Overview
This visualization correlates two independent paleoclimate temperature reconstructions -- Loehle (2007) and Ljungqvist (2010) -- with 132 major historical events spanning 1 AD to 2000 AD across six world regions and ten event categories. The dataset covers empires, wars, plagues, explorations, inventions, collapses, trade networks, and agricultural revolutions on every inhabited continent.
The project began as a question: does the temperature record align with when civilizations flourished and when they collapsed? The answer, visible in the chart above, is that the alignment is striking -- though not deterministic. Climate did not cause history. But it shaped the conditions within which history happened, raising the carrying capacity of populations during warm periods and compressing it during cold ones.
The dataset is version 1 of an ongoing research project. Each event carries a confidence rating, source citations, and -- where scholarly literature supports it -- an explicit climate note connecting the event to the temperature record. 56 of 132 events carry such notes.
Why This Matters
The question of whether climate drives civilization is one of the oldest and most contested in historical scholarship. This visualization does not resolve that debate. What it does is make the question visible and explorable in a way that static timelines and academic papers cannot. When you see the Medieval Warm Period and the Crusades, the Norse settlement of Greenland, the Song Dynasty technological revolution, and the Polynesian settlement of New Zealand all clustering on the same warm anomaly in the chart, the hypothesis earns scrutiny it might not otherwise receive.
The Little Ice Age cluster is the most telling. Between 1300 and 1350 AD -- a single half-century of climate descent -- the Great Famine of Europe, the Black Death, the Hundred Years War, the end of Norse Greenland, and China's maritime retreat all appear in the dataset. No single thesis explains all of them. But the temperature record is the one thing they share.
Understanding this relationship has immediate relevance. The Current Warm Period is the steepest temperature rise in the 2000-year record. Whether human civilization can manage that expansion without triggering the same collapse dynamics that ended previous warm-period civilizations is the defining question of this century.
Events
Correlations
Research synthesis based on peer-reviewed paleoclimate literature and historical scholarship. Climate connections are noted where scholarly consensus supports them. Not all historical events correlate with climate -- institutional, political, and cultural factors matter independently.
The Warm Period Pattern
Roman Warm Period (1--300 AD): The Imperial Window
The Roman Climate Optimum (roughly 200 BC to 150 AD) enabled the agricultural surpluses that fed Roman legions and urban populations. Historian Kyle Harper in The Fate of Rome (2017) draws a direct line between stable, warm, wet Mediterranean conditions and Rome's peak prosperity under Trajan around 100 AD. When the climate began cooling after 150 AD, the first plagues arrived -- the Antonine Plague (165 AD) and Plague of Cyprian (250 AD) -- and the structural stresses that would eventually destroy the empire began compounding.
The same period saw the Han Dynasty in China reach its peak, Teotihuacan rise to dominate Mesoamerica, and the Kingdom of Aksum control Red Sea trade. The Roman Warm Period was not a European phenomenon -- it appears in the global Loehle reconstruction as a sustained positive anomaly correlating with civilizational peaks across multiple continents simultaneously.
Dark Age Cold Period (300--900 AD): Cascading Collapse
The transition out of the Roman Warm Period tracks almost exactly with the fall of the Western Roman Empire (476 AD) and the simultaneous weakening of the Han Dynasty. The 536 AD event -- a volcanic winter lasting 18 months that tree ring records identify as the coldest decade in 2,300 years -- appears in the dataset as the single most dramatic short-term climate shock. The Plague of Justinian followed five years later in 541 AD.
The late Dark Age Cold Period (750--900 AD) shows a partial recovery in the temperature record corresponding with the founding of the Tang Dynasty in China (618 AD), the Islamic Golden Age beginning under the Abbasid Caliphate (750 AD), and the early Viking expansion (793 AD onward). The warming trend that would become the Medieval Warm Period was already creating conditions for expansion before reaching its peak.
Medieval Warm Period (900--1300 AD): The Great Expansion
The Medieval Warm Period contains the densest cluster of expansion events in the 2000-year dataset. Norse settlement of Greenland (985 AD), Leif Erikson reaching North America (1000 AD), the Crusades (1095 AD), the Song Dynasty technological revolution, Cahokia in North America, Polynesian settlement of Hawaii, New Zealand, and Easter Island, the Mali Empire, Angkor Wat, and the European cathedral-building boom all fall within this warm window.
PNAS research (2014) on Polynesian migration provides the most precise climate-exploration correlation in the dataset. Paleoclimate wind field reconstructions show that the Medieval Climate Anomaly shifted Pacific wind patterns in ways that opened downwind sailing routes to New Zealand and Easter Island that would not exist under modern or Little Ice Age conditions. The Polynesian expansion was, in a direct physical sense, enabled by the Medieval Warm Period.
The Cold Period Pattern
Little Ice Age (1300--1850 AD): The Great Contraction
The Little Ice Age descent produces the most striking clustering in the dataset. Between 1300 and 1350 AD -- within a single generation of climate deterioration -- the Great Famine of Europe (1315-1322), the Black Death arrival (1347), the Hundred Years War (1337), and the beginning of Norse Greenland's terminal decline all appear. The Wikipedia article on the Crisis of the Late Middle Ages explicitly states this crisis period coincides with the shift from the Medieval Warm Period to the Little Ice Age.
The Maunder Minimum (1645-1715), the coldest phase of the Little Ice Age, corresponds almost exactly with the Thirty Years War (1618-1648), the English Civil War, the Qing Dynasty's replacement of the Ming following catastrophic famines, and what historian Geoffrey Parker in Global Crisis (2013) calls a simultaneous collapse of states across Eurasia.
Current Warm Period (1850--2000 AD): Unprecedented Acceleration
The Current Warm Period shows the steepest temperature rise in the 2000-year dataset. It coincides with the Industrial Revolution, two world wars, the Green Revolution, the space age, and the World Wide Web -- a compression of civilizational advancement without historical precedent. The Green Revolution (1966) inverts the typical climate-civilization relationship: Norman Borlaug's high-yield wheat varieties decoupled food production from climate variability for the first time in human history, effectively making it the dataset's most powerful counterexample to climate determinism.
Cross-Cutting Observations
Plague as a Climate Amplifier
All five major plague events in the dataset follow periods of climate stress. Nutritional stress from poor harvests weakens immune systems, population displacement from famine moves people into contact with disease vectors, and cold wet conditions favor certain pathogen transmission routes. The Black Death arriving within 30 years of the Great Famine, in a Europe already weakened by cold-weather crop failures, is the textbook case.
Retreat and Isolation as Cold-Period Responses
Three of the most historically significant withdrawals in the dataset cluster in the Little Ice Age: China ends maritime exploration (1433), the Norse abandon Greenland (1400), and Japan enters 250 years of isolation under the Edo shogunate (1603). All three represent civilizations turning inward under resource pressure. The pattern suggests cold periods drive not just collapse but contraction -- civilizations abandoning costly extensions and consolidating around cores.
Where Climate Correlation Breaks Down
The Italian Renaissance flourished during the Little Ice Age. The Islamic Golden Age peaked during the Dark Age Cold Period. The Mughal Empire reached its height under Akbar during the Little Ice Age. Climate sets a ceiling on what civilizations can sustain, but within that ceiling, institutions, ideas, trade networks, and individual leadership produce wildly divergent outcomes. The correlations are real and significant. They are not destiny.
Concepts
Climate Eras
Roman Warm Period (1--300 AD)
A period of stable, warm, wet conditions across the northern hemisphere broadly corresponding to the height of the Roman Empire and Han Dynasty. Loehle (2007) shows positive temperature anomalies peaking around 100 AD. Characterized by agricultural surpluses, population growth, and urban expansion across Eurasia.
Dark Age Cold Period (300--900 AD)
A prolonged cooling following the Roman Warm Period, with a particularly severe trough around 536--660 AD linked to volcanic eruptions. Both reconstructions show sustained negative anomalies. Corresponds broadly with the fall of Rome, the Migration Period, and the Plague of Justinian.
Medieval Warm Period (900--1300 AD)
Also called the Medieval Climate Anomaly. Elevated temperatures across the northern hemisphere peaking roughly 950--1150 AD. The Ljungqvist reconstruction shows this as the warmest pre-industrial period in the 2000-year record. Corresponds with Norse expansion, the Crusades, and the Polynesian settlement of the Pacific's most remote islands.
Little Ice Age (1300--1850 AD)
Not a single continuous cold period but a series of cooling phases separated by partial recoveries, with the most severe cold during solar Grand Minima (Sporer, Maunder, Dalton Minimums). Characterized by advancing alpine glaciers, frozen rivers, shortened growing seasons, and recurring harvest failures. Corresponds with the Black Death, the 17th century global crisis, the collapse of Ming China, and Norse Greenland's abandonment.
Current Warm Period (1850--present)
The warming trend beginning around 1850 as the Little Ice Age ended. The Ljungqvist reconstruction shows the steepest rise in the 2000-year dataset in the 20th century. Corresponds with the fastest period of technological advancement, population growth, and civilizational complexity in human history.
Data Sources
Loehle (2007): Global Non-Tree-Ring Reconstruction
Craig Loehle's 2000-year global temperature reconstruction uses 18 non-tree-ring proxy records including lake sediments, stalagmites, coral, and ice cores. Developed to avoid the controversy surrounding tree-ring proxies. Shows a clear Medieval Warm Period peaking around 1000 AD and a Little Ice Age trough.
Ljungqvist (2010): Extratropical Northern Hemisphere
Fredrik Ljungqvist's 2000-year reconstruction covers the extratropical northern hemisphere using 30 proxy records. Generally shows higher amplitude variability and a more pronounced Medieval Warm Period. The two reconstructions together bracket the plausible range of northern hemisphere temperature history.
Event Classification
Categories and Confidence Levels
Each event is classified into one or more of ten categories and assigned a confidence level. High confidence indicates well-documented events with unambiguous dates in mainstream scholarly literature. Medium confidence indicates events where dates are approximate or scholarly debate exists. Climate notes are included only where peer-reviewed literature explicitly supports the connection.
Implementation
Practical guidance for using this visualization as a research and analysis tool.
Start with the chart, not the event list
The climate curve is the primary axis of this visualization. Before filtering events, look at the temperature record as a continuous story. Identify the peaks and troughs -- the Roman Warm Period plateau, the 536 AD crash, the Medieval Warm Period double peak, the Little Ice Age descent, and the Current Warm Period rise. Understanding the shape of the temperature record makes the event correlations visible when you add them back in.
Click dots on the chart to explore events
Each dot on the chart is a historical event. Gold-outlined dots have explicit climate connections in the scholarly literature. Click any dot to open a modal with the full event description, climate note, and source citation. Events with a gold-bordered climate connection block in the modal have the strongest temperature-record ties.
Use the era filter on the Events tab to isolate a period
Filter by a single climate era to see all events within it in chronological order. The Little Ice Age filter (46 events) is the most revealing -- it shows the density of collapse, war, and famine events concentrated in the cooling period while also showing the Renaissance and Reformation, reminding you that climate correlation is not determinism.
Use the climate connection as a research lead
Events with gold climate notes point to specific papers and historians -- Kyle Harper's The Fate of Rome, Geoffrey Parker's Global Crisis, Pederson et al. PNAS on Mongol expansion, Woodruff's lake sediment study on kamikaze typhoons. These develop the climate-civilization connection in far more depth than the event cards can.
Look for the counterexamples
Events without climate notes are as analytically important as those with them. The Italian Renaissance (1450), the Islamic Golden Age (750), and the Mughal Empire under Akbar (1556) all flourished during cold periods. These cases demonstrate that institutional quality, trade networks, and intellectual culture can sustain civilizational achievement independently of climate conditions.
Use the AI Prompt tab to explore events outside the database
The AI Implementation Prompt is designed specifically for this. Copy it, paste it into any capable AI assistant, then describe an event not in the database. The AI will place it within the climate era framework, assess the temperature anomaly at the time, identify whether scholarly literature supports a climate connection, and flag related events already in the dataset.
Cite the primary sources, not this page
This visualization is a synthesis and exploration tool. When using any climate-civilization correlation from this page in research or publication, verify against the primary source. The event source citations are starting points, not exhaustive bibliographies.
Tools & Resources
Primary Climate Sources
| Resource | Description |
|---|---|
| Loehle (2007) | A 2000-year global temperature reconstruction based on non-tree-ring proxies. Energy and Environment. Primary global temperature source for this visualization. |
| Ljungqvist (2010) | A 2000-year extratropical northern hemisphere temperature reconstruction. Geografiska Annaler. Provides the second curve and a regional northern hemisphere perspective. |
| Ruter et al. / PNAS (2014) | Climate windows for Polynesian voyaging to New Zealand and Easter Island. Direct evidence for climate-driven exploration windows during the Medieval Climate Anomaly. |
Key Books
| Resource | Description |
|---|---|
| Harper, K. (2017) | The Fate of Rome: Climate, Disease, and the End of an Empire. Princeton University Press. Foundational text connecting Roman climate data to the empire's collapse. |
| Parker, G. (2013) | Global Crisis: War, Climate Change and Catastrophe in the Seventeenth Century. Yale University Press. Definitive account of the 17th century global crisis and its climate drivers. |
| Fagan, B. (2008) | The Great Warming. Bloomsbury. Broad survey of the Medieval Warm Period and its civilizational correlates across multiple regions. |
| Fagan, B. (2001) | The Little Ice Age: How Climate Made History 1300-1850. Basic Books. Standard popular account of Little Ice Age impacts on European civilization. |
| Diamond, J. (2005) | Collapse: How Societies Choose to Fail or Succeed. Viking. Covers Easter Island, Norse Greenland, Maya, and other collapse cases. Partially superseded by subsequent research on Easter Island but still a useful framework. |
Suggested Further Research
| Resource | Description |
|---|---|
| NOAA Paleoclimate Database | Primary archive for paleoclimate proxy records including the Loehle and Ljungqvist datasets. Allows download of raw data for independent analysis. |
| World History Encyclopedia | Peer-reviewed open-access historical encyclopedia. Primary verification source for event dates, especially for ancient and medieval events. |
| The Conversation (2015) | Climate and the rise and fall of civilizations. Accessible overview of climate-civilization research with links to primary literature. |
Source Material
Dataset Structure
Each event carries: id, year, title, description, region (Europe / Asia / Americas / Middle East / Africa / Global), category (empire, war, collapse, invention, science, exploration, trade, religion, plague, agriculture), climate_era (rwp / dacp / mwp / lia / cwp), climate_note (explicit scholarly connection, or null), confidence (high / medium), and source (short citation traceable to primary literature).
The dataset is stored as JSON at claytonsmith.ca/pages/data/climate_civ_timeline_v20260621-01.json and fetched by the page on load. The version suffix follows the YYYYMMDD-NN convention.
Event Count by Region (v20260621-01)
Europe: 45 • Asia: 29 • Americas: 19 • Middle East: 16 • Africa: 13 • Global: 10 • Total: 132 events. 56 events carry explicit climate notes. 126 events rated high confidence, 6 medium confidence.
Event Count by Climate Era
Roman Warm Period: 10 • Dark Age Cold Period: 27 • Medieval Warm Period: 26 • Little Ice Age: 46 • Current Warm Period: 23
Known Gaps and Planned Additions
Underrepresented in version 1 and planned for future dataset versions: Oceania and Pacific Island civilizations beyond Polynesian expansion; Japan -- Sengoku period, Russo-Japanese War, WWII Pacific theater; South and Southeast Asia -- Chola Empire, Khmer collapse in depth, Vietnamese kingdoms; Ottoman decline and dissolution (1700--1922); Cold War proxy conflicts and decolonization wave in Africa and Asia; Latin American independence movements; pre-1 AD foundational civilizations for context band.
AI Prompt
Custom AI implementation prompt for exploring climate-civilization correlations and placing events within the 2000-year temperature record.
AI Implementation Prompt -- Climate and Civilization
CONTEXT You are working with a research dataset called Climate and Civilization: 2000 Years of Human History. The dataset correlates two independent paleoclimate temperature reconstructions with 132 major historical events spanning 1 AD to 2000 AD. The two temperature sources are: -- Loehle (2007): A global reconstruction using 18 non-tree-ring proxy records (lake sediments, stalagmites, coral, ice cores). Shows temperature anomaly relative to the 1961-1990 baseline. -- Ljungqvist (2010): An extratropical northern hemisphere reconstruction using 30 proxy records. Generally shows higher amplitude variability and a more pronounced Medieval Warm Period. The dataset covers six regions (Europe, Asia, Americas, Middle East, Africa, Global) and ten event categories (empire, war, collapse, invention, science, exploration, trade, religion, plague, agriculture). FIVE CLIMATE ERAS Roman Warm Period (1-300 AD): Stable, warm, wet conditions. Positive temperature anomaly peaking around 100 AD. Corresponds with peak Roman Empire, Han Dynasty, and Teotihuacan. Dark Age Cold Period (300-900 AD): Prolonged cooling with a severe trough around 536-660 AD (Late Antique Little Ice Age caused by volcanic eruptions). Corresponds with fall of Rome, Migration Period, Plague of Justinian, Han fragmentation. Medieval Warm Period (900-1300 AD): Elevated temperatures peaking 950-1150 AD. Corresponds with Norse expansion, Crusades, Polynesian Pacific settlement, Song Dynasty technology revolution, European population boom. Little Ice Age (1300-1850 AD): Series of cooling phases with coldest during the Maunder Minimum (1645-1715). Corresponds with Black Death, Great Famine, Hundred Years War, Norse Greenland abandonment, China maritime retreat, 17th century global crisis, French Revolution. Current Warm Period (1850-2000 AD): Steepest rise in the 2000-year dataset. Corresponds with Industrial Revolution, two world wars, Green Revolution, space age, information age. KEY PRINCIPLES 1. Climate sets a ceiling on what civilizations can sustain, but does not determine outcomes within that ceiling. 2. Warm periods correlate with agricultural surplus, population growth, and expansion events. Cold periods correlate with harvest failures, plague vulnerability, and contraction events. 3. The plague-famine-war cluster is a recurring pattern: climate stress produces food shortages, weakened immunity, population displacement, and conflict -- often within a single generation. 4. Retreat and isolation appear as cold-period responses: China maritime closure (1433), Norse Greenland abandonment (1400), and Japan Edo isolation (1603) all cluster in the Little Ice Age. 5. Counterexamples matter: the Italian Renaissance, Islamic Golden Age, and Mughal Empire under Akbar all flourished during cold periods. Climate is not destiny. 6. The 536 AD volcanic winter is the most dramatic single climate shock in the dataset -- the coldest decade in 2,300 years, followed within 5 years by the Plague of Justinian. 7. Polynesian Pacific expansion (1100-1300 AD) is one of the most precisely documented climate-exploration correlations: PNAS research shows Medieval Warm Period wind pattern shifts opened downwind sailing routes to New Zealand and Easter Island that did not exist under modern or Little Ice Age conditions. IMPLEMENTATION MODES When a user brings you an event not in the database, help them by: ERA PLACEMENT: Identify which of the five climate eras the event falls within. Describe the approximate temperature anomaly based on the Loehle and Ljungqvist reconstructions. Note whether the event falls during a warming or cooling phase within its era. CLIMATE CONNECTION ASSESSMENT: Evaluate whether peer-reviewed literature supports a climate connection. Rate as: well-documented (explicit scholarly literature), plausible (logical mechanism exists, limited direct research), or speculative (no strong evidence). Be honest about the limits of the evidence. DATASET COMPARISON: Identify events already in the database that are geographically, temporally, or thematically related. Note whether the new event extends an existing pattern. PATTERN IDENTIFICATION: When the user describes multiple events or asks about a region or time period, identify whether the events cluster in ways that align with the climate eras. Distinguish clustering from coincidence by asking whether a plausible causal mechanism exists. SKEPTIC MODE: On request, challenge a proposed climate-civilization connection. What are the non-climate explanations? What would the evidence look like if climate were not a factor? DATASET EXPANSION: Help the user draft a new event entry in the dataset JSON format, including year, title, description, region, category, climate_era, climate_note, confidence, and source fields. RESEARCH EXPANSION: Suggest specific peer-reviewed papers, historians, or datasets that would help the user investigate a particular climate-civilization connection more deeply. AI OPERATING INSTRUCTIONS Stay grounded in the paleoclimate literature. Distinguish between what Loehle and Ljungqvist actually show and what is inferred from other proxy records. When making climate-civilization connections, specify the scholarly basis. Avoid speculative connections presented as established facts. For event placement, be specific about temperature: instead of saying it was a cold period, say the Ljungqvist reconstruction shows a temperature anomaly of approximately X degrees C relative to the 1961-1990 baseline during this period. Challenge weak causal claims. A correlation between a cold period and a war does not establish that climate caused the war. Ask: what is the mechanism? What does the historical record say about food prices, harvest failures, or population displacement at the time? GUIDED DISCOVERY Ask me up to three questions, one at a time, to determine: (1) what event or topic I want to explore and whether it is inside or outside the existing database, (2) whether I am looking to understand a climate connection, place an event in the temperature record, or expand the dataset, (3) what level of scholarly rigor I need. Once you understand my situation, help me build a practical analysis or dataset expansion plan.