ChapterPDF Available

Precipitation and Streamflow Variability in Tekeze River Basin, Ethiopia

Authors:

Abstract and Figures

This research is carried out using nonparametric Mann–Kendall test and indicators of hydrologic alteration (IHA) to determine the potential trend and variability of precipitation and streamflow in Tekeze River basin, Ethiopia. Precipitation trend analysis result showed increasing trends for Annual, Kiremt, and Belg seasons whereas a decreasing trend in the Bega season throughout the basin. Tekeze River streamflow analysis also showed a decreasing trend in annual and dry season (October to January), and increasing trend in the wet season months. The IHA and range variability approach methods are used to evaluate the pre- and postimpact hydrologic regimes due to Tekeze River dam construction. The result showed that Tekeze hydropower reservoir operation significantly changed the hydrological regime downstream of the dam. Multiday minima streamflow increased, multiday maxima streamflow decreased, high pulse count decreased, and fall and rise rates decreased, number of annual hydrograph reversals increased, and the number and duration of high and low pulses increased. Hence, investigation of trends in the hydro-climatic variables of Tekeze basin revealed many significant trends, both increasing and decreasing. The findings may assist water managers in better planning and management of water resources under climate variability and change.
Content may be subject to copyright.
Chapter 10
Precipitation and streamflow variability
in Tekeze River basin, Ethiopia
Fikru Fentaw*, Assefa M. Melesse
, Dereje Hailu*and Agizew Nigussie*
*School of Civil and Environmental Engineering, Addis Ababa Institute of Technology (AAIT), Addis Ababa University, Addis Ababa, Ethiopia
Department of Earth and Environment, Florida International University, Miami, FL, United States
10.1 Introduction
Water resources problems are becoming more challenging
and complex worldwide. The complexity of water resources
planning and management is due to the contribution of
climate variability, social and environmental considerations,
transboundary nature of rivers, and population growth. The
stress on water resources diversely increased due to the rapid
increase in population (Wu et al., 2013). Further, water man-
agers have been concerned about the anticipated impacts of
climate variability and change on water resources (Dawadi
and Ahmad, 2013;Vedwan et al., 2008). Changes in climate
variability enhance the uncertainties in the availability of
fresh water for the future (Middelkoop et al., 2001). As a
result, water resources managers face challenges to meet
future water demands on existing water infrastructure that
may be inadequate in the future and stress is increasing to
meet environmental flow requirements and provide water
for energy needs (McCartney and Menker Girma, 2012;
Qaiser et al., 2011). Changes in the hydrological cycle
can result from both climate variability and anthropogenic
interference. These changes may be gradual (trend) or
abrupt (shift). Changing hydrology may lead to under-
designed or over-designed projects (Mosquera-Machado
and Ahmad, 2007), which may not meet long-term needs;
thus, the traditional assumption of stationarity for hydraulic
designs requires review (Milly et al., 2008).
Rainfall and river flow in Africa display high levels of
variability across a range of spatial and temporal scale, with
significant consequences for the management of water
resource systems (Conway et al., 2009). Throughout Africa,
this variability brings significant implications for society
and causes widespread acute human suffering and economic
damage. Examples of variability include prolonged periods
of high flows for rivers draining large parts of East and
central Africa (Conway, 2002). Most parts of East Africa
cover the Nile river basin and are sensitive to climate vari-
ations as shown in previous studies (Beyene et al., 2010;
Kim and Kaluarachchi, 2009;Setegn et al., 2011) implying
that climate change will have a considerable impact on the
resource. About 70% of the Nile flow is from the three major
subbasins in the Eastern Nile Basin [the Blue Nile, the
Tekeze-Setit-Atbara, and the Baro-Akobo (Sobat) Rivers],
and located at the North-eastern portion of the Eastern Nile
Basin (Arsano and Tamrat, 2005;Swain, 1997). As the
clear majority of agriculture is rain-fed in the Eastern Nile,
precipitation plays a pivotal role (Block et al., 2008). Pre-
cipitation also plays another equally important role in the
Ethiopian highlands, feeding the headwaters of the Blue
Nile and Atbara Rivers, which eventually supply the Nile
River. Policy and planning tools, including water man-
agement, economic, hydropower, and irrigation models
for Ethiopia and other downstream countries rely heavily
on precipitation and streamflow as critical parameters.
Any rise or fall in the annual and seasonal rainfall in
this region leads to stress on average yearly runoff flow
in the area with consequent implication on the planning
of water resource, water allocation, and overall integrated
development of the basin. Changes in precipitation directly
affect water resources management, hydrology, agri-
culture, natural ecosystem, food security, water quality
and quantity, and human health. For this reason, it is vital
to understanding and predicting trends and variability of
precipitation and streamflow to improve water resources
management strategies (Block et al., 2008;Cannarozzo
et al., 2006).
Many studies conducted to characterize trends and var-
iabilit of rainfall and streamflow time series at various loca-
tions across the globe (Casanueva et al., 2013;Melesse
et al., 2011;Moges et al., 2014). These studies have adopted
several statistical techniques to quantify increasing or
decreasing trends in annual and monthly rainfall.
Most studies used trend detection techniques like Mann
Kendal, Spearman rho, and linear regression tests to under-
stand rainfall and streamflow trend and variability in the
Extreme Hydrology and Climate Variability.https://doi.org/10.1016/B978-0-12-815998-9.00010-5
©2019 Elsevier Inc. All rights reserved. 103
Eastern Nile Basin. Many studies on Blue Nile basin have
been conducted to detect changing pattern and amounts of
rainfall and streamflow in the last decades (Conway and
Hulme, 1993;Mengistu et al., 2014;Tabari et al., 2015;
Taye and Willems, 2012). All these studies on rainfall do
not show any consistent pattern or trends. Conway and
Hulme (1993) reported declining annual rainfall over the
Blue Nile, and Tekeze-Atbara basins resulting in a reduction
of river flows between the years of 1945 and 1984. In con-
trast, recent investigations by Tabari et al. (2015) reported
that rainfall over the Upper Blue Nile basin did not show
a statistically significant trend. Whereas streamflow showed
both statistically significant increasing and decreasing
trends in annual and seasonal time scale at the different Blue
Nile gauging stations for the last 40 years (19642005).
Mengistu et al. (2014) showed that annual and seasonal
rainfall for 19812010 in the Upper Blue Nile River basin
showed statistically nonsignificant increasing trends except
spring season that shows declining trend in the northeastern
part of the basin.
In Tekeze River basin part of Eastern Nile, there are little
local level studies about rainfall and streamflow trends. Long-
term trend analysis of rainfall in some stations covering a
small part of Tekeze basin was studied (Gidey et al., 2013;
Kiros et al., 2017;Seleshi and Zanke, 2004;Tesfay et al.,
2017). Seleshi and Zanke (2004) investigated the pattern of
rainfall over the upper part of Tekeze River basin by consid-
ering only one climatic station. Their result demonstrated that
the amount of rainfall remained constant for the past 40 years
(19622002). Kiros et al. (2017) investigated a decline of
rainfall at Geba catchment, a tributary of Tekeze River,
using MannKendall trend for the last 40 years. Gidey
et al. (2013) indicate that there was a slight decrease in
rainfall on upper Tekeze basin considering only part of
Tigray region from 1954 to 2008. However, all the above
studies are specific to catchment level without including
rainfall stations in the southern parts of Tekeze basin.
In pursuit of detecting the trend and the shift of trend in
hydro-meteorological variables, various statistical methods
have been developed and used over the years in Ethiopia
(Cheung et al., 2008;David et al., 1999;Tabari et al.,
2015;Tesemma et al., 2010). There are various parametric
and nonparametric tests used for identifying trends in hydro-
meteorological time series. However, in recent studies,
nonparametric tests are mostly used for non-normally dis-
tributed and censored data, including missing values, which
frequently encountered in hydro-climatological time series.
The nonparametric MannKendall statistical test (Kendal,
1975;Mann, 1945) has commonly been used to quantify the
significance of monotonic trends in hydro-meteorological
time series (Huang et al., 2014;Shifteh Some’e et al.,
2012). The MannKendall test does not provide an estimate
of the magnitude of the trend itself. For this purpose, another
nonparametric method referred to as the Sen’s slope estimator
approachis very popular by researchers to quantifyslope of the
pattern (magnitude) (Gocic and Trajkovic, 2013;Sen, 1968).
Sen’s slope estimator uses more robust slope estimate than
least-squares method because it is insensitive to outliers or
extreme values and competes well against simpleleast squares
even for normally distributed data in a time series. However,
these cited trend detection methods are not designed to detect
duration or timing of changes. The timing of a shift and
change-point tests have been performed in association with
a trend test, using various methods, such asthe nonparametric
Pettitt test (Pettitt, 1979). The Pettitt test can detect potential
change points in the mean of time series data; it has been
widely used with precipitation and streamflow data (Mu
et al., 2007;Villarini et al., 2011). Change-point tests are also
performed separately from trend tests to provide information
on both the significance and timing of the change. However,
there is no mean for those studies to differentiate between
abrupt (a step change) and gradual (a trend) change rigorously
and consistently unless the nature of the shift is visually
obvious.
Natural streamflow of rivers worldwide are signifi-
cantly altered due to construction of reservoirs, weirs
and other hydraulic structures for irrigation, hydropower,
industry, and domestic uses (Wang et al., 2016). Reservoir
operation has significant impacts on river hydrology,
primarily through changing the magnitude, frequency,
duration, and timing of flow regime. Indicators of hydro-
logic alteration (IHA) are widely used to assess river
regime streamflow alteration due to dam construction
(Richter et al., 1997). IHA is statistical software developed
by the US Nature Controversy to assess the change in
hydrologic conditions (e.g., River flows, lake levels) over
time including changes caused by water management
activities (dams, diversions, and ground water pumping),
climate shifts and land use change due to human activities
(Maingi and Marsh, 2002). The IHA method embodies the
range of variability approach (RVA) proposed by Richter
et al. (1997),whichhasbeensuccessfullyusedtoevalu-
ate hydrological disturbance in many regulated rivers
around the world ( Jiang et al., 2014). The RVA assess
the extent to which flow conditions are altered after
dam construction (disturbance) lies with in a user defined
target range of flow conditions that typified the hydro-
logical regime in preimpact periods. The IHA assess
hydrological alterations based on 33 parameters charac-
terized by streamflow magnitude, timing, frequency,
duration, and rate of change which is essential for under-
standing and predicting the impact of altered flow
regimes in the river.
The objective of this study was to investigate the change
and variability of long-term historical precipitation and
hydrological data in the Tekeze basin and to evaluate
Tekeze River streamflow regime change caused by Tekeze
hydropower reservoir dam operation.
104 Extreme hydrology and climate variability
10.2 Datasets used
The basic datasets that are required for the trend analysis test
and IHA for trend, change point detection, and streamflow
regime change for Tekeze basin are precipitation and
streamflow data.
Examination of climate trends and variability needs
long and high quality records of climatic variables. There
are now many precipitations recording stations located in
Tekeze basin. However, only a few stations have con-
tinuous records essential for hydro-climatic studies. In this
study, long periods of daily, monthly, and annual precip-
itation recorded data corresponding to 11 selected sites
out of several stations available for Tekeze River basin
collected by the Ethiopian National Metrological Service
Agency (ENMSA) were used. The data was analyzed to
determine whether there is evidence of specific trends in
the characteristics of interannual and annual rainfall events
in the basin. The period of record was from 1953 to 2013
with varying record length. The length of record varies due
to differences in site establishment and data gaps. To check
on the spatial coherence of the variability results across the
study area, length of rainfall data record was considered.
Some potential data problems, for instance, missing values,
data entry errors, outliers, etc., were solved by careful
inspection. Spatial distributions of the 11 stations are
shown in Fig. 10.1, and their characteristics and data avail-
ability is presented in Table 10.1.
There are >20 streamflow gauging stations in Tekeze
basin covering small tributaries. Streamflow data was
obtained from the Department of Hydrology—Ethiopian
Ministry of Water Irrigation and Electricity. But all stations
except Embamadre station cannot be used for the analysis
because of massive data gaps during the civil war, particu-
larly from the end of 1970s to mid-1980s. After this period,
most stations were not in operation for long periods of time.
The remaining station which are operational after 1990,
have unreliable discharge data mainly due to the short
period of record and high missing data. The minimum
length of streamflow record is 21 years (19942014) at
Embamadre and <15 years (19982015) for the remaining
stations. Based on the quality of the data, time series length,
influence of infrastructure (dams), and spatial distribution,
Embamadre station was selected for detailed analysis.
The temporal resolution of available data is daily.
10.3 Methods used for trend and variability
analysis
10.3.1 Trend analysis methods
In hydro-climatic time series analysis trend detection tests
are classified as parametric and nonparametric methods.
Parametric trend tests require data to be independent and
normally distributed, while nonparametric trend tests require
FIG. 10.1 Location of Tekeze basin and the stations selected for the study.
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 105
only that the data be independent. In this study, trend detec-
tion was carried out by the MannKendall test for deter-
mining the approximate year of the beginning of the
significant trends at the 95% confidence level, the Theil
Sen’s estimator for the trend magnitudes, and the Pettit test
to determine the change point detection. Brief explanations
of these methods are as follows:
10.3.1.1 Mann–Kendall test
The nonparametric MannKendall test (Kendal, 1975;
Mann, 1945) is widely used to evaluate statistically signif-
icant trends in hydro-meteorological time series (Gocic and
Trajkovic, 2013;Shifteh Some’e et al., 2012). Like many
other trend methods, the MannKendall test assumes that
the time series data are stable, independent, and random
with equal probability distributions. It has the advantage
of being a simple calculation that assumes no special data
distribution.
In the MannKendall test, the null hypothesis H0 states
that the data x1; x2; ;xn is a sample of n random variable
independent and distributed identically without considering
seasonal changes. The alternative hypothesis H1 of a two-
sided test is that the distributions of x
k
and x
j
are not iden-
tical for all k,jnwith kj. The test statistic Sis given by.
S¼X
n1
k¼1X
n
j¼k+1
sgn xjxk
 (10.1)
where nis number of data points, x
i
and x
j
are the data values
in time series iand j(j>i), respectively and sgn (x
j
x
i
)is
the sign function as.
sgn xjxk

¼
+1 if xjxk>0
0ifxjxk¼0
1ifxjxk<0
8
>
<
>
:
(10.2)
The variance is computed as.
Var SðÞ¼
nn1ðÞ2n+5ðÞ
X
m
i¼1
titi1ðÞ2ti+5ðÞ
"#
b
18
(10.3)
where nis number of data points; mis the number of tied
groups which has a set of sample data with same value
and t
i
is the number of ties for the ivalue. When the sample
size n>10, the standard normal variable, Z, is computed
from the following equation (Douglas et al., 2000):
Z¼0,
SL
ffiffiffiffiffiffiffiffiffiffiffiffiffi
Var SðÞ
p,ifS>0
S+L
ffiffiffiffiffiffiffiffiffiffiffiffiffi
Var SðÞ
p,ifS<0
if S¼0
8
>
>
>
<
>
>
>
:
(10.4)
A hypothesis test based on normalized Kendall’s sta-
tistics for a significance level of ais used to analyze all vari-
ables in the MannKendall test. The null hypothesis, H0, is
accepted at the significance level of ain the two-sided trend
test if Za/2 <Z<Za/2, where Za/2 are standard normal
deviates. Alternatively, H0 is rejected or Zis statistically
significant if Z>Za/2 or if Z<Za/2. Moreover, positive
values of Zindicate an increasing trend while a negative
Zreflects a decreasing trend. In most of the researches
around the globe, changing trends are tested at 0.01, 0.05,
TABLE 10.1 Location of weather stations and lengths of precipitation series used in this study
Station Lat. (o) Long. (o) Altitude (masl) P
mean
(mm) Period of record
Axum 14.12 38.73 2105 723.94 19922012
Debre Tabor 11.53 38.02 2690 1439.04 19882013
Gonder 12.33 38.02 1967 1175.18 19532004
Hager Selam 13.39 39.09 2000 692.49 19942012
Hawzen 13.58 39.26 2242 531.90 19712012
Korem 12.31 39.31 3000 980.50 19852012
Lalibela 12.31 39.03 2500 799.07 19762004
Maichew 12.48 39.32 2400 733.03 19712012
Mekele 13.3 39.29 2070 603.68 19802012
Nefas Mewucha 11.44 38.27 3000 1103.41 19862004
Wukro 13.46 39.36 2070 581.29 19922012
106 Extreme hydrology and climate variability
and 0.1 significance levels. The null hypothesis of no trend
is rejected if [Z]>1.65, [Z]>1.96, and [Z]>2.57 at the
10%, 5%, and 1% significance levels, respectively.
In this study, significance level a¼0.05 is used as in
the Nile Basin hydro-climatic trend analysis (Tabari
et al., 2015). The Zvalues are approximately normally dis-
tributed, and a positive Zvalue larger than 1.96 (based on
normal probability tables) denotes a significant increasing
trend at the significance level of 0.05, whereas a negative
Zvalue lower than 1.96 shows a significant decreasing
trend.
10.3.1.2 Sen’s slope estimator
The Sen’s slope method is a nonparametric, linear slope
estimator that works most effectively on monotonic data.
Sen’s slope method is used to determine the magnitude of
the trend line. Sen’s slope calculates the slope as a change
in measurement per change in time. The slope of a trend in a
sample of N pairs of data estimated as.
Qi¼xjxk
jkfor i¼1,,N, (10.5a)
where Q
i
is slope between data points, x
j
and x
k
are the data
values at times j and k (j>k), respectively. If there is only
one datum in each period, then.
N¼nn1ðÞ
2(10.5b)
where nis the number of time periods. If there are multiple
observations in one or more-time periods, then.
N<nn1ðÞ
2(10.5c)
where nis the total number of observations.
The Nvalues of Q
i
are ranked from smallest to largest,
and the median of slope or Sen’s slope estimator calculated
as
Qmed ¼
QN+L
2
ðÞ if Nis odd
QN
2
ðÞ
+QN+2
2
ðÞ
2if Nis even
8
>
<
>
:
(10.6)
The Q
med
sign reflects data trend reflection, while its
value indicates the steepness of the trend. To determine
whether the median slope is statistically different than zero,
one should obtain the confidence interval of Q
med
at specific
probability.
The confidence interval about the time slope can be
computed as follows:
Ca¼ZLa=2ffiffiffiffiffiffiffiffiffiffiffiffiffi
Var SðÞ
p(10.7a)
where Var(S) defined in Eq. (10.3) and Z
1a/2
is obtained
from the standard normal distribution table. In this study,
the confidence interval was computed at significance level
a¼0.05. Then M
1
and M
2
are computed as follows:
M1¼NCa
2(10.7b)
M2¼N+Ca
2(10.7c)
The lower and upper limits of the confidence interval,
Q
min
, and Q
max
are the M
1
th largest and the (M
2
+ 1)th
largest of the Nordered slope estimates (Gilbert, 1987).
The slope Q
med
is statistically different than zero if
the two limits (Q
min
and Q
max
) have a similar sign. When
hypothesis of no trend is rejected by MannKendall test,
the Sen’s slope is used to quantify the trend. Sen’s slope esti-
mator is widely used in hydro-meteorological time series
(Huang et al., 2014;Shifteh Some’e et al., 2012).
10.3.1.3 Pettit’s test for change-point detection
This test, developed by Pettitt (1979) is a nonparametric test,
which is useful for evaluating the occurrence of abrupt
changes in hydrological and climatic records with continuous
data. One of the reasons for using this test is that it is more
sensitive to breaks inthe middle of the time series and mostly
used to detect change point in climatic records (Smadi
et al., 2006). This method detects a significant change in
the mean of a time series when the exact time of the change
is unknown. The test uses a version of the MannWhitney
statistic U
t,N
, that tests whether two sample sets X1, ,Xt
and Xt+1, ,XN are from the same population. The test
statistic U
t,N
is given by:
Ut,N¼Ut1,N+X
N
J¼1
sgn XtXj
(10.8)
where t¼2, 3, ,Nand
if XtXj

>0, sgn XtXj

¼1
if XtXj

¼0, sgn XtXj

¼0
if XtXj

<0, sgn XtXj

¼1
(10.9)
The test statistic counts the number of times a member of
the first sample exceeds a member of the second sample.
The test statistic KN and the associated probability (P) used
in the test is given as.
KN¼max1tNUt,N
jj (10.10)
P2exp 6KN
ðÞ
2
N3+N2
ðÞ
() (10.11)
The significance probability of K
N
is approximated by
p0.05.
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 107
10.3.2 Indicators of hydrologic alteration
and range of variability approach
The Nature Conservancy developed IHA method to enable
rapid processing of daily hydrologic records to charac-
terize natural flow conditions and evaluate human induced
changes to flow regimes (Richter et al., 1996;Yu et al.,
2016;Zhang et al., 2014). The program was designed to
calculate the value of 33 hydrologic parameters that char-
acterize daily, monthly, and annual flow regime by five
hydrologic features: (i) magnitude of monthly discharge,
(ii) magnitude and duration of annual extreme flows,
(iii) timing of annual extreme discharge, (iv) frequency
anddurationofhighandlowpulses,and(v)rateandfre-
quency of discharge change (Table 10.2). Range vari-
ability approach (RVA) method is used to assess the
hydrological regime alterations for regulated rivers based
on IHA (Yu et al., 2016). In the RVA analysis, hydro-
logical parameters were calculated using parametric (mean
and standard deviation) or nonparametric (median and per-
centile) statistics. For most situations, nonparametric sta-
tistics are a better choice, because of the skewed nature
of many hydrologic datasets. But parametric statistics
may be preferable for certain situations such as flood fre-
quency or average monthly flow volumes. IHA detail
description can be found in Richter et al. (1996, 1997,
1998).
In this study, Embamadre gauge records were analyzed
using the IHA methods to determine hydrologic shifts of
Tekeze River basin streamflow in response to Tekeze
hydropower reservoir dam construction and operation. Ana-
lyses were conducted on mean daily discharges for the water
year (JuneMay) for the period of record prior to Tekeze
hydropower dam construction (reference) and then again
after dam construction completed (disturbance). For the
case of Tekeze River the historical (19942008) and post-
dam construction (20092014) hydrologic conditions were
evaluated. As the analysis focused on two-time intervals of
dam’s preimpact and postimpact periods, the RVA was used
to evaluate hydrological change (Gao et al., 2013;Richter
et al., 1997;Wang et al., 2016). The RVA uses the pre-
impact variation of IHA parameter values as references
for defining the extent to which the flow regime has been
altered by dam construction (Richter et al., 1998). Based
on these references, RVA analysis generates a series of
hydrologic alteration (HA) factors, which quantify the
degree of alteration of 33 IHA parameters.
The HAs in the RVA analysis are evaluated by com-
paring the frequency with which preimpact and postimpact
variables (usually the IHA) fall within the three categories.
TABLE 10.2 Summary of hydrologic parameters used in the IHA and their characteristics
IHA statistics group
Hydrologic
characteristics
Parameters used in Tekeze
basin streamflow
Total no. of (m
3
/s)
parameters
1. Magnitude of monthly discharge
condition
Magnitude, timing Median discharge for each
calendar month
12 parameters
2. Magnitude and duration of annual
extremes discharge condition
Magnitude,
duration
Annual maxima 1-, 3-, 7-, 30-, and
90-day means
12 parameters
Annual minima 1-, 3-, 7-, 30-,
and 90-day means
Number of zero days, base flow
index
3. Timing of annual extremes discharge
condition
Timing Julian date of each annual 1-day
maximum
2 parameters
Julian date of each annual 1-day
minimum
4. Frequency and duration of high and low
pulses
Magnitude,
frequency
Number of high pulses each year
Number of low pulses each year
Duration (days) Median duration of high pulses
each year
4 parameters
Median duration of low pulses
each year
5. Rate and frequency of hydrograph
changes
Frequency, rate of
change
Number of rises and falls 3 parameters
Number of reversals
108 Extreme hydrology and climate variability
Rechiter et al. (1998) divide the range of HAs in to three
classes of equal ranges with a distinct pattern as no
alteration (0%33%), moderate alteration (34%67%),
and high degree of alteration (68%100%). HA assumed
to occur if the number of postimpact values falling in the
central interval (34th to 67th percentile) differs from the
expected ones, that is, the number of the preimpact values.
The RVA target range for each parameter is bracketed by
25th and 75th percentile values of the preimpact daily flow.
The deviation of the postimpact flow regime from the pre-
impact period is quantified using HA of Tekeze River. The
degree of HA, is calculated for each variable as.
HA ¼NoNe
Ne
x100 (10.12)
Ne¼pxNT(10.13)
where N
o
is observed number, N
e
is expected number and p
is percentage of postimpact years for which the values of
hydrologic parameters falls within the RVA target range,
and N
T
is total number of postimpact years. When the
observed frequency of postimpact annual values falling with
in the RVA target range equals the expected frequency, HA is
equal to zero. A positive HA factor means that the frequency
of values in the category has increased from the preimpact to
postimpact period (maximum is 2), while a negative value
means that the frequency of values has decreased (minimum
is 1). The coefficient of dispersion (CD) is a commonly
used indicator to evaluate the variability of daily streamflow.
It is calculated as follows:
CD ¼75th_P25th_P
50th_P(10.14)
where 75th_P is 75th percentile; 25th_P is 25th percentile,
and 50th_P is 50th percentile.
10.4 Result and discussion
10.4.1 Preliminary analysis
The precipitation data screened and comparisons between
stations were made using the statistical metrics mean,
standard deviation (STD), coefficient of variation (CV),
skewness (Cs), and kurtosis (Ku). The mean annual
rainfall varied between 581.29mm in the Northern part
of the Tekeze River basin (Wukro) and 1439.04mm in
the Southwest part (Nefas Mewucha) of this basin. The
skewness (Cs), which is a measure of asymmetry in a fre-
quency distribution around the mean, varied between 1.47
and 2.41, positive skewness indicating that annual precip-
itation during the period is asymmetric and it lies to the
right of the mean of all the stations. Kurtosis (Ku) is a sta-
tistic describing the shape of the peak of a symmetrical
frequency distribution, for Tekeze basin it varied from
0.65 to 5.44 (Table 10.3). For time series data to be con-
sidered normally distributed, the coefficient of skewness
and kurtosis must be equal to 0 and 3, respectively.
Table 10.3 shows that annual precipitation in Tekeze basin
is positively skewed, not normally distributed. The CV, a
statistical measure of the dispersion of data points in a data
series around the mean, was computed for all stations to
investigate spatial pattern of interannual variability of
annual precipitation over the study area. The CV varied
between 1.18 (Maichew station) and 1.86 (Wukro station).
Table 10.3 shows that stations in the Northern part of
Tekeze basin (Wukro, Mekele, Hawzen, and Axum) have
more interannual precipitation variability than the stations
observed in the South and Southwest of the basin. It can be
concluded from the results that the areas of usually heavy
precipitation are the zone of least variability and areas of
lowest precipitation are the zone of highest variability.
TABLE 10.3 Annual precipitation time series basic statistical properties of the study area climate stations
Station P
mean
(mm) P
max
(mm) STD CV Cs Ku
Axum 723.94 1067.20 88.24 1.46 1.91 3.47
Debre Tabor 1439.04 1998.40 148.15 1.24 1.32 0.65
Gonder 1175.18 1772.80 120.62 1.23 1.41 1.26
Hager Selam 692.49 900.00 82.41 1.43 1.90 3.50
Hawzen 531.90 768.50 66.91 1.51 2.07 3.90
Korem 980.50 1272.20 103.00 1.26 1.68 2.14
Lalibela 799.07 1100.10 96.12 1.44 1.93 3.47
Maichew 733.03 1051.00 72.05 1.18 1.47 1.58
Mekele 603.68 917.90 84.18 1.67 2.21 4.47
Nefas Mewucha 1103.41 1105.70 113.82 1.24 1.90 4.13
Wukro 581.29 757.81 90.07 1.86 2.41 5.44
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 109
The highest mean annual precipitation was at Debre
Tabor station (1439.04 mm) and the lowest was registered
at Hawzen station (531.90 mm). These two rainfall stations
also recorded the maximum (148.15 mm) and minimum
(66.91 mm) STDs respectively. However, maximum annual
rainfall (1998.40 mm) was recorded at Debre Tabor station
while minimum (757.81 mm) was recorded at Wukro
station. The annual rainfall series are positively skewed
for all 11 stations.
10.4.2 Precipitation trend
10.4.2.1 Annual precipitation trend
Analysis of the Tekeze basin annual precipitation time-
series using MannKendall test showed that 64% of the sta-
tions have a positive trend and the rest with a negative trend
at 95% confident level (Fig. 10.2). Annual precipitation
trend magnitude, direction, and significance in the Tekeze
River basin are shown in Figs. 10.3 and 10.4A and B. In
the Tekeze basin selected stations, the level of significance
using Zvalue shows as positive and negative nonsignificant
trends.
The Theil Sen’s Slope estimator (Qmm/yr) summarizes
the results of change per unit time of the trends detected in
the basin. The highest decreasing trend detected in the
northern part (Axum, Hawzen, and Hagere Selam stations)
whereas an increasing trend was identified in the stations in
the south (Lalibela, Nefas Mewucha, Debre Tabor, and
Korem stations) and the east (Maichew and Mekele stations)
part of Tekeze basin (Fig. 10.3). Sen’s slope indicates that
the magnitudes of the nonsignificant positive trends at
95% confident level vary in the range of 8.12mm/yr at
Lalibela station to 2.32 mm/yr at Nefas Mewucha station.
A negative nonsignificant trend in the basin varies from
1.26 mm/yr at Gonder station to 6.22 mm/yr at Axum
station. Positive trends mostly are in the eastern part of
Tekeze basin whereas the negative trends occurred in the
northern and southwestern parts of the basin. MannKendal
test results in all selected stations of Tekeze basin show a
nonsignificant increasing and decreasing annual precipi-
tation trend observed at the 5% significance level as reported
in other studies done in the Eastern Nile (Bewket and
Conway, 2007;Conway, 2000;Gebremicael et al., 2013;
Mengistu et al., 2014;Tabari et al., 2015).
10.4.2.2 Seasonal trends of precipitation
For all selected stations of Tekeze basin, MannKendall
test and Sen’s slope estimator method were also applied
to detect the temporal trends of seasonal precipitation time
series during 19532013. Similar to the annual preci-
pitation series, the seasonal time series in the Tekeze basin
showed a mix of positive and negative trends (Figs. 10.4A
and B and 10.6).
2.0
1.5
0.5
–0.5
–1.5
–2.5
–1.0
–2.0
1.0
0
Z statistics
Axum
Debre Tabor
Gonder
Hawzen
Korem
Lalibela
Maichew
Mekele
Wukr o
Nefas Mewucha
Hager Selam
FIG. 10.2 Results of MannKendall annual precipitation time series
trend test at 95% confidence interval.
10
8
4
2
0
Axum
Gonder
Hager Selam
Hawzen
Korem
Lalibela
Maichew
Mekele
Wukro
Nefas Mewucha
Debre Tabor
–4
–2
–6
–8
6
Trend magnitude (mm/year)
FIG. 10.3 Annual precipitation trend magnitudes at different stations of
Tekeze basin.
120
100
80
60
Station (%)
Station (%)
18
0
55
73
82
18 1002745
0
00
0
91
91
36
64
82
(A)
(B)
9
9
Bega
Bega
Kiremt
Kiremt
Belg
Belg
Annual
Annual
40
20
20
40
60
80
100
120
0
0
[–]
Sign [–]
Sign [+]
Non Sign
[+]
FIG. 10.4 (A) Percentage annual and seasonal trend test results overall
negative and positive trends. (B) Percentage annual and seasonal trend test
results significant positive and negative trends at 95% confidence level.
110 Extreme hydrology and climate variability
MannKendal test result for Bega season (October
January) showed negative trends (10 of 11 stations, 91%)
observed than the positive trends (Fig. 10.4A and B) in most
of the stations in Tekeze basin. However, 73% of stations
were found to have negative significant trend at 95% confi-
dence level and only Korem station was significant for
positive trend (Fig. 10.5). Based on the results of the
MannKendall test (Fig. 10.5), significant negative trends
are detected in the north and southwestern parts of Tekeze
basin. Significant negative trends in Bega season precipi-
tation varies between 0.129 mm/yr at Maichew station
to 1.33 mm/yr at Nefas Mewucha and Debre Tabor sta-
tions. But the magnitude of significant variation in the
northern part of Tekeze basin stations of Axum, Hager
Selam, Hawzen, Wukro, and Mekele is very small. In
general, Bega season has small declining precipitation
trends in the basin over the past 50 years.
Kiremt (Wet season, JuneSeptember) precipitation,
82% of the stations and 18% of the stations show positive
and negative trends, respectively (Fig. 10.4A). Only 55%
(6 stations) have significant positive trends and none with
significant negative trends (Fig. 10.4B). These significant
positive trends are mostly observed at the source of
Tekeze basin at stations Debre Tabor, Lalibela, Nefas
Mewucha in the southern mountainous areas and stations
like Maichew and Korem in the eastern part of the basin.
Kiremt precipitation trend magnitude varied between
1.19 and 3.79 mm/yr. In general, it can be stated that
the Kiremt season has been experiencing mild increasing
precipitation trend over the past 50 years. Kiremt season
trend was similar to the annual precipitation trend, which
indicates that Kiremt precipitation has high contribution
for annual precipitation in Tekeze basin.
The Belg (Small rainy season: FebruaryMarch) precip-
itation trends show an opposite nature when compared with
that of Bega season but similar to the Kiremt series, most of
the trends in the Belg precipitation time series were positive
accounting for about 91% of the stations except Korem
station (Figs. 10.6 and 10.4A). Nevertheless, the significant
positive trend in Belg precipitation was higher compared
with those in the other seasonal series. Nine significant pos-
itive trends (82%) are detected in the Belg time series
(Fig. 10.4B). This is due to the south-easterly winds from
the Indian Ocean, and the Gulf of Aden that produce the
Belg rains to the east-central part of the northwestern high-
lands of Ethiopia (Seleshi and Zanke, 2004). The magnitude
of the significant increasing trend varies from 0.47 mm/yr at
northern part (Wukro station) to 2.41 mm/yr at Southern part
(Debre Tabor station) of Tekeze basin. According to these
results, significant increasing trend in Belg precipitation
Axum
Debre Tabor
Gonder
Hager Selam
Hawzen
Korem
Lalibela
Maichew
Mekele
Nefas Mewucha
Wukro
–8
–6
–4
–2
0
MK statistics
2
4
6
8
Kiremt BelgBega
FIG. 10.5 MannKendall Z values at different Tekeze
basin stations.
Trend maganitude (mm/year)
Be
g
a Kiremt Bel
g
Debre Tabor
Axum
Gonder
Hager Selam
Hawzen
Korem
Lalibela
Maichew
Wukro
Mekele
Nefas Mewucha
6
5
4
3
2
1
0
–1
–2
7FIG. 10.6 Results of seasonal precipitation trend mag-
nitude at different weather stations.
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 111
is observed throughout Tekeze basin for the last 50 years.
The result of this study in higher variability and increasing
trend for Belg precipitation in Tekeze basin is also reported
by Cheung et al. (2008).
10.4.2.3 Change point results
Since the MannKendall tests showed significant trend in
seasonal precipitation at the significance level of 0.05, the
Pettit test was further used to detect the change points of
transitional years. For annual precipitation, there is no
change point year that could be detected at 95% confidence
level as all stations annual precipitation shows nonsignif-
icant trend. Most of the station’s precipitation in all Kiremt,
Bega, and Belg seasons shift occurred around the 1970s or
1980s either positive or negative. This is mainly due to the
drought period of Tekeze basin that lasted from 1978 to
1986. This result is confirmed by the study of Conway
(2000) and Seleshi and Zanke (2004). During the late
1970s to mid-1980s extremely low precipitation was
recorded in Ethiopia and significant shift occurred during
that period.
10.4.3 Streamflow trend and variability
Streamflow is a very useful indicator of long-term hydro-
climatic changes. From a water resources management
perspective, the identification of trend and variability in
streamflow are critical for planning purposes. Trend analysis
is useful for understanding dynamics and behaviors of
hydrological and climatic variables over a long-term period.
The MannKendall test was applied to the annual and sea-
sonal streamflow data at Embamadre station over the period
1994 to 2014. The Zstatistic of streamflow was 0.62 and
showed a nonsignificant decreasing trend at the 5% confi-
dence level. The annual streamflow declined at a rate of
8.67 m
3
/yr. The observed decrease in yearly streamflow
was primarily from Bega season runoff.
To analyze trend and variability in seasonal streamflow
following precipitation characteristics, each year is divided
into wet season (Kiremt, JuneSeptember), dry season (Bega,
OctoberJanuary), and small rainy season (Belg, February
March). Kiremt (wet) and Belg seasons streamflow shows
significantly increasing trend with Z¼4.43 and Z¼8.14,
respectively. Bega (dry) season showed significant
decreasing trend with Z value 10.35. The magnitude of
the increasing trend in Kiremt varies up to 4.69m
3
/yr
andBelgseasonupto0.14m
3
/yr, and the decline in Bega
season is up to 0.59m
3
/yr.
This study of streamflow variability at Embamadre
gauging station is significant as all streamflows generated
from the high altitude and mountainous regions usually
reaches its maximum value at this station. It should be
pointed out that hydrological regime of the Tekeze River
mainstream is strongly affected by human activities like
Tekeze hydropower reservoir and planned irrigation, hydro-
power, and water conservation projects. Although annual
precipitation exhibited an increasing trend for the past five
decades, which in theory should result in more runoff,
streamflow, and water availability in the Tekeze main-
stream at Embamadre has decreased and the environmental
situation has been severely impaired because of limited
water resources. It may get dried-up during dry periods.
Therefore, human activities, as well as the change and var-
iability of climate all may contribute to the trends of
streamflow detected in this study. Overall, this study pro-
vides an elaborate view of past precipitation and streamflow
trends in the upper part of Tekeze basin which should be
useful for further research.
10.4.4 Influence of Tekeze reservoir
operation on the streamflow regime
Tekeze River at Embamadre gauging station’s all the 33
hydrologic parameters of medians, CD, and measure of
HA were calculated with the IHA software. The 25th and
75th percentile values were calculated based on the
available preimpact streamflow records considering low
and high boundaries of the RVA target range. The RVA
analysis showed that the natural flow regime in the middle
Tekeze River at Embamadre streamflow gauging station
significantly changed after the operation of Tekeze hydro-
power reservoir (Table 10.4).
10.4.4.1 Magnitude of monthly streamflow
The result in Fig. 10.7 indicate that the river flow has
become smoother in the postimpact period by two major
changes, a decrease in high flow and an increase in low
flow. Flow regime alterations were closely related to Tekeze
hydropower reservoir operation that stores more water in the
rainy season (July through September) and release water
downstream for low flow season power production. Tekeze
hydropower reservoir operation has altered the original
hydrologic process, smoothing peak flow, and increasing
dry season discharge of Tekeze River.
The magnitude of monthly flow from February to June,
the normal low flow period, increased after 2009 when the
reservoir behind the dam stored water for power pro-
duction. Median discharge for all months after Tekeze
hydropower dam construction differs significantly from
pre-dam construction period discharges. Thus, during the
low flow period of the year, operation of the dam increases
median discharges while decreases high flow months’
median flow. Fig. 10.8A shows that an increase in
November median flow and a Fig. 10.8B shows a decrease
in August median flow.
112 Extreme hydrology and climate variability
TABLE 10.4 Nonparametric RVA scores at Embamadre station of Tekeze River
Hydrologic parameters
Preimpact period
(1994–2008)
Postimpact period
(2009–2014) RVA targets
Medians CD Medians CD Low High IHA (%)
Parameter Group #1
July 477.7 0.64 488.6 0.84 296.3 526.2 29
August 1322 0.53 849 1.08 897.8 1400 57
September 320.3 0.52 443.1 0.46 297.3 375.5 57
October 110.8 0.79 264.9 0.65 86.87 139.4 57
November 42.9 1.07 239.8 0.86 35.82 58.96 100
December 32.5 0.84 211.3 0.67 25.77 37.46 57
January 23.1 1.23 201.5 0.94 9.72 30.16 100
February 21.3 1.02 304.1 0.48 9.1 24.9 100
March 20.2 0.91 273.7 0.42 16.6 21.9 100
April 16.1 1.5 293.2 0.76 7.95 29.15 100
May 19.1 0.85 320 0.75 13.87 26.44 100
June 58.55 1.13 312 0.69 31.4 74.95 100
Parameter Group #2
1-day minimum 3.7 1.92 64.7 2.17 2.15 8.06 100
3-day minimum 9 1.42 154.9 0.97 2.27 9.37 100
7-day minimum 9.81 1.33 184.4 0.82 2.47 13.28 100
30-day minimum 14.21 1.01 201.5 0.8 4.71 17.22 100
90-day minimum 17.12 1.13 201.5 0.88 9.97 22.7 100
1-day maximum 3033 0.5 1719 0.65 2213 3130 64
3-day maximum 2249 0.5 1451 0.43 1724 2470 14
7-day maximum 1951 0.48 1240 0.46 1447 1995 57
30-day maximum 1344 0.68 1015 0.5 1088 1553 57
90-day maximum 748.4 0.91 723.5 0.22 665.5 958 71
Base flow index 0.03 2.08 0.44 0.93 0.02 0.07 57
Parameter Group #3
Date of minimum 130 0.21 337 0.46 108.1 154.2 14
Date of maximum 224 0.05 239 0.05 220.6 233.9 29
Parameter Group #4
Low pulse count 4 1 0 0 3 5.72 64
Low pulse duration 4.5 3.11 3.5 1.43 3 12.16 69
High pulse count 4 1.25 3 3.33 3 4.72 64
High pulse duration 4.5 2.44 34.75 3.26 2 9 20
Parameter Group #5
Rise rate 39.3 1.2 33.34 0.35 23.48 45.74 114
Fall rate 3.4 0.82 18.83 0.5 4.67 2.2 100
Number of reversals 87 0.3 138 0.25 80.84 96.16 100
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 113
Preimpact period (1994–2008)
2500
2000
1500
500
0
January
February
March
April
May
June
July
August
September
October
November
December
1000
Stream flow (m3/s)
Postimpact period (2009–2015)
FIG. 10.7 Comparison of median monthly streamflow before and after Tekeze hydropower reservoir dam construction.
0
(A) 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
20
40
60
80
100
140
120
160
180
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
25th percentile
November median flow (m3/s)
Median
200
220
240
260
280
300
320
340
2200
2000
1800
1600
1400
1200
1000
800
600 Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
Median
25th percentile
400
August median flow (m3/s)
200
B
0
1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
FIG. 10.8 (A) November monthly median flow change before and after Tekeze hydropower reservoir constructionin Tekeze River. (B) August monthly
median flow change before and after Tekeze hydropower reservoir construction in Tekeze River.
114 Extreme hydrology and climate variability
10.4.4.2 Timing of annual extreme streamflow
change
The median Julian dates of each annual 1-day minimum
move forward from the 130th day in preimpact period to
the 337th day in the postimpact period, with the low alter-
ation of 14%. The median Julian dates of each annual
1-day maximum also moved forward from the 224th day
in preimpact period to the 239th day in the postimpact
period, with the low alteration of 29%.
10.4.4.3 Rate and frequency of streamflow
changes
Table 10.4 showed that there was a change in the medians of
rise rate, fall rate and number of reversals in the preimpact
and postimpact periods. The median of rise rate decreased
from 39.3 m
3
/s day in the preimpact period to 33.34 m
3
/s day
in postimpact period with high hydrological alteration of
114%. The median of fall rate also decreased by 3.4 m
3
/s
day in preimpact period to 18.83 m
3
/s day in postimpact
periods with hydrological alteration of 100%. These
changes indicate that the dam significantly decreased the rate
of rise of hydrographs, due to storage effects of the reservoir
and led to many more reversals between rising and falling
stages of flow in the river. The median of number of reversals
also has been significantly altered and increased from 87 in
preimpact to 138 in postimpact with high hydrological alter-
ation of 100%.
10.4.4.4 Magnitude and duration of annual
extreme
Tekeze River time series of 1-day, 3-day, 7-day, 30-day, and
90-day maxima and minima medians flow in the preimpact
and postimpact periods together with the median value
(dashed line) and the boundaries of the middle category
(i.e., from 25th and 75th percentile), computed with reference
to the preimpact period are shown in Figs. 10.9AD and
10.10AD. In the RVA analysis, significance differences
were observed in the annual maximum and minimum flows
in the postimpact periods. The medians of annual 1-, 3-, 7-,
30-, and 90-day minimum flow for the postimpact period
increased due to the reservoir capturing high flood season
flow for later release in the dry season for hydropower pro-
duction. By contrast, as the medians of annual 1-, 3-, 7-,
30-, and 90-day maximum flow for the postimpact period
decreased greatly due to the dampening of high magnitude
flood by storage in the reservoir.
Except for low alteration in the 90-day annual maxima,
the others were high. The hydrological alteration of annual
1-day, 3-day, 7-day, and 30-day parameter values fell within
the RVA target value whereas the minima 1-day, 3-day,
7-day, 30-day, and 90-day maxima medians reached
100%, which means most values of these five parameters
fell out of the RVA target value (Table 10.4). The dispersion
coefficients of annual minima and maxima flows in the
postimpact period ranging from 0.22 to 0.97 are generally
lower than those in the preimpact period ranging from
0.48 to 1.92. The base flow index is larger in postimpact
period due to the effect of low flow season water released
from reservoir for hydropower production when natural
flow is at its minimum. This is shown by a higher persis-
tence of annual base flow index HA ¼1.57 for the upper cat-
egory and accordingly lower persistence in lower and
middle categories by negative HA index of 1.0 and 0.57,
respectively. Therefore, the result showed that daily,
weekly, monthly, and annual maxima/minima flow cycles
were positively influenced by Tekeze hydropower reservoir
operation.
10.4.4.5 High and low pulses
Low pulse count, low pulse duration, high pulse count and
high pulse duration have been changed, with HA of 64%,
69%, 64%, and 20%, respectively. Except high pulse
duration, the median of low pulse count, low pulse duration,
and high pulse count in the postimpact period were lower
than those in the preimpact period (Table 10.4). The coef-
ficient of dispersion in the low pulse count and low pulse
duration were higher in the preimpact period, in contrast,
the high pulse count and high pulse duration value were
higher in the postimpact period. This indicates that the fre-
quency and duration of low and high flow pulses in
the Tekeze River are influenced by Tekeze hydropower
reservoir construction.
10.5 Summary and conclusion
In this study, the nonparametric MannKendall trend test
and Pettit test were used to investigate the spatiotemporal
trends and variability of precipitation data from 11 stations
in Tekeze basin on annual and seasonal timescales for the
period 19532013 and streamflow at Embamadre station
for the period 19942014.
The nonparametric MannKendall test shows that
annual precipitation has an increasing trend in southern
and south-eastern regions of Tekeze basin varying from
0.98 mm/yr to 8.12 mm/yr whereas the decreasing trend pre-
vails in the northern part of the basin ranging from 6.22 to
2.35 mm/yr. There was no significant positive or negative
trend detected in annual precipitation at a¼0.05 significant
level in this basin. The analysis of the seasonal precipitation
time series showed a mix of positive and negative trends. In
Belg and Kiremt season >80% of the precipitation stations
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 115
showed positive trend whereas Bega showed a similar per-
centage but in a decreasing trend. All the three seasons show
statistically significant increasing and decreasing precipi-
tation trends with abrupt change detected in the late 1970s
and mid-1980s. In Belg and Kiremt seasons significant pos-
itive trends were found in 82% and 55% of the stations
respectively, whereas in Bega, 73% of the stations show neg-
ative trends. The strongest positive trend of 2.14 and
6.41 mm/yr was detected in Belg and Kiremt season at Debre
Tabor station, respectively and negative trend of -
1.33 mm/yr detected in Bega season at Nefas Mewucha
station.
220
200
180
160
140
120
100
3-day minimum flow (m3/s)
80
60
40
20
0
(A) 1994 1996 1998
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
25th percentile
Median
2000 2002 2004 2006 2008 2010 2012 2014
240 Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
25th percentile
Median
220
200
180
160
140
120
7-day minimum flow (m3/s)
100
60
80
40
20
0
(
B
)
1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
FIG. 10.9 (A) Hydrological alterations of 3-day minimum flow before and after Tekeze hydropower reservoir construction in Tekeze River. (B) Hydro-
logical alterations of 7-day minimum flow before and after Tekeze hydropower reservoir construction in Tekeze River. (C) Hydrological alterations of 30-
day minimum flow before and after Tekeze hydropower reservoir construction in Tekeze River. (D) Hydrological alterations of 90-day minimum flow
before and after Tekeze hydropower reservoir construction in Tekeze River.
116 Extreme hydrology and climate variability
Annual streamflow in Tekeze basin showed a nonsignif-
icant decreasing trend at a rate of 8.67 m
3
/yr. The seasonal
streamflow showed similar projections like precipitation
both increasing and decreasing trend. Streamflow trend
increases in the Kiremt and Belg seasons whereas decreases
in Bega (dry) periods. Statistical analysis performed using
IHA at Embamadre streamflow station shows an increase
in minimum flow duration and also decrease in maximum
flow duration, fall, and rise rate. After the construction of
Tekeze Dam, the hydrology was altered with significant
decline of high flows and increase of low flows, mainly
attributed to storage in the rainy season and release in the
dry season.
The findings of this research can provide some infor-
mation to the government and community on the variability
of precipitation and streamflow for current and future
planned dams and irrigation projects. Such information also
can be used by policy makers and managers for water
resource management, hydrology, agriculture, and eco-
system management in the Tekeze River basin.
1994
0
(
C
)
20
40
60
80
100
120
140
30-day minimum flow (m3/s)
160
180
200
220
240
260
280
1996 1998
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
25th percentile
75th percentile
Median
2000 2002 2004 2006 2008 2010 2012 2014
1994
0
(
D
)
20
40
60
80
100
120
140
90-day minimum flow (m3/s)
160
180
220
200
240
280
260
320
300
1996 1998
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
25th percentile
75th percentile
Median
2000 2002 2004 2006 2008 2010 2012 2014
FIG. 10.9 CONT’D
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 117
3000
2800
2600
2400
2200
2000
3-day maximum flow (m3/s)
1800
1600
1400
1200
1000
800
400
200
1994 1996 1998 2000 2002
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
25th percentile
Median
2004 2006 2008 2010 2012 2014
0
(
A
)
600
2800
2600
2400
2200
2000
7-day maximum flow (m3/s)
1800
1600
1400
1200
1000
800
400
200
1994 1996 1998 2000 2002
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
25th percentile
Median
2004 2006 2008 2010 2012 2014
0
(
B
)
600
FIG. 10.10 (A) Hydrological alterations of 3-day maximum flow before and after Tekeze hydropower reservoir construction in Tekeze River.
(B) Hydrological alterations of 7-day maximum flow before and after Tekeze hydropower reservoir construction in Tekeze River. (C) Hydrological alter-
ations of 30-day maximum flow before and after Tekeze hydropower reservoir construction in Tekeze River. (D) Hydrological alterations of 90-day
maximum flow before and after Tekeze hydropower reservoir construction in Tekeze River.
118 Extreme hydrology and climate variability
Acknowledgment
The research presented was financially supported by the Ethiopian
Government through Addis Ababa University (Ph.D. program for
the lead author). We acknowledge the data support of Ethiopian Min-
istry of Water, Irrigation, & Electricity and the Ethiopian National
Meteorological Service Agency.
References
Arsano, Y., Tamrat, I., 2005. Ethiopia and the Eastern Nile basin. Aquat.
Sci. 67, 1527. https://doi.org/10.1007/s00027-004-0766-x.
Bewket, W., Conway, D., 2007. A note on the temporal and spatial
variability of rainfall in the drought-prone Amhara region of
Ethiopia. Int. J. Climatol. 27, 14671477. https://doi.org/10.1002/
joc.1481.
Beyene, T., Lettenmaier, D.P., Kabat, P., 2010. Hydrologic impacts of
climate, change on the Nile River basin: implications of the 2007 IPCC
scenarios. Clim. Change 100, 433461. https://doi.org/10.1007/
s10584-009-9693-0.
Block, P.J., Strzepek, K., Rosegrant, M.W., Diao, X., 2008. Impacts of
considering climate variability on investment decisions in Ethiopia.
Agric. Econ. 39, 171181. https://doi.org/10.1111/j.1574-0862.
2008.00322.x.
2400
2200
2000
1800
1600
1400
1200
1000
30-day maximum flow (m3/s)
800
600
400
200
0
(
C
)
1994 1996 1998 2000 2002
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
25th percentile
75th percentile
Median
2004 2006 2008 2010 2012 2014
1200
1100
1000
900
800
700
600
90-day maximum flow (m3/s)
500
400
300
200
100
(D) 0
1994 1996 1998 2000
Preimpact flows (1994–2008)
Postimpact flows (2009–2015)
75th percentile
Median
25th percentile
2002 2004 2006 2008 2010 2012 2014
FIG. 10.10 CONT’D
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 119
Cannarozzo, M., Noto, L.V., Viola, F., 2006. Spatial distribution of rainfall
trends in Sicily (19212000). Phys. Chem. Earth Parts ABC
31, 12011211. https://doi.org/10.1016/j.pce.2006.03.022.
Casanueva, A., Rodriguez-Puebla, C.D., Frı
´as, M., Gonza
´lez-
Reviriego, N., 2013. Variability of extreme precipitation over Europe
and its relationships with teleconnection patterns. Hydrol. Earth Syst.
Sci. 10, 1233112371. https://doi.org/10.5194/hessd-10-12331-2013.
Cheung, W.H., Senay, G.B., Singh, A., 2008. Trends and spatial distri-
bution of annual and seasonal rainfall in Ethiopia. Int. J. Climatol.
28, 17231734. https://doi.org/10.1002/joc.1623.
Conway, D., 2000. Some aspects of climate variability in the north-east
Ethiopian highlands—Wollo and Tigray. Ethiop. J. Sci. 23, 139161.
https://doi.org/10.4314/sinet.v23i2.18163.
Conway, D., 2002. Extreme rainfall events and Lake level changes in east
Africa: recent events and historical precedents. In: The East African
Great Lakes: Limnology, Palaeolimnology and Biodiversity, Advances
in Global Change Research. Springer, Dordrecht, pp. 6392. https://
doi.org/10.1007/0-306-48201-0_2.
Conway, D., Hulme, M., 1993. Recent fluctuations in precipitation and
runoff over the Nile sub-basins and their impact on main Nile dis-
charge. Clim. Change 25, 127151.
Conway, D., Persechino, A., Ardoin-Bardin, S., Hamandawana, H.,
Dieulin, C., Mah
e, G., 2009. Rainfall and water resources variability
in sub-Saharan Africa during the twentieth century. J. Hydrometeorol.
10, 4159. https://doi.org/10.1175/2008JHM1004.1.
David, S., Randolph, B., Upali, A., 1999. Water scarcity in the twenty-first
century. Int. J. Water Resour. Dev. 15, 2942. https://doi.org/10.1080/
07900629948916.
Dawadi, S., Ahmad, S., 2013. Evaluating the impact of demand-side man-
agement on water resources under changing climatic conditions and
increasing population. J. Environ. Manage. 114, 261275. https://
doi.org/10.1016/j.jenvman.2012.10.016.
Douglas, E.M., Vogel, R.M., Kroll, C.N., 2000. Trends in floods and low
flows in the United States: impact of spatial correlation. J. Hydrol.
240, 90105. https://doi.org/10.1016/S0022-1694(00)00336-X.
Gao, B., Yang, D., Yang, H., 2013. Impact of the three gorges dam on flow
regime in the middle and lower Yangtze River. Quat. Int. 304, 4350.
Gebremicael, T., Mohamed, G., Betrie, P., van der Zaag, Teferi, E., 2013.
Trend analysis of runoff and sediment fluxes in the upper Blue Nile
basin: a combined analysis of statistical tests, physically-based models
and land use maps. J. Hydrol. 482, 5768.
Gidey, T., Gebrehiwot, T., van der, V.A., 2013. Assessing the evidence of
climate variability in the northern part of Ethiopia. J. Dev. Agric. Econ.
5, 104119. https://doi.org/10.5897/JDAE12.056.
Gilbert, R.O., 1987. Statistical Methods for Environmental Pollution. John
Willey & Sons, New Jersey.
Gocic, M., Trajkovic, S., 2013. Analysis of changes in meteorological vari-
ables using Mann-Kendall and Sen’s slope estimator statistical tests in
Serbia. Global Planet. Change 100, 172182. https://doi.org/10.1016/j.
gloplacha.2012.10.014.
Huang, J., Sun, S., Xue, Y., Li, J., Zhang, J., 2014. Spatial and temporal var-
iability of precipitation and dryness/wetness during 19612008 in
Sichuan province, West China. Water Resour. Manag. 28, 16551670.
https://doi.org/10.1007/s11269-014-0572-8.
Jiang, L., Ban, X., Wang, X., Cai, X., 2014. Assessment of hydrologic alter-
ations caused by the three gorges dam in the middle and lower reaches
of Yangtze River, China. Water SA 6, 14191434. https://doi.org/
10.3390/w6051419.
Kendal, M., 1975. Rank Correlation Methods, fourth ed. Charles Griffin,
London (WWW Document). https://www.abebooks.com/book-search/
title/rank-correlation-methods/. [(Accessed 1 May 2018)].
Kim, U., Kaluarachchi, J.J., 2009. Climate change impacts on water
resources in the upper Blue Nile River basin, Ethiopia. J. Am. Water
Resour. Assoc. 45, 13611378. https://doi.org/10.1111/j.1752-
1688.2009.00369.x.
Kiros, G., Shetty, A., Nandagiri, L., 2017. Extreme rainfall signatures under
changing climate in semi-arid northern highlands of Ethiopia. Cogent
Geosci. 3https://doi.org/10.1080/23312041.2017.1353719.
Maingi, J., Marsh, S., 2002. Quantifying hydrologic impacts following dam
construction along the Tana River, Kenya. J. Arid Environ. 50, 5379.
Mann, H.B., 1945. Nonparametric tests against trend. Econometrica
13, 245259. [WWW Document]. Eur. Environ. Agency, https://
www.eea.europa.eu/data-and-maps/indicators/nutrients-in-freshwater/
mann-h.b.-1945-nonparametric-tests. [(Accessed 1 May 2018)].
McCartney, M.P., Menker Girma, M., 2012. Evaluating the downstream
implications of planned water resource development in the Ethiopian
portion of the Blue Nile River. Water Int. 37, 362379. https://doi.
org/10.1080/02508060.2012.706384.
Melesse, A., Bekele, S., McCornick, P., 2011. Hydrology of the Niles in the
face of land-use and climate dynamics. In: Melesse, A.M. (Ed.), Nile
River Basin: Hydrology, Climate and Water Use. Springer.
Mengistu, D., Bewket, W., Lal, R., 2014. Recent spatiotemporal temper-
ature and rainfall variability and trends over the Upper Blue Nile River
basin, Ethiopia. Int. J. Climatol. 34, 22782292. https://doi.org/
10.1002/joc.3837.
Middelkoop, H., Daamen, K., Gellens, D., Grabs, W., Kwadijk, J.C.J.,
Lang, H., Parmet, B.W.a.H., Sch
adler, B., Schulla, J., Wilke, K.,
2001. Impact of climate change on hydrological regimes and water
resources management in the Rhine basin. Clim. Change 49, 105128.
https://doi.org/10.1023/A:1010784727448.
Milly, P., Betancourt, J., Falkenmark, M., Hirsch, R.M., Kundzewic, Z.W.,
Lettenmaie, D.P., Stouffer, R.J., 2008. Stationarity is dead: whither
water management? Science 319, 573574.
Moges, S.A., Taye, M.T., Willems, P., Gebremichael, M., 2014. Excep-
tional pattern of extreme rainfall variability at urban centre of Addis
Ababa, Ethiopia. Urban Water J. 11, 596604. https://doi.org/
10.1080/1573062X.2013.831914.
Mosquera-Machado, S., Ahmad, S., 2007. Flood hazard assessment of
Atrato River in Colombia. Water Resour. Manag. 21, 591609.
https://doi.org/10.1007/s11269-006-9032-5.
Mu, X., Zhang, L., McVicar, T.R., Chille, B., Gau, P., 2007. Analysis of the
impact of conservation measures on stream flow regime in catchments
of the Loess Plateau, China. Hydrol. Process. 21, 21242134. https://
doi.org/10.1002/hyp.6391.
Pettitt, A.N., 1979. A non-parametric approach to the change-point
problem. J. R. Stat. Soc. C Appl. Stat. 28, 126135. https://doi.org/
10.2307/2346729.
Qaiser, K., Ahmad, S., Johnson, W., Batista, J., 2011. Evaluating the impact
of water conservation on fate of outdoor water use: a study in an arid
region. J. Environ. Manage. 92, 20612068. https://doi.org/10.1016/j.
jenvman.2011.03.032.
Richter, B., Baumgartner, J., Powell, J., Braun, D., 1996. A method for
assessing hydrologic alteration within ecosystems. Conserv. Biol.
10, 11631174.
Richter, B., Baumgartner, J., Wigington, R., Braun, D., 1997. How much
water does a river need? Freshwater Biol. 37, 231249.
120 Extreme hydrology and climate variability
Richter, B., Baumgartner, J., Braun, D., Powell, J., 1998. A spatial
assessment of hydrologic alteration within a river network. Regul.
Rivers: Res. Manage. 14, 329340.
Seleshi, Y., Zanke, U., 2004. Recent changes in rainfall and rainy days in
Ethiopia. Int. J. Climatol. 24, 973983. https://doi.org/10.1002/
joc.1052.
Sen, P.K., 1968. Estimates of the regression coefficient based on Kendall’s
tau. J. Am. Stat. Assoc. 63, 13791389. https://doi.org/10.1080/
01621459.1968.10480934.
Setegn, S.G., Rayner, D., Melesse, A.M., Dargahi, B., Srinivasan, R.,
2011. Impact of climate change on the hydroclimatology of Lake
Tana basin, Ethiopia. Water Resour. Res. 47. https://doi.org/
10.1029/2010WR009248.
Shifteh Some’e, B., Ezani, A., Tabari, H., 2012. Spatiotemporal trends and
change point of precipitation in Iran. Atmos. Res. 113, 112. https://
doi.org/10.1016/j.atmosres.2012.04.016.
Smadi, M.M., Zghoul, A., Smadi, M.M., Zghoul, A., 2006. A sudden
change in rainfall characteristics in Amman, Jordan during the
mid-1950s. Am. J. Environ. Sci. 2, 8491. https://doi.org/10.3844/
ajessp.2006.84.91.
Swain, A., 1997. The Nile River dispute: Ethiopia, the Sudan, and Egypt. J.
Mod. Afr. Stud. 35 (4), 675694. December 1997.
Tabari, H., Taye, M.T., Willems, P., 2015. Statistical assessment of precip-
itation trends in the upper Blue Nile River basin. Stoch. Environ. Res.
Risk A. 29, 17511761. https://doi.org/10.1007/s00477-015-1046-0.
Taye, M.T., Willems, P., 2012. Temporal variability of hydroclimatic
extremes in the Blue Nile basin. Water Resour. Res. 48, https://doi.
org/10.1029/2011WR011466 W03513.
Tesemma, Z.K., Mohamed, Y.A., Steenhuis, T.S., 2010. Trends in rainfall
and runoff in the Blue Nile Basin: 19642003. Hydrol. Process.
24, 37473758. https://doi.org/10.1002/hyp.7893.
Tesfay, G., Mohamed, Y., van der Zaag, P., Yazew, E., 2017. Temporal and
spatial changes of rainfall and streamflow in the Upper TekezeAtbara
river basin, Ethiopia. Hydrol. Earth Syst. Sci. 21 (4), 21272142.
https://doi.org/10.5194/hess-21-2127-2017.
Vedwan, N., Ahmad, S., Miralles-Wilhelm, F., Broad, K., Letson, D.,
Podesta, G., 2008. Institutional evolution in Lake Okeechobee man-
agement in Florida: characteristics, impacts, and limitations. Water
Resour. Manag. 22, 699718. https://doi.org/10.1007/s11269-007-
9187-7.
Villarini, G., Smith, J.A., Ntelekos, A.A., Schwarz, U., 2011. Annual
maximum and peaks-over-threshold analyses of daily rainfall accumu-
lations for Austria. J. Geophys. Res. Atmos. 116. https://doi.org/
10.1029/2010JD015038.
Wang, Y., Rhoads, B.L., Wang, D., 2016. Assessment of the flows regime
alterations in the middle reach of the Yangtze River associated with
dam construction: potential ecological implication. Hydrol. Process.
30, 39493966. https://doi.org/10.1002/hyp.10921.
Wu, G., Li, L., Ahmad, S., Chen, X., Pan, X., 2013. A dynamic model for
vulnerability assessment of regional water resources in arid areas: a
case study of Bayingolin, China. Water Resour. Manag.
27, 30853101. https://doi.org/10.1007/s11269-013-0334-z.
Yu, C., Yin, X., Yang, Z., 2016. A revised range of variability approach for
the comprehensive assessment of the alteration of flow regime. Ecol.
Eng. 96, 200207.
Zhang, Q., Xiao, M., Liu, C.-L., Singh, V.P., 2014. Reservoir induced
hydrological alterations and environmental flow variation in the East
River, the Pearl River basin, China. Stoch. Environ. Res. Risk A.
28 (8), 21192213.
Further reading
Tabari, H., Taye, M.T., Willems, P., 2005. Statistical assessment of precip-
itation trends in the upper Blue Nile basin. Stoch. Environ. Res. Risk A.
29 (7), 17511761.
Precipitation and streamflow variability in Tekeze River basin, Ethiopia Chapter 10 121
This page intentionally left blank
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Statistical analysis of continuous daily climate data and extensive weather records are important to evaluate possible long-term, hydrologic/climatic changes at local and regional scales. In the present study, daily rainfall data recorded in the period 1971–2013 (43 years) at seven meteorological stations distributed in the Geba River basin, northern Ethiopia were used to study trends in extreme rainfall indices at different temporal scales. The selected rainfall indices focus on intensity, frequency, and duration of extreme rainfall measures. The Mann– Kendall trend test results show that decreasing tendencies in the rainfall indices have predominantly observed in several stations, even though most of the stations did not show statistically significant trend over time at 95% significance level during the study period. In majority of the extreme rainfall indices, station Abiadi which is located at the downstream of the basin showed statistically significant increasing trend, while decreasing trend in very heavy rainfall days (R20mm) for Adigrat station, and consecutive dry days (CDD) and highest rainfall amount in one-day period (RX1day) for Mek’ele showed statistically significant at the 95% level of significance. Results of this study contribute to climate change research in the region and provide inputs for better planning toward adapting to changing climate.
Article
Full-text available
The Upper Tekezē–Atbara river sub-basin, part of the Nile Basin, is characterized by high temporal and spatial variability of rainfall and streamflow. In spite of its importance for sustainable water use and food security, the changing patterns of streamflow and its association with climate change is not well understood. This study aims to improve the understanding of the linkages between rainfall and streamflow trends and identify possible drivers of streamflow variabilities in the basin. Trend analyses and change-point detections of rainfall and streamflow were analysed using Mann–Kendall and Pettitt tests, respectively, using data records for 21 rainfall and 9 streamflow stations. The nature of changes and linkages between rainfall and streamflow were carefully examined for monthly, seasonal and annual flows, as well as indicators of hydrologic alteration (IHA). The trend and change-point analyses found that 19 of the tested 21 rainfall stations did not show statistically significant changes. In contrast, trend analyses on the streamflow showed both significant increasing and decreasing patterns. A decreasing trend in the dry season (October to February), short season (March to May), main rainy season (June to September) and annual totals is dominant in six out of the nine stations. Only one out of nine gauging stations experienced significant increasing flow in the dry and short rainy seasons, attributed to the construction of Tekezē hydropower dam upstream this station in 2009. Overall, streamflow trends and change-point timings were found to be inconsistent among the stations. Changes in streamflow without significant change in rainfall suggests factors other than rainfall drive the change. Most likely the observed changes in streamflow regimes could be due to changes in catchment characteristics of the basin. Further studies are needed to verify and quantify the hydrological changes shown in statistical tests by identifying the physical mechanisms behind those changes. The findings from this study are useful as a prerequisite for studying the effects of catchment management dynamics on the hydrological variabilities in the basin.
Article
Full-text available
Hydrological regimes strongly influence the biotic diversity of river ecosystems by structuring physical habitat within river channels and on floodplains. Modification of hydrological regimes by dam construction can have important consequences for river ecosystems. This study examines the impacts of the construction of two dams, the Gezhouba Dam and the Three Gorges Dam, on the hydrological regime of the Yangtze River in China. Analysis of hydrological change before and after dam construction is investigated by evaluating changes in the medians and ranges of variability of 33 hydrological parameters. Results show that the hydrological impact of the Gezhouba Dam is relatively small, affecting mainly the medians and variability of low flows, the rate of rise, and the number of hydrological reversals. The closure of the Three Gorges Dam has substantially altered the downstream flow regime, affecting the seasonal distribution of flows, the variability of flows, the magnitude of minimum flows, low-flow pulses, the rate of rise, and hydrological reversals. These changes in flow regime have greatly influenced the aquatic biodiversity and fish community structure within the Yangtze River. In particular, populations of migratory fish have been negatively impacted. The results help to identify the magnitudes of hydrological alteration associated with the construction of dams on this important large river and also provide useful information to guide strategies aimed at restoration of the river's ecosystems.
Article
Full-text available
Hydrologic regime plays a major role in structuring biotic diversity within river ecosystems by controlling key habitat conditions within the river channel and floodplain. Daily flow records from seven hydrological stations and the range of variability approach were utilized to investigate the variability and spatial pattern of the hydrologic alterations induced by the construction of the Three Gorges Dam (TGD) in the middle and lower reaches of the Yangtze River, China. Results show that the impoundment of the TGD disturbed the hydrologic regime downstream and directly affected the streamflow variations. The rate of changes and the annual extreme conditions were more affected by the TGD, particularly the low-flow relevant parameters. The alterations in the hydrologic regime were mainly caused by the TGD storing water during early autumn and releasing water during winter and spring. The effects on spatial patterns decreased as the distance from the dam increased, which was mainly attributed to the inflows from large tributaries along the Yangtze River as well as the interaction with the two largest natural lakes (i.e., Dongting Lake and Poyang Lake). These hydrologic alterations not only break the natural balance of eco-flow regimes but also result in undesirable ecological effects, particularly in terms of habitat availability for the fish community.
Article
Full-text available
This paper presents a review of climate variability in the northeast Ethiopian Highlands, particularly Wollo and Tigray, during the last 10000 years (the Holocene) and an analysis of rainfall variability during the historical period. To date little work has been done on climate reconstruction in Tigray and Wollo, however, a number of ongoing studies in the area using palaeosoil analysis, archaeological techniques and lake level reconstruction will help reveal the magnitude of past climate variability in the region. On recent time scales, rainfall over Wollo is characterised by a distinctive bi-modal pattern with Belg rains in April-May preceding the main wet season Krempt, July to September, which is typical of the study region along much of the eastern escarpment. Further north over Tigray the Belg rains are less marked and the main Krempt rains dominate the seasonal pattern. Only three stations in the region, Dessie, Combolcha and Mekele, possess relatively unbroken records back to the 1950s. In all three, 1984 stands out as the driest year on record due to very low rainfall during the Krempt season, whilst 1999's Belg season was the driest on record. Although the succession of dry years between the late 1970s and late 1980s produced the driest decade in the Ethiopian Highlands this century there is no evidence for a long-term trend or change in the region's annual rainfall regime. Rainfall during the 1990s has seen a return to more humid conditions since the dry 1980s. The paper ends with a discussion of the major influences on rainfall variability over the Ethiopian Highlands: the El Niño-Southern Oscillation - El Niño events tend to be associated with lower than average rainfall; Atlantic, Indian and Pacific Ocean sea surface temperatures play a varying role in different seasons; the strength of the Indian Summer Monsoon; and the frequency of several tropical depressions over the Southwest Indian Ocean. SINET: Ethiopian Journal of Science Vol. 23, No. 2 (December 2000), pp. 139-161
Article
Full-text available
The East River basin is the major source of water supply for megacities in the Pearl River Delta and Hong Kong. Intensifying development of water resources and reservoir-induced hydrological alterations negatively affect ecological hydrological requirements. In this study, hydrological alterations and environmental flow variation are determined. Results indicate that: (1) multi-day maxima have reduced, while multi-day minima have increased, due to hydrological regulations of water reservoirs; (2) hydrological regimes of the East River have also been severely affected by hydropower generation, leading to a greater frequency of high and low pulses of lesser duration, and these effects are increasingly evident from the upper to lower East River basin; (3) owning to the water being released rapidly for hydropower generation or flood protection, the number of hydrologic reversals have increased after reservoir operations, also with increasing rise and fall rate; and (4) the alteration of three different types of environmental flow components have been shown in the study, which can be used to support the determination of environmental flow requirements in the East River basin.
Article
Assessment of flow regime alteration is significant for the scientific protection and restoration of rivers. The range of variability approach (RVA) and its revised versions, based on a suite of Indicators of Hydrological Alteration (IHA), have been widely used for flow regime alteration assessment. In these methods, the IHA of each specific year are treated separately, and their individual alterations are used to reflect the alteration of flow regime. However, it is known that one hydrological year is an important hydrological period, and the alteration of inner characteristics of one hydrological year (ICOHY), collectively characterized by the 32 IHA values of the year, can also lead to serious ecological degradation of rivers. In other words, treating the 32 IHA individually and separately in RVA and its revisions can not reflect the alteration of ICOHY. To overcome this limitation, a revised RVA is proposed. In the revised method, the 32 IHA values for each specific year are considered as a set, and the alteration of the set (SA) is measured by the set pair analyses method. The assessment of individual alteration (IA) of each IHA is carried out according to RVA. SA and IA are incorporated into a single index to reflect the overall flow regime alteration (OA). The proposed method is applied to a case study of the Xi Dayang Reservoir in China to test its effectiveness. The results suggest that the RVA underestimates the alteration degree and that the new method could give a more comprehensive assessment.
Article
This study investigates long-term trends in annual and seasonal precipitation at 16 stations in the upper Blue Nile River basin. The non-parametric Mann–Kendall test modified by effective sample size is used to detect linear trends in the precipitation time series. The trends magnitudes and starting time of significant trends are determined using the Sen's slope approach and the sequential Mann–Kendall test, respectively. Albeit annual precipitation shows a tendency to decrease in more than 80 % of the stations, statistically significant trends are found at only two stations. The significant decreasing trends of-40.3 and-168.1 mm/year per decade in annual precipitation at Debre Brihan and Gore stations started in the late 1970s and early 1980s respectively, which is consistent with the devastating droughts and famine during that period in the region. Owing to the great contribution of the main rainy season's (June to September: JJAS) precipitation to annual precipitation in the upper Blue Nile basin, the variation pattern of the JJAS precipitation is very similar to that of annual precipitation. Insignificant decreasing/increasing trends in the short rainy season's (March to May: MAM) precipitation are clearly predominant in the basin, where only one significant decreasing trend is detected in the time series.
Article
The Three Gorges Dam (TGD) started its regular operation in 2006, and its impacts on flow regime in the middle and lower Yangtze River are receiving attention worldwide. Based on daily river discharge data of seven gauges, flow regime changes in the middle and lower Yangtze River between the pre- and post-TGD periods have been investigated. Water storage by TGD has been found to have significantly reduced the mean flow in October. At the same time, the decrease of discharge from the Dongting and Poyang lakes due to climate variability has also significantly aggravated the reduction of river discharge. According to estimates, the river discharge at the Datong gauge during the TGD impoundment period decreased by 18-40% after the reservoir started full-capacity operations in 2008, at which time TGD contributed nearly 8-18% of this flow reduction. By contrast, the water release by TGD significantly increased the river discharge in February. Meanwhile, TGD changed the magnitude of extreme flows, leading to increase of annual minimum flows and reduction of annual maximum flows. Close attention needs to be given to the possible consequences of TGD-induced flow regime changes in the complex river lake ecosystem in the middle and downstream sections of the Yangtze River in the future.