Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Chapter 22.On the problem of water supply in the Hai-Luan plain

Huang Rangtang
Commission for Integrated Survey of Natural Resources, Academia Sinica, Beijing

Is THE PLAIN of the Hai He and Luan He short of water? Do we need to transfer Chang Jiang water into the plain? How can we optimize the use of local water resources? These are important and controversial questions which we are all concerned about. The following views are based on my own field work and an investigation of relevant documentary materials.


Water Sources Cannot Satisfy Agricultural Requirements

The 126,000 km Hai-Luan Plain lies between the Huang He in the south, the Taihang Shan in the west, the Yan Shan in the north and the Bo Hai in the east, at an altitude of 0 to 100 m above sea level. Though this region has a long history of cultivation, the level of agricultural production has consistently been very low. Before 1949 the grain yield was little more than 0.75 t/ha. Since then there has been a certain amount of improvement, but yields still fluctuate between 1.5 and 2.25 t/ha, The main reason for these low yields is water deficiency.

Light and heat conditions are adequate for the growth of many crop varieties and in most areas allow two crops to be reaped in a year. The land is flat and contiguous in large tracts, highly suited for mechanized cultivation. The only deficiency is in moisture.

First of all, precipitation is scant. The average annual precipitation varies from 400 mm in the vicinity of Hengshui to 600 mm in the piedmont and coastal areas. Moreover, only a small proportion of the catchment areas of the plain's rivers lies in the mountainous areas, so the inflow from this source is not very significant. The mean annual precipitation on the plain is 73.5 km and the runoff only 7.4 km. The mountainous areas provide an additional runoff of 20.95 km (from a precipitation of 104 km). Runoff per cultivated hectare is only 2,775 m in the basin, 10 per cent of the national average.

Second, precipitation is uneven, both seasonally and from year to year. (For specifics, see Wei Zhongyi and Zhao Chunian, Chapter 7).

Third, potential evaporation is high, (See Cheng Weixin, Chapter 19). Runoff depth is 50 mm in most of the plain and only 10 to 25 mm in the central part. The runoff coefficient is 0.101, far less than the national average of 0.452 for exterior drainage basins.

The insufficiency and variability of available moisture combine with the flat topography (1:10,000 to 1:20,000) to produce alternating drought, flooding and salinization, which are the basic causes of low yields. At present, virtually every locality on the plain which has good water control also has high and stable agricultural yields. For example, Quzhonzhuang Village was a saline area in the past where grain yields were only a few hundred kg per hectare. After land shaping and soil improvement centering on water control, especially well irrigation and drainage, the salts were leached out of the soil and drained off, causing output to increase steadily until now grain yields have reached 7.5 t/ha. There are numerous examples like this.

Water Sources Cannot Satisfy Industrial and Municipal Requirements

Mineral resources are abundant in the Hai-Luan plain, including the Fenggeng, Jingxing,Jingxi and Tangshan coalfields, as well as the Datong, Shnoxian and Yangchuan fields in eastern Shanxi whose water use is related to that on the plain; the oilfields of Shengli, Dagang, Renqiu and Bo Hair and the iron mines of Qian'an, Luanxian, Xingtai and Handan. The exploitation of these resources furnishes favourable conditions for the development of thermal power generation, coal chemical industry, oil refining, petrochemicals and iron and steel smelting, but these industries are relatively large consumers of water. The required investment in water sources is therefore quite large for factories on the Hai-Luan Plain such as the newly built petrochemical base in Tianjin or the Dongfanghong petrochemical plant in Beijing. The lack of water sources often makes it difficult to come to a decision even on industrial siting. For example, the construction of a power station has been considered at the pit entrance of the Datong mine to use its coal as a means of solving the problem of power shortages in the Beijing-Tianjin-Tangshan system. The project has been postponed time and again because of the inability to fix the water sources and because of conflicts with agricultural water use.

Water Sources Cannot Satisfy Economic Requirements

The area under irrigation in the Hai-Luan Plain has reached 4.4 x 106 ha, 52 per cent of the arable land. Water supply is insufficient in most of the irrigation districts, and some can only be irrigated once or twice a year. Great changes have occurred in surface runoff and groundwater conditions in many areas because of large-scale diversions of surface water and excessive extractions of groundwater. For example, in 1975 Hebei Province extracted nearly 9 km of groundwater. Although an increasing number of wells have been dug since, the total yield has not increased and the water tables of the aquifers in many places are being lowered year by year because the rate of extraction has exceeded the recharge. Lowering of the water tables was registered in 22 places in the province in 1975, occupying an area of 11,600 km. In some cases, this has developed very quickly.

The overexploitation of the deep aquifers has brought with it a number of problems. First, because of the lowering of the water table, per-well discharge has declined. Not only has the efficiency of irrigation fallen but irrigation costs have increased due to the increase in pumping lift. Second, the original pumping implements must be replaced because they do not meet the new requirements. In some cases, pumps have had to be replaced twice. Third, with the drop in the water table a number of old wells become dry each year and new replacement wells have to be drilled. In some regions and some years the number of abandoned wells exceeds that of new ones. Fourth, increasing mineralization has led to deterioration in water quality. In littoral areas, salt water intrusion can take place if the water table is too low. Fifth, there is land subsidence.

Tianjin is located in the lower- reaches of the Hai He, and has serious water supply problems. In the past, the municipality's inland navigation was highly developed. The Wei Canal, Ziya He and Daqing He provided important links between Tianjin and Baoding, Handan and other municipalities in Hebei Province. The coal produced at the Fengfeng and Jingxing mines could be transported directly by river to Tianjin.

At present, however, the situation is quite different. Water is scant or absent in some reaches of those rivers. In other reaches locks have been built to store water, forming reservoirs. Not only is navigation impossible, but even the supply of several hundred million cubic metres of municipal water is hard to assure.

One after another the Guanting, Wangkuai, Xidayang, GangnanHuangbizhuang and Yucheng reservoirs, which supplied water to Tianjin in the early 1960s, have stopped doing so as the result of increases in water use in the upper and middle reaches of the Hai He. Water supply from the remaining reservoir at Miyun is unreliable. Moreover, the flow loss is very serious, and water actually delivered to Tianjin is often less than one-third that released from the reservoir. At present, per capita domestic water use averages only 60+ litres per day, one-third that of Shanghai. As a result of the severe 1980 north China drought, Tianjin's water shortage during the following winter was even more serious. Industrial water use was affected for months and the situation in agriculture was even more strained.

In the 1950s Tianjin used the water of the Hai He to grow about 60,000 ha of paddy rice. At that time 82 per cent of grain output was rice and Tianjin's xinozhan variety was well-known throughout the country. Beginning in the early 1960s, however, the insecurity of water sources forced a changeover to upland crops and the area sown to paddy rice was reduced repeatedly. By the beginning of the 1970s, paddy rice could only be assured on about 700 ha. Because of the absence of water resources, Tianjin has been forced to continue to overexploit the deep aquifers within the municipality, and land subsidence has continued to develop as a result.

Conditions in Beijing are quite similar. Following the rapid growth in industry and agriculture and the expansion of urban construction, water use has increased daily. The supply of running water in the municipality and that of self-provided water sources in 1978 were 45 and 100 times the respective figures for 1949. At present industrial, agricultural, municipal and domestic water use in the municipality totals nearly 4.8 km/annum. The excessive exploitation of the groundwater that has occurred in Hebei and Tianjin has also appeared to varying degrees in Beijing. In the case of surface water, large reservoirs have been built at Guanting and Miyun over the past three decades and the runoff of the mountainous areas has been basically controlled. At present the reservoirs are responsible for a very large portion of Beijing's industrial water, but the flow into some of them has decreased due to increases in water use in the upper reaches. Although the inflow of the Miyuan reservoir is relatively stable, it supplies Tianjin and Hebei as well as Beijing, and the resultant conflicts are extremely acute.

In a word, the Hai-Luan Plain lacks water. The conflicts between supply and requirements are becoming ever more serious. The solution of this problem deserves our serious consideration.


This conclusion is drawn from the following two aspects:

The Necessity of Chang Jiang Water Transfer

From the viewpoint of economic development in the Hai-Luan Plain, the shortage of water is beyond any doubt. It would be difficult to satisfy requirements even with best use of local water sources, including the Huang He. During dry years, the Plain's water resources have already reached a high degree of exploitation. The average flow into the sea of the Hai He during 1950-1972 was 9.24 km/annum. Within this period, however, the average flow for 1960- 1972 was only 6.65 km/annum. The water flowing into the Hai He in the mountainous areas was similar, 9 km, in both 1952 and 1968, but the flow into the sea in 1968 was only 0.347 km, less than 7 per cent of the 1952 figure of 5.07 km. This is due to the higher level of utilization of the water of the Hai in the later year.

The degree of utilization is much higher in dry years such as 1968, when the flow into the sea was only 3.86 per cent of the amount produced in the mountains, leaving very little potential. Thus future water source development will mainly be a matter of storing and utilizing the runoff of average and wet years. This is not a simple matter. Since the maximum discharge into the sea from the Hai basin was 33.7 km in 1963, nearly a hundred times that of 1968, and since wet years occur only once every several years and sometimes at intervals exceeding a decade, extensive evaporation losses may preclude using one year's water for a number of years even if there were sufficient storage capacity.

In addition, it is necessary to maintain a certain amount of river runoff to carry the salts in the soil of the plain into the sea and to prevent silt accumulation in the river channel and estuary which would affect flood paths. Nutrients required by littoral aquatic life could vanish if the runoff is exhausted. Consequently, both the comprehensive control of flood, drought and salinization of the plain and ecological equilibrium require a certain portion of water to be discharged into the sea. This means that it will not be long before the limits are reached in the exploitation of local water resources. The northward transfer of Chang Jiang water must be carried out as soon as possible.

The Formidable Nature of Chang Jiang Transfer

Although the development of water use on the Hai-Luan Plain and the natural conditions for diversion make it both necessary and feasible to transfer the water of the Chang Jiang, the formidable nature of the project should never be underestimated because the conditions of the Hai-Luan Plain, sited on the final half of either the Middle Route or the East Route, are considerably more complex than those south of the Huang He.

First, the project would be expensive. The East Route would divert water 660 to 1,150 km and require a pump lift of 65 to 70 m, making unit project investment costs far greater than for diversion into northern Jiangsu. At present, conveyance system investment costs on the Hai-Luan Plain are about 7 yuan per irrigated hectare. To combat salinity mainly requires constructing a drainage system which corresponds with that of irrigation. On the Plain, project investment for drainage is not much less than for irrigation. Calculating on the basis of the project magnitude as roughly estimated by the relevant departments, the total investment for the East Route, including the main canal, pumping stations, Huang He crossing and conveyance systems in the irrigation districts, will be 10 to 12 x 109 yuan for water transfer alone, of which two-thirds would be for the Hai-Luan Plain.

Investment on the Middle Route would also be quite high, even though the water would flow by gravity and pumping equipment would not be used, because the diversion works would be more complex than on the East Route. A new channel would have to be dug virtually the entire route from the Danjiangkou Reservoir to Beijing and would have to cross 167 rivers with catchment basins over 200 km each, including the main rivers and tributaries of the Han Shui, Huai He, Huang He and Hai He. Investment on the main channel would be greater than on the East Route as would be the amount of land occupied. If surface irrigation canal systems and salinity control projects are included, total investment would be at least as high as on the East Route.

In terms of the present economic strength of China, this project is an enormous investment item, requiring close to the country's total capital investment in water control over the past five years. Total agricultural income is still quite low, so it is impossible in the short term to rely on the accumulated funds of the rural collectives (people's communes, production brigades and production teams) to build the surface conveyance systems for the northward transfer project. Likewise, it is impossible to rely on state investment in the near future. On the one hand, a certain ratio must be maintained between capital investment in water control and in other sectors; in addition, a certain ratio must be maintained between capital investments in water control between the Hai-Luan Plain and other parts of the country. Clearly, northward water transfer cannot occupy too much of the state's investment.

Second, the returns to investment are not striking. The northward transfer of Chang Jiang water is mainly to serve agriculture, which is at present chiefly under collective ownership. The operating expenses of the transfer project must be recompensed in the form of water fees paid to higher levels. These payments would be obtained from their agricultural and subsidiary produce by beneficiary units such as communes and brigades. Even though agricultural output may require irrigation by the water of the Chang Jiang, if the water fees are too high, higher yields will not elicit higher income and the collective units will not be able to use the water. The operating costs of diverting the water of the Chang Jiang to the northern bank of the Huang He will be 0.020 to 0.025 yuan/m. North of the Huang He, the average additional cost of transporting water to the consumer, including 30 per cent loss due to evaporation and seepage from the reservoirs and canals, will be an additional 0.020 to 0.025 yuan/m, making a total of 0.04 to 0.05 yuan/m. The cost of water transfer increases if we also include deductions from profits of enterprises built with bank loans. This estimate yields a rather high water fee, over 10 times greater than the price of water paid at present in the surface water irrigation districts of the Hai-Luan Plain.

The existing state-set prices for agricultural produce and by-products are relatively low. This would make it very difficult to have a net increase in income from irrigating with interbasin water. Certainly water transfer is necessary to alleviate shortage on the Hai-Luan Plain, but if the project is too large and the costs too high, the water so diverted will not necessarily find a market.

Third, there is a significant danger of secondary salinization. The success or failure of the proposed transfers depends on whether they aggravate the danger of salinization. For a time in 1958, drought was fought by damming rivers, building large numbers of plains reservoirs and diverting the water of the Huang He. This resulted in widespread secondary salinization. Since then, as the result of many years of treatments the saline area has been greatly reduced. To date, however, the situation remains quite unstable and in a state of frail equilibrium.

The threat of salinization would become stronger with a future mass transfer of Chang Jiang water and the utilization of lowlands for storage. Of course, secondary salinization is not an inevitable consequence of water diversion for irrigation. Theoretically, it can be averted by combining drainage with irrigation and using both wells and canals so as to control the water table below the critical level. But at present there are a number of problems with this in practice, such as the lack of complete conveyance systems to the fields, inadequate management and the non-implementation of water control policies. Some of these problems involve the economic capabilities of the state and some await a reform in the management institutions of society. The transfer of Chang Jiang water would have some of the most complex management problems of any irrigation project in the world. In addition to requiring a series of measures to prevent salinization on the project itself and a set of effective management systems for water sources, the mass diversion has no assurance of being realized until the comprehensive control of flooding, drought and salinization on the plain in general has reached a certain level.

Therefore, since the northward transfer of Chang Jiang water is both necessary and feasible, scientific research, surveying and designing should continue to be carried out actively so that the project can be realized smoothly as soon as China's economic conditions have matured sufficiently. On the other hand, because of the enormous magnitude of the project and the difficulty in carrying it out in the near future, the problem of water shortage on the Hai-Luan Plain must be solved by tapping local resources and making the best use of them.


Although the potential of water resources in the Hai-Luan Plain, including the Hai, Luan and Huang He, is not great in a dry year, in normal years several km are still untapped. While the conditions for their exploitation are not favourable, perunit capital investment and operating costs are not likely to be greater than interbasin water transfer. Moreover, the scope of the projects is smaller and their implementation easier than diversion from the Chang Jiang. In particular, some projects can yield additional benefits in preventing flood, excess surface water and salinization. Priority should be given to exploiting the water resources in normal years, along the following lines.

Fully Intercept Mountain Runoff

Although reservoirs control 60 per cent of the mountain inflow area, this is not enough. For example, in Hebei Province the runoff from the mountains averages 20.95 km/annum, yet the total capacity of existing reservoirs is only 9.37 km/annum. Because precipitation is concentrated and varies greatly from year to year, and meteorological forecasts are not accurate, there is a tradeoff between flood control, the primary purpose of the reservoirs, and water storage.

Actively Promote Underground Storage

Water produced on the Hai-Luan Plain constitutes about one-third of the drainage into the sea in the Hai-Luan basin. This water source has not been effectively used so far. Utilizing the aquifers of the plain for storage is one means of regulating the water. In actuality, pumping water from shallow wells for irrigation may lower the phreatic water table, vacating the pores of the soil and sand layers so they may absorb rain and river water during the flood season. This method of storage can not only increase the utilizable water sources but also help prevent flooding and salinization. Originally the plain in front of the mountains in Hebei was subject to frequent flooding due to rainfall, but in the past dozen or so years the development of well irrigation has brought about a major transformation in the runoff discharge situation and greatly reduced the area of flooding due to rain. In general, the phreatic water table has fallen from 1.5 to 2.5 m to about 8 m and the drainage standard of ditches and canals has increased from a 5-year event to a 10 to 20 year event.

In addition, underground water storage can also cut losses through evaporation. It is easy to see the desirability of this form of storage (see Wu Chen and Wu Jinxiang, Chapter 21).

Build Plains Reservoirs Suited to Local Conditions

In central Hebei there are over 40 major lowlands each exceeding 10,000 mou (667 ha) in area. In most of the counties of the Tianjin-Cangzhou-Baoding plain, 20 to 30 per cent of the cultivated area is made up of lowlands prone to flooding. In recent years an average of more than 70,000 ha (1 x 106 mou) of arable land in Hebei have been inundated each year. Most of this area consists of lowlands in the middle and lower reaches of the various rivers. Grain yields are extremely low in some lowlands seriously affected by flooding and salinization.

Because of the difficulties in increasing yields in the lowlands, it would be more straightforward to use them to store water. For example, the Baiyang Dian collects the inflow of the Daqing He drainage system and is closely linked together with the Dong Dian and Wen'an Wa, controlling a large portion of various drainage basins and extremely effectively reducing the discharge into the sea of those river systems through its storage.

If the necessary projects are built in all the lowlands of the Hai-Luan Plain, taking into account their differing conditions, and reservoir construction in the mountainous upper reaches is combined with groundwater storage in the plains, thereby intercepting the runoff at each level, it may be possible to utilize the water sources of the plain to their fullest extent. Problems of salinization and swampiness can be overcome when lowlands are used as plains reservoirs for storage, provided that appropriate engineering measures are adopted (see Yu Fenglan and Wang Wenkai, Chapter 20).

Appropriately Divert the Huang He

In recent years the average annual discharge of the Huang He into the sea has been approximately 48.6 km. One might expect this amount to decline gradually as a result of increases in water use in the upper and middle reaches, but even in the extraordinary dry period of 1970-1977 the annual discharge still reached 31.0 km; 14.3 km flowed into the sea during the non-irrigation period (October to March); and 6.1 km was discharged in the driest year. It is rather difficult to divert large amounts of water from the Huang He in the upper and middle reaches because most of the cultivated land has relatively poor conditions for diversion. Following the completion of the Longyang Gorge and other reservoirs in the upper reaches, the low water flow will increase in the lower reaches. The construction of the Xiaolangdi Reservoir would be even more favourable to Huang He diversion. It is estimated that in the coming decade the annual discharge into the sea of the Huang He will still be 30 km, indicating that it will be possible to increase diversions by several km, but the diversion must be done appropriately.

Silt deposition occurs in the lower reaches mostly during the non-flood season when the rate of flow is low and the silt-carrying capacity is small. Bed scouring occurs primarily during the flood season when the flow is high and the silt-carrying capacity great. This has positive implications for diversion during the non-flood season when water use is high. If all the water of the Huang He is diverted, the small flow which produces silt deposition would not pass through and diversion would be beneficial to river control. When the water is not consumed entirely, the remaining portion can be adjusted by making concentrated discharges into the sea from artificial flood peaks released by the Sanmen Gorge Reservoir, precluding bed silt deposition in the lower reaches. Of course, there would still be a considerable amount of silt to take care of, but this northward conveyance project would be much simpler than diverting Chang Jiang water across the Huang He. Diversion expenses, including silt control, would not necessarily be higher than the transfer of Chang Jiang water and the period of construction would be much quicker.

Use Surface and Underground Reservoirs Jointly and Manage all Water Sources in a Unified Manner

If we are to solve the problem of water shortage on the Hai-Luan Plain, all of the above measures, which focus on opening up new sources, must be supplemented by the joint operation of surface and underground reservoir storage in accordance with the principle of mutual benefit. We must carry out unified management of all water sources, beginning by requiring that projects be constructed with sound storage, irrigating and drainage systems. In the plain, if there is no storage there can be no irrigation, and storage is impossible without drainage. With a sound water system, imbalances in water sources between areas and between surface and subsurface can be artificially regulated, controlled and supplemented through management which is done in accordance with the requirements of ecological balance and the development of production. This would turn harmful water into helpful water and make optimal use of water sources. This type of unified management requires the establishment and perfection of institutions governing water source management.

It is first of all necessary to utilize the shallow groundwater and effectively control the water table. Then irrigation must be done from the plains and mountain reservoirs according to the level of the water table and the severity of the drought. Only in an extraordinary drought should the deep groundwater reserves be utilized, and only for short periods.

Water should be stored in major flood periods in the following order: first, the mountain and plains reservoirs, and then the underground reservoirs. In addition to the water they absorb directly from rainfall, shallow aquifer resevoirs may, under proper conditions, be recharged from the mountain and plains reservoirs by providing water to the canals in excess of the water duty. This would allow the water table to rise continuously until it reaches the critical level at the end of the flood period.

Contents - Previous - Next