Quantitative Analysis of Climate Change and Human Crises in History - Space-Time Integration in Geography and GIScience - page 235

Our results reveal that high war frequencies echoed more stringent resource
conditions. For instance, the correlation coefficients for total wars, rebellions, and
Central wars became stronger from the annual scale to phase scale (Table 14.1 ),
because prolonged cooling exhausted the stored livelihood resources and eventually
brought wars. In humid tropical and sub-tropical South China, the influence of
cooling in cold phases on agricultural production might have a subdued effect on
resource reduction because of a rich endowment in heat and moisture in the coastal
region. Even if cooling were severe enough to affect cropping, the more flexible
farming system in the South, with a wide range of domesticated species, could
adopt alternative crops. Therefore, human reaction to cooling in South China was
not so sensitive and severe. In contrast, the climate in Central China is controlled by
the monsoons: cold-dry air masses move in from Siberia in the winter and warm-
humid marine air masses come in from the southeast and southwest in the summer.
It should be noted that many studies of China's paleo-climate indicated that the cold
periods were dominated by winter monsoons from Siberia and hence were drier than
warm periods (An 2000 ;Lietal. 2000 ). Thus, cold periods could have imposed
a double jeopardy in terms of coldness and dryness, bringing a highly stressed
condition for agriculture in Central China. North China, where the main sustenance
was grazing, was sensitive to cooling which could reduce short-term production
and trigger long-term loss of ecosystem productivity due to land degradation and
desertification. In addition, unlike crop produce, pastoral animal resources could
not be stored for a long time. Thus in the north, the onset of a cold period would
soon be followed by wars. It should be noted that although most of the war peaks
occurred in cold phases, the correlation between temperature and wars was not
significant. At the time when the whole of China or part of Central China was
ruled by northern nomadic tribes (South Song, Yuan, and Qing Dynasties), a period
which covered more than 500 years and over half of the study period, people in
the north could freely shift to the south or acquire their livelihood from the south;
thus, the war frequency in North China was reduced in the cold phases (respectively
in the thirteenth, fourteenth, seventeenth, and nineteenth centuries). At the annual
scale correlation analysis, if we disregarded the occupation years, the correlation
between temperature and wars in North China (i.e., r D 0.208, P < 0.01, not shown
in Table 14.1 ) was more significant than those in other areas (cf. Table 14.1 ).
We extended our exploration of the climate-war relationship from China to
different geographic regions in the Northern Hemisphere (Zhang et al. 2007a ). We
found that in the same manner as the Northern Hemisphere temperature variations 6
(Fig. 14.2 a), the incidence of warfare in the Northern Hemisphere, Europe, Asia, and
the arid areas of the Northern Hemisphere (i.e., the arid zone from Eurasia to North
Africa) in AD 1400-1900 7

occurred in a cyclic pattern, with a turbulent period

6 Our temperature data come from Mann and Jones' ( 2003 ) Northern Hemisphere annual tempera-
ture trajectory, supplemented by the Southern Hemisphere and global temperature trajectories.
7 Our war data was elicited from Brecke's ( 1999 ) Conflict Catalog , which is the most inclusive
global war dataset so far, documenting a total of 2,912 wars fought in AD 1400-1900. The