Personal tools
You are here: Home » Topics » INSAM homepage texts from 2002 onwards (also in Spanish and some in Farsi) » INSAM homepage until April 2011
INSAM Navigation

INSAM homepage until April 2011

Last modified March 31, 2011 13:18

INSAM homepage until April 2011.

In meteorology we have long ago lost trust in arbitrary correlations and other hinted statistical relationships. We have often been warned against apparent connections and associations that were due to much more complex and sometimes very different cause and effect relationships than suggested by simple links. But not so in some social sciences.

A year ago, Marshall B. Burke and colleagues published a paper in the Proceedings of the National Academy of Sciences (of America), PNAS, under the title “Warming increases the risk of civil war in Africa”. Not only do they away, for the purpose of forecasting, with the more logical indirect relationship, via resource endangerment, between drought and conflicts, because “climate models disagree on both the sign and magnitude of future precipitation changes over most of the African continent”. They also argue that few studies have explicitly considered the role of temperature, without considering that this could be the case because of the obvious simplifications involved. If you want to see their in my eyes extremely confusing statistical approaches to human suffering from war in Africa, see:

Then last September, Halvard Buhaug attacked the above approach, also in PNAS. He comes with other statistical approaches that show virtually no correlation between climate change indicators and frequency of civil wars over the past 50 years in sub-Saharan Africa. He argues that climate variability is a poor predictor of armed conflict and that “the primary causes of civil war are political, not environmental” ( These opposite statements may not be surprising because the first group of authors are resource economists and political scientists while Buhaug is what we used to call in the seventies a “polemologist”, a “study of war” researcher.

Before I challenge also Buhaug’s last statement, I make use of a discussion on the two above mentioned contradictory results in Nature, dated 6 September 2010 ( This starts reporting that the two authors particularly question each others handling of the data sets. But then the director of the Potsdam Institute for Climate Impact Research in Germany is quoted in stating that “even if the data and methods were up to the task - which they aren’t - the “causal noise” would be too loud to discern the currently still weak climate signals in civil wars”. That same discussion in Nature also quotes various studies that show that conflicts in sub-Saharan Africa and elsewhere include components related to changing levels of scarcity of resources, that we know is definitely climate dependent.

I feel that the main problem of these studies and discussions indeed is that they compare “apples” with “pears”, that is they compare very dissimilar things, even when both are fruits. There are no two similar conflicts in sub-Saharan Africa nor in other parts of the world, but this is only understood when looking at the essentials of each conflict separately. But in the statistical approaches they are all thrown into one heap to make it possible to find common environmental factors involved that even if and when involved do so in such different ways that their weights and actual influences cannot be compared in any way. But statistics does so and is therefore simply a bad approach for such matters. However, this does indeed not mean that scarcity of and competition for resources are not involved in conflicts and wars.

I found this well illustrated in Eliza Griswold’s “The tenth parallel” (Farrar, Straus and Giroux, New York, 2010) which has the sub-title “Dispatches from the fault line between Christianity and Islam”. This book tells us that religious conflicts along the tenth parallel north of the equator, in Africa (Nigeria, Sudan, Somalia) and Asia (Indonesia, Malaysia and the Philippines), are also conflicts about land, water, oil and other natural resources. “Faith is bound up with (people’s) struggle for resources and survival” (p. 12). Moreover, I can confirm from our work on desertification in the eighties and the nineties in Africa, particularly Sudan, Nigeria and Kenya, that social problems between religious, ethnic and political groups are, as Griswold states, “fomented by a drastic lack of education and services” (p. 41). And here is where all of us working in applied sciences, also in agrometeorology, come in!

“When there is perceived (social and economic) injustice, the conflict will continue” (p. 64). “The elders (in Sudan) also understood that their people were not simply competing for grazing land, or even oil; they were pawns in a larger conflict” (p. 123). “(……) Christians were competing day to day with Muslims over fertile forestland for ebony and cacao. Yet simultaneously, the two struggled (among as well as between them KS.) over which worldview and way of life would dominate (p. 176). “The conflict here and in the rest of the world isn’t just political”, he said, “it’s economic, social, everything a person needs” (p. 260).

We should all remain convinced that you can’t catch this in any statistics. For the calendar year to come I wish that we can have more influence with our agrometeorology and related practices on education and services applied to satisfying basic needs of all, independent of our believes and backgrounds. And for all of you personally, I wish that some of your dreams may come true in a way that harms nobody and strengthens our common goals, again independent of our backgrounds and believes. Jointly we can make this world a significantly better place for all, also statistically!

Document Actions
  • Share on Facebook
  • Print this