Cover Image
close this bookIrrigation-Induced Salinity - A Growing Problem for Development and the Environment (World Bank, 1993)
close this folderGlobal magnitude of irrigation-induced salinity
View the document(introduction...)
View the documentA global survey of affected areas
View the documentThe world bank experience

(introduction...)

Salinity buildup is a long degenerative process; initial manifestations may take as long as 15 or more years to appear after the introduction of irrigation. Because the advent of salinity, to the level that is clearly injurious to plant growth, often takes years, monitoring for its presence is often neglected by irrigation authorities. The general lack of monitoring of salinity-affected areas is clearly evidenced by the dearth of field level information on this topic. This section attempts review the available information in the extent of irrigation-induced salinity problems worldwide.

A global survey of affected areas

A compilation of publicly available data on areas affected by irrigation-induced salinity in selected countries is presented in Table 3.1. It should be noted that these figures cover various years and since most of the available figures apply to the early to mid-80's, the extent of the problem may changed thereafter. Nonetheless, the FAO report, An International Action Programme on Water and Sustainable Agricultural Development, A Strategy for the Implementation of the Mar del Plata Action Plan of the 1990s (1990, p. 15) estimates that "...about 20 to 30 million hectares are severely affected by salinity and an additional 60 to 80 million hectares are affected to some extent.. Estimates of the area affected have ranged from 10 to 48 percent of total irrigated area. Clearly, the countries in the arid and semi-arid regions that ranked high in irrigation investments have extensive salinity problems: salinity-affected areas as a percentage total irrigated area amount to 11 percent in India, 21 percent in Pakisan, 10 percent in Mexico, 23 percent in China, and 28 percent in the United States. In the case of India, for example, 11 percent translates to 4.7 million ha, and given the generally small farm sizes, this translates to thousands of farm households: 100,000 households assuming an average farm size of 5 ha. The newly-formed Central Asian Republics in the Aral Basin-Kazakhstan (17%), Turkemenistan (48%), and Uzbekistan (24%)-also exhibit serious salinity problems.


Table 3.1: Cultivated, irrigated and saline areas in selected countries


Table 3.1: Cultivated, irrigated and saline areas in selected countries (continued)

As of 1985, 10 percent of irrigated area in Mexico is affected by salinity and approximately 64 percent of the affected area has salinity levels greater than 8.1 dS/m (Table 3.2). Over 80 percent of the affected areas is located in the northern regions of Mexico (Table 3.3). In California, Backlund and Hoppes (1984) found that most of the salt affected areas were situated in the San Joaquim Valley and totalled 890,000 ha or 22% of total irrigated area (Table 3.4).

In the People's Republic of China, the extent of irrigation-induced salinity and its causal factors varied by region (International Program for Technology Research in Irrigation and Drainage [IPTRID] 1992). In the north, most of the salinity problems are found in the Huang-Huai-Hai Plain or North China Plain. These problems began to be manifested in the fifties when large scale irrigation was introduced in the area without providing for adequate drainage to remove the excess water. The areas with improved drainage have already reduced the salinity problem and at present, it is mostly restricted to low lying areas underlain by saline groundwater at shallow depth. Total affected cultivated areas in North China in 1991 is estimated at 2.1 million ha. In North-East China, the total salt-affected area is estimated at 6.0 million ha in 1991. Soil salinity became a serious problem due to indiscriminate reclamation and over-grazing of natural pasture. The introduction of irrigation without providing for adequate canal seepage control and drainage led to large scale waterlogging and capillary salinization of the upper soil layers. The western section of the North-East Plain is also seriously affected. Similarly, irrigation development without adequate drainage provision resulted in significant groundwater table rise and to capillary salinization in North-West China. The area affected in the North-West is estimated at 3.0 million in 1991.

Table 3.2: Land area subject to irrigation-induced salinity in Mexico, 1985

SALINITY LEVEL

SALINITY AFFECTED AREA

PERCENT OF TOTAL

(dS/m)

(ha)

(%)

4.0 to 8.0

180,000

36

8.1 to 12.0

135,000

27

12.1 to 16.0

85,000

17

16.1 to 20.0

45,000

9

Greater than 20.0

55,000

11

Total

500,000

100

Source: Aceves-Navarro 1985.

Table 3.3: Regional distribution of salinity affected areas in Mexico, 1985

REGION

AFFECTED AREA(ha)

PERCENT(%)

Northwest

250,000

50

North

80,000

16

Northeast

75,000

15

Central

90,000

18

Southeast

5,000

1

Total

500,000

100

Source: Aceves-Navarro 1985.

A survey conducted by Pakistan's Water and Power Development Authority of salinity-affected areas in the Punjab, Sind, NWFP, and Baluchistan in 1981 revealed that 31 percent of salinity-affected areas exhibits strong salinity, 25 percent has moderate salinity, and 44 percent has slight salinity (Table 3.5). Ahmad and Kutcher (1992) used the Indus Basin Model Revised (IBMR), which models the groundwater and salt flows in the Indus Basin, to project salt accumulation in the Punjab and Sind regions in Pakistan in both the fresh and saline areas by the year 1995.9 The study estimated that the salt added to the groundwater in the fresh water areas by 1995 will amount to 1.16 mt per ha in Punjab and 1.95 mt per ha in Sind. In addition, the amount of salt added to the soil will total 0.10 and 0.30 mt per ha

Table 3.4: Salinity and drainage problems by major irrigated areas in California, 1984 (000 ha)

AREA

REGION

Irrigated

Saline

Hi. Water Table

Water Quality

San Joaquin Valley

2,300

890

610

930

Sacramento Valley

850

80

160

120

Imperial Valley

200

80

200

200

Other Areas

770

120

120

120

Total

4,120

1,170

1,090

1,370

Source: Backlund and Hoppes 1984.

Table 3.5: Extent of salinity in Indus Basin, Pakistan, 1981

SURFACE SALINITY


Slight

Moderate

Strong

Total

PROVINCE

(000 ha)

(%)

(000 ha)

(%)

(000 ha)

(%)

(000 ha)

(%)

Punjab

698

50.0

399

28.6

299

21.4

1,396

100

Sind

1,028

39.6

595

22.9

974

37.5

2,597

100

NWFP

50

69.4

11

15.3

11

15.3

72

100

Baluchistan

60

65.2

18

19.6

14

15.2

92

100

Source: Survey and Research Organization, Planning Division, Water and Power Development Authority 1981 (cited by the Economic and Social Commission for Asia and the Pacific 1989). in Punjab and Sind respectively. In the saline areas, 0.30 and 1.24 mt per ha will be added to the ground water and 0.54 and 0.84 mt per ha will be added to the soil in Punjab and Sind respectively.

In India, the salt-affected area in the country as reported by the Central Soil Salinity Research Institute (CSSRI), Karnal, India as of 1991 was 7 million hectares of which 2.4 million ha were inland saline soils of the arid and semi-arid regions, 2.5 million ha were alkali soils of the Indo-Gangetic plains and 2.1 million ha were coastal saline soils. This amounts to 2.3 percent of total geographical area or about 4 percent of total cultivable land. The leaching of the saline dessert soils by irrigation was the major cause of the secondary (capillary) salinization (Smedema 1990a).

As an indication of the magnitude of salinity incidence at the field level, Table 3.6 presents some figures on waterlogged and salinized areas in 10 command sites in India. It should be noted, however, that the project areas overlap to a certain extent. Furthermore, due to the lack of appropriate information, it is also not known whether all waterlogging and salinization are irrigation-induced. Smedema (1990a) notes that irrigation-induced waterlogging and salinization in India are spreading. This conclusion is supported by information from the field and by the fact that the recharge of groundwater in the command areas from seepage and normal deep percolation from irrigation continues mostly unabated. Although the spread of waterlogging and salinity is monitored in some command areas, no statistics at the state and national levels are available.

Table 3.6: Incidence of waterlogging and salinity In selected irrigation command areas In India, 1990

AREA (000 HA)

PROJECT AREA

STATE

Waterlogged

Saline

Shards Sahayak

Uttar Pradesh

7.0

6.7*

Ran Gangaa

Uttar Pradesh

9.7

17.6*

Gandak

Bihar

20.1

-

Rama Sagar

Andhra Pradesh

27.6

0.8

Tungabhadra

Andhra Pradesh, Karnataka

1.3

6.7

Ukai Kakarpar

Gujarat

4.3

2.2

Mahi Kadana

Rajasthan, Gujarat

16.8

7.3

Chambal

Madhya Pradesh, Rajasthan

20.3

8.2

Tawa

Madbya Pradesh

-

3.8

Rajasthan Canal

Rajasthan

8.0

5.4

Note: * - soil is highly alkaline. Source: Director, Central Soil Salinity Research Institute Karnal, Haryana, cited In Smedema 1990a.

IPIRID (forthcoming) reports that soils affected by salinity represent close to 20 percent of irrigable land in the arid Tifalalet and Ouarzazate areas of southern Morocco. The climatic (hot and dry) and soil conditions (shallow, saline water table, and salt bearing land) and the external salt addition (excessive application of irrigation water) are factors that have contributed to the problem. Furthermore, traditions relating to water rights favored upstream users over those downstream, and the return water naturally drained from the upstream areas resulted in increasing salinity of irrigation water and accumulating concentrations of salt in the aquifers downstream. Similarly, increasing accumulation of salts have been observed in the Moulouya area (Triffa and Bou Areg). The IPTRID report also noted that in Algeria, 1.1 million ha are affected by salinity, whether due to poor management of the saline groundwater or due to the use of saline irrigation water.

In many cases, irrigation has resulted in rising water tables that subsequently contributed to salinity problems. Table 3.7 presents a list of irrigation projects compiled by Smedema (1990b) which contributed to rising water tables. In the most extreme case, the water table rose over 3 meters per year.

The world bank experience

The Operations Evaluation Division (QED) of the World Bank conducts impact evaluation studies by revisiting Bank-supported operations some five to twelve years after completion. As instruments of evaluation, these studies are uniquely placed to measure trends and effects that are outside the purview of audits undertaken during the time of completion. They examine issues such as the social impact of fairly large development projects, the distributional consequences (i.e. poverty alleviation) of the design and implementation, the condition of the infrastructure created under the operation, the experience with the production of goods and services aided by the operation, the long term performance of the implementing organizations, and the ultimate effects of the project operation on the environment and natural resources.

In 1989, OED reviewed 21 World Bank projects that were approved between 1961 and 1978 and completed in 1970-86. These comprised medium- and large-scale public irrigation systems which were typical of the Bank's lending policy for irrigation projects in the 1960s and 1970s. All were designed to: (1) raise food production to meet the demand of growing populations, and (2) earn or save muchneeded foreign exchange by reducing the need for food imports and/or creating surplus food for export. Most of the projects also aimed to reduce rural-urban migration by creating additional employment opportunities and increasing the incomes of the project beneficiaries.

Table 3.7: Irrigation induced watertable rise In selected projects

WATER TABLE

IRRIGATION PROJECT

COUNTRY

Original Depth (meters)

Increase(mters/yr)

Bhatinda, Punjab

India

40-50

0.45

Pre-SCARP 1 area

Pakistan

15-30

0.30-0.50

Pre-SCARP VI area

Pakistan

10-15

0.20-0.40

Khaipur Command, Sind

Pakistan

4-10

0.10-0.30

State Farm 2°, Xinjang

China

5-10

0.35-0.70

Murray-Darling Basin,




New South Hales

Australia

30-40

0.50-1.50

Noubaria, Western Desert

Egypt

15-20

2.00-3.00

East Ghor, Jordan Valley

Jordan

10-15

nil

Beni Amir

Morocco

15-30

1.50-3.00

Gezira Scheme

Sudan

20-50

nil

Salt Valley, Arizona

USA

15

0.60

Amibara, Hiddle Ahwaz Valley Ethiopia 10-15 1.00
Source: Smedema 1990b.

The OED evaluation found that more than half the 21 projects were very successful but had some degree of adverse impact on the enviromnent (World Bank 1991b). Increasing waterlogging problems were found in 11 projects (Table 3.8). Soil salinity problems were found in four (Pyongtaek-Kumgang Irrigation Project in Korea, the Seyhan Irrigation Project in Turkey, the San Lorenzo Project Irrigation and Land Settlement project in Peru and the Rio Sinaloa Project in Mexico) and were caused mostly by poor drainage. The drainage network was insufficient or incomplete in half of the 21 projects. The most striking cases were the Sinaloa project in Mexico and the San Lorenzo project in Peru where 17 and 20 percent of the project area respectively were uncultivable at the time of the impact evaluation. In the Sinaloa Project, the impact evaluation found a typical case of damage due to poor drainage; the area affected by salinity and waterlogging increased from about 800 ha in 1973 to about 11,000 ha in 1987. About 850 families were found to have incurred serious economic losses as a result.


Table 3.8: OED irrigation and drainage performance ratings for selected projects, 1989

Despite the knowledge gained through the centuries regarding the causes of soil salinity and the availability of technologies to prevent or reverse its deleterious effects, salinity remains a serious problem in present day irrigated areas. The following section examines the economic and political factors that, deliberate or otherwise, foster the advent of salinity.