Quantitative Analysis of Climate Change and Human Crises in History - Space-Time Integration in Geography and GIScience - page 242

Table 14.4 Pearson's correlation coefficients between temperature anomaly and population
growth rate in different geographic regions in the world in the Little Ice Age. All data have been
smoothed by 40-year Butterworth low-pass filter prior to statistical analysis
Region AD 1400-1900 AD 1500-1900 AD 1400-1800 AD 1500-1800

Worl d 0.387*** 0.518*** 0.584*** 0.754***

N. Hemisphere 0.355*** 0.528*** 0.523*** 0.765***

Asia 0.434*** 0.593*** 0.550*** 0.751***

Europe 0.151*** 0.362*** 0.228*** 0.587***

North America 0.230*** 0.326*** 0.428*** 0.565***

China 0.388*** 0.574*** 0.471*** 0.718***

***Significant at 0.001 level (2-tailed) ( P < 0.001)

in AD 800-1900 19 (Fig. 14.4 )(Zhangetal. 2011b ). There were 88 population
collapses at the country/regional scale, of which 38 were caused by wars, 33
by epidemics, 6 by famines, 5 by war-epidemics synthesis, 4 by cross-oceanic
colonization, 1 by famine-migration synthesis, and 1 by famine-war synthesis. There
were 7, 70, and 11 (the average number of population collapses per century was 4,
16, and 2) in the Medieval Warm Period, cold phases (C1-C4), and mild phases,
respectively. Most of the collapses happened in C1 (41) and C3 (23) - the two
long cold phases. Nearly 70 % of the collapses clustered in AD 1040-1050 (7),
AD 1340-1350 (36), and AD 1592-1620 (17) - although the three short climate-
deteriorating periods spanned less than 2 % of the study period (Fig. 14.4 b, grey
bars). In contrast, the mild phases spanned over 50 % of the study period, but the
associated number of population collapse incidents was only 11. Furthermore, the
average population growth rate and temperature in the Northern Hemisphere moved
in the same direction. The long cold phases were associated with population decline;
after-cooling mild phases were associated with rapid population growth, while warm
phases were associated with diversified population growth rates in different climatic
zones 20 (Fig. 14.4 b).

19 See footnote no. 12 for centennial climate variability in the Northern Hemisphere from AD 800 to
1900. Based on the temperature series, we established warming and cooling thresholds according to
the averaged temperature anomalies of the starting century of the Medieval Warm Period and Little
Ice Age, respectively. A period in which the temperature anomaly was > 0.3 ı C was classified as

a warm phase; a period in which the temperature anomaly was < 0.42 ı C was classified as a cold

phase. It should be noted that the average AD 1961-1990 temperature anomaly is 0 ı C, which is
c. 0.3 ı C higher than that of the Medieval Warm Period. In line with these criteria, we identified
a warm phase in the Medieval Warm Period (AD 954-1114) and four cold phases in the Little Ice
Age [C1 (AD 1236-1359), C2 (AD 1459-1510), C3 (AD 1554-1741), C4 (AD 1804-66)]. These
represent periods of climate deterioration. The remaining years were classified as mild phases
(Fig. 14.4 a).
20 The delineation of major climatic zones in the Northern Hemisphere is made according to the
modified Köppen classification system (see London Times 2007 ). Four climate zones are delineated
as follows: (1) tropical humid - rainy climate with no winter, coolest month above 18 ı C; (2)
warmer humid - rainy climate with mild winter, coolest month above 0 ı C, but below 18 ı C,