Making maps to counter prevailing assumptions and beliefs is a well established tradition. Counter mapping, radical mapping, protest mapping … the map proposes an alternative. Bolstered by its authoritative aura, the map can be quite convincing.
Geographers John Agnew, Thomas Gillespie, and Jorge Gonzalez, with Political Scientist Brian Min (all of UCLA) propose an alternative to the mantra – repeated by just about all on the political Right and Left – that the Iraq “Surge” has succeeded.
Agnew and his colleagues argue that the celebrated decline in violence in Baghdad is actually the result of inter-ethnic cleansing which began prior to the “Surge.” And this counter-proposal about the “Surge” is bolstered by a garrison of maps.
Counter-mapping the “Surge” depends on a relatively mundane set of meteorological satellite data, ironically generated by the Defense Meteorological Satellite Program – Operation Linescan System (KMSP-OLS). Nighttime light is one kind of data collected by this program.
Nighttime light certainly suggests population patterns – we have all seen the global maps of nighttime light – and also access to electricity.
Agnew and his colleagues asked a relatively simple question that can be answered with a series of maps based on the KMSP-OLS data: how has emitted nighttime light in Baghdad changed as U.S. Military strategy in Iraq changed?
The study area consists of the ten security districts in Baghdad, here indicated on a Landsat ETM satellite image.
Nighttime light imagery was selected and analyzed for dates after the U.S. invasion of Iraq (November 16, 2003, 9pm), before the “Surge” (March 20, 2006, 9pm), and after the “Surge” (March 21 and December 16, 2007, both 9pm).
The results seem to contradict proclamations of the success of the “Surge.” In general, Baghdad’s nighttime light increased between the initial U.S. invasion and mid 2006, then begins a rapid decline prior to the implementation of the “Surge” strategy.
Even more interesting, the mid-2006 decrease in nighttime light is not evenly distributed in Baghdad. The areas of declining nighttime light correspond with areas of ethnic violence and cleansing as documented in the Jones Report and its maps.
The greatest decline is in East and West Rashid – historically mixed Sunni and Shia – but also Adhamiya (Sunni), Kadamiya (Shia), Rusafa, and Karada (mixed and Sunni). No change was observed in Sadr City (Shia), New Baghdad (Shia), Karkh (Green Zone), and Al Mansour (historically mixed but heavily Sunni by late 2007). This is certainly easier to see on a map:
Agnew and his colleagues conclude:
Our findings suggest that … the surge has had no observable effect, except insofar as it has helped to provide a seal of approval for a process of ethno-sectarian neighborhood homogenization that is now largely achieved but with a tremendous decline in the extent of residential intermixing between groups and a probable significant loss of population in some areas.
Furthermore, the nighttime light signature of Baghdad data when matched with ground data provided by the report to the US Congress by Marine Corps General Jones and various other sources, makes it clear that the diminished level of violence in Iraq since the onset of the surge owes much to a vicious process of inter-ethnic cleansing.
Disagree? Raise your own army of data and maps to counter this counter-“Surge” proposition.
The text of Agnew, Gillespie, Gonzalez, and Min’s article “Baghdad Nights: Evaluating the U.S. Military ‘Surge’ Using Nighttime Light Signatures” is, for review and educational purposes, here.
What a waste of time…I can’t believe I just read that pile of garbage. I am not sure what pisses me off more – the outlandish assumptions or the claim the data was “high resolution spatial and temporal data”.
Click to access StatSci%20Aug2005%20(Lying%20with%20Maps).pdf
http://www.press.uchicago.edu/presssite/metadata.epl?mode=synopsis&bookkey=47431
I strongly question the correlation between nighttime lighting and violence. Baghdad had/has erratic power transmission, inadequate power generation, and lines simultaneously being repaired and destroyed. Assuming that lights = no violence or vice versa is amazingly simplistic and overlooks that there is no good reason to assume that lighting would *only* be influenced by violence, in contrast to say rising temperatures over time.
Also the timings for the “Surge” in Baghdad is wrong, the surge started in Anbar and for the most part finished in Sadr city and Mosul five months later, so looking at Sadr city when the surge started is meaningless- there were no surge troops deployed there. At best anything that happened them happend in anticipation of troops coming to Baghdad at a later point in time.
Also the map doesn’t take into consideration why violence was happening. Approximately one thousand Mahdi militamen were killed when Maliki had his brief 3 day war with Moqtada’s militia. These people are technically civilian casualties, but civilians with an asterix next to them.
In the end I think that woodward got it right when he stated that things in Iraq became better in light of:
-Sunni’s becoming disillusioned with Al Qaeda’s violence
-Sunni’s realizing that they could not win a civil war
-American troops being deployed in number to Sunni areas
-American troops having a proper counterinsurgency strategy that protects Sunnis
-Magic technology (Woodward won’t say what this is, but I could probably guess)
-Iraqi’s wanting peace
In the meantime your map isn’t worth the digital ink it was printed on.
You’ve also confused correlation with causation (which is popular to do nowadays, no worries, it just makes a mockery of the scientific method). Not that you explained how light correlates to violence (you just stated it).
It is good to see that I am not the only one that is perturbed by the fact that this came out of a university. It is no wonder that our education system is failing. I want to point out that this was a peer-reviewed publication. God help us…
At least UCLA is pretty good at sports:)
As both a U.S. Army combat-arms officer currently serving in Baghdad and college graduate with a B.S. in Geospatial Information Science, I cannot help but be absolutely stunned at the simplistic analysis of this study. This is a product generated above the high-school level but is completely based on a correlation between lights and violence. I might have respect for a study that links to the real spatial distribution of violence, but this is absurd.
1) The underlying assumption is not scientific. Even though there is data stating that some of the areas of declining light correspond with violence, that study does not link violence with light distribution. Therefore, the change in light distribution over time has no relevance.
2) Power distribution changes every few hours in Baghdad. A satellite photo taken at 1900 could show completely difference light patterns than one taken at 2300 the same day. Comparing one image from a given day to another image on another day does not indicate and increase or decrease in power allotment. You can physically see the distribution of electricity changing while flying over Baghdad; whole districts light up and others turn off in an instant.
I am not defending the surge and it is not my role to judge whether it worked or not. I can say that any claim about the surge “working” is misleading as well because of the complexity of the issue. In my area, most violence has declined because community leaders stepped up to speak and act against violence. Perhaps having more troops helped facilitate that action by the community. Ultimately, it is too complex an issue to even begin to analyze whether or not the surge did anything. It will be many years before we know the true impact if we ever do.
This project is just one example of how external motivations can skew the results of a study. I know my education focused heavily on the scientific method; it is unfortunate to see that other institutions do not use such a method as their basis.
Stratman…..thank you for your service!
KoS
What makes me sad is that when people disagree on this page, most of them have been quite rude. It’s shameful that we are like this as a people.
The hard left never ceases to amaze me in it’s creative thinking.
Pathetic.
I love maps and map blogs and I hurried over here to look when I saw the link from Strange Maps. But I won’t be back, because this isn’t really a map blog — it’s a fact-challenged leftie political screed. Buh-bye!
I see a number of fatal weaknesses with this study.
First, the author wears his politics, and thereby his agenda, on his sleeve. While an ad-hominem observation proves nothing, when a researcher produces exactly those results to which he has displayed considerable emotional investment, the results become that much more suspect. An example is the use of the “quotes” and the “so-called” sneer: whether you accept the results or you do not, the surge existed, the increase in the combat troops was real, the increase in offensive operations was real, the use of the sneer- “so-called” – does nothing but demonstrate the bias of the speaker.
Second, the timeline of the observations fails to conform to the timeline of the military actions of the surge, and of the results. It is equivalent to saying that the “so called” allied invasion of Europe in 1944 achieved nothing because over the next six months, the amount of destroyed buildings and bridges in Europe had actually increased. In the latter half of 2007, operations against insurgents in Bagdad were in full swing, we would not expect a recovery in infrastructure- and a resumption in nighttime lighting- until early 2008. Utility crews have an understandable reluctance to work with bullets whizzing about. At the same time, areas that had seen earlier success do not appear in this study. Why not? What about lighting in Fallujah, where combat operations ceased some time before? Has it been passed over because it does not support the author’s conclusions?
Finally, he attributes the security in Iraq to ethnic cleansing! Yes, we know that many Iraqi neighborhoods were ethnically homogenized during their civil war. But this alone cannot account for the decrease in violence. The rival sectarian groups are still within walking distance of each other, and to a religious fanatic, all a homogenous neighborhood means is that when you explode your car bomb, you’ll kill fewer of your own people. In Europe, substantial ethnic cleansing followed WWII as many populations were relocated, voluntarily or otherwise, to their corresponding nations. But to claim that this caused the decline in violence in Europe would be to reverse the cause and effect. Most of the population movements occurred after the defeat of Germany. Here, the author appears to be making a case in favor of ethnic cleansing, crediting it, and not increased physical security, with the decline in violence in Iraq!
Swami
I respect people who wear their politics on their sleeve. One’s biases do not change simply because one does not disclose them publicly. Honestly, a little more author honesty helps in the proper digestion of a map (and where one should stick it, should the map suck).