Quantitative Analysis of Climate Change and Human Crises in History - Space-Time Integration in Geography and GIScience - page 255

Table 14.7 Granger causality analysis for each of the linkages as shown in Fig. 14.7
Causal linkage (null hypothesis) F P
Climate change does not Granger-cause bio-productivity 207.485 0.000***
Bio-productivity does not Granger-cause agricultural production 7.440 0.007**
Agricultural production does not Granger-cause food supply per capita 9.834 0.002**
War does not Granger-cause population 391.805 0.000***
Epidemics do not Granger-cause population 103.054 0.000***
Famine does not Granger-cause population 155.736 0.000***
Population does not Granger-cause agricultural production 5.731 0.017*
Population does not Granger-cause food supply per capita 67.664 0.000***
Food supply per capita does not Granger-cause famine 10.307 0.000***
Famine does not Granger-cause nutritional status 2.139 0.009**
Nutritional status does not Granger-cause epidemics 2.345 0.004**
Food supply per capita does not Granger-cause social disturbance 1.971 0.024*
Social disturbance does not Granger-cause war 3.256 0.000***
Social disturbance does not Granger-cause migration 1.786 0.037*
War does not Granger-cause migration 2.250 0.006**
Food supply per capita does not Granger-cause migration 2.164 0.008**
Migration does not Granger-cause epidemics 1.835 0.031*
*Significant at 0.05 level (2-tailed) ( P < 0.05); **Significant at 0.01 level (2-tailed) ( P < 0.01);
***Significant at 0.001 level (2-tailed) ( P < 0.001)
past few decades only (i.e., especially AD 1950 and forward) or even shorter time
periods (AD 1980 and forward) (Burke et al. 2010 ;Hsiangetal. 2011 ; Hendrix and
Glaser 2007 ; Hendrix and Salehyan 2012 ; Jensen and Gleditsch 2009 ; Meier et al.
2007 ;Migueletal. 2004 ; Raleigh and Kniveton 2012 ; Koubi et al. 2012 ;Theisen
2012 ), their results for the climate-crisis relationship are diverse. Indeed, in those
studies, there is some confusion whether it is the effect of 'climate' or 'weather'
that is being tested, because climate is usually defined as 'average weather' over
a 30-year period (see Gleditsch 2012 ). It should be emphasized that short-term
variations in climate may not have a significant effect on human crises, but the
long-term variations do, because worsening 'climate' conditions contribute to a
gradual build-up of tension. Crises occur only if the tension exceeds the buffering
capacity of human societies (in terms of migration, economic change, innovation,
trade, peaceful resource redistribution, and so on) for a significantly long period
(Zhang et al. 2007a , 2011a ).
On the other hand, there are also some recent studies based on selective individ-
ual cases that question the role of climate change in triggering human crises. For
instance, Butzer ( 2012 ) argues that between the fourteenth and eighteenth centuries,
Western Europe responded to extreme climatic perturbations by innovation and
intensification; such modernization was decentralized, protracted, flexible, and
broadly based. Under extreme stress, people try new solutions. After overcoming
initial, ideological dissonance, people can indeed come together to support change.