East-Central China Winter Half-year Temperature Reconstruction ----------------------------------------------------------------------- World Data Center for Paleoclimatology, Boulder and NOAA Paleoclimatology Program ----------------------------------------------------------------------- NOTE: PLEASE CITE CONTRIBUTORS WHEN USING THIS DATA!!!!! NAME OF DATA SET: East-Central China Winter Half-year Temperature Reconstruction LAST UPDATE: 12/2003 (Original Receipt by WDC Paleo) CONTRIBUTORS: Zheng Jingyun and Ge Quansheng, Institute of Geographic Sciences and Natural Resources Research (IGSNRR), Chinese Academy of Sciences IGBP PAGES/WDCA CONTRIBUTION SERIES NUMBER: 2003-090 SUGGESTED DATA CITATION: Ge, Q., et al., 2003, East-Central China Winter Half-year Temperature Reconstruction, IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series # 2003-090. NOAA/NGDC Paleoclimatology Program, Boulder CO, USA. ORIGINAL REFERENCE: Ge, Q., J. Zheng, X. Fang, Z. Man, X. Zhang, P. Zhang, and W.-C. Wang. 2003. Winter half-year temperature reconstruction for the middle and lower reaches of the Yellow River and Yangtze River, China, during the past 2000 years. The Holocene 13,6 pp. 933-940. GEOGRAPHIC REGION: East-Central China PERIOD OF RECORD: 0-2000 AD DESCRIPTION: Winter half-year temperature reconstruction at 10 or 30 year resolution for the region between the middle and lower reaches of the Yellow and Yangtze Rivers. Reconstruction is based on phenological events (plant budding or flowering dates) from historical and documentary records, supplemented by winter snow day records from historical documents for some locations. ABSTRACT: Phenological cold/warm events recorded in Chinese historical documents are used to reconstruct, at 10-30 years' resolution, winter half-year (October to April) temperatures for the past 2000 years in the central region of eastern China. Because of the uneven spatial and temporal distribution of the phenological records, the reconstruction of the regional mean temperature involves two steps: reconstruction for individual sites within the region and calculation of the regional mean. For a single site, the reconstruction involves: identifying the difference in dates in phenological events for both historical and modern records; establishing the conversion function between the date difference and temperature change from the modern records; and converting the historical records into temperature variation. The spatial representativeness of the individual sites is studied by examining the correlation between individual sites and regional mean temperature from modern instrumental data. The correlation is then used as the basis for constructing the regional mean winter half-year temperature for the past 2000 years. From the beginning of the Christian era, climate became cooler at a rate of 0.17°C per century, and around the AD 490s temperature reached about 1°C lower than that of the present (the 1951- 80 mean). Then, abruptly, temperature entered a warm epoch from the AD 570s to 1310s with a warming trend of 0.04°C per century; the peak warming was about 0.3-0.6°C higher than present for 30-year periods, but over 0.9°C warmer on a 10-year basis. After the AD 1310s, temperature decreased rapidly at a rate of 0.10°C per century; the mean temperatures of the four cold troughs were 0.6-0.9°C lower than the present, with the coldest value 1.1°C lower. Temperature has been rising rapidly during the twentieth century, especially for the period 1981-99, and the mean temperature is now 0.5°C higher than for 1951-80. The most interesting aspect over the past 2000 years has been the rapid transitions between cold and warm periods. DATA: 1. 30-year resolution reconstructed temperature anomalies AD 15 - AD 1995. Column 1: Time Period, midpoint of 30-year interval, year AD Column 2: Reconstructed Temperature Anomaly, Degrees C. Column 3: Three point moving average of reconstructed temperature anomaly Column 4: Maximum estimated error range (degrees C) Period(AD) T.Anom. 3 pt. Avg. Error 15 0.2 - 1 45 -0.2 0.03333333 1.1 75 0.1 0.1 1 105 0.4 0.266667 1 135 0.3 0.3 1.1 165 0.2 0.166667 0.8 195 0 -0.03333 0.7 225 -0.3 -0.1 0.9 255 0 -0.33333 1.1 285 -0.7 -0.4 0.9 315 -0.5 -0.6666667 1.2 345 -0.8 -0.4333333 1.2 375 0 -0.333333 0.9 405 -0.2 -0.166667 0.9 435 -0.3 -0.4 0.9 465 -0.7 -0.66667 0.7 495 -1 -0.83333 1.2 525 -0.8 -0.7 1.1 555 -0.3 -0.26667 0.7 585 0.3 0 0.9 615 0 0.1666667 0.7 645 0.2 0.1666667 1.1 675 0.3 0.3333333 1.1 705 0.5 0.3666667 1.1 735 0.3 0.2666667 1.2 765 0 -0.1 0.9 795 -0.6 -0.4333333 0.9 825 -0.7 -0.4666667 1 855 -0.1 -0.4666667 0.7 885 -0.6 -0.4 1.2 915 -0.5 -0.3666667 1.2 945 0 -0.06666667 0.9 975 0.3 0.1333333 1.2 1005 0.1 0.2333333 1 1035 0.3 0.2666667 1.2 1065 0.4 0.4 1.3 1095 0.5 0.2 1.2 1125 -0.3 -0.1 0.9 1155 -0.5 -0.3333333 0.7 1185 -0.2 -0.2 0.7 1215 0.1 0.2666667 0.8 1245 0.9 0.5666667 1 1275 0.7 0.5333334 1 1305 0 0.03333333 1 1335 -0.6 -0.3666667 0.9 1365 -0.5 -0.3333333 0.9 1395 0.1 -0.3333333 0.9 1425 -0.6 -0.4 0.9 1455 -0.7 -0.6 1 1485 -0.5 -0.3666667 1.1 1515 0.1 -0.1333333 1 1545 0 -0.06666667 1.2 1575 -0.3 -0.18667 1.3 1605 -0.26 -0.22 1.3 1635 -0.1 -0.4866 1.2 1665 -1.1 -0.4866 1.4 1695 -0.2 -0.4333333 1.1 1725 0 -0.03333333 0.9 1755 0.1 -0.1 1.1 1785 -0.4 -0.3666667 1.2 1815 -0.8 -0.6333333 1.3 1845 -0.7 -0.8 1.3 1875 -0.9 -0.7 1.3 1905 -0.5 -0.4333333 1.2 1935 0.1 -0.1333333 1.1 1965 0 0.2 1995 0.5 - 1999 2. 10-year resolution reconstructed temperature anomalies AD 961 - AD 1110. Column 1: Time Period, midpoint of 10-year interval, year AD Column 2: Reconstructed Temperature Anomaly, Degrees C. Column 3: Three point moving average of reconstructed temperature anomaly Column 4: Maximum estimated error range (degrees C) Period(AD) T.Anom. 3 pt. Avg. Error 965 0.6 - 1.2 975 0.3 0.2666667 1.1 985 -0.1 0.2666667 1 995 0.6 0.1666667 1.2 1005 0 0.1 1 1015 -0.3 0.03333333 1.1 1025 0.4 0.1333333 1.1 1035 0.3 0.3 1.1 1045 0.2 0.2 1.1 1055 0.1 0.4333333 1 1065 1 0.4333333 1.3 1075 0.2 0.5666667 1 1085 0.5 0.3666667 1.2 1095 0.4 0.4666667 1.2 1105 0.5 - 1.2 3. 10-year resolution reconstructed temperature anomalies AD 1501 - AD 1999. Column 1: Time Period, midpoint of 10-year interval, year AD Column 2: Reconstructed Temperature Anomaly, Degrees C. Column 3: Three point moving average of reconstructed temperature anomaly Column 4: Maximum estimated error range (degrees C) Period(AD) T.Anom. 3 pt. Avg. Error 1505 0 - 0.9 1515 0.2 0.1333333 1.1 1525 0.2 0.03333333 1.1 1535 -0.3 0.06666667 1.2 1545 0.3 0 1.2 1555 0 -0.03333333 0.9 1565 -0.4 -0.2 1.3 1575 -0.2 -0.2666667 1.1 1585 -0.2 -0.2333333 1.1 1595 -0.3 -0.2 1.2 1605 -0.1 -0.2666667 1 1615 -0.4 -0.266667 1.3 1625 -0.3 -0.26667 1.2 1635 -0.1 -0.1333333 1 1645 0 -0.4666667 0.9 1655 -1.3 -0.8666667 1.5 1665 -1.3 -1.1 1.5 1675 -0.7 -0.9666666 1.4 1685 -0.9 -0.5666667 1.3 1695 -0.1 -0.2333333 1.2 1705 0.3 0.06666667 1.2 1715 0 0.03333333 0.9 1725 -0.2 -0.02266667 1.1 1735 0.1 -0.01266667 0.7 1745 0 0.03 0.6 1755 -0.1 0.096 0.6 1765 0.3 0.122 0.7 1775 0.1 0.09 0.6 1785 -0.2 -0.384 0.7 1795 -1.1 -0.644 1.1 1805 -0.7 -0.876 0.8 1815 -0.9 -0.808 0.9 1825 -0.9 -0.84 0.9 1835 -0.8 -0.858 0.9 1845 -0.9 -0.658 0.9 1855 -0.3 -0.424 0.7 1865 -0.1 -0.574 0.6 1875 -1.4 -0.914 1 1885 -1.3 -1.228 1 1895 -1 -0.928 0.9 1905 -0.5 -0.482 0.8 1915 0 -0.152 0.6 1925 0 -0.032 0.6 1935 -0.1 0.102 0.6 1945 0.4 0.07453175 0.7 1955 -0.1 0.09030159 1965 -0.1 9.25186E-18 1975 0.2 0.09664286 1985 0.2 0.4056455 1995 0.9 -