Skip to main content
v2026.11,986 entries · CC-BY 4.0

Paleoecology: Definition, Proxies and Data Sources

Paleoecology reconstructs past environments from proxies. See proxy types, where Neotoma and NOAA NCEI hold data, and a sampling-error checklist.

Written and maintained by CASRAI Editorial Board

Published · Last updated

On this page

Short answer

Paleoecology is the study of past ecosystems and environments, reconstructed from the preserved remains and chemical signatures that organisms and environments left in sediments, ice, tree rings and similar archives. Researchers read these records, called proxies, to infer past vegetation, climate and fire, and they depend on shared databases such as Neotoma and NOAA’s paleoclimatology archive to store and reuse them.

Definition and scope

Paleoecology sits between paleontology and ecology. Paleontology is the broader study of life in the geological past, while paleoecology asks a narrower question: what were the living conditions, communities and environments at a particular place and time, and how did they change? The answer is never observed directly. It is inferred from proxies, which are measurable traces that stand in for conditions nobody recorded.

The two largest public data services show how the field defines its material. The Neotoma Paleoecology Database describes itself as an open, community-curated data resource for global-change research, covering primarily the Pliocene to Quaternary part of the geologic record, and it states that it can accommodate virtually any type of fossil data. NOAA’s National Centers for Environmental Information (NCEI) describes its paleoclimatology holdings as data derived from natural sources such as tree rings, ice cores, corals and stalagmites, together with ocean and lake sediments, and as including reconstructed variables such as temperature and precipitation. Taken together, they place the field’s practical centre of gravity on recent geological time, on cored or excavated archives, and on quantities that can be compared across sites.

Two points about scope matter for research offices and lab managers. First, paleoecology overlaps with paleoclimatology, and the same archive may serve both: a lake-sediment pollen record can be read as a vegetation history or, with additional inference, as a climate signal. Second, the discipline is as much a data practice as a field practice, because a single core is only useful once its measurements, taxonomy and age control are documented well enough for someone else to reuse them.

How paleoecology is done in practice

From archive to proxy record

A typical workflow begins with an archive: a lake or bog sediment core, a plant-remains deposit, a packrat midden, a tree-ring series or an ice core. The core or deposit is sub-sampled at chosen intervals, the proxy of interest is extracted and counted or measured in the laboratory, and an age model assigns dates to depths. The results are then interpreted as vegetation, climate or disturbance history.

NCEI’s pollen guidance illustrates the inference step. Pollen preserved in sediments can provide a record of the vegetation that grew on or adjacent to the shore, and past climate can be inferred by determining what conditions were favourable to the pollen types found in the sediment. That is an inference chain, not a measurement, and each link in it (sampling, counting, dating, calibration to modern conditions) adds uncertainty.

Where the data are kept

Two public services are especially relevant, and they overlap. Neotoma is organised as a database plus a software ecosystem and a community: it standardises variable names and taxonomy, and provides data models for sites, proxies, cores and age controls. Its tools include an Explorer, a web-service API, the Tilia software, an R package and a Hierarchical Taxonomy Viewer. Governance is community-based, with a Leadership Council and nearly 100 database stewards worldwide; stewards are described as analogous to editors of peer-reviewed journals and can upload and correct data.

NCEI Paleoclimatology organises holdings by product category and offers a Paleo Data Search, an interactive map and an API. Its pollen listings draw on the World Data Service for Paleoclimatology and on Neotoma, so a pollen dataset can be reachable through both routes. For fossil-occurrence records (what lived where, rather than a downcore proxy series) the relevant resource is different: see the guide to the Paleobiology Database.

For project planning, the practical consequence is that data deposition belongs in the data management plan from the start. Choosing a community repository early means the variable names, taxonomy and age-control fields are known before the first sample is counted, and it makes the eventual deposit a matter of formatting rather than reconstruction. The data repository term and the metadata term explain the general vocabulary.

Types of proxy and where each is catalogued

The data table below cross-walks proxy types against the category names that Neotoma and NCEI themselves use, so a researcher can see where a given kind of record is most likely to be found. It is compiled only from the data types and product categories each service lists on its own pages; a proxy marked “Not named” is simply not named in that service’s list, which is not a statement that no such data exist there.

Three patterns are worth noting. Biological remains that fossilise in sediments (pollen, plant macrofossils, diatoms, ostracodes, testate amoebae, vertebrates) are listed in Neotoma’s data types. Archives defined by their physical medium (ice cores, tree rings, corals, speleothems, boreholes) appear as NCEI product categories. Plant macrofossils and pollen appear in both. Researchers therefore often have to search more than one service, and a dataset’s presence under one category name does not guarantee the same label elsewhere.

Where it matters for labs, projects and data stewardship

An uncertainty checklist for sampling and counting

A 2026 paper in Climate of the Past by Asena, Perry and Wilmshurst, first released as an EGUsphere preprint, uses a virtual-ecology approach: the authors generated synthetic, error-free core-type pseudoproxy samples, then added uncertainty from three sources and tested the effects on two statistical methods, Fisher Information and principal curves. Its results translate into a practical checklist for anyone designing a coring or counting protocol.

Source of error tested What the authors report Design question for a protocol
Sub-sampling interval Increasing the interval had the most substantial influence of the sources tested, on both statistical methods. Is the spacing between sampled depths fine enough for the change you expect to detect?
Proxy counting from sub-samples Counting fewer proxies mattered mainly in combination with wider intervals; that combination had the strongest influence on both methods. Is the count target per sample documented, and was it applied consistently?
Core mixing Tested as one of three uncertainty sources, individually and in combination. Is disturbance during coring or in the archive itself recorded in the field notes?

The same paper reports that principal curves were less influenced by simulated proxy counting and sub-sampling than Fisher Information, so the choice of analysis method interacts with the design of the protocol. The authors argue that quantifying information loss helps researchers design studies and make more robust inferences, and the paper’s review process recorded a request to archive the method protocol on Zenodo alongside code and data. That points to a general stewardship habit: record the sub-sampling interval, the counting target and the age-control method as structured fields, not as prose in a methods section, because these are exactly the parameters that determine how much signal survives.

The paper simulates data rather than analysing a particular real core, so its findings describe how error propagates in modelled cores. They are best used as a prompt for protocol design, not as a quantitative correction for any one site.

Why curation and standardisation matter

Neotoma’s stewardship model, with named stewards who can upload and correct data, and its standardised variable names and taxonomy, exist because proxy records are only comparable when the labels match. For a research office, that makes the repository choice a compliance and reuse decision, not only a scientific one: deposit in a community-curated service keeps the dataset findable and consistent with other records in the same region. NCEI notes that data contributors can reach the paleo team through the contact address on its pollen page.

For related ecological context, the ecosystem ecology guide covers the study of present-day ecosystem processes that paleoecological records are often used to extend backwards in time.

Common misunderstandings

Paleoecology is not the same as paleontology

Paleontology covers life in the geological past in general. Paleoecology uses fossils and other proxies to reconstruct environments and ecological conditions. The two share material (fossils) but ask different questions, and they are catalogued in different resources: occurrence records in the Paleobiology Database, proxy series in services such as Neotoma and NCEI.

A proxy is not a direct measurement

Pollen counts are not temperatures, and tree rings are not rainfall gauges. NCEI’s own phrasing, that climate is inferred by determining which conditions were favourable to the pollen types present, signals that reconstructions rest on a calibration step. Reported reconstructed variables such as temperature and precipitation are products of that inference.

More data points do not automatically fix a coarse design

The simulation study shows that wide sub-sampling intervals combined with low counts had the strongest combined effect. Adding statistical sophistication downstream does not restore information that the sampling design lost.

Deep time is not the main target of these databases

Neotoma states that it covers primarily the Pliocene to Quaternary. Researchers working on older intervals should expect to use other resources, such as occurrence-based databases, and should not assume a proxy database spans all of geological history.

Frequently asked questions

What is the difference between paleoecology and paleontology?

Paleontology is the study of ancient life generally. Paleoecology is the part concerned with reconstructing past environments and ecological conditions from proxies such as fossil pollen, plant macrofossils and sediment chemistry. See the paleontology guide for the wider field.

Which proxies do paleoecologists use?

Neotoma lists fossil pollen, vertebrates, diatoms, ostracodes, testate amoebae, stable isotopes, plant macrofossils and packrat middens, among others. NOAA NCEI adds categories such as tree rings, ice cores, corals, speleothems, insects, fire history and loess and paleosols. The data table on this page maps them.

Where can I find paleoecological data?

Neotoma (neotomadb.org) and NOAA NCEI Paleoclimatology (ncei.noaa.gov/products/paleoclimatology) are the two public services described on this page. NCEI offers a search tool, interactive map and API, and Neotoma offers an Explorer, API and R package. Fossil occurrence records are held by the Paleobiology Database.

What can go wrong with a paleoecological record?

Environmental processes and observer or laboratory choices can mask the true signal. In a simulation study, widening the sub-sampling interval had the most substantial influence, and wide intervals combined with fewer proxies counted had the strongest combined effect.

How far back do these databases go?

Neotoma covers primarily the Pliocene to Quaternary. NCEI’s paleoclimatology categories span a variety of natural archives; consult each product category for its time range.

Sources

Data: Paleoecological proxy types cross-walked to Neotoma data types and NOAA NCEI Paleoclimatology product categories

Compiled by comparing the data types listed on the Neotoma Paleoecology Database About page with the product categories listed on NOAA NCEI's Paleoclimatology page. 'Not named' means the service's own list does not name that proxy; it does not mean no such data exist there.

Paleoecological proxy types cross-walked to Neotoma data types and NOAA NCEI Paleoclimatology product categories
Proxy typeNeotoma data type listedNCEI Paleoclimatology product category
Fossil pollenFossil pollenPollen
Plant macrofossilsPlant macrofossilsPlant Macrofossil
DiatomsDiatomsNot named
OstracodesOstracodesNot named
Testate amoebaeTestate amoebaeNot named
Stable isotopesStable isotopesNot named
Packrat middensPackrat middensNot named
Vertebrate remainsVertebratesFauna
Insect remainsNot namedInsect
Tree ringsNot namedTree Ring
Ice coresNot namedIce Core
Corals and sclerospongesNot namedCoral and Sclerosponge
Speleothems (stalagmites)Not namedSpeleothem
Fire historyNot namedFire History
Loess and paleosolsNot namedLoess and Paleosols

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about Paleoecology: Definition, Proxies and Data Sources

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →