The Tigris-Euphrates basin is the largest and most important river system between the Nile and the Indus. Details of its biology can be found in Rzóska (1980) but comparatively little is based on the Iranian part of this basin although Nümann (1966) gives some limited data on chemical and physical parameters. Studies on limnology and pollution were restricted mostly to waters of Iraq, but probably apply equally well to Iran, certainly as far as those marshes which cross the border are concerned and for the Shatt al Arab, part of which forms the southern border of Iran and Iraq. Such studies include Cressey (1958a), Jacobsen and Adams (1958), Al-Hamed (1966c), Mohammed (1965; 1966), Salonen (1970), Al-Saadi and Arndt (1973), Al-Sahaf (1975), Al-Saadi et al. (1975), Arndt and Al-Saadi (1975), Antoine and Al-Saadi (1982), Maulood et al. (1979; 1981; 1993), Sarker et al. (1980), Saad (1978a; 1978b), Saad and Antoine (1978a; 1978b; 1978c; 1982; 1983), Saad and Kell (1975), Kell and Saad (1975), Al-Hamed (1976), Al-Daham et al. (1981), Huq et al. (1981), Schiewer et al. (1982), Antoine (1983), DouAbul et al. (1987; 1987; 1988), Abaychi and Al-Saad (1988), Abaychi et al. (1988), Mohamed and Barak (1988), Al-Saadi et al. (1989), Hussain et al. (1991), Kassim and Al-Saadi (1995), Partow (2001), among others. Ionides (1937) describes the river regimes of the Tigris-Euphrates basin, MacFadyen (1938) the water supplies, El Kholy (1952) the hydrology of the Tigris River, Buringh (1957) the physiographic regions, shores and irrigation systems on the lower Mesopotamian plain, and Al-Khashab (1958) the water budget of the Tigris-Euphrates basin, mainly referring to waters in Iraq. Scott (1995) gives details of wetlands in Iraq, some of which border and/or are contiguous with Iranian wetlands, and whose general ecological features are very similar. Shapland (1997) reviews water disputes in the Middle East although western Iranian rivers flow out of the country and are not likely to be affected apart from any losses in shared habitats or refuges in border areas.

Por and Dimentman (1989) regard the Tigris-Euphrates or Mesopotamian basin as a cradle for inland aquatic faunas. A proto-Euphrates collected water from the Levant and had contacts with the Black and Caspian sea drainages before the Pliocene orogeny. Berg (1940) points out that the upper reaches of the Tigris-Euphrates basin today lie on a plateau close to the upper reaches of the Caspian Sea basin. The basin acted as an area where African and Asian species could meet or transit such as the cichlid Iranocichla. These connections were interrupted in the early Pliocene by orogeny, rifting and desert formation. Banarescu (1977) and Por and Dimentman (1989) regard the area to be a zoogeographic crossroads with elements from the Palaearctic such as the cyprinid genera Leuciscus (= Squalius) and Chondrostoma, Mediterranean genera such as the cyprinid Acanthobrama (although Krupp (1987) refers to this genus as Palaearctic, of Mesopotamian origin), and Oriental genera such as the cyprinid Garra and the spiny eel Mastacembelus.

Khalaji-Pirbalouty and Sari (2004) studied the biogeography of amphipods crustaceans in the central Zagros Mountains. They consider habitat diversification and climatic fluctuations to be the principal factors influencing species diversity and endemism in this area, with the mountains acting as a barrier to species distribution. Endemism is evident in lizards, plants and amphipods as well as fish.

An analysis by Coad (1996f) shows that this basin is mainly Black-Caspian sea basin in its connections, with minor links to Asia and possibly Africa. Numbers of families, genera and species shared between the Tigris-Euphrates and neighbouring basins are summarised in this analysis. Relatively few taxa appear to have made the transition between Asia and Africa or survived subsequent climatic and habitat changes.

Certain families are absent from the Tigris-Euphrates but are found in the Indus and the Nile (Notopteridae, Schilbeidae, Clariidae, Anabantidae, Channidae). These are assumed to be of Gondwanic origin and are separated today by plate tectonic movements. A representative of the Channidae is found in eastern Iran but this species is at the western limit of its range there. Only two families are shared between the three basins but are not found to the north, Bagridae and Mastacembelidae, and the relationships of the two species in these families are with the Indus (Travers, 1984).

At the generic level, some have dispersed into eastern Iran from the Indus and other eastern basins but have not reached the Tigris-Euphrates basin, presumably for reasons of time or lack of suitable environmental conditions, e.g. Aspidoparia, Crossocheilus, schizothoracines. However two genera have reached the Tigris-Euphrates (Glyptothorax, Barilius) and Howes (1982) considers Cyprinion to be related to the eastern genus Semiplotus. Barilius resembles Indus and other eastern species superficially although its relationships have not been fully worked out. Assuming that these taxa dispersed westward from the Indus and the east, the route must be determined. All but Cyprinion are absent from much of Iran, including the bagrid Mystus and the mastacembelid Mastacembelus referred to at the family level above (Mastacembelus is not found in eastern Iran and hence does not have a continuous range throughout the Orient (pace Travers (1984)). It is unlikely that rivers of the Tigris-Euphrates basin were once tributary to the Indus when sea levels were lower during glaciations as the Gulf of Oman descends to an abyssal plain at 3340 m. I suspect, but cannot prove, that these taxa reached the Tigris-Euphrates basin across the Iranian land mass and subsequently became extinct as desiccation increased. Many of the rivers in southern and eastern Iran today are very small, regularly dry up and some are highly saline. They may be unsuitable for these taxa. Barilius, it should be noted, appears to prefer, in Asia and the Tigris-Euphrates basin, large lowland rivers and its dispersal across Iran is difficult to envisage by headwater capture (the other genera can be found in small streams at higher altitudes as well as lowland rivers). However Berg (1940) suggested that fish dispersal across this region was facilitated by the coastal rivers of Iranian and Pakistani Baluchestan being part of a single river system in the Pliocene, since submerged by subsidence. The presence of Mastacembelus and Barilius in western Iranian basins is attributed to headwater capture and/or colonisation from the Tigris-Euphrates basin when Gulf rivers were tributary to an expanded Tigris-Euphrates basin during lowered sea levels in glacial times. This distribution of these genera is not, therefore, a remnant of the dispersal across Iran from Asia.

At the generic level, only Garra is found from the Indus to the Nile and in the Tigris-Euphrates basin. Menon (1964) suggests that Garra reached the Tigris-Euphrates basin and Africa in two "waves" from Asia, the first wave being in the Miocene to the Tigris-Euphrates basin, the second through southern Arabia to Africa during the Pliocene. Karaman (1971) disputes Menon's Garra waves based on anatomy and zoogeography. Garra presumably dispersed from Asia to Africa via the Tigris-Euphrates basin and the Levant. The apparent continuous distribution of Garra across southern Arabia is not borne out in systematic analyses by Krupp (1983). Garra (and Cyprinion) species of southeastern Arabia are clearly related to southern Iranian species, having crossed the Persian Gulf when it was drained during the Pleistocene and part of an extended Tigris-Euphrates basin. Southwestern Arabian species (and a Barbus species) are a mixture of African and Levantine elements. Krupp (1983) found no evidence in his studies for the Arabian Peninsula serving as a transition area in an exchange of freshwater fishes between Asia and Africa.

Nemacheilus also has a similar wide distribution but is probably polyphyletic and requires a detailed revision to enable adequate zoogeographical analyses to be made. The systematics of loaches in the Middle East is a contentious subject (Por and Dimentman, 1989). The absence of Nemacheilus species from southern Arabia also argues for a dispersal route through the Tigris-Euphrates basin as these cryptic fishes are found today in many small streams throughout Southwest Asia and are unlikely to have been eliminated from southern Arabia through desiccation.

The only Nile (or east African) genus present in the Tigris-Euphrates basin is Barbus. Certain members of this polyphyletic genus in Southwest Asia are characterised by sharing 6 branched anal fin rays, last unbranched dorsal fin ray a smooth spine, large scales, few gill rakers, high dorsal fin ray counts, reduced barbel numbers, compressed body, and other characters which set them apart from European Barbus as a monophyletic group, probably related to east African species (suggested by Banister (1980)). These Barbus species are found from southwestern Arabia (but not southeastern Arabia), through the Levant and the Tigris-Euphrates basin to rivers at the Strait of Hormuz in Iran. They may represent an African element in the fauna of the Tigris-Euphrates and may reflect the route of the cichlid Iranocichla or its ancestor from Africa to the Strait of Hormuz. Bănărescu (1992b) considers African elements in Southwest Asia to be the oldest, of at least Miocene age.

A significant proportion of the families and genera in the Tigris-Euphrates basin is also found in the Black-Caspian sea basin. Such widespread, northern cyprinid genera as Alburnoides, Alburnus, Aspius, Alburnus, Chondrostoma, and Leuciscus (= Squalius) reach their southern limit in the Tigris-Euphrates basin (and neighbouring Iranian basins) suggesting that they reached the Tigris-Euphrates basin from the north.

The presence of Glyptothorax in the Black Sea basin of Anatolia (Coad and Delmastro, 1985) is a recent event through headwater capture from the Tigris-Euphrates basin and thus far is the only example of a clearly-defined Indus genus reaching the Black-Caspian seas basin. It is probably an example, in reverse, of the colonisation of the Tigris-Euphrates basin in recent times from the Black-Caspian seas basin. Headwaters of a number of Tigris-Euphrates basin rivers interdigitate with the upper reaches of Black-Caspian seas basin rivers, e.g. the Aras River of the Caspian Sea and the Kizilirmak of the Black Sea with the Euphrates near Erzurum and Sivas respectively; the Qezel Owzan of the Caspian Sea with Tigris River tributaries. Headwater capture is common in the Zagros Mountains (Oberlander, 1965) and in Anatolia and pluvial conditions in the past would have facilitated fish dispersal. Por and Dimentman (1989) mention direct connections of a proto-Euphrates with Black Sea and Caspian sea fluviatile drainages before the Pliocene orogeny which would serve to allow entry of taxa to the Tigris-Euphrates basin. Direct connections were interrupted by the early Pliocene as orogeny, rifting and desertification took hold. Almaça (1990) has reviewed possible routes for Barbus species into Iran and the Tigris-Euphrates basin from the north via what is now Anatolia and east of the Caspian Sea dating from the early Oligocene. A continuous route for exchange of taxa has been possible since the upper Miocene, almost 12 million years ago. These routes have been variously available down to modern times for Barbus and other taxa as exemplified by some species being in common between the Black-Caspian seas basin while others are distinct but related at the generic level. Bănărescu (1992b) considers that northern or European elements penetrated to the Tigris-Euphrates basin earlier than Asian ones, and that this partially explains their prevalence.

Iranian internal and Gulf basins and the Levant show evident affinities with the Tigris-Euphrates basin. The ichthyogeography of the Levant has been dealt with by Krupp (1987) and will not be reviewed here. Krupp considers that parts of the Levant were colonised separately via branches of the Tigris-Euphrates river system. Iranian basins to the west of the Tigris-Euphrates basin have a very similar fauna to that of the Tigris-Euphrates at the species level. The diversity falls off rapidly with distance (Coad, 1987). Headwater capture in the Zagros Mountains is an evident route for species found in common with the Tigris-Euphrates basin but not in Iranian rivers draining separately to the Gulf. The draining of the Gulf during Pleistocene lowering of sea levels enabled Tigris-Euphrates basin fishes to colonise tributary Iranian rivers now separated by a rise in sea level. The melting of the Laurentide ice sheet and drainage of Lake Agassiz in Canada caused this rise in sea level world-wide, including the shallow Persian Gulf (Perkins, 2002). By about 11,500 years B.P., the Gulf was filled with present shorelines attained shortly before 6000 B.P. and exceeded by 1-2 m (Lambeck, 1996).

Por and Dimentman (1989) regard the Mesopotamian subregion or Tigris-Euphrates basin as one of the most isolated major freshwater areas in the world. However, as Coad (1997f) points out, endemism is only at the species level and diversity is low with only about 52 primary division species in 7 families, 34 species of which are Cyprinidae.

The Zagros Mountains form the western flank of Iran and store water as snow. The higher peaks are snow-capped even in summer. Zard Kuh, for example, reaches 4548 m (32°22'N, 50°04'E). Rivers drain south and west to become tributaries of the Tigris River in Iraq or its confluence with the Euphrates River, the Shatt al Arab (known as the Arvand (= swift) Rud in Iran). The Shatt al Arab has a course of 190 km to the head of the Persian Gulf and is navigable by ocean-going ships. It forms part of the Iran-Iraq border. The origin of the Tigris River is the Hazar Gölü of Elazig (38°41'N, 39°14'E) between the Murat Nehri and the Euphrates. It flows south-east, forming a short section of the border of Syria with Turkey, before entering Iraq to parallel, roughly, the course of the Euphrates River. It is a larger and swifter river than the Euphrates because of its left bank tributaries from Iran. The Tigris is over 1900 km long (1851 km and 2032 km are extremes cited in the literature). It is the 81st river in size in the world. The Tigris-Euphrates basin encompasses 784,500 sq km of which 19% or 146,000 sq km lies in Iran (Gleick (1993) gives 238,500 sq km and 27% for Iran and 884,000 sq km for the whole basin; the Iraqi Government in the same publication gives 378,834 sq km for the Tigris basin alone with Iran's share 28.8%). Iran contributes 7% of the water supply of this immense basin. The Tigris catchment is 166,155 sq km.

The Tigris is an alkaline river (pH 7.8-8.2) with a total hardness of 200-350 mg/l. Water temperatures range from 8.5°C in January to 31.4°C in August. The flow pattern of the Tigris and its tributaries has a sharp peak in April at about 9 billion cu m, falling rapidly to about 1 billion cu m from August to October before beginning to rise again. The water level may fall by as much as 2 m over the summer. Interannual variation in spring flood levels are marked. Approximate streamflows over the past 6000 years are given by Kay and Johnson (1981) based on proxy data from paleoenvironmental sources. They found an increase in streamflow over this period. The southern province of Khuzestan in the Tigris river basin with 9% of Iran's surface area has an estimated 37% of its surface water flow.

The Shatt al Arab is under tidal influence up to 110 km from the mouth. Its waters are therefore strongly mineralised. Salinity varies with distance from the sea. Crops are irrigated by means of the tidal rise which is used to push fresh water into the fields (Harrison, 1942; Gholizadeh, 1963; Gholizadeh and Fatemi, 1969). This has obvious effects for the fish fauna and its composition as well as for increased salinisation of habitats. There are appreciable diurnal and seasonal fluctuations in physico-chemical conditions. Tidal waters probably penetrated far inland through the Holocene as evidenced by faunal remains in boreholes of the Hammar Formation (MacFadyen and Vita-Finzi, 1978). Al-Hassan and Hussain (1985) describe the hydrological parameters affecting the penetration of marine fishes into the Shatt al Arab. Recently an increase in the Tigris River discharge has decreased salinity in the Shatt al Arab: previously marine species were common at Basrah in Iraq but they became rare, Carassius auratus appeared in Basrah fish market and Cyprinus carpio was caught in large numbers down to the estuary (N. A. Hussain, in litt., 1994). Pollution is widespread in the Shatt al Arab from industrial, agricultural and untreated human wastes. Hussain et al. (2001) evaluate environmental degradation in the Iraqi portion of the Shatt Al-Arab and its effects on the fish fauna.

The principal Iranian tributaries of the Tigris River are the Little Zab River (= Zab-e Kuchek) which drains a small stretch of mountains south of Lake Orumiyeh, and the Diyala River (= Sirvan River) which drains the western mountains of Kordestan. The Diyala River is 442 km long. A principal tributary of the Diyala in Iran is the Qeshlaq River which flows through Sanandaj (35°19'N, 47°00'E). Lake Zaribar, Zarivar or Zeribar is a permanent freshwater body with fringing reed beds and extensive marshes lying at 1435 m in the Diyala River drainage just west of Marivan at 35°32'N, 46°08'E. It has an area of 8 sq km and a maximum depth of 6 m and an average depth of 2.5-3.5 m. Reputedly the lake is fed by 600-700 springs. At high water it overflows into a small river at its southern end. In winter it often freezes over. It was damaged in the Iran-Iraq War suffering rocket and missile hits and chemical warfare (Scott, 1995; 1997). There is a small resort at the southeast corner of the lake, the surrounding land has livestock grazing and agriculture with drainage channels for the peripheral marshes, forests are cut for fuel, and there is waterfowl hunting and fishing. Exotic fish species have been introduced by a government organization, including Alburnus alburnus, Ctenopharyngodon idella, Cyprinus carpio (in two varieties), Hypophthalmichthys molitrix, H. nobilis, Pseudorasbora parva and Gambusia holbrooki. Native fish include Capoeta buhsei (sic),Leuciscus (= Squalius) cephalus and Mastacembelus mastacembelus (Scott, 1997). Jalali et al. (2002) add the species Capoeta damascina (probably the correct identification of the C. buhsei listed above), Carassius auratus and Chalcalburnus (= Alburnus) sp.

A number of minor streams also cross the Iran-Iraq border, but the principal rivers drain through anticlines in spectacular gorges or tangs, funnelling the waters of the Zagros onto the Khuzestan plains through a narrow gap near Dezful (32°23'N, 48°24'E). Stream flows in late winter are at least ten times that of summer and 116,500 sq km of mountains and three big rivers debouch onto 38,800 sq km of plain. Lowlands may be inundated for more than 100 days. Early accounts of floods in Mesopotamia, dating back to Sumerian times almost 5000 years ago, are discussed by Mallowan (1964). Floods can encompass close to 100,000 sq km in Iran and Iraq at the head of the Persian Gulf (Naff and Matson, 1984). Progressive clearing of woodland over the last 7000 years increased runoff, causing higher and more severe floods, soil erosion, increased turbidity in streams and higher sedimentation (Wagstaff, 1985). Erosion is three times the world standard rate at 30 tonnes/hectare and will rise twofold over the next ten years (IRNA, 20 December 1998). All these must have, and continue, to affect the fishes in this and other basins, favouring those species able to cope with these conditions. Even artificial habitats such as small dam reservoirs in Chahar Mahall and Bakhtiari are affected by high sedimentation rates and their utility as fish habitat must be affected (Mousavi and Samadi-Boroujeni, 1998).

The main river is the Karun, with a catchment of 67,340 sq km (Naff and Matson, 1984) and a length of 820 km. It now drains to the Shatt al Arab but once drained directly into the Persian Gulf. The Karun is also connected to the Gulf via the Bahmanshir River, paralleling the Shatt al Arab, and enclosing Abadan Island. The Bahmanshir is the only river along the Persian Gulf coast to have a significant fishery. The Karun headwaters are extensive and lie near both the Esfahan and Kor River basins. The environmental conditions in a headwater dam, the Hanna Reservoir, in the Karun basin are described by Esteky (2001), two-thirds of the reservoir being covered by macrophytes. The Dez River is a Karun tributary and is 400 km long. The Karkheh River (with the Cherdavel, Kashkan, Qareh Su, Gav Masiab and Simareh in its upper reaches) is 320 km long, but is lost in the Hawr-al-Azim marshes of the Tigris after draining 43,000 sq km. Sutcliffe and Carpenter (1967) described runoff from the Karkheh basin. The Karkheh and Dez flows were depleted by 70% in 2001 during a drought and it was thought that these rivers might dry completely (Foltz, 2002). The marshes along the Karkheh and Dez rivers, with oxbow lakes and riverine forest, are a habitat now rare in southern Iran and Iraq outside protected areas. The severe drought of the year 2000 dried up the natural Dez reservoirs south of Dezful (www.irna.com/newshtm/eng/08130315.htm, IRNA, 29 July 2000). The Karkheh Dam, 20 km northwest of Andimeshk, has a crest 3030 m long, a height of 127 m and is the sixth largest dam in the world with a capacity of 7.8 billion cu m, nearly a third of the total dam capacity for the country. The dam is meant to produce electricity, for fish farming and to control floods and drought (IRNA, 17 April 2001; 19 April 2001; Aftab Yazd, Tehran, 346(18 April 2001, 7 pp.; Sadegi, 2003). The Qareh Su near Kermanshah is about 30 m wide and less than a metre deep at its deepest. The Qareh Su or "black water" derives its name from its transparency over a dark, pebbly bed, distinguishing it from the muddy rivers of the lowlands. The Qareh Su is the Classical Choaspes, the water of which the ancient monarchs of Persia carried with them on their military expeditions for its taste, a superiority confirmed by Buckingham (1829). The Gawshan Dam is located at Kamyaran near Kermanshah on the "Gaweh" River and is scheduled for completion in 2002. The dam will be 136 m high and the complex includes a 19 km long tunnel for water transfer (http://netiran.com/news/IranNews/html/94111305INEC.html). Other dams include the 40 million cu m Zarivar Dam in Marivan and the 563 million cu m Kavoshan Dam 35 km south of Sanandaj (http://netiran.com/news/TehranTimes/html/95111803TTPL.html).

Partow (2001) lists 18 dams in the Tigris basin of Iran, either constructed or planned, and these will affect the environment markedly in changing flow regimes, impounding water and eliminating fluvial habitat, removing silt, affecting temperature downstream, causing salinisation as return water from irrigation projects flows back into rivers, and so on. The Karkheh Dam is planned to carry water via pipeline over land (330 km in length) and under the sea (210 km) to Kuwait. The supply rate would be 200 million gallons per day (Partow, 2001) or 300 million cu m (www.irna.com, downloaded 29 January 2003).

The Dez (formerly Mohammed Reza Shah Pahlavi) Dam on the Dez River at 32°38'N, 48°28'E contains 3350 million cu m of water (another source states 60 billion cu m) and has a maximum surface area of 4000 ha. Surface water temperatures can exceed 30°C while at 50 m plus depths it is 15-16°C in summer. Its original life span was estimated at 100 years but this had to be reduced to less than 50 years because of the rapid accumulation of sediment from erosion. Sediment prevents development of a bottom fauna and steep banks with water fluctuations limit vegetation. Nümann (1966, 1969) gives some limnological information on this reservoir. Nümann (1966) recommended introduction of Acanthobrama terraesanctae and Tilapia galilaea from Israel to the reservoir, and later Sander lucioperca and even trout. Sabzalizadeh (2006) gives a description of the ecology of this reservoir and Eskandary et al. (2007) a description of fish populations. Capoeta trutta, Barbus grypus and Barbus esocinus were the most numerous species and the fauna includes the exotics, Carassius auratus, Hypophthalmichthys molitrix and Oncorhynchus mykiss. There is also a diversion dam, the Sadd-e Gotvand. The Gotvand Dam is under construction and will be 180 m high with a reservoir capacity of 4,500 million cubic metres making it the second largest dam in Iran (sic) (IRNA, 25 January 2000). The 205 m high Karun-3 Dam near Izeh, to be completed in the year 2001 (filling actually started in 2003 - www.netiran.com, downloaded 15 November 2004), is a major hydroelectrical plant as is the Karun-4 Dam (Shahid Abbaspour) near 25 km northeast of Masjed-e Soleyman (http://netiran.com/news/IranNews/html/95040822INPL.html). A major dam is also planned at Shushtar (IRNA, 26 September 1998). A tunnel is planned from the Dez River to Golpayegan to supply water to Markazi Province in central Iran (www.iranmania.com, downloaded 19 January 2004).

Some literature refers to the Seymarreh-Karasu-Gamasiab (and variant spellings) as an important complex of rivers. These are the Simareh, Qareh Su and Gav Masiab in gazetteers. A giant dam is planned for the Simareh (IRNA, 26 September 1998). Nümann (1966) notes pollution in these rivers from an oil refinery and sugar factory which decreased fish populations, a condition exacerbated through the use of explosives, insecticides and herbicides by local people to catch fish. He also lists explosive usage on the rivers Khairabad and Zohreh.

Lake Mirabad lies in the basin of the Karkheh at 33°05'N, 47°43'E. While it measures only 100 by 200 m it is important for establishing past vegetation and environments based on sediment cores (Griffiths et al., 2001). The Hashelan or Hashilan Marsh at 34°33'N, 46°55'E occupies 260 to 400 ha (accounts differ) northwest of Kermanshah at about 1310 m. It is a complex of permanent spring-fed pools and associated marshes with much submerged, floating and emergent vegetation. The surrounding plains are heavily grazed and cultivated and ducks are hunted in the marshes. The Sabz Ali spring feeding the marsh has an average annual discharge of 323.4 l/sec, range 208.3-442.5 l/sec, highest in March and lowest in September. The total average volume of water in the marsh is estimated at 1.02 x 10⁷ (Karami et al., 2001). Local people and those from Kermanshah fish in the marsh. A drought in 2008 severely affected the Hashilan Marsh (www.payvand.com/news/08/aug/1152.html, downloaded 5 July 2009).

A truck carrying diethyl hexanoyl plunged into the Kashkan River, a Karkheh tributary in Lorestan, 15 km from Pol-e Dokhtar resulting in the poisoning of thousands of fish on 13 April 1998 (IRNA, 14 April 1998; Brief on Iran, 880, 16 April 1998). The river suffered an oil slick in October 2001 when the Khuzestan-Tehran pipeline fractured 4 km from Pol-e Dokhtar. Oil pollution caused a fish kill numbering about 70,000 fish in the Kambel River near Gachsaran, a centre of oil production (Tehran Times, 24 November 2002). Varkouhi and Sobhani (2005) studied the presence of various pollutants in the livers of fishes in the Khorramabad River.

The Jarrahi River is a southern Karun tributary from the east. The Marun River is a major Jarrahi tributary. The Marun and Jarrahi feed the Shadegan Marshes, the largest Iranian wetland according to Kurdistani and Bajestan (2004). The Marun Reservoir Dam northeast of Behbahan was scheduled for completion in 1996 with a crest of 345 m (IRNA, 11 November 1998) but was to be completed in 2004 with a crest of 175 m and containing 1.2-1.3 billion cu m of water (IRNA, 12 January 1999; IRNA, 5 February 2002). There are also four diversion dams on the Marun and one of these, the Jazaeen, has a fishway but fish are trapped downstream of it during their migration (sic)(Kurdistani and Bajestan, 2004). Other dams in this system lack a fishway. Later Kurdistani and Bajestan state that there are no migratory fishes in the Marun, only resident species (which presumably undergo local movements blocked by dams). They mention Barbus grypus and B. pectoralis as the affected species. The Jareh Dam on the Zard River northeast of Ramhormoz dates back to the Sassanid era and is still in use (IRNA, 26 June 2000).

The Karun has the greatest mean discharge, followed by the Dez and Karkheh. The Karun mean discharge is the largest in Iran. The Karun carries a heavy silt load with a hundredfold increase during flood. The Karun discharge ranges from 207 cu m per second to 2225 cu m/sec, average 1100 cu m/sec, while the Dez is 63-1227 cu m/sec, average 288 cu m/sec. The Jarrahi range is 8-770 cu m/sec, average 78 cu m/sec. These figures vary among different sources indicating fluctuations between years and gauging stations; however the relative importance of these rivers is shown. The peak discharge of the Karun is in April, with high values also in March and May; the lowest discharge is in October when flow is only about a ninth of the peak. The combined Tigris-Euphrates-Karun in flood carries five times the load of the Nile (Fisher, 1968). Most of this is deposited north of Basrah (30°30'N, 47°47'E) and much is lost to evaporation in the marshes, e.g. of 27 cu km of discharge into the Persian Gulf through the Shatt al Arab, 22 cu km is from the Karun River. 22 million metric tons of dissolved chemicals are deposited each year and hence there are siltation and salinity problems in the lower parts of this basin.

The Karun River on the Khuzestan plains was examined in 1992 for various parameters and at various localities (courtesy of the Iranian Fisheries Research and Training Organization, Ahvaz). It has a pH of 7.07-8.85, mean 8.17, dissolved oxygen 5.6-12.38 p.p.m., mean 9.29 p.p.m., bicarbonate 79.3-214.72 p.p.m., mean 154.4 p.p.m., carbonate 0.6-21, mean 5.53 p.p.m., total alkalinity 1.9-3.8 meq/l, mean 2.84 meq/l, carbonate hardness 5.32-10.64 p.p.m., mean 7.95 p.p.m., total hardness 168-474 p.p.m., mean 287 p.p.m., ash residue 40-1142 p.p.m., mean 425 p.p.m., chloride 45.4-518.3 p.p.m., mean 207.28 p.p.m., total dissolved solids 226-1374 p.p.m., mean 696 p.p.m., sulphate 43.75-325 p.p.m., mean 101.73 p.p.m., calcium 33.63-101.7 p.p.m., mean 61.8 p.p.m., magnesium 16.8-78.24 p.p.m., mean 33.8 p.p.m., phosphate 0.05-4 p.p.m., mean 0.24 p.p.m., iron trace to 0.32 p.p.m., mean 0.069 p.p.m., manganese trace to 0.657 p.p.m., mean 0.483 p.p.m., and nitrate trace to 0.657 p.p.m., mean 0.039 p.p.m. Esmaili et al. (1999) report heavy metals in water, sediments and fish from the Karun River and Jafarzadeh-Haghiehi et al. (2005) report on the poor water quality of the river.

As lowlands at the head of the Persian Gulf receive waters from this vast drainage basin, floods occur, increasing the depth and extent of marshes. Flood waters may increase depths by 1-1.5 m, with 2-3.5 m in more permanent basins. Most of the large marshes lie in Iraq, but the Hoveyzeh or Hawr-al-Azim marshes are on the border, and occupy 3000 sq km at high water. They are fed by the Karkheh and other rivers from Iran. Construction of the Karkheh Dam in Iran (and pipeline water transfer to Kuwait) will reduce input of water to this marsh, compounded by canal construction and draining of marshes in Iraq. Additionally, irrigation return waters will be salinised (Partow, 2001). A dam has been built by Iran across the Hoveyzeh marsh to retain water on the border with Iraq. Marsh temperatures range from 15°C in January to 31°C in August and fish may retreat to deeper areas or move upriver at the higher temperatures. Flooded marshes tend to be warmer than rivers in winter. The Shatt al Arab has temperatures of 32°C in July and 16°C in December.

Floods are often a feature of these southern rivers and some loss of fish stocks must occur as they recede. For example, the Jarrahi and Zohreh rivers overran their banks in November 1994 after torrential rains causing widespread flooding (http://netiran.com/news/IranNews/html/94112109INEV.html).

The Zagros Mountains consists of tightly packed ranges in the Tigris basin trending north-west to south-east. A trellis drainage pattern is imposed on this. The tangs, their formation and the drainage pattern are described by Harrison (1937) and Oberlander (1965; 1968a). These deep defiles may exceed 2400 m in depth with vertical walls of 300 m splitting anticlinal mountain ranges instead of taking apparently easier routes around their ends. They may well be barriers to the movement of less vagile fish species or a highway into the interior for those with some dispersal ability. Tangs formed because an antecedent drainage over lower relief was gradually uplifted at a rate slow enough to permit streams to cut through ridges and retain the original pattern of drainage once the softer material was washed out of the valleys between the anticlines.

The uppermost parts of the basin show evidence of headwater captures and this orogenic zone is very unstable. The divide between endo- and exo-rheic basins is not the snowline of the Zagros but is east of it, so streams must first cross the Zagros peaks to start on their journey to the Persian Gulf.

Springs are important in the mountains, tapping aquifers and helping to maintain river flow. The Karun River traditionally has its source in springs. Keivany et al. (1992) surveyed 72 springs in Chahar Mahall va Bakhtiari Province, in the upper Karun River basin, and found them suitable for trout culture with a potential production of about 6000 tonnes per year. Flows varied from 50 to 4000 l/second, temperature from 6 to 15°C, pH from 6.2 to 7.8, conductivity from 128 to 570 mMoh/cm, total alkalinity from 20-220 meq/l, total hardness from 140-250 mg/l, oxygen from 7 to 11 mg/l, carbon dioxide from 5 to 20 mg/l (falling rapidly to less than 2 mg/l within a few tens of metres of the spring source), H₂S 0 mg/l, Cl^- 1-28 mg/l, SO₄^-- 14-135 mg/l, PO₄^-- 0.1-0.3 mg/l, Ca⁺⁺ 16-82 mg/l, Mg⁺⁺ 3-34 mg/l, K⁺ 0.2-1.0 mg/l, Na⁺ 0.5-1.5 mg/l, Fe⁺⁺⁺ 0-0.06 mg/l, Fe⁺⁺ 0 mg/l, NO₂^- 0-0.2 mg/l, NO₃^- 0-13 mg/l, NH₄⁺ 0-0.5 mg/l, and HCO₃^- 48-220 mg/l. Springs (or sarabs) in Kermanshah Province have been described by Khatami and Shayegan (2003) and are regarded as a significant water supply for rivers. Sarabs are used for drinking water and irrigation, and are threatened by pollution and fish farms. Qanats are also found, in drier parts of the basin, but they are not as significant for fish habitat as in other parts of Iran.

Marshes and ponds as well as seasonally flooded arable land along the Karun River in the lowlands of Khuzestan provide temporary and permanent habitats for fishes. Some are reviewed below.

The "Hawr-e Bmdej"? or "Sadi Shavour" Marshes lie between the Karkheh and Dez rivers northwest of Ahvaz at 31°45'N, 48°36'E and encompass 12,000 ha. This is the most extensive permanent freshwater marsh with tall reeds of Phragmites and Typha in Khuzestan. There is relatively little open water. Some parts are being drained for agriculture, a continuing trend for marshes with concomitant loss of fish habitat. The "Hamidieh" plains at 31°20'N, 48°20'E comprise 20,000 ha of seasonally flooded (winter) plain and arable land along the Karkheh River. Hamidieh Lake, an old oxbow of the Karkheh, is included in this area and is permanent fresh water. The lake is 3 ha and has extensive reed beds.

The "Susangerd" Marshes or Hawr-e Susangerd at 31°45'N, 47°55'E are northwest of Ahvaz near the Iraqi border and form the extreme eastern edge of the Hawr-al-Azim, most of which lies in Iraq. The marshes occupy about 30,000 ha and are made up of permanent and seasonal fresh and brackish marshes and seasonally flooded arable land. The marshes are on the floodplain of the Karkheh River. Irrigation projects, grazing by livestock, reed cutting and fishing all occur here. Parts of the marsh were damaged by the Iran-Iraq War. The Iran-Iraq marshes declined in area from 1089 sq km to 758 sq km from 2000 to 2002 and was predicted to dry up in 5 years from 2002 because of the Karkheh Dam. Reports conflict since once the dam was full, a relatively normal flow regime would help maintain the marshes. Restocking with 490,000 Barbus sharpeyi and and B. xanthopterus took place in this marsh (www.shilat.com, downloaded 12May 2006).

The Shadegan Marshes at 30°20'N, 48°20'E occupy 282,500 ha (Jones, www.ramsar.org/lib_dir_2_3.htm, downloaded 4 April 2000) gives 296,000 ha) and form the southern part of the seasonal floodplain of the Dez, Karun and other rivers at the head of the Persian Gulf. There are adjacent tidal mudflats. The central and southern part of the marshes are part of a Ramsar Site, along with the mudflats (World Conservation Monitoring Centre, 1990). Sabzalizadeh and Amirineia (2003) give some physical and chemical characteristics of 5 sample stations in this marsh. Range of pH was 7.2-9.4, maximum water temperatures occurred in July and August. Maximum levels of dissolved oxygen were found in November and February but were more than 5 p.p.m in most cases, optimum for fish growth and reproduction. The water quality was hard and brackish. The whole area may dry out in late summer, a natural condition exacerbated by dams and irrigation schemes on the major inflowing rivers. In a November 2000 visit, much of this area was dry although it had been flooded in 1999. When the marsh dries, fish concentrate in the deeper pools where they are easily caught, even the smaller ones. The marsh is re-colonised from the rivers. The fishes of this marsh in order of abundance are kopur, shirbot, touyeni, esbele, binni, berzem, biah and very few himri and gattan (Y. Mayahi, pers. comm., 2000). Rice paddies occupy part of the Ramsar Site and reed cutting, fishing and grazing goes on. There is extensive reed cutting, some livestock grazing, some rice paddies and potential pollution from main roads, shipping and oil terminals. Over 100,000 ha were contaminated with oil from a leaking pipeline in 2000 and 35,000 cu m of refinery wastes were dumped in the marsh in 2004 (www.payvand.com, downloaded, 5 September 2006). Chemical weapon use occurred here in the Iran-Iraq War and acid rain fell from the burning of the Kuwaiti oilfields in the Gulf War. About 10% of the marshes were destroyed (Anonymous, 1988b; Scott, 1993; Jones, www.ramsar.org/lib_dir_2_3.htm, downloaded 4 April 2000). Fishing is important. The Shadegan Wildlife Refuge, encompassing 296,000 ha, is on the threatened list for National Parks since it was substantially damaged in the Iran-Iraq War, both physically and by chemical agents.

The principal fishes appearing on fish stalls in Ahvaz from marshes such as Shadegan are Barbus xanthopterus, Liza abu, Barbus sharpeyi and Cyprinus carpio as well as cultured Hypophthalmichthys molitrix as escapes or plantings. Farm ponds in Khuzestan have Barbus barbulus, Ctenopharyngodon idella, Hypophthalmichthys molitrix and Cyprinus carpio. Hawr-al-Azim, Hawr-al-Hoveyzeh and the Shadegan marshes are important refuges for fishes in Khuzestan (Korki, 1992; N. Najafpour, pers. comm., 1995). 490,000 fingerlings of Barbus sharpeyi and B. xanthopterus were stocked in this marsh in 2005, a 40% increase over the previous year (www.iranfisheries.net, downloaded 30 November 2005).

Izeh and "Shiekho" lakes at 31°52'N, 49°54'E occupy 1400 ha in the Zagros foothills. These small freshwater lakes are shallow with extensive sedge marshes. Izeh is the deeper of the two with much more open water. They are fed by runoff and springs. Shiekho, the larger lake, is almost overgrown with emergent vegetation except where cattle have grazed and trampled areas leaving some open water. Some fishing occurs in the lakes and water is abstracted for irrigation.

"Choghakor" or "Chaghakhour" Marsh or Wetland at 31°55'N, 50°54'E lies in upper Karun River drainage in the Zagros Mountains in Chahar Mahal va Bakhtiari at ca. 2100-2270 m and occupies 1600 ha. Maximum depth in spring and winter is 2 m but in summer it is almost entirely dry and overgrown with emergent vegetation. Construction of a dam may enable a more permanent marsh to exist (Taqvaie and Ramezani, 2002) although others consider dam construction to be a threat to the habitat and its diversity as the habitat changes from a wetland to a lake (Ebrahimi and Moshari, 2006). After the dam was built, water depth increased from 1.5 m to 6 m or more.

"Gandoman" Marsh at 31°50'N, 51°07'E at 2250 m and occupying 1500 ha (or 1200 ha, Khan et al. (1992) or 3510 ha Taqvaie and Ramezani (2002)) is a similar habitat but it has a stream running through it. "Sulegan" wetland or marsh in the same area encompasses 164 ha and is spring fed. These marshes have been proposed as a Ramsar Site although not yet formally designated (Scott and Smart, 1992).

The southern areas of this basin are areas with high temperatures and large cities (Abadan in Iran and Basrah in Iraq). Adjacent waters are highly polluted with sewage, agricultural waste and other chemicals (e.g. see DouAbul et al., 1988; Diagomanolin et al., 2004; Karamouz et al, 2005; Afkhami et al., 2007). The increased use of motor boats has led to oil pollution. DDT is still sprayed against malarial mosquitos on stagnant pools adjacent to the main river course leaving a brown stain on the rocks (observations in 1995; a letter of complaint to the appropriate agency carrying out this spraying by the Iranian Fisheries Research and Training Organization elicited no response). Scott (1995) records sale of Chloridrin, a persistent insecticide, to residents of the Hawr-al-Hoveyzeh in Iran as a means of poisoning large numbers of fish for sale. Phytoplankton blooms are common and in canals the chlorosity increases, transparency decreases and pH is reduced because of the dying plant material. The Shatt al Arab is more affected by physical factors as it is an estuary. Historical problems with salinisation of soils (and presumably water) extend back 5000 years in southern Mesopotamia including Khuzestan, a consequence of over-irrigation and inadequate drainage (Goldsmith and Hildyard, 1984). The irrigation systems rose and fell with the vicissitudes of history. There was a large-scale irrigation network in Khuzestan during the Sassanian period (A.D. 226-639), lost through conflict and natural disasters after this date and reconstructed in modern times (Adams, 1962).

A theory has been advanced that the silt-laden discharge of the Tigris-Euphrates-Karun rivers has built out a delta into the Persian Gulf. The head of the Gulf would have reached Baghdad and Samarra about 7000-6000 B.P. and since then the land area is supposed to have extended some 200 km southward. The present plains would not then have been as extensive and rivers from Iran would have entered directly into the Gulf. The Admiralty Naval Staff (1918), Mason et al. (1944), Adams (1962), Hansman (1978), Maltby (1994) and Lambeck (1996) provided illustrations of this recession of the head of the Persian Gulf in historic times along with details of historical and archaeological evidence. The sea coast was then supposedly as far inland as Ahvaz in Iran for example. Lees and Falcon (1952) proposed that in fact downwarping occurs under the weight of sediment. Certainly the silt load has not built up a land surface. The coastline, under this theory as interpreted by Fisher (1968), has been constant since the end of the Pliocene and presumably as a marsh habitat for fishes too. However Lees and Falcon did state that there were advances and retreats through historic and prehistoric time. Ionides (1954), Larsen (1975) and Nützel (1975) refuted Lees and Falcon and maintained that marine clays and silts indicate a marine embayment as far inland as Amara in Iraq (31°50'N, 47°09'E) and that the third millennium cities of Ur and Eridu have left cuneiform sources placing them on the sea although now they are 100 km from the head of the Persian Gulf. Lees and Falcon did not take into account sea level changes such as the postglacial rise of 100 m and interglacial rises of 30-100 m. Active growth of a delta at the head of the Gulf over the last 20,000 years may only have occurred from 10,000 to 2000 B.P. and again in the last 300 years. Subsidence levels are probably not as great as postulated (Vita-Finzi, 1978). Nevertheless, there were probably marshes to the north and they may have just become more available and extensive in recent centuries (Aqrawi, 2001). As Larsen and Evans (1978) and Wagstaff (1985) point out, the Persian Gulf shoreline at the head of the Gulf has been affected by, and rendered difficult to interpret by, a complex of factors including confusion of marine and freshwater fossils in an estuarine environment, subsidence, eustatic sea level fluctuations, local seismic activity, climate and therefore hydrologic changes, and cultural changes such as irrigation. Jacobsen (1960) detailed some of the changes in the courses of rivers and canals, based on evidence of ancient settlements which were presumed to be linearly arranged along water courses. Mallowan (1964) also maps some ancient river courses. The fish fauna has evidently had to cope with a changing availability of habitat through the post-glacial period. Floods and changes in river courses over this time have no doubt facilitated movement of fishes between Iran and the Tigris-Euphrates basin. It seems unlikely that the separate entry of rivers from Iran into the Gulf would have led to isolation of the faunas to any significant degree.

Canals and other irrigation structures have long been a feature of the Mesopotamian plains, forming habitats for fishes dating back thousands of years (Bagley, 1976). Their loss through natural and man-made disasters must have affected fish populations but sufficient natural habitat no doubt remained to ensure survival. The construction of dams upstream in Turkey and the large scale, modern drainage programmes in Iraq bordering Iran such as the "Three River Project" are drying up the extensive marsh systems and these are regarded as an eco-disaster leading to desertification in Iraq and adjacent regions of Iran (North, 1993; Pearce, 1993, 2001; Ryan, 1994; National Geographic, 185(4):unnumbered page, 1994; Scott, 1995; Munro and Touron, 1997; Maltby, 1999; Partow, 2001; www.amarappeal.com/documents/Draft_Report.pdf, downloaded 15 November 2001). The 32 km long "Fish Lake" was constructed as a barrier to Iranian attacks on Basrah. The Iranians dug several drainage ditches from "Fish Lake" northeast of Basrah to the Karun River, to dry up land for infantry attacks on Basrah. This whole marsh area of about 17,000 sq km, is the most important wetland in the Middle East and one of the top ten in the world. The Central and Al-Hammar marshes in Iraq by 2001 have had 97% and 94% of their land converted into bare ground and salt crusts. Less than one-third of the Hawr al Hawizeh (= Hawr al Azim in Iran) survives. It was estimated in the 1990s that the marsh area would be a desert within a decade and this seems to be an accurate assessment. The effects on the fishes in Iran are unknown but much habitat is being lost which could have served as a reserve against loss in Iran through natural and man-made changes.

The Iran-Iraq War of 1980-1988 severely damaged the Hawr-al-Hoveyzeh in Iraq, and presumably to some extent in Iran. Bombs and shells, chemical weapons, pollution, burning of reed beds, reed cutting and armoured boats used to smash through obstructing reeds all had deleterious effects (Scott, 1995). The Iraqi shores of this hawr have been drained by dyke construction and river control presumably for military reasons in this border area. Some marsh will survive in Iran because it is fed from wholly Iranian rivers but Iran News (19 February 1995) reports that draining of Iraqi marshes will lead to desertification inside Iran.

The Southeast Anatolia Project (known as GAP after its Turkish acronym) incorporates 21 dams and 19 hydroelectric facilities including the massive Ataturk Dam on the Euphrates completed in 1993. It plans to draw off one-third of the waters originating in Turkey and will also use water from the Tigris River (Ottawa Citizen, 10 November 1994; Morris, 1992; Biswas, 1994; Beaumont, 1998). The reduction in flow for Iraq may reach 60%, especially when water is taken from the Euphrates or ath-Thawrah Dam (its reservoir is Lake Assad) at Tabqa in Syria (Vesiland, 1993). This will have major downstream effects, less so in Iran than in Syria and Iraq, but flow into the Shatt al Arab shared between Iran and Iraq will be greatly decreased perhaps allowing greater penetration of saline water and restricting migrations of fishes.

Between 20 and 15 thousand years ago, the Persian Gulf was dry as water was locked up in ice-caps and sea level was 110-120 m lower than today (Sarnthein, 1972; Kassler, 1973; Nützel, 1975; Al-Sayari and Zötl, 1978; Vita-Finzi, 1978). The floor of the Gulf was then thought to be a generally waterless, flat depression with a few swampy tracts rather than a "Garden of Eden" as has been proposed. A marine transgression occurred between 12 to 8 thousand years ago and by 6 thousand years ago the present sea-level was attained. Streams now isolated from the Tigris River basin by the sea in the Gulf and Hormozgan basins would have been tributary to an extended Shatt al Arab, extending 800 km down the gulf to form an estuary at the shelf margin in the Sea of Oman, now under 110 m of sea. Earlier regressions no doubt occurred and facilitated the movement of fishes.

Construction of fish farms is widespread throughout this basin in Iran. For example in Lorestan Province, 772 tonnes were produced by the Lorestan Province Fishery Company in 1997, 50 fish farms were under construction and 125 pools built for aquaculture uses. The long-term aim was to increase fish production to 20,000 tonnes worth 156 billion rials and employing 10,000 people (Tehran Times, 22 September 1998). In Chahar Mahall va Bakhtiari Province in the highlands of this basin, 4360 tons of trout fingerlings were produced with plans to produce 8000 tons in future years (Tehran Times, 14 March 2005). The Indian carps Cirrhinus mrigala, Labeo rohita and Catla catla are being reared in aquaculture stations and are potential escapees into the natural environment (Gilkolaei, 2007).

Berg (1940) places this basin in the Mesopotamian Transitional Region, since the boundaries of three zoogeographical regions meet here, namely the Holarctic (i.e. its Palaearctic part), Sino-Indian (= Oriental) and the African (= Ethiopian). The Mesopotamian Transitional Region includes the Tigris and Euphrates basins and the Quwayq River, Syria, forming a single Mesopotamian Province. The province is transitional between the Mediterranean Subregion and the Indian Subregion. Genera such as Leuciscus (= Squalius) Aspius, Chondrostoma and Alburnus point to a Mediterranean or European association while such genera as Glyptothorax, Barilius, Mystus and Mastacembelus point to an Indian association.

Top