three axes (i.e., red color) indicate higher priority for continuity.

Figure 5 .
Figure 5. Schematic illustration of a hybrid space architecture (HSA) that provides multiple new dimensions of opportunities for sustained space-based observations of essential climate and other environmental variables.The upper (lower) part of the diagram includes traditional (non-traditional) approaches to Earth science and application mission implementation, while the left (right) side of the diagram denotes the international (national) approaches to mission implementation.This four-quadrant view highlights the opportunity space for implementation of sustained observations.In this depiction, the priorities and capabilities of individual nations are combined to produce global observations not otherwise possible, and large, government sponsored missions are combined with non-profit and commercial data products to deliver sampling densities not otherwise affordable.An efficient and effective HSA would adjust the boundaries between these four quadrants to optimize the strengths of each.

Figure 6 .
Figure 6.A US framework for carefully coordinated, end-to-end stewardship would increase the value of past, current and future Earth observations for all Americans.Sustained Earth observations, made up of evolving spacecraft and sensors, are represented by the gray bars at the bottom.Key elements of modern data stewardship are represented by the multi-color shaded box.The benefits to society, supported by the sustained observations and stewardship elements, are represented by the green lines and the simplified form of the societal benefit graphic from Figure 4. Red (green) lines represent investments in (societal benefits from) data stewardship.Solid (dotted) lines represent sustained (intermittent, piecemeal) investments in stewardship.

Finding 4. 1 .
A framework for successful stewardship of sustained Earth observations requires end-to-end planning with a long-term horizon in mind (i.e., well beyond individual satellite mission lifetimes), a suite of technical attributes that support open and easy access, interoperability of related observations, as well as carefully coordinated and sustained programmatic structures that provide the needed shepherding and support.Data production elements Critical continuity considerations Sensor Technology and Characteristics • Representativeness of the sensor measurements (e.g., spectral channels, sampling, etc.) • Known sensor characteristics and uncertainties Satellite Observing System Architecture • Time-of-day or viewing geometry impacts • Stability of satellite orbit • Swath width and revisit rate • Launch cadence and gap risk posture Algorithm Development, Updates, Documentation • Commonality in radiometric and geophysical algorithms (e.g., forward radiative models and inversion techniques, ancillary data sets, etc.) Calibration and Validation • Documentation of calibration techniques, inclusive of pre-launch characterization and on-orbit characterization • Consistency across validation protocols and ground truths (Re-)Processing Demands and Cadence • Consistent geolocation and grids • Lineage and preservation of original (i.e.Level 1) data for retrospective reprocessing • Compute, storage and access capability for re-processing • Compute, storage and access capabilities for utilizing data from multiple missions and programs.Data and Information Product Quality Control • Documentation and evaluation of systematic impacts from the chosen filtering approaches • Uncertainty quantification and traceability for data products Data Archive and Dissemination • Management of interfaces under common APIs • Common data and metadata formats • Provenance tracking • Unique digital object identifier Usability and User Ecosystem • Co-location of data across missions, timeseries, and programs • Curation of community access and development tools • Interoperability of use across different observation types • Permissive licensing regimes for commercially acquired data sets

Figure 7 .
Figure 7.An example of the importance of climate observation continuity.The figure on the left shows incoming radiative energy from the Sun, a fundamental climate data record, from the set of space-based observations since 1978.The observations are shown at each instrument's native calibration scale.Because the individual observation records overlap, it enables the construction of the composite record on the right where all observations are on a common scale.This helps reveal trends in the long-term data record that may have impacts on climate.Figure from Kopp (2023).

Figure 8 .
Figure 8. Schematic representation of United States and European approaches to satellite missions for advancing science (left shaded boxes), sustained satellite programs for weather forecasting (right shaded boxes), and other missions outside these two areas that require sustained commitments and continuity across multiple missions (middle shaded boxes).

Finding 1.4. The
Many quantities measurable from satellites that have been shown to have scientific and/or decision-support value do not have a plan for sustained observations.US does not have a systematic, overarching plan or framework for identifying, prioritizing, funding, and implementing additional sustained Earth observations to support our nation's science, policy, and societal resilience goals.

Table 1
Summary of Current and Potential Future Support Models for the Acquisition of Earth Observations via Satellite and the Associated Data Distribution Systems 23284277, 2024, 2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023EF003757by Nasa Jet Propulsion Lab, Wiley Online Library on [12/02/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Table 1 Continued
An example is the recent success of Spire and GeoOptics Radio Occultation observations, that include NOAA and NASA as customers for the data 23284277, 2024, 2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023EF003757by Nasa Jet Propulsion Lab, Wiley Online Library on [12/02/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License space agencies.

Table 2
Key Elements of the Data Production Value Chain of Satellite-Based Earth Observations 23284277, 2024, 2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023EF003757by Nasa Jet Propulsion Lab, Wiley Online Library on [12/02/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License In certain cases, this has been recognized by US agencies.For instance, since 2013, the NASA Earth Science Division Research and Analysis program has funded algorithm continuity development and assessments through competed activities to bridge NASA's legacy Earth Observing System (EOS) assets to NOAA's next-generation Suomi-National Polar-Orbiting Operational Environmental Satellite System Preparatory Project (Suomi NPP) . US government agencies have made laudable progress in enhancing the use, usability and utility of Earth observations data.As indicated above, USGS's commitment to free and open data, long-term data continuity and consistency, as well as the development of standardized APIs and analysis-ready data sets have greatly expanded the utility of the 50-year Landsat record.NOAA's Open Data Dissemination Program (NOAA-ODD, 2023) has democratized data access and discovery by making high-interest data sets available in public commercial clouds.NASA's Transform to Open Science (TOPS) (NASA-TOPS, 2023) initiative will accelerate innovation by reducing the barrier to access NASA data and enhance transparency and reproducibility of science products.Likewise, US agencies have begun to grapple with many of the systematic, technical and cultural challenges for managing this new era of data accessibility, such as through NASA's Strategy for Data Management and Computing for Groundbreaking Science Weatherhead et al. (2018)observations required to deliver the public benefits of the Copernicus Program-organized along the themes of Atmosphere, Marine, Land, Emergency, Climate, and Security Services(European Commission, 2023).For the most part, ESA-along with Europe's space industry-is responsible for developing the space component of Copernicus, and ESA, EUMETSAT and ECMWF are largely responsible for producing and delivering the information to support its Services component.Copernicus costs from its initiation in 1998-2020 are estimated at €6.7 B with around €4.3 B spent in the period 2014 to 2020 (or about $600 M US per year), with the benefits of the program to the EU economy estimated at €30 billion through 2030(European Commission, 2014).This factor of a ∼five-fold return on investment (ROI) is on the same order as that estimated by the Global Center for Adaptation for investments in infrastructure, water management resources and early warning systems to support climate change resilience (Global Commission on Adaptation, 2019).In addition,Weatherhead et al. (2018)provide what they considered as a conservative estimate of $50 return for every $1 invested in a significantly improved climate observing system.More recent efforts organized by Resources for the Future, summarized in the Valuables Project Report, reinforce those values and indicate that higher ROI may well be justified for the more general category of Earth Observations (Kuwayama The European UnionJutz & Milagro-Pérez, 2020;Thépaut et al., 2018)e development of its Copernicus Program(Copernicus, 2023;Jutz & Milagro-Pérez, 2020;Thépaut et al., 2018).Copernicus is the EU's Earth observation program (see Figure8top) that looks "at our planet and its environment to benefit all European citizens.It offers information services that draw from satellite Earth Observation and in-situ (non-space) data…to help service providers, public authorities, and other international organizations improve European citizens' quality of life and beyond.The European Commission manages the Program; it is implemented in partnership with the Member States, the ESA, EUMET-SAT, the European Centre for Medium-Range Weather Forecasts (ECMWF), and EU Agencies and Mercator Océan.The information services provided by Copernicus are free and openly accessible to users."Copernicus' satellite missions, the Sentinel satellites,