Uncovering Earth’s Past Climate: The Science of Palaeoclimatic Data

We have been measuring temperature and rainfall with instruments for barely 150 years – a blink in Earth’s 4.5-billion-year history. To understand what the climate was like thousands or millions of years ago, scientists rely on a diverse set of clues called palaeoclimatic data. These are records preserved in nature and in human history that act as indirect indicators, or proxies, of past climate conditions. According to the U.S. Geological Survey, palaeoclimate reconstructions establish the baseline of natural climate variability before instrumental records began – and they are essential for predicting how climate may change in the future.

The role of historical and archaeological records 🔗

Before scientific instruments, humans were still observing and recording the world around them. These observations, scattered across manuscripts, diaries, administrative documents, and archaeological sites, have become an important category of palaeoclimatic evidence.

Written and documentary sources 🔗

Historical documents provide qualitative information about past climates through human observation rather than precise measurement. Ship logs, farmers’ diaries, and travellers’ journals contain weather descriptions, harvest dates, and seasonal observations that help scientists piece together climate conditions from centuries past. Chinese imperial records, some dating back over 3,000 years, document floods, droughts, and unusual weather patterns with remarkable consistency. Medieval European monasteries kept meticulous records of harvest dates, wine quality, and severe weather events – data that have helped establish the existence of the Medieval Warm Period (roughly 900-1300 CE) and the subsequent Little Ice Age (approximately 1300-1850 CE).

A well-known example is the Thames River in London. Historical accounts describe “Frost Fairs” held on the frozen Thames during cold winters in the 17th and 18th centuries – concrete evidence of how much colder London’s winters were during the Little Ice Age compared to today.

Archaeological evidence 🔗

Archaeological sites preserve environmental information in less obvious ways. Rock layers at excavation sites reveal the geological context of past human settlements, while plant and animal remains indicate what species thrived under different climate conditions. Scientists analysing pottery from ancient cooking vessels have discovered hydrogen isotopes in animal fats that reflect precipitation patterns from over 8,000 years ago. Charred plant materials found at sites can indicate periods of drought or vegetation change.

Settlement patterns are also telling. Norse settlements in Greenland, established during the Medieval Warm Period, were eventually abandoned as cooling temperatures made agriculture unsustainable. Similarly, the cliff dwellings of the American Southwest were abandoned during severe droughts in the 13th century – a conclusion supported by both archaeological evidence and tree ring data. Even the types of crops grown, storage facilities built, and farming techniques used all reflect the climate of their era. For instance, the discovery of grape presses in northern England from Roman times suggests warmer conditions than the region experiences today.

While historical and archaeological records are invaluable, they cover only a tiny fraction of Earth’s history – at most a few thousand years. To go further back, scientists turn to the geological record.

Geological evidence: rocks, fossils, and ice cores 🔗

Palaeoclimate archives consist of geological and biological materials that preserve evidence of past changes in climate. These span everything from sedimentary rocks billions of years old to ice frozen just tens of thousands of years ago. Each archive contains different proxies – substances or features that can be sampled and analysed to reconstruct past conditions.

Sedimentary rocks and fossil records 🔗

Sediment is deposited in layers in lakes, wetlands, estuaries, and oceans over time. Because the oldest sedimentary rocks are about 3.9 billion years old, sediment records provide a means to study past climates throughout most of Earth’s history. Within these sediments, scientists find a variety of climate proxies: pollen grains, charcoal, algae, plant remains, and the shells of microscopic marine organisms called foraminifera and diatoms.

Foraminifera are particularly useful. These shelled organisms record evidence of past environmental conditions in their calcium carbonate shells. Stable oxygen isotope ratios contained in the shells can be used to infer past water temperatures, because warmer water causes shells to incorporate more of the lighter oxygen isotope. Measurements from hundreds of deep-sea cores around the world have been used to map past surface and bottom water temperatures across geological time.

Fossils more broadly – of plants, insects, and animals – tell scientists which species existed under which conditions, extending climate knowledge deep into Earth’s past. On a larger scale, rock features such as fossilised sand dunes, glacial striations (scratches left by moving ice), and evidence of ancient sea-level changes all signal major shifts in past climate.

Tree rings 🔗

Each year, trees add a new layer of growth known as a tree ring. These rings record changes in temperature and precipitation, as well as more localised changes such as fire and insect attacks. In years with favourable growing conditions – warm temperatures and adequate moisture – rings are wide. In harsh years, they are narrow. By analysing patterns of ring thickness, scientists have reconstructed annual variability in moisture and temperature going back thousands of years. Individual events such as forest fires also leave distinctive dark arcs in the sequence of rings.

Ice cores: frozen climate archives 🔗

Among the most powerful sources of palaeoclimatic data are ice cores – cylinders of ice drilled from glaciers and polar ice sheets. Ice is formed by snow that accumulates over hundreds of thousands of years; when layers of snow compact, tiny air bubbles are trapped inside – snapshots of the atmosphere at the moment they were sealed in.

These air bubbles provide direct samples of past atmospheres, preserving ancient concentrations of greenhouse gases like carbon dioxide (CO₂) and methane (CH₄). Ice cores help scientists compare atmospheric conditions before and after the Industrial Revolution, when humans began burning large quantities of fossil fuels, and Antarctic ice cores preserve up to 800,000 years of climate history. Ice core data confirm that current CO₂ concentrations are higher than at any point in that record.

Modern techniques in palaeoclimatic reconstruction 🔗

Collecting the raw material – a sediment core, an ice cylinder, a fossil shell – is only the first step. Extracting meaningful climate information from these materials requires sophisticated analytical techniques, many of which have advanced dramatically in recent decades.

Isotopic analysis 🔗

The most widely used modern technique in palaeoclimatology is stable isotope analysis, particularly the measurement of oxygen isotope ratios. Oxygen isotopes are the most commonly used palaeoproxy to reconstruct long records of past climates. The principle relies on a physical property of water: the lighter isotope of oxygen (¹⁶O) evaporates more readily than the heavier isotope (¹⁸O). As a result, the ratio of ¹⁸O to ¹⁶O in ice cores, sediment shells, and other materials shifts predictably with temperature.

Once an air mass reaches high latitudes, it contains fewer heavy oxygen isotopes – so ice cores extracted from polar regions reveal past temperatures through the ratio of heavy to light oxygen isotopes. The fewer heavy isotopes in a polar ice core, the lower the ancient temperature. Using mass spectrometers – instruments capable of measuring isotope ratios with extreme precision – researchers measure the ratio of oxygen isotopes in ice core samples and compare the result with an isotopic ratio standard known as SMOW (Standard Mean Ocean Water). A similar approach is applied to fossil shells from marine sediment cores, where oxygen isotope shifts reveal past ocean temperatures and ice volumes.

The same logic applies to hydrogen isotopes. Deuterium (D, or ²H) is measured alongside oxygen-18 in ice cores, and the relationship between these stable isotopes and surface temperature is consistent and linear over Antarctica, making it a reliable palaeothermometer across deep time.

Radiometric dating 🔗

Knowing that a proxy record shows a warm or cold period is only useful if scientists can determine when it occurred. Radiocarbon dating (carbon-14) is widely used for organic materials up to about 50,000 years old, while uranium-thorium (U-Th) dating extends accurate chronologies to cave formations (speleothems) hundreds of thousands of years old. Together, these dating methods allow scientists to build precise timelines and correlate records from different parts of the world.

High-resolution scanning and other emerging tools 🔗

New technologies continue to expand the palaeoclimatic toolkit. Advanced mass spectrometry allows analysis of increasingly small samples with greater precision. High-resolution scanning techniques can analyse sediment and ice cores at microscopic scales, revealing seasonal and even monthly climate variations. Ancient DNA extracted from lake sediments now provides information about past ecosystems and how they responded to climate changes. Satellite imagery helps identify optimal sites for drilling new cores.

Crucially, no single proxy is used in isolation. Palaeoclimatology data derived from tree rings, ice cores, cave deposits, pollen, coral growth bands, and ocean sediments together extend the climate archive by hundreds to millions of years. By combining multiple proxies and cross-checking findings, scientists build increasingly robust reconstructions of past climate – ones that can then be fed into computer models to test hypotheses about how the climate system works and how it may evolve in the future.

What do you think? Given that each source of palaeoclimatic data has its own strengths and limitations, how confident should we be in reconstructions that combine multiple proxies – and does that confidence change your view of how reliable our predictions of future climate change are? Also, if historical records like diaries and harvest calendars can yield genuine climate data, what other everyday human records from the past might scientists have overlooked?

References

1. https://www.usgs.gov/programs/ecosystems-land-change-science-program/science/paleoclimate-research
2. https://climatechange.academy/introduction-to-climate-change/unlocking-past-palaeoclimatic-data-sources/
3. https://www.usgs.gov/programs/ecosystems-land-change-science-program/science/paleoclimate-archives
4. https://serc.carleton.edu/microbelife/topics/proxies/paleoclimate.html
5. https://www.whoi.edu/know-your-ocean/ocean-topics/climate-weather/paleoclimatology/
6. https://nsidc.org/learn/ask-scientist/core-climate-history
7. https://itssedimentary.com/2016/10/26/direct-measurements-for-global-temperature-only/
8. https://www.scientificamerican.com/article/how-are-past-temperatures/
9. https://www.antarcticglaciers.org/glaciers-and-climate/ice-cores/ice-core-basics/
10. https://www.climate.gov/maps-data/dataset/natural-records-past-climate-maps-visualizations-and-descriptive-information