Hydrological Changes in the upper Tuul River Basin
Davaa Gombo1, Erdenetuya Magsar2
1Institute of Meteorology and Hydrology, 2Information and computer center, Juuljinii St. 5, Ulaanbaatar-46, Mongolia
Fig. 1. Location and geographical features of the Tuul River, Mongolia
Forest degradation
Total forest area in Mongolia is 17 mln hectar, 500 thousand hectar is the Tuul river basin.
Wood cut is started since 1940-th and 100-200 thousand ì3 of wood have been cut annually. Wild fire occurred significantly in 1960, 1996.
Total deforested area is 19484 hectar and forest resources has decreased by 35 percent in the basin.
Forest area affacted by forest insects 41000 hectar /D. Enkhsaikhan, 2000/.
ÃÃàäààäàðãðãûíûí áîëîíáîëîí ããàçàçððûíûí äîîäîîðõ ðõ óñíûóñíû õõààððèëöàíèëöàíóÿëäààóÿëäàà
0
5
10
15
20
25
30
1 3 5 7 9 11 Cap
Q T
uul-U
B,
0
1
2
3
4
5
6
7
8
9
H w
ell,
Q Tuul Central w ell No 14, 2002,04Central w ell No 14, 1996 Industrial w ell No 14,1996Upper source w ell No 19, 1996
0
10
20
30
40
50
60
70
Songino Zaisan Amgalan Railwaybridge
Bayanzurkhbridge
Gachuurt Terelj bridge Confluenceof Tuul and
Terelj
21,0
21,2
21,4
21,6
21,8
22,0
22,2
22,4
22,6
22,8
Q, 2002,04 Q, 1998,09 Q1994, 03 cap Q 1978,aver.
Runoff loss dynamics along the Tuul river
Interaction ground and surface waters in the Tuul river basin
Methods and data
1. Annual storage ratio /ϕ/ series of the basin
∫=
=
=1
0
k
k
Tdkϕ
where: T is days, k is the ratio of daily average discharge to yearly average discharge.
2. Analyses of autumn recession curves:
Qt = Qo exp(-at)
where: Qt – autumn recession flow at time t, m3/s, Qo - autumn flow at the beginning of the recession period (t=0), m3/s, a- storage variation or drainage factor, 1/day, t - duration recession flow, day.
0
Tdkf 0,10=ϕ
I II III IV V VI VII VIII IX X XI XII
k=1
k
3. Runoff coefficient /C/
Cnatural = hnatural regime /Pbasin
where, C natural is runoff coefficient in the period of natural flow regime, hnatural is annual runoff depth, mm, Pbasin is annual and basin average precipitation, mm in the same period.
Natural flow rate in the period of increased anthropogenic pressures is determined using the runoff coefficient in the period of natural flow regime and precipitation occurred in the period of increased anthropogenic pressures, as follows.
hnatural in anthropogenic period = Cnatural Panthropogenic period
Then, change in runoff will be the difference runoff in the period of natural flow regime /hnatural in anthropogenic period / and runoff, observed in the period of increased anthropogenicpressures hanthropogenic period.
∆ h = h natural in anthropogenic period - hanthropogenic period
a = 0,0003 Year - 0,5807
0
0.02
0.04
0.06
0.08
0.1
0.12
1945 1955 1965 1975 1985 1995 Years
Dra
inag
e fa
ctor
, 1/d
ay
Results and discussions
Drainage factor trend of the Tuul river basin /Tuul-Ulaanbaatar/
Drainage factor values range from 0,040 to 0,01. Increase in drainage factor or storage ratio “a” indicate that size of active drainage decreases in the basin /fig.2/.
-6
-4
-2
0
2
4
6
8
1943 1953 1963 1973 1983 1993 2003
year
sum
(k-1
)/Cv
Cyclic pattern of annual flows of theTuul river at Ulaanbaatar/Sum(Qi/Qaverage-1)/Cv verses with time, where Cv is coefficient variation of annual mean discharges (Qi)/
P basin = 0.5052 P effective + 186.49R2 = 0.5669
0
50
100
150
200
250
300
350
400
0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0
P effective or runoff depth,mm
P U
B, m
m
There are 2 cycles. First cycle is observed in the period of 1945-1976 and second one is in the period of 1977-2003. These 2 periods are highly distinguished by anthropogenic pressures.
Relating annual precipitation observed in Ulaanbaatar (Buyant-Ukhaa meteo. station) with effective rainfall or runoff depth of the Tuul river at Ulaanbaatar, annual basin average precipitation in last cycle has been approximated by the following regression equation
0.0
50.0
100.0
150.0
200.0
250.0
300.0
350.0
400.0
1945 1955 1965 1975 1985 1995 2005
YearPrec
ipita
tion
(P),
runo
ff de
pth
and
evap
orat
ion
(P-h
), m
m
h, Tuul-UB P Basin E=P-h± S
Basin average evapotranspirationhas been estimated by simple water balance equation, assuming that change in water storage in year is near zero. Annual flow series are remarkably and basin average precipitation are slightly increasing, while evapotranspiration orevapotranspiration and change in water storage are drastically decreasing in last 60 years. y = -0.0003x2 + 1.2902x - 1274.9
R2 = 0.1988
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1940 1950 1960 1970 1980 1990 2000 2010
Year
Run
off c
oefic
ient
= ru
noff/
prec
ipita
tion
K runoff Poly. (K runoff)
Analyses of annual runoff coefficients, expressed as ratio of annual runoff depth to the basin average precipitation show its increasing trend, indicating increase in overland flow or direct runoff and decrease in groundwater recharge
Basin average precipitation was 251 mm in the period of 1945-1976 and its slight increase (257 mm) is observed in the next period of 1977-2003. However, annual and long term mean flow of the Tuul River was 128 mm in the first and increased till 139 mm in the second periods. Following these changes annual evapotranspiration and water storage in the basin have been decreasing from 123 mm in the first and till 118 mm in the second periods. This evidences that annual runoff coefficient has increased by 0.03 value, indicating degrease in roughness in the basin. Decrease of roughness indicates degradation of vegetation cover in the basin. With increasing runoff coefficient decreases annual storage ratio /ϕ/ series and it indicates flow regime regulating capability of the basin has been decreasing. Change of runoff coefficient of 0.03 corresponds to the change of storage ratio by 0.005. It means that runoff coefficient is sensitive indicator than storage ratio, which is function of flow regime regulators as forest, wetland and lakes in the basin.
= -0.1708 C + 0.5387R2 = 0.4926
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.2 0.4 0.6 0.8 1
Runoff coefficient
Sto
rage
ratio
ϕ
Relationship between runoff coefficient and storage ratio of the Tuul river at Ulaanbaatar in the second period
ϕ
Therefore, storage ratio is somewhat function of vegetation cover dynamics. However, between NDVI and storage ratio wasn’t revealed good relationship in the particular basin. However, with increasing NDVI the water regime regulating capability in the basin increases.
Pi= 0.2205 NDVI(VII) + 0.3848R2 = 0.0877
0.350.370.390.410.430.450.470.490.510.530.55
0.3 0.4 0.5 0.6
NDVI in last decade of July
Stor
age
ratio
Decrease in active drainage area, degradation of vegetation cover, leading to the deterioration of flow regime, regulating capability lead to increase in vulnerability to flooding. Annual maximum discharge series show that it tends to increase and duration of occurrence of rainfall floods decreases from 15 days to 12 days in the basin
T, days
Q max = 0,3792 OH - 379,77 T = -0,0687 OH + 149,16
0
200
400
600
800
1000
1200
1400
1600
1800
1941 1951 1961 1971 1981 1991 2001OH
Q m
ax,
0
5
10
15
20
25
30
35
40
T, x
o
Y Y
Year (Y)
Annual maximum discharge (Qmax) and flood duration (T, days) series and their tendency of change in last years (Y)
0,000
0,050
0,100
0,150
0,200
0,250
0,300
0,350
0,400
1980 1985 1990 1995 2000 2005
OH
NDV
I
NDVI VI-IX average Linear (NDVI VI-IX average) Poly. (NDVI VI-IX average)
0,000
0,050
0,100
0,150
0,200
0,250
0,300
0,350
0,400
0,450
1980 1985 1990 1995 2000 2005
OH
NDV
I
NDVI VI-IX average Poly. (NDVI VI-IX average) Linear (NDVI VI-IX average)
0,20
0,25
0,30
0,35
0,40
0,45
0,50
0,55
1980 1985 1990 1995 2000 2005
OH
NDV
I
0,300
0,350
0,400
0,450
0,500
0,550
1980 1985 1990 1995 2000 2005
OHND
VI
NDVI VI-IX average Linear (NDVI VI-IX average) Poly. (NDVI VI-IX average)
a) b)
c) d)
Year
Dynamics of average values of NDVI in June-August period and in river basins of various region a- Bulgan river at Bulgan soum, b- Khalkh river at Sumber soum, c-Kherlenriver at Undurkhaan, d-Ider river at Zurkh bag
y = 12,85x - 1,2844R2 = 0,4643
00,5
11,5
22,5
33,5
44,5
0,000 0,050 0,100 0,150 0,200 0,250 0,300 0,350 0,400
M, l/s,km2 Linear (M, l/s,km2)
y = 10.368x - 0.9329R2 = 0.7019
0
0.5
1
1.5
2
2.5
3
0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400
M. l/s. sq.km
NDVI
a) b)
The relationship between average values of NDVI in June-August and annual average of specific discharge of the river Bulgan at Bulgan soum in 1982-2001, a- in 20 years, b- after 1991
Trend of annual maximum flow series, observed during rainfall floods shows that they it is increased by 20 cub.m/s per a year. However, maximum discharge values of spring floods tend to decrease by 3.1 cub. m/s and its duration of occurrence prolonged by 14 days per a year due to early occurrence of discontinuous snow melting resulted from air temperature variation in spring.
Log(Qmax) = -4.4011 Pi + 4.3915R2 = 0.5678
0
0.5
1
1.5
2
2.5
3
3.5
0.35 0.40 0.45 0.50 0.55 0.60
Coefficient of natural regulation
Log(
Q a
nnua
l max
.)
Q max = 23324 (Bmax*Pi)1.1254
R2 = 0.9857
0
100
200
300
400
500
600
700
800
900
0 0.01 0.02 0.03 0.04 0.05 0.06
Bmax*Pi, m3/s sq.kmQ
max
, m3/s
ϕ
ϕ
Relationships between annual maximum discharge and storage ratio (a), and specific discharge (b)
For the Tuul river, Qmax can be estimated by following empirical equation:
Qmax = 23324.(Bmax.ϕ)1.1254
where, Qmax is annual maximum discharge, cub.m/s, Bmax is specific discharge, which is function of climate variables, cub.m/s sq.km, ϕ is storage ratio of the Tuul River.
Concluding remarks
Analysis of storage ratio, drainage factor and runoff coefficient series show that water regime changes occurring in the Tuul river basin. Changes in natural components such as forest, vegetation and soil cover leading to predominantly changes in hydrological regime. For the estimation of annual maximum flows, it is important to include flow regime regulating capability of a basin. This is getting one of the basis for regional streamflow estimation in this changing environment.