The Civilization Record
The Civilization Record, methodology
The long-run planetary substrate the whole site rests on, across six dimensions from 10,000 BCE to today: population and the economic baseline, land and surface, energy, climate and atmosphere, water, matter and biosphere. The population, GDP and materials series are inherited from historical_series.json; the energy, land-cover, climate, water and biomass series are sourced here.
All series use the same 14 era years: -10000, -8000, -6000, -4000, -2000, 1, 1000, 1700, 1800, 1900, 1950, 1970, 2000, 2025. Each is marked measured, estimate or qualitative; the ancient end of almost every series is a wide-band reconstruction. Population, GDP and material extraction are reused from historical_series.json; land, energy, climate, water and biosphere are compiled here.
1Population
From four million foragers to a species past eight billion: population, urbanisation, lifespan and output per head all trace the same curve, flat for ten thousand years, then steep.
| Series | Source | Type / note |
|---|---|---|
| World population | historical_series.json · nufus_milyon (HYDE 3.3; McEvedy & Jones 1978 / Biraben 1980; UN World Population Prospects 2024) | estimate · For 10,000 BCE the band is 1-10 million, central value near 4 million. From 1950 the figures are UN World Population Prospects measurement; before 1800 they are reconstruction. |
| Annual population growth rate | türetilmiş: dünya nüfusu serisinin ardışık noktaları arası bileşik oran (historical_series.json · nufus_milyon); 2025 değeri UN World Population Prospects 2024 cari hız | estimate · Each point is the average compound growth rate over the interval from that year to the next; for 2025 the current annual rate is given. The intervals are not of equal length. |
| Urban population share | UN World Urbanization Prospects 2018 (1950 sonrası); Reba, Reitsma & Seto 2016 (Scientific Data 3:160034) ve HYDE 3.3 (sanayi öncesi) | estimate · The first cities date to around 4000-3500 BCE (Uruk). The pre-industrial urban share is below one tenth; the threshold chosen (5,000 versus 20,000 residents) moves the figure. |
| Life expectancy at birth | Riley 2005 (Population and Development Review 31:537-543); UN World Population Prospects 2024; Our World in Data | estimate · Life expectancy at birth stays near 30 years until 1800; high infant mortality pulls the mean down, adult lifespans were longer. From 1950 the figures are UN measurement. |
| World GDP per capita | historical_series.json · gsyih_kisi_basi_usd (Maddison Project Database 2023 reel büyüme endeksi; IMF WEO 2025 nominal çıpa) | estimate · The subsistence line: until 1800, annual output per head corresponds to a living standard of one to two dollars a day. Then the hockey stick. Values are 2025 USD equivalent, not purchasing-power parity. |
| Total world GDP | historical_series.json · gsyih_tusd_2025esdeger (Maddison Project Database 2023 reel büyüme endeksi; IMF WEO 2025 nominal çıpa) | estimate · Sectoral structure, wealth and market flow do not sit in this dimension; they live in Layers 1-3. Only the scale of total output is given here. |
| Workforce by layer | layer_workforce.json · ILO modelled estimates, ILOSTAT (2024 update), employment by ISIC Rev.4 section, world, 2023 | L0 = ISIC A,B; L2 = ISIC K,L; L1 = everything else. measured (L0, L1), estimate (L2, L3) |
| Sector share over time | Our World in Data, Labor Force by Sector (uzun dönem); Herrendorf, Rogerson & Valentinyi 2014 (Handbook of Economic Growth 2:855-941); ILO modelled estimates, ILOSTAT (2024 update) için 2023 | estimate · For the long run only three coarse groups exist: farming and extraction (hunting-gathering, agriculture, forestry, fishing, mining), industry (manufacturing, construction, utilities), services (everything else). This is a different lens from the layer split: here industry and services are separate, in the layer split both are L1. Shares are employment-based; in high-income economies service employment exceeds 70%. 10,000 BCE to 2000 BCE is a coarse reconstruction. |
The 10,000 BCE population is a wide-band reconstruction: 1-10 million for HYDE 3.3 and McEvedy & Jones, central value near 4 million.
Working age is defined as 15-64 (ILO).
Unpaid household and care work is outside the System of National Accounts production boundary; the person-equivalent is 16.4 billion hours a day divided by an 8-hour working day.
The layer mapping is the ISIC Rev.4 cut in the layer_workforce.json methodology: L0 = ISIC A+B, L2 = ISIC K+L, L1 = everything else; L3 (core market roles) is an estimate carved out of the finance component of L2.
GDP per capita and total GDP are 2025 USD equivalent (Maddison real growth index x 2025 nominal anchor); not purchasing-power parity.
The historical sector share (farming-extraction / industry / services) is a separate lens from the layer split; it is employment-based, and the value-added services share is higher.
2Land and surface
The planet's solid surface is fixed: 149 million km² of land, 130 million of it ice-free. What changes is how much of that land is ploughed, grazed and built on, from nothing toward half.
| Series | Source | Type / note |
|---|---|---|
| Earth total surface area | USGS; NOAA, Earth's surface area | measured · Constant. Oceans 361 million km², land 149 million km². Two endpoints are given; it does not change at the intermediate years. |
| Ice-free land | layer_horizontal_matrix.json (buzsuz kara 130 milyon km², bugün); HYDE 3.3 (Klein Goldewijk et al. 2017, ESSD 9:927-953); Peltier 2004 (ICE-5G buzul geri çekilmesi) | estimate · At the end of the last glacial the ice sheets were larger; ice-free land approached its present value around 6000 BCE. Antarctic ice is outside 'usable' land. |
| Forest area | HYDE 3.3; Ramankutty & Foley 1999 (Global Biogeochem. Cycles 13:997-1027); Ellis et al. 2010 (Global Ecol. Biogeogr. 19:589-606); FAO, Global Forest Resources Assessment 2020 (yakın dönem) | estimate · Early-Holocene forest cover is around 6 billion hectares. Pre-industrial loss concentrated in the temperate zone; the post-1900 acceleration is tropical. FAO 2020: 4.06 billion hectares. |
| Cropland | HYDE 3.3 (Klein Goldewijk et al. 2017, ESSD 9:927-953); Ramankutty & Foley 1999 (Global Biogeochem. Cycles 13:997-1027); FAOSTAT, Land statistics (2021, yakın dönem) | estimate · Zero at the start of agriculture; 1.6 billion hectares today (arable land + permanent crops). The largest expansion is between 1800 and 1950. |
| Pasture and grazing land | HYDE 3.3 (otlak alanı); FAOSTAT, daimi çayır ve mera (2021, yakın dönem) | estimate · The most contested land-cover series. HYDE 'grazing land' and FAO 'permanent meadows and pasture' are not the same thing; the line between naturally grazed rangeland and sown fodder land is disputed. Band 3.0-3.5 billion hectares. |
| Built-up and infrastructure area | HYDE 3.3 (yapılı alan); Gong et al. 2020 (Remote Sensing of Environment 236:111510, GAIA yapay geçirimsiz yüzey 0,80 milyon km², 2018, daha geniş tanım) | estimate · HYDE built-up is narrowly defined (around 0.5 million km² in 2017); the GHSL/GAIA artificial-impervious-surface measurement is 0.80 million km² (2018) and covers only the recent period. The definition gap is large. |
| Human land footprint as share of ice-free land | türetilmiş: (ekili alan + otlak + yapılı alan) ÷ buzsuz kara; bileşenler HYDE 3.3 + FAOSTAT | estimate · This series measures intensive use (cropland + pasture + built-up) and reaches 42% of ice-free land today. On the broader 'anthrome' definition, which includes used forests and populated rangelands, the human-influenced share is about half. |
The Antarctic ice sheet is kept outside 'usable' land.
'Pasture' is the most contested land-cover series: HYDE grazing land and FAO permanent meadows and pasture do not match; the split between natural rangeland and sown fodder land is unclear.
Built-up: HYDE's narrow definition is around 0.5 million km² in 2017; the GHSL/GAIA artificial-impervious-surface measurement is 0.80 million km² (2018) and covers only the recent period.
Layer mapping: L0 = cropland + pasture + mining; L1 = industry/logistics/rail; L2 = built-up and titled land; L3 = data-centre floor area (layer_horizontal_matrix.json).
The human-footprint series measures intensive use; on the 'anthrome' definition the human-influenced share is about half.
3Energy
A forager burned only food and firewood: a few gigajoules a head. The world average today is twenty times that, fifty times in rich countries, and three quarters of it is fossil.
| Series | Source | Type / note |
|---|---|---|
| Total primary energy consumption | Smil 2017, Energy Transitions (2nd ed.) ve Energy and Civilization (2017); Krausmann et al. 2009 (Ecological Economics 68:2696-2705) uzun dönem; Energy Institute, Statistical Review of World Energy 2024 (1965 sonrası); Our World in Data (1800 sonrası) | estimate · Primary energy on the substitution method. The forager level is food plus firewood. From 1965 the figures are Energy Institute measurement; before 1900 they are Smil's physical reconstruction. On a like-for-like physical basis the recent total is lower. |
| Primary energy per capita | türetilmiş: toplam birincil enerji ÷ nüfus (historical_series.json · nufus_milyon); taban değerler Smil 2017 | estimate · Most pre-industrial energy is muscle (human and draft animal) and firewood; whether food and fodder energy is counted materially changes the early figures. In rich countries today it exceeds 250 GJ per head. |
| Fossil fuel share of primary energy | Smil 2017, Energy Transitions (2nd ed.); Our World in Data, Energy Mix; Energy Institute, Statistical Review of World Energy 2024 | estimate · Fossil energy overtook biomass as the primary source around 1900. The share peaked near 87% in the 2000s and has been declining slowly since. |
| Electricity generation | Smil 2017; Our World in Data, Electricity Generation; Energy Institute, Statistical Review of World Energy 2024; IEA | estimate · Zero before 1880. Electricity is not a separate primary source but a carrier; it is not double-counted in the primary-energy total. |
Headline: primary energy on the substitution method. Final energy is around 440 EJ; useful ('exergy') work is about a third of primary energy.
Pre-industrial energy is mostly muscle (human and draft animal) and firewood; whether food and fodder energy is counted materially changes the early totals.
Incoming solar energy (around 173,000 TW) is excluded; it is named as a scale anchor, humanity uses about one ten-thousandth of it.
Electricity is not a separate primary source but a carrier.
Figures from 1965 are measured from the Energy Institute Statistical Review; before 1900 they are Smil's physical reconstruction.
4Climate and Atmosphere
For ten thousand years atmospheric CO₂ held between 260 and 280 ppm. In the industrial age it passed 420 ppm, a level not seen since the Pliocene, and is rising a hundred times faster than the fastest natural increase.
| Series | Source | Type / note |
|---|---|---|
| Atmospheric CO₂ | Antarktika buz çekirdekleri, EPICA Dome C, Law Dome (Lüthi et al. 2008, Nature 453:379-382; MacFarling Meure et al. 2006, GRL 33:L14810); NOAA Global Monitoring Laboratory, Mauna Loa (1958 sonrası) | measured · Early Holocene 260-265 ppm; 278 ppm in 1750; ice core and direct measurement are spliced. The 2024 annual mean is 422.7 ppm. |
| Atmospheric CH₄ | Antarktika ve Grönland buz çekirdekleri, Law Dome (MacFarling Meure et al. 2006, GRL 33:L14810); NOAA Global Monitoring Laboratory (1983 sonrası) | measured · Pre-industrial methane is around 700 ppb; around 1,930 ppb in 2024. |
| Global mean temperature anomaly | PAGES 2k Consortium 2019 (Nature Geoscience 12:643-649); Kaufman et al. 2020 (Scientific Data 7:201) Holosen için; IPCC AR6 WG1 (2021), HadCRUT5 ve NASA GISTEMP aletli dönem için | estimate · The pre-industrial baseline is 1850-1900. The magnitude of the early-to-mid Holocene thermal maximum is disputed: proxies suggest around 0.5 °C, climate models less. From 1900 the figures are instrumental. 2024: 1.3 °C above 1850-1900. |
| Annual anthropogenic CO₂ emissions (fossil and industry) | Global Carbon Project 2024 (Friedlingstein et al. 2024, Earth Syst. Sci. Data); Boden, Marland & Andres / CDIAC tarihsel seri | measured · Fossil fuel and industry (cement) emissions only. Land-use change adds around a further 4 Gt CO₂ a year and carries high uncertainty. Before 1900 the figures are reconstruction. |
| Cumulative anthropogenic CO₂ (since 1750, fossil and industry) | Global Carbon Project 2024; IPCC AR6 WG1 (2021) | measured · Fossil and industry only. Including land-use change, the total since 1750 is around 2,500 Gt CO₂. The series starts in 1700; it is zero before then. |
The basket is CO₂, CH₄, N₂O; CO₂-equivalent is computed with the 100-year global warming potential (GWP-100).
Pre-industrial baseline: 1850-1900 for temperature, 1750 for CO₂.
Land-use-change emissions carry large uncertainty (around 4 Gt CO₂ a year, wide band).
The magnitude of the early-to-mid Holocene thermal maximum is disputed between proxies and climate models.
The remaining carbon budget for 1.5 °C is around 200-250 Gt CO₂ (start of 2024; IPCC AR6).
5Water
The planet holds 1.386 billion km³ of water; 2.5% is fresh, and the accessible part is under 1%. Human withdrawal has gone from zero to 4,000 km³ a year, and agriculture takes 70% of it.
| Series | Source | Type / note |
|---|---|---|
| Total water volume | USGS, Where is Earth's Water?; Shiklomanov 1993 | measured · Constant. Two endpoints are given. Freshwater is 2.5% of this (around 35 million km³); of freshwater, 69% is ice and snow, 30% groundwater, 0.3% rivers and lakes. |
| Annual renewable freshwater flux | Oki & Kanae 2006 (Science 313:1068-1072); Shiklomanov 2000 (Water International 25:11-32); UN World Water Development Report | estimate · Land precipitation minus evapotranspiration; in the band 40,000-45,000 km³/yr. Taken as constant; two endpoints are given. |
| Human freshwater withdrawal | Shiklomanov 2000 (Water International 25:11-32); FAO AQUASTAT; UN World Water Development Report (yakın dönem) | estimate · Before 1900 the series is coarse and low-confidence; irrigation begins around 6000 BCE. From 1900 it rests on Shiklomanov and AQUASTAT measurement. |
| Agriculture share of total withdrawal | FAO AQUASTAT; Shiklomanov 2000 (Water International 25:11-32) | estimate · Today agriculture is 70%, industry 19%, municipal 11%. The series starts in 1900; the sector split is not reliable before then. |
| Groundwater net depletion | Wada et al. 2010 (Geophysical Research Letters 37:L20402); Konikow 2011 (GRL 38:L17401); Famiglietti 2014 (Nature Climate Change 4:945-948) | estimate · Withdrawal in excess of recharge; in the band 150-300 km³/yr. The series starts in 1900. |
Stocks are taken as constant; only the withdrawal and groundwater-depletion series change over time.
Headline: withdrawal. Consumption and 'water footprint' are separate concepts and lower.
Blue water (rivers and aquifers) differs from green water (soil moisture, rainfed agriculture); the headline is blue water, green water is larger for agriculture.
The pre-1900 withdrawal series is coarse; irrigation begins around 6000 BCE but a reliable global series exists only after 1900.
The freshwater stock split is the USGS convention (ice 69%, groundwater 30%, surface 0.3%).
6Matter and Biosphere
Annual material extraction has gone from a few megatonnes to 85 Gt. Over the same span living plant mass has halved, and human-made mass, concrete, asphalt, metal, plastic, passed all living biomass around 2020.
| Series | Source | Type / note |
|---|---|---|
| Annual material extraction | historical_series.json · malzeme_gton (Krausmann et al. 2009, Ecological Economics 68:2696-2705; UNEP IRP Global Material Flows Database 2024; civilization-dynamics Layer 0 modeli 2025 çıpası 85,06 Gton) | estimate · The headline is 'used domestic extraction': biomass, fossil fuels, metal ores, industrial and construction minerals; waste rock and overburden excluded. The UNEP IRP current series gives around 100 Gt for 2024; the civilization-dynamics Layer 0 anchor is 85.06 Gt for 2025, a difference of year and boundary. |
| Construction minerals share of extraction | historical_series.json · malzeme_insaat_mineral_gton ÷ malzeme_gton (Krausmann et al. 2009; UNEP IRP Global Material Flows Database 2024) | estimate · 2025 split: construction minerals 51.36 Gt (60%), fossil 17.43 Gt (21%), biomass 13.12 Gt (15%), metal ores 2.62 Gt (3%), industrial minerals 0.53 Gt (1%). Sand, gravel and crushed stone dominate. |
| Global living biomass | Bar-On, Phillips & Milo 2018 (PNAS 115:6506-6511); Erb et al. 2018 (Nature 553:73-76) sanayi öncesi ayakta biyokütle kaybı için | estimate · Carbon mass (the Bar-On convention). Today 550 Gt carbon; of that, 450 is plants, 70 bacteria, 12 fungi, 2 animals, 0.06 humans, 0.10 livestock. Erb et al. 2018: standing biomass is about half of its pre-industrial potential. |
| Technosphere, human-made mass | Elhacham, Ben-Uri, Grozovski, Bar-On & Milo 2020 (Nature 588:442-444) | estimate · Dry weight, 'in-use' stock. Around 30 Gt in 1900, around 1,100 Gt in 2020, doubling about every 20 years. Concrete and aggregate are three quarters of the mass. The wet ('ever produced') variant is about twice this. The series starts in 1 CE. |
| Technosphere ÷ living biomass (dry weight) | türetilmiş: Elhacham et al. 2020 (insan yapımı kütle) ÷ kuru biyokütle (Bar-On et al. 2018 karbon kütlesinin iki katı). Kesişim 2020 dolayı, belirsizlik 6 yıl | estimate · Both sides are dry weight. Bar-On's 550 Gt is carbon mass; dry weight is about twice that. Human-made mass passed living biomass around 2020. |
Extraction: 'used' (biomass + fossil + metal ores + minerals) differs from 'total' (including waste rock and gangue, about twice as much). The headline is used extraction.
The civilization-dynamics Layer 0 anchor is 85.06 Gt for 2025; the UNEP IRP current series gives around 100 Gt for 2024, a difference of year and boundary.
Biomass is given as carbon mass (the Bar-On convention); wet and dry weight differ, the comparison with the technosphere is on a dry-weight basis.
Technosphere: 'in-use' stock differs from 'ever produced'; Elhacham et al. 2020 give both variants.
The pre-industrial biomass and technosphere series are wide-band reconstructions.
7Primary sources
- civilization-dynamics, historical_series.json (population, GDP, materials back to 10,000 BCE at 14 era years); layer_workforce.json; layer_learned_extras.json; layer_horizontal_matrix.json
- McEvedy & Jones 1978, Atlas of World Population History (Penguin)
- HYDE 3.3, Klein Goldewijk, Beusen, Doelman & Stehfest 2017 (Earth System Science Data 9:927-953)
- UN DESA, World Population Prospects 2024
- Riley 2005, Estimates of Regional and Global Life Expectancy, 1800-2001 (Population and Development Review 31:537-543)
- Reba, Reitsma & Seto 2016, Spatializing 6,000 years of global urbanization (Scientific Data 3:160034)
- Maddison Project Database 2023, Bolt & van Zanden 2024 (Journal of Economic Surveys)
- ILO, modelled estimates, ILOSTAT (2024 update); ILO 2018, Care work and care jobs for the future of decent work; ILO, World Employment and Social Outlook: Trends 2024
- Ramankutty & Foley 1999 (Global Biogeochemical Cycles 13:997-1027); Ellis et al. 2010 (Global Ecology and Biogeography 19:589-606); Ellis et al. 2021 (PNAS 118:e2023483118)
- FAO, Global Forest Resources Assessment 2020; FAOSTAT, Land statistics 2001-2023 (Analytical Brief 107, 2025)
- Maus, Giljum, da Silva et al. 2022, An update on global mining land use (Scientific Data 9:433)
- Smil 2017, Energy and Civilization: A History (MIT Press); Smil 2017, Energy Transitions: Global and National Perspectives (2nd ed.)
- Energy Institute, Statistical Review of World Energy 2024; IEA, World Energy Balances 2024
- Lüthi et al. 2008 (Nature 453:379-382); MacFarling Meure et al. 2006 (Geophysical Research Letters 33:L14810); NOAA Global Monitoring Laboratory (Mauna Loa)
- IPCC AR6 WG1 2021; Global Carbon Project 2024, Friedlingstein et al. 2024 (Earth System Science Data)
- PAGES 2k Consortium 2019 (Nature Geoscience 12:643-649); Kaufman et al. 2020 (Scientific Data 7:201)
- Shiklomanov 2000, Appraisal and assessment of world water resources (Water International 25:11-32); FAO AQUASTAT; USGS, Where is Earth's Water?
- Wada et al. 2010 (Geophysical Research Letters 37:L20402); Konikow 2011 (Geophysical Research Letters 38:L17401)
- Krausmann et al. 2009 (Ecological Economics 68:2696-2705); UNEP International Resource Panel, Global Material Flows Database (2024 update)
- Bar-On, Phillips & Milo 2018, The biomass distribution on Earth (PNAS 115:6506-6511); Erb et al. 2018 (Nature 553:73-76)
- Elhacham, Ben-Uri, Grozovski, Bar-On & Milo 2020, Global human-made mass exceeds all living biomass (Nature 588:442-444)
- Population Reference Bureau 2022, How Many People Have Ever Lived on Earth?