Study on the extremely cold winter of 1670 over the middle and lower reaches of the Yangtze River

JunHui Yan, MingQi Li, HaoLong Liu, JingYun Zheng*, Hui Fu

1. Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China

2. College of Urban and Environmental Sciences, Xinyang Normal University, Xinyang, Henan 464000, China

3. College of Geography and Tourism, Chongqing Normal University, Chongqing 400047, China

Study on the extremely cold winter of 1670 over the middle and lower reaches of the Yangtze River

JunHui Yan1,2, MingQi Li1, HaoLong Liu1, JingYun Zheng1*, Hui Fu3

1. Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China

2. College of Urban and Environmental Sciences, Xinyang Normal University, Xinyang, Henan 464000, China

3. College of Geography and Tourism, Chongqing Normal University, Chongqing 400047, China

The snow-cover days over the middle and lower reaches of the Yangtze River (MLRYR) in the winter of 1670 were extracted from Chinese historical documents. By these records, the winter temperature anomalies (compared to the mean of 1961-1990) recorded at seven meteorological stations and the regional mean winter temperature were estimated. The results show that: (1) There was an average of about 30 snow-cover days over the MLRYR region in 1670, ranging from 11-20 days in Shanghai and eastern Zhejiang to 51-60 days in eastern Hunan Province. The snow-cover days averaged about 40 days in Anqing and Nancheng, and ranged from 30 to 40 days in Quzhou, Jingdezhen, and Nanchang; and (2) the regional mean winter temperature in 1670 was estimated to be approximately 4.0 °C lower than that of 1961-1990. The maximum negative anomaly of 5.6 °C occurred in Nanchang and the minimum anomaly of ?2.8 °C was detected in Quzhou. Both of these were lower than that of the coldest winter during the instrumental observation period of 1951-2010. This research could not only provide a method to estimate historical climate extremes, but also provide a background to understand the recent instrumentally climate extremes.

extreme cold winter; snow-cover days; Chinese historical documents; the middle and lower reaches of the Yangtze River; the winter of 1670

1 Introduction

The Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) pointed out that the globally averaged combined land and ocean surface temperature increased by 0.85 °C (0.65 °C to 1.06 °C) over the period of 1880 to 2012 (IPCC, 2013). This warming trend has also been observed in China (SCNAR Editorial Committee, 2011). Against this warming background, there is increasing concern that extreme events may be changing in frequency and intensity as a result of human influences on climate (Trenberthet al., 2007). Recently, climate extremes have become a hot topic in climate change research because the occurrence of climate extremes has more severe effects on nature and human society than do climatic conditions within the normal range of variation (Pfisteret al., 1999; Brázdilet al., 2010). Anomalous cold winters can have disastrous effects on natural environments and human society. For example, the most severe cold winter during the last 30 years occurred in 2008 in southern China. This extremely cold winter, which was characterized by low temperatures, deep snowfall accumulation, and long durations of frozen days, caused direct economic losses of more than 150 billion Yuan (Dinget al., 2008; Wanget al., 2008; Chenet al., 2009). Similar cold winters occurred many times in historical periods (Haoet al.,2011a, 2011b; Zhenget al., 2012, 2014).

During recent decades there have been a growing body of studies using instrumental data to investigate these climate extremes (Zhaiet al., 1999; Meehlet al., 2000); however, due to the limited instrumental observation data, climate extremes in the historical period have not been fully studied. In China, there exist large amounts of historical documents containing abundant records of extreme events, such as snow/ice storms, frozen rivers and lakes, and the durations of frost and snow (Geet al., 2003). Using these records, many studies have qualitatively described severe meteorological disasters and extreme climate events (e.g., Gonget al., 1987; Zhanget al., 1997; Wanget al., 2004), but quantitative reconstructions of temperature anomalies during these extreme events were scarce. In this study, using the temperature measurements and numbers of snow-cover days extracted from Chinese historical documents, we attempted to reconstruct winter temperatures for the middle and lower reaches of the Yangtze River (MLRYR) in 1670. Our study can not only reveal the strength of historical extremely cold winters, but also provide a background to understand more-recent climate extremes.

2 Data sources

Two types of data were used in this study, the snow-cover records from historical documents and modern meteorological data. The modern meteorological data included monthly mean temperature and daily snow-depth data.

2.1 Snow-cover records from historical documents

Snow-cover records for the winter of 1670 were mainly compiled from local gazettes, which are comprehensive books that record both natural and social conditions of administrative units (Geet al., 2008). The government paid much attention to editing local gazettes, predefining a fixed format and setting up a professional department to supervise and check the quality of the gazettes. Therefore, the derived snow-cover days were homogeneous and reliable. In this study, we extracted a total of 77 records about the extremely cold winter of 1670 fromA Compendium of Chinese Meteorological Records of the Last 3,000 years(Zhang, 2004). The original sources of these records (from historical documents saved in 75 libraries and archives from 37 cities across China) had been carefully checked, and the errors of records regarding time, place, and descriptions were corrected as much as possible (Zhang, 2004). Of these 77 stations, data from 7 stations with exact snow-cover days both in 1670 and modern times were used to reconstruct winter temperatures (shown in figure 1).

Figure 1 Study area and locations of sites mentioned in this study

2.2 Modern meteorological data

The modern instrumental monthly mean temperatures from 1961 to 2010 were derived from the Chinese Meteorological Administration (CMA). The means of individual site measurements were calculated to represent the regional mean temperature of the MLRYR region. The daily snow-depth measurements were available in theChina Monthly Meteorological Report(the Chinese Meteorological Administration, 1961-2010). In order to be consistent with historical records, a snow-cover day was defined as a day when half of the observation field was covered with snow, even though the snow depth was less than 0.5 cm (Cuiet al., 2005;Anet al., 2009).

3 Methods

3.1 Estimation of snow-cover days in the winter of 1670

There are three kinds of snow-cover descriptions in historical documents:

? Directly recorded snow-cover days. For example, "It snowed heavily in December of 1670, with snow cover on the ground for more than 40 days. The rivers were frozen and many people were frozen to death" (quoted from gazettes of the Fuzhou Prefecture). There were a total of 29 such records (accounting for 37.7%). Using these records, we could directly extract the snow-cover days.

? Recorded start and end dates of snow cover. For example, "It snowed heavily in winter. The heavy snow started on December 10 and continued until February 3 of the next year…" (quoted from gazettes of Xiangtan County). There were only 14 such records (accounting for 18.2%). Based on these records, we could calculate the snow-cover days.

? General descriptions of snow-cover events. For example, "The ground surface was covered by 4?5 chi (about 132?165 cm) of snow after continuous heavy snowfall. Houses were buried and many people froze to death on the road" (quoted from gazettes of Yifeng County). There were 34 such records (accounting for 44.2%). Although we could not obtain exact snow-cover days from these records, we could estimate the general situation of snowfall and snow cover.

3.2 Relationship between the winter temperature and snow-cover days at seven stations

There are two kinds of snow in the world, permanent snow and seasonal snow. Permanent snow is mainly distributed in high-altitude and high-latitude areas, and the snow-cover days of permanent snow have little inter-annual variability. The snow in the MLRYR region, which is located in middle latitudes, is seasonal snow; it mostly occurs in winter and the snow-cover days have large inter-annual variability and can be closely related to temperature change (Panget al., 2006; Qinet al., 2006). In this study the inter-annual winter temperatures correlated significantly with the snow-cover days (Table 1,n= 50). The correlation coefficients ranged between 0.52 and 0.72 for the seven stations. The regression coefficients ranged from ?0.194 in Nanchang to ?0.115 in Anqing, which suggests that the winter temperatures decreased by 1.15-1.94 °C if snow-cover days increased by 10 days.

Table 1 Linear regression relationship between winter temperature and snow-cover days in the MLRYR region

3.3 Regional temperature reconstruction

To reconstruct the regional mean temperature of the MLRYR region in 1670, we first performed a regression analysis between winter temperature and the number of snow-cover days using modern instrumental data at the seven stations where the snow-cover data were available for 1961-2010. Subsequently, winter temperature anomalies at those seven stations were estimated based on the derived regression equation. Finally, applying a stepwise regression model between the regional winter temperature and the mean winter temperature of those seven stations, we could calculate the regional mean winter temperature in 1670 over the MLRYR region. Through stepwise regression, there were only four representative sites retained. As shown by equation(1), the four sites explained 99.6% of the variance of the regional mean winter temperature:

4 Results

4.1 Spatial pattern of snow-cover days

In the winter of 1670, heavy snow occurred in the MLRYR region under the influence of a cold wave. Rain and snow froze on the ground surface for longperiods due to persistent low-temperature days and continuous heavy snowfalls. Figure 2 shows the spatial pattern of snow-cover days across the MLRYR region in the winter of 1670. The regional mean number of snow-cover days was about 30 days, with large spatial variability. Generally, there were more snow-cover days in the western part of the study area than in the east. The maximum snow-cover days (about 70) was found in Zixing County, Hunan Province. As recorded in historical documents, the heavy snow in Zixing County started at the end of December of the Chinese lunar calendar and continued until the third lunar month of the next year. Secondarily, 51-60 days of heavy snow-cover were found in southeastern Hubei and western Hunan. For instance, in Puqi County, Hubei Province, ice and snow froze on the ground surface for more than 50 days (quoted from gazettes of Puqi County). The snow-cover days in Anqing, Jingdezhen, Nanchang, Nancheng, and Quzhou ranged from 30 to 50 days. The fewest snow-cover days (less than 20) were found in Shanghai and eastern Zhejiang Province.

4.2 Estimates of winter temperatures for the seven stations

We reconstructed the winter temperatures for each station using the above regression models. For example, the gazettes of Linchuan County stated "Heavy rain and snow occurred in winter at Linchuan County, with snow covering the ground for 40 days. Rivers were frozen and many homeless people were frozen to death." Linchuan is very close to Nancheng, so, based on the relationship between winter temperature and snow-cover days in Nancheng (Table 1), the winter temperature of Nancheng was estimated to be 5.19 °C lower than that of 1961-1990. Table 2 shows the reconstructed winter temperature anomalies for the seven stations in the winter of 1670. We found that the winter of 1670 was extremely cold, with temperature anomalies ranging from ?6 °C to ?3 °C with respect to the 1961-1990 mean. The maximum winter temperature anomaly of ?5.6 °C was found in Nanchang, Jiangxi Province. The next lowest were Changsha and Nancheng, with winter temperature anomalies of?5.26 °C and ?5.19 °C, respectively. In Wuhan and Jingdezhen, the winter temperature anomalies ranged between ?5 °C and ?4 °C. In comparison, the winter temperature in Anqing and Quzhou showed smaller negative anomalies ranging from ?4 °C to ?3 °C, but they were still lower than the coldest value during the modern observation period.

4.3 Regional mean winter temperature in 1670

The regional mean winter temperature correlated significantly with the modern individual site-based temperatures; the correlation coefficients were more than 0.90, corresponding to explaining the variance of more than 80% (P<0.001). Thus, it was feasible to estimate the regional mean temperature using modern site-based temperatures. Since the winter temperature anomalies in Anqing, Nancheng, Wuhan, and Quzhou were ?3.42, ?5.19, ?4.81, and ?2.79 °C, respectively, the regional winter temperature anomaly in 1670 was estimated to be 4.0 °C lower than the 1961-1990 mean. In the modern instrumental period from 1951 to 2010, 1967 had the coldest winter temperature with an anomaly of ?1.94 °C. Obviously, the regional mean winter temperature in 1670 was much lower than the coldest value in the modern instrumental period.

Figure 2 Snow-cover days for the middle and lower reaches of the Yangtze River in 1670

Table 2 Reconstructed winter temperatures of 1670 in the MLRYR region

5 Conclusions and discussion

In this study, we collected 77 records about snow-cover days from historical documents from the MLRYR region. Using these records as well as modern meteorological observations, we analyzed the spatial characteristics of snow-cover days and the degree of cold in the winter of 1670 over the MLRYR region. Our main conclusions are:

? In the winter of 1670, there was an average of 30 snow-cover days in the MLRYR region, ranging from 11-20 days in Shanghai and eastern Zhejiang to 51-60 days in eastern Hunan Province. The maximum snow-cover days (about 70) was found in Zixing County, Hunan Province, while snow-cover days in Anqing, Jingdezhen, Nanchang, Nancheng, and Quzhou ranged from 30 to 50 days. The fewest snow-cover days (less than 20) were found in Shanghai and eastern Zhejiang Province.

? The winter of 1670 was extremely cold, and the regional winter temperature was about 4.0 °C lower than the 1961-1990 mean. The maximum (coldest) winter temperature anomaly of ?5.6 °C occurred in Nanchang and the minimum (warmest) winter temperature anomaly of ?2.8 °C occurred in Quzhou. The coldness of the winter in 1670 was much more severe than that in any other years of the modern instrumental period (1951-2010).

It should be noted that snow-cover days at some stations were estimated using the start and end dates of snow-cover rather than the direct records of snow-cover days. Actually, there may be several snow-free days from the beginning to the end date of snow-cover. Therefore, to address these uncertainties, we used the daily snow depth over the MLRYR region and we selected the years with more than 10 total snow-cover days. We then calculated the proportion of snow-cover days in the snow-cover duration (days from the beginning to the end date of snow-cover) after removing the outliers. We found that, on average, the snow-cover days accounted for 65.2%, 79.4%, and 93.3% of snow-cover duration in the years with more than 10, 20, and 25 days of snow-cover, respectively. Given that there were as many as 30 snow-cover days at most of the stations over the MYRYR in 1670, there would have been very few snow-free days during the snow-cover duration.

Acknowledgments:

We would like to thank the anonymous reviewers for their constructive comments and suggestions that helped to improve the quality of this study. This research was supported by grants to the Institute of Geographic Sciences and Natural Resources Research (IGSNRR) from the Chinese Academy of Sciences (No. XDA05080100), the Ministry of Science and Technology of the People’s Republic of China (No. 2010CB950101), the Basic Research Project of the Ministry of Science and Technology (No. 2011FY120300), and the National Natural Science Foundation of China (Nos. 41271124, 41071029).

An D, Li DL, Yuan Y,et al., 2009. Contrast between snow cover data of different definitions. Journal of Glaciology and Geocryology, 31(6): 1019-1027.

Brázdil R, Demaree GR, Deutsch M,et al., 2010. European floods during the winter 1783?1784: Scenarios of an extreme event during the "Little Ice Age." Theoretical and Applied Climatology, 100(1-2): 163-189. DOI: 10.1007/s00704-009-0170-5.

Chen HB, Fan XH, 2009. Some extreme events of weather, climate and related phenomena in 2008. Climatic and Environmental Research, 14(3): 329-340.

Cui CX, Yang Q, Wang SL, 2005. Comparison analysis of the long-term variations of snow cover between mountain and plain areas in Xinjiang Region from 1960 to 2003. Journal of Glaciology and Geocryology, 27(4): 486-490.

Ding YH, Wang ZY, Song YF,et al., 2008. Causes of the unprecedented freezing disaster in January 2008 and its possible association with the global warming. Acta Meteorologica Sinica, 66(5): 808-825.

Editorial Committee for Second China National Assessment Report on Climate Change (SCNAR Editorial Committee), 2011. Second China National Assessment Report on Climate Change, Science Press, Beijing, China, pp. 38-42.

Ge QS, Zheng JY, Fang XQ,et al., 2003. Winter half-year tem-perature reconstruction for the middle and lower reaches of the Yellow River and Yangtze River, China, during the past 2,000 years. The Holocene, 13(6): 933-940. DOI: 10.1191/0959683603hl680rr.

Ge QS, Zheng JY, Tian YY,et al., 2008. Coherence of climatic reconstruction from historical documents in China by different studies. International Journal of Climatology, 28(8): 1007?1024. DOI: 10.1002/joc.1552.

Gong GF, Zhang PY, Zhang JR, 1987. Chilly winter of 1892?1893 and its effect. Collected Papers of Geography, 18: 129?138.

Hao ZX, Zheng JY, Ge QS,et al., 2011a. Historical analogues of the 2008 extreme snow event over central and southern China. Climate Research, 50(2-3): 161?170. DOI: 10.3354/cr01052.

Hao ZX, Zheng JY, Ge QS,et al., 2011b. Variations of extreme cold winter events in southern China in the past 400 years. Acta Geographica Sinica, 66(11): 1479-1485.

Intergovernmental Panel on Climate Change (IPCC), 2013. Summary for policymakers. In: Stocker TF, Qin D, Plattner GK,et al.(eds.). Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, UK, pp. 1?10.

Meehl GA, Karl T, Easterling DR,et al., 2000. An introduction to trends in extreme weather and climate events: Observations, socioeconomic impacts, terrestrial ecological impacts, and model projections. Bulletin of the American Meteorological Society, 81(3): 413-416. DOI: 10.1175/1520-0477(2000)081< 0413:AITTIE>2.3.CO;2.

Pang HX, He YQ, Lu AG,et al., 2006. Influence of Eurasian snow cover in spring on Indian Ocean dipole. Scientia Geographica Sinica, 26(6): 682-686.

Pfister C, Brázdil R, 1999. Climatic variability in sixteenth century Europe and its social dimension: A synthesis. Climatic Change, 43(1): 5?53. DOI: 10.1023/A:1005585931899.

Qin DH, Xiao CD, Ding YJ,et al., 2006. Progress on cryospheric studies by international and Chinese communities and perspectives. Journal of Applied Meteorological Science, 17(6): 649-656.

The Chinese Meteorological Administration, 1961-2010. The China Monthly Meteorological Report. China Meteorological Press, Beijing, China.

Trenberth K, Jones P, Ambenje P,et al., 2007. Observations: Surface and atmospheric climate change. In: Solomon S, Qin DH, Manning M,et al. (eds.). Climate Change 2007: The Physical Science Basis. Cambridge University Press, UK, pp. 300-301.

Wang PH, Fang XQ, He LX, 2004. Historical records on cold events and their influence during 1328-1330 AD in China. Journal of Palaeogeography, 6(4): 480-484.

Wang ZY, Zhang Q, Chen Y, 2008. Characters of meteorological disasters caused by the extreme synoptic process in early 2008 over China. Advances in Climate Change Research, 4(2): 63-67.

Zhai PM, Sun AJ, Ren FM,et al., 1999. Chances of climate extremes in China. Climatic Change, 42(1): 203?218. DOI: 10.1023/A:1005428602279.

Zhang DE, 1997. A case study of the reality recovery on the historical severe winter and cool summer in China. Chinese Science Abstracts: Technical Bulletin, 3(4): 487?488.

Zhang DE, 2004. A Compendium of Chinese Meteorological Records of the Last 3,000 years. FengHuang Press, Nanjing, China, pp. 1847?1858.

Zheng JY, Ding LL, Hao ZX,et al., 2012. Extreme cold winter events in southern China during AD 1650?2000. Boreas, 41(1): 1?12. DOI: 10.1111/j.1502-3885.2011.00225.x.

Zheng JY, Hao ZX, Fang XQ,et al., 2014. Changing characteristics of extreme climate events during past 2,000 years in China. Progress in Geography, 33(1): 3-12.

: Yan JH, Li MQ, Liu HL,et al., 2014. Study on the extremely cold winter of 1670 over the middle and lower reaches of the Yangtze River. Sciences in Cold and Arid Regions, 6(6): 0540-0545.

10.3724/SP.J.1226.2014.00540.

Received: March 20, 2014 Accepted: June 30, 2014

*Correspondence to: JingYun Zheng, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. 11A Datun Road, Chaoyang District, Beijing 100101, China. E-mail: zhengjy@igsnrr.ac.cn