Iraq and Kuwait
1972, 1990, 1991, 1997

These images show some effects of economic, political, and military events on the landscape in and around Kuwait. These include the destruction of marshes in Iraq between 1972 and 1997, and the burning of hundreds of oil wells during the Gulf War of 1990-1991. In these images, vegetation looks red and desert looks bright. Clear, deep water looks almost black, but shallow or silty water looks lighter.

Irrigating in Mesopotamia

The Tigris and Euphrates are like twins, starting near each other in the mountains of Turkey, and flowing parallel through dry flat land until they meet near the Gulf. The ancient name Mesopotamia means "between the rivers," and this is where many historians believe the first "civilizations" developed. This means the societies were settled rather than nomadic, they were farmers rather than hunters, and they had central authority. Rainfall in southern Mesopotamia, especially, was insufficient for farmers to rely on, so they used the river water flowing by to irrigate their crops. This had social and political effects. Irrigation required a coordinated system of embankments, canals, etc., so a centralized bureaucracy developed to manage the waterworks, keep written records, and live off the farmer's surplus grain. The Tigris and Euphrates have been manipulated for a very long time.1

Unlike rainwater, river water contains salt. Irrigated fields often become waterlogged with salty water which kills the crops, so people build drainage systems to carry away the salt and excess water. In the mid-1900s, American and European engineers designed such a drainage system for the newly-independent country of Iraq. Much of the recent engineering actually follows these plans. The largest element of this engineering is the "Third River" (labeled "Main Outfall Drain" on the CIA map), a canal from Baghdad to the Gulf, built from 1953 to 1992, designed to carry irrigation drainwater from smaller side-drains. It runs between the Tigris and Euphrates, then it crosses under the Euphrates riverbed via three large pipes, to take a more southerly route to the Gulf.2

The building of the Third River and its related engineering works has generated much debate-- not about draining fields to help agriculture, but about draining marshes to eliminate political opposition.

Draining marshes in Iraq

Around the area where the Tigris and Euphrates met, much of their water flowed through a large triangle of marshes. The Ma'dan, or Marsh Arabs, lived in the marshes for thousands of years, building reed houses on artificial floating islands of reeds, moving around by boat, selling reed mats, and living on fish, water buffalo, and rice. Wildlife also lived in the marshes; a 1979 survey found 81 species of waterfowl, including birds which were rare or endemic. The marshes were said to be one of the most important wintering grounds in southwest Asia.3

The midcentury engineers drew up plans to drain these marshes, as wasteful evaporators of potential irrigation water. But in 1972 they were still plainly visible to Landsat 1. In the 1970s, Turkey dammed the Euphrates, reducing water to the marshes far downstream. In 1985 the Iraqi government built levees and drained part of the marshes, to develop the oilfield there. Without water the reeds die; in the Landsat images the bright red color, representing vegetation in these areas, is gone.4

The Euphrates has been brandished as a "water weapon" several times. In 1989 Turkey threatened to restrict Syria's water supply, and during the Gulf conflict of 1990-1991 the same leverage was contemplated against Iraq. After the war the Iraqi government defeated a Shiite Muslim revolt. Many cite this as the government's motivation to finally remove the marshes. The Marsh Arabs are themselves Shiite, and reportedly maintained links with Iraqi dissidents and with Iran. Army deserters and Shiite refugees also hid in the marshes, as well as rebel fighters who slipped back and forth by boat over the Iranian border.5

It took the government almost a year, 4,500 workers, and round-the-clock shifts to complete the Third River in December, 1992. They also dammed the Euphrates near where it passed over the Third River, so that most or all of its water was diverted into the Third River. Iraqi officials said this was to keep salty water from the marshes, but critics say it was to keep any water from the marshes. The marshes' other water source was the Tigris flowing in from the north, through a delta-like maze of distributaries. Apparently during 1992, these were either blocked at the Tigris or diverted into a long, wide canal which forms an obtuse "7" in the 1997 image. This canal empties into the Euphrates near the old Tigris-Euphrates confluence. The Euphrates got a second dam, just upstream from the confluence, allegedly so that Tigris/canal water could not flow up the Euphrates bed and reach the marshes.6

The "draining" of the marshes is actually a combination of draining from below and cutting off from above, and there were more engineering works than described here. The Iraqis point to even other factors, such as increasing water use all along the rivers (perhaps even more irrigation since Iraq has been under international sanctions), and alleged damming of streams by Iran. But the basic premise is that the Tigris and Euphrates used to percolate through the marshes, but now they are piped through the marshes or diverted around them. There has been a shortage of information on the marshes since about 1994, perhaps because the story is over and the marshes are gone. The 1997 images show little sign of them.

Many marsh inhabitants fled to Iran, from where accounts-- accurate or not-- reached the press. They describe the former marshes as a landscape where people struggle through deep mud, water is underground and unsafe, waste accumulates, and food is scarce. They say that in addition to dikes and ditches, the government secured the marshes using troops, mines, poison, and artillery. Other critics accuse the Iraqi government of draining unusable land, leaving good land unused, and ruining other land by stripping its topsoil to build dikes.7

Trenching along the border

These images also show water used for military purposes. In the 1980s Iran-Iraq War, Basrah was a major target, as Iraq's second city, a major port, and an industrial center. Much of the war was fought with tanks, and the Iraqis built several obstructions to protect Basrah. The 1990 Landsat image shows water still standing along the right-angled border. Nearer Basrah, the remains of a long, diagonal moat can be seen, built by Iraq in the mid-1980s. Beyond its ecological roles, water can play various military roles, aiding guerrillas, but blocking tanks.8

Spilling and burning in Kuwait

The Landsat images also show oil spills and oil fires in Kuwait from the Gulf War. During the air and ground war of January-February 1991, 700 oil wells were damaged, of which more than 600 were set on fire.

The August 1990 image shows the capital city Kuwait in the upper part of the image. In the February 1991 image the Kuwaiti coastline south of the city is obscured by smoke plumes from burning oil wells. The November 1991 image was acquired after the fires had been extinguished.

Besides affecting the oceans and atmosphere, this oil had a severe impact on Kuwait's landscape. The sand and gravel on the land's surface combined with oil and soot to form a layer of hardened "tarcrete" over almost 5% of the country's area. Over 300 oil lakes also formed; though covering only about 0.1% of Kuwait's area, these lakes were estimated by the Kuwaiti Oil Minister to hold 25 to 50 million barrels of oil. They are visible in the images as black pools within the dark gray tarcrete.

Military movements also disturbed sand fields in Kuwait and elsewhere. Spatially this is a much larger problem, affecting about 17% of Kuwait's area. The sand is normally held in place by a thin layer of pebbles known as "desert pavement". Troops also dug in the sandy soil, and some sparse vegetation was destroyed. The prevailing winds are from the northwest, which you can see by the drifting sand in the post-fire Landsat image.9


Look at the oil field during the fires. What is represented by the red line left of center? (See the answer below.)


1. Clive Ponting, "Historical Perspectives on Sustainable Development": Environment v. 32, no. 9, November 1990, p. 6-7.

2. Andrew North, "Saddam's Water War", Geographical Magazine, July 1993, p. 11-12. Colbert C. Held, Middle East Patterns: Places, Peoples and Politics, (Boulder, Westview, 1994), p. 278. Fred Pearce, "Draining Life from Iraq's Marshes", New Scientist 17 April 1993, p. 11-12.

3. North, p. 14.

4. North, p. 12. Pearce, p. 12.

5. North, p 14. Peter H. Gleick, "Water, War and Peace in the Middle East": Environment v.36 no. 3, April 1994, p. 14. Pearce, p. 11.

6. Pearce, p. 12. North, p. 11-12. Held, p. 278.

7. North, p. 12, 14. Pearce, p. 12.

8. North, p. 14. Pearce, p. 12.

9. Magaly Koch and Farouk El-Baz, "Identifying the Effects of the Gulf War on the Geomorphic Features of Kuwait by Remote Sensing and GIS": Photogrammetric Engineering & Remote Sensing July 1998, p. 739-743. Richard S. Williams Jr. and others, Environmental Consequences of the Persian Gulf War: special issue of Research & Exploration, National Geographic Society, Committee for Research and Exploration, 1991, p. 11.

Other references

Edward Barnes, "Sanctuary Under Siege": Time, v. 141 n. 13, 29 March 1993, p. 32.

Julie Flint, "The Terror in Iraq: Massacre in the Marshes": World Press Review, v. 40, n. 12, December 1993, p. 22.

Iraqi National Congress, "Drained of Life: Saddam's Marshes" (18 January 1999).

Andrew North, "Eco-genocide in the Marshes": The Middle East, n. 235, June 1994, p. 34.

Satellite images

Landsat scene LT4165040009024310 (Kuwait, Landsat TM, 31 August 1990)

Landsat scene LT4165040009105410 (Kuwait, Landsat 4 TM, 23 February 1991)

Landsat scene LT5165040009131810 (Kuwait, Landsat 5 TM, 14 November 1991)

1972 Landsat MSS mosaic: mostly 1/2 August 1972: Landsat scenes LM1178039007221490, LM1178038007221490, LM1178040007221490, LM1179039007320990, LM1179038007221590.

1990 Landsat TM mosaic:

path  row     date
163   40,42   10 Sep 1990
164   38      17 Sep 1990
164   39       1 Sep 1990
164   40       9 Sep 1987
164   41-42   16 Aug 1990
165   38      19 Jan 1987
165   39-42   15 Aug 1990
166   38-42   14 Aug 1990
167   38-39   13 Aug 1990    
21 February 1997 Resurs image: from the Russian Resurs-01 satellite, MSU-SK sensor, 160 m pixel size. Scene ID: SK970221N30E049.

1990 AVHRR image: a mosaic of August and September AVHRR scenes from NOAA satellites, with country borders overlaid.

1997 AVHRR image: AVHRR scene AL14083097105346 (NOAA 14 satellite, 30 August 1997, with country borders from the 1990 mosaic)


Defense Mapping Agency, 1991 (compiled 1983, partial revision 1991), Tactical Pilotage Chart H6B: scale 1:500,000.

U. S. Central Intelligence Agency, 1994, Former Marshes and Water Diversion Projects in Southeastern Iraq, June 1994: original scale 1:2,100,000. Available in February 1999 from The Perry-Casta´┐Żeda Library Map Collection at The University of Texas at Austin, at Included in: U. S. Central Intelligence Agency, 1994, "The Destruction of Iraq's Southern Marshes": color foldout, advertised at map number IA 94-10020, NTIS number PB-94-928014.

Defense Mapping Agency, 1957 (compiled 1957, partial revision 1988), Global Navigation and Planning Chart 11: scale 1:5,000,000.

Defense Mapping Agency, no date, The World: Mercator projection, scale 1:30,000 at the Equator, stock number 1145XTHEWORLD.

Answer to the question above

This red line is a product of the sensor, not a real part of the landscape.

As the satellite passes southward, its sensor quickly scans east-and-west. In this image, it was scanning east-to-west as it passed over a very hot oil fire (the orange dot at the end of the red line). Even at an altitude of about 700 km, the intensity of the fire made the near-infrared band (band 4, shown here as red) saturate and "stick" at the maximum value for a short time after the scanner had passed on to the west. These sensors are built to measure faint reflected sunlight, not to stare directly into oil fires. Only one band saturated, though-- that's why the oil fire itself looks bright orange but the "tail" is just red.

The specific name for this effect is "A/D Converter Saturation". The TM detectors react to detected energy by giving off analog voltage, and then this analog voltage is converted to a digital number (DN) between 0 and 255 by a piece of electronics called the Analog-to-Digital (A/D) Converter. So technically, it isn't the case that the detector "heated up" and took a while to "cool down"-- it's that the brief overload from the detector made the converter jam at DN = 255.

How to cite this article

Campbell, Robert Wellman, ed. 1999. "Iraq and Kuwait: 1972, 1990, 1991, 1997." Earthshots: Satellite Images of Environmental Change. U.S. Geological Survey. This article was released 14 February 1997 including Kuwait, and revised 14 February 1999 adding Iraq.