Only a few records come close to meeting these criteria, and most of them suffer from multiple deficiencies, usually because the instruments were shifted to nearby, but different sites. Thus, most records are, in fact, composite series, containing data from a number of sites in one area. One notable exception is the record of the Armagh Observatory in Northern Ireland 1796-2002. The resulting historical climate record is the longest for any single site in the United Kingdom and Ireland (although a gap 1825-33 was filled by data from two stations in Dublin). Only two other European stations have records as long at the Armagh series: Stockholm and Uppsala, both in Sweden.
The central England time-series of air temperature is the longest instrumental record in the world. It puts together data from a number of climate stations in England, namely, Ringway, Malvern, Squires Gate, and Rothampsted, for 1772-present. It was not until 1840 that the first official national archive of climate data was set up, housed at the Royal Greenwich Observatory near London.
A formal system of weather observations did not begin in the United States until 1818. However, climate data have been gleaned from private sources for several years prior to this. Of note is the reconstruc tion by R.D. Erhardt for the southern United States, using data from the private journals of Winthrop Sargent and William Dunbar. Measurements were taken near Natchez, Mississippi, commencing in 1798 and 1799, respectively. The Dunbar record includes daily measurements of air temperature, barometric pressure, wind direction, precipitation, and state of the sky for the hours of 0600, 1500, and 2100 local time 1799-1818. It is regarded as one of the best early instrumental records in the United States.
H.E. Landsberg and others assembled a composite temperature series from 1795 for the eastern United States. The main sources of climate data 1820-70 are archived by the Smithsonian Institute and the U.S. Army Medical Corp. Data. Records became plentiful following the establishment of the U.S. Signal Service in 1870 and the U.S. Weather Service in 1891. L.J. Darter has assembled a comprehensive listing of archived U.S. data.
Globally, the total number of climate stations with records of daily air temperature grew steadily from about 100 in 1880, to a peak of approximately 1,500 in 1970. A decline set in after that, with a sudden loss of over half the stations around 1990, leaving a total of just over 5,000 in 2000. The loss in stations varied around the world. The biggest losses occurred in China, the former Soviet Union, Africa, and South America.
The number of stations used as the global historical climate network for calculation of mean global temperature is reduced by about two-thirds to remove records with insufficient continuity or quality control. The maximum number of acceptable station records over land regions of the globe was just over 3,000 1951-90. Prior to the 1940s, there are problems of accuracy and lack of standardization in the way climate variables are measured, making it difficult to reliably compare measurements made in different places. For marine regions, sea-surface temperature measurements taken on board ships are used. As most of them are made voluntarily, coverage is mainly along main shipping lanes and is sparse in the southern oceans. There has been increased reliance on highly-accurate satellite measurements of global temperature since measurements began in 1979.
It is common to use proxy data to extend the historical record of actual measurements. A common source of proxy data are written records, such as diaries, chronicles, and ships' logs, or accounts of things such as floods, droughts, lake levels, grape harvests, and wheat prices from which climate conditions can be inferred. Some of these records are quite long. For example, there are records of the variation of the level of Lake Victoria in equatorial Africa since 1860, wheat prices in Europe since 1200, the occurrence of sea ice off Iceland since 860 c.E., date of first bloom of cherry tress in Kyoto, Japan, since 812 c.E., and water level in the Nile at Cairo, since 622 c.E. Icono-graphic proxy data is assembled from old paintings and carvings that depict the type of vegetation and the position of glaciers and frozen rivers and lakes that give an indication of climate of the time.
Archaeological remains can provide proxy climate data and evidence of past migrations linked to climatic change. For example, from about 900 c.E., Vikings settled in southern Greenland and farmed soil, that by 1200 c.E., became too cold for agriculture. The remnants of these settlements provide evidence of what agricultural activities took place during this time, known as the Medieval Warm Period, which was brought to an end by the onset of a cold period known as the Little Ice Age.
Tree rings of long-lived trees also provide useful proxy data for extending the historical climate record. Tree rings are wider in wet years for trees growing in semi-arid climates, and wider in warm years for trees growing in cold climates. Reconstruction of past climates prior to the historical record requires the use of the paleoclimatic proxy data from the geological past based on investigations of fossils, sedimentary rocks, and ice and sea cores.
SEE ALSO: Climatic Data, Atmospheric Observations; Climatic Data, Nature of the Data; Cave Records; Climatic Data, Ice Observations; Climatic Data, Instrumental Records; Climatic Data, Climatic Data, Oceanic Observations; Climatic Data, Proxy Records; Climatic Data, Tree Ring Records.
BIBLIOGRApHY. L.J. Darter, List of Climatological Records in the National Archives (National Archives, 1942); R.D. Erhardt, "Reconstructed Annual Minimum Temperatures for the Gulf States, 1799-1988," Journal of Climate (v.3, 1990); H.H. Lamb, Climate, History and the Modern World (Methuen, 1982); H.E. Landsberg, C.S. Yu, and L. Huang, Preliminary Reconstruction of Long Time Series of Climatic Data for the Eastern United States (University of
Maryland Institute of Fluid Dynamics and Applied Mathematics, 1968).
C.R. de Freitas University of Auckland, New Zealand
Was this article helpful?