The National Ecological Observatory Network (NEON) collects and provides free and open high-quality continental and decadal-scale data for fundamental research on ecological systems (Keller et al. 2008). NEON is solely funded by the US National Science Foundation (NSF) and operated by Battelle. Data and samples are collected from 81 terrestrial and aquatic field sites, distributed throughout the continental US, Alaska, Hawaii, and Puerto Rico (see neonscience.org for more details). NEON provides >180 open datasets that capture environmental, biogeochemical, atmospheric, and taxonomic data from various ecosystems that are subject to an array of land management and other environmental impacts.

The NEON design is intended to enable scientific research that furthers our understanding of the interactions between the drivers of environmental change and its biological responses, including biodiversity (Schimel et al. 2011). NEON’s sampling design is distinguished by the use of standardized methods across space and time and the coordinated, collocated collection of diverse and complementary datasets using instrumented, observational, and airborne remote sensing approaches (Fig. 1) (Meier et al. 2023). The standardized, spatially balanced designs underpinning NEON’s data collection efforts are necessary for enabling regional to continental-scale analyses, as distinct from most other networks, such as the US Long-Term Ecological Research (LTER) network that was designed to investigate the mechanisms and dynamics of local-scale ecosystem processes (Hobbie et al. 2003). The differing approaches of NEON and other networks like LTER are highly complementary and critical to informing global solutions to ongoing ecological challenges.

Figure 1. Schematic of idealized collocated aquatic and terrestrial field sites within the US National Ecological Observatory Network, illustrating how the three collection systems, the observational sampling, instrument systems, and airborne remote sensing, are spatially integrated. Image credit: National Ecological Observatory Network/Battelle. By Williams C.

NEON provides both data and samples that are suited to addressing many aspects of biodiversity and that span all taxonomic domains of life (Fig. 2). Indeed, biodiversity was one of the ‘Grand Challenges in Environmental Sciences’ (National Research Council 2001) that served as the foundation for NEON’s design. In aquatic freshwater ecosystems (represented by 34 field sites), NEON collects periphyton, phytoplankton, plants, macroinvertebrates, zooplankton, fish, and microbial samples from benthic and surface water that are sequenced for archaea, bacteria, and fungi (Parker and Utz 2022). Terrestrial sampling at 47 field sites targets plant phenology, diversity, biomass, and productivity, as well as the abundance, diversity, and (for select groups) pathogen status of birds, small mammals, ticks, mosquitoes, ground beetles, and soil microbes (Thorpe et al. 2016). Remotely sensed observations capture structural and spectral features of landscapes at 1 m resolution over the field sites (Musinsky et al. 2022).

Figure 2. Schematic of the organisms collected by National Ecological Observatory Network at terrestrial sites (in green, top row) and at aquatic sites (in blue, middle row), and the types of data and samples provided for each of the taxa represented by the smaller icons, as described by the legend at the bottom. Image credit: National Ecological Observatory Network/Battelle. By Williams C.

Samples collected from all sites include environmental samples (e.g., soil, sediment), nontraditional organismal samples (e.g., ground roots, mammal feces, bulk bycatch samples), and traditional natural history voucher specimens (e.g., pinned mosquitoes, pressed herbarium samples). Most samples are curated at the NEON Biorepository, managed by Arizona State University’s (ASU) Biodiversity Knowledge Integration Center in Tempe, Arizona, USA; nearly two-thirds of samples there are cryo-preserved. Unique compared to conventional natural history collections, a portion of most NEON Biorepository collections may be consumptively or destructively analyzed for research (Tazik et al. 2022). All data associated with the organismal samples curated at the NEON Biorepository are available from the Symbiota (Gries et al. 2014; https://symbiota.org/) portal at biorepo.neonscience.org, the Environmental Data Initiative (EDI) at https://edirepository.org/ (Gries et al. 2023) and from the Global Biodiversity Information Facility at gbif.org (GBIF 2023b). NEON also shares its datasets for birds (Thibault et al. 2023) and ticks (Paull et al. 2022) via GBIF for enhanced discoverability and interoperability.

The data and samples provided by NEON and other global networks are key to yielding the depth of understanding required to achieve the ambitious goals of the global community to ‘to halt and reverse biodiversity loss’ as most recently codified in the Kunming-Montreal Global Biodiversity Framework (GBF) adopted by the 15th Conference of Parties of the United Nations Convention on Biological Diversity in December of 2022 (Convention on Biological Diversity 2022). The purpose of this study is to review the composition and use of NEON resources to date as a whole and specific to biodiversity, as an example of the impact of national-scale research infrastructure on global efforts such as the GBF.

We compiled data to summarize NEON’s contributions to biodiversity science to date and four categories of use of NEON: web portal visitation, data downloads, NEON Biorepository sample loans, and scientific publications. Here we describe the methods used to generate the usage metrics.

Web portal visitation

Google Analytics (GA) was used to track web portal usage across NEON’s three public-facing websites: (1) https://www.neonscience.org supplies all general information about NEON, including news, events, partnerships, and data collection and management documentation; (2) https://data.neonscience.org is the data portal for accessing all of NEON’s data products as well as data visualization tools; and (3) https://biorepo.neonscience.org is a Symbiota portal dedicated to the NEON Biorepository collections hosted by ASU. As metrics of web portal visitation, we tallied both the mean number of unique IP addresses of web portal visitors per month and the mean page views per month for each of the portals in 2022. As GA does not provide actual IP addresses, it is not possible to verify if users are unique across the three portals. To examine the global reach of this US infrastructure, we also assessed the distribution of unique IP addresses visiting the portals from October 1, 2022, through September 30, 2023, across the countries recognized by GA. Google Analytics - Universal Analytics was used for October 1, 2022 through a) November 30, 2022 for www.neonscience.org, b) April 30, 2023 for biorepo.neonscience.org, and c) May 31, 2023 for data.neonscience.org. Google Analytics 4 - GA4 was used for all other dates through September 30, 2023. Although users can intentionally manipulate their IP addresses (e.g., via use of a virtual private network, or VPN) to yield inaccuracies in the associated location data, this metric is the best available. Note that GA uses the International Organization for Standardization (ISO) 3166-1 standard for country names (ISO 2020).

Data downloads

Users can download NEON data from NEON’s data portal (data.neonscience.org), through NEON’s application programming interface, directly from storage buckets, or through various externally hosted repositories. As the external repositories use different approaches for packaging data and tools for tracking use, here we focus only on the direct NEON download options (i.e., all options except the external repositories). We use Google Cloud Storage bucket logs to determine the mean number of download requests per month, mean unique IP addresses associated with data downloads per month, and the mean volume of data downloaded per month for Oct 1, 2022, through September 31, 2023. Data file transfers associated with known IP addresses from internal servers or Google services were filtered out.

Sample loan requests

Samples and specimens at the NEON Biorepository have been available for research and educational use since 2019. NEON Biorepository staff record all inquiries and requests for sample loans as they are received. To measure use of this unique resource, we tallied the total number of researchers who have engaged in sample use requests (i.e., have borrowed or used samples, are actively using samples, or are awaiting funding to use samples), as well as the total number of samples involved in these requests, since 2019.

Scientific publications

We compiled data on scientific publications from two sources. First, the Dimensions dataset of publications that use NEON resources (obtained on 7 September 2023; Digital Science & Research Solutions, Inc. 2018) is based on a query for either the phrase, “National Ecological Observatory Network,” or one of the digital object identifiers associated with NEON’s released data products in the abstract, acknowledgements, or references within their database. Second, the GBIF dataset of publications obtained on 28 September 2023 includes publications that reference datasets containing NEON occurrence records obtained from GBIF (and are not captured in the Dimensions data) (GBIF 2023a). Occurrence records document the presence of a taxon in a location at a specified time (Wieczorek et al. 2012). We analyzed the distribution of countries represented by the author organizations within these two datasets. We also manually tagged the publications within the Dimensions dataset with the NEON data product identifiers (IDs) included in the publication and other applicable keywords (https://www.zotero.org/groups/2665225/neon_publications). To determine the number of publications related to biodiversity science, we then tallied the number of publications in the Dimensions data that had been tagged with one of the biodiversity-relevant product IDs (see Table S1) or one of the following keywords (“phenocam”, “biomass”, “vegetation”, “biodiversity”, “tree”, “forest”, “NEON Samples”).

NEON’s contributions to biodiversity science

First and foremost, NEON provides open and freely available data products, producing an average of 1.4 million records and over 32 TB per year (based on 2020–2022 data). The data catalog includes 85 products that represent taxonomic or other organismal information (Table S1). These data products span NEON’s three data collection systems, observational, sensor-based, and remotely-sensed. As of 15 October 2023, published NEON data includes 13,708 unique taxonomic names, excluding the archaea, bacteria, and fungi contained within NEON’s microbial data products that lack Linnaean binomials. Associated with these data is a suite of open, standardized field sampling protocols, sensor maintenance procedures, analytical laboratory procedures, and documents that describe the data processing algorithms NEON uses (see https://data.neonscience.org/documents for the complete library). These documents are freely available to the global public, and NEON encourages adoption of these methods by other researchers, institutions, and networks to expand the comparability and interoperability of data around the world.

Each year, NEON collects and archives approximately 100,000 biological, genomic, and environmental samples and specimens from its 81 terrestrial and aquatic sites. These samples and specimens complement the field observations and automated measurements collected at these same field sites. As of September 2023, NEON has collected and curated more than 503,000 samples and specimens; a total of more than 3 million are expected to be collected during the Observatory’s anticipated 30-year lifespan. These samples include individually curated voucher specimens representing more than 3,600 taxa; however, the samples also include pooled individuals collected as bulk or bycatch. Thus, NEON samples actually contain many more taxa than have been identified. We estimate that the terrestrial invertebrate bycatch likely will contain 10,000 to 15,000 arthropod species and 20+ million individual specimens over the 30-year lifetime of NEON.

Throughout its construction and ongoing operations, the broader impacts of the Observatory include development and adoption of information and data management standards, development and delivery of data skills training resources (https://www.neonscience.org/resources/learning-hub), and supporting interoperability with data provided by partner networks. For biodiversity data, these partners include GBIF, EDI, the Barcode of Life Data System (BOLD) (Ratnasingham and Hebert 2007), the PhenoCam Network (Brown et al. 2016), and the USA-National Phenology Network (Elmendorf et al. 2016). Interoperability efforts include sharing NEON data on other platforms; for example, GBIF contains nearly 1M records provided by NEON as of October 2023, with growth in these contributions planned throughout NEON’s lifespan, enabling NEON data to be discovered and used alongside the billions of other GBIF records. Interoperability efforts also include building of tools to enable harmonization of NEON data with other datasets, such as the ecocomDP data model and R package that harmonizes community ecology data from NEON, LTER, and EDI (O’Brien et al. 2021).

Regarding standards development and adoption, NEON has developed a broad and well-described internal vocabulary for ingested and published data streams, and it also employs community supported vocabularies (e.g., Darwin Core [Wieczorek et al. 2012]), Minimum Information about any (x) Sequence (MixS) (Yilmaz et al. 2011). Metadata are structured using community schemas (e.g., Ecological Metadata Language [Jones et al. 2019], schema.org). Through the genomic standards consortium, NEON provided input to the development of the MIxS environmental package. NEON has also partnered with the Environment Ontology (Buttigieg et al. 2013) to describe land use terms and is leveraging this standard to share NEON metagenomics data with the US National Microbiome Data Collaborative (Eloe-Fadrosh et al. 2022).

Use of NEON resources to advance science

Web portal visitation, data downloads, and sample use requests are considered leading indicators, or impact measures that are predictive of eventual scientific and educational outcomes that are key to the success of NEON. Tables 1, 2 report recent values for the primary leading indicators that NEON tracks. Ideally, these values could be interpreted relative to similar metrics from equivalent organizations or activities. However, the diversity of NEON data products is unparalleled within the environmental research network community, and organizations do not typically share these usage data publicly (with GBIF as a notable exception). Even in the absence of benchmarks, it is clear that there is substantial interest from the global community in NEON, with more than 47,000 unique IP addresses visiting its three portals each month (Table 1), more than 1.7 TB of data requested on average each month, and over 200 researchers engaged in sample use requests from the NEON Biorepository (Table 2). Moreover, approximately 61% of the IP addresses visiting NEON portals are affiliated with countries outside of the United States (Fig. 3).

Table 1 . Usage metrics of the three web portals for accessing NEON resources Oct 2022–Sep 2023.

NEON portal	Mean number of unique web portal visitors per month	Mean sessions per month
www.neonscience.org	47,152	67,235
data.neonscience.org	4,847	7,446
biorepo.neonscience.org	1,287	1,654

NEON: National Ecological Observatory Network.

Table 2 . Usage metrics of NEON data products and samples archived at the NEON Biorepository.

Category of use and time period	Usage metric	Value
NEON data downloads per month (Oct 2022–Sep 2023)	File downloads	853,000
	Unique IP addresses/hosts	1,331
	Data download volume (TB)	1.77
Biorepository loan requests (Jan 2019–July 2023)	Number of researchers	205
	Number of samples	13,673*

NEON: National Ecological Observatory Network.

*Count does not include several thousand NEON samples loaned prior to the initiation of the Biorepository.

Figure 3. Map of locations derived from IP addresses requesting data downloads from data.neonscience.org, Oct 2022–Sep 2023, using Google Analytics.

The Dimensions (Digital Science & Research Solutions, Inc. 2018) dataset includes 590 publications that use NEON resources, with increasing rates of publication observed since NEON entered its full operations phase in May 2019 (Fig. 4). The Dimensions dataset also provides citation counts for each publication, which also shows increasing rates in recent years, with 79% of publications in the dataset cited at least once and a cumulative total exceeding 17,000 citations. Within this dataset, we identified 265 publications (45%) that used at least one of NEON’s biodiversity data products (Table S1), samples, or other NEON resources to address a biodiversity-relevant question. These publications are complemented by the GBIF dataset of an additional 288 publications that reference datasets that include NEON occurrence records (GBIF 2023a) and are also biodiversity research products. These counts are likely underestimates of all product use, as data users often fail to cite data properly (Vannan et al. 2020).

Figure 4. Count of publications produced each year from January 2013 through September 2023 that use National Ecological Observatory Network resources, as tracked by each of two sources. GBIF: Global Biodiversity Information Facility.

After only a few years of full operations for NEON, the resulting literature includes several impactful and different uses of NEON resources. From a biodiversity perspective, the findings generally fall under four themes. First, NEON data and samples have been used as a baseline against which to track changes. These include leveraging both the unplanned disturbances, such as hurricanes (e.g., Kenney et al. 2021) and wildfire (e.g., Choi et al. 2023), and planned disturbances, such as cattle grazing (e.g., Gaffney et al. 2021), that NEON sites have experienced to date. Second, NEON data and samples have yielded discoveries of new species (e.g., Ciugulea et al. 2019; Will and Liebherr 2022) and range expansions for known species (Gorris et al. 2021). NEON data have played a key role in advancing the emerging field of ecological forecasting, including efforts to forecast changes in species abundance and community composition through time (Averill et al. 2021; Quinn Thomas et al. 2023). NEON samples have aided both the development of machine learning methods to characterize biodiversity (Blair et al. 2020) and the elucidation of mosquito evolutionary history (Soghigian et al. 2023). Finally, NEON data and samples have been used to reveal large-scale patterns of species richness, sometimes with unexpected results (e.g., Carrasco et al. 2019; Weiser et al. 2022).

NEON is fully funded by the US NSF and collects data and samples from US field sites, but its mission is global, to engage the international community to address the grand environmental challenges of our time. To that end, access to NEON data and samples is open to the global public, and NEON has active partnerships with many global networks and valuable insights are gained from international researchers who contribute to NEON’s advisory groups. NEON has forged formal partnerships with, for example, the International LTER Networks (iLTER), Integrated Carbon Observation System, Australia's Terrestrial Ecosystem Research Network, and Research Data Alliance, among others. NEON is a key member of international projects, including FLUXNET, the Global Lakes Ecological Observatory Network, GBIF, BOLD, the Forest Global Earth Observatory, Biodiversity Information Standards, the European Union’s (EU) Copernicus Programme, the Omic Biodiversity Observation Network of the Group on Earth Observations (Meyer et al. 2023), and the Global Carbon Project. Partnership activities include data sharing, standards and methods co-development, research collaboration, conference support, and advisory services.

Further evidence of NEON’s global relevance can be found in the portal visitation data and the author affiliations across the Dimensions and GBIF publication datasets (Digital Science & Research Solutions, Inc. 2018; GBIF 2023a). 232 countries, territories, or areas of geographical interest are represented in the user base for www.neonscience.org, for the period October 1, 2022, to September 30, 2023, with 55 countries associated with more than 1,000 users. Across the publication datasets, authors are affiliated with organizations representing 84 countries. The most represented countries among the organizations include the US (59%), the UK and Germany (4% each), and Canada and China (3% each).

In conclusion, the future health, welfare, and security of the planet depends upon diverse ecological systems that are undergoing poorly understood transitions due to climate change. NEON’s development and application of standards across diverse ecological regions and types of environmental data are unique in the field of ecology. These standardized biological and environmental data and samples enable unprecedented interoperability, which in turn enables a comprehensive and predictive understanding of ecosystem functions in relation to change. Because many of these standards are developed by and/or shared with national and international partners, NEON data are actively integrated and aggregated by a variety of resources to serve a global user base. NEON has democratized access to these environmental data through the application of FAIR data principles (Wilkinson et al. 2016), with cyberinfrastructure that provides free and open access to data and analytic tools, thus empowering a community of practice to develop open-source code, analytic tools, standardized protocols, and derived ecological data products. Because data and/or metadata are shared regularly between NEON, its global partners, and third-party aggregators, world-wide use of the data is enabled and science further accelerated.

NEON is presented here as an exemplar of the global research infrastructure needed to achieve the ambitious targets set by the global community through the GBF, a key part of the “new biodiversity science landscape centered on big data integration” (Heberling et al. 2021). However, NEON must be complemented by many more global networks that collect and provide free, open, and standardized data to achieve our shared goals. NEON has served as a founding member of the Global Ecosystem Research Infrastructure (GERI) network to facilitate the necessary collaboration and interoperability (Loescher et al. 2022). Still in its early phase of development, this group is actively working to strengthen existing connections, pilot the potential to address a global use case centered on ecological drought, and expand the reach of the network to additional countries. Such a strong global network is key to unlocking the as yet unrealized potential of NEON and the broader GERI network, with new frontiers in artificial intelligence, microbial gene discovery, synthetic biology, and sensor technologies ahead.

Supplementary information accompanies this paper at https://doi.org/10.5141/jee.23.076.

Table S1 NEON biodiversity data products that include taxonomic or other relevant organismal information (n = 85).

The National Ecological Observatory Network is a program sponsored by the US NSF and operated under cooperative agreement by Battelle. This material is based in part upon work supported by the US NSF through the NEON Program.

BOLD: Barcode of Life Data System

EDI: Environmental Data Initiative

EU: European Union

FAIR: Findable, Accessible, Interoperable, and Reusable

GBIF: Global Biodiversity Information Facility

GBF: Global Biodiversity Framework

GERI: Global Ecosystem Research Infrastructure

ISO: International Organization for Standardization

LTER: Long-Term Ecological Research Network

MixS: Minimum information about any (x) sequence

NEON: US National Ecological Observatory Network

NSF: US National Science Foundation

CML collated and analyzed data on web portal visitation and data downloads. KMT collated and analyzed data on publications and author affiliations. KMY collected and analyzed data on sample loan requests. KMT, PMM, NMF, KMY, and CML all contributed to writing the manuscript. All authors read and approved the final manuscript.

Funding for this work was provided by the US NSF Award #1724433. The NSF did not participate in any aspect of the study; the authors are solely responsible for the content.

The datasets used and/or analyzed during the current study are available online (Digital Science & Research Solutions, Inc. 2018; GBIF 2023a) or available from the corresponding author on reasonable request.

Not applicable.

Not applicable.

The authors declare that they have no competing interests.