Contents - Previous - Next

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

Chapter 17. Preliminary estimation of natural runoff in the Huai He basin

Tang Qicheng, Cheng Tianwen and Luan Lukai
Institute of Geography, Academia Sinica, Beijing, China

THE HUAI HE is a large river in China with a total length of about 1,000 km and a basin of 185,700 km≤. It rises in the Tongbai Shan of Henan Province and flows through Henan, Anhui and Jiangsu provinces. Originally the Huai flowed into the sea at Yuntiguan in northern Jiangsu, but when the Huang He changed course in 1938-1947, this outlet was plugged up by a huge amount of sediment. Now the Huai flows in its middle reaches into Lake Hongze, whereupon it is entirely under human control until it finally empties into the Chang Jiang at Sanjiangying in the vicinity of Yangzhou.

Either the East Route or the Middle Route of China's south-to-north water transfer project must pass through the Huai He basin. Therefore, the water resources in the basin proper, especially the volume of stream runoff and its fluctuations, bear importantly upon the planning and design of the project and its future operation. Because the Huai was the earliest major river on which water projects were built after 1949, most of the hydrological data available from 1954 are affected by human activities, making calculations and analysis extremely difficult.

The present method of reconstructing runoff data is mainly to carry out extensive year-to-year investigations and analyses of farm irrigation water use and to make comparisons with representative stations where the impact of human activities is relatively minor.

This chapter attempts a different approach, to use the principle of the water balance to determine the long-term annual average runoff. Because this method has not yet been utilized widely and is not very highly developed, the present estimates must be regarded as preliminary.


The Huai He basin is bordered on the south by the Dabie Shan, Huo Shan and Zhangba Ling as well as the Chang Jiang basin and on the north by the Huang He (Figure 1). The upper reaches extend from the river source to the mouth of the Hong He on the border between Henan and Anhui provinces, with undulating mountains and hills on either side of the basin. The river here is 364 km long with an average gradient of 0.049 per cent. The middle reaches stretch from the mouth of the Hong He to Lake Hongze, a distance of 490 km with an average gradient of 0.027 per cent. There are numerous lakes and depressions in the vicinity of the main course. Along the northern edge is a plain and there are numerous hills along the south. The lower reaches are below Lake Hongze and are 146 km long with an average gradient of 0.036 per cent. (Survey and Design Institute of Huai He Commission, 1957).

The Huai discharges through two channels. Most of the water joins the Chang Jiang at Sanjiangying after passing through the Sanhe gate on the south bank of Lake Hongze and then crossing Lake Gaoyou. The other channel exits Lake Hongze through the Gaoliangjian gate on the east bank, passes through the North Jiangsu Canal and empties into the Huang Hai at Biandan'gang. In a particularly heavy flood year the river may also pass through the Erhe gate at the northeast corner of Lake Hongze through a channel which flows into the Yi, and then into the Xin Yi He, which empties into the Huang Hail The terrain in the lower reaches of the Huai is flat with a crisscrossing drainage network, numerous sluice gates and dikes, and myriad lakes.

There are numerous tributaries in the upper and middle reaches of the Huai. Of these there are 29 larger ones which converge directly into the Huai in Henan and Anhui.


The bulk of data on precipitation in the Huai He basin has been accumulated since the late 1950s. Most gauging stations did not begin taking observations until 1958. In our statistical analysis we have consistently used actual measurements. Where the figures from the hydrological and meteorological stations at one site are not identical, we have generally taken the data from the meteorological stations as standard and used those from the hydrological stations for reference.

The annual precipitation series used are from 1958 to 1978. Analysis shows the mean value for these 21 years to be approximately identical with the long-term average. For example, the 74-year mean annual precipitation from 1905 to 1978 at Nanjiang station was 1000.4 mm, while the 1958-1978 mean was 1,018.0 mm, a difference of only 2 per cent. The average at Ji'nan station for the 63 years from 1916 to 1978 was 649.3 mm compared with 697.6 for 1958-1978, a difference of 7 per cent. Figure 1 presents an isogram of annual precipitation in the Huai basin, based on statistics from about 60 stations, sketched on a 1:1,500,000 topographic map.

By taking measurements from Figure 1, total annual precipitation of 167.5 km≥ (average depth, 902 mm) can be calculated for the basin. Precipitation decreases progressively from 1,400 mm in the south to 650 mm in the north.

We compare the annual average value of precipitation in the period 1951-1970 with that for 1958-1978 at eleven stations-Ji'nan, Nanjing, Kaifeng, Xinanyang, Hefei, Qingjiang, Bengbu, Linan, Haoxian, Lianyungang, and Shouxian. These stations were chosen for their relatively uniform distribution in the basin. The mean values for each of the individual stations differ less than 3 per cent between the two time intervals. The aggregate values for the eleven stations were almost the same during the two time intervals.

Figure 1. Annual Precipitation in the Huai He Basin


There are quite a number of methods available for estimating annual evaporation. Among those which are used frequently in China and which have a relatively solid theoretical foundation in water-heat balance equations based on the energy budget are the Budyko (1960), Penman, Olidekop, Bagrov and Schreiber (Gao Guodong et al., 1980) formulae. The suitability of each of these depends upon climatic conditions. We have compared them using meteorological data from 1951-1970 from a number of stations both within the basin and its periphery to estimate long-run average annual evapotranspiration.

The results are listed in Table 1. They are analysed below.

1. The evaporation estimated by the Penman formula (E0) is with a humid underlying surface, that is, when water supply is sufficient. During most months and in most parts of the Huai He basin, actual evaporation is lower than Eo owing to water deficit. Only some of the mountainous areas south of the Huai He and a small number of downstream networks can meet the requirements of the formula for wetness in most months. In most cases it is necessary to multiply by a coefficient of conversion (Zhu Gangkun et al. 1955). We determined different coefficients of conversion by referring to the "Map on Current Land Use in China" (Institute of Geography 1980) as well as to the crop types in the vicinity of the stations involved. This coefficient is 1.0 for the rice growing regions of Nanjing, Hefei, and Gaoyou in the south; 0.6-0.8 in the grasslands and forests of Kaifeng and Ji'nan in the north; and 0.6-1.0 in most dry farming regions within the basin. Isolines plotted for evaporation estimated by means of the Penman formula modified by the coefficients of conversion tend not to conform to the general patterns of natural conditions in the basin. In addition, evaporation tends to be greater in the centre than in the east part of the basin. Because we have neither a complete method nor sufficient experience in choosing coefficients of conversion, we have not adopted the Penman formula in China to estimate evaporation in the Huai He basin.

Table 1 Evaporation Calculated by Different Formulae at Representative Stations in the Huai He Basin

Estimated long-term average annual evaporation (mm)

Evaporation formulae Zhengzhou Shangqiu Xuzhou Xuchang Boxian Suxian Qingjiang Zhumadian Fuyang Bengbu Xinyang Nanyang Huoshan
Bagrov(n=0.8) 487.4 535.1 602.7 528.2 584.3 603.1 569.1 590.8 598.8 585.0 672.4 656.0 759.0  
Schreiber (equivalent n = 1.0) 509.3 552.1 626.3 547.6 597.3 624.2 618.7 634.3 618.6 610.6 664.5 656.7 723.5
Budyko (equivalent n=4/3) 535.2 584.8 671.6 586.3 644.1 676.6 681.1 692.0 676.7 664.8 734.9 723.2 812.6
Bagrov (n= 1.5) 588.9 642.1 734.1 647.8 703.3 751.3 706.5 723.2 743.4 726.2 794.7 793.6 882.1
Olidekop (n = 2.01 568.6 620.7 723.2 627.9 692.5 730.1 745.5 753.7 733.1 726.2 815.0 793.6 902.6
Penman(revisedviaconver- sion coefficient) 525.7 621.6 633.6 597.5 634.4 596.8 534.9 527.5 610.7 598.0 600.4 723.3 604.1
Penman 698.7 726.0 742.5 684.6 740.6 721.9 660.0 699.0 711.4 698.4 701.9 723.3 703.9

2. In the Bagrov formula, the estimated value n = 1.0 is equivalent to that of the Schreiber formula; the estimated value n = 1.33 is equivalent to that of Budyko formula; and the estimated value n = 2.0 is equivalent to that of the Olidekop formula. Table 1 shows that the results from these formulas vary in an orderly way, that is to say, evaporation increases with n when n is 0.8-1.5, and it is the highest when n = 1.5. But evaporation increases when the Olidekop formula (n=2.0) is utilized. Isolines plotted with evaporation estimated by the Bagrov formula (n = 0.8, 1.5) and the Budyko formula are similar in tendency and have great regularity. Basin evaporation increases progressively from northwest to southeast, corresponding to the natural conditions, that is, evaporation in the southeastern plains exceeds that in the northwest and the mountains. The annual evapotranspiration, annual runoff and runoff coefficient of the basin are 604 mm, 298 mm, and 0.33 mm respectively when n = 1.0. The runoff coefficient approximates that of the Hangjia lake region south of the Chang Jiang. The estimated evaporation is generally too small. By computing the Bagrov integral using
and n= 1. 5, we obtained estimates for four representative stations (Table 2).The estimated results differ greatly from P/Eo when n= 1.5, but are relatively close
This situation corresponds to the study by Gao Guodong et al (1955), which concludes that using the Bagrov formula with
to estimate evaporation over the Jiangsu plain and other places approximates the Budyko formula in terms of numerical values.

Table 2.Estimated results of Bagrov formula with Different n values

Zbengzhou Xuzhou Qingiiang Xinyang
P/E0 0.627 0.794 1.001 1.087
Estimates derived from
the Bagrov formula
when n = 1.33
0.737 0.884 1.049 1.130
Estimates derived from the
Bagrov formula when
n= 1.5
0.247 0.225 0.286 0.359

3. Evaporation estimated by means of the Budyko formula mainly depends on the radiation equilibrium value and precipitation. The magnitude of the radiation equilibrium value in turn depends on the available evaporative capacity. Actual evaporation depends on the quantity of convertible moisture, i.e. the amount of precipitation. Table 3, presenting the ratio of precipitation to evaporativity

at various stations in the basin indicates that the available evaporative capacity is often far greater than precipitation. This means that either the radiation equilibrium or precipitation determines the magnitude of evaporation in the basin. Thus the Budyko formula may be used for the Huai He basin.

It should be pointed out that annual evaporation estimated by using the Budyko formula for the Huoshan station located in the mountains south of the Huai He is 812.6 mm. This figure is 90 mm higher than that at the Nanjing station in the southeast. It is clearly unreasonable so we have not used it.

Table 3 Ratio of Precipitation to Evaporativity (R/L) for Various Stations in the Huai He Basin

Station: Zhengzbou Shangqui Xuzhou Xuchang Boxian Suxian Qingjiang
PL/R: 0.627 0.661 0.794 0.742 0.756 0.855 1.001
Station: Zhumadian Fuyang Bengbu Xinyang Nanyang Huoshan  
PL/R: 0.958 0.894 0.904 1.087 0.970 1.383  

These results led us to adopt the Budyko formula to calculate the long-term average evapotranspiration at each station and from this to determine the value for the entire basin. The results are presented in Figure 2. The distribution is basically the same as that of precipitation, with the lowest total evaporation of about 550 mm in the northwest portion of the basin and the highest, over 700 mm, in the water network region of the lower reaches in the east. Evapotranspiration is relatively homogeneous over the extensive plains area in the centre of the basin, ranging between 650 mm and 700 mm. Measurements from Figure 2 indicate a long-term annual average evapotranspiration in the entire Huai basin of 673 mm, or a total of 12.498 km≥.

In order to correspond to other water balance elements, the 1951- 1970 period is used. We compared the means for 1951-1970 with those for 1958-1978 at 11 stations which are fairly evenly distributed throughout the basin: Ji'nan, Nanjing, Kaifeng, Xinyang, Hefei, Qingjiang, Bengbu, Linan, Haoxian Lianyungang and Shouxian. The difference between the two periods for each station is within +3 per cent and the aggregate mean is virtually the same.

Figure 2. Annual Evapotranspiration in the Huai He Basin

Figure 3. Annual Runoff in the Huai He Basin

The same data were employed in the calculation of the radiation balance value. The evapotranspiration results for 13 stations are listed in Table 1.


The isogram of average long-term annual runoff (Figure 3) is derived from the isograms for precipitation and evapotranspiration. From Figure 3 an annual runoff depth of 221.8 mm (total natural runoff: 35.08 km≥) can be derived for the basin above Lake Hongze. Since annual precipitation averages 881.0 mm (total: 139.34 km≥) in this area (from Figure 1), the annual runoff coefficient in the upper and middle reaches is 0.25.

At present most estimates of the natural runoff in the basin above Lake Hongze are about 35 km≥, virtually identical to our estimate. Nonetheless, the inadequacies in the methods used here to estimate evapotranspiration and incompletely studied problems such as the variations in total evaporation caused by extensive irrigation combine to make the present estimates rather tentative.


Budyko, M. I., 1960, Surface heat balance, Science Press, Beijing

Gao Guodong et al, 1955, "Research on Land Surface Evaporation and Heat Consumption Through Evaporation in China", Journal of Meteorology, Vol. 26, No. 1-2.

Institute of Geography, Chinese Academy of Sciences, ea., " 1 :6,000,000 Current Land Use Map of China", Cartography Press, Beijing.

Survey and Design Institute of the Huai He Commission, Ministry of Water Conservancy, 1957, "Summary of Huai He basin planning," China's Water Resources, No. 4.

Zhu Gangkun, et al, 1955, "The Application of Meteorological Records in Economic Construction (II): A Preliminary Study of Evaporation in Different Parts of China", Journal of Meteorology. Vol. 26, Nos 1-2.

Contents - Previous - Next