July 22, 2010

Not Warming-Up to War?

Filed under: Climate Politics

Literally thousands of websites pound home the idea that global warming is a threat to our national security and that violent conflicts will result from disruptions caused by climate change. Many of the websites point to a study released several years ago by the CNA Corporation which is a nonprofit institution that conducts in-depth, independent research on complex public interest challenges. Their study entitled “National Security and the Threat of Climate Change” was prepared with 11 retired generals and admirals, and it is widely quoted by those insisting global warming will increase the threat of war. The executive summary of the report states “Projected climate change poses a serious threat to America’s national security. The predicted effects of climate change over the coming decades include extreme weather events, drought, flooding, sea level rise, retreating glaciers, habitat shifts, and the increased spread of life-threatening diseases. These conditions have the potential to disrupt our way of life and to force changes in the way we keep ourselves safe and secure.”

The executive summary also states “Climate change acts as a threat multiplier for instability in some of the most volatile regions of the world. Projected climate change will seriously exacerbate already marginal living standards in many Asian, African, and Middle Eastern nations, causing widespread political instability and the likelihood of failed states.” And at home they claim “Projected climate change will add to tensions even in stable regions of the world. The U.S. and Europe may experience mounting pressure to accept large numbers of immigrant and refugee populations as drought increases and food production declines in Latin America and Africa.”

Before you enlist in the military or start shining up combat boots, there is a recent article in the journal Climatic Change that might change your mind about global warming and war. The research was conducted by Richard Tol and Sebastian Wagner from The Netherlands and Germany, respectively. The last sentence of their abstract caught our attention as they conclude “it appears that global warming would not lead to an increase in violent conflict” in mid-latitude locations such as China or Europe. We don’t see this study getting a lot of press coverage, so we decided to feature it on World Climate Reportjust as we did an earlier study which contradicted the global warming=more war claims.

Tol and Wagner begin noting “In the gloomier scenarios of climate change, violent conflict plays a key part. War would break out over declining water resources, and millions of refugees would cause mayhem. The Nobel Peace Prize of 2007 was partly awarded to the IPCC and Al Gore for their contribution to slowing climate change and thus preventing war. Scenarios of climate-change-induced violence can be painted with abandon, because there is in fact very little research to either support or refute such claims.”

Are these two doubting Al Gore by suggesting there is little research to support any claim that global warming will exacerbate violent conflict? This could get interesting!

Furthermore, they reviewed a paper published a few years ago in Climatic Change in which a research team from China examined the “warmer equals more war” hypothesis. Regarding that study, Tol and Wagner state “They construct a dataset of climate and violent conflict for China for the last millennium, and show that the Chinese are more inclined to fight each other when it is cold.”

Tol and Wagner assembled the data for Figure 1 showing a time series of conflict in Europe back about 1,000 years. To compare with temperature and precipitation, they assembled reconstructed values for Europe available back to 1500 AD. These are gridded data that come from meteorological observations as well as proxy information found in Europe – the climate data had been quality checked for inconsistencies. They even assembled climate model simulation data from Europe based on solar and volcanic forcing as well as greenhouse gas concentrations.

Figure 1. Annual number of violent conflicts in Europe according to www.warscholar.com (from Tol and Wagner, 2010)

The map below (Figure 2) tells us what we want to know – it shows the correlation coefficient between annual temperatures and the overall state of violent conflict in Europe, and all the blue indicates the coefficients are negative, indicating more war in cold periods, not hot ones. And if your eye has been attracted to the red regions where warmer weather seems to produce more conflict, be aware that Tol and Wagner warn “positive correlations are evident over the Balkans. These correlations should however not be overinterpreted, because the Balkans are largely excluded from the violent conflict data base.”

The two authors recognized some statistical issues dealing with the violent conflict time series, most notably the high level of autoregression in the data (the value of any year is highly related to the value the previous years). They addressed this unwanted property a variety of different ways, and in each case, they continued to find an overall negative association between temperature and conflict. With respect to the model generated climate data they report “Correlations between the simulated temperatures and European wars also show negative correlations, consistent with results obtained for reconstructions based on observational data and proxy data.”

Figure 2. Correlation map between war conflicts and temperature during the period 1500–1900. The “HW 10” indicates that a 10-year smoothing was applied to the data (from Tol and Wagner, 2010)

Obviously, Europe changed over the 1500-1900 time period, and indeed, Tol and Wagner observed “that the correlations are stronger in the more distant past. This confirms the agricultural hypothesis. Agriculture became progressively less important over the period, because of economic development, and agriculture became less dependent on the weather, because of improved cultivation methods and better fertilizers.” Fair enough.

In their conclusions, the authors state “We present some evidence that periods with lower temperatures in the pre-industrial era are accompanied by violent conflicts”, consistent with what others had found in China. Furthermore, they note “If anything, lower temperatures imply violence, and this effect is much weaker in the modern world than it was in pre-industrial times. This implies that future global warming is not likely to lead to (civil) war between (within) European countries.”

Another popular claim about global warming is once again not supported by what has been observed for centuries – sound familiar?


Tol, R.S.J, and S. Wagner. 2010. Climate change and violent conflict in Europe over the last millennium. Climatic Change, 99, 65–79.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress