Medieval Chronicles Turned Climate Data Reveal Centuries of Drought and Flood in the Levant

Long before thermometers and rain gauges, the scribes of Damascus, Aleppo, Jerusalem and Cairo recorded the weather that shaped their world. A chronicler noting that the winter rains failed, that a flood swept through a market, or that a drought drove up the price of bread left behind something more valuable than he could have imagined: a distributed network of environmental observations stretching back nearly two millennia. A new study published in Regional Environmental Change has now mined that archive, compiling thousands of historical hydrometeorological records from the Levant between 200 and 1850 CE and testing whether these centuries-old written accounts can stand up against the tree-ring reconstructions that modern paleoclimatologists rely on. The answer, the researchers report, is a qualified but encouraging yes.

The team, led by Mohammad Merheb of Institut Agro and INRAE in Rennes, France, together with colleagues from Montpellier and Imperial College London, assembled a database of 3,507 individual hydrometeorological records drawn from existing historical databases and published compilations. The sources behind these records are remarkably diverse. They include the annals of court chroniclers such as Ibn Al-Qalanisi of Damascus, who died in 1160, and Ibn Kathir, who wrote in the fourteenth century, as well as diaries like that of Ibn Tawq, a fifteenth-century Damascene whose daily jottings yielded more than 7,000 climate and environmental observations for the years 1481 to 1501 alone. Other records come from compilations of natural disasters in the medieval Levant and Egypt, and from coded datasets such as the Grotzfeld collection, which spans the Near and Middle East from 801 to 1821 CE.

Raw historical quotations, however, are not climate data. The core methodological achievement of the study lies in harmonization. Each of the 3,507 records was standardized according to three dimensions: the type of event described, whether drought, flood, heavy rainfall, or another hydrometeorological phenomenon; an intensity index that translates qualitative descriptions into a semi-quantitative scale; and the geographic location of the observation. This unified structure allowed the researchers to aggregate event frequencies by decade and to assess how well documented each period and place actually was. That coverage assessment matters enormously, because historical archives are notoriously biased toward times and places where literate elites had both the incentive and the leisure to write things down.

The spatial and temporal unevenness the team uncovered is striking. Documentation is densest and most continuous in Damascus, a city whose long tradition of chronicling, from Ibn Al-Qalanisi through Ibn Tulun in the early sixteenth century, produced an exceptionally rich written memory of weather and water. Other parts of the Levant appear only sporadically, often only when a disaster severe enough to merit mention struck. This unevenness shapes what the dataset can and cannot do. For most of the region and most of the period, the records offer snapshots rather than continuous series. But for Damascus between 1250 and 1520 CE, the density of long-term precipitation and flood records is sufficient to analyze variability at the decadal scale, and in some cases even at the seasonal scale.

The crucial test came when the researchers compared their documentary evidence with an independent, physically based source of paleoclimate information: the Old World Drought Atlas, or OWDA, a tree-ring-based reconstruction that provides summer Palmer Drought Severity Index values, or PDSI, across Europe, the Mediterranean and the Middle East for the past two millennia. Tree rings encode moisture stress in their growth rings, giving annually resolved reconstructions that are completely independent of what any medieval scribe wrote. If the two archives disagree systematically, the documentary records might reflect local anecdote, literary convention, or copying errors rather than real climate. If they agree, both are likely capturing genuine hydroclimatic signal.

To make the comparison rigorous, the team derived a dryness index directly from the historical descriptions, classifying episodes of drought and water scarcity recorded in the texts, and then evaluated the relationship between the documentary index and the OWDA-derived PDSI for Damascus using conditional probability. The analysis showed relative coherence between the two datasets: when the chronicles reported dry conditions, the tree rings tended to indicate drought as well, and vice versa. This agreement is significant because it validates, for the Levant, an approach that historical climatologists have applied successfully in Europe and China but that has remained comparatively underdeveloped in the eastern Mediterranean, where the documentary traditions, languages and archival conditions differ substantially.

One of the most intriguing findings emerged from the seasonal analysis of the historical records. Most of the dry episodes documented in the texts occurred during the winter and autumn rainy seasons, the very months on which the Levant depends for nearly all of its annual precipitation. In a region with a Mediterranean climate, where summers are reliably dry and agriculture hinges on the October-to-May rainy season, a failure of autumn or winter rains is not a minor inconvenience but a threat to harvests, water supplies and, historically, to political stability. The chronicles’ emphasis on these events suggests that medieval observers were recording precisely the hydrometeorological anomalies that mattered most to their societies, which in turn makes their testimony especially valuable for reconstructing drought variability.

The study’s implications reach beyond medieval history. The Levant is one of the world’s most water-stressed regions today, and climate projections generally indicate increasing aridity and greater hydrological variability in the decades ahead. Understanding how precipitation and drought behaved over the past sixteen centuries provides a baseline against which modern changes can be measured, and it extends the observational record far beyond the instrumental era, which in this region barely stretches back a century. Documentary records can also capture extreme events, such as destructive floods, that may be poorly represented in tree-ring reconstructions, which respond primarily to moisture stress rather than to individual episodes of heavy rainfall.

The authors are candid about the limitations that remain. The spatial distribution of surviving documents is highly uneven, concentrating evidence in a handful of urban centers while leaving much of the rural Levant in darkness. Many potential sources remain inaccessible or unexamined, and unlocking them requires proper curation by historians with the linguistic and paleographic skills to read Arabic, and other historical manuscripts in their original forms. The study is, in this sense, as much an argument for interdisciplinarity as it is a climate reconstruction: without collaboration between hydrologists and historians, a large fraction of the region’s environmental memory would stay locked in archives that scientists rarely visit.

Even with those caveats, the research demonstrates that fragmented documentary evidence, when systematically harmonized and analyzed, can yield partially consistent, semi-quantitative indicators of past hydroclimatic variability. The 3,507 records compiled here represent a foundation that can be expanded as more chronicles are transcribed, translated and coded, and as datasets like tambora.org continue to grow. For a region whose water future is uncertain, the meticulous observations of medieval scribes, cross-checked against the silent testimony of tree rings, offer a rare and precious long view of drought and flood, one written in ink but verified in wood.

Subject of Research: Historical documentary records of hydrometeorological events in the Levant compared with paleoclimate reconstructions

Article Title: Characterization of hydrometeorological events from historical records in the Levant (200–1850 CE) compared to paleoclimate reconstructions

Article References: Merheb, M., Moussa, R., Hamelin, G., Lemoine, N., & Cudennec, C. (2026). Characterization of hydrometeorological events from historical records in the Levant (200–1850 CE) compared to paleoclimate reconstructions. Regional Environmental Change, 26(4), Article 200. https://doi.org/10.1007/s10113-026-02693-w

Image Credits: AI Generated

DOI: 10.1007/s10113-026-02693-w

Keywords: historical climatology, Levant, Damascus, drought, floods, Palmer Drought Severity Index, Old World Drought Atlas, documentary sources, paleoclimate reconstruction, medieval chronicles, hydrometeorology, Regional Environmental Change