Impact of Rainfall Variability, Land Use and Land
Cover Change on Stream Flow for Hydropower
Generation in the Black Volta Basin
AKPOTI Komlavi
MRP Climate Change and Energy
Impact of Rainfall Variability, Land
Use and Land Cover Change on
Stream Flow for Hydropower
Generation in the Black Volta Basin
First Supervisor:
Author:
Dr. Eric Ofosu Antwi
Komlavi AKPOTI
Second Supervisor:
Dr. Amos T. Kabo-bah
November 30, 2015
Acknowledgement
First of all, I thank the Almighty God for His grace and Benedictions upon my life
through the master program and during my thesis work.
I acknowledge with thanks the scholarship and financial support provided to me by
BMBF, the Federal Ministry of Education and Research of Germany and the West
African Science Service Centre on Climate Change and Adapted Land Use (WASCAL).
There were numerous of people whose advice, criticism, and support have greatly helped
during this project. I will thank my main supervisor Dr. Eric Ofosu Antwi, University
of Energy and Natural Resources, Ghana, for all his useful instructions, comments and
criticism. His help was determinant from the very beginning of the conception of this
project.
I will say a big thank to my co-supervisor, Dr. Amos Kabo-Bah, University of Energy
and Natural Resources, for all his help and encouragements.
My acknowledgements are also extended to Dr. Benjamin L. Lamptey, Deputy Director
General, ACMAD for his advice and guidance through this project.
I am grateful to my dear friend Collins Owusu from the KWAME NKRUMAH University
of Science and Technology. In fact, he has explained to me the basic aspects of the Soil
and Water Assessment Tool (SWAT).
A big thank to Dr.
Boubakar IBRAHIM, hydro climatologist at the WASCAL
competence center for reviewing this thesis work.
I will finally say thanks to all my colleagues in the Climate Change and Energy master
program for their support. I will specially say thanks to my dear colleague Alima Dajuma
for her useful suggestions and advice. I will also say thanks to my colleague Sawadogo
Windmanagda for facilitating the acquisition of meteorological data for the Burkina Faso
section of the Black Volta.
i
Abstract
Potential implications of rainfall variability along with Land Use and Land Cover Change
(LULC) in the Black Volta basin on the Bui hydropower plant have been assessed.
The aridity index profile of the Black Volta catchment was first developed showing
three climatic conditions in the basin: Semi-arid zone, dry sub-humid and humid zone.
The spatio-temporal variability of rainfall over the Black Volta was assessed using the
Mann-Kendall monotonic trend test and the Sen’s slope for the period 1976-2011. The
statistics of the trend test showed that 61.4% of the rain gauges (8 stations out of a
total of 13) presented an increased precipitation trend whereas the rest of the stations
showed a decreased trend. However, the test performed at the 95% confidence interval
level showed that the detected trends in the rainfall data were not statistically significant.
Only rainfall data at Boura in dry sub-humid zone revealed a statistically significant
increase (pvalue less than 0.05). LULC is an important factor controlling the hydrology
of a basin. Three Landsat satellite images were collected including Landsat 5 Thematic
Mapper for the year 1987, Landsat 7 SLC-on for the year 2000 and Landsat 8 OLI for the
year 2013. Land use trends between the year 2000 and 2013 show that within thirteen
years, land use classes like bare land, urban areas, water bodies, agricultural lands, forest
deciduous and forest evergreen have increased respectively by 67.06%, 33.22%, 7.62%,
29.66%, 60.18%, and 38.38%. Only grass land has decreased by 44.54% within this
period. To evaluate the combined effects of rainfall variability and land use change on
the discharge at Bui, the hydrological model, SWAT has been selected. A first calibration
was performed for the period 2000-2005 and validation for the period 2006-2010. During
calibration, the model performance was qualified as very good with N S = 0.9 and R2
= 0.91. The strong correlation between measured and simulated flows showed that the
physical processes implicated in the generation of the stream flow in the basin are well
captured by the model. For the validation period, the model performance was good with
N S = 0.7 and R2 = 0.8. However, the graphical representation of the observed and the
simulated flows during validation revealed that there was a delay in peak flows starting
from 2008. It was assumed this effect may be due to the Bui reservoir construction. To
confirm the hypothesis, the model performance was tested for a period prior to the Bui
dam construction.This second calibration was performed for the period 1990-1995 and
validated for the period 1996-2000. The calibration was qualified as very good (N S =
ii
0.9 and R2 = 0.82) and validation as good (N S = 0.7 and R2 = 0.85). The graphical
representation of the observed and the simulated flows during calibration and validation do
not show important delay in the peak flows proving that the construction of the Bui dam
may effectively affect the dynamic of the river system. Changes in seasonal stream flow
due to LULC was assessed by defining dry season (February, March and April) and wet
season (August, September and October). The results showed that from year 2000 to year
2013, the dry season discharge has increased by 6% whereas the discharge of wet season
has increased by 1%. The changes in stream flows components such us surface run-off
(SURF Q), lateral flow (LAT Q) and ground water contribution to stream flow (GW Q)
and also on evapotranspiration (ET) changes due to LULC was evaluated. The results
showed that between the year 2000 and 2013, SURF Q and LAT Q have respectively
increased by 27% and 19% while GW Q has decreased by 6%. At the same time, ET
has increased by 4.59%. The resultant effects is that the water yield to stream flow has
increased by 4%. We believe that the overall impacts of rainfall variability and LULC
may benefit the Bui hydropower plant.
iii
Key Words
Rainfall Varibility
Aridity Index
Land Use and Land Cover Change
SWAT Model
Black Volta
Stream Flow
iv
Acronyms
AI
Aridity Index
BPA
Bui Power Authority
CN
Curve Number
CO2
Carbon Dioxide
CH4
Methane
DEM
Degital Elevation Model
ESIA
Environmental and Social Impact Assessment
FAO
Food and Agricultural Organization
GHGs
Green House Gases
GIS
Geographic Information System
GMA
Ghana Meteorological Agency
GW
Gigawatt
HRUs
Hydrological Response Units
ICOLD
International Commission on Large Dams
IEA
International Energy Agency
IPCC
Intergovernmental Panel on Climate Change
K
Kappa Coefficient
LULC
Land Use and Land Cover Change
MW
Megawatt
NRCS
National Resource Conservation Service
NS
Nash-Sutcliff Coefficient
R2
Coefficient of Determination
v
SCS
Soil Conservation Service
SRTM
Shuttle Radar Topography Mission
SWAT
Soil and Water Assessment Tool
SWAT-CUP Soil and Water Assessment Tool-Calibration and Uncertainty Programs
SUFI-2
Sequential Uncertainty Fitting
TWh
Tera Watt hour
PET
Potential Evapo-Transpiration
UNEP
United Nations Environment Programme
WEC
World Energy Council
WGN
Weather Generator
vi
Contents
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ii
Key Words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iv
Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v
List of Figures
x
List of Tables
xii
1 INTRODUCTION
1
1.1
Background of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1.2
Statement of the problem . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
1.3
Objective of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
1.4
Research questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
1.5
Significance of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
1.6
Thesis structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2 LITERATURE REVIEW
7
2.1
Rainfall variability, climate change and runoff influence on hydropower . .
2.2
Hydrological models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3
The SWAT model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.1
Application of the SWAT model worldwide . . . . . . . . . . . . . . 15
2.3.2
Application of the SWAT model in the Volta Basin . . . . . . . . . 15
3 STUDY AREA
3.1
7
17
Study area presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.1
The Black Volta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
vii
3.1.2
The Bui dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4 MATERIALS AND METHODS
4.1
4.2
4.3
4.4
4.5
23
Rainfall variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.1
Aridity Index map development . . . . . . . . . . . . . . . . . . . . 23
4.1.2
Rainfall data acquisition . . . . . . . . . . . . . . . . . . . . . . . . 24
4.1.3
Trend analysis through the Mann-Kendall statistics . . . . . . . . . 26
4.1.4
Sens slope estimator . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Hydrology of the Black Volta catchment . . . . . . . . . . . . . . . . . . . 30
4.2.1
Brief description of the SWAT Model . . . . . . . . . . . . . . . . . 30
4.2.2
Surface runoff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
SWAT input data collection and analysis . . . . . . . . . . . . . . . . . . . 34
4.3.1
Digital Elevation Model (DEM) . . . . . . . . . . . . . . . . . . . . 35
4.3.2
Soil Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.3
Landsat data acquisition for LULC analysis . . . . . . . . . . . . . 37
4.3.4
LULC classes definition
4.3.5
Accuracy assessment of the developed LULC Maps . . . . . . . . . 39
4.3.6
Hydrological data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
. . . . . . . . . . . . . . . . . . . . . . . . 38
SWAT model setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.4.1
Watershed delineation . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.4.2
HRUs analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4.3
Weather generator . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Sensitivity analysis, calibration and validation in the SWAT-CUP . . . . . 46
4.5.1
Sensitivity analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.5.2
Calibration and validation . . . . . . . . . . . . . . . . . . . . . . . 47
4.5.3
Model performance evaluation . . . . . . . . . . . . . . . . . . . . . 49
5 RESULTS AND DISCUSION
5.1
5.2
51
Aridity index and rainfall variability over the Black Volta . . . . . . . . . . 51
5.1.1
Aridity index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.1.2
Rainfall variability through trend analysis . . . . . . . . . . . . . . 52
LULC analysis in the Black Volta . . . . . . . . . . . . . . . . . . . . . . . 58
5.2.1
Accuracy assessments (confusion matrices and statistics on LULC)
viii
58
5.3
5.2.2
LULC maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
5.2.3
LULC 1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2.4
LULC 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.2.5
LULC 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.2.6
Land use trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Hydrology of the Black Volta catchment . . . . . . . . . . . . . . . . . . . 66
5.3.1
Sensitivity Analysis (to identify most sensitive parameters in the
basin, for calibration of SWAT) . . . . . . . . . . . . . . . . . . . . 66
5.4
5.5
5.3.2
Calibration and validation . . . . . . . . . . . . . . . . . . . . . . . 68
5.3.3
Model uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
LULC impacts on stream flow in the Black Volta catchments
. . . . . . . 73
5.4.1
Changes in seasonal stream flows due to LULC . . . . . . . . . . . 74
5.4.2
Changes in stream flow components due to LULC . . . . . . . . . 75
Combined potential implications of rainfall variability and LULC for the
Bui hydropwer plant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6 CONCLUSIONS AND RECOMMENDATIONS
77
6.1
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.2
RECOMMENDATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
A Additional tables
82
B Additional graphs
84
References
89
List of Figures
1.1
Flow Chart of Rainfall, Temperature and Land Use Land Cover Change
effects on hydropower. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
3.1
Study area map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2
Digital elevation map of Black Volta . . . . . . . . . . . . . . . . . . . . . 19
3.3
Bui Resevoir Plan Trajectory (Jan 1st-June 30th 2014) . . . . . . . . . . . 22
4.1
Double Mass Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.2
Schematic representation of the hydrologic cycle . . . . . . . . . . . . . . . 31
4.3
Flow chart of the steps in the SWAT model application for the Black Volta 35
4.4
Soil Map of the Black Volta . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.5
The fourteen scenes covering the entire Black Volta . . . . . . . . . . . . . 37
4.6
Bui monthly average flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.7
Black Volta watershed as defined in the SWAT model.
The numbers
represent each sub-basin (total of 17 sub-basins) . . . . . . . . . . . . . . . 42
5.1
Aridity Index and rainfall map of the Black Volta Basin . . . . . . . . . . . 52
5.2
Bui rainfall 1954-2005
5.3
Wa rainfall 1976-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.4
Boura rainfall 1970-2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.5
Boromo rainfall 1970-2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.6
Bobo-Dioulasso rainfall 1960-2012 . . . . . . . . . . . . . . . . . . . . . . . 56
5.7
LULC maps of the Black Volta . . . . . . . . . . . . . . . . . . . . . . . . 59
5.8
LULC occupied area in 1987 . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.9
percentage of each land use type in 1987 . . . . . . . . . . . . . . . . . . . 62
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.10 LULC occupied area in 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . 62
x
5.11 percentage of each land use type in 2000 . . . . . . . . . . . . . . . . . . . 63
5.12 LULC occupied area in 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.13 percentage of each land use type in 2013 . . . . . . . . . . . . . . . . . . . 65
5.14 Monthly calibration and validation graphs for the Bui station (2000-2010) . 68
5.15 Average simulated monthly discharge vs average observed monthly
discharge for the Bui station during calibration and validation (2000-2010)
70
5.16 Monthly calibration and validation graphs for the Bui station ((1990-2000)
71
5.17 Average simulated monthly discharge vs average observed monthly
discharge for the Bui station during calibration and validation (1990-2000)
B.1 Batie rainfall 1960-2013
72
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
B.2 Bomborokuy 1962-2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
B.3 Dano rainfall 1970-2013
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
B.4 Bondoukuy rainfall 1970-2013 . . . . . . . . . . . . . . . . . . . . . . . . . 86
B.5 Diebougou rain fall 1970-2013 . . . . . . . . . . . . . . . . . . . . . . . . . 86
B.6 Bole rainfall 1976-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
B.7 Sunyani rainfall 1976-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
B.8 Wenchi rainfall 1976-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
B.9 Flow in the first calibration with the 95PPU . . . . . . . . . . . . . . . . . 88
xi
List of Tables
3.1
Some features of the Bui hydropower plant . . . . . . . . . . . . . . . . . . 21
4.1
Climate classes according to AI values . . . . . . . . . . . . . . . . . . . . 24
4.2
General Characteristics of Rainfall Stations included in the variability
analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.3
General statistics of the Gauging stations in the basin . . . . . . . . . . . . 27
4.4
Weather generator for the Black Volta catchment . . . . . . . . . . . . . . 45
4.5
Most sensitive parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.6
General performance ratings for recommended statistics for a monthly time
step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.1
The Mann-Kendall trend statistics with the Sens Slope estimator . . . . . 57
5.2
Overall Accuracy and Kappa coefficient the LULC maps . . . . . . . . . . 58
5.3
LULC characteristics in the basin . . . . . . . . . . . . . . . . . . . . . . . 60
5.4
LULC trends in the catchment . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.5
Final Parameter range and their sensitivity rank . . . . . . . . . . . . . . . 67
5.6
Summary results for SWAT model calibration and validation (2000-2010) . 69
5.7
Summary results for SWAT model calibration and validation (1990-2000) . 71
5.8
Intra-annual flow changes due to LULC . . . . . . . . . . . . . . . . . . . . 74
5.9
Stream flow Components and ET Changes due to LULC . . . . . . . . . . 75
A.1 Confusion matrix LULC 1987 . . . . . . . . . . . . . . . . . . . . . . . . . 82
A.2 Confusion matrix LULC 2000 . . . . . . . . . . . . . . . . . . . . . . . . . 82
A.3 Confusion matrix LULC 2013 . . . . . . . . . . . . . . . . . . . . . . . . . 82
A.4 General Statistics on LULC 1987 . . . . . . . . . . . . . . . . . . . . . . . 83
A.5 General Statistics on LULC 2000 . . . . . . . . . . . . . . . . . . . . . . . 83
xii
A.6 General Statistics on LULC 2013 . . . . . . . . . . . . . . . . . . . . . . . 83
xiii
Chapter 1
INTRODUCTION
1.1
Background of the study
Hydropower, a mechanical power of falling water, is one of the most important sources
of renewable energy worldwide (Kaunda et al., 2012). As clean energy, hydropower
potential is estimated to be around 15000TWh/year in the world(Bartle, 2002). It is
one of the most proven technology for electricity generation and has been used for long
in the world. According to the International Energy Agency (IEA), it represents 16.3%
of the world’s total electricity supply in 159 countries that is about 3500 TWh in 2010
(Technology Roadmap, 2011) with 24 countries depending on it for more than 90% of
their supply according to World Commission on Dams (World Commission on Dams,
2000).
Hence hydropower is the first renewable energy for electricity production as
defined by The International Commission on Large Dams (ICOLD).
Hydropower has been and is still occupying an important part of electricity production
in Africa. Its technical hydropower potential is around 1 750 TWh which is about 12% of
the global capacity with only 5% of this technically feasible potential exploited (Zealand,
2008). The challenge of low development of hydropower in Africa may be due to many
factors including the lack of finance. Recently, hydropower development in Africa has
involved foreign investors among which we have China according to the Word Energy
Council (WEC, 2013). Despite the fact that Africa has a large amount of hydropower
potential, the level of access to electricity that comes mainly from hydropower and fossil
fuels is low. The access to electricity remains the main challenges for economic growth
1
1.1. BACKGROUND OF THE STUDY
in Africa. In sub-Saharan Africa, access to energy is a big challenge, with an estimated
51% of urban population and only 8% of rural population having access to electricity
(Boko et al., 2007)
Hydropower will be able to play the greatest role in the future in economic development
of Africa because a number of countries describe their hydro-potential as one of the most
valuable resources, and the backbone to future social and economic development (Bartle,
2002). Africa’s electricity consumption is expected to grow at a rate of 3.4% per year
over the period 1999 to 2020. Of the total 20.3 GW of hydropower currently installed
in Africa, about 23% is located in North Africa, 25% in West Africa and the remaining
51% located in South/Central/Eastern Africa (Kalitsi, 2003).
Hydropower promises
enormous contribution to boosting the energy base of the continent (Kalitsi, 2003). At
the African Ministerial Conference on Hydropower and Sustainable Development, held
in Johannesburg in March 2006, there was general agreement on the need to accelerate
the implementation of dam-building projects throughout Africa. Given this background,
the key challenge is to determine how dams are best able to contribute to African
countries growth when obtaining reliable and sustainable sources of water, food and
energy security, whilst simultaneously avoiding and mitigating harmful impacts on the
environment as far as possible (McCartney, 2007).
Hydropower is the most important electricity source in Ghana and many other
countries in West Africa. Ghana’s electricity supply is highly dependent on hydropower
which accounts for about two-thirds of total electricity supply (Arndt et al., 2014).
Ghana currently has three large hydroelectric dams including Akosombo, which
was commissioned in 1965 and has a capacity of 1,012MW, and Kpong which was
commissioned in 1982 and has a capacity of 160MW (Hensengerth, 2011).The Bui dam
is its third dam to be constructed and second large hydroelectric dam with a maximum
generation capacity of 400 MW, a net average energy production of 980 GWh.year-1
(ESIA, 2007). The latest is the focus of this research.
Hydropower, due to its dependence on stream flow, is subjected to challenges including
changes in precipitation, temperature, and land use.
2
Changes in water cycle have
CHAPTER 1. INTRODUCTION
therefore implications for hydropower generation. The complexity of the combined effects
of all these changes might sometimes be difficult to identify. It is therefore interesting,
at a basin level to study carefully each element’s susceptibility to change, in order to
identify their degree of change in a certain time bound and their possible implications
for the hydropower plant.
Figure 1.1: Flow Chart of Rainfall, Temperature and Land Use Land Cover Change effects
on hydropower.
Blue colour indicates effects that may have negative impacts on hydropower
productions while green colour indicates potential positive impacts for hydropower
production. The Flow chart has been adapted from an existing one on the website
https://sites.google.com/a/marence.at/climate-change-and-hydropower/
The approach to understand the impacts of precipitation, temperature, land use and
3
1.2. STATEMENT OF THE PROBLEM
land cover change (LULC) on hydropower production can be summarized as in figure
1.1. Land use practices in a river basin may have important implications for hydrology
and therefore affect water availability. For instance, LULC may induce a decrease in
evaporation over the basin as well as produce the reverse effect. This may directly and/or
indirectly affect river flow and consequently the hydropower. It is therefore meaningful
to have a well managed watershed in term of land use.
Although the causes of rainfall and temperature variability may come from more global
phenomena, its important to know and understand their historical patterns or trends
for better decision making. Changes in temperature and/or precipitation may affect
positively or negatively evaporation over the basin with the corresponding effect on
stream flow. Rainfall changes may also affect directly river flow. The variability as used
in the present work should be considered as temporal (in terms of years, seasons or
simply time) but also spatial (in term of location). This is justify by the fact that the
Black Volta is a trans-boundary watershed extending from Mali, Burkina Faso to Ivory
Coast and Ghana (see figure 3.1) with different climate conditions.
The present project is a multidisciplinary approach with combination of statistics,
geographical information systems (GIS) and hydrological modelling. In ideal conditions, it
may imply collaboration among land users (farmers for examples), soil and crop scientists,
as well as hydrologists.
1.2
Statement of the problem
Rainfall variability has a large influence on a water resource project.
Hydropower
depends heavily on rainfall and runoff. The eco-hydrological processes that influence
the evolution of the surface ecosystem, depend on the spatio-temporal patterns of the
precipitation and evapotranspiration (Oguntunde et al., 2006). Furthermore, land use
and cover are changing fast in the in basin due to population growth and economical
activities development in the basin. The newly developed Bui hydropower plant, which
is one of the major water resources project on the Black Volta, is very important for
the economic growth of Ghana. The competition for water in the basin is expected
4
CHAPTER 1. INTRODUCTION
to increase due to rapid population growth. According to (Green Cross International,
2001), the population growth rate in the basin is about 3% per year. There is also a
concern about a global climate variability and changes that will put more pressure on
water availability.
Therefore, a strong need is identified for studying the rainfall variability, LULC and its
potential impacts on the Bui hydropower plant.
1.3
Objective of the study
The present study tries to evaluate the rainfall variability and LULC of the Black Volta
and its potential impact on hydropower generation at the Bui power plant in Ghana. The
specific objectives are as follow:
1. Assessment of the spatial and temporal variability of rainfall over the Black Volta
2. Assessment of LULC trends in Black Volta
3. Assessment of the potential rainfall variability and LULC on stream flow in the
basin.
1.4
Research questions
This study addresses the following questions:
1. What are the patterns of observed rainfall change and variability over the Black
Volta?
2. How is the land use evolution in the basin?
3. How is the combined effect of potential rainfall variability and LULC influence
discharge in the basin?
5
1.5. SIGNIFICANCE OF THE STUDY
1.5
Significance of the study
The hydropower plays a major role in the economy of Ghana.
With its two-third
contribution to electricity generation in the country, hydropower remains the main
source of electricity in Ghana. The overall challenge of the hydropower production is
the availability of water for future energy production in the context of global climate
variability and change coupled with increasing competing water uses due to population
growth in the river basin. This study is motivated by the fact that rainfall is a key
parameter in hydropower generation and understanding its patterns is necessary for
planning. In addition, LULC implications on the basin hydrology neeed to be addressed.
Since several studies have shown that the Volta Basin is climate sensitive (Oguntunde et
al., 2006; Goulden et al., 2009; Van de Giesen et al., 2010), there is a need to pay a close
attention to the rainfall regime over the Black Volta. This is for the benefit of the future
planning of hydropower generation at the Bui dam, which was completed in earlier 2014
within the Black Volta of West Africa.
1.6
Thesis structure
The thesis is structured in six chapters. The first chapter deals with a general introduction
to this thesis including the background of the study, the problem statement, the objectives
of the study, the research questions, the significance of the research, and the thesis
structure. The literature review follows chapter one. The third chapter briefly presents
the study area. The materials and the methods used are presented in chapter four. Results
are presented and discussed in chapter five while chapter six is the conclusion of the thesis
with some recommendations.
6
Chapter 2
LITERATURE REVIEW
2.1
Rainfall variability, climate change and runoff
influence on hydropower
Assessing temporal trends and their spatial distribution pattern of precipitation remains
a difficult task owing to their complex and non-linear nature in different regions. It is
an important step for water resource projects in a basin. Rainfall distribution in Ghana,
and West Africa in general, is influenced by the moist south-west monsoon and the dry
north-east trade wind.
In West Africa, most of the electricity generation comes from hydropower that relies
on rainfall. (Conway et al., 2009) studied rainfall and water resources variability in
sub-Saharan Africa during the twentieth century.
Their analysis of rainfall-runoff
relationships reveals varying behavior including strong but non stationary relationships
particularly in West Africa with rainfall accounting for around 60%-70% of river flow
variability. (Mahe et al., 2001) assessed the trends and discontinuities in regional rainfall
of West and Central Africa from 1951-1988. Their results showed that the whole of
West Africa, West of the Atakora Mountains experiences the more severe drought that
has been observed in the majority of the stations. In Ghana for instance, their analysis
revealed that there was a significant decreasing trend in standardized annual rainfall
over the period 1951-1989 (Mahe et al., 2001). Using gridded monthly precipitation data
available at 0.5 deg intervals over the period 1951-2000, (Owusu & Waylen, 2009) showed
7
2.1. RAINFALL VARIABILITY, CLIMATE CHANGE AND RUNOFF INFLUENCE
ON HYDROPOWER
that West Africa has undergone a period of diminished rainfall with an apparent shift in
the rainfall regime towards a longer dry season.
(Logah et al., 2013) analysed rainfall pattern in Ghana showing high and low rainfall
distribution in the country. Their results showed that between the period 1981-2010,
there was a general decline in mean annual rainfall with high rainfalls shifting to the
south-western corner of the country . Consequently, the agricultural production potential
in the northern Ghana is diminished by high rainfall variability while the mean annual
rainfall totals in all agro-ecological zones experienced a decline in precipitation (Owusu
& Waylen, 2009).
(Oguntunde et al., 2006) have used a Mann-Kendal statistics to assess trends and
variability of hydro climatology of the Volta basin in West Africa from 1901 to 2002.
The results showed that a 10% relative decrease in precipitation resulted in a 16%
decrease in runoff between 1936 and 1998. A water balance study of the Volta basin by
(Andreini et al., 2000) showed that runoff is extremely sensitive to rainfall therefore has
significant impact on the hydropower generation. The Mann- Kendal test has also been
used by (Lacombe et al., 2012) when assessing drying climate in Ghana over the period
1960-2005. The analysis showed that no significant changes in the annual rainfall were
observed but reduction in the number of wet season days, a delay in the wet season onset
at several locations throughout the country and lengthening of rainless periods during
wet season have shown significant changes.
Several studies have shown that hydropower is sensitive to the state of the environment,
and climate change all over Africa (Bunyasi, 2012; Kaunda et al., 2012; Stanzel & Kling,
2014). It is important to know that the statistical significance of climate variability can
allow climate change detection in the basin. As such, climate change refers to a change
in the state of the climate that can be identified (e.g. using statistical tests) by changes
in the mean and/or the variability of its properties, and that persists for extended
period, typically decades or longer (Barker et al., 2007). Climate change is driven by
greenhouse gases (GHGs) such as carbon dioxide (CO2), methane (CH4) and many
others that dominate the radiation forcing of the climate system. Warming of the climate
8
CHAPTER 2. LITERATURE REVIEW
system is univocal and most of the observed increase in global average temperature
since the mid-20th century is very likely due to the increase in the anthropogenic GHG
concentrations (Barker et al., 2007). According to Intergovernmental Panel on Climate
Change (IPCC) Fifth Assessment Report (AR5), evidence of warming over land regions,
consistent with anthropogenic climate change, has increased (Pachauri et al., 2014)
including Africa. In addition, the AR5 estimated with high confidence that climate
change will amplify existing stress on water availability in Africa.
Hydropower resource potential is sensitive to climate change because of its dependence
on run-off water, a resource which is dependent on climate driven hydrology (Kaunda et
al., 2012). Runoff in a given river basin is a function of some meteorological parameters
including precipitation and temperature. Changes in the quantity and timing of river
runoff, together with increased reservoir evaporation will have a number of effects on the
production of hydropower (Harrison, 1998). In the context of climate change and global
warming, losses of water through evaporation will be significant. For instance (Stanzel
& Kling, 2014) has reported changing climatic conditions might impair the benefits
of hydropower development meaning warmer temperature will tend to decrease inflow
due to higher land surface evapotranspiration and will increase reservoir evaporation.
According to (Mukheibir, 2007), the change in temperature and rainfall has the potential
to affect hydroelectric installations in four major ways: surface water evaporation,
reduced run-off due to drought, increased run-off due to flooding, Siltation deposits.
Clearly, climate change will impact hydropower as it strongly relies on river discharge.
This reality may affect African countries and Ghana’s economy which rely on water
resources and hydropower for electricity generation. In fact, it is widely acknowledged
that developing countries will suffer some of the greatest impacts of climate change
due to their greater reliance on climate-dependent natural resources (McSweeney et al.,
2010). With high confidence, the Fourth Assessment Report of the IPCC (AR4) has
shown that Africa is one of the most vulnerable continent to climate variability and
change, a situation aggravated by the interaction of multiple stress occurring at various
levels, and low adaptive capacity (Boko et al., 2007). Therefore, climate change and its
variability have the potential to impose additional pressure on water availability and
9
2.1. RAINFALL VARIABILITY, CLIMATE CHANGE AND RUNOFF INFLUENCE
ON HYDROPOWER
demand in Africa.
It is likely to experience increasing stress in the near future as a consequence of both
greater water demand and climate change. According to (McCartney et al., 2012), up
to the middle of the twenty-first century, the Volta basin wide average annual runoff
and mean ground water recharge will all decline. Despite the planned diversification,
hydropower is expected to remain a major energy source in Ghana over the coming
decades and this raises concerns about the effect of climate on river flows and generation
capacity(Arndt et al., 2014).
Unfortunately, across much of sub-Saharan Africa, climate variability and change is
given low priority. This means that in many countries, there has been no systematic
evaluation of the possible implications of climate change for water resources and it is
given little consideration in the planning of future water development (McCartney et
al., 2012). This might be the case of the Bui hydropower project, while it is serious
oversight that the impact of global climate change was not taking into account when
the plans for the construction of the Bui dam were evaluated and finalized (Laube
et al., 2008). In the review of the Bui hydropower project Environmental and Social
Impact Assessment (ESIA), (Raschid et al., 2007) have revealed that no drought flow
analysis was undertaken to determine the impact of recurrent drought conditions on
the operation of the dam. They considered that it is a serious issue for the future of
the dam while looking at the fact that, extreme low flow in the past (including 2006)
had put the Akosombo dam under severe stress and have been a major concern in the
country; the omission of such analysis is rather serious. Also competing water needs
for hydropower generation and agriculture will affect each other. It is therefore very
important and should have been taken into consideration when the Bui dam was designed.
However, at the second Ghana Dam forum, held in Accra on 26th-27th of February, 2008
the participants suggested the assessment of the impacts of changing climate on the Bui
dam (Raschid Sally et al., 2008). This stage is important because dam construction and
water development projects create complex wide ranging ecological and environmental
effect while the design and characteristics of each reservoir to the climate change effects
10
CHAPTER 2. LITERATURE REVIEW
are vital in determining reproach strategies (Bunyasi, 2012).
Taking the whole Volta basin, several studies have shown the climate sensitivity of water
resources, not only regarding the changing climate but also population growth in the
basin. The Volta basin population is estimated a little over 14 million inhabitants with
an annual growth rate estimated at almost 3% placing increased pressure on land and
water resources (Green Cross International, 2001). This condition will put constraints on
the quantity and the quality of water with time (Kasei, 2009). This means also that the
availability of water for hydropower in the future is questionable regarding flow regimes.
Changes in flow regimes in arid areas can have particularly adverse impacts, leading
to the loss or degradation of many essential services provided by rivers and wetlands
(Watson et al., 2005), since the Volta basin includes some arid and semi-arid zones.
The Volta basin runoff exhibits higher temporal variability than the rainfall input due to
non-linear response and threshold effects, making the extend to utilizable water resource
in the basin highly sensitive to variation in precipitation and to alteration in land use
and cover (Rodgers et al., 2006). Even though the per capita water availability of the
basin may be perceived as normal, deforestation, land degradation, and high population
growth coupled with global climate change promises to exacerbate the growing scarcity
on water resource as water supplies are unreliable and insufficient to meet demand of the
growing population (Kasei, 2009). Time series analysis demonstrates that small changes
in precipitation lead to proportionately larger changes in runoff, which is an issue of great
concern given the likelihood of alteration in rainfall pattern as consequence of changes
in global circulation (Rodgers et al., 2006). For instance, (Kasei, 2009) has demonstrate
that, various stations data from the Volta river system indicate that the months in which
precipitation exceeds the evapotranspiration to generate runoff and direct recharge are
usually June, July, August and September. The others months therefore experience no
or little rainfall resulting in low flows.
One prediction for the Volta basin on which various climate change scenarios agree, is the
overall increase of seasonal river flow availability due to the change of rainfall pattern,
with less water available in the dry season, while river flow will increase in the rainy
11
2.2. HYDROLOGICAL MODELS
season (Laube et al., 2008). According to (Martin, 2006), the Volta river system annual
recharge vary from 13% to 16% of the mean annual precipitation.
Considering the Black Volta sub basin, little research has been conducted on climate
variables and their implications for the Bui hydropower. However, (Kabo-bah et al.,
2014) have used the remote sensing approach to estimate the evapotranspiration over
the Black Volta. The satellite observation data has also been used by (Beck et al., 2012)
to simulate runoff for the Black Volta. The impact of climate change on the Black Volta
river flow for instance has been studied by (Sawai et al., 2014).
2.2
Hydrological models
The need of assessing land use changes and rainfall variabilities effects on stream flow
generation in the Black Volta implies also the review of the most well known hydrological
models capable of doing this specific task with respect to the available data. Hydrological
models have become increasingly an important tool for water resource management in a
river basin. A hydrological model can be considered as a mathematical representation
of a hydrological cycle of an entire river basin or a part of it (Lundin et al., 1998).
According to (Cunderlik, 2003), hydrological models can be classified in to three main
categories:
i. Lumped models.
Parameters of lumped hydrological models do not vary
spatially within the basin and thus, basin response is evaluated only at the
outlet, without explicitly accounting for the response of individual sub basins.
Parameters of lumped models often do not represent physical features of
hydrologic processes and usually involve certain degree of empiricism. The
impact of spatial variability of model parameters is evaluated by using certain
procedures for calculating effective values for the entire basin.
IHACRES (Identification of Unit Hydrographs and Component flows from
Rainfall, Evaporation and Stream flow data) can be considered as on of the
lumped model developed by (Jakeman et al., 1990). The lumped hydrologic
12
CHAPTER 2. LITERATURE REVIEW
modelling is based on a transfer function/hydrograph and the model is
used to characterise the dynamic relationship between rainfall and stream
flow. The model requires data such us rainfall, temperature, or potential
evapo-transpiration (PET) to predict stream flow. the model can be used on
minute, daily or monthly time step. The model can be used to assess impact
of climate change and identify effects of land use changes.
ii. Semi-distributed models.
Parameters of semi-distributed (simplified
distributed) models are partially allowed to vary in space by dividing the basin
into a number of smaller sub basins. The main advantage of semi-distributed
models is that their structure is more physically-based than the structure
of lumped models, and that they are less demanding on input data than
fully distributed models. As examples of semi-distributed models, we have
HEC-HMS (William & Matthew, 2010), TOPMODEL and SWAT(J. G. Arnold
et al., 1998).
HEC-HMS (Hydrologic Engineering Center -Hydrlogic Modelling System) is a
hydrological model developed by the U.S Army Corps of Engineering. It is a
public domain model designed to simulate the precipitation-runoff processes
of dentritic watershed systems. As semi-distributed physically based model,
it is capable of modeling continuous processes and events. According to the
HEC-HMS website (http://www.hec.usace.army.mil/software/hec-hms/), the
model is able of representing physical features and the hydrologic elements
(subbasin, reach, junction, reservoir, diversion, source, and sink) of the basin.
The model meteorologic data analysis is based on short wave radiation,
precipitation, evapotanspiration, and snowmelt. However, not all these data are
requiered during simulation. Along with hydrologic simulation, the model can
perform optimisation, forecast stream flow, assess model uncertainties as well
as sediment and water quality analysis. Finally, the model has a GIS connection
known as Geospatial Hydrologic Modelling Extension (HEC-GeoHMS).
TOPMODEL (TOPographical based hydrological MODEL) is based on the
use of the digital terrain data. It has been wideky used due to it simplicity
based on th topographical index application and the possibility of visualising
13
2.3. THE SWAT MODEL
the predictions of the model in the spatial context (Beven, 1997).
The SWAT model as part of the tools used in the present project is fully
discussed in the following sections.
iii. Distributed models. Parameters of distributed models are fully allowed to vary
in space at a resolution usually chosen by the user. Distributed modeling
approach attempts to incorporate data concerning the spatial distribution
of parameter variations together with computational algorithms to evaluate
the influence of this distribution on simulated precipitation-runoff behavior.
Distributed models generally require large amounts of (often unavailable)
data for parameterization in each grid cell. However, the governing physical
processes are modeled in detail, and if properly applied, they can provide the
highest degree of accuracy.
As an example of distributed model, we can cite HYDROTEL model.
According to (Fortin et al., 2001), it is a spatially distributed hydrological
model with physical bases specially developed to facilitate the use of remote
sensing data and GIS data. HYDROTEL can be used to simulate stream flow,
spatial distribution of hydrological variables.
2.3
The SWAT model
Soil and Water Assessment Tool or SWAT (J. G. Arnold et al., 1998) is a comprehensive,
semi-distributed river basin model that requires a large number of input parameters,
which complicates model parameterization and calibration (J. G. Arnold et al., 2012).
As semi-distributed physically based simulation model, SWAT can predict the impacts
of LULC and management practices on hydrological regimes in watersheds with varying
soils, land use and management conditions over long periods and primarily as a strategic
planning tool (S. Neitsch et al., 2005). SWAT operates on a daily time step and is
designed to predict the impact of land use and management on water, sediment, and
agricultural chemical yields in ungauged watersheds (J. G. Arnold et al., 2012). All these
characteristics of the model along with the fact that the data needed for setting up the
model for this project in the study area are available, make the choice of the SWAT model
very appropriate.
14
CHAPTER 2. LITERATURE REVIEW
2.3.1
Application of the SWAT model worldwide
The SWAT model has gained reliability and consistency. It has been used worldwide
because of its applicability in many aspects of water resource projects. Many models are
being developed but physically based semi-distributed models such as SWAT are well
established models for analyzing the impact of land management practices on water,
sediment, agricultural chemical yields in large complex watershed (Setegn et al., 2008).
For instance, in the case of watershed management program for conservation and
development, runoff is one of the most important phenomena of hydrological cycle.
SWAT Model has been used for the estimation of surface runoff in India (Malunjkar et al.,
2015). In the case of multiple analysis, SWAT has performed runoff and sedimentation
analysis in the Blue Nile in Ethiopia (Easton et al., 2010). It has also been applied
to model a hydrological water balance (Setegn et al., 2008). Sensitivity analysis has
been performed by using SWAT for daily stream computation (Spruill et al., 2000)
in Kentucky watershed. SWAT was used to investigate the impact of climatic change
on stream ow in western Kenya (Githui et al., 2009). Results showed that significant
increase in stream flow may be expected in the coming decades as a consequence of
increased rainfall amounts. Clearly, SWAT can handle a large number of phenomenon in
watershed for planning and decision making.
2.3.2
Application of the SWAT model in the Volta Basin
The Volta Basin is one of the major river systems in Africa. Due to the importance of its
contribution to the riparian countries economy, the use of hydrological model has become
important in its water management. SWAT model has been applied for some case studies
within the basin. For instance, (Awotwi, Kumi, et al., 2015; Awotwi, Yeboah, & Kumi,
2015) used the SWAT model to assess the impact of land cover and climate changes on
water balance components of the White Volta. The results of this study suggested that
change in rainfall leads to corresponding change in all the water balance components.
The results also showed that land use plays a dominant role in changing basin hydrology
and stream flow in the White Volta Basin. SWAT was used to estimate groundwater
recharge in the White Volta river basin (Obuobie et al., 2008). The results of the study
15
2.3. THE SWAT MODEL
showed that the White Volta water balance indicates an important increase in the mean
annual discharge, surface runoff and shallow groundwater recharge. According to the same
results, SWAT shows that about 11% of the annual precipitation in the White Volta Basin
becomes discharge, which consists of 4% surface runoff and 7% base flow. On a larger
scale, SWAT was used to address calibration and uncertainties in West Africa including
the basins of river Niger, Senegal and Volta (Schuol & Abbaspour, 2006). The results
showed that SWAT can be used for large-scale water investigations.
16
Chapter 3
STUDY AREA
3.1
3.1.1
Study area presentation
The Black Volta
The Black Volta river basin is a transnational river system that lies from Mali, through
Burkina Faso, Ivory Coast and Ghana.
The Black Volta basin, the largest in the
catchments of the Volta basin, has a total area of 142056 km 2 of which 33.302 km 2
is located in Ghana representing 23.5% of the total (Barry et al., 2005). It contributes
about 18% of the annual flows of the Lake Volta (Andreini et al., 2000).
17
3.1. STUDY AREA PRESENTATION
Figure 3.1: Study area map
Referring to the Figure 3.1, the Black Volta basin lies between latitude 7°00 000 N and
15°00 000 N and longitude 5°300 000 W and 1°300 000 W. The present study will focus more on
the Ghana part of the Black Volta with a particular interest in the Bui hydropower plant.
The Black Volta Basin within Ghana (Figure 3.2) is located in the north west of the
country, between latitude 7°00 000 N and 11°00 000 N and longitude 2°00 000 W and 1°300 000 W.
18
CHAPTER 3. STUDY AREA
The elevation varies from 61 m to 784 m.
In the year 2000, the Black Volta was home to about 4.5 million people in Burkina Faso,
Ghana, Ivory Coast and Mali. The population density in the basin ranges from 8 to 123
people/Km2 with Lawra district in Ghana having the highest density. This population
is estimated to be about 8 million by the year 2025 (Allwaters Consult, 2012). According
to (Green Cross International, 2001), the population growth rate in the basin is about
3% a year.
Figure 3.2: Digital elevation map of Black Volta
Figure 3.2 shows the Black Volta Basin location in West Africa. It stretches from Mali,
Burkina Faso, Ivory Coast through Ghana. The elevations vary from 61m to 784 m.
The Shuttle Radar Topography Mission (SRTM) 90m Digital Elevation Data (DEM),
produced by NASA has been downloaded from http://srtm.csi.cgiar.org/
19
3.1. STUDY AREA PRESENTATION
In the Black Volta, the major land use is agriculture. The major farming system in
the sub-basin is bush fallow which is extensive. In the north of the basin, particularly
the Lawra district, lands are highly degraded both in terms of physical status and
fertility levels and can therefore hardly support meangfull crop cultivation (Barry
et al., 2005). Previous study showed that most of the land in Ghana is under tree
cover and crop land whereas artificial surface and associated areas are few (see figure
5.7 for land cover map). This gives an opportunity for development of agricultural
activitiesin the Ghana section. In contrast, most of the land in Burkina Faso seem to
be developed for food production with very few tree cover in the southern part. In
fact, the vegetation zones are oriented from north to south, from the sparsely vegetated
Sahel, to savannah regions, and the Guinea forest zone or rainforest in the extreme South.
In the Black volta basin, the rainfall is highly variable in time and space. According
to (Shaibu et al., 2012), it ranges respectively from 400 mm/year in the North to 1500
mm/year in the South. Over 70% of the annual total rainfall occurs in the months of
July, August and September, with a little or no rainfall in the months of November to
March in most parts of the basin. Mean monthly potential evapotranspiration exceeds
mean monthly rainfall for most of the year for the entire basin.
3.1.2
The Bui dam
The Bui hydropower project is located on the Black Volta River (see figure 3.2) at the
border of the Bole (Nothern Region), and Wenchi (Brong-Ahafo Region) districts in
north weastern Ghana. As the Black Volta in Ghana is more of forest land, portion of
the bui dam fall in the Bui national Park. According to the Bui Hydropower Authority
(http://www.buipowerauthority.com/node/102), the the projet reservoir, which occupies
444 Km2 of the bui gorge at full supply level, represents 21% of the area of the Bui
National Park. The Environmental and Social Impact Assessment of the Bui Hydropower
Project (ESIA) report stated that the land occupied by the project represents 50%
grassdland, 25% savannah woodland and 25% water and riverine gallery forest.
The Bui hydrpower plant comprises the main dam, two saddle dams and the reservoir.
20
CHAPTER 3. STUDY AREA
As reported by (ESIA, 2007), the main dam is a gravity dam made of roller compacted
concrete, located at the Bui gorge, at 150 Km upstream of the Lake Volta. The dam’s
height above ground level is 90 m. The first saddle dam is a rockfill dam, located
approximatively 500 m from the main dam, at the right bank of the river. The second
saddle dam, a zoned earth-fill, located at north of Bui Camp along right bank of river,
approximatively 1 Km from the the main dam. It is at 7 m above ground level. The
reservoir maximum operating level is 183 m. At full supply level, the reservoir has 440
km2 area, 40 km length with an average depth of 29 m. At minimum operating level
(167.2 m), it represents 288 Km2 . The storage volume at full operating level is 12,350
million m3 while at minimum operational level, it represents 6,600 million m3 . The main
characteristics of the dam is presented in the Table 3.1 below:
Table 3.1: Some features of the Bui hydropower plant
Project facility
Description
Installed capacity
400 MW
Unit Type
3 Francis Turbines/Generators of 133MW each
Guaranteed Peak Efficiency
>94%
Reservoir Area and Dimensions at Full Supply Level
Area: 440 km2 (21% of Bui National Park)
Length: 40 km
Average depth: 29 m
Minimum Operation Level
167.2 m
Full Supply Level
183 m
Storage at Minimum Operational Level
Total: 6,600 million m3
Storage at Full Supply Level
Total: 12,350 million m3
Active: 6,000 million m3
source:(ESIA, 2007)
According to the BPA, the power production at Bui started in May 2013 when the reservoir
water level had risen from about 100 meters above sea level to more than 168 meters above
sea level. This level was below the maximum operation level. Even in october 2013, that
is rainy season, the reservoir level had risen only up to 177.9 m. The PBA has therefore
been forced to plan a draw down strategy of the Bui reservoir as shown on the figure 3.3
where the generation of power was therefore to be restricted to an average of about 1.44
million kWh/day .
21
3.1. STUDY AREA PRESENTATION
Source: BPA
Figure 3.3: Bui Resevoir Plan Trajectory (Jan 1st-June 30th 2014)
22
Chapter 4
MATERIALS AND METHODS
4.1
Rainfall variability
Rainfall variability assessment is an important part of the present work. The first objective
is to use the available rain gauge data in the Black Volta basin to check possible change
in rainfall pattern at the basin level. The second is to use the rain gauge data to
drive the SWAT hydrological model in order to quantify possible implications of these
potential changes on stream flow for the benefit of the Bui hydropower plant. In order to
achieve these specific objectives, we first establish an aridity index map of the Black Volta
catchment using ArcMap sofware in order to define the climatic zones using the Global
Aridity Index dataset collected from Global Aridity and Potential Evapotranspiration
(PET) Database. Mann-Kendall trend test and the Sen’s slope estimator were then used
for possible monotonic trend detection in the Black Volta rainfall data and the statistical
significance of this trend.
4.1.1
Aridity Index map development
The Aridity Index (AI) map was developed for this project using the Global Aridity
Index dataset collected from Global Aridity and Potential Evapotranspiration (PET)
Database (http://www.cgiar-csi.org/data/global-aridity-and-pet-database) developed by
(Trabucco & Zomer, 2009). Precipitation availability over atmospheric water demand
can be quantified using an Aridity index (UNEP, 1997). The data is a mean of AI from
1950-2000 period at 30 arc second (1 arc second equal 1/3600 degree) spatial resolution.
23
4.1. RAINFALL VARIABILITY
The aridity is usually expressed as function of rainfall, temperature and PET. (Zomer et
al., 2008) expressed AI as:
(AI) =
M AP
M AE
(4.1)
Where M AP is the Mean Annual Precipitation and M AE is the Mean Annual PET.
The FAO defined PET as the as the evapotranspiration of a reference crop in optimal
conditions having the following characteristics: well watered grass with an assumed
height of 12 cm, a fixed surface resistance of 70 s/m and an albedo of 0.23 (Allen
et al., 1998). Note that FAO preferred the utilisation of the denomination Reference
crop evapotranspiration (ETo) to PET. The map is classified based on the generalized
classification scheme of (UNEP, 1997) for Global Aridity values (Table 4.1)
Table 4.1: Climate classes according to AI values
Value
Climate Class
<0.03
Hyper Arid
0.03-0.2
Arid
0.2-0.5
Semi-Arid
0.5-0.65
Dry Sub-Humid
>0.65
Humid
Source: (UNEP, 1997)
4.1.2
Rainfall data acquisition
The rainfall data for thirteen stations was collected from the Ghana Meteorological Agency
(GMA), the Direction General de la Meteorologie du Burkina Faso and Global Climate
Data (http://en.tutiempo.net/climate). Table 4.2 shows the general characteristics of the
study stations and their country locations since the Black Volta is a trans-boundary basin.
24
CHAPTER 4. MATERIALS AND METHODS
Table 4.2: General Characteristics of Rainfall Stations included in the variability analysis
Country
Gauging Station
Lat (N)
Lon (W)
Elevation (m)
Ghana
Sunyani
7.3328
-2.3281
296
Ghana
Wenchi
7.75
-2.1
322
Ghana
Bui
8.2361
-2.2772
176
Ghana
Bole
9.0319
-2.4762
295
Ghana
Wa
10.0667
-2.5
262
Burkina Faso Batie
9.8739
-2.9213
298
Burkina Faso Diebougou
10.9667
-3.25
298
Burkina Faso Boura
11.0333
-2.5306
Burkina Faso Dano
11.1490
-3.2931
304
11.75
-2.9333
260
Burkina Faso Bondoukuy
11.8450
-3.7639326
Burkina Faso Bobo-Dioulasso
11.1605
-4.3298
374
Burkina Faso Bomborokuy
12.9972
-3.9496
344
Burkina Faso Boromo
In climate data analysis, several methods have been developed for continuity and
consistency checking (Peterson et al., 1998; Reeves et al., 2007). For this study, the
double mass curve method has been applied to the in-situ rainfall data. The double
mass curve is used to check the consistency of rainfall data by comparing data for a
single station with that of a pattern composed of several other stations in the study area
(Searcy & Hardison, 1960). The point plotted in the double mass curve were fitted closely
in a straight line showing that there was no or was minimum error in data processing,
and there was no change in data collection method (see figure 4.1 as an example. In this
example, the Bui data was set as the reference data). In general, one can say that the
data is good enough for the investigations that were undertaken in this project.
25
4.1. RAINFALL VARIABILITY
Figure 4.1: Double Mass Curve
Basic statistics including average, standard deviation and coefficient of variation were
applied to the data used in this study in order to capture the general behaviour of rainfall
pattern over the respective period considered in the basin. The results are presented in
Table 4.3.
4.1.3
Trend analysis through the Mann-Kendall statistics
In hydrology and climatology, the tests for trend detection in time series data can be
considered as parametric test and non-parametric test. The parametric test requires the
data to be normally distributed and independent while the data is required to be only
independent for non-parametric test. For this project, two ranked non-parametric tests
have been used for detecting monotonic trends in the rainfall time series data. They
are namely Mann-Kendall trend test and Sens slope estimator. The Mann-Kendall test
has been widely used in hydrology and climatology for trend detection (Westmacott &
Burn, 1997; Cannarozzo et al., 2006; Oguntunde et al., 2006; Liu et al., 2008; Zhang et
26
CHAPTER 4. MATERIALS AND METHODS
Table 4.3: General statistics of the Gauging stations in the basin
Gauging Station
Period for rainfall
Average of
Average of
Coefficient of
variability
annual total
annual monthly mean SD
Variation (CV)
analysis
rainfall (mm)
rainfall (mm)
based on annual total
Sunyani
1976-2011
1179.8
98.714.9
16%
Wenchi
1976-2011
1231.0
102.814.8
14%
Bui
1954-2005
1141.1
95.1 19.7
21%
Bole
1976-2011
1090.2
90.814.0
15%
Wa
1976-2011
1020.4
86.414.7
18%
Batie
1970-2013
1095.6
91.316.8
18%
Diebougou
1970-2013
1030.6
85.914.6
17%
Dano
1970-2013
902.7
75.210.4
14%
Boromo
1970-2012
872.3
72.611.3
16%
Boura
1970-2013
893.5
74.510.8
15%
Bondoukuy
1970-2013
850.9
70.911.5
16%
Bobo-Dioulasso
1960-2012
1034.7
86.213.8
16%
Bomborokuy
1962-2001
724.7
60.410.1
17%
al., 2011; Gocic & Trajkovic, 2013). There are some advantages using this test. It does
not required data to be normally distributed. The results of the test are not affected by
the existence of outliers in the data (Lanzante, 1996) or inhomogeneous time series. The
test was originally used by (Mann, 1945) and the test statistic distribution was derived
by (Kendall. . . , 1962).
In the trend, the null hypothesis H0 assumes that there is no trend in the rainfall time
series data over a given period and it is tested against the alternative hypothesis H1
which considers that there is increasing or decreasing trend (Ahmad et al., 2015). The
theoretical calculation of the Mann-Kendall statistic S is given by:
S=
n−1 X
n
X
Sgn(xj − xi )
(4.2)
i=1 j=i+1
Where n is the number of data points, xi and xj are the data values in time series i and
j (j > i). Sgn(xj − xi ) is the sign function defined as:
27
4.1. RAINFALL VARIABILITY




−1, if xj − xi < 0



Sgn(Xj − Xi ) = 0, if xj − xi = 0





+1, if xj − xi > 0
(4.3)
The variance of the statistic S,V(S) is defined as:
n(n − 1)(2n + 5) −
q
P
ti (ti − 1)(2ti + 5)
i=1
V (S) =
18
(4.4)
Again n is the number of data points, ti is the number of the tied groups and q the group
of extent. Tied group can be defined as group of sample data which have the same value.
When n > 10, a standardized statistic Zs is calculated as below:



√S−1 , if S > 0



 V (S)
Zs =
0, if S = 0





 √S+1 , if S < 0
(4.5)
V (S)
Positive value of Zs indicates upward trend and negative value of Zs indicates downward
trend. The testing trend is done at a specific α level. The alpha level considered in this
study is α = 0.05 which corresponds to 95% confidence interval level. The null hypothesis
H0 is rejected when | Zs | > Z1−α/2 implying significant trend in the time series data
(Either increase when Zs > Z1−α/2 or decrease when Zs < −Z1−α/2 ); H0 is accepted
which implies there is no trend in the rainfall time series data (Mann, 1945; Kendall. . . ,
1962). Z1−α/2 can be found in the standard normal distribution table. However, α = 0.05
is considered in this study with its corresponding value Z1−α/2 =1.96. Another parameter
considered in the Mann-Kendall test is the is Kendalls tau. The tau value is considered
as the slope and varies between -1 and +1. A negative value of tau indicates decreasing
trend and positive value indicates increasing trend (A. McLeod & McLeod, 2011).
Before the application of Mann-Kendall test for trend detection in the rainfall time series
data at annual level, it is important to investigate if the data is serial correlated. For this
28
CHAPTER 4. MATERIALS AND METHODS
study, the serial correlation of the data was investigated using the acf (autocorrelation)
and pacf (partial autocorrelation) function is R (A. I. McLeod, 2005).
4.1.4
Sens slope estimator
The Sens slope estimator has been widely used in hydro-climatology to estimate the true
slope in time series data wherever it exists (Hirsch et al., 1991; Bayazit et al., 2004; Tabari
et al., 2011). Assuming that the existing trend is linear, the Sen’s slope can be used. The
trend in the data can theregore be represented by the equation 4.6:
g(t) = Qt + B
(4.6)
where Q is the slope and B a constant. The true slope, which is the change per unit time
can be computed based on the non-parametric method developed by (Sen, 1968). For N
pair of data in the sample, the slope is:
Qi =
xj − xk
j−i
f or i = 1, ...N
(4.7)
Where xi and xj are data values at time j and k (j > k) respectively. The median Qmed
of N values of Qi is the Sens slope estimator. The median is calculated as the equation 4.8:
Qmed =


Q(n + 1)/2
if N is odd

 1 (QN/2 + Q(N + 2)/2 ) if N is even
2
(4.8)
The two tailed test is used to test the significance of the median Qmed at a specific alpha
level. In this study, α = 0.05 is used. The value and sign of Qmed represent the steepness
and the trend of the data. The confidence interval for this value is computed as:
Cα = Z1−α/2
p
V (S)
(4.9)
With V (S) the variance of the Mann-Kendall statistic defined in the equation 4.4 while
Z1−α/2 is obtained from the standard normal distribution table.
29
4.2. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
Furthermore, M1 =
N −α
2
and M2 =
N +α
2
are computed. The lower and upper limits of
the confidence interval, Qmin and Qmax , are the M1 th largest and M2 th largest of the N
ordered slope estimates (Hollander et al., 2013). For this work, the Sens slope estimator
was computed in R software using the Package wq (Jassby & Cloern, 2015).
To estimate the value of B, the difference xi − Qti is calculated from the equation 4.6 for
the n values. The value of B is then considered as the median of the values obtained from
the differences.
4.2
Hydrology of the Black Volta catchment
In order to evaluate how rainfall variability over the Black Volta could impact the
hydropower production at the Bui dam, a hydrological model was used to evaluate the
stream flow. The stream flow consists of surface run-off, lateral flow and return flow.
Therefore, the soil type, the land cover and land use (LULC) changes as well as climate
conditions in the basin are likely to impact the dynamics of the hydrology in the basin. For
this project, Soil and Water Assessment Tool (SWAT), a comprehensive semi- distributed
model was chosen to assess the steam flow of the basin. The rationale behind this selection
has been discussed in the section of literature review.
4.2.1
Brief description of the SWAT Model
The SWAT is a semi-distributed physical based model. It was developed by USDA (U.S.
Department of Agriculture) Agricultural research Service (ARS) to predict the impact
of land use practices on water, sediment and agricultural chemical yields in large and
complex watersheds with varying soils, land use and management conditions over long
period (S. L. Neitsch et al., 2011). SWAT is a continuous time model, in other words
long term yields model where the model is not designed to simulate detailed, single-event
flood routing. The model simulation is daily time step based.
30
CHAPTER 4. MATERIALS AND METHODS
Figure 4.2: Schematic representation of the hydrologic cycle
The values given in the figure 4.2 are real figures for the Black Volta for this project
according to the input data used in the SWAT model. All the values are in millimeters
(mm). The Sun has been manually added.
The spatial complexity of the watershed is taken into account in the model by combining
the information from DEM, soil and land use. The SWAT model divides the watershed
into sub-basins. The sub-basins are further subdivided into hydrological response units
based on the land used and soil distribution. The model is divided into eight major
components including hydrology, weather, sedimentation, soil, temperature, crop growth,
nutrients, pesticides and agricultural managements.
Water balance is the driving force behind everything that happens in the watershed.
According to (S. L. Neitsch et al., 2011), the hydrology of the watershed can be separated
31
4.2. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
into two major divisions: the first division is the land phase of the hydrologic cycle and
the second division is the water or routing of the hydrologic cycle. The land phase of
the hydrologic cycle is shown in figure 4.2 which is the schematic representation of the
hydrologic cycle in the Black Volta basin. It controls the amount of water, sediments,
nutrients and pesticides loading to the main channel in each sub-basin. The movement of
water, sediments, nutrients and pesticides through the channel network to the watershed
outlet is considered as the routing phase.
The SWATs simulation of the hydrologic cycle is based on the following water balance
equation (S. L. Neitsch et al., 2011):
SWt = SWo +
t
X
(Rday − Qsurf − Ea − Wseep − Qgw )
(4.10)
i=1
Where:
SWt is the final soil water content (mmH2 O)
SWo is the initial soil water content (mmH2 O)
t is the time (days)
Rday is the amount of precipitation on day i (mmH2 O)
Qsurf is the amount of surface runoff on day i (mmH2O)
Ea is the amount of tevaporation on day i (mmH2O)
W seep is the amount of water entering in the vadose zone from the soil profil on day i (mmH2O)
Qgw is the amount of return flow on day i (mmH2O)
32
CHAPTER 4. MATERIALS AND METHODS
4.2.2
Surface runoff
Overland flow or surface runoff consists of flow along sloping surface. SWAT provides
two methods for modeling surface runoff: The Soil Conservation Service (SCS) curve
number method (SCS, 1972) and the Green and Ampt infiltration method (Green &
Ampt, 1911). SWAT simulates surface runoff volume and peak rates for each HRU based
on these two methods.
The curve number method is daily based time step when used for computing surface
runoff in the SWAT and it is unable to compute directly infiltration. Rather, the amount
of water seeping into the soil is computed as the difference between the amount of rainfall
and the amount of surface runoff. In the other hand the Green and Ampt infiltration
method compute directly the infiltration in the model but is requires data in sub-daily
increments (S. L. Neitsch et al., 2011).
The method used for this work is the curve number method due to the daily based data
that we have for the project. The curve number equation is given by (SCS, 1972) as:
Qsurf =
(Rday − Ia )2
(Rday − Ia + S)
(4.11)
Where: Qsurf is the accumulated runoff (rainfallexcess) in (mmH2O)
Rday is the rainfall depth for the day (mmH2O)
Ia is the initial abstraction which includes surface storage , interception and infiltration prior to runoff
S is the retention parameter (mm H2O) and it is defined as:
S = 25.4(
1000
− 10)
CN
(4.12)
Where CN is the curve number for the day. Ia is commonly given as 0.2S. Hence equation
4.11 becomes:
33
4.3. SWAT INPUT DATA COLLECTION AND ANALYSIS
Qsurf =
(Rday − 0.2S)2
(Rday + 0.8S)
(4.13)
Runoff is generated only when Rday > Ia . Typical curve number for moisture conditions
are classified into four hydrological groups such as: A for high infiltration, B for moderate
infiltration, C for slow infiltration and D for very slow infiltration. According to the U.S
National Resource Conservation Service (NRSC) Soil Survey Staff, a hydrologic group
is defined as a group of soil having similar runoff potential under the similar storm and
cover conditions (NRCS, 1986). The full description of the other components of the SWAT
model can be found in the theoretical documentation of the SWAT model (S. L. Neitsch
et al., 2011)
4.3
SWAT input data collection and analysis
The methodology used in this project can be summarized as in the flow chart
below figure 4.3. The details in the data collection and processing as well as model
setup, model simulation, calibration and validation are discussed in the following sections.
34
CHAPTER 4. MATERIALS AND METHODS
Figure 4.3: Flow chart of the steps in the SWAT model application for the Black Volta
4.3.1
The
Digital Elevation Model (DEM)
DEM
data
for
this
study
was
90
m
resolution
(http://srtm.csi.cgiar.org/SELECTION/inputCoord.asp).
from
the
SRTM
The DEM was used for
watershed delineation including stream definition, outlets and inlets definitions as
well as calculation of the sub-basins parameters. The slopes definition in the basin
were also based on the DEM. The DEM map of the study area can be seen in the figure 3.2.
4.3.2
The
Soil Data
soil
data
was
collected
from
the
FAO
(http://www.fao.org/geonetwork/srv/en/metadata.show?id=14116)
35
soil
(FAO,
database
2003).
4.3. SWAT INPUT DATA COLLECTION AND ANALYSIS
The importance of the soil data in the model is due to the fact that it allows the
determination of the soil texture, available water content, hydraulic conductivity, bulk
density and organic carbon content for different layer according to each soil type. For the
Black Volta, the dominant soil types are illustrated in figure ?? according to the SWAT
database code and the FAO soil unit classification.
Figure 4.4: Soil Map of the Black Volta
The data is the FAO soil data for Africa 2003. The legend presented in figure ?? is the
labeling of the dominant soil types in the Basin. In bracket are different soil texture of
each soil types as they were defined for the SWAT database. C= Clay, L=Loam, S=
Sand, S-C-L = Sandy Loam Clay, S-L= Sandy Loam, C-L= Clay Loam, L-S= Loamy
Sand.
36
CHAPTER 4. MATERIALS AND METHODS
4.3.3
Landsat data acquisition for LULC analysis
In order to assess land use land cover change in the Black Volta catchment, three land
cover maps where developed based on satellite image bands combination. The years
1987, 2000 and 2013 Landsat images were collected corresponding to 28 years period.
The Landsat 5 thematic mapper (TM) sensor data was processed for the year 1987, the
Landsat 7 SLC-on was processed for the year 2000 while the year 2013 data was collected
from the Landsat 8 Operation Land Image (OLI). In order to minimize classification error
and differences in the vegetation growth throughout the year, all the data were collected
for the month of January and the cloud cover is set to zero. All Landsat data were at
30 meters resolution. These satellite data are produced by the United States Geological
Survey (USGS) and freely available from the USGS Global Visualization Viewer (USGS
GLOVIS) platform (http://glovis.usgs.gov/).
For the Black Volta, 14 scenes were
necessary to cover the entire basin which correspond to the following Path/Row: 194/54,
195/55, 195/54, 195/53, 195/52, 195/51, 196/54, 196/53, 196/52, 196/51, 196/0, 197/53,
197/52, 197/51 (Figure 4.5).
Figure 4.5: The fourteen scenes covering the entire Black Volta
Each square in the figure 4.5 represents one scene. We have the combination of Landsat
8 bands 7, 5, 4, 3, 2, and 1 in Red, Green and Blue bands.
The bands were combined in to three bands (Red, Green Blue) for the image classification.
37
4.3. SWAT INPUT DATA COLLECTION AND ANALYSIS
For this study, the supervised maximum likelihood classification method was used to
develop the land use classes. The method is time consuming due to the training sample
selections and the possibility to confuse some classes. The images were processed in
ArcGIS software. The color balancing is also identified as main challenge are the mosaic
of all combined bands. Indeed, due to the difference in climatic conditions in a big
watershed like the Black Volta (14 scenes), the same type of land use can be represented
by two difference colors.
4.3.4
LULC classes definition
The classification of land use and land cover types in different classes was based on
the SWAT 2012 global land use and crops databases. The classification was performed
in comparison to previous land use classes identified in the basin (Allwaters Consult,
2012). The classification was also validated using Google Earth since it can give better
visualization of the land cover types on the ground. For this work, seven classes of land
use and land cover classes were selected or identified:
a) Water bodies: This class represents the stream networks (Black Volta river
and its tributaries), the reservoirs and any identifiable water bodies on the
ground.
b) Bare land: it represents all areas that lack vegetation throughout the year.
The classification bare land in ArcGIS is a real challenge due to the fact the
reflectance of some identified barren areas are being confused to some other
urban areas. It was noticed that these areas are mainly in the extreme north
of the basin.
c) Urban Areas: Urban land use considered for the SWAT project are high
populated areas in the basin. The urbanization rate due to population
growth is very significant according to the result of the land use land cover
change analysis.
d) Forest Evergreen: It corresponds to a group of trees holding their leaves
permanently all over the year. The decrease of such forest in the basin is
also significant probably due to deforestation in the basin.
38
CHAPTER 4. MATERIALS AND METHODS
e) Forest deciduous: this type of forest is not perennial all the year.
f) Grass Land: it represents all grasses, shrubs and all types of small vegetation.
g) Agricultural land: these types of lands are used for crop cultivation. The
agricultural practices in the basin are mainly rain-fed agriculture. However,
some irrigated areas are identified in the basin and are all grouped under
agricultural land.
4.3.5
Accuracy assessment of the developed LULC Maps
Classification map or thematic map is an efficient approach to derive information from
an image However, it is also subjected to error which can be geometric errors, incorrectly
training sample labelling before supervised classification, unidentified classes etc.
A statistical approach to quantify these errors is the random selection of pixels from
the classification map to be compared to the reference map which produces a confusion
matrix. Confusion matrix has been widely used for accuracy assessment of the land use
land cover maps (Stehman & Czaplewski, 1998; Foody, 2002; Zăvoianu et al., 2004; van
Vliet et al., 2011).
The main statistical information derived from the confusion matrix are: overall accuracy,
commission error, omission error, the producers accuracy, the users accuracy and Kappa
Coefficient K. The overall accuracy is the total accuracy of the classification.
The
producers accuracy is considered as the probability that a land use of an area is classified
as such while users accuracy is the probability that a pixel labeled as a land cover class
in the developed map is as such. Commission error results from the fact that pixels
that belong to another class (the true class) are labeled as belonging to the class under
consideration. The omission represents pixels that belong to the real class but failed to
be classified into proper class. The Kappa coefficient is a discrete multivariate technique
of use in accuracy assessment. K>0.8 usually considered as strong agreement and good
accuracy (Carletta, 1996). The results in the case of thematic maps developed for this
work are under the section of results and discussion.
39
4.4. SWAT MODEL SETUP
4.3.6
Hydrological data
The stream flow data used in this work corresponds to a monthly stream flow data of
the Black Volta measured at Bui from 1954 to 2010. The data was collected from the
Bui Power Authority (BPA). The maximum flows are observed in the months of August,
September and November. The annual average, maximum, minimum and coefficient of
variation of flow for all the months of the period 1954-2010 are respectively 227.6m3 /s,
612.2m3 /s, 70.7m3 /s and 40%.
Figure 4.6: Bui monthly average flow
4.4
SWAT model setup
4.4.1
Watershed delineation
As a hydrological model, the starting point of the SWAT database creation is the
watershed definition. The watershed definition consists of streams definition, outlets inlets
definition and calculation of the sub-basins parameters. During the streams definition
phase, flow direction is computed for each cell in the DEM raster to determine the water
destination in it. For the Black Volta basin in this project, the model gives 451105.6
40
CHAPTER 4. MATERIALS AND METHODS
Ha as threshold for calculating the sub-basins parameters. A total of 17 outlets and 17
sub-basins are defined for the Black Volta with total area of 130785.06 Km2 (figure 4.7).
This area must only be considered as for the watershed with outlet defined at Bui. The
real Black Volta area is bigger than the one stated above (155076.4 Km2 ). The slopes
and elevations were also generated as a topographical parameter of the watershed using
the DEM as an important factor for the basin. The slopes in combination with soil and
land use control the surface runoff. In this work, five slopes classes were defined as follow:
0-2%, 2-4%, 4-10%, 10-15% and greater than 15%. The highest elevation of the basin is
784 and the smallest is 79 m above mean sea level.
4.4.2
HRUs analysis
Each sub-basin in the watershed is divided into units called hydrological response units
(HRUs). The HRU is the smallest unit in the SWAT model and it is used to simulate
processes such as rainfall, infiltration, plant dynamics, erosion, nutrient cycling, leaching
of pesticides etc.(S. L. Neitsch et al., 2011). HRUs are portions of the sub-basins having
unique land use, management and soil attributes. For the Black Volta, 382 HRUs were
created for the 17 sub-basins.
In the SWAT model, there are three types of HRU definition options including dominant
land use, soils, slope, dominant HRU and multiple HRU option. In this work, multiple
HRUs option was chosen. Land use class percentage over sub-basin areas was set to 15%,
soil class percentage over land use area was 5% and slope class percentage over soil area
was 15%.
41
4.4. SWAT MODEL SETUP
Figure 4.7: Black Volta watershed as defined in the SWAT model. The numbers represent
each sub-basin (total of 17 sub-basins)
4.4.3
Weather generator
SWAT input data are daily based. The Black Volta basin is a data scarce watershed
where very few stations have all the climate data required by the SWAT model. Even
when the data is available, some of the weather stations present gaps in the data
recorded sometimes up to years. The SWAT model provides a useful approach to provide
comprehensive representative data for the basin: a weather generator in the WGN file.
It contains the statistical data needed to generate representative daily climate data for a
sub-basin. If the measured data contains missing values (-99 represents missing data in
42
CHAPTER 4. MATERIALS AND METHODS
the SWAT model), the weather generator is activated to produce data to replace missing
values. Weather generator is also used when user specify that simulated weather should
be used. For this project, only rainfall and temperature data for 13 stations (see figure
3.1) were provided as observed data. The remaining climate variables such as solar
radiation, wind speed and relative humidity were all simulated based on the weather
generator we built for the Black Volta.
According to (J. Arnold et al., 2012), ideally, at least 20 years of records are used to
calculate parameters in the WGN file. For the Black Volta, data from 5 climate stations
such as Sunyani, Bole, Wa, Boromo and Bomborokuy were used to build the weather
generator. Apart from the latitude, the longitude and elevation that are needed for each
station, 15 other statistical parameters are needed to build an independent representative
weather generator such as:
• RAINYRS: the number of year of maximum monthly 0.5 h rainfall data used to
defined values for RAINHHMX(month).
• TMPMX (month): average daily maximum air temperature for a month (◦ C).
• TMPMN (month): average daily minimum air temperature for a month (◦ C).
• TMPSTDMX (month): standard deviation for daily maximum air temperature in
month (◦ C).
• TMPSTDMN (month): standard deviation for daily minimum air temperature in
month (◦ C).
• PCPMM (month): average precipitation for a month (mm H2O).
• PCPSTD (month): standard deviation for daily precipitation in month (mm
H2O/day).
• PCPSKW (month): skew coefficient for daily precipitation in a month.
• PRW (1, month): probability of wet day following a dry day in a month.
• PRW (2, month): probability of wet day following a wet day in a month.
43
4.4. SWAT MODEL SETUP
• PCPD (month): average number of precipitation in a month.
• RAINHHMX (month): maximum 0.5 rainfall in entire period of record
• SOLAR (month): average daily solar radiation for a month (MJ/m2/day).
• DEWP (month): average daily dew point for each month (◦ C).
• WNDAV (moth): average daily wind speed in month (m/s).
Solar
the
radiation
NASA
and
Prediction
wind
of
speed
data
Worldwide
of
22
Energy
years
Resource
(http://power.larc.nasa.gov/cgi-bin/cgiwrap/solar/agro.cgi).
were
collected
(POWER)
from
website
It is a Climatology
Resource for Agro climatology with a global coverage on a 1 latitude by 1 longitude grid
resolution data. The rest of the parameters were computed based on station data of 36
years (1976-2011) except for RAINHHMX were the number of years varying according
to the station. Stations such as Sunyani, Bole and Wa, 36 years of data were used
to compute the RAINHHMX. For Boromo and Bomborokuy, 43 and 40 years of data
respectively were used (see table 4.4). Most of the parameters were computed based on a
program developed by (Liersch, 2003). The theoretical background of all the parameter
can be found in the SWAT theoretical documentation (S. L. Neitsch et al., 2011).
44
CHAPTER 4. MATERIALS AND METHODS
Table 4.4: Weather generator for the Black Volta catchment
STATION
TMPMX1
TMPMX2
TMPMX3
TMPMX4
TMPMX5
TMPMX6
TMPMX7
TMPMX8
TMPMX9
TMPMX10
TMPMX11
Sunyani
38.3
38.6
42.2
37
38
33.9
62
34.8
32.7
33
36.2
TMPMX12
35
Bole
39.6
41.2
41.5
41
38.5
36.1
34.4
33.8
33.8
35.5
38.1
39
Wa
40
42
41.5
29.4
28.3
27
25.6
25
24.9
31
25
26
Boromo
37.07
38.86
40.75
40.86
39.87
36.7
33.58
32.93
33.63
37.6
37.83
36.89
Bomborokuy
37.49
33.21
35.28
38.67
38.64
36.48
36.56
40.81
41.35
42.3
41.13
40.43
STATION
TMPMN1
TMPMN2
TMPMN3
TMPMN4
TMPMN5
TMPMN6
TMPMN7
TMPMN8
TMPMN9
TMPMN10
TMPMN11
TMPMN12
Sunyani
11.1
15.6
16.1
2.7
18.9
Bole
11
13.5
15.6
19.2
18.7
18.5
18.7
18
17.3
18.5
17.3
16.5
13
18.7
18.4
17.2
15
9.5
10.1
15.5
Wa
14
17.2
17.9
17.5
17.4
18.4
17.9
18.5
18.3
19.1
16
Boromo
14.29
17.38
20.54
23.9
23.62
21.65
21.24
21.1
20.57
20.89
15.96
14.7
Bomborokuy
21.24
21.12
21.34
21.48
19.2
16.05
15.19
17.61
20.82
24.4
22.93
21.89
TMPSTDMX3 TMPSTDMX4
TMPSTDMX5
TMPSTDMX6 TMPSTDMX7 TMPSTDMX8
STATION
TMPSTDMX9
TMPSTDMX10
Sunyani
TMPSTDMX1 TMPSTDMX2
1.928754
1.59978
1.916871
1.690159
1.487787
1.489963
1.431596
1.528457
1.587358
1.347414
TMPSTDMX11 TMPSTDMX12
1.108275
1.389257
Bole
1.940902
1.784403
1.970821
2.493682
2.103159
1.828475
1.678322
1.644368
1.77286
1.816264
1.254957
1.447916
Wa
2.170333
2.054959
1.926866
2.068584
1.904353
1.61451
1.227699
1.082989
1.222844
1.159577
1.761242
1.822124
Boromo
1.50633848
1.11285359
0.92367809
1.001449
1.218718
1.072818
0.798733
0.794106
0.816292
1.023734
0.813436
1.17384362
1.73784151
1.08416778
1.66205534
2.363019
1.755186
1.418057
1.58357
1.809507
1.508802
1.723288
2.217413
2.52452632
TMPSTDMN3 TMPSTDMN4
TMPSTDMN5
TMPSTDMN6 TMPSTDMN7 TMPSTDMN8
TMPSTDMN9
TMPSTDMN10
Bomborokuy
STATION
TMPSTDMN1 TMPSTDMN2
TMPSTDMN11 TMPSTDMN12
Sunyani
2.648393
1.964232
1.404579
1.457291
1.113235
1.052576
0.847941
0.767028
0.858563
0.907864
1.093428
Bole
2.639039
2.740039
2.254144
1.618123
1.303127
1.195202
1.037339
0.985781
1.067948
0.987957
2.342245
2.429144
2.9333
Wa
2.19145
2.87758
1.918698
2.068584
1.883504
1.61451
1.218834
1.082989
1.222844
1.16021
1.761242
1.822124
1.12424139
Boromo
1.34943548
1.50838839
1.28934408
0.96789
0.937109
0.749007
0.530571
0.455992
0.584499
0.701879
1.275175
Bomborokuy
0.9062698
0.52444072
0.51729237
0.9225
1.056645
1.297663
1.705204
1.67778
1.409262
0.81425
1.277607
1.30318171
STATION
PCPMM1
PCPMM2
PCPMM3
PCPMM4
PCPMM5
PCPMM6
PCPMM7
PCPMM8
PCPMM9
PCPMM10
PCPMM11
PCPMM12
Sunyani
10.18
45.74
101.12
142.88
141.9
183.36
95.55
73.48
159.76
170.47
45.03
15.64
Bole
2.36
14.03
51.11
102.66
133.44
144.61
147.65
164.87
212.8
94.08
17.08
5.46
Wa
8.26
7.4
23.3
82.49
119.7
135.75
152.4
210.33
192.44
87.76
9.9
5.33
Boromo
0.97
6.68
42.19
73.7
114.33
175.39
251.07
157.67
44.68
4.2
0.52
0.89
Bomborokuy
0.2
0.28
4.59
14.51
50.31
104.28
178.87
218.33
117.47
31.51
2.92
1.38
STATION
PCPSTD1
PCPSTD2
PCPSTD3
PCPSTD4
PCPSTD5
PCPSTD6
PCPSTD7
PCPSTD8
PCPSTD9
PCPSTD10
PCPSTD11
PCPSTD12
Sunyani
2.95059
7.915
9.5076
11.5416
10.1674
12.8363
9.4473
7.8553
11.6175
10.741
4.9468
3.412
Bole
1.1403
4.8277
7.0725
9.8296
10.8699
10.1237
11.6467
12.3723
13.5479
8.2321
4.0034
1.8522
Wa
2.6508
2.4215
4.0202
8.9577
10.081
10.6446
10.3525
13.8381
11.3539
7.529
1.5185
1.5132
Boromo
0.5033
1.9854
6.6567
7.8165
8.4925
11.6435
14.0921
10.2366
5.4537
1.8095
0.4655
0.6815
Bomborokuy
0.1971
0.2002
1.5691
3.3629
6.4563
8.8255
12.8346
13.4358
9.0428
4.3881
1.2113
1.2954
STATION
PCPSKW1
PCPSKW2
PCPSKW3
PCPSKW4
PCPSKW5
PCPSKW6
PCPSKW7
PCPSKW8
PCPSKW9
PCPSKW10
PCPSKW11
PCPSKW12
Sunyani
11.3749
8.0049
3.9529
3.5396
3.2534
3.446
4.914
7.0382
3.8516
3.4129
4.5907
9.9571
Bole
22.3037
15.2527
7.1082
4.1291
3.8288
2.9994
3.9575
3.7805
3.7778
4.4744
10.9512
14.434
Wa
15.1531
22.524
7.4912
4.7305
4.5455
4.1443
3.0779
3.3311
2.6075
4.1022
9.1012
19.0499
33.7808
Boromo
20.5001
16.8678
9.4888
5.4683
3.5642
2.9776
2.8112
3.5282
5.5908
19.3229
33.1014
Bomborokuy
34.6179
21.1973
14.1417
11.853
6.9585
3.4992
3.5945
2.8879
3.5346
6.6331
14.2431
33.9248
STATION
PR W1 1
PR W1 2
PR W1 3
PR W1 4
PR W1 5
PR W1 6
PR W1 7
PR W1 8
PR W1 9
PR W1 10
PR W1 11
PR W1 12
Sunyani
0.0258
0.1085
0.2396
0.3462
0.3768
0.4657
0.2865
0.2874
0.4426
0.3997
0.1427
0.0415
Bole
0.0118
0.0421
0.1345
0.2594
0.32
0.3988
0.345
0.4111
0.5818
0.2375
0.0505
0.0174
Wa
0.0065
0.0239
0.0767
0.2335
0.3129
0.3708
0.3865
0.4856
0.501
0.219
0.0338
0.0084
0.0045
Boromo
0.0053
0.0355
0.1305
0.2574
0.3581
0.4814
0.5596
0.4047
0.1375
0.0092
0.0031
Bomborokuy
0.0016
0.0027
0.0139
0.0537
0.1571
0.2909
0.3916
0.4693
0.3193
0.0994
0.0059
0.0016
STATION
PR W2 1
PR W2 2
PR W2 3
PR W2 4
PR W2 5
PR W2 6
PR W2 7
PR W2 8
PR W2 9
PR W2 10
PR W2 11
PR W2 12
Sunyani
0.0667
0.187
0.2563
0.2535
0.3848
0.4791
0.423
0.3804
0.5464
0.5451
0.3284
0.1607
Bole
0.0714
0.0682
0.172
0.2086
0.3115
0.3381
0.419
0.5038
0.4684
0.3513
0.1714
0.1852
Wa
0.8
0.5273
0.1919
0.2088
0.2951
0.34
0.4423
0.4964
0.5251
0.4049
0.6442
0.7234
0.4
Boromo
0.2222
0.2459
0.2094
0.2914
0.2789
0.441
0.5583
0.4837
0.2715
0.375
0
Bomborokuy
0.3333
0
0.1429
0.0462
0.1531
0.1688
0.2905
0.3813
0.3038
0.1888
0.1
0.6
STATION
PCPD1
PCPD2
PCPD3
PCPD4
PCPD5
PCPD6
PCPD7
PCPD8
PCPD9
PCPD10
PCPD11
PCPD12
Sunyani
0.83
3.42
7.69
9.86
12.06
14.61
10.64
10.22
15.25
15.08
5.67
1.56
Bole
0.39
1.22
4.36
7.72
10.17
11.75
12
14.44
16.25
8.78
1.94
0.75
Wa
1.11
1.53
2.75
6.92
9.69
11.19
13.25
15.61
16.03
9.06
2.89
1.31
0.23
Boromo
0.21
1.42
4.44
8.14
10.67
14.35
17.74
14.28
5.14
0.56
0.12
Bomborokuy
0.08
0.08
0.52
1.63
4.9
8
11.27
13.9
9.88
3.58
0.25
0.13
STATION
RAINHHMX1
RAINHHMX2
RAINHHMX3
RAINHHMX4
RAINHHMX5
RAINHHMX6
RAINHHMX7
RAINHHMX8
RAINHHMX9
RAINHHMX10
RAINHHMX11
RAINHHMX12
Sunyani
15.576
38.973
25.113
28.38
24.288
40.194
31.68
42.933
32.274
33.858
15.048
18.48
Bole
12.243
26.334
36.399
26.829
30.558
33.363
42.933
40.26
45.936
31.251
24.42
14.949
12.21
Wa
18.447
22.407
14.157
32.109
41.58
37.455
28.38
43.857
25.839
22.44
8.052
Boromo
5.148
19.272
30.624
64.119
67.056
95.799
154.869
99.297
43.032
15.015
5.346
9.141
Bomborokuy
2.508
1.485
18.447
25.311
45.573
75.867
103.092
125.169
97.548
29.865
12.078
15.048
STATION
SOLARAV1
SOLARAV2
SOLARAV3
SOLARAV4
SOLARAV5
SOLARAV6
SOLARAV7
SOLARAV8
SOLARAV9
SOLARAV10
SOLARAV11
SOLARAV12
Sunyani
19.487
20.44
20.45
19.42
18.255
15.93
13.97
12.99
13.48
15.54
17.29
18.119
Bole
20.6461
21.69
22.3185
21.23
17.979
20.406
15.87
14.958
16.555
19.08
19.07
20.456
Wa
20.77
21.82
22.45
21.86
21.25
19.69
17.51
16.5
17.9975
20.248
20.28
20.65
Boromo
18.94355
18.10323
19.15516
20.80935
20.32645
20.04129
20.22774
21.75129
22.32806
22.36839
21.9929032
20.6087097
Bomborokuy
19.96806
19.00387
20.02484
20.9471
20.22968
19.2071
19.57968
21.8329
22.61903
22.70226
22.3683871
21.0664516
STATION
DEWPT1
DEWPT2
DEWPT3
DEWPT4
DEWPT5
DEWPT6
DEWPT7
DEWPT8
DEWPT9
DEWPT10
DEWPT11
DEWPT12
Sunyani
17.4
22.08
25.66
25.4
25.14
24.32
23.41
23.12
23.75
23.8
23.3
20.23
Bole
9.08
13.47
21.06
24.89
24.85
24.27
23.7
23.39
23.36
22.51
18.38
11.44
Wa
5.82
9.29
17.07
23.73
24.53
24.54
24.19
23.78
23.36
21.89
15.94
8.42
Boromo
6.12
6.89
11.25
18.91
22.39
22.75
22.65
22.67
22.9
22.2
15.89
9.07
Bomborokuy
22.49
22.72
22.13
19.38
9.74
2.65
0.28
0.25
5.95
14.93
20.1
21.81
STATION
WNDAV1
WNDAV2
WNDAV3
WNDAV4
WNDAV5
WNDAV6
WNDAV7
WNDAV8
WNDAV9
WNDAV10
WNDAV11
WNDAV12
Sunyani
1.81
1.96
2.167
2.335
2.08
2.069
2.285
2.189
1.7889
1.71
1.7
1.639
Bole
2.53
2.6057
2.53
2.7947
2.52
2.685
2.6667
2.43669
1.915
2.0822
2.013
2.234
45
Wa
2.895
2.88
2.65
2.9
2.91
2.69
2.74
2.39
1.95
2.216
2.011
2.44
Boromo
1.360408
1.448163
1.566735
1.802245
1.892449
1.689592
1.443265
1.154694
0.906939
0.967551
0.90877551
1.1344898
Bomborokuy
2.76871
2.350645
2.067419
2.092903
2.257419
3.15871
3.567097
3.602581
3.087419
2.906452
3.21322581
3.10322581
The table was built based on the available data but also we tried to make sure that it
4.5. SENSITIVITY ANALYSIS, CALIBRATION AND VALIDATION IN THE
SWAT-CUP
4.5
Sensitivity analysis, calibration and validation in
the SWAT-CUP
Soil and Water Assessment Tool-Calibration and Uncertainty Programs (SWAT-CUP)
is an automated calibration model which provides link between the input/output of a
calibration program and the model. It is a generic interface that was developed for
calibrating the SWAT model.
4.5.1
Sensitivity analysis
Measured monthly river flow at Bui from 1954 to 2010 was used for calibration and
validation of the SWAT model applied to the Black Volta catchment. As semi-distributed
model, SWAT has several parameters and it quasi impossible to calibrate all of them. A
sensitivity analysis is therefore needed to determine the most sensitive parameter in the
basin for the calibration process. The sensitivity analysis was performed in two ways:
first by varying one parameter at a time while keeping the others constant, second by
varying all the parameters simultaneously. Finally, twelve parameters were used during
the calibration (see table 4.5)
46
CHAPTER 4. MATERIALS AND METHODS
Table 4.5: Most sensitive parameters
Parameter Name Definition
Absolute SWAT Values
Values
0 − 2000
r SOL K
Saturated hydraulic conductivity
v MSK CO2
Calibration coefficient used to control impact of the storage time constant for low flow
v SURLAG
Surface runoff lag time
v MSK C01
Calibration coefficient used to control impact of the storage time constant for normal flow
r SOL AWC
Available water capacity of the soil layer
r CN2
SCS runoff curve number f
a GWQMN
Threshold depth of water in the shallow aquifer required for return flow to occur (mm)
v ALPHA BF
Base flow alpha factor (days)
v GW DELAY
Groundwater delay (days)
v RCHRG DP
Deep aquifer percolation fraction
0-1
v ESCO
Soil evaporation compensation factor
0-1
v CH N2
Manning’s ”n” value for the main channel
0-10
0.05 − 24
0-10
0-1
−0.2 − 0.2
0-5000
0-1
0-500
−0.01 − 0.3
The type of change to be applied to the parameter in the model are defined by the code
r , v , and a . According to (Abbaspour et al., 2007), r
means the existing parameter
value is multiplied by (1+ a given value), v means the default parameter value is replaced
by a given value, a
4.5.2
means a given quantity is added to the default value.
Calibration and validation
The calibration was performed based on the parameters in table 4.5. According to Xu
(2002), model calibration are usually based on the following objectives:
47
4.5. SENSITIVITY ANALYSIS, CALIBRATION AND VALIDATION IN THE
SWAT-CUP
a) A good agreement between the of average simulated and observed catchment
runoff volume (i.e. good water balance)
b) A good overall agreement of shape of the hydrograph
c) A good agreement of the peak flow with respect to timing, rate and volume
d) A good agreement for low flows
To fulfill the above objectives, the Sequential Uncertainty Fitting (SUFI- 2) model by
(Abbaspour et al., 2004) for optimization and uncertainties analysis was used in the
SWAT-CUP for calibration and validation. The Nash-Sutcliff (SN) coefficient (Nash
& Sutcliffe, 1970) was assigned as the objective function.
In SUFI-2, a parameter
uncertainty is propagated (as uniform distribution) through a Latin Hypercube
(statistical method for generating a sample of plausible collections of parameter values
from a multidimensional distribution) sampling (Schuol & Abbaspour, 2006).
It is
referred to as the 95% depicting prediction uncertainty or 95PPU (known as P-factor)
calculated at 2.5% and 97.5% levels for each parameter. The 95PPU is the degree to
which all uncertainties are accounted for (Abbaspour, 2013). The average thickness of
the 95PPU band divided by the standard deviation of the measured data quantify the
strength of the calibration and uncertainty analysis is known as R-factor. The perfect
situation would be 100% of the observed data bracketed in the 95PPU while at the same
time R-factor is close to zero (Abbaspour, 2013). SUFI-2 account for all uncertainties
including uncertainties in driving variable like rainfall, conceptual model, parameters
and measured data.
For this project, two sets of calibration and validation were performed based on the land
use of the year 2000 and 2013. The first calibration covered the period 2000-2005 and the
validation for the period 2006-2010 using the 2013 land use data. The 2013 Landsat 8 data
was used assuming that within 3 yeas, there was not significant changes in the basin. The
Landsat 8 was the latest among all landsat satellites and gives better images.The second
calibration was performed for the period 1990-1995 and the validation for the period
1996-2000 using the land use data of the year 2000. A model validation is comparing the
model output to an independent set of data without making any further adjustment in
the parameters.
48
CHAPTER 4. MATERIALS AND METHODS
4.5.3
Model performance evaluation
For the model evaluation, two most widely used statistical test in hydrology are considered:
i. Goodness of fit or coefficient of determination (R2 ) between the observation
and the final best simulation:
n
P
(Qobs,i Qobs )(Qsim,i − Qsim )
i=1
2
R = P
n
n
P
[ (Qobs,i − Qobs )2 ]0.5 [ (Qsim,i − Qsim )2 ]0.5
i=1
(4.14)
i=1
ii. And the Nash-Sutcliffe coefficient(NS) Nash & Sutcliffe (1970):
n
P
NS = 1 −
(Qsim,i − Qobs,i )2
i=1
n
P
(4.15)
(Qobs,i − Qobs )2
i=1
Where:
Qobs,i is the observed flow at time i (m3/s)
Qobs is the mean flow at time (m3/s)
Qsim,i is the stimulated flow at time i (m3/s)
Qsim is the mean of stimulated flow at time (m3/s)
The Nash-Sutcliffe coefficient rating (table 4.6) on monthly time step is given by (Moriasi
et al., 2007).
49
4.5. SENSITIVITY ANALYSIS, CALIBRATION AND VALIDATION IN THE
SWAT-CUP
Table 4.6: General performance ratings for recommended statistics for a monthly time
step
Performance rating
NS
Very Good
0.75 <NS≤1.0
Good
0.65 <NS≤0.75
Satisfactory
0.50<NS≤0.65
Unsatisfactory
NS≤0.5
Source: (Moriasi et al., 2007)
(Awotwi, Kumi, et al., 2015) stated that for the model calibration to be accepted, R2
should be greater than 0.6 and the NS greater than 0.5.
50
Chapter 5
RESULTS AND DISCUSION
5.1
Aridity index and rainfall variability over the
Black Volta
5.1.1
Aridity index
The aridity index (AI) map of the Black Volta in this project is based on the global
aridity database which consists of an average of AI from 1950-2000 period at 30 arc
second (1 arc second equal 1/3600 degree) spatial resolution. Precipitation availability
over atmospheric water demand can be quantified using an AI, an important factor to
understand the rainfall variability since it reveals the climatic conditions in the basin.
The results for AI in the catchments are presented in Figure 5.1. The AI of Black Volta
shows three climatic conditions: Semi-arid, dry sub-humid and humid. The AI of the
semi-arid zone varies from 0.2 to 0.46 and it is located in the northern part of the
catchment (Burkina and Mali). For the dry sub-humid zone, the AI varies from 0.46
to 0.58 and it is located in the middle of the basin (Burkina Faso and Ghana). The
humid zone that is the southern of the basin (Ghana and Ivory Coast) has AI between
0.58 and 0.84.
51
5.1. ARIDITY INDEX AND RAINFALL VARIABILITY OVER THE BLACK VOLTA
Figure 5.1: Aridity Index and rainfall map of the Black Volta Basin
5.1.2
Rainfall variability through trend analysis
In order to understand the rainfall variability for possible climate change detection
in the basin, the Mann-Kendall trend test and the Sens estimator were applied to
thirteen rainfall stations throughout the entire basin. The statistical indicators from
the trend test are the Kendall tau, the standardized statistic Zs , the pvalue, and
the Sens slope. The Kendall tau is a measure of correlation; therefore measures the
strength of the relationship between the two variables (time and rainfall) based on
the rank of the data. The statistic Zs and the pvalue measures the significance of the
trend while the Sens slope estimates the magnitude of the trend (similar to the tau value).
The test was performed based on the annual average for different periods that correspond
to reliable data availability. Along with these trend tests, the moving average method
(for four years) was also applied to show graphical representation of the general trend in
52
CHAPTER 5. RESULTS AND DISCUSION
the time series data. The results are summarized in table 5.1 and figure 5.1. The results
showed that about 62% of the stations in the basin (that is eight out of thirteen) presented
an increased trend in precipitation while 38% presented a decrease in precipitation (five
stations out of thirteen). However very few of these trends are statically significant.
In the humid zone of the Black Volta (that is Ghana and Ivory Coast), all the stations
presented an increase in rainfall except for Bui. The analysis for Bui correspond to
the period 1954-2005. According to the Mann Kendall trend test, the tau value is
-0.0724 while the Sens slope is -0.1283 suggesting a decrease in rainfall since 1954. The
pvalue that is 0.4535 (greater than 0.05) suggested that although there is a decrease in
rainfall, the trend is not statistically significant. This was confirmed by the statistic Zs
equals -0.7497 (weak compared to 1.96). Let’s recall that as discussed in the theoretical
background of the Mann-Kendall statistics, for the trend to be statistically significant,
Zs should be greater than 1.96. Results in figure 5.2 show that four years moving average
for the Bui station is in a wave form with a decreasing peak. The rainfall pattern at Bui
also showed that most of the annual average are within the band mean plus one standard
deviation and mean minus one standard deviation with very few years outside this band.
Figure 5.2: Bui rainfall 1954-2005
The rest of the stations studied in the humid zone showed an increased trend in rainfall
and this correspond to the period 1976-2011 (see figures B.6, B.8, B.7 in appendix B).
53
5.1. ARIDITY INDEX AND RAINFALL VARIABILITY OVER THE BLACK VOLTA
Although the overall statistics of these stations presented an increased trend in rainfall
during the period 1976-2011 (none were statistically significant), some of the stations
presented some years of continuous decreased trend in rainfall (which might be considered
as drought years). Wa station is an example. The four years moving average (figure
5.3) applied to the Wa rainfall revealed three distinctive patterns in rainfall variability:
1976-1985, 1986-1997, 1998-2005. The period 1976-1985 showed as continuous decrease
in rainfall while the period 1986-1997 showed a recovery with continuous increased trend.
Finally, the period 1998-2005 presented a kind of stabilized years in rainfall. The overall
Mann-Kendall trend test suggested an increased trend in rainfall for the Wa station with
the pvalue equals 0.2819 and the statistic Zs equals 1.076 proving the non-significance of
the trend.
Figure 5.3: Wa rainfall 1976-2011
In the dry sub-humid zone (in Burkina part of the basin), all the stations presented
an increased trend in rainfall during the period 1976-2011 (see figures B.3, B.5 in B)
but statistically non-significant except Boura. In fact, Boura rainfall (figure 5.4) is the
only station which presented a statistically significant trend in the basin for all period
considered.
54
CHAPTER 5. RESULTS AND DISCUSION
Figure 5.4: Boura rainfall 1970-2013
The Mann-Kendall tau is 0.279 and the Sens slope 0.3218 reveals the magnitude of the
upward precipitation for the corresponding period. The pvalue that is 0.0171 (less than
alpha 0.5) and the statistics Zs equals 2.38 (greater 1.96) reveal that the increased trend
is statistically consistent. The moving average (figure 5.4) showed continuous increased
trend. More importantly, it showed that from the year 1995, the trend is more consistent.
This was confirmed by the fact that the magnitude of dry years (peaks below the green
line or mean one standard deviation) are being reduced. Similar results are also found
for Diebougou (figure B.5) and Dano (figure B.3) showing a logical pattern in rainfall for
the dry sub humid zone.
In the semi-arid zone (that is Burkina and Mali part of the catchment),all the stations
(B.4 in appendix B) presented a downward precipitation pattern except the Boromo
(figure 5.5) which showed an increased trend in precipitation.
55
5.1. ARIDITY INDEX AND RAINFALL VARIABILITY OVER THE BLACK VOLTA
Figure 5.5: Boromo rainfall 1970-2012
The trend test for Boromo in the semi-arid zone for the period 1976-2011 showed an
upward evolution but note statistically significant. The tau value 0.113 and the Sens slope
0.1754 showed a positive trend while the big pvalue 0.3403 suggested the non-significance
of this tendency.
Figure 5.6: Bobo-Dioulasso rainfall 1960-2012
Just as the Bobo station, all the other stations presented a decrease in precipitation.
Figure 5.5 shows that considering the years prior to 1976, the decrease was important
while after 1976, there was a sort of decrease but stabilized pattern in precipitation.
56
CHAPTER 5. RESULTS AND DISCUSION
Table 5.1: The Mann-Kendall trend statistics with the Sens Slope estimator
Gauging Station
Period for Rainfall
tau
pvalue
Zs
Variability Analysis
Qs med or
Constant B
Sens Slope estimator
Sunyani
1976-2011
0.156
0.1864
1.321
0.3437
93.12
Wenchi
1976-2011
0.073
0.5399
0.6129
0.1197
101.68
Bui
1954-2005
-0.0724
0.4535
-0.7497
-0.1283
94.11
Bole
1976-2011
0.18
0.1271
1.55
0.3961
82.17
Wa
1976-2011
0.127
0.2819
1.076
0.1749
83.29
Batie
1976-2011
-0.0635
0.5953
-0.531
-0.1559
88.69
Diebougou
1976-2011
0.106
0.3686
0.899
0.1178
83.47
Boura
1976-2011
0.279
0.0171
2.38
0.3218
70.02
Dano
1976-2011
0.143
0.2254
1.21
0.222
72.03
Boromo
1976-2011
0.113
0.3403
0.954
0.1754
68.17
Bondoukuy
1976-2011
-0.1320
0.6728
-0.422
-0.0936
71.85
Bobo-Dioulasso
1976-2011
-0.0635
0.5952
-0.5312
-0.1683
84.75
Bomborokuy
1962-2001
-0.126
0.2584
-1.130
-0.2106
62.73
57
5.2. LULC ANALYSIS IN THE BLACK VOLTA
5.2
5.2.1
LULC analysis in the Black Volta
Accuracy assessments (confusion matrices and statistics
on LULC)
The objective of the accuracy assessment of the land use maps for this project is to
assess how well and accurate the maps are. Details on the methodology used can be seen
under the section 4.3.5. A total number of 70 randomly selected points for each land
use type were used on the original mosaic images during the accuracy assessment. The
confusion matrices for the years 1987, 2000 and 2013 are presented in tables A.1, A.2 and
A.3 while A.4, A.5 and A.6 present the general statistics on the LULC for these years in
the appendix A.
According to (Anderson et al., 1976), the minimum level of interpretation accuracy in
the identification of LULC categories from remote sensing data should be at least 85%
while (Carletta, 1996) states that the Kappa coefficient should be greater than 80%.
Based on the results in table 5.2, the classification used in this study fulfils the minimum
acceptability level as defined above.
Table 5.2: Overall Accuracy and Kappa coefficient the LULC maps
LULC Years
Overall Accuracy (%)
Kappa Coefficient (%)
1987
90.20
88.57
2000
93.06
91.90
2013
99.18
99.05
The results in table 5.2 shows that the Landsat images has been improved since 1987
(recent Landsat gives better results). In fact, the overall accuracy of LULC 1987 (Landsat
5 TM) is 90.2%, 93.06% for LULC 2000 (the Landsat 7 SLC-on) and 99.18% for LULC
2013 (Landsat 8 OLI). Similar results for the Kappa coefficient which varies from 88.57%
(1987) to 99.05% (2013). During the images classification, it was difficult some times
to distinguish between bare land in the extreme north of the catchment with semi-arid
climatic condition from the urban areas in the catchment or agricultural land from grass
58
CHAPTER 5. RESULTS AND DISCUSION
land. In conclusion, we can say that the land use maps are acceptable enough and can
be used for land use changes analyses and projections in the basin.
5.2.2
LULC maps
Figure 5.7: LULC maps of the Black Volta
Figure 5.7 shows LULC maps developed for the years 1987, 2000 and 2013 for the Black
Volta in this project. The land use classes considered in this work are water bodies, bare
land, urban areas, forest evergreen, forest deciduous, grass land and agricultural land.
The definitions of each land use type as considered in this project are specified in the
section 3.3.1.2.3.1. In general, agricultural and bare lands are dominantly in the Burkina
Faso and Mali section of the basin while forest evergreen and deciduous are dominantly
in the Ghana and Ivory Coast sections of the catchment. This agreed with previous work
(Allwaters Consult, 2012) in the basin. Table 5.3 summarizes the characteristics of each
land use type according to the year.
59
5.2. LULC ANALYSIS IN THE BLACK VOLTA
Table 5.3: LULC characteristics in the basin
1987
LULC classes
2000
2013
Area (Km2)
Area (%)
Area (Km2)
Area (%)
Area (Km2)
Area (%)
Bare Land
13231.9
8.5
11279.4
7.3
18842.9
12.15
Urban Areas
9619.8
6.2
16537.3
10.7
22030.3
14.21
Water Bodies
101.7
0.1
872.7
0.6
939.1
0.61
Agricultural Land
36125.7
23.3
36796.5
23.7
47710.4
30.77
Grass Land
76253.9
49.2
74198.4
47.9
41151.4
26.54
Forest Deciduous
14034.6
9.1
14398.3
9.3
23063.6
14.87
Forest Evergreen
5708.9
9.1
967.4
0.6
1338.6
0.86
60
CHAPTER 5. RESULTS AND DISCUSION
5.2.3
LULC 1987
Figure 5.8: LULC occupied area in 1987
The dominant land use types in the Black Volta catchment in the year 1987 (figure
5.8) were grass land (76253.9 Km2 ) and agricultural land (36125.7 Km2 )) representing
respectively 47% and 22% (figure 5.9) of the total area of the basin. Forest evergreen and
forest deciduous represent respectively 9% and 8%. Urban areas in 1987 occupied 6% of
the Black Volta. Water bodies represents the smallest portion of the basin with an area
of 101.7Km2 ) representing only 0.1% of the basin.
61
5.2. LULC ANALYSIS IN THE BLACK VOLTA
Figure 5.9: percentage of each land use type in 1987
5.2.4
LULC 2000
Figure 5.10: LULC occupied area in 2000
In the year 2000, the Black Volta had grass land as dominant land cover type with 74198.48
Km2 ) (figure 5.10 and 5.11) representing 48% of the catchment followed by agricultural
62
CHAPTER 5. RESULTS AND DISCUSION
land (24%). Between 1987 and the year 2000, forest evergreen has decreased by 1% while
water bodies has increased from 0.1% in 1987 to 0.6% in year 2000. Urban areas also
have increased significantly.
Figure 5.11: percentage of each land use type in 2000
63
5.2. LULC ANALYSIS IN THE BLACK VOLTA
5.2.5
LULC 2013
Figure 5.12: LULC occupied area in 2013
The year 2013 was characterized by the dominance of the agricultural land with an area of
47710.41 Km2 )(figure 5.12 and 5.13) representing 31% of the catchment area while grass
land has decreased up to 41151.37 Km2 ) (26%). We notice also a significant increase in
urban areas (14%) and forest deciduous (15%).
5.2.6
Land use trends
Table 17 presents the land use evolution in the Black Volta catchment from 1987 to 2013.
It is based on the land use proportions as developed in the sections above. The objective
is to present the rate at which each land use class changes over time. In fourteen years
period (1987-2000), urban areas have increased by 71.91% that is a growth rate of 5.14%
each year. Subsequently, agricultural lands have increased by 1.86% (0.13%/year). The
changes observed in these land use types can be explained by the population growth
observed in the basin. According to (Green Cross International, 2001), the population
growth rate in the basin is about 3% a year. Deforestation is an important factor
impacting land use mainly in Ghana and Ivory Coast where the forest is mostly located.
64
CHAPTER 5. RESULTS AND DISCUSION
Figure 5.13: percentage of each land use type in 2013
Between 1987 and 2000, forest evergreen has decreased about 83.05% (5.93%/year).
If we consider the period 2000-2013, agricultural land has increased by 29.66%
(2.12%/year) while urban areas have increased by 33.22% corresponding to +2.37%
each year. If we consider the evolution of the land use classes conserved in this study
from 1987 to 2013, the results showed that grass land and deep forest proportions
have decreased at a rate of 1.70% and 2.84% each years. This makes sense when we
observed for the same period (1987-2013), an increase of urban areas (4.78%/year) and
agricultural land (1.19%/year). We also have noticed an important evolution in water
bodies (30.5%/year). This important increased trend might be due to the Bui dam
development. Note that this important development of water bodies represents only
0.61% of the total basin area in the year 2013.
In conclusion, we can say that urbanization and deforestation are taking place very fast
in the Black Volta catchment with possible implications for hydrology.
65
5.3. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
Table 5.4: LULC trends in the catchment
1987-2000
LULC classes
% of Change % Change/year
2000-2013
1987-2013
% of Change
% Change/year
% of Change
% Change/year
Bare Land
-14.76
-1.05
67.06
4.79
42.40
1.57
Urban Areas
71.91
5.14
33.22
2.37
129.01
4.78
Water Bodies
758.10
54.15
7.62
0.54
823.45
30.50
Agricultural Land
1.86
0.13
29.66
2.12
32.07
1.19
Grass Land
-2.70
-0.19
-44.54
-3.18
-46.03
-1.70
Forest Deciduous
2.59
0.19
60.18
4.30
64.33
2.38
Forest Evergreen
-83.05
-5.93
38.38
2.74
-76.55
-2.84
5.3
Hydrology of the Black Volta catchment
In order to assess the implications of rainfall variability, land use and land cover changes
on the stream flow generation in the basin, the Soil and Water Analysis (SWAT) model
was used. It was also important to test the performance of the model in the catchment.
5.3.1
Sensitivity Analysis (to identify most sensitive parameters
in the basin, for calibration of SWAT)
Let us recall that the SWAT has several parameters and it is quasi impossible to calibrate
all of them. The sensitivity analysis was therefore needed to determine the most sensitive
parameters in the Black Volta basin for the calibration process. The sensitivity analysis
was performed in two ways: first by varying one parameter at a time while keeping the
others constant, second by varying all the parameters simultaneously. The definitions
and the significance of each parameter were presented in the table 4.5.
66
CHAPTER 5. RESULTS AND DISCUSION
Table 5.5: Final Parameter range and their sensitivity rank
Final Parameter Range
Parameter Name
Fitted Value
Minimum value
Maximum value
Sensitivity Rank
v GW DELAY
1.087874
0.994085
1.122037
9
r SOL AWC
0.201121
0.145896
0.2093
5
v SURLAG
22.728235
22.493849
22.75284
3
v MSK CO1
10.726233
9.479877
14.147874
4
v MSK CO2
3.117479
2.870465
5.470611
2
r SOL K
-0.999737
-1.009404
-0.966056
1
v CH N2
0.267461
0.242134
0.271966
12
v ALPHA BF
-0.155831
-0.160484
-0.153358
8
a GWQMN
73.4655
73.35611
74.378479
7
v RCHRG DP
0.159342
0.148055
0.176487
10
v ESCO
0.599675
0.595305
0.646715
11
r CN2
-0.537993
-0.552411
-0.505139
6
Although several parameters were tested during the modeling of the Black Volta
catchment, twelve were finally found to be the most sensitive.
Table 5.5 gives the
summary of the most sensitive parameters; their final range gave by the last iteration in
SUFI-2, the fitted values and the sensitivity rank. Sensitivity of parameters is calculated
by a multiple regressive system using the Latin hypercube generated parameters against
the objective function (Abbaspour, 2013). The objective function in this study was the
Nash-Sutcliffe coefficient.
Parameter ranking is based on the t-test and the p-value in the SUFI-2 program. T-test
gives a measure of the sensitivity meaning the larger the t-test in absolute value, the more
sensitive the parameter is while the p-value determined the significance of the sensitivity
(Abbaspour, 2013). P-value closed to zero are more sensitive. Among the twelve final
parameters, the six most sensitive including saturated hydraulic conductivity (SOLK),
time constant for low flow (MSKCO2), time constant for normal flow (MSKCO1), Surface
runoff lag time (SURLAG), Available water capacity of the soil layer (AWC), and the curve
number (CN2). Most of these parameters are soil and land use land cover related, hence
it calls for paying particular attention to soil and land use management practices in the
67
5.3. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
basin.
5.3.2
Calibration and validation
Although the SWAT model was set up based on daily data, the calibration and validation
were based on average monthly data. The model was calibrated and validated at Bui.
Two sets of calibration and validation were performed. The final parameters and their
fitted values are summarized in table 5.5. Although the minimum and the maximum
values of these parameters as presented in table 5.5 are close, these parameters agreed
in general with previous SWAT modeling in the Black Volta basin in particular and in
West Africa in general (Schuol & Abbaspour, 2006; Adjei et al., 2014; Awotwi, Kumi, et
al., 2015) .
The first calibration was performed for the period 2000-2005 and the first validation for
the period 2006-2010 (figure 5.14). The performance of the model for this first calibration
and validation is summarized in table 5.6.
Figure 5.14: Monthly calibration and validation graphs for the Bui station (2000-2010)
The model performance during calibration and validation can be qualified as very good
based on the general performance ratings for recommended statistics for a monthly
time step (table 4.6) given by (Moriasi et al., 2007). The calibration shows that, the
objective function which is the Sutcliffe coefficient (NS) is 0.90 while the goodness of
68
CHAPTER 5. RESULTS AND DISCUSION
fit between the measured and the simulated R2 is also 0.91. The strong correlation
between the measured flow and the simulated (figure 5.15a) showed that the physical
processes implicated in the generation of stream flow in the Black Volta catchment are
well captured by the model during calibration. However, the model overestimated the
simulated average monthly flow (252.70 m3 /s) compared to the measured (223.16 m3 /s)
during calibration while it’s underestimated simulated standard deviation (274.3 m3 /s)
compared to the measured standard deviation (306.38 m3 /s).
Table 5.6: Summary results for SWAT model calibration and validation (2000-2010)
Average Monthly Flow (m3 /s)
Standard Deviation (m3 /s)
Model Performance
Period
Measured
Simulated
Measured
Simulated
p-factor
r-factor
R2
NS
Calibration (2000-2005)
223.16
252.7
306.38
274.3
0.64
0.66
0.91
0.9
Validation (2006-2010)
447.99
371.23
730.94
433.84
0.38
0.33
0.8
0.7
Model performance during validation (figure 5.14b) can be qualified as good. The NS
was 0.70 while R2 was 0.80 (figure 5.15b). Contrary to the calibration, the model during
validation underestimates the average monthly simulated flow (371.23 m3 /s) compared
to the observed flow (447.99 m3 /s). Similar results are observed for standard deviations.
The simulated standard deviation (433.84 m3 /s) was underestimated compared to the
observed one (730.94 m3 /s). A close look at the figure 5.14a revealed that the peak flows
during calibration were correctly modeled while figure 5.14b during validation showed a
delay in peak flows from August 2008 (which corresponds to months 31 on the figure
5.14b). The simulated peak flows are much lower than the observed peaks. One of the
major reasons behind this result is the assumption that the construction of the Bui Dam
disturbed the dynamism of the flows.
69
5.3. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
Figure 5.15: Average simulated monthly discharge vs average observed monthly discharge
for the Bui station during calibration and validation (2000-2010)
According to the Bui Power Authority (BPA), the diversion of the Black Volta
River was completed in December 2008 and the construction of the main dam began
in December 2009.
The dam was commissioned for operations in December 2013
(http://www.water-technology.net/projects/bui-dam-hydro-power-ghana/).
Similar
effects have been reported by (Schuol & Abbaspour, 2006) on the Niger river where
construction of a reservoir is said to delay the river flows. To confirm this hypothesis in
our case, a second set of calibration and validation have been performed prior to the Bui
dam construction.
70
CHAPTER 5. RESULTS AND DISCUSION
Figure 5.16: Monthly calibration and validation graphs for the Bui station ((1990-2000)
The second calibration and validation (figure 5.16a and b) were performed respectively
for the period 1990-1995 and 1996-2000 with the same number of years as it was for the
first calibration and validation (five years for calibration and four years for validation).
The results are summarized in table 5.7. The performance of the model during calibration
can be qualified as very good since the objective function NS = 0.90 and the fit of
goodness between simulation and observation R2 is 0.82 (figure 5.17a). The average
measured flow is 172.28 m3 /s while the simulated was underestimated (161.79 m3 /s). At
the same time, simulated standard deviation (212.24 m3 /s) was lower than the observed
standard deviation (234.95 m3 /s).
Table 5.7: Summary results for SWAT model calibration and validation (1990-2000)
Average Monthly Flow (m3/s)
Standard Deviation (m3/s)
Model Performance
Period
Measured
Simulated
Measured
Simulated
p-factor
r-factor
R2
NS
Calibration (1990-1995)
172.28
161.79
234.95
212.24
0.97
0.83
0.82
0.9
Validation (1996-2000)
229.18
254.37
338.39
394.68
0.64
0.9
0.85
0.7
For the validation performance of the model,the NS was 0.7 and the R2 =0.85 (figure
34b) and can be classified as good. The mean monthly simulated flow was overestimated
71
5.3. HYDROLOGY OF THE BLACK VOLTA CATCHMENT
(254.37 m3 /s) compared to the observed (229.18 m3 /s). There is also difference in the
standard deviation between simulated and observed flow. However, the difference is
smaller compared to the first validation (figure 5.14b). This can be a confirmation that
the construction in the Bui dam impact the dynamics of the river flow. In general, we
can say that our model has performed well despite all the uncertainties related to data
quality or the model itself and the parameter used during the calibration processes.
Figure 5.17: Average simulated monthly discharge vs average observed monthly discharge
for the Bui station during calibration and validation (1990-2000)
5.3.3
Model uncertainties
The SWAT model can be uncertain in its prediction due to the fact that all inputs data
(example of rainfall and temperature) as well as the processes involved in the model have
been measured with some error. The degree to which all uncertainties are accounted
for is quantified by a measure referred to as the P-factor, which is the percentage of
measured data bracketed by the 95% prediction uncertainty (95PPU). In oder words,
the 95PPU is the degree to which all uncertainties are accounted for (Abbaspour, 2013).
SUFI-2 account for all uncertainties including uncertainties in driving variable like
rainfall, conceptual model, parameters and measured data (stream flow). P-factor is the
percent of observations that are within the given uncertainty boundaries. The perfect
situation would be 100% of the observed data bracketed in the 95PPU while at the same
72
CHAPTER 5. RESULTS AND DISCUSION
time the R-factor is close to zero (Abbaspour, 2013). SUFI-2 account for all uncertainties
including uncertainties in driving variable like rainfall, conceptual model, parameters
and measured data.
During the first calibration (2000-2005), 64% of the observed flow were bracketed in the
95PPU (see figureB.9 in appendix B) while the R-factor was 0.66 suggesting important
uncertainties in the model. For the validation period (2006-2010), only 38% of the
observed flow were barracked in the 95PPU and 0.33 as R-factor (table 5.6).
The uncertainties in the model during the second calibration and validation are reported
in table 5.7. During calibration (1990-1995) about 97% of the observed discharge were
bracketed in the 95PPU suggesting a very good results or minimal uncertainties in the
model. However, the R-factor was relatively high (0.83) meaning that the uncertainties
cannot be neglected. The validation performance showed that uncertainties in the model
are important (p-factor =64% and R-factor = 0.90). This may explain why the peak
flows were not well captured during validation. Unfortunately, one can not identified the
source of uncertainties in the model.
5.4
LULC impacts on stream flow in the Black Volta
catchments
In order to quantify the influence of land use land cover changes on stream flow in
the catchments, the land use maps of the years 2000 and 2013 were used in two
different simulations using the SWAT model. The length of each simulation was 36
years (1976-2011). The other parameters (input data and settings) in the model were
unchanged during these simulations. Only land use data have been changed. Two sets
of analysis were performed on the results on each output of the simulations. The first
analysis was done looking at intra-annual or seasonal variability of stream flow due to
land use change considering dry (February, March, April) and wet (August, September,
October) months. Dry and wet months were defined based on the flows measured at
Bui (see figure 4.6). The second analysis was performed based on annual average for the
73
5.4. LULC IMPACTS ON STREAM FLOW IN THE BLACK VOLTA
CATCHMENTS
entire period of each components of the stream flow such us surface run-off (SURF-Q),
lateral flow (LAT Q) and ground water contribution to stream flow (GW Q) and also
on evapotranspiration (ET). The combination of all these components represents the
water yield to stream flow. The values used in this work were converted from mm to m3 /s.
5.4.1
Changes in seasonal stream flows due to LULC
The results of the intra-annual or seasonal variability analysis are shown in table 5.8.
The total flow of three months such as January, February and March were considered as
dry period over the year for both LULC 2000 and 2013. The modeling of the dry period
flow using the LULC of the year 2000 was 238.62 m3 /s while the dry period flow using
LULC of the year 2013 was 253.42 m3 /s. The results showed that there is an increase of
6% in dry period flows due to LU practices in the Black Volta basin.
Comparatively, wet period was defined considering the total flow of three months such
as August, September and October. The stream flow changes were also assessed due
to land use changes in the catchments. The total flow for the wet period according to
LULC 2000 was 2969.02 m3 /s and 2995.35 m3 /s for 2013 LULC. The results indicated
that there was an increase by 1% in wet period stream flow. This stream flow assessment
showed that it is important to monitor the land use practices in the basin since it has
huge impact on river flow and by extension on the hydro-power generation at the Bui dam.
Table 5.8: Intra-annual flow changes due to LULC
Mean Monthly Flow 1976-2011 (m3/s)
LULC 2000
LULC 2013
Mean Monthly Flow Change
Dry Months
Wet Months
Dry Months
Wet Months
(Jan, Feb, Mar)
(Aug, Sep, Oct)
(Jan, Feb, Mar)
(Aug, Sep, Oct)
Dry
Wet
238.62
2969.02
253.42
2995.35
6%
1%
74
CHAPTER 5. RESULTS AND DISCUSION
5.4.2
Changes in stream flow components due to LULC
Primary investigation on the changes in the stream flow due to LULC showed strong
impacts on stream flow.
Further analysis focused on the stream flow components
variability due to land use.
The stream flow components including surface runoff
(SURF Q), lateral flow (LAT Q), ground water contribution to stream flow (GW Q)
contribute to water yield. The effects of LULC on these elements are depicted in table
5.9.
The results showed that with LULC 2000, the SURF Q, LAT Q and GW Q were
respectively 376.7 m3 /s, 5.7 m3 /s and 828 m3 /s while those with the LULC 2013 are
respectively 477.3 m3 /s, 6.8 m3 /s and 775.1 m3 /s. We have noticed that surface runoff
and ground water are the big contributors to stream flow in the Black Volta catchment.
We noticed that the components of stream flow are not affected in the same manner. For
instanced, between 2000 and 2013, the
Table 5.9: Stream flow Components and ET Changes due to LULC
Stream Flow Components (m3 /s)
LULC2000
LULC2013
Changes (%)
376.7
477.3
27%
LAT Q
5.7
6.8
19%
GW Q
828
775.1
-6%
WATER YIELD
1210.4
1259.2
4%
ET
521.6
546.7
4.59%
and ET (mm)
SURF Q
SURF Q has increased by 27% while LAT Q has increased by 19%. In contrary to
75
5.5. COMBINED POTENTIAL IMPLICATIONS OF RAINFALL VARIABILITY
AND LULC FOR THE BUI HYDROPWER PLANT
the two other components, for the same period, GW Q has decreased by -6%. The
evapotranspiration (ET) has increased at the same time by 4.59%. These results can
be explained by the fact that between the year 2000 and 2013, the urbanization rate
and bare lands have increased respectively by 33.22% and 67.06%. At the same time,
agricultural land has increased by 29.66% as an implications for the reduction in grass
land by 44.54%. These results call for paying particular attention to ground water in the
basin. The change in land use in terms of increasing urbanization and and bare land have
resulted in increasing surface runoff and reduced groundwater. The reduction in ground
water contribution to stream flow may be due to the increase in ET when grass lands have
been converted in crop lands. This may be due to difference in land use that controls ET.
5.5
Combined
potential
implications
of
rainfall
variability and LULC for the Bui hydropwer
plant
It would have been ideally interesting if we have had the power production data from the
Bui power station since the plant was operating. Unfortunately, all our attempts to get
these data have been unsuccessful. However based on the previous analysis, one can draw
some conclusions in terms of possible implications of the combined potential implications
of both rainfall and LULC on the power production at Bui.
The rainfall variability analysis through the Mann-Kendall monotonic trend detection has
shown that about 61.4% of the stations in the basin (that is 8 stations out of a total of 13
stations) have presented an increase in rainfall and 38.6% of the stations have decreasing
rainfall pattern over more than 30 years of data that were analysed. However, the results
were not statistically significant at 95% confidence interval level. This result may induce
an overall increase in stream flow. In fact, the combination of both precipitation and land
use in the SWAT model has shown an increase in the water yield to discharge at Bui by
4% between the years 2000 and 2013. One should notice that, in thirteen years, the basin
has undergone important land use changes.
76
Chapter 6
CONCLUSIONS AND
RECOMMENDATIONS
6.1
CONCLUSIONS
The primary objective of the present study is to evaluate the possible implications of
rainfall variability along with land use land cover changes (LULC) in the Black Volta
basin on the Bui hydropower plant. To do so, several approaches have been used.
First of all, it has been found useful to develop an aridity index profile of the Black
Volta catchment because precipitation availability over atmospheric water demand can
be quantified using the aridity index (AI). Using the AI dataset from the Global Aridity
and PET Database, it has been found that the Black Volta basin is a complex watershed
that is divided in three climatic zones according to the UNEP aridity index classification
scheme: Semi-arid zone with AI varying between 0.2 and 0.46, dry sub-humid with
AI between 0.46 and 0.58 and humid climate zone with AI varying between 0.58 and
0.84. The identification of these different climatic zones helped to better understand the
spatial variability of rainfall in the basin.
In order to assess the spatio-temporal variability of rainfall over the Black Volta, the
Mann-Kendall monotonic trend test and the Sen’s slope estimator were applied to
thirteen station data in the basin on the annual average basis for more than thirty years
rainfall data for each station. The statistics of the trend test showed that 61.4% of the
rain gauges (8 stations over a total of 13) presented an increased precipitation trend
77
6.1. CONCLUSIONS
whereas the rest of the stations showed a decreased trend. However, the test that was
performed at 95% confidence interval level showed that the detected trends in the rainfall
data were not statistically significant. Only rainfall data at Boura in dry sub-humid zone
revealed a statistically significant increased trend (pvalue less than 0.05). One can also
noticed that most of the stations that have shown an increased precipitation trend are
located in the humid and dry sub-humid zones of the basin. This aspect of the rainfall
variability in the basin call for paying a particular attention when choosing a site for
water resource infrastructure.
Land use land cover change is an important factor controlling the hydrology of a basin.
In the recent years, remotely sensed data have been useful and are widely used to assess
land use changes in a watershed. It is a very important tool to assess land use trend
in a complex, trans-boundary and wide watershed like the Black Volta basin. Three
Landsat satellite images were collected including Landsat 5 Thematic Mapper for the
year 1987, Landsat 7 SLC-on for the 2000 and Landsat 8 OLI for the year 2013. The
land use trends in the basin were analysed based on these data. One emphasized more
on the recent years land use in the analysis meaning between the year 2000 and 2013.
The results show that within thirteen years, land use classes like bare land, urban areas,
water bodies, agricultural lands, forest deciduous and forest evergreen have increased
respectively by 67.06%, 33.22%, 7.62%, 29.66%, 60.18%, and 38.38%. Only grass land
has decreased by 44.54% within this period. The land use trend may be explained by
population growth and infrastructure development in the basin. It has been shown that
the population growth rate in the Black Volta basin is about 3%/year. Increasing water
bodies may be explain mainly by the Bui reservoir construction. The decrease in grass
land has resulted in agricultural land extension.
To evaluate the combined effect of rainfall variability and land use change on the
discharge at Bui, the hydrological model, SWAT has been selected. A sensitivity analysis
was first performed in order to come up with representative parameters of the Black
Volta basin. The results showed that parameters related to soil and land use are those
controlling the hydrological processes of the basin. The SWAT model was then tested
for its performance in the basin for this particular project using the Nash-Sutcliffe
coefficient (N S) and and the coefficient of determination (R2 ). A first calibration was
performed for the period 2000-2005 and validation for the period 2006-2010. During
78
CHAPTER 6. CONCLUSIONS AND RECOMMENDATIONS
calibration, the model performance was qualified as very good with N S = 0.9 and R2
= 0.91. The strong correlation between measured and simulated flows showed that the
physical processes implicated in the generation of the stream flow in the basin are well
captured by the model. For the validation period, the model performance was good with
N S = 0.7 and R2 = 0.8. However, the graphical representation of the observed and the
simulated flows during validation revealed that there was a delay in peak flows starting
from 2008. We assumed this effect may be due to the Bui reservoir construction. To
confirm the hypothesis, the model performance was tested for a period prior to the Bui
dam construction.This second calibration was performed for the period 1990-1995 and
validated for the period 1996-2000. The calibration was qualified as very good (N S =
0.9 and R2 = 0.82) and validation as good (N S = 0.7 and R2 = 0.85). The graphical
representation of the observed and the simulated flows during calibration and validation
do not show important delay in the peak flows proving that the construction of the Bui
data may effectively affect the dynamic of the river system.
After the evaluation of the model performance, land use land cover changes impacts on
stream flow was then assessed. The land use maps of the years 2000 and 2013 were used
in in two different simulations using the SWAT model. The length of each simulation was
36 years (1976-2011). The other parameters (input data and settings) in the model were
unchanged during these simulations. Only land use data have been changed. Two sets of
analysis were performed on the results on each output of the simulations.
First, changes in seasonal stream flow due to LULC was assessed by defining dry season
(February, March and April) and wet season (August, September and October). The
results showed from year 2000 to year 2013, the dry season discharge has increased by
6% whereas the discharge of wet season has increased by 1%. This land use effects on
seasonal variation of stream flow may benefit the Bui hydropower plant.
The second analysis focused on the changes in stream flows components such us surface
run-off (SURF Q), lateral flow (LAT Q) and ground water contribution to stream flow
(GW Q) and also on evapotranspiration (ET) changes due to LULC. The results showed
that between the year 2000 and 2013, SURF Q and LAT Q have respectively increased
by 27% and 19% while GW Q has decreased by 6%. At the same time, ET has increased
by 4.59%. The resultant effects is that the water yield to stream flow has increased by
79
6.2. RECOMMENDATIONS
4%. These results call for paying particular attention to ground water in the basin. The
change in land use in term of increasing urbanization and and bare land have resulted
in increasing surface runoff and reduce groundwater. The reduction in ground water
contribution to stream flow may be due to the increase in ET when grass lands have
been converted in crop lands. This may due to difference in land use that control ET.
We believe that the overall impacts of rainfall variability and LULC may benefit the Bui
hydropower plant.
In general, one noticed that rainfall pattern is changing in the basin although it not
statistically significant.
The land use in the basin is changing fast with increasing
urbanisation, agricultural land but also extension of bare land particularly in the semi
arid zone. The SWAT model has then been found as a useful tool for combining rainfall
variability and LULC in the analysis of stream flow. This stream is somehow increasing
due to land use but also changing rainfall pattern.
6.2
RECOMMENDATIONS
At the end of this particular project, we find that it will be useful to include the following
recommendations for better water management in the basin:
• Due to the fact that surface runoff has increased because of LULC, important
erosion may take place in the basin. Therefore , it will be important to use the
SWAT model or another useful software to assess the erosion rate in the basin and
establish the erosion vulnerability map of the basin.
• We recommand the assessment of the sedimentation rate of the Bui reservoir in the
near future in order to avoid reservoir volume loss looking at the increase in surface
runoff.
• The increase in surface runoff may also induce the degradation of surface water
quality. We then recommend that surface water quality be regularly controlled.
• To improve the present work, we suggest that in future for similar work in the basin,
one should include the Bui reservoir in the modelling process.
80
CHAPTER 6. CONCLUSIONS AND RECOMMENDATIONS
• For better water resource planning in the basin, the present work can also be done
looking at future climate change scenarios. We aslo recommend that rainfall trend
analysis be done at monthly basis and more detailed work be done on land use
classes in the basin in order to improve the model performance.
81
Appendix A
Additional tables
Table A.1: Confusion matrix LULC 1987
LULC Classes
C Bare Land
C Water Bodies
C Forest Evergreen
C Urban Areas
C Agricultural Land
C Forest Deciduous
C Grass Land
Bare Land
43
0
0
0
0
0
0
Water Bodies
0
68
0
0
0
0
0
Forest Evergreen
0
0
69
0
0
0
0
Urban Areas
0
0
0
64
0
0
1
Agricultural Land
0
2
0
0
62
0
2
Forest Deciduous
0
0
0
0
0
69
0
Grass Land
27
0
1
6
8
1
67
Table A.2: Confusion matrix LULC 2000
LULC Classes
C Bare Land
C Forest Deciduous
C Grass Land
C Urban Areas
C Water Bodies
C Agricultural Land
C Forest Evergreen
Bare Land
45
0
0
0
0
0
0
Forest Deciduous
0
69
0
0
0
0
0
Grass Land
0
0
70
2
1
0
0
Urban Areas
25
1
0
66
0
0
0
Water Bodies
0
0
0
0
67
0
0
Agricultural Land
0
0
0
2
0
70
1
Forest Evergreen
0
0
0
0
2
0
69
Table A.3: Confusion matrix LULC 2013
LULC Classes
C Urban Areas C Water Bodies
C Forest Deciduous
C Forest Evergreen
C Agricultural Land
C Grass Land
C Bare Land
Urban Areas
70
0
0
0
0
0
2
Water Bodies
0
69
0
0
0
0
0
Forest Deciduous
0
0
70
1
0
0
0
Forest Evergreen
0
0
0
69
0
0
0
Agricultural Land
0
1
0
0
70
0
0
Grass Land
0
0
0
0
0
70
0
Bare Land
0
0
0
0
0
0
68
82
APPENDIX A. ADDITIONAL TABLES
Table A.4: General Statistics on LULC 1987
1987
LULC Classes
Comision (%)
Omission (%)
Producers Accuracy (%)
User Accuracy (%)
Bare Land
0
38.57
61.43
100
Water Bodies
0
2.86
97.14
100
Forest Evergreen
0
1.43
98.57
100
Urban Areas
1.54
8.57
91.43
98.46
Agricultural Land
6.06
11.43
88.57
93.94
Forest Deciduous
0
1.43
98.57
100
39.09
4.29
95.71
60.91
Grass Land
Table A.5: General Statistics on LULC 2000
2000
LULC Classes
Comision (%)
Omission (%)
Producers Accuracy (%)
User Accuracy (%)
Bare Land
0
35.71
64.29
100
Forest Deciduous
0
1.43
92.86
94.2
Grass Land
4.11
0
100
95.89
Urban Areas
28.26
5.71
94.29
71.74
Water Bodies
0
4.29
95.71
100
Agricultural Land
4.11
0
100
95.89
Forest Evergreen
2.82
1.43
98.57
97.18
Table A.6: General Statistics on LULC 2013
2013
LULC Classes
Comision (%)
Omission (%)
Producers Accuracy (%)
User Accuracy (%)
0
2.9
97.1
100
Forest Deciduous
1.4
0
100
98.6
Grass Land
4.11
0
100
95.89
Urban Areas
2.8
0.0
100.0
97.2
Water Bodies
0
1.4
98.6
100
Agricultural Land
1.4
0
100
98.6
Forest Evergreen
0
1.4
98.6
100
Bare Land
83
Appendix B
Additional graphs
Figure B.1: Batie rainfall 1960-2013
B.4B
84
APPENDIX B. ADDITIONAL GRAPHS
Figure B.2: Bomborokuy 1962-2001
Figure B.3: Dano rainfall 1970-2013
85
Figure B.4: Bondoukuy rainfall 1970-2013
Figure B.5: Diebougou rain fall 1970-2013
86
APPENDIX B. ADDITIONAL GRAPHS
Figure B.6: Bole rainfall 1976-2011
Figure B.7: Sunyani rainfall 1976-2011
87
Figure B.8: Wenchi rainfall 1976-2011
Figure B.9: Flow in the first calibration with the 95PPU
88
References
Abbaspour, K. C. (2013). Swat-cup 2012: Swat calibration and uncertainty programs-a
user manual. Eawag: Dübendorf, Switzerland , 103 .
Abbaspour, K. C., Johnson, C. A., & Van Genuchten, M. T. (2004). Estimating uncertain
flow and transport parameters using a sequential uncertainty fitting procedure. Vadose
Zone Journal , 3 (4), 1340–1352.
Abbaspour, K. C., Vejdani, M., & Haghighat, S. (2007). Swat-cup calibration and
uncertainty programs for swat. In Modsim 2007 international congress on modelling
and simulation, modelling and simulation society of australia and new zealand.
Adjei, K. A., Ren, L., Appiah-Adjei, E., & Odai, S. N.
(2014).
Application of
satellite-derived rainfall for hydrological modelling in the data-scarce black volta
trans-boundary basin. IWA Publishing.
Ahmad, I., Tang, D., Wang, T., Wang, M., & Wagan, B. (2015). Precipitation trends
over time using mann-kendall and spearmans rho tests in swat river basin, pakistan.
Advances in Meteorology, 2015 .
Allen,
R. G.,
Pereira,
L. S.,
Raes,
D., & Smith,
M.
(1998).
Crop
evapotranspiration-guidelines for computing crop water requirements-fao irrigation and
drainage paper 56. FAO, Rome, 300 (9), D05109.
Allwaters Consult. (2012). DIAGNOSTIC STUDY OF THE BLACK VOLTA BASIN IN
GHANA Final Report.
Anderson, J. R., Hardy, E. E., Roach, J. T., & Witmer, R. E. (1976). A Land Use
and Land Cover Classification System for Use with Remote Sensor Data. In Usgs
professional paper 964 (pp. 138–145).
89
REFERENCES
Andreini, M., van de Giesen, N., van Edig, A., Fosu, M., & Andah, W. (2000). Volta
basin water balance. zef-discussion papers on development policy, no. 21. Center for
Development Research, Bonn, Germany.
Arndt, C., Asante, F., & Thurlow, J. (2014). Implications of climate change for ghana’s
economy (Tech. Rep.). WIDER Working Paper. Retrieved from http://hdl.handle
.net/10419/96301
Arnold, J., Kiniry, J., Srinivasan, R., Williams, J., Haney, E., & Neitsch, S. (2012). Soil
and water assessment tool input/output file documentation: Version 2012. Texas Water
Resources Institute TR vol. 436 .
Arnold, J. G., Moriasi, D. N., Gassman, P. W., Abbaspour, K. C., White, M. J.,
Srinivasan, R., . . . others (2012). Swat: Model use, calibration, and validation.
Transactions of the ASABE , 55 (4), 1491–1508.
Arnold, J. G., Srinivasan, R., Muttiah, R. S., & Williams, J. R.
(1998).
Large
Area Hydrologic Modeling and Assessment Part I: Model Development. Journal of
the American Water Resources Association, 34 (1), 73–89. Retrieved from http://
doi.wiley.com/10.1111/j.1752-1688.1998.tb05961.x
doi: 10.1111/j.1752-1688
.1998.tb05961.x
Awotwi, A., Kumi, M., Jansson, P., Yeboah, F., & Nti, I. (2015). Predicting hydrological
response to climate change in the white volta catchment, west africa. J Earth Sci Clim
Change, 6 (249), 2.
Awotwi, A., Yeboah, F., & Kumi, M. (2015). Assessing the impact of land cover changes on
water balance components of white volta basin in west africa. Water and Environment
Journal , 29 (2), 259–267.
Barker, T., Davidson, O., Davidson, W., Huq, S., Karoly, D., Kattsov, V., . . . others
(2007). Climate change 2007: Synthesis report. Valencia; IPPC .
Barry, B., Obuobie, E., Andreini, M., Andah, W., & Pluquet, M. (2005). The volta river
basin. Comparative study of river basin development and management. Rapport, IWMI,
CAWMA.
90
REFERENCES
Bartle, A. (2002, November). Hydropower potential and development activities. Energy
Policy, 30 (14), 1231–1239.
Retrieved from http://linkinghub.elsevier.com/
retrieve/pii/S0301421502000848 doi: 10.1016/S0301-4215(02)00084-8
Bayazit, M., Önöz, B., Yue, S., & Wang, C. (2004). Comment on ”Applicability of
prewhitening to eliminate the influence of serial correlation on the Mann-Kendall test”
by Sheng Yue and Chun Yuan Wang. Water Resources Research, 40 (8), 1–5. doi:
10.1029/2002WR001925
Beck, M. W., Claassen, A. H., & Hundt, P. J. (2012). Environmental and livelihood
impacts of dams:
common lessons across development gradients that challenge
sustainability. International journal of river basin management, 10 (1), 73–92.
Beven, K. (1997). Topmodel: a critique. Hydrological processes, 11 (9), 1069–1085.
Boko, M., Niang, I., Nyong, A., Vogel, C., Githeko, A., Medany, M., . . . Yanda, P. (2007).
Africa. Cambridge University Press.
Bunyasi, M. M. (2012). Vulnerability of hydro-electric energy resources in kenya due to
climate change effects: The case of the seven forks project. Journal of Agricultural and
Environmental Sciences, 1 , 36–49.
Cannarozzo, M., Noto, L., & Viola, F. (2006). Spatial distribution of rainfall trends
in sicily (1921–2000). Physics and Chemistry of the Earth, Parts A/B/C , 31 (18),
1201–1211.
Carletta, J. (1996). Assessing agreement on classification tasks: the kappa statistic.
Computational linguistics, 22 (2), 249–254.
Conway, D., Persechino, A., Ardoin-Bardin, S., Hamandawana, H., Dieulin, C., & Mahé,
G. (2009, February). Rainfall and Water Resources Variability in Sub-Saharan Africa
during the Twentieth Century. Journal of Hydrometeorology, 10 (1), 41–59. Retrieved
from http://journals.ametsoc.org/doi/abs/10.1175/2008JHM1004.1
doi: 10
.1175/2008JHM1004.1
Cunderlik, J. (2003). Hydrologic model selection for the cfcas project: Assessment of
water resources risk and vulnerability to changing climatic conditions. Department of
Civil and Environmental Engineering, The University of Western Ontario.
91
REFERENCES
Easton, Z., Fuka, D., White, E., Collick, A., Biruk Ashagre, B., McCartney, M., . . .
Steenhuis, T. (2010). A multi basin swat model analysis of runoff and sedimentation
in the blue nile, ethiopia. Hydrology and Earth System Sciences, 14 (10), 1827–1841.
ESIA. (2007). Environmental and Social Impact Assessment of the Bui Hydropower
Project.
FAO. (2003). (Food and Agriculture Organization of the United Nations). Digital Soil Map
of the World (DSMW): based on the FAO/UNESCO SoilMap of theWorld, Original
Scale 1:5 000 000. FAO-AGL (Land and Water Development Division), Rome.
Foody, G. M. (2002). Status of land cover classification accuracy assessment. Remote
Sensing of Environment, 80 (1), 185–201. doi: 10.1016/S0034-4257(01)00295-4
Fortin, J.-P., Turcotte, R., Massicotte, S., Moussa, R., Fitzback, J., & Villeneuve, J.-P.
(2001). Distributed watershed model compatible with remote sensing and gis data. i:
Description of model. Journal of Hydrologic Engineering, 6 (2), 91–99.
Githui, F., Gitau, W., Mutua, F., & Bauwens, W. (2009). Climate change impact on swat
simulated streamflow in western kenya. International Journal of Climatology, 29 (12),
1823–1834.
Gocic, M., & Trajkovic, S. (2013). Analysis of changes in meteorological variables
using Mann-Kendall and Sen ’ s slope estimator statistical tests in Serbia. Global
and Planetary Change, 100 , 172–182. Retrieved from http://dx.doi.org/10.1016/
j.gloplacha.2012.10.014 doi: 10.1016/j.gloplacha.2012.10.014
Goulden, M., Conway, D., & Persechino, A. (2009). Adaptation to climate change in
international river basins in africa: a review/adaptation au changement climatique
dans les bassins fluviaux internationaux en afrique: une revue. Hydrological Sciences
Journal , 54 (5), 805–828.
Green, W. H., & Ampt, G. (1911). Studies on soil physics, 1. the flow of air and water
through soils. J. Agric. Sci , 4 (1), 1–24.
Green Cross International. (2001). Burkina Faso.Trans-boundary Basin Sub-Projects:
The Volta River Basin. Retrieved from www.gci.ch/GreenCrossPrograms/waterres/
pdf/WFP Volta
92
REFERENCES
Harrison, G.
(1998).
Climate change impacts on hydroelectric power.
UNIV POWER ENG . . . (1).
PROC
Retrieved from http://www.energybc.ca/cache/
climatechange/www.see.ed.ac.uk/ gph/publications/GPH-Upec98.pdf
Hensengerth, O. (2011). Interaction of chinese institutions with host governments in dam
construction: the bui dam in ghana. DIE-Deutsches Institut für Entwicklungspolitik .
Hirsch, R. M., Alexander, R. B., & Smith, R. a. (1991). Selection of methods for the
detection and estimation of trends in water quality. Water Resources Research, 27 (5),
803–813. doi: 10.1029/91WR00259
Hollander, M., Wolfe, D. A., & Chicken, E. (2013). Nonparametric statistical methods.
Jakeman, A., Littlewood, I., & Whitehead, P. (1990). Computation of the instantaneous
unit hydrograph and identifiable component flows with application to two small upland
catchments. Journal of hydrology, 117 (1), 275–300.
Jassby, A. D., & Cloern, J. E. (2015). wq: Exploring water quality monitoring data.
Kabo-bah, A. T., Anornu, G., Ofosu, E., Andoh, R., & Lis, K. J. (2014). Spatial-temporal
estimation of evapotranspiration over black volta of west africa. International Journal
of Water Resources and Environmental Engineering, 6 (12), 295–302.
Kalitsi, E.
(2003).
PROBLEMS AND PROSPECTS FOR HYDROPOWER
DEVELOPMENT Prepared by The Workshop for African Energy Experts on
Operationalizing the NGPAD Energy Initiative Problems and Prospects for
Hydropower Development in Africa. (June).
Kasei, R.
in
the
(2009).
volta
basin,
Modelling impacts of climate change on water resources
west
africa,
number
34.
PhD thesis,
Erlangung
des Doktorgrades (Dr. rer. nat), Mathematisch-Naturwissenschaftlichen Fakultät,
Rheinischen Friedrich-Wilhelms-Universität Bonn in Germany.
Kaunda, C. S., Kimambo, C. Z., & Nielsen, T. K. (2012). Hydropower in the Context of
Sustainable Energy Supply: A Review of Technologies and Challenges. ISRN Renewable
Energy, 2012 , 1–15. Retrieved from http://www.hindawi.com/journals/isrn/2012/
730631/ doi: 10.5402/2012/730631
93
REFERENCES
Kendall. . . , M. (1962). Rank correlation methods.
Lacombe, G., McCartney, M., & Forkuor, G. (2012, November). Drying climate in Ghana
over the period 19602005: evidence from the resampling-based Mann-Kendall test at
local and regional levels. Hydrological Sciences Journal , 57 (8), 1594–1609. Retrieved
from http://www.tandfonline.com/doi/abs/10.1080/02626667.2012.728291 doi:
10.1080/02626667.2012.728291
Lanzante, J. R. (1996). Resistant, Robust and Non-Parametric Techniques for the
Analysis of Climate Data: Theory and Examples, Including Applications To Historical
Radiosonde Station Data. International Journal of Climatology, 16 (11), 1197–1226.
doi: 10.1002/(SICI)1097-0088(199611)16:11h1197::AID-JOC89i3.0.CO;2-L
Laube, W., Leemhuis, C., & Amisigo, B. (2008). GLOWA Volta Policy Brief : Impact of
Climate Change on the Black Volta Basin and the Bui Dam. , 1–3.
Liersch, S. (2003). The programs dew. exe and dew02, exe user’s manual.
Liu, Q., Yang, Z., & Cui, B.
(2008).
Spatial and temporal variability of annual
precipitation during 1961–2006 in yellow river basin, china. Journal of Hydrology,
361 (3), 330–338.
Logah, F., Obuobie, E., Ofori, D., & Kankam-Yeboah, K. (2013). Analysis of rainfall
variability in ghana. International Journal of Latest Research in Engineering and
Computing, 1 , 1–8.
Lundin, L.-c., Bergström, S., Eriksson, E., & Seibert, J. (1998). Hydrological models and
modelling. , 129–140.
Mahe, G., L’Hote, Y., Olivry, J. C., & Wotling, G.
(2001, April).
Trends and
discontinuities in regional rainfall of West and Central Africa: 19511989. Hydrological
Sciences Journal , 46 (2), 211–226. Retrieved from http://www.tandfonline.com/
doi/abs/10.1080/02626660109492817 doi: 10.1080/02626660109492817
Malunjkar, V. S., Shinde, M. G., Ghotekar, S. S., & Atre, A. A. (2015). Estimation of
Surface Runoff using SWAT Model. (4), 12–15.
94
REFERENCES
Mann, H. B. (1945). Nonparametric tests against trend. Econometrica: Journal of the
Econometric Society, 245–259.
Martin, N. (2006). Ecology and Development Series No . 41 , 2006. (41).
McCartney, M. (2007). Decision support systems for large dam planning and operation
in africa. international water management institute. 47 p. IWMI Working Paper 119),
Sri Lanka, Colombo.
McCartney, M., Forkuor, G., Sood, A., Amisigo, B., Hattermann, F., & Muthuwatta, L.
(2012). The water resource implications of changing climate in the volta river basin.
IWMI , 146 .
McLeod, A., & McLeod, M. A. (2011). Package kendall.
McLeod, A. I. (2005). Kendall rank correlation and mann-kendall trend test. R Package
Kendall .
McSweeney, C., Lizcano, G., New, M., & Lu, X. (2010). The undp climate change country
profiles: Improving the accessibility of observed and projected climate information
for studies of climate change in developing countries.
Bulletin of the American
Meteorological Society, 91 (2), 157–166.
Moriasi, D., Arnold, J., Van Liew, M., Bingner, R., Harmel, R., & Veith, T. (2007). Model
evaluation guidelines for systematic quantification of accuracy in watershed simulations.
Trans. Asabe, 50 (3), 885–900.
Mukheibir, P. (2007). Possible climate change impacts on large hydroelectricity schemes
in southern africa. Journal of Energy in Southern Africa, 18 (1), 4–9.
Nash, J., & Sutcliffe, J. V. (1970). River flow forecasting through conceptual models part
i discussion of principles. Journal of hydrology, 10 (3), 282–290.
Neitsch, S., Arnold, J., Kiniry, J., Srinivasan, R., & Williams, J. (2005). Soil and water
assessment tool input/output file documentation. Blackland Research Center, Temple,
Texas.
95
REFERENCES
Neitsch, S. L., Arnold, J. G., Kiniry, J. R., & Williams, J. R. (2011). Soil and water
assessment tool theoretical documentation version 2009 (Tech. Rep.). Texas Water
Resources Institute.
NRCS. (1986). Urban Hydrology for Small Watersheds TR-55. USDA Natural Resource
Conservation Service Conservation Engeneering Division Technical Release 55 , 164.
doi: TechnicalRelease55
Obuobie, E., et al. (2008). Estimation of groundwater recharge in the context of future
climate change in the white volta river basin, west africa. ZEF .
Oguntunde, P. G., Friesen, J., van de Giesen, N., & Savenije, H. H.
(2006,
January). Hydroclimatology of the Volta River Basin in West Africa: Trends and
variability from 1901 to 2002. Physics and Chemistry of the Earth, Parts A/B/C ,
31 (18), 1180–1188. Retrieved from http://linkinghub.elsevier.com/retrieve/
pii/S147470650600297X doi: 10.1016/j.pce.2006.02.062
Owusu, K., & Waylen, P. (2009, May). Trends in spatio-temporal variability in annual
rainfall in Ghana (1951-2000). Weather , 64 (5), 115–120. Retrieved from http://
doi.wiley.com/10.1002/wea.255 doi: 10.1002/wea.255
Pachauri, R. K., Allen, M., Barros, V., Broome, J., Cramer, W., Christ, R., . . . others
(2014). Climate change 2014: Synthesis report. contribution of working groups i, ii
and iii to the fifth assessment report of the intergovernmental panel on climate change.
IPCC .
Peterson, T. C., Easterling, D. R., Karl, T. R., Groisman, P., Nicholls, N., Plummer, N.,
. . . Parker, D. (1998). Homogeneity adjustments of in situ atmospheric climate data:
A review. International Journal of Climatology, 18 (13), 1493–1517. doi: 10.1002/
(SICI)1097-0088(19981115)18:13h1493::AID-JOC329i3.0.CO;2-T
Raschid, L.-s., Koranteng, R. T., Kyei, E., & Danso, A. (2007). Capacity building for the
sustainability of dam development with special reference to the bui dam project.
Raschid Sally, L., Akoto-Danso, E. K., Twum-Korangteng, R., et al. (2008). Sustaining
the multi-stakeholder dialogue process for dams and development: final report of phase
96
REFERENCES
2 of the ghana dams forum, april 2007-march 2008. International Water Management
Institute.
Reeves, J., Chen, J., Wang, X. L., Lund, R., & Lu, Q. Q. (2007). A review and comparison
of changepoint detection techniques for climate data. Journal of Applied Meteorology
and Climatology, 46 (6), 900–915. doi: 10.1175/JAM2493.1
Rodgers, C., van de Giesen, N., Laube, W., Vlek, P. L. G., & Youkhana, E.
(2006, September). The GLOWA Volta Project: A framework for water resources
decision-making and scientific capacity building in a transnational West African
Basin.
Water Resources Management, 21 (1), 295–313.
Retrieved from http://
link.springer.com/10.1007/s11269-006-9054-y doi: 10.1007/s11269-006-9054-y
Sawai, N., Kobayashi, K., Takara, K., Ishikawa, H., Yokomatsu, M., Samaddar, S., . . .
Kranjac-berisavljevic, G. (2014). Impact of Climate Change on River Flows in the
Black Volta River. Journal of Disaster Research, 9 (4), 432-442.
Schuol, J., & Abbaspour, K. (2006). Calibration and uncertainty issues of a hydrological
model (swat) applied to west africa. Advances in geosciences, 9 (9), 137–143.
SCS. (1972). Section 4: Hydrology. In National engineering handbook (pp. 10–1 – 10–22).
Searcy, J. K., & Hardison, C. H. (1960). Double-Mass Curves. WaterSupply Paper 1541B ,
66. doi: http://udspace.udel.edu/handle/19716/1592
Sen, P. K. (1968). Estimates of the Regression Coefficient Based on Kendall’s Tau. Journal
of the American Statistical Association, 63 (324), 1379–1389. doi: 10.2307/2285891
Setegn, S. G., Srinivasan, R., & Dargahi, B. (2008). Hydrological modelling in the lake
tana basin, ethiopia using swat model. The Open Hydrology Journal , 2 (2008), 49–62.
Shaibu, S., Odai, S. N., Adjei, K. A., Osei, E. M., & Annor, F. O. (2012). Simulation of
runoff for the black volta basin using satellite observation data. International journal
of river basin management, 10 (3), 245–254.
Spruill, C., Workman, S., & Taraba, J. (2000). Simulation of daily and monthly stream
discharge from small watersheds using the swat model. Transactions of the ASAE ,
43 (6), 1431–1439.
97
REFERENCES
Stanzel, P., & Kling, H. (2014). Future hydropower production in the lower zambezi
under possible climate change influence. Water SA, 40 (4).
Stehman, S. V., & Czaplewski, R. L.
(1998).
Design and Analysis for
Thematic Map Accuracy Assessment - an application of satellite imagery.
Remote Sensing of Environment,
64 (January),
331–344.
Retrieved from
http://www.ingentaconnect.com/content/els/00344257/1998/00000064/
00000003/art00010$\delimiter"026E30F$nhttp://dx.doi.org/10.1016/
S0034-4257(98)00010-8
Tabari, H., Marofi, S., Aeini, A., Talaee, P. H., & Mohammadi, K. (2011). Trend analysis
of reference evapotranspiration in the western half of Iran. Agricultural and Forest
Meteorology, 151 (2), 128–136. doi: 10.1016/j.agrformet.2010.09.009
Technology Roadmap. (2011). Technology Roadmap —hydropower—International Energy
Agency. doi: 10.1007/SpringerReference\ 7300
Trabucco, A., & Zomer, R. (2009). Global aridity index (global-aridity) and global
potential evapo-transpiration (global-pet) geospatial database. CGIAR Consortium
for Spatial Information. Published online, available from the CGIAR-CSI GeoPortal
at: http://www. csi. cgiar. org/(2009). Global Aridity Index (Global-Aridity) and
Global Potential Evapo-Transpiration (Global-PET) Geospatial Database. In. CGIAR
Consortium for Spatial Information.
UNEP. (1997). United Nations Environment Programme,World atlas of desertification
2ED. UNEP, London.
Van de Giesen, N., Liebe, J., & Jung, G. (2010). Adapting to climate change in the volta
basin, west africa. Current science, 98 (8), 1033–1037.
van Vliet, J., Bregt, A. K., & Hagen-Zanker, A. (2011). Revisiting Kappa to account
for change in the accuracy assessment of land-use change models. Ecological Modelling,
222 (8), 1367–1375. Retrieved from http://dx.doi.org/10.1016/j.ecolmodel.2011
.01.017 doi: 10.1016/j.ecolmodel.2011.01.017
Watson, R., Rosswall, T., Steiner, A., Töpfer, K., Arico, S., & Bridgewater, P. (2005).
Ecosystems and human well-being. Ecosystems, Millenium Assessments, 5 , 1–100.
98
REFERENCES
Westmacott, J. R., & Burn, D. H. (1997). Climate change effects on the hydrologic regime
within the Churchill-Nelson River Basin. , 202 , 263–279.
William, A., & Matthew, J. (2010). Hydrologic modeling system hec-hms users manual.
US Army Corps Of Engineers Hydrologic Engineering Center (HEC), Washington, DC ,
20314–1000.
World Commission on Dams. (2000). Dams and development: A new framework for
decision-making: the report of the world commission on dams. Earthscan.
Xu, C.-y. (2002). Hydrologic Models. Report, 72 , 1–168.
Zăvoianu, F., Caramizoiub, A., & Badeaa, D. (2004). Study and accuracy assessment
of remote sensing data for environmental change detection in romanian coastal zone of
the black sea. In Proceeding of ISPRS , 778–783.
Zealand. (2008). Executive summary Global overview of hydropower.
Zhang, Y., Guan, D., Jin, C., Wang, A., Wu, J., & Yuan, F. (2011). Analysis of impacts
of climate variability and human activity on streamflow for a river basin in northeast
China. Journal of Hydrology, 410 (3-4), 239–247. Retrieved from http://dx.doi.org/
10.1016/j.jhydrol.2011.09.023 doi: 10.1016/j.jhydrol.2011.09.023
Zomer, R. J., Trabucco, A., Bossio, D. a., & Verchot, L. V.
(2008).
Climate
change mitigation: A spatial analysis of global land suitability for clean development
mechanism afforestation and reforestation. Agriculture, Ecosystems and Environment,
126 (1-2), 67–80. doi: 10.1016/j.agee.2008.01.014
99