Ice Core Analysis Simulator: 800,000 Years of Climate Data

simulator intermediate ~10 min
Loading simulation...
Age ≈ 18,500 yr — Last Glacial Maximum

At 1000 m depth with 8 cm/yr accumulation, the ice dates to approximately 18,500 years ago — the Last Glacial Maximum when global temperatures were 5°C cooler and ice sheets covered much of North America and Europe.

Formula

age(z) ≈ (H/A) × ln(H/(H-z)) (Nye model)
δ¹⁸O ≈ 0.69 × T_surface - 13.6 ‰ (Dansgaard relation)
ρ(z) = ρ_ice - (ρ_ice - ρ_snow) × exp(-z/z_c)

Frozen Archives

Every snowflake that falls on an ice sheet traps a tiny sample of the atmosphere. As snow compresses into ice over centuries, these air bubbles become sealed capsules of ancient atmosphere. By drilling kilometers into Antarctic and Greenland ice sheets, scientists extract continuous records of past climate stretching back 800,000 years — the longest direct archive of Earth's atmospheric composition. Each layer is a page in a book of climate history written by nature itself.

Isotopes as Thermometers

Water molecules containing the heavier oxygen-18 isotope condense more readily than those with oxygen-16. Colder temperatures mean more heavy isotopes rain out before reaching the ice sheet, so the ratio δ¹⁸O in ice decreases with temperature. This relationship, quantified by Willi Dansgaard in 1964, converts isotope measurements into a continuous temperature record. The technique reveals glacial-interglacial cycles with Antarctic temperature swings of 8–10°C over tens of thousands of years.

Trapped Gas Bubbles

At about 60–100 m depth, compressed snow (firn) seals into solid ice, trapping air bubbles that preserve the atmospheric composition at the time of closure. Mass spectrometry measures CO₂, CH⁴, N₂O, and other trace gases directly. The ice core record shows CO₂ oscillating between 180 ppm (glacials) and 280 ppm (interglacials) with remarkable regularity — until the industrial era shattered this pattern with a spike to 420 ppm, unprecedented in 800 millennia.

Reading the Layers

Beyond gases and isotopes, ice cores contain volcanic ash layers (tephra) that provide precise chronological markers, dust concentrations that reveal wind patterns and aridity, sea salt that tracks storm intensity and sea ice extent, and even pollen grains that record vegetation changes. Together, these proxies create a multi-dimensional portrait of past climate. This simulation lets you explore how depth, accumulation rate, and temperature interact to determine the age and climate signals preserved in the ice.

FAQ

What do ice cores tell us about past climate?

Ice cores are the most detailed archive of past climate. Trapped air bubbles preserve ancient atmosphere, allowing direct measurement of CO₂, methane, and other gases. Isotope ratios (δ¹⁸O, δD) record temperature, while dust, volcanic ash, and sea salt reveal atmospheric circulation, eruptions, and ocean conditions — all with annual to decadal resolution.

How far back do ice cores go?

The longest ice core (EPICA Dome C, Antarctica) reaches 800,000 years. Greenland cores reach about 130,000 years. The Beyond EPICA project aims to recover ice dating to 1.5 million years. At the deepest levels, annual layers are compressed to millimeters, limiting resolution.

How do scientists date ice cores?

Ice cores are dated by counting annual layers (visible in snow accumulation, chemistry, and isotopes), matching volcanic ash to known eruptions, using trapped gas concentrations correlated to other records, and flow models that account for layer thinning with depth. Multiple methods cross-check each other for reliability.

What is the CO₂-temperature relationship in ice cores?

Over the past 800,000 years, CO₂ and temperature track each other closely through glacial cycles. CO₂ ranges from 180 ppm (glacials) to 280 ppm (interglacials), with temperature swings of 8-10°C in Antarctica. Today's 420 ppm is far outside this natural range and rising 100× faster than any change in the ice core record.

Sources

Embed

<iframe src="https://homo-deus.com/lab/glaciology/ice-core-analysis/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub