Wetlands in the Mesopotamian Plain in southern Iraq were extensively drained in the 1990s. Re-flooding of drained areas commenced in 2003, and included parts of the Central marsh between the Euphrates and Tigris Rivers. New vegetation in the re-flooded areas of the Central marsh was studied in 2006. Most of the wetland plant species and communities widely distributed prior to drainage have reappeared, but there were some species and communities that did not re-establish. Aboveground plant biomass is recovering in some communities, but in most of the new communities, biomass and diversity were low. Postflooding sites were characterized by higher concentrations of chloride and bicarbonates in surface water and higher percent organic matter in sediment than those prior to drainage. Comparisons among the three study sites in the re-flooded areas and those between pre-drainage and postflooding sites suggests that differences in water quality, including more saline conditions in the re-flooded wetlands, might be hindering the biomass recovery. The water source being limited to only the Euphrates River, a much more subdued seasonal fluctuation in the quantity of water input and output, and inputs of contaminated waters appear to be responsible for the delay in vegetation recovery in the Central marsh.

A vast ecosystem of wetlands and lakes once covered the Mesopotamian Plain of southern Iraq. Widespread drainage in the 1990s nearly obliterated both components of the landscape. This paper reports the results of a study undertaken in 1972-1975 on the vegetation of the wetlands prior to drainage and provides a unique baseline for gauging future restoration of the wetland ecosystems in Mesopotamia. Five representative study sites were used to assess the flora, three of which were wetlands. A total of 371 plant species were recorded in the five sites, of which approximately 40% represent obligate or facultative wetland species. The wetland vegetation was classified into five major physiognomic forms (submerged, floating, herbaceous tall emergent, herbaceous low emergent and woody low emergent), which was further subdivided into 24 fresh and halophytic communities. Water levels greatly fluctuated across the different types of wetlands, and mean surface water depth ranged from below to greater than 2 m above the sediment surface, reflecting permanently, seasonally or intermittently wet habitats. Aboveground biomass was also highly variable among the communities. The Phragmites australis community, which was the most extensive community type, had the greatest biomass with an average value of approximately 5,000 g m -2 in summer. Distribution and community composition were largely controlled by water levels and saline-freshwater gradients. Canonical correspondence analysis showed that salinity and water depth were the most important factors to explain species distribution. Environmental variables related to soil salinity separated halophytic species in woody low emergent and herbaceous low emergent forms (Tamarix galica, Cressa cretica, Alhagi mannifera, Aeluropus lagopoides, Juncus rigida, and Suaeda vermiculata) from other species. Their habitats were also the driest, and soil organic matter content was lower than those of other species. Habitats with deepest water were dominated by submerged aquatic and floating leaved species such as Nymphoides peltata, Ceratophyllum demersum, and Najas armata. Such diverse environmental Electronic supplementary material The online version of this article (

No abstract