Ships' Logbooks and “The Year Without a Summer”

Abstract

Weather data extracted from the logbooks of 227 ships of opportunity are used to document the -state of the global climate system in the summer of 1816 (?The Year Without a Summer?). Additional land-based data, some never before used, supplement the marine network. The sources of the data are given and briefly discussed. The main highlights of the global climate system in the 3-year period centered on the summer of 1816 include: ? a cold-phase Southern oscillation (50) (La Niña) event in the Northern Hemisphere (NH) winter of 1815?16, which was preceded and followed by warm-phase SO (El Niño) events in the winters of 1814/15 and 1816/17; ? strong Asian winter and summer monsoons, which featured anomalous cold in much of south and east Asia in the winter of 1815/16 and near-or above-normal rains in much of India in the summer of 1816;. ? below-normal air temperatures (1°?2°C below 1951?80 normals) in parts of the tropical Atlantic and eastern Pacific (in the Galapagos Islands), which imply below-normal sea surface temperatures in the same areas; ? a severe drought in northeast Brazil in 1816?17; ? an active and northward-displaced intertropical zone in most areas from Mexico eastward to Africa; ? generally colder-than-normal extratropical temperature anomalies in both hemispheres; ? an area of anomalous warmth (1°?2°C above 1951?80 normals) in the Atlantic between Greenland and the Azores during at least the spring and summer of 1816; and ? an active Atlantic hurricane season in both 1815 and 1816. A general circulation model simulation of the spatial patterns of high latitude NH temperature anomalies in the first winter following a major volcanic eruption (Graf et al. 1993) is not fully supported by the results in the North American sector where warming in Greenland was observed in 1816, as the GCM indicates cooling. The area of maximum cooling over North America near 50°N, 90°W in 1816 is north of the GCM results. This second difference may be partly attributed to the effects of the cold-phase SO (La Niña) event superimposed on the volcanic signal. Elsewhere in North America, Asia, and Europe, there is generally good agreement between the observed patterns and the GCM results.