CC BY-NC-ND
0Available online at www.sciencedirect.com
SciVerse ScienceDirect PfOCSCliQ
Environmental Sciences
Procedia Environmental Sciences 13 (2012) 1354 - 1362 ^^^^^^^^^^^^^^^^^^^
The 18th Biennial Conference of International Society for Ecological Modelling
Analysis of the Characteristics of Runoff in Manasi River Basin in the Past 50 Years
X. L. Tang a*, J. F. Li b, X. Lvc, H. L. Longd
"Normal College of Shihezi University, Shihezi 832003, P.R. China bWater Resources and Architectural Engineering College of Shihezi University, Shihezi 832003, P.R. China 'Production and Construction Corps in Xinjiang Key Laboratory of Oasis Ecological Agriculture, Shihezi 832003, P.R.
dSchool of Life Resource Environment, Ylli Normal College ,Kuitun, 832003, P.R. China
Abstract
Runoff generated in the upper Manasi River Basin is water resource supplied for the middle and lower reaches. Based on the monthly runoff series (1957-2006) of the two hydrological gauging stations, Kensiwate, Hongshanzui and precipitation series at 4 precipitation stations of the upper Manasi River Basin, multiple indexes and a series of methods and wavelet methods are adopted to analyze the inter-annual, annual and periodic variation characteristics of the streamflows in the watershed for the last 50 years. The results show that: showed that: (1) Distribution of annual runoff was uneven, and mainly focuses on June and july and August. The inter-annual variations were great, with significant difference between the wet year and dry year; (2) annual runoff changes had obvious stages and the tendency of annual runoff increasing was significant; and (3) annual runoff changes had a main period of about30 years and 15 years, with an abrupt point happening in 1995. The increasing of precipitation and Climate warming are the key factors influencing the runoff.
© 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of School of Environment, Beijing Normal University.
Keywords: Runoff; Manasi River Basin; Variation characteristics
1. Introduction
The change of watershed hydrologic cycle has been very obvious theses years because of the global climate and human activity. Dynamic binary cycle process, which includes natural water circulation and
* Corresponding author. Tel.: 0993-2057131 or 18999330738.
E-mail address: txling@sina.cn.
1878-0296 © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of School of Environment, Beijing Normal University. doi:10.1016/j.proenv.2012.01.128
artificial water circulation, has been formed[1]. Runoff is a key part of hydrological process.It directly reflects the effect of the change of area climate and human activity have on hydrological process.Annual runoff volume shows some certain regularity, but also shows strong Randomness[2,3]. Because of the dry climate and human activity recent years, river runoff has decreased tremendously. This caused a series of ecological environment problem, such as channel shrinkage, estuarine sedimentation, downstream river drying up, Wetland degradation and loss of function of the river. Analyzing the characteristics of river runoff has very important theoretical and practical meanings. It helps getting further knowledge of the variation trend of Surface water resources. It also provides science evidence for rational development of water resources and protection of Ecological Environment.
Manas River water system is the lifeblood of the Manas oasis survival, it is the melting glaciers and precipitation are mixed Nourished rivers, and its richness and poorness relate to the change of climate closely[4]. Manas River basin upstream of the formation of mountain runoff is the water supply of middle and lower reaches. It keeps the development of the economic scientific system and ecology system of the area. A lot of work has been done about the effect hydrological characteristics, climate change have on the water resource, and a lot of scientific result has been reached so far, but the article about the runoff processes and hydrological extreme events and the frequency of flood peak discharge of the quantitative relation between are yet hard to find. This article analyzes the Manas River runoff for the year, inter-annual variation and runoff trends in recent 50 years. The result of the research has very important practical meaning of rational use of water resources, water resources planning, and so on. It relates to not only the development of the oasis itself, but also the protection of ecology environment of the whole area.
2. Profile of study area
Manas river is located at the southwest part of Southwest of Junggar Basin in Xinjiang, east path is 85°0l' — 86°32' , north latitude is 43°27' — 45°2l' . South of it is the famous the TianShan mountains, and west of it is the edge of Junggar Basin. The north part is low and the south part is high. Mountains and plains equally share the area. The area is 5156km2, the average height is 3022m, and the area of the glacier is 608.25 km2 . The main source of the runoff are rain and meltwater, among which 34.5% is meltwater.
There are 5 inland rivers within the area, which all stared at the north foot of Tianshan Mountain with Ha Bier Ga Mountains. They are Tacitus Creek, Manas River, Ning Jia river, Jingou river, and Ba Yinggou river from east to west. They all flow from south to north to Junggar basin, Manas river flows into the south edge of the basin, and then turns northwest to Manas lake. It appears as the typical comb water. The mountain pass as the boundary, mountain areas is runoff, out of the mountain pass is losing of district runoff. There are Honggou coal pit, Qingshui river, Ken Siwate and Hongshan hydrographic station at upstream of Manas river. Ken Siwate is trunk of river, Control station of the confluence of tributary. Its height is 910m and its controlling area is 4637km2' Hong Shanzui station is the control station of mountain pass of Manas River. Its height is 610m. Water diversion [5] is located in 300m of downstream.
3. Materials and Methods
3.1 Data Sources and Data Processing
Considering the geographic characteristics of the area and the completeness, consistency, and the representation of the sources from hydrological stations, I chose the following sources to analyze: the
average monthly runoff from 1956 to 2006 of Ken Siwate hydrological station , Hong Shanzui hydrological station in this area, and the source of the raining amount of Shihezi,Manas,Hong Shanzui and Ken Siwate rainfall stations.
When analyzing the data, I used the SPSS software to get the average runoff of Spring (March to May), Summer ( June to August), Autumn ( September to November), winter ( December to February next year), and the average runoff of the whole year.
3.2 Research Methods
3.2.1 Wavelet analysis[6]
Wavelet analysis has been used in this scientific area recent years. Wavelet analysis has the quality of localization characteristic of time and frequency. It can show the changing of the sequence's instantaneous promotion along with the time. It can analyze the periodic abrupt of the same frequency. The change of runoff in different time domains have multi-level time scale features of the structure and localization, so wavelet analysis is being used more and more widely. In this article, I chose wavelet functions Mexican hat[7,8] to do the periodicity analysis of annually runoff serial of Manas river. Using different time scales the square of the wavelet coefficients in b area by all points, so I get wavelet variance:
From formula (1) can test out the main periods of time series. 3.2.2 Mutation test statistics
I used Mann-Kendall's non-parametric rank correlation test method[9] to do test of the runoff, rainfall and evaporation trends and mutation. The advantage of this method is that it covers a big checking area, and there are less man-made interference. It is also highly quantified. There are a lot of examples of the use of Mann-Kendall GQM in the last 20a around the world. It has been widely used in the area of Inspection of the trend component of hydro meteorological data, which includes the quality of water, flow, temperature, rainfall series, and so on.
4. Changes in runoff for the year
The amount of the runoff of the whole year much depends on supplement. According to upper reaches of Manas river, distribution curve of precipitation during the year, we can see that runoff flow during the year of hydrographic station increases gradually from January to July, and decreases from August to December. The amount of runoff is relatively little from January to April, and it is rich from May to September. The average flow duration curve appears as single peak. There is a flood season which starts from May and reach to the top in July. Rain mainly falls from April to July, which equals to 47.7% of the rain of the whole year. The main supplement of runoff of Manas river is rain and glacier melting. Runoff changes during the year because of the rain, but this change is not very obvious. Compare precipitation line to flow duration curve, we can see that there is a obvious supplement from melt water during June, July, and August. Flood water in summer flood is because of melting snow and the thaw. Rain supplement contributes less to runoff during that time.
The 2 Hydrographie station of Manas river have 2 factors for the change of runoff within a year and the change of rain within a year. The 2 factors are 0.319 and 0.312, which both fail to pass the level of significance test a=0.01. This concludes that rain is not the main factor that affects the change of runoff within a year.
irecipitaticnfnrO —« kenSiïiaïe .
H'j[]iï Si id;i2ui /
140 120 10D 80 60 40 20 0
Ol £ CJ
nionth
Fig. 1. Distribution of annual runoff and precipitation in Manasi River Basin
According to the upper and midddle reaches of Manas river for the proportion of the average monthly flow distribution (graphic 1), we can see that the amount of runoff varies tremendously during the year. Runoff mainly happens from June to August. During that time, the rain is rich, and the temperature is high, The amount of runoff is large too, and it equals 66.9%~70.3% of the average amount of many years' runoff. The amount of rain is largest of the whole year during April and May, but the amount of glacier meltwater is still little because of the low temperature, so runoff quantity is little too. The amount of runoff is largest during summer. Autumn seconds to summer, and spring is after autumn. Winter has the least amount because of the low temperature and river of ice cover.To conclude, summer contributes the most of a year to Manas River.
Table 1 The percentage of monthly runoff in annual runoff in Manasi River Basin (%)
Hydrological station Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
KenSiWaTe 1.62 1.41 1.45 2.02 4.76 15.61 28.75 25.97 9.76 4.11 2.57 1.96
HongShanZui 2.11 1.97 2.17 2.47 4.76 14.73 27.53 24.62 9.61 4.44 3.05 2.51
5. Inter-annual variation of runoff
According to the variation curve of annual runoff at hydrological stations, Ken Siwate and Hong Shanzui hydrographic station's runoff curve synchronously, and they both appear to be increasing in all. The Increased rate are 0.56*108m3/10a and 0.43*108m3/10a.
с м g <в
fH 2 10
■KenSiWaTe ■HongShanZui
Fig. 2. The variation curve of annual runoff at hydrological stations in Manasi River Basin 5.1 Inter-annual variation of runoff
The usual method we use to measure the annual runoff is annual runoff uniformity coefficient and ratio of yearly extremum (ratio of Maximum and minimum of annual runoff). Yearly variation of runoff of drought region closely relates to the exploitation of water resource. It directly affects the reliability of stage of water supply and project storage capacity.
(K -1)2
(n: number of years for the observation; Kt For the i year variability of annual runoff)
Uniformity coefficient of runoff v is one of the most important indicator to measure the inter-annual variation of river flows« This factor directly shows the fluctuating quantity of the annual change of runoffs. It means that annual change of the runoff is big if the number of Cv is big. ratio of yearly extremum P is the times relation between the 2 extreme value of annual runoff in the basin. It shows the degree of uneven of the amount of runoff. The bigger P is, the more uneven of the inter-annual variation of the runoff[10].
Table 2 The inter-annual variation characteristics value of runoff in Manasi River Basin
Hydrological station
Average runoff(m3*108) Cv
Measured maximum
Flow /m3*108
Measured minimum
Extreme than the
annual
Flow Year P
/m3*108
KenSiWaTe HongShanZui
12.43 13.29
0.14 0.13
20.08 19.16
1999 1999
9.39 10.5
1992 1984
2.138445 1.824762
According to formula (4-1), we get the inter-annual variation characteristics value of runoff in Manas river basin (Figure 2). We can tell from Figure 2 that uniformity coefficient and ratio of yearly extremum
of Manas river basin are both small. The main supply of Manas river are rain and meltwater, so the amount of Cv is small because of the coupling relationship of the rain and the temperature.
5.2 Interannual variability of runoff stage
We can use the difference accumulation curve of annual runoff to analyze the variation tendency of multiple years' runoff[11]. The difference accumulation curve of annual runoff usually takes the following steps to draw: First, calculate msr mode suppression ratio of annual runoff according to the data of the amount of runoff. (msr mode suppression ratio of annual runoff of some year is shown by the ratio of annual runoff and mean annual runoff); Second, calculate difference of msr mode suppression ratio according to formula (3); Third, calculate difference-product of msr mode suppression ratio according to formula (4); Forth, draw the difference accumulation curve of annual runoff according to the numbers calculated.
In the formula, AKj , Ki represent the K-years difference of msr mode suppression ratio of the Kth year and factors of msr mode suppression ratio; Qt is the amount of runoff of the ith year; Q0 is the average amount of runoff of many years. Ht is the difference-product of msr mode suppression ratio of the tth year. Because difference of msr mode suppression ratio can be positive or negative, when the difference of msr mode suppression ratio keeps increasing, it means that the amount of runoff keeps increasing during this period of time. The amount of water is large when the number of difference of msr mode suppression ratio increases. When the difference of msr mode suppression ratio keeps decreasing, it means that the amount of runoff keeps decreasing during this period of time. The amount of water is little when the number of difference of msr mode suppression ratio decreases. With this method, we can get the result of the inter-annual variation of runoff more directly and accurately.
AKi = QL_Qo = Q _ i = k _1
—KenSiWaTe-HongShanZui
Fig. 3. The difference accumulation curve of annual runoff of Manas River
We can tell from Figure 3 that the changing process of difference accumulation curve of annual runoff of Ken Siwate and Hong Shanzui are similar. The sequence of the amount of annual runoff from the year
1957 to 2006 in Manasi river are divided in the following periods: an obvious lacking of water period from 1972 to 1980, an obvious rich in water period from 1995 to 2002, a steady period from 1961 to 1971, and an obvious unsteady period from 1981 to 1995, during which the time of lacking water is longer than the time of rich in water.
5.3 Runoff of the cyclical variation
Based on wavelet equation of variation of Manas River annual runoff change (fig. 4), In 1957 to 2006 within the time domain, the main time periods of annual runoff is 30a, oscillation energy is gradually enlarge at 1~30a
Fig. 5. Variance curves of the real parts of wavelet transformation coefficient in 30 years and 15 years time scales at hydrological stations in the Manasi River Basin
According to Wavelet variance test results , design The first and second principal period of annual runoff about image wavelet coefficients ( fig. 5) .Can be seen annual runoff characteristics of the 15-year time scale of annual runoff, runoff was changed about 2 conversions of wet and dry, the average cycle of about 14 years, From the short time scale 2008^2015, Manas river's annual runoff in a stable
period; characteristics of the 30-year time scale , runoff was changed about 1 conversion of wet and dry, he average cycle of about 33 years,, 1957^1966 years are high flow periods, 1967^1985 years are low water period, 1986^1999 years are high flow periods, 2000 starts into the low water period« As 30-year cycle of the largest energy scale, it makes a contribution to variance of o riginal runoff series is also the largest« Therefore, the upper and midddle reaches of Manas river is in low water period of large time scale, estimates until 2019, Annual runoff period will be in low water period about 2000 to 2019.
5.4 Mut/tion /n/lysis of /nnu/l runoff
In this paper, adopt most people think that preferable theoretical basis and application results-- Mann-Kendall, make a study of mutation test in runoff of Manas river. From the analysis result of Mann-Kendall, draw the curve of UF and UB, if Uf or UB greater than 0, indicates that there is an upward trend in the number of sequence, if Uf or UB less than 0, indicates that there is an moderating trend of sequence. When they exceed the critical value line, upward or downward trend that was significantly.the area exceeds the critical value line is mutation of the time zone. Such like UF and UB these two lines is across, and point of intersection is in the critical region, so the point is mutations in the beginning of time« From fig. 6, the annual average runoff was increased in the early 1990s is a mutations, that means from 1995, although trend of annual runoff began to become profusion from 1995, but increase was not significant.
6. Conclusion
In this article, I used multiple methods to analyze the regulation of the inter- annual variation of runoff in Manasi River Basin, and get the following conclusions:
The annual distribution of the runoff in Manasi River Basin is very uneven. The runoff mainly happens from June to August, which equals 66.9%~70.3% of the average annual flow. Seasonal distribution of runoff in summer is the most abundant runoff, autumn is secondly, minimum flow in winter; Changes in runoff and precipitation during the year have no significant correlation between changes, It means precipitation is not crucial for summer flood, but summer temperatures is the main contribution for Manas river.
The past 50 years, the total amount of runoff in Manas River Basin has been increasing, especially the time from late 1960s to early 1980s, the runoff increased obviously. The variation of runoff has the
obvious mutation characteristics. There was a whole basin runoff increased mutations happened in 1995, but it was not a significantly. The increase of rain and temperature and human activity are the main reasons of the variation of runoff.
I discovered that the inter-annual variation of runoff in Manasi River Basin has the obvious stage and cyclical using Wavelet analysis. sequence of runoff in 50 years experienced a period of lacking water, a period of rich in water, a steady period, and an obvious unsteady period. The amount of runoff appeared 30years, and 15 years significant cyclical change. From changes in runoff are the primary cycle, We concluded that Manasi River Basin is experiencing the period of lacing water now, and it will probably last till the year 2019.
Acknowledgements
This work was supported by the National 973 Key Project of China (2010CB951004) and the National Natural Science Foundation Project of China (41001020). The authors are grateful to JunFeng Li and Chao Ming for their help in preparing this article.
References
[1] Huang Q, Jiang XH, Liu JP, et al. Binary mode variation of annual runoff of the Yellow River . Progress in Natural Science, 2002,12 (8) :874-7.
[2] Wang GX, Cheng GD. The last 50 years in the Hexi Corridor region variation of the ecological environment and comprehensive countermeasures. Natural Resources, 2002, 17 (1): 78 - 86.
[3] Huang F, Zhang SW, de Leeuw J. 20th century, 70 years Wuyu'er Songnen Plain wetlands evolution River Basin [J]. Natural Resources, 2006,21 (6): 949 - 56.
[4] Liu DQ, Xu SG. Zhalong wetland water balance analysis . Natural Resources, 2006,21 (3): 341 - 8.
[5] Jiang YC . Manas River "96,7" flood of .Hydrology, 1999 (6) :57-8.
[6] Wang WS, Ding J, Li YQ. Hydrological wavelet analysis. Beijing: Chemical Industry Press, 2005.
[7] Liu XZ, Qing S, Liu DL. Runoff characteristics of the evolution of multi-time scale. Water Resources and Power, 2008, 26 (1): 21- 5.
[8] Lu XN, Deng W, Zhang SQ. Huolinhe River runoff sequence characteristics and effects of multiple time scale analysis. Natural Resources, 2006, 21 (5): 819 - 26.
[9] Fu CB , Wang Q. Climate definition and detection of mutations. Atmospheric Science, 1992, 16 (4): 482 -93.
[10] Wu Y, Cheng WM, Ren LL. In the Hotan River basin runoff characteristics of time series. Natural Resources, 2006, 21 (3): 375 - 81.
[11] Li SX, Lang H. Representative of annual precipitation series in Hebei Province Haihe River Water Resources, 2004, (5):43-5.
