EFA Namibia Geomorphology
Okavango River Basin: Transboundary
Diagnostic Analysis Project:
Environmental Flow Assessment Module
Specialist Report
Country: Namibia
Discipline: Geomorphology
Colin Christian & Associates CC
Colin Christian Soils
E.B. Simmonds
February 2009
1
EFA Namibia Geomorphology
Colin Christian & Associates CC
Environmental Consultants
3 Gordon Day St
PO Box 81182
Olympia
Windhoek
Namibia
Tel: 061 302296
Fax: 061 - 302297
Cellphone: 081 1490037
Email: colinchr@iway.na
Reg. No: CC/2006/1209
Okavango River Basin: Transboundary Diagnostic Analysis Project:
Environmental Flow Assessment Module
Specialist Report: Geomorphology Country: Namibia
Author: Colin Christian Soils input: E.B. Simmonds
C022b
February
2009
2
EFA Namibia Geomorphology
EXECUTIVE SUMMARY
INTRODUCTION
An Environmental Protection and Sustainable Management of the Okavango River Basin
(EPSMO) Project is being implemented under the auspices of the Food and Agriculture
Organization of the United Nations (UN-FAO). There are two linked components of this
project. These are:
a. A Transboundary Diagnostic Assessment (TDA) -an analysis of current and future
possible causes of transboundary issues between the three countries of the basin:
Angola, Namibia and Botswana.
b. An Environmental Flow Assessment (EFA) -to predict possible development-driven
changes in the flow regime of the Okavango River system and the impacts on
ecosystems, as well as consequent impacts on resources used by people.
Teams of specialists in all three basin states contributed to these studies for various
disciplines. This report serves to fulfil the requirements for the EFA and TDA for the
Geomorphology and Soils component in Namibia.
STUDY AREA
The study area for the EFA and TDA is the entire Okavango River catchment and Delta, but
excluding the Makgadikgadi Pans and Nata River. "Fossil river" systems in Namibia are
theoretically included but in practise they play no significant role in the functioning of the
Okavango River system.
Since the entire river cannot be studied in detail, specific sites were selected for more detailed
study. These sites were considered to be representative and relatively homogeneous in terms
of both biophysical and socio-economic characteristics. Angola has three representative sites,
Namibia two sites, and Botswana three sites. In Namibia, the two sites are Kapako and Popa
Falls. A general overview of these two sites is provided, fol owed by a detailed account of the
geomorphology and sediment processes. Site specific data on soils could not be obtained, but
data on soils from similar sites are provided in the literature review in a later section of this
report. A third site was nominated in Namibia the ecologically important and relatively
pristine islands and riverine forests of the MukweAndara-Divundu section. However, time and
budget constraints did not allow for this area to be included in the study.
IDENTIFICATION OF INDICATORS & FLOW CATEGORIES
For each discipline, indicators were selected that can be expected to change in response to
potential changes in the river flow regime. The selected indicators are usually important as
habitat for various species (which will be considered by other specialists) and/or they have
implications for people and their use of natural resources. A maximum of ten indicators per site
per discipline was permitted. The following table lists the ten indicators selected for
geomorphology, nine of which apply in Namibia. (No.2 applies only in Angola.)
3
EFA Namibia Geomorphology
Indicator
Indicator name
Sites represented
Number
Kapako 4
Popa Falls 5
1
Extent of exposed rocky habitat in main channels.
No
Yes
2
Extent of coarse sediments on the bed
No
No
3
Cross sectional area of bank full channel
Yes
Yes
4
Extent of backwater areas (slow/no flow areas)
Yes
No
5
Extent of exposed sandbars at low flow
No
Yes
6
Extent of vegetated islands
No
Yes
7
Percentage silt & clay in the top 30cm of the
Yes
No
floodplain
8
Extent of the floodplain flooded each wet season
Yes
No
9
Extent of inundated pools/pans on floodplain at the
Yes
No
end of the dry season
10
Extent of cut banks along the active channel
Yes
No
Hydrological data were supplied to the specialists in each discipline for each study site. The
hydrograph was divided into important flow categories: -namely the dry season, transitional
season 1 (rising limb), flood season, and transitional season 2 (falling limb).
LITERATURE REVIEW
A review of the limited available literature was conducted relating to geomorphology,
sedimentology and soils along the Okavango River in Namibia. In this context it was found that
there was almost no literature on fluvial geomorphology. A recent study of sediment transport
at Divundu provides valuable information on sediment transport below the confluence of the
Cuito River. There is, however, a large volume of literature on geomorphology, and
sedimentology in the Okavango Delta. Limited data on soils exists for the Namibian reach, but
very few relate soils to the flow of the river.
The Okavango River is an extremely unusual river in terms of the nature of its sediment load,
and the ecological implications of this fact are often not understood. Most importantly, the
major load that the river carries is fine sand (0.25 0.43mm) of aeolian origin that has been
reworked by water. This is transported as bedload, by saltation and not in suspension.
(Saltation is the process whereby sand grains are picked up and dropped repeatedly.) This
sand will be trapped in any weirs or dams and will not be easily moved through such
impoundments. The sand has important ecological implications for the Delta an issue that
has been well studied and documented in Botswana.
The much smaller load of silt and clay particles has important implications for the fertility of
soils on floodplains. These fine particles will not be replaced on floodplains or terraces that no
longer flood. This would not bode well for the sustainability of agriculture, because the soils
can be expected to deteriorate rapidly over time. The inherently poor nutrient status of these
quartzitic soils also means that considerable amounts of artificial fertilisers must be added,
which can adversely affect water quality in the river.
4
EFA Namibia Geomorphology
PREDICTED RESPONSES OF INDICATORS TO CHANGE IN FLOW REGIME
For every indicator that was applicable at each site, an analysis was made concerning the
expected response of that indicator to potential changes in the flow regime.
FLOW-RESPONSE RELATIONSHIPS FOR USE IN THE OKAVANGO EF-DSS
At the Knowledge Capture Workshop in April 2009, an effort was made to enter descriptive
data (in terms of the predicted responses of indicators) into the EF-DSS. The Environmental
Flows Decision Support System is a computer programme developed on a base of
hydrological data for each study site. It enables some form of "quantification" to be made
based on the available knowledge and experience of the specialist teams. The output is a
series of graphs that indicate how each indicator can be expected to respond to potential
changes in river flow. The flow-response curves are not included in this report as it was
decided at the workshop that they would be collated and compiled in a separate report by the
EFA project leaders.
An important limitation of the EF-DSS is that it was not able to take into account the effects of
sediment trapping in dams or weirs. Due to constraints of the study scope and budget, only the
effects on river flow and the associated secondary impact on local sediment transport will be
considered under various development scenarios. However, in reality, dams and weirs will trap
sediment and have a geomorphological impact downstream that is more severe than that
expected from the hydrological changes alone. This is due to the release of sediment free
water that would rapidly erode the beds and banks of the river downstream of the dam. In
areas where easily erodible extensive floodplains are present, such as at Kapako, or where
the driving riverine processes are critically dependent on inflowing bedload, such as Okavango
Delta, the sediment trapping effects of upstream dams are likely to outweigh the hydrological
impacts. Sediment trapping by dams or weirs should thus be carefully evaluated in any future
studies.
5
EFA Namibia Geomorphology
Contents
EXECUTIVE SUMMARY ............................................................................................ 3
ABBREVIATIONS ....................................................................................................... 9
ACKNOWLEDGEMENTS ......................................................................................... 10
1 INTRODUCTION ................................................................................................... 11
1.1 Background ......................................................................................................... 11
1.2 Okavango River Basin EFA: Project Objectives ............................................. 11
2 STUDY AREA ........................................................................................................ 12
2.1 Description of the Okavango Basin ................................................................. 12
2.3.1 Site 4: Kapako -Overview ......................................................................... 16
2.3.2 Site 5: Popa Falls -Overview ................................................................... 20
2.4 Geomorphological Description : Site 4 Kapako ............................................... 23
2.4.1 Methods ........................................................................................................ 23
2.4.2 Geomorphological Description & Data for Kapako ....................................... 24
2.5 Geomorphological Description : Site 5 Popa Falls .......................................... 35
2.5.1 Methods ........................................................................................................ 35
2.5.2 Geomorphological Description & Data for Popa Falls .................................. 35
2.6 Habitat integrity of the sites in Namibia ........................................................... 44
3 IDENTIFICATION OF INDICATORS AND FLOW CATEGORIES ........................ 45
3.1 Indicators ............................................................................................................ 45
3.1.1 Introduction ................................................................................................... 45
3.2 Flow categories river sites ............................................................................... 51
4 LITERATURE REVIEW ......................................................................................... 53
4.1 Introduction ......................................................................................................... 53
4.2 Literature on Geomorphology and Sedimentology ............................................. 53
4.3 Information on Soils ............................................................................................ 55
4.4 Summary & Information Gaps ............................................................................ 66
5 PREDICTED RESPONSES OF INDICATORS TO CHANGE IN FLOW REGIME
............................................................................................................................... 68
5.1 Site 4: Kapako Indicator responses to changes in flow regime ....................... 68
5.3 CONCLUSION .................................................................................................... 83
6 REFERENCES ...................................................................................................... 84
6
EFA Namibia Geomorphology
TABLE OF FIGURES
Figure 2. 1: Upper Okavango River Basin from Sources to the Northern end of the
Delta .............................................................................................................................. 13
Figure 2. 2: The Okavango River Basin, Showing Drainage into the Okavango Delta 14
Figure 2. 3: Kapako Village ........................................................................................... 17
Figure 2. 4: Satellite images of the Kapako Flood Plain ............................................... 18
Figure 2. 5: Site 5, Popa Falls. the majority of the population lives along the river ..... 21
Figure 2. 6: Satellite image of Popa Falls showing wet and dry transects and main
sampleing sites ............................................................................................................. 22
Figure 2. 7: Kapako Map with Countours ..................................................................... 25
Figure 2. 8: Kapako Cross Section ............................................................................... 26
Figure 2. 9: Popa Falls Map with Contours .................................................................. 37
Figure 3. 1:Three representative years for Site 4: Okavango River at Kapako
(hydrological data from Rundu), illustrating the approximate division of the flow regime
into four flow seasons ................................................................................................... 51
Figure 3. 2:Three representative years for Site 5: Okavango River at Popa
(hydrological data from Mukwe), illustrating the approximate division of the flow regime
into four flow seasons ................................................................................................... 52
Figure 4. 1: Schematic diagram (not to scale) of Pterocarpus angolensis
Schinziophyton Rautanenii woodland and associated soil families on the floodplain
and terrace system of the Okavango River.
62
7
EFA Namibia Geomorphology
TABLE OF TABLES
Table 2. 1: Location of the eight EFA sites ................................................................... 15
Table 2. 2: Grain Size distribution of the dominant sediments ..................................... 31
Table 3. 1: List of indicators ......................................................................................... 45
Table 3. 2: Questions addressed at the Knowledge Capture Workshop, per indicator
per site. In all cases, `natural' embraces the full range of natural variability ................ 52
Table 4. 1: Vegetation Types and Soil Associations Aggregated by Land Syatem ..... 61
Table 5. 1: Predicted response to possible changes in the flow regime of cross
sectional area of the channel ........................................................................................ 69
Table 5. 2: Predicted response to possible changes in the flow regime of slow/no flow
backwaters .................................................................................................................... 70
Table 5. 3: Predicted response to possible changes in the flow regime of percentage
silt and clay in the top 30cm on floodplains .................................................................. 72
Table 5. 4: Predicted response to possible changes in the flow regime of the extent of
inundated floodplains .................................................................................................... 74
Table 5. 5: Predicted response to possible changes in the flow regime of inundated
pools and pans .............................................................................................................. 76
Table 5. 6: Predicted response to possible changes in the flow regime of cut banks .. 77
Table 5. 7: Predicted response to possible changes in the flow regime of exposed
rocky habitat .................................................................................................................. 79
Table 5. 8: Predicted response to possible changes in the flow regime of cross
sectional area of the channels ...................................................................................... 80
Table 5. 9: Predicted response to possible changes in the flow regime of sandbars at
low flow ......................................................................................................................... 81
Table 5. 10: Predicted response to possible changes in the flow regime of vegetated
islands ........................................................................................................................... 82
TABLE OF PHOTOGRAPHS
Photos 1: 1-3 ................................................................................................................. 27
Photos 2: 4-6 ................................................................................................................. 28
Photos 3: 7-9 ................................................................................................................. 38
Photos 4: 10-12 ............................................................................................................. 39
Photos 5: 13-15 ............................................................................................................. 40
Photos 6: 16-18 ............................................................................................................. 56
Photos 7: 19-21 ............................................................................................................. 57
8
EFA Namibia Geomorphology
ABBREVIATIONS
ABBREVIATION MEANING
DSS
Decision Support System
DTM
Digital Terrain Model
EFA
Environmental Flows Assessment
EIA
Environmental Impact Assessment
EPSMO
Environmental Protection and Sustainable Management of the
Okavango River Basin
FAO
Food and Agriculture Organization
IUAs
Integrated Units of Analysis
m.a.s.l.
Metres above sea level
OBSC
Okavango Basin Steering Committee
OKACOM
Okavango River Basin Commission
TDA
Trans-boundary diagnostic Assessment
UN
United Nations
CEC
Cation Exchange Capacity
9
EFA Namibia Geomorphology
ACKNOWLEDGEMENTS
Ms. E.B. (Sophie) Simmonds provided input on soils and soil indicators and compiled part of
Section 4.3 on soils.
Sections 2.3.1 and 2.3.2, being the overviews of the Kakapo and Popa Falls sites, were written
by Ms. Shirley Bethune and are shared in common with the other specialist reports.
Mr. Mark Rountree provided discussions on defining geomorphological indicators. He also
helped us at the Knowledge Capture Workshop in Windhoek to interface with the computer
model in drawing up the response curves, and provided comments on a draft of this report.
Ms Jackie King provided helpful comments on a draft of this report.
Ms. Celeste Espach from the Ministry of Agriculture kindly assisted with GIS work and maps.
Prof. Terence McCarthy kindly supplied reprints of research papers by the Okavango Working
Group at WITS University.
10
EFA Namibia Geomorphology
1 INTRODUCTION
1.1
Background
An Environmental Protection and Sustainable Management of the Okavango River Basin
(EPSMO) Project is being implemented under the auspices of the Food and Agriculture
Organization of the United Nations (UN-FAO). One of the activities is to complete a
transboundary diagnostic assessment (TDA) for the purpose of developing a Strategic Action
Plan for the basin. The TDA is an analysis of current and future possible causes of
transboundary issues between the three countries of the basin: Angola, Namibia and
Botswana.
The Okavango Basin Steering Committee (OBSC) of the Okavango River Basin Water
Commission (OKACOM) noted during a March 2008 meeting in Windhoek, Namibia, that
future transboundary issues within the Okavango River basin are likely to occur due to
developments that would modify flow regimes. The OBSC also noted that there was
inadequate information about the physico-chemical, ecological and socioeconomic effects of
such possible developments. OBSC recommended at this meeting that an Environmental Flow
Assessment (EFA) be carried out to predict possible development-driven changes in the flow
regime of the Okavango River system, the related ecosystem changes, and the consequent
impacts on people using the river's resources.
The EFA is a joint project of EPSMO and the Biokavango Project. One part of the EFA is a
series of country-specific specialist studies, of which this is the Geomorphological Report for
Namibia.
1.2 Okavango River Basin EFA: Project Objectives
The goals of the EFA are:
To summarise all relevant information on the Okavango River system and its users, and
collect new data as appropriate within the constraints of the EFA
a. to use these to provide scenarios of possible development pathways into the future
for consideration by decision makers, enabling them to discuss and negotiate on
sustainable development of the Okavango River Basin;
b. to include in each scenario the major positive and negative ecological, resource-
economic and social impacts of the relevant developments;
c. to complete this suite of activities as a pilot EFA, due to time constraints, as input to
the TDA and to a future comprehensive EFA
.
The specific objectives are:
a. to ascertain at different points along the Okavango River system, including the
Delta, the existing relationships between the flow regime and the ecological nature
and functioning of the river ecosystem;
b. to ascertain the existing relationships between the river ecosystem and peoples'
livelihoods;
c.
to predict possible development-driven changes to the flow regime and thus to the
river ecosystem;
d. to predict the impacts of such river ecosystem changes on people's livelihoods.
11
EFA Namibia Geomorphology
e. To use the EFA outputs to enhance biodiversity management of the Delta.
f.
To develop skills for conducting EFAs in Angola, Botswana, and Namibia.
Layout of this report
Chapter 2 contains an overview of the Okavango Basin in Sections 2.1 and 2.2. Section 2.3
contains overviews of the two sample/study sites in Namibia -Site 4: Kapako and Site 5: Popa
Falls. Sections 2.4 and 2.5 provide some details of the geomorphology of these two Namibian
sites.
Chapter 3: Section 3.1 provides a list of indicators used in relation to the two sample/study
sites and explanations of these indicators. Section 3.2 provides selected graphs showing of
river discharge and the four stages in the annual cycle.
Chapter 4 is a brief account of the very limited literature available for geomorphology and soils
along the Okavango River. Unfortunately none of the available literature relates directly to the
flow indicators.
Chapter 5 provides some limited descriptive data and sketchmaps based on aerial photos and
the fieldwork. Due to the limitations of time and budget, it was only possible to attend the flood
season fieldtrip in February 2009.
Chapter 6 was to have presented the response curves that were developed at the Knowledge
Capture Workshop in March/April 2009. However, it was decided at that workshop that the
response curves would be handled outside of the specialist reports. Therefore Chapter 6 is
limited to a few remarks that must be taken into account in further consideration of the
response curves and any further development of the model.
2 STUDY AREA
2.1 Description of the Okavango Basin
The Okavango River Basin consists of the areas drained by the Cubango, Cutato, Cuchi,
Cuelei, Cuebe, and Cuito rivers in Angola, the Okavango River in Namibia and Botswana, and
the Okavango Delta ( Figure 2.1). This basin topographically includes the area that was
drained by the now fossil Omatako River in Namibia. Outflows from the Okavango Delta are
drained through the Thamalakane and then Boteti Rivers, the latter eventually joining the
Makgadikgadi Pans. The Nata River, which drains the western part of Zimbabwe, also joins
the Makgadikgadi Pans. On the basis of topography, the Okavango River Basin thus includes
the Makgadikgadi Pans and Nata River Basin ( Figure 2.2). This study, however, focuses on
the parts of the basin in Angola and Namibia, and the Panhandle/Delta/Boteti River complex in
Botswana. The Makgadikgadi Pans and Nata River are not included.
12
EFA Namibia Geomorphology
Figure 2. 1: Upper Okavango River Basin from Sources to the Northern end of the Delta
13
EFA Namibia Geomorphology
Figure 2. 2: The Okavango River Basin, Showing Drainage into the Okavango Delta
Delineation of the Okavango Basin into Integrated Units of Analysis
Within the Okavango River Basin, no study could address every kilometre stretch of the river,
or every person living within the area, particularly a pilot study such as this one. Instead,
representative areas that are reasonably homogeneous in character may be delineated and
used to representative much wider areas, and then one or more representative sites chosen in
each as the focus for data-collection activities. The results from each representative site can
then be extrapolated over the respective wider areas.
Using this approach, the Basin was delineated into Integrated Units of Analysis
(EPSMO/Biokavango Report Number 2; Delineation Report) by:
a. dividing the river into relatively homogeneous longitudinal zones in terms of:
· hydrology;
· geomorphology;
· water chemistry;
· fish;
· aquatic invertebrates;
· vegetation;
14
EFA Namibia Geomorphology
b. harmonising the results from each discipline into one set of biophysical river
zones;
c. dividing the basin into relatively homogeneous areas in terms of social systems;
d. harmonising the biophysical river zones and the social areas into one set of
Integrated Units of Analysis (IUAs).
The 19 recognised IUAs were then considered by each national team as candidates for the
location of the allocated number of study sites:
Angola: three sites
Namibia: two sites
Botswana: three sites.
The sites chosen by the national teams are given in Table 2.1.
Table 2. 1: Location of the eight EFA sites
EFA Site No
Country
River
Location
1
Angola
Cuebe
Capico
2
Angola
Cubango
Mucundi
3
Angola
Cutio
Cuito Cuanavale
4
Namibia
Okavango
Kapako
5
Namibia
Okavango
Popa Falls
6
Botswana
Okavango
Panhandle at
Shakawe
7
Botswana
Khwai
Xakanaka in Delta
8
Botswana
Boteti
Chanoga
Overview of Units of Analysis and Study sites
In the Namibian section of the Okavango River, the majority of the human population lives
along the river and the main road, with several hot spots such as Rundu, Divundu and
Nkurenkuru which have a high population density. The river can be divided into four clear units
of analysis:
1) The longest unit extends from where the river enters Namibia at Katwitwi to the Cuito
confluence. This unit is typified by the meandering mainstream and large seasonally-flooded
floodplains on either side of the river. Kapako, site 4, was chosen as a typical meandering
floodplain site to represent this unit;
2) This unit is immediately downstream of the Cuito confluence and is characterised by
permanent swamp areas and large islands. No sites were selected in this preliminary
survey, but it would be essential to include the assessment of this unit in any future more
detailed EFA studies;
15
EFA Namibia Geomorphology
3) The unit from Mukwe to just below the Popa Falls is a southward flowing, rocky, braided
section. The river is largely confined to the mainstream and flows around several sand
and rock based islands. Popa Falls Site 5, was chosen as a typical rocky river site to
represent this section.
4) The protected section of the river downstream of Popa Falls to the border with Botswana at
Mohembo that lies within the newly declared Bwabwata National Park is the last unit
within Namibia. Two core conservation areas on either side of the river, the Buffalo core
area on the west bank and the Muhango core area on the east bank. This area has
extensive floodplains and represents the beginning of the transition to the Panhandle part
of the Okavango Delta.
2.3.1
Site 4: Kapako -Overview
The main focus point for socio-economic work at the Kapako floodplain site 4 is Kapako
village: S-17.94 E19.56, situated some distance inland from the river on the other side of the
main road (Figure 2.3). Kapako is approximately 20km west of Rundu by road.
The main villages close to Kapako village are Mupini to the east (downstream), Mukundu to
the south, Ruugua and Sinzogoro to the west (upstream).
The floodplain site itself is situated on the Okavango River and three main sites on the
floodplain and the mainstream were used for sampling. They were:
a. Kapako site 1 S-17.87775 E-19.58200 (start south bank) S-17.87850 E-19.58211 (end
of site 1)
b. Kapako site 2 S-17.86557 E-19.58057 (start at floodplain only 3 observations due to
flooding)
c. Kapako site 3 S-17.86209 E-19.57855 (deep pool).
The riverine landscape includes the main Okavango River channel or mainstream, the
annually flooded floodplains with several braided side channels and deeper pools or
backwaters, as well as the higher fluvial terrace with alluvial deposits that are very seldom
flooded. There is a steep, well vegetated bank at the edge of the floodplain close to the main
road that rises to several meters above the floodplain.
Kapako area has a population of approximately 2,500 people within 10 km of Kapako village.
The greatest density of people (over 100 per km2) live alongside the river in the area just west
of the Kapako study site whilst at the site itself the density varies from no people on the
floodplain, 6 25 / km2 at the Ebenezer mission, to a density of 25 50 /km2 closer to the road
and 51 100 /km2 on the other side of the main road, rapidly decreasing again with distance
inland. (See Map 3 in Populations Demographics Report prepared by Celeste Espach). We
can assume that some of these people make some use of the floodplain site at Kapako and
elsewhere along this stretch of river.
16
EFA Namibia Geomorphology
Figure 2. 3: Kapako Village
17
EFA Namibia Geomorphology
Figure 2. 4: Satellite images of the Kapako Flood Plain
During the focus group discussion held at Kapako village, the basin residents mentioned that
the flooding starts when the rising river waters push out over flat surrounding ground and the
biggest floodplains form in years when river levels are highest. They said that the most
important feature of the flooded areas is that they are rich in nutrients. The floodplains also
offer the young fish refuge from larger, predatory species and thus offer the greatest survival
of young fish. They had noted that an overall increase in fish population occurs in years when
18
EFA Namibia Geomorphology
water levels are high and flooding lasts longest. Local people have recognised that water
quality and fish resources are decreasing in the Okavango River. Fish and fishing remain
significant features in the lives of people at Kapako, who fish for food or to earn incomes by
selling their catches. In addition some earn money by providing trips for tourists. They estimate
fish stocks in the floodplains to be four times higher than in the main channel.
About 47% of households at Kapako catch fish, and each person consumes an average of 10-
20 kilograms of fish per year. September to December is the peak fishing period at Kapako
when the river is at its lowest and fish are concentrated in the mainstream. The kinds of traps
or gear used to catch fish are separated into traditional and modern methods. The most used
traditional gear are fish funnels, kraal traps, scoop baskets, push baskets, bows and arrows,
set fish hooks and spears. Modern gear consists of line and hooks, wire mesh fykes, illegal
mosquito nets, and gill and seine nets. The use of fish for recreational angling forms part of the
tourism value associated with the river. Biophysical response curves for the angling species
would feed into the tourism values for the river reducing them partially. Only a small part of
tourism value is attributable to angling.
At Kapako, as elsewhere along the Namibian section of the river, the ever -increasing human
population and clearing for crops and livestock has put increasing pressure on the natural
resources along the main channel. The vegetation along the river bank is overgrazed and in
some areas depleted, thus at Kapako the residents graze their livestock across the river on the
Angolan floodplain. Cattle were routinely seen being swam across the river at this site during
fieldwork. Associated with this population growth, has been an increase in livestock, fire
frequency as well as the area of land cleared for crops and fuel. These associated land use
changes are an undeniable factor of increasing settlement and development at a Kapako and
indeed all along the Okavango.
The road westwards from Rundu has been upgraded and is currently being tarred. It runs
parallel to the Okavango River all the way to the border post with Angola at Katwitwi. This has
opened up the region allowing people to exploit the land alongside the road. As expected,
highest densities are alongside the road parallel to the river. As the population continues to
increase, exploitation of the land that new roads have opened up should disperse the pressure
on the Okavango River floodplains and its resources to land further inland from the river,
although the river will always remain the main source of water even for livestock watering.
The extent of erosion and clearing and thus of bare ground has also increased, yet the people
perceive the overall water quality not to have declined substantially. The only exceptions
mentioned were an increase in phosphate concentrations, a decrease in water clarity and a
related increase in suspended sediments. There are more short term, seasonal variations in
water quality particularly in the floodplain pools, than any long term water quality change. So
far there does not seem to have been an excessive exploitation of the water resources in the
main channel, although the basin further inland has some serious water shortages at times
and a lack of deep boreholes.
The Kalahari sands that overlay the area are deep.
19
EFA Namibia Geomorphology
2.3.2
Site 5: Popa Falls -Overview
The Popa Falls site is approximately 5km south of Divundu bridge by road.
The main focus for the socio-economic work at the Popa Falls Site 5 was the village of Popa and the
Popa Falls Rest Camp run by Namibia Wildlife Resorts. The main transect used for the physical and
biological field survey work was a transect across the river immediately above the Popa Falls from the
irrigation water draw-off point used by the Prison Services on the eastern bank (West Caprivi) where a
gauge plate was put up to the protected section close to the Popa Falls Rest Camp on the western bank.
Popa Falls rest camp: S 18.15316
E 21.6045 Popa Falls gauge plate:
S 18.11603 E 21.57900
Figure 2.5 shows the main villages. Figure 2.6 shows two satellite images of Popa Falls.
About 3,000 people live in the area surrounding Popa. The highest population density in the
area is immediately upstream of the Popa Falls at the Bagani/Divundu settlement, within an
area of over 12 km2. At the Popa Falls site itself the population density is much lower at 6 25
people/km2 and it must be remembered that the Popa camp is within an 8 km2 park, the
islands are uninhabited and the opposite bank supports a community campsite reserved for
tourists. Immediately downstream of Popa camp the riverside population increases to 26-50
people/km2 and includes several lodges. See Map 3 in the Population Demographics report
prepared by Celeste Espach for the TDA.
At the Popa Falls, the entire width of the river cascades down four meters before resuming its
normal slow and leisurely flow. The quartzite rocks were formed from sediments deposited in rift
valleys about 900 million years ago, (Mendelsohn el Obeid, 2004).
20
EFA Namibia Geomorphology
Figure 2. 5: Site 5, Popa Falls. the majority of the population lives along the river
21
EFA Namibia Geomorphology
Figure 2. 6: Satellite image of Popa Falls showing wet and dry transects and main sampleing
sites
22
EFA Namibia Geomorphology
During the focus group discussion, it was mentioned that due to the Popa Falls and rocky
areas, it's difficult for the local fishermen to catch fish as desired. Therefore, only a few
individuals that own local mukoros, hook and line, and gill fish nets have access to fish
catches in the main channel. Thus fishing is a secondary activity for most people at the Popa
area, contributing little to the overall cash or in-kind incomes of the majority of households.
People also pay much less attention to fishing than to farming and business activities. Each
household depends on a different mix of incomes derived from wages, business earnings,
pensions and remittances.
Papyrus cyperus (papyrus) dominates the deepest water margins alongside the main
channels. Water can seep through the walls of papyrus to the reedbeds behind the papyrus
and in places into backwaters and side channels. The sandy sediments are confined to the
channels. These are flanked by reed beds of Phragmites, Typha capensis or bulrushes and
the sedge Miscanthus junceus in the shallower waters. The residents do not experience floods
as there are no floodplains in this area. They depend in the main channel for most of their
water and wetland resources. Most houses at Popa village are thatched with grass and reeds,
while reeds are used extensively to make sleeping mats, walls, palisades, courtyards and
fences.
Farming activity is an important source of income; households are engaged in both crop and
livestock farming. Planting is staggered through the raining reason and is initiated only after a
good rainfall event. This increases the chance of crop survival during the hot dry periods.
Livestock farming is dominated by cattle and goats, not kept within fields but are moved for
grazing and between water sources, mainly the Okavango River.
Tourism is a major source of income to the Popa residents; most of them are employed within
the lodges around the Popa area. They value tourism as their major source of income.
The maps below (Figure 2.6) show the Popa Falls Site 5 in both wet and dry season. The main
field survey transect and sampling sites are indicated by red dots.
2.4 Geomorphological Description : Site 4 Kapako
2.4.1 Methods
The site visit to Kapako was undertaken on 8-9 February 2009. A high water mark on banks
and vegetation showed that the flood had already peaked and then subsided by about 0.5m.
The flood was overflowing the banks by about half a metre and almost the entire active
floodplain was inundated at the time. (Note that a much higher peak followed in March but we
were not present.)
The water level was too high to allow walking on most of the active floodplain, and not high
enough to allow the floodplains to be traversed by boat. Therefore the inspection of the area
was restricted mainly to a boat trip on the main channel and viewing from there. Where
possible we got out onto higher banks and viewed from there. Photographs were taken where
possible, but the complete lack of high vantage points was a limitation. Thus the extent of
flooding could not be properly determined.
Photos were also taken by Mark Paxton at the Kapako site from December 2008 to February
2009 in order to show the rising flood levels in relation to the floodplains. However, the lack of
high vantage points from which to view substantial areas of the floodplain limited the
usefulness of these photos in showing the extent of inundation in relation to the rising stage of
the river.
23
EFA Namibia Geomorphology
Depths were measured from the boat or on foot in channels and backwaters using a
surveyor's pole at locations shown by red dots in Figure 2.4. Details and cross sections are
provided in a report by Celeste Espach.
Other information provided in the following sub-section was drawn from the author's
experience of the Okavango River in Namibia, topographic maps, and the satellite images
provided as Figure 2.4. Interpretation of the landforms was made based on the satellite
images, with rough contours superimposed, and observations and photographs in the field.
This information was mapped in Figure 2.7. Areas of water features were measured by a
simple squares method on a much larger scale print of the same map. Schematic cross
sections were made and presented as Figure 2.8.
Information from the literature also contributed to the following section.
2.4.2 Geomorphological Description & Data for Kapako
Figure 2.7 is an orthophoto of the Kapako site, with contours generated in a GIS programme.
The contours are not accurate because the small contour interval of 1m, which was chosen in
order to define such a flat area, is too fine for the satellite images. However, the contours give
some indication of the topography and elevations of the landscape.
Figure 2.8 shows two schematic Cross Sections through the floodplains. The location of these
cross sections is shown in Figure 2.7 by the lines marked A A1 and B -B1.
Photos 1 6 show parts of the Kapako site as seen from a boat on the river. The floodplains
were mostly too deeply inundated to allow access on foot, but not deeply enough for a boat to
move about.
24
EFA Namibia Geomorphology
Figure 2. 7: Kapako Map with Countours
25
EFA Namibia Geomorphology
Figure 2. 8: Kapako Cross Section
26
EFA Namibia Geomorphology
Photos 1: 1-3
27
EFA Namibia Geomorphology
Photos 2: 4-6
28
EFA Namibia Geomorphology
The floodplains
The flood plains at Kapako are extensive on the Namibian side. At its northern and southern
extremities within the study site, the river cuts into the dry banks, but for the most part,
ongoing changes in the river course occur within the floodplain. Many scroll bars mark the
former courses of the river across the floodplain (Figure 2.7).
The floodplains are covered in wetland grasses and reeds which gives some stability to the
landforms in the study area. Patches of trees are very limited in extent, often confined to river
banks of the main channel or old channels.
Local human impacts on the geomorphology of the floodplain appear to be very limited. The
population is moderately dense on the Namibian side, but sparse on the Angolan side. Most of
the human disturbance, such as agriculture occurs outside of the dry banks of the river, and
mostly on the Namibian side. Aerial photos show little or no evidence of cultivation on the
active floodplain. Cattle graze on the floodplains during the dry season, and fish are caught on
the floodplains and main channel. However, there are no human activities on the floodplains
that would result in significant impacts on the geomorphological features here. At very limited
localities, vehicular access to the river has led to deep erosion of the sandy material over very
small areas.
The overall floodplain on the Namibian side of the river varies in width from almost nothing to
approximately 2,2km. The exact extent is difficult to determine because the images supplied
were cut off such that they do not show the entire length of the dry banks on either side of the
river. This is particularly true for the Angolan side. Based on the Namibian 1: 250,000
topographic map (2002) and Figure 2.4, the floodplain is shown to be almost as wide in Angola
as on the Namibian side but only slightly further upstream.
The dry banks of the valley are outlined in orange in Figure 2.7. On the Namibian side these
banks rise at least 10 metres above the floodplain (typically 1070 m.a.s.l. at the base rising to
1080 m.a.s.l. at the top). The material that comprises these dry banks is Kalahari sand.
Because the entire area is comprised of these paleo-dunes, it is reasonable to assume that
the dry banks on the Angolan side are of a similar elevation. The dry banks vary in gradient,
becoming steepest where they are being actively eroded at the base. In that case the gradient
can be as steep as 8%. In Namibia, in the far west of Figure 2.7 it can be seen that the main
river channel comes close to eroding the dry banks.
At the base of the dry banks, deposits of "calcrete" are common overlain by the dune sand.
These deposits would have been formed in situ. The river water, although very pure,
nevertheless contains low concentrations of salts, silicates and carbonates. The margins of the
floodplain (the bases of the dry banks) are subject to seasonal wetting and drying. In this
region where average evaporation exceeds average rainfall in every month of the year
(Mendelsohn & el Obeid, 2004) there is a net loss of water from the floodplains to the dry
banks, followed by evaporative and transpirative losses to the atmosphere. Before intense
human settlement and clearing of woodlands, transpiration by large trees probably accounted
for water being actively drawn out of the floodplain and into the dry banks even more so than
today. McCarthy, Ellery & Dangerfield (1998) described a similar process on islands in the
Okavango Delta. The dissolved load in that water is left behind in the sandy dry banks as
carbonate and silicate deposits, loosely referred to here as "calcrete". Outcrops of this
material, which is whitish in colour, are sometimes visible and are sometimes exploited for
building material or road construction. It is the author's observation that these deposits may
play a role in limiting the horizontal erosion of the dry banks of the river. However whether or
not these calcrete deposits are continuous beneath the dry banks has not been established.
29
EFA Namibia Geomorphology
The dry banks vary in condition. Near Kapako they are mostly disturbed to a degree by
grazing livestock, people walking down to the river, and in places where they are less steep,
they are sometimes cultivated in the rainy season.
In elevation, the floodplain is very flat. From dry bank to dry bank, the elevation typical y varies
by only 4 metres (from 1066 m.a.s.l. to 1070 m.a.s.l.). Within the floodplain two zones can be
identified. This was confirmed during the field trip in the high flow season, February 2009. A
comparison between Photo 2 and Photo 5 shows the banks overflowing in the former case
and a bank of about 1m in the latter case reflecting the two discrete levels of floodplains.
The active floodplain zones are outlined in blue in Figure 2.7 and will be referred to as the
"Blue Zones". These zones vary in elevation by only 2m (typically 1066 m.a.s.l. 1067
m.a.s.l. but occasionally reaching 1068 m.a.s.l) and are repeatedly being reworked by the river
during every flood season. Numerous scroll bars mark the progress of former river channels
as the river meandered over the floodplain. Blue Zone 1 (Figure 2.7) covers an area of
approximately 288ha. The extent of the other active floodplain zones cannot be determined
from the cropped images that are available. Some of the deeper old courses become active
channels during flooding e.g. the obvious channel that borders Blue Zone1 on its southern to
eastern sides in Figure 2.7.
The older passive, relict floodplain zones are outlined in green in Figure 2.7 and will be
referred to as the "Green Zones". In some places these abut the river, but usually they are
separated from the river by an active floodplain zone (indicated in blue). The relict floodplains
are at an elevation of at least a metre or two higher than the active floodplain zones (typically
1067 m.a.s.l. to 1068 m.a.s.l. but reaching up to 1069 or 1070 at the base of the dry banks).
The relict floodplains appear more homogeneous, the scroll bars having been levelled by
rainfall, wind and livestock. These relict floodplains also have a few depressions in them of 1m
or 2m depth. In some places the contours suggest these depressions are as low as the level of
the main river channel. They can fill with water even if the river does not flood into them. In the
case of pools that are somewhat remote from any active channels, the filling of such pools is
probably explained by a combination of rainfall and seepage from an elevated groundwater
table during the flood season. In other words water seeps through the sandy substrate into the
pools, either from rainfall or from the river, or both while the elevated groundwater table
prevents drainage through the sandy substrate.
Various hypotheses are suggested by the author to explain the older and higher "relict"
floodplains.
Hypothesis A: In a period of much higher rainfall and greater seasonal extremes, the active
floodplains may have been far more extensive. Reduced rainfall and flooding would have
shrunk the area of active floodplain. According to regional modelling research at Oxford
University, the Kavango Region lies within a climate change zone where warming and drying
is the current trend (BBC News / Science-Nature / African sands `set for upheaval' (30 June
2005). However, this trend may be only about 100 years old, which may be too recent to
explain the existence of the older floodplain. The ages of these floodplain zones are not
known.
Hypothesis B: Tectonic processes probably also played a role. It is known that the Delta area
is seismically active and that the Delta lies between two faults trending south-west to northeast
while the panhandle is a graben structure between two parallel faults (McCarthy, Green and
Franey, 1993; McCarthy et al, 1997; Gumbricht, McCarthy and Merry, 2001; McCarthy et al,
2002). Downward movement of the lower Okavango Basin would also have resulted in slight
downward cutting of the river.
30
EFA Namibia Geomorphology
Hypothesis C: Changes in the sediment yield from the catchment may also explain the older,
higher floodplains which may represent a period of much greater transport and deposition.
This would have to have been followed by a period of reduced transport and deposition so that
the River cut into the older floodplain.
Sediment: origins and transport
Both the active and passive floodplains are all made of the same fine sand as the dunes. The
Kalahari dunes were deposited under a paleo-climate that was much drier than the present.
These dunes are visible on large scale aerial or satellite photos, trending roughly east-west in
the northern Kavango Region (Mendelsohn & el Obeid, 2004). They cover the whole Kavango
Region, much of southern Angola, Zambia and Botswana. Almost the entire catchment of the
Okavango River in Angola is covered in a deep layer of Kalahari sand, and these dunes still
form the dry banks of the river val ey.
Because of its aeolian origin, the particle size is in a narrow band. Previous studies of the
sediment load at Popa Falls (EcoPlan, 2003) indicate that almost 90% is fine sand (Table 2.2)
making the Okavango an extremely unusual river with regard to its sediment load. The river
carries very little silt-or clay-sized particles a fact that explains the clarity of the water. What
little clay there is in the system has important implications for soils and soil fertility, but very
little importance for landforms because there is so little of it.
Table 2. 2: Grain Size distribution of the dominant sediments
Grain size (mm)
Percentage of sample
>0.43
0.0%
0.25 0.43
89.6%
0.13 0.25
1.4%
0.11 0.13
0.8%
These results concur closely with the study by McCarthy (2003) who found that the sediments
were predominantly in the range 0.25 0.39mm and are transported as bedload. The sand
grains are too large to be carried along in suspension, so they are moved along the bed of the
river in a process known as saltation. McCarthy (June, 2003) has described the relationship
between bedload sediment transport and flow velocity by the following equation.
Qs = 0.15U3.4 where Qs is bedload discharge in kg/m/s and U is average flow velocity in m/s.
No study of sediment transport on the Cubango/Okavango River has been found. All the
available studies have been done below the confluence with the Cuito River. The most up to
date study of sediment transport was carried out at Divundu / Popa Falls. McCarthy (2003)
measured and estimated annual bedload sediment discharge at 117,000 tonnes (about
70,000m3). With regard to suspended sediment, McCarthy & Ellery (1998) estimated the
volume passing through Mohembo into the delta to be only 39,000 tonnes.
Since there are no figures available for sediment transport in the Cubango/Okavango River
upstream of the confluence with the Cuito, one can only get an indication from the following
analysis. The Cubango /Okavango River contributes 55% and the Cuito River 45% of all the
water flowing to the Delta (Mendelsohn & el Obeid, 2004). River discharge and gradient are
both related to flow velocity. Flow velocity, in turn, has been shown (McCarthy, 2003) to be
related to the third power of the average flow velocity. Two implications of these facts need to
31
EFA Namibia Geomorphology
be considered in making an educated guess as to the relative volumes of sediment contributed
by each major tributary.
a. The gradient of the Cubango/Okavango River is steeper than the Cuito for
most of its length. Therefore greater discharge and gradient imply higher
flow velocity and therefore higher sediment transport.
b. The Cubango/Okavango River has much higher peak discharges than the
Cuito. Because bedload sediment discharge is related to the cube of the
flow velocity, most of the sediment discharge will occur in the flood season.
c. The Cuito River has more extensive wetlands in its lower reaches. This may
act to reduce sediment discharge due to greater deposition.
These three factors all act in the same direction, so it is reasonable to assume that the
Cubango/Okavango River carries considerably more sediment than the Cuito River.
If the above equation is applied to historical flow data, it should be possible to estimate how
much sediment is coming down each of these major tributaries. However, the above equation
should be tested by means of direct measurement of sediment transport in these tributaries.
The river channel, levees & scroll bars
The river channel itself varies in width from 60m to 140m at the Kapako site. In February 2009,
depths across the channel were recorded by sounding from a boat. It reached 4.2m deep with
about 0,5m overflowing its banks at "Site 2" (Figure 2.4).
Figure 2.4 shows satellite images of the Kapako study area in wet and dry seasons. Three
locations are indicated by red dots where depths were measured. These were: "Site 1" A
transect of the old channel to the south of the floodplain, "Site 2" The main channel and
adjacent active floodplain on the south side, "Site 3" A major backwater area upstream of the
oxbow loop.
Figure 2.8 shows two schematic cross sections of the floodplain, the location of which is
shown in Figure 2.7 by the lines marked A A1 and B -B1. In cross section A A1 note the
long slope from the main channel to an old channel, with numerous scrol bars that mark the
former positions of the old channel. The formation of scroll bars is explained as follows.
Since flow velocity is the most important factor, most of the bedload transport of sand occurs
in the flood season. In high flow conditions, when the river is overflowing onto the floodplains,
sand is picked up in the water column briefly before being deposited on the bed again. As
soon as the water leaves the deep channel and enters the floodplain, its flow velocity drops
substantially (due to reduced depth and vegetation). As a result any sand that it is carrying
gets deposited immediately on the banks next to the main channel. The river banks are thus
slightly higher than the adjacent floodplain. These raised banks are called levees.
In any cross section of a river bend, the water flows faster on the outside of the bend and
slower on the inside of the bend. The result is erosion of the outside of the bend and
deposition on the inside of the bend. Thus the bend is extended outwards. In this process the
32
EFA Namibia Geomorphology
levees that were on the inside of the bend remain as scrol bars. These are very evident on the
active floodplains (Blue Zones) in Figure 2.7.
During the flood season field trip in February, water was seen overflowing strongly out of the
main channel onto the active floodplain (at, Blue Zone 1 in Figure 2.7). A bit further upstream,
the over-bank flow has been sufficient, over a number of successive floods, to obliterate the
old scrol bars to form a "reverse fan" shape. The water that leaves the main channel on the
south side, eventually reaches the old channel that fringes the active floodplain. This flows
northwards, and eventual y discharges into the main channel downstream of the oxbow loop.
Minor flow probably continues in the old channel long after the flood has receded, fed by
seepage out of the floodplain. The levees on the south side of the river appear to have been
lowered over a bank length of about 1,6km -resulting in overbank flow to feed the old channel
mentioned above. If this process continues, it is possible that the old channel could be
reactivated.
Backwaters and pools
Remnants of old channels that remain connected to the main channel are visible as
backwaters that are filled by river flow backing up into them. Only during the flood season,
when the active floodplain is inundated, do these experience some through-flow. The depth of
the backwaters will initial y be the same as that in the original channel. With time these
features become filled in so that their depth gradually decreases.
From west to east in Figure 2.7, these are found at the following locations:
A. A long backwater is found at the base of the dry bank in the south west of the map. Its area
is estimated at approximately 50,000 m2.
B. A deep backwater is adjacent to the upstream side of the oxbow loop. This is of particular
interest due to the process of sedimentation that is in progress, which will be discussed
later in this section. The area is estimated at 36,000 m2, and the maximum depth was
measured at 4m.
C. Two remnants of an old channel, which is still partially active, intersect the main channel
just downstream of the oxbow loop. Their combined area is approximately 31,000 m2.
D. East of Blue Zone 1 in Figure 2.7 and within the relict floodplain (Green Zone) there is a
small pool of approximately 7,000 m2. This is also a remnant of an old channel but it is
no longer connected by surface flow. We will refer to this type of feature as a pool rather
than a backwater.
Numerous small pools are scattered about on the active floodplain as well. It has not been
possible to determine how many of these might hold water throughout the dry season.
Sandbanks and sand movement in the main channel
Figure 2.7 and shows large sandy bedforms that are of interest, particularly upstream of the
oxbow loop. These bedforms are the result of sand being moved as "dunes" along the bed of
the river. During the dry season some of these bedforms may be exposed as sand banks, but
they are dynamic features in the main channel constantly changing and moving on.
An interesting sedimentary process is evident in backwater B (mentioned above) adjacent to
the oxbow loop. Here five or six lobes of sand can be clearly seen pushing into the pool from
33
EFA Namibia Geomorphology
the southeast. There is a submerged sandbank on the inside of the bend to the southeast, and
from there, sand overflows into that backwater during the flood season. This is clear visual
evidence of rapid sedimentation in progress.
Slightly further upstream, more extensive submerged sandbanks are seen in Figure 2.7.
These are adjacent to the "reverse fan" mentioned above. These banks are due mainly to the
large loss of water to the floodplain mentioned above, which would reduce the river's carrying
capacity at that point. A slight widening of the river here would also contribute to reduced flow
velocities here.
There are no islands in the main channel within the Kapako study site.
Two alternative scenarios are possible concerning the next stage in channel migration.
In Scenario 1, the oxbow loop would continue to be accentuated by the erosion on the outside
of the bends that is in progress. Ultimately the loop would be cut off and the river course
shortened here. That is the normal progress of such an oxbow loop.
In the alternative, Scenario 2, the large volumes of water that overflow the east bank into the
active floodplain may progress and continue to erode that bank. The old channel that borders
Blue Zone 1 (in Figure 2.7) may thus become reactivated as the main channel due to an
increasing volume of water that it receives.
At the upstream side of the oxbow loop, active sedimentation of backwater B is in progress as
explained above. The loss of water here aggravates the deposition of sediment in the channel.
This could have the effect of aggrading the bed of the main channel, reducing flows through
the oxbow, and favouring reactivation of the old channel to the east.
Whether the oxbow loop will progress, or the old channel be reactivated, is very hard to
predict. A very detailed and precise survey of the whole area using Differential GPS would
help to make an assessment of the most likely trends into the next few decades, but predicting
changes in such a dynamic landscape can never be an exact science.
Erosion and deposition are continually in a state of dynamic equilibrium erosion more-orless
balancing deposition in the floodplain locally with a net transport of sand from upstream
onwards to the Okavango Delta.
Soils on the floodplain
Soils on the floodplain are comprised almost entirely of sand. The river, however does carry
very low concentrations of clay and silt sized particles. These remain suspended in the main
channel, but when flow velocities are sufficiently reduced, these particles are differentially
deposited. For example in reeds or thick grass on the floodplain, where flow velocities are
reduced due to friction, muddy patches can be found. Silt, clay, organic matter and dissolved
minerals enrich the floodplains each year that they are inundated. In time these get mixed with
the sand and the nutrients are leached out and need to be replaced by the next flood. Any
activity upstream that prevents flooding or reduces the extent of floodplain inundation will have
a detrimental effect on the soils of the floodplain.
Rock structures across the channel
Along the section of the Okavango River that forms the border with Angola, linear rock
structures occur at irregular intervals across the river. Elsewhere, these have the appearance
34
EFA Namibia Geomorphology
of igneous dykes, but in fact are made of hard sandstone. These usually form riffles or minor
rapids in the low flow season but are often drowned during the annual flood. Given the high
flow conditions at the time of the field trip in February 2009 it was not possible to locate such
structures in the area of the Kapako site. However, where they occur, these structures limit the
downward erosion of the river bed.
Further research
A detailed land survey with Differential GPS is recommended as a basis for measuring
changes in future. This would need to cover the whole study site, from the top of the dry bank
on one side to the top of the dry bank on the other side. Highly accurate contours at 0,5m
intervals (or better) would be needed. That can only be done in the low-flow season. At the
same time records of the processes evident in the field need to be mapped.
Measurement of sediment transport in the Cubango/Okavango River is recommended. It may
be possible to estimate the sediment transport from the known relationship between transport
and flow velocity on one hand, and historical flow records at Rundu on the other hand.
However the equation relating bedload sediment discharge to flow velocity should be verified
for the flow conditions at or near Kapako.
2.5 Geomorphological Description : Site 5 Popa Falls
2.5.1 Methods
The site visit to Popa Falls was made on 10-11 February 2009.
Both banks were visited on foot e.g. at the pump station for the Correctional Services Farm,
and within the Popa Falls resort, where small side streams were flowing strongly.
A trip was also made by boat upstream to the riffles above the first group of big islands and
along both major channels. Photographs were taken. The water level was high but small areas
of rock at the top of Popa Falls were still exposed. Later photos by Mark Paxton on 14 March
2009 showed that the flood had risen much higher and the rocks at the top of the falls were
completely covered by water, as well as parts of the islands above and below the falls.
Depths were measured from the boat across the two main channels above the falls. The
locations of these depth measurements are shown by the red dots in Figure 2.6. Details are
provided in a report by Celeste Espach.
Other information was provided from the author's experience of the area, which was studied
for the Popa Falls Hydro Power Project (Eco.plan 2003), from the satel ite images larger
scale prints of the same map as Figure 2.9. The areas of individual islands were measured by
a simple squares method. Relevant information from the literature is also summarised in the
following sub-section.
2.5.2 Geomorphological Description & Data for Popa Falls
Figure 2.9 shows part of the study site, with the river flowing from west to east. The sandstone
structure that forms Popa Falls is orientated north-south in the middle of the satellite photo.
GIS generated contours at 1m intervals are provided but they are not accurate because the
small contour interval was not supported by the resolution of the satellite photos. For this
reason making a useful cross section was not possible.
35
EFA Namibia Geomorphology
A digital laser survey with contours at 0,5m intervals was carried out for Water Transfer
Consultants (WTC, 2003) for the Popa Falls Hydro Power project. If the study proceeds to a
greater level of detail it is recommended that these be obtained for the purpose of accurate
contours.
Popa Falls itself is a ridge of hard sandstone that is exposed for almost 1km. Gaps that have
eroded through the ridge allow the water to pass so that the entire ridge is completely
overtopped only during above-average floods. The drop across the falls is only about 4m
during the low flow season and 2,5m during the high flow season. Immediately above and
below them are large islands.
Photos 7 10 are aerial views of the Popa Falls study site in high and low flow seasons (see
photo captions). Photo 11 is a view of the foot of the falls in the low flow season. Photo 12 is
an aerial view of the river between Popa Falls and below Divundu Bridge during the flood
season, illustrating the bedload transport of sediment.
Photos 13-15, taken from a boat on 10 February 2009, offer closer views of the banks, islands
and falls from the river upstream of the falls.
Photos 9 and 15 show that the islands have a base of bedrock. Other photos by Mark Paxton
during low flow conditions in October and November 2008 confirm this observation (not
included in this report).
The original riverine vegetation was Riverine Forest, which is now scarce along the Okavango
River in Namibia. At Site 5, remnants of this habitat are found on the south bank around the
Popa Falls wildlife resort and around lodges below the falls. On the north bank more extensive
woodlands occur as far as the Divundu Bridge. A small Community Campsite exists close to
the falls but the small clearings have made for little disturbance there.
Elsewhere, the banks have been considerably disturbed by cultivation and grazing. Cattle
have even grazed most of the fringing reedbeds adjacent to fields. North of the falls a farming
project operated by the Department of Correctional Services can be seen on the aerial photos.
36
EFA Namibia Geomorphology
Figure 2. 9: Popa Falls Map with Contours
37
EFA Namibia Geomorphology
Photos 3: 7-9
38
EFA Namibia Geomorphology
Photos 4: 10-12
39
EFA Namibia Geomorphology
Photos 5: 13-15
40
EFA Namibia Geomorphology
Soils outside the dry banks are the typical reddish Kalahari sands. The dry banks support
Kalahari woodlands where they have not been cleared for agriculture. Only on the north side,
below the fal s, there is an old floodplain where exposure to water has reduced the iron and
the soil is whitish in colour (refer Photo10).
Upstream from the falls:
Channels
Because of the impounding effect of the sandstone ridge that creates the falls, the river
channels are at their deepest just upstream of the falls, even in the low flow season. Because
the river is contained by fairly steep dry banks, the lateral spread of water between low and
high flow conditions varies little upstream from the falls.
Above the falls there are two main channels and a small distributary channel that flows
adjacent to the Popa Falls wildlife resort on the south bank. The widths of these three
channels from north to south are typically 50m, 100m, and 30m respectively. They all
converge just above the falls.
Depth measurements were made on 10 February 2009 across the two larger channels at the
locations indicated by the red dots in Figure 2.6. Depths of up to 4.5m were measured in the
largest (middle) channel, and of only 2.2m in the northern channel.
An additional, much smaller channel also exists at Popa Falls wildlife resort but it is concealed
on Figure 2.6 by overhanging trees. This small side-stream has much steeper gradients and
flows over rocky substrates with small rapids, joining the river again only below the falls.
There are no big pools in the Popa Falls study site. A few small pools, alternating with rapids,
occur on the small side-stream through the wildlife resort that was mentioned above. There
are also no real backwaters all the channels are active.
Levels are very difficult to determine from the inaccurate contours on the orthophoto. However,
the (low flow) water level was at approximately 1002 1003 m.a.s.l. Upstream of the falls, the
tree line on the north bank suggests that the dry banks are at about 1007 1008 m.a.s.l. i.e.
about 5m above the low flow level. The dry banks rise to 1020 m.a.s.l. within 400m on the
south side and reach that level within 800m on the north side.
Islands
The islands above the falls are very stable for two reasons. Firstly they have a base of hard
sandstone, with sand on top. Secondly, they are stabilised by vegetation trees in the interior
and sometimes on the margins, while reeds and/or papyrus stabilise the margins of the
islands.
The four large islands upstream from the falls measure, in approximate area, 23.0ha, 3.1ha,
1.9ha, and 7.0ha. These areas of the islands are listed in sequence from north to south
respectively.
41
EFA Namibia Geomorphology
Floodplains
Unlike Kapako, there is very little area that can be called floodplain in this study site. The
exceptions are a few small areas of bright green floodplains in Photos 7, 8 & 9 that are more
or less permanently wet. These are covered in papyrus, or reeds and papyrus. There are no
grassy floodplains here. The tree line suggests an upper limit of historical flooding.
Conditions above the falls appear to be in a state of equilibrium. No evidence was found of
progressive (net) erosion or deposition in this area. The rate of change at the study site is
slow, compared to Kapako.
Downstream from the falls:
Islands
The sandstone ridge that forms Popa Falls also forms the base of islands immediately below
the falls. Channels, incised into the sandstone, separate a number of islands. Sand has been
deposited on top of these islands and is stabilised by vegetation, including reeds and trees.
The three largest islands immediately below the falls measure approximately 8.2ha, 5.8ha,
and 7.0ha.
Channels, Sandbars & Floodplains
The channels below the falls are less well defined than upstream as the deposition of sand
results in constantly changing bedforms (see Photo 10).
Extensive sandbanks form below the falls as a result of the drop in flow velocity. In this area
these sandbanks are seldom exposed. However, more extensive banks form downstream,
where they become more exposed in the low flow season and are important for breeding
African skimmers.
Below the falls, the river is contained on its south bank by wooded banks a few metres high,
but the north bank has small areas of both active and passive floodplains. In the foreground of
Photo 10, the active floodplain is indicated by the area that is obviously wet, without bush or
trees. A slightly higher floodplain adjacent to that is indicated by whitish soil with scattered
bush and trees. The fact that the trees there are still small suggests that a major flood event
removed larger trees at some stage in the last decade or two. The dry bank is indicated by a
clear transition from white soil and sparse trees to reddish soil and densely spaced trees
next to the wing strut in Photo 10. The area of this floodplain was not determined because it is
omitted from the satellite images that were supplied.
Sediment transport
The Okavango River is well known for the unusual nature of its sediment load. It is
characterised by low dissolved solids, low turbidity (even in the flood season) and bedload
transport of most of its solid load. Photo 10 shows how clear the water is at low flow, and
Photo 12 shows the river bed some 3 4m deep during an average flood.
42
EFA Namibia Geomorphology
McCarthy and Ellery (1998) studied the load carried by the River into the Delta at Mohembo.
They estimated the following figures, which can be assumed to be the same as for Popa Falls
as there are no significant inflows between Popa and Mohembo:
a. Total dissolved solute load: 380,000 tonnes per year,
b. Total bedload transport: 170,000 tonnes per year,
c. Total suspended sediment: 39,000 tonnes per year.
The dissolved load is of little importance to geomorphology in Namibia and it simply passes
through the system. (It does become important in landforms in the Delta, however).
The low level of suspended load is very unusual for such a large river, consisting mainly of
clay-sized particles, with some organic matter. A little of the clay and mud sized particles gets
trapped in reedbeds, where muddy patches can be found. However there is no major
deposition of clay or silt-sized particles in progress in the Popa Falls study area.
By far the major part of the solid load is fine Kalahari sand in the range 0.25 0.43mm.
Particles of this size are too large to be carried in suspension. As evidence of this, Photo 12
shows sand moving as "dunes" on the bed of the river during an average flood (April 2003).
Where flow velocities increase locally around rocks, the sand is stirred up and carried in
suspension for about 50 before settling out again and resuming its progress as bedload.
McCarthy and Ellery (1998) measured bedload transport and estimated a total transport of
170,000 tonnes entering the Delta at Mohembo. This estimate was later revised by McCarthy
(2003) during a study for the Popa Falls hydro power project, where longer term hydrological
records were used.
As part of the Premilinary Environmental Assessment for the proposed hydro power project at
Popa Falls (Eco.plan 2003), McCarthy (2003) undertook a specialist study and measurement
of sediment transport from 24 to 27 April 2003 close to the peak flow period for that year.
The site used was approximately 300m downstream from the Divundu bridge and
approximately 4,4km upstream from Popa Falls. This was a relatively straight section of river,
with few rocks and an active channel width of 152m. The depth across the channel at the time
varied from about 3m to 4m, with an average depth of 3.4m. The bedload sediment was
comprised mostly of fine sand with particle size in the narrow range 0.25 0.43mm.
Suspended sediment was at a very low level, such that visibility was 2 3 metres below the
surface of the river. Flow velocity was measured at intervals and depths across the channel
and found to average 0.60 m3/s across the entire active channel. The discharge was
calculated at 313 m3/s.
McCarthy, Stanisstreet and Cairncross (1991) had previously established a relationship
between flow velocity and bedload discharge of sediment, which was further refined by
McCarthy, Ellery, and Stanistreet (1992). As mentioned under the section on Kapako, above,
this relationship is defined by the equation:
Qs = 0.154U3.40 where Qs is bedload discharge per unit width (kg/m/s) and U is the flow
velocity (m/s).
McCarthy (2003) found that this relationship held for the study upstream from Popa Falls as
well. He used 30-year historical flow data provided by the Namibian Directorate of Water
Affairs to calculate the annual bedload sediment discharge. The result was a revised estimate
of:
43
EFA Namibia Geomorphology
Total bedload transport: 117,000 tonnes (or about 70,000 m3)
Bedload sediment transport is notoriously difficult to measure reliably. Yuquian (1989) noted
that the Helley-Smith bedload sampler (used by McCarthy) is one of the better methods to
achieve this, but recommended that the instrument be calibrated in situ. In order to gain an
independent measurement, Eco.plan (2003) commissioned the Marine GeoScience Unit,
Council for GeoScience, South Africa to undertake separate measurements using side-scan
sonar and high resolution bathymetry. Coles (2003) presented the results. The method
enabled the bedforms to be imaged, and measured to an accuracy of 2cm. Three sets of data
were made over a period of 28 hours. Then computer programmes were used to compare
these data sets and arrive at a measurement of sediment transport. The measured transport
was 49.94 m3/day during the measurement period 25 to 27 April 2003. However, the main
disadvantage of this method is that it can only measure transport in that part of the river where
large bedforms exist. The data had to be interpolated across the rest of the channel, where
bedforms were not prominent. The resulting estimate was 113m3/day for the river discharge at
the time. In the end, the limitations of this method proved to be a significant disadvantage in
measuring sediment transport, but it did provide an accurate bathymetric survey of the
bedforms. For comparison, Mc Carthy (2003) measured 197m3/day but this was over the full
width of the active channel.
The deep channels immediately upstream of the falls are thought to be the result of an
increased hydraulic gradient there so that the bed is locally scoured by turbulence during high
flows. The main channels reached depths of 4.5m in February, 2003. Below the falls, however,
the hydraulic gradient drops and deposition occurs forming the large bedforms as explained
above.
Where bedforms can be seen (e.g. below the falls in Photo 10, and upstream in Photo 12)
organic sediment (dark material appearing like shadows) collects on the downstream side of
bedforms.
Both bedload and suspended sediment appear to be in a state of dynamic equilibrium in the
Popa Falls area. Erosion of sediment is balanced by deposition, while erosion of bedrock is
very slow due to the hard nature of the sandstone.
2.6 Habitat integrity of the sites in Namibia
The geomorphological features of both the Kapako and Popa Falls sites have been very little
affected by human activities to date. One exception may be the relict floodplains at Kapako
where cultivation and livestock has probably accelerated the natural process of flattening of
the original scroll bars due to rainfall, wind and soil organisms.
Vegetation, however will have been modified as a result of cultivation on the higher relict
floodplains, and cutting of firewood and timber (stands of larger trees may have been more
common years ago).
The river and floodplains still support a wide diversity of bird species including some red data
species (Paxton, pers comm.), and populations appear to be in a relatively good state
considering that there is a substantial human population living adjacent to the floodplains.
44
EFA Namibia Geomorphology
3 IDENTIFICATION OF INDICATORS AND FLOW CATEGORIES
3.1 Indicators
3.1.1 Introduction
Biophysical indicators are discipline-specific attributes of the river system that respond to a
change in river flow by changing in their:
abundance;
concentration; or
extent (area).
Social indicators are attributes of the social structures linked to the river that respond to
changes in the availability of riverine resources (as described by the biophysical indicators).
The indicators are used to characterise the current situation and changes that could occur
with development-driven flow changes.
Within any one biophysical discipline, key attributes can be grouped if they are expected to
respond in the same way to the flow regime of the river. By example, fish species that all
move on to floodplains at about the same time and for the same kinds of breeding or feeding
reasons could be grouped as Fish Guild X.
3.1.2 Indicator list for Geomorphology
In order to cover the major characteristics of the river system and its users many indicators
may be deemed necessary. For any one EF site, however, the number of indicators is
limited to ten (or fewer) in order to make the process manageable. The full list of indicators
was developed collaboratively by the country representatives for the discipline in Namibia
this was done by Colin Christian with input from Sophie Simmonds and Mark Rountree - and
is provided in Table 3.1 below.
Indicat
Indicator name
Sites represented no more
or
than ten indicators per site
Numbe
r
Kapako 4
Popa Falls 5
1
Extent of exposed rocky habitat in main
No
Yes
channels.
2
Extent of coarse sediments on the bed
No
No
3
Cross sectional area of bank full channel
Yes
Yes
4
Extent of backwater areas (slow/no flow
Yes
No
areas)
5
Extent of exposed sandbars at low flow
No
Yes
6
Extent of vegetated islands
No
Yes
7
Percentage silt & clay in the top 30cm of the
Yes
No
floodplain
8
Extent of the floodplain flooded each wet
Yes
No
season
9
Extent of inundated pools/pans on floodplain
Yes
No
at the end of the dry season
10
Extent of cut banks along the active channel
Yes
No
Table 3. 1: List of indicators
3.1.3 Description and location of indicators
45
EFA Namibia Geomorphology
(Geomorphology) Indicator 1
Name: Extent of exposed rocky habitat in main channels
Description: Along the length of the Namibian reach of the river, there are numerous rocky
structures that cross the river. They are seldom or never exposed on floodplains. These low
rocky ridges form riffles or rapids in the channels.
At the Kapako site a minor riffle suggested such a feature on the east side of the oxbow loop
during the high flow field trip in February 2009. However, this was not considered a
significant feature and was further ignored.
Popa Falls is a very prominent exposure of one of these rocky ridges, accentuated by
faulting. Smaller ridges occur both above and below the falls, where they form riffles and are
visible on aerial photos, at least in low flow conditions. Rock pratincoles use this habitat for
breeding in the low flow season, and for feeding for as long as the water is low enough.
Flow-related location: The rocky sandstone outcrops have similar directional trends e.g.
north-east to south-west. This suggests that they are the product of geological processes on
a large scale. It is therefore reasonable to assume that they underlie the floodplains and
Kalahari dunes as well. In the main channels they are often covered by water during the high
flow season, but partially exposed during the low flow season. In small side channels they
may be seasonally covered by water or sediment.
Known water needs: Low flow seasons are necessary to expose these rocky structures e.g.
for nesting Pratincoles. Medium and high flow seasons are necessary to inundate large areas
of rock for certain species of catfish that prefer rapids.
(Geomorphology) Indicator 2
Name: Extent of coarse sediments on the bed
Description:
The sediment in the Namibian reach of the river is typically fine sand of aeolian origin
(Kalahari sand) that has been reworked by water. The particle size range is 0.25 to 0.4mm.
Coarser sediments are almost entirely absent.
Occasional large pebbles may occur where
they have become lodged in holes in the river bed, but we have not found any place where
these form a different habitat from the usual sandy bed. Coarse sediments were therefore
rejected as a potential indicator in the Namibian reaches. They may occur only in the
headwaters in Angola.
(Geomorphology) Indicator 3
Name: Cross sectional area of bank full channel
Description: The cross sectional area of the active channels between clearly defined banks
that mark the edge of the floodplain, or in some cases the edge of dry banks. For most of the
Namibian reaches of the river there is a single channel about 150m wide and an estimated
maximum depth of some 4 metres at bank full capacity, shallowing towards the inside of the
bends.
Flow related location: Edges of clearly defined main channels.
46
EFA Namibia Geomorphology
Known water needs: Channel cross section is a function of high flow conditions. If peak flows
are regularly increased, channel enlargement can be expected. This can happen rapidly if
the peak flows exceed the historical maxima. If peak flows are regularly decreased, for
example, due to dams upstream, the channel will gradually get narrower, or a misfit river will
form. Vegetation (e.g. reeds) will probably stabilise the banks and sandbanks. During the
(reduced) high flow seasons, sediment will get trapped by this vegetation, thus building up
part of the bed of the river. In time the channel will get shallower and narrower.
(Geomorphology) Indicator 4
Name: Extent of backwater areas (slow / no flow areas connected to the main channels)
Description: The backwaters are sections of old channels that are directly connected to the
main channel.
As water levels in the main channel rise and fall, the backwaters respond immediately in the
same way. The backwaters are usually quite wel defined at Kapako, and their slopes being
originally river banks are fairly steep. Therefore, the area of the backwaters does not
increase a lot when the floods are high.
(Note that there are significant backwater pools in bedrock in the Mukwe-Andara-Divundu
section of the river and some of minor lesser extent in the side channels just upstream of Popa
Falls. However, these are excluded from our definition here.)
Known water needs: Since the backwaters are, by our definition, connected to the main
channels, this connection needs to be maintained. If that connection is blocked, the
backwaters would become pools as defined under indicator No.9, below.
(Geomorphology) Indicator 5
Name: Extent of exposed sandbars at low flow
Description: Sandbars are sometimes exposed in the river channels during the low flow
season. The extent of such sandbars could increase or decrease in response to changes in
flow of water and sediment discharge.
Flow related location: Sandbars form where a reduction in flow velocity occurs for some
reason. Most often they form on the inside of bends, but extensive bars also occur
immediately below Popa Falls and downstream near Mahangu. These are exposed during the
low-flow season. They are important for breeding African skimmers in the low flow season.
The high season field trip at Kapako indicated that the sandbars are inundated at high flows,
but aerial photos show some exposed sand bars during low flow conditions.
If sediment is impounded upstream, e.g. by a dam or weir, but the flow of water is unaltered,
then erosion of the river bed and loss of sand bars can be expected.
Known water needs: Sandbars are a function of the complex interaction of water flows and
sediment transport. It would be impossible to make predictions based on changes in water
flow alone. If sediment is impounded without a change in water flow, the sediment-hungry river
will pick up sediment until its carrying capacity is reached. The extent of erosion downstream
depends on distance, flow velocity, slope and the amount of sediment available downstream
from the impoundment. If water flow increases without an increase in the supply of sediment,
then erosion can be expected downstream, resulting in the removal of sandbars. Conversely, if
water flows decrease without the supply of sediment decreasing, deposition will occur
downstream of the perturbation and sandbars can be expected to grow.
NB. For the purposes of generating the response curves, the supply of sediment was ignored
because the model was unable to take that factor into account independently of the flow of
water. The response curves therefore do not reflect the impacts of sediment-trapping dams or
47
EFA Namibia Geomorphology
weirs. This is a significant shortcoming of the approach that would need to be addressed in
future studies.
(Geomorphology) Indicator 6
Name: Extent of vegetated islands
Description: In areas dominated by depositional features, such as extensive floodplains,
islands are usually comprised only of sediment (mainly fine sand). However, the islands
immediately above and below Popa Falls are comprised of sand on a base of bedrock. (More
extensive islands of this type occur in the Mukwe-Andara-Divundu section of the river, which
are not considered here.)
Flow related location: Islands, whether all-sand or sand-on-bedrock occur in the main channel.
They are, however absent from the Kapako site.
Known water needs: Islands that consist entirely of sand probably started as sandbanks that
were shal ow enough for long enough so that vegetation such as reeds could become
established on them. Thereafter, their elevation could be increased due to floods that
deposited sand on top of the island where it was trapped by vegetation. Thus the island grew
in elevation, and possibly in area as well. All-sand islands may need normal flooding in order
to be maintained as islands. Conversely, excessively high floods would erode such islands,
especial y if the high levels are maintained for longer than normal.
The origin of sand-on-bedrock islands is less clear.
They may have formed in a similar way to that described above. Alternatively, they may be
remnants of dunes that were colonised by riverine woody vegetation and never eroded. They
appear to be higher relative to water levels and they support woodlands, which suggests that
they are not fully inundated by floods. These islands would be at risk of erosion if peak flows
are significantly increased above historical levels. If the islands were partially submerged, e.g.
in an impoundment, then the margins of the islands would also be eroded due to water
loosening the sand and possibly due to die-off of some less hygrophilous trees along the
margins.
Reduction in flow should not affect the sand-on-rock islands, as long as there was enough
moisture from the river to sustain the vegetation. In the case of all-sand islands reduction in
flow consistently in high and low flow seasons can be expected to result in island growth.
(Geomorphology) Indicator 7
Name: Percentage of silts and clays in the top 30cm of the floodplain on active
floodplains
Description: An increase or decrease in the percentage of clay and silt-sized particles in the
top 30cm of soil on the floodplains. The floodplains are comprised mainly of sand in the
particle size range approximately 0.25 0.4mm. Clays (< 2µ) and silt (2 -53µ) also occur, but
in very low concentrations in the Okavango River. Some of this fine material is deposited on
the floodplains during overbank flooding. Despite low concentrations, the volumes of clay and
silt can be considerable due to the large volume of water moving slowly over a floodplain. If
flooding does not occur, or is reduced in extent, then one could expect a gradual loss of these
fine particles in the top 30cm as this material is mixed in by cattle, soil organisms, or blown
away by wind in the dry season.
Flow related location: The fine sand that comprises most of the River's solid load moves by the
process of saltation and is not truly suspended in the water column. On the other hand, the
finer clay and silt-sized particles are easily suspended in the water column. For them to be
48
EFA Namibia Geomorphology
deposited requires a substantial reduction in flow velocity. This does not occur in the main
channel except where eddies occur in pockets or reedbeds along the channel margins.
However, as the water spreads out over the floodplains, its flow velocity drops and vegetation
such as grasses help to trap the clay particles.
Known water needs: It is assumed that the percentage of silts and clays in the top 30cm of soil
on the floodplains will decrease during the dry season as it is mixed downwards by soil
organisms or blown away by wind. They would only be replenished during subsequent
overbank flooding onto the floodplains.
(Geomorphology) Indicator 8
Name: Extent of the floodplain inundated during each wet season
At Kapako there are extensive floodplain areas, which can be divided into two zones. The
active floodplains display distinct scroll bars and are understood to have been flooded every
year for which flow records exist. There are also extensive floodplains at a level of 0.5m -1m
higher than the active floodplains. These higher relict floodplains do not display scroll bars,
which suggests that they are no longer flooded, with the possible exception of the highest
floods on record. Higher floodplains are also common elsewhere along the River in Namibia.
They could represent the result of tectonic processes, or reduced sediment yield in the
catchment for a period that resulted in downcutting of the river by 0.5m 1m.
The following discussion relates to the active floodplains. The most extensive one lies to the
south of the oxbow loop. Here the floodplain drops away from the present channel by
approximately 2m over about 1.3km to an older channel. The river is contained by levees
along its banks. As the flood rises and overtops those levees, the water flows over almost the
entire floodplain, with the exception of a few higher localities.
On the floodplains at Kapako, it is assumed that most of the active floodplain will be inundated
once the levees overtop. More accurate contour maps would be needed to determine the
exact extent of inundation for any given flow scenario.
There are no significant floodplain areas at Popa Falls, only a few small areas covered by
reeds or papyrus.
(Geomorphology) Indicator 9
Name: Extent of inundated pools / pans on floodplains
Description: An increase or decrease in the extent of inundated pools on the floodplain at the
end of the dry season.
Flow related location: Perennial pools occur mainly on the active floodplain, where they are
understood as remnants of old river channels. Small pools may be found between scroll bars,
or large pools found in old channels that have been abandoned by the river. In the case of the
latter, the width of the old river channel is reflected in the width of the pool. On active
floodplains, these pools fill with water as a result of river level rising and spilling over onto the
floodplains. It is assumed that the pools contain some water throughout the dry season due to
the permeability of the fine sand, allowing ingress of groundwater from the river channel into
the floodplain throughout the year. This would need to be confirmed by accurately surveying
49
EFA Namibia Geomorphology
the surface levels of pools in relation to the river stage at any time in the low flow season, and
doing groundwater surveys.
In the case of pools on higher and older relic floodplains filling is more likely to be from rainfall,
but it is not known whether these pools persist throughout the dry season. Even if infiltration is
impeded by a clay or calcrete horizon, the pools are likely to dry out due to high evaporation
rates.
Such pools do not occur at Popa Falls due to the absence of significant floodplain areas.
Known water needs: If the assumption is correct that the perennial pools on the active
floodplain are maintained by seepage, then their levels will be related to the level of the river
(probably lower than the river level due to evaporation). However, it may be that the pools
need to be filled during overbank flooding, and that they then retain water due to an
impermeable lining of fine material.
To prevent stagnation, annual flushing of pools by flooding is probably desirable. However,
flooding also means that sediment is deposited in these pools, gradually reducing their depth.
In time all pools will be in-filled, but others will have formed where the river has meandered.
The filling in of pools and creation of new ox-bows and cut-off meanders is likely to occur over
a longer timescale than is considered for this study.
(Geomorphology) Indicator 10
Name: Extent of cut banks along the active channel
Description: Active erosion of banks produces steep (near-vertical) banks, which gradually
retreat. The horizontal length of such cut banks is the proposed indicator of increased erosion.
Flow related location: Bank erosion occurs on the outside of bends, where the flow velocity is
higher. Tighter bends are more prone to bank erosion. Cut banks occur where there is no rock
control present. There are cut banks in several places at Kapako. However, at Popa Falls, the
banks are well protected by vegetation reeds, papyrus or trees, so that bank cutting is rare.
Known water needs: Bank erosion is primarily a function of high flow periods. Bank collapse
may also occur during the Transition stage 2, if the water level drops rapidly. This is because
the sandy banks are infiltrated with water, and as the water level drops rapidly, the water
flowing out at the base of the bank combined with the saturated weight of the bank soil causes
bank failure and slumping.
50
EFA Namibia Geomorphology
3.2 Flow categories river sites
One of the main assumptions underlying the EF process to be used in the TDA is that it is
possible to identify parts of the flow regime that are ecologically relevant in different ways and
to describe their nature using the historical hydrological record. Thus, one of the first steps in
the EFA process, for any river, is to consult with local river ecologists to identify these
ecologically most important flow categories. This process was followed at the Preparation
Workshop in September 2008 and four flow categories were agreed on for the Okavango
Basin river sites:
Dry season Transitional Season 1 Flood Season Transitional Season 2.
Tentative seasonal divisions for river Sites 4 and 5 are shown in Figure 3.1 and 3.2. These
seasonal divisions were formalised by the project hydrological team in the form of hydrological
rules in the hydrological model. In the interim they provide useful insights into the flow regime
of the river system. Figure 3.1 shows the flow of the Cubango / Okavango River at Rundu (just
downstream from Kapako). Figure 3.2 shows the flow of the Okavango River downstream from
the confluence with its major tributary the Cuito River.
1000
900
Wet
We
800
Dry
Dr
Tra
Tr n
a s
n 1
s
Tra
Tr n
a s
n 2
s
Dry
Dr
700
600
Year 1
500
Year 2
400
Year 3
300
200
100
0
O
N
D
J
F
M
A
M
J
J
M
J
A
S
Figure 3. 1:Three representative years for Site 4: Okavango River at Kapako (hydrological data
from Rundu), illustrating the approximate division of the flow regime into four flow
seasons
51
EFA Namibia Geomorphology
1800
180
1600
160
We
W t
e
1400
140
Dr
D y
r
Tra
Tr n
a s
n 1
s
Tra
Tr n
a s
n 2
s
Dr
D y
r
1200
120
1000
Yea
Ye r
a 3
1000
r
Yea
Ye r
a 2
r
800
80
Yea
Ye r
a 1
r
600
60
400
40
200
20
0
O
N
D
J
F
M
A
M
J
J
M
J
A
S
Figure 3. 2:Three representative years for Site 5: Okavango River at Popa (hydrological data
from Mukwe), illustrating the approximate division of the flow regime into four flow
seasons
Table 3.2 poses the nine main questions that are initially expected to be related to these
flow seasons. These questions were considered in the Knowledge Capture Workshop in
Windhoek in March/April 2009 and where relevant, the answers were provided in the form of
completed response curves (not included in this report).
This exercise was carried out for each indicator, first at Kapako (Site 4) and then at Popa Falls
(Site 5).
Question
Season
Response of Indicator 1... if:
number
1
Onset is earlier or later than natural mode/average.
Dry
2 Season
Water levels are higher or lower than natural mode/average.
3
Extends longer than natural mode/average
Transition
Duration is longer or shorter than natural mode/average i.e. hydrograph is steeper
4 1
or shallower.
5
Flows are more or less variable than natural mode/average and range.
Flood
Onset is earlier or later than natural mode/average synchronisation with rain may
6 season
be changed.
7
Natural proportion of different types of flood year changed.
Transition
8 2
Onset is earlier or later than natural mode/average.
Duration is longer or shorter than natural mode/average i.e. hydrograph is steeper
9
or shallower.
Table 3. 2: Questions addressed at the Knowledge Capture Workshop, per indicator per site. In
all cases, `natural' embraces the full range of natural variability
52
EFA Namibia Geomorphology
4 LITERATURE REVIEW
4.1 Introduction
The literature review for this report was intended to concentrate mainly on the identified
indicators and relate mainly to the Namibian section of the River. In fact, none of the
literature found relates directly to the indicators. General references were found that relate to
the nature of sediments, and sediment transport rates but only for the river below the
confluence with the Cuito River.
The literature search for this report was conducted by various means:
a. Discussions with an expert in geomorphology and sedimentology relating to
the Okavango Basin -Prof T. McCarthy of WITS University,
b. Review of books on the Okavango River by Mendelsohn and el Obeid
(2004) and Mendelsohn (2006),
c. Review of papers of the Okavango Research Group, WITS University,
d. Searches in Libraries in Windhoek, especially the DRFN library,
e. Internet web search,
f.
Review of original field data relating to soils.
The results turned up very little of direct relevance to this component report. No detailed
studies were found on the subject of geomorphology on the Namibian portion of Okavango
River.
4.2 Literature on Geomorphology and Sedimentology
Mendelsohn et al (2002), Mendelsohn & el Obeid (2004) provide very useful general data on
the flows of the various river tributaries, climatic information (e.g. regarding evaporation
relative to rainfall), maps, diagrams and photographs. However there is no specific information
available on geomorphology or the processes that give rise to the landforms observed along
the river.
There is a great deal of literature available on the geomorphology of the Okavango Delta and
panhandle, some of which is applicable to the lower reaches of the river in Namibia. The work
of the Okavango Research Group at the University of Witwatersrand (WITS) is particularly
valuable in this regard. Their research papers have been published by WITS in seven volumes
(ORG, 1986 to 2005). It is expected that the Botswana team for the TDA and EFS will cover
this literature, so only that which is of specific relevance to the Namibian reaches of the river
will be covered here.
McCarthy, Green and Franey (1993), McCarthy et al (1997), Gumbricht, McCarthy and Merry
(2001), and McCarthy et al (2002) have described the tectonic processes and faulting affecting
the Delta and Panhandle, which explains the existence of the inland Okavango Delta.
McCarthy, Ellery & Dangerfield (1998) have described the role of trees on islands in removing
dissolved load from the waters of the Delta, concentrating this load in islands as a result of
transpiration processes, and the resulting growth in small islands. This may be of some
relevance in the margins of the floodplains in Namibia at the foot of the dry banks, where
calcrete is formed, but there is no research available on this. Islands in the river, for instance
near Andara, and above and below Popa Falls, appear to be built of sand on a base of
53
EFA Namibia Geomorphology
bedrock, but no studies have been found that investigate the possible concentration of salts
here.
The author's experience in the field has found a hint of such processes in the form of slightly
white accretions of salts on floodplains. The most striking example is below Rundu, where a
plantation of Eucalyptus trees exists (Eco.plan, 2001). The trees are some 20 years old but
are not growing as well as this species normally does. A clue to the reason for this is found
around the bases of these trees. Here white accretions of salts are found evidence of strong
transpirative action that concentrates carbonates and silicates in the soil around the trees.
Eucalyptus trees are known for high rates of transpiration. Whether trees and transpiration
may play a role on some of the islands near Andara and Popa Falls is unknown.
The likely effects of climate change in the Kavango Region and near regions in neighbouring
countries has been modeled and reported in BBC News / Science-Nature / African sands `set
for upheaval' (30 June 2005).
The most recent research on the Namibian portion of the river was carried out for the Popa
Falls Hydro Power Project Environmental Assessment. (Eco.plan, 2003). The appendices
include three specialist reports that relate to geomorphology and sediment transport. McCarthy
(2003) gave an account of the measurement and estimation of sediment transport at Divundu,
about 4km upstream from Popa Falls. McCarthy (2003) provided a very useful, concise
account of the ecological importance of sediment transport for the Delta. This report drew on
work done by the McCarthy, Ellery and others in the Okavango Research Group, WITS
University (ORG, 1986 to 2005).
McCarthy & Ellery (1998) measured sediment transport at Mohembo and estimated that on
average 170,000 tonnes of bedload sediment (fine sand in the range 0.25 0.4mm) was
transported into the Delta each year. The study also estimated that about 39,000 tonnes of
suspended sediment (e.g. clay, mud and organic material) and about 380,000 tonnes of
dissolved load entered the Delta each year.
McCarthy (2003), on the basis of work done for the Popa Falls Hydro Power EIA, revised this
estimate of bedload transport down to 117,000 tonnes (or about 70,000 m3).
No other literature was found that has particular relevance to the geomorphology on the
Namibian portion of the Okavango River.
Some general references were found on the Internet, which may help to provide some
understanding of the subject, but applicability may be limited.
54
EFA Namibia Geomorphology
4.3 Information on Soils
4.3.1 Soils in relation to geomorphological units and vegetation
A brief description of soils is given by CCA (2007) for the Kavango Biofuel project proposal.
The soils along the Namibian section of the River are very weakly developed and consist
mainly of fine sand. They are all derived from Kalahari sands, which were aeolian deposits
during a much drier palaeo-climate. This sand is comprised mainly of quartz and is very low in
nutrients. In the current wetter era, some of that material was re-worked by the River, while
vegetation colonized the sand dunes and stabilized them.
Outside of the dry banks of the River, the Kalahari dunes are/were covered dry Kalahari
Woodland. This woodland receives its water by rainfall only and is drought-deciduous. The fine
sand here is often reddish in colour due to coating of the sand grains with a little iron oxide
(refer Photos 16 & 17).
55
EFA Namibia Geomorphology
Photos 6: 16-18
56
EFA Namibia Geomorphology
Photos 7: 19-21
57
EFA Namibia Geomorphology
On a narrow margin along the river banks and islands, evergreen Riverine Forest was
dominant, but has been largely cleared by people. The most intact section of this habitat lies in
the Mukwe-Andara-Popa Falls section of the River. This type of forest derives much of its
water from the river and seepage so that evergreen species are able to survive. Close to the
river and on the floodplains, the soil is whitish the iron oxide having been reduced as a result
of contact with water (Photo 17).
Soil in woodlands may also be whitish in colour. Here the loss of iron oxides is assumed to be
the result of organic acids produced by plants and detritus in the soil (Photo 18).
On floodplains and omurambas, where trees were not able to grow due to flooding, various
grasses and hygrophilous grasses dominate (Photo 19). Omurambas, where water stands for
up to a few weeks during the wet season, also do not permit the growth of trees (Photo 20).
Mixed with the fine Kalahari Sand are small amounts of clay that were contributed by wind or,
close to the River, by over-bank flooding. Within the dry woodlands and riverine forests the
nutrients in the soil are continually being recycled by plants and animals. Nutrients are
returned to the soil by leaf litter, droppings of animals, birds and insects, and fires (in dry
woodland). Organic content also helps to maintain the fertility of the soil.
Where indigenous vegetation has been cleared for agriculture, the fertility rapidly declines.
Nutrients are leached out of the soil or taken up by crops, and clay particles are blown away
so that fields are abandoned after a few years of cultivation (Photo 21). In the case of the
limited irrigation projects, artificial fertilizers are essential to maintaining the fertility of the
irrigated soils.
Mendelsohn & el Obeid (2004, p.74) stated that all the water in the Namibian reaches of the
River comes from Angola. The rainfall in Namibia contributes virtual y no runoff to the river.
This is true also for the "fossil tributary" the Omatako, which barely reaches the mainstem
river. Mendelsohn (2006, p.27) stated that "There is also little surface run-off or erosion of
these porous soils. Sand, or more correctly grains of quartz, makes up the bulk of the soil,
which contains limited humus or organic matter...". It therefore follows that sediment yield to
the river from the Namibian side will be almost negligible, under any development scenario.
Wind erosion of finer particles from cleared lands may contribute slightly to a decline in water
quality and clarity, but not to a significantly increased solid load. Fine sand comprises the bulk
of the river's solid load. Thus under present climatic conditions, changes in land use in
Namibia are not expected to significantly alter the sediment loading in the river.
However, regional model ing of the effects of climate change has predicted that the Kalahari
dunes that are now vegetated could become re-mobilised within several decades (BBC News /
Science-Nature / African sands `set for upheaval' (30 June 2005)). In a scenario where the
Okavango Basin gets warmer and drier, woodlands may die off leaving the dunes exposed to
winds. Concurrent increases in wind speeds would accompany such change resulting in
mobilization of Kalahari dunes that would surely encroach on the river's course. Climate
change therefore represents a "wild card" that will have serious impacts but the exact nature,
extent and timing of those impacts are not yet sufficiently understood.
Excerpts from the soil and vegetation study (Simmonds & Burke, 2001) which contributed to
the Kavango Region Environmental Profile provide the following insights on the soils,
landforms and vegetation of the Okavango floodplain and terrace system.
Genetic Inheritance
The majority of soils in Kavango have formed on either sandy or loamy substrates. The older
Tertiary parent materials are completely recycled eroded and re-deposited. As a result the
58
EFA Namibia Geomorphology
soils developed on these substrates have not inherited the end-products of in situ weathering
and individual soil particles are therefore not commonly coated by oxidized iron. The only soils
in which oxidized iron coatings would give the soils a red colour are the younger red aeolian
sands blown in on top of the older Tertiary aeolian deposits, and the soils developed on top of
the alluvial deposits of the Okavango river terraces.
Between these two soil settings the genetic chemical conditions favouring iron coatings are
very different. In the case of the younger red aeolian sands, it is possible that previous
weathering conditions did not completely remove the iron pigment which consequently
remains coated on quartz grains giving a pale pink to red colouration. In the case of soils on
alluvial terraces, the immature sediments forming the parent material would have originated in
humid paleo-environmental conditions of central Angola and provide the source of iron.
Furthermore, the mixture of materials comprising the alluvial deposits of the Okavango river
terraces derive their bases from the same weathered origins in central Angola.
Fluvisols: Soils of Alluvial Lowlands
FL FLUVISOLS
FLd Dystric Fluvisols
FAO Soil Classification:
FL FLUVISOLS Soils showing fluvic properties and having no diagnostic horizons other than
an ochric, a mollic or an umbric A horizon, or sulphidic material within 125cm of the surface.
FLd Dystric Fluvisols Fluvisols having a base saturation rate of less than 50% at least between
20 and 50cm below the surface; lacking a sulphidic horizon and sulphidic material within
125cm of the surface.
Fluvisols have developed in recent fluviatile deposits on the floodplain of the Okavango River.
On the banks and active floodplain of the Okavango River these soils are periodical y wet in all
or part of the profile due to the presence of seasonal flood water.
The floodplain area, two to six kilometres wide, can be divided into two zones in terms of soil
development and modification. A broad area adjacent to the present course of the river
actively receives fresh sediments during regular seasonal periods of inundation and hence the
soils are regularly rejuvenated. The soils of this zone, although used for wet season cropping
and dry season grazing, are not profoundly modified by agricultural activities and can therefore
be classified by their fluvic properties.
Soil profiles and auger holes show stratified layers of coarse and fine materials with a
predominance of fine to very fine sands and silts in shallower horizons and an increase in clay
content with depth. Significantly, the deposition of clay layers is uncommon and clay contents
even at depth are consequently lower than would be expected in these alluvial soils. This
attribute can be linked to the scarcity of clays in the provenance areas (Minader, 1996).
Surface clay contents are lower than subsurface horizons with an average of 6%, increasing
irregularly to 14% in lower horizons. Low clay contents combined with low and irregular levels
of organic matter are also linked to relatively low nutrient concentrations and CEC (total)
levels. Whereas these soils are not infertile, they are also not highly productive.
59
EFA Namibia Geomorphology
Anthrosols: Soils Conditioned by Human Influence
AT ANTHROSOLS
ATd Dystric Anthrosols
FAO Soil Classification:
AT ANTHROSOLS Soils in which human activities have resulted in profound modification or
burial of the original soil horizons through removal or disturbance of surface horizons, cuts and
fills, secular additions of organic materials, long-continued irrigation, etc.
ATd Dystric Anthrosols Anthrosols which have a base saturation of less than 50% at least from
20 to 50cm below the surface. Moderately deep to deep; moderately well to well drained; dark
brown fine sands to sandy loams. This unit of classification does not appear in the FAO
Revised Legend. Based on field experience, however, it best describes the soil characteristics
modified by man on the old and now drying (no longer periodically inundated) Okavango
floodplain. It is suggested as a refinement to the Legend.
Away from the main Okavango River channel in a broad zone of variable width, the floodplain
is no longer seasonally inundated. In this area soils resembling buried Fluvisols at depth and
Arenosols nearer to the surface have developed on colluvial sands lying over older dry
fluviatile deposits.
Intensively used for both dryland and irrigated cultivation, these soils have been significantly
modified. Evidence from borehole records and sample analyses (Weirenga, 1999) indicate that
the original morphology of these soils would have included buried accumulation horizons of
stratified coarse and fine materials under moderately deep fine sands of colluvial and aeolian
origin. Analytical and profile records from Mashere Agricultural College indicate that the
surface horizons have been physically mixed by ploughing, chemically altered by the addition
of organic materials, leached by irrigation water and generally deficient in potassium.
These soils therefore have been classified as Anthrosols to indicate the degree to which
modification by agricultural use has altered a number of their diagnostic properties. Judging by
the inherently low CEC status of the Arenosol group, by the relatively low nutrient
concentrations of the underlying Fluvisols, and by the fact that organic additives are needed to
increase the concentration of base cations, these soils have been categorized as Dystric
Anthrosols.
60
EFA Namibia Geomorphology
Soils, Landforms and Vegetation Sequences on Okavango Floodplains and Terraces
LAND
LAND
VEGETATION
VEGETATION TYPES
SOIL
REGION
SYSTEM
MAPPING UNIT
ASSOCIATIONS
Dystric
Northern Okavango
Okavango river
Acacia nigrescens Peltophorum africanum
Fluvisols
floodplain and
valley fields and
Sandplain terraces
shrublands
riverine forest
Combretum imberbe Acacia erioloba
Dystric
shrubland
Anthrosols
Haplic
Terminalia sericea Bauhinia petersiana
Arenosols
shrubland
Ferralic
Catophractes alexandri shrubland
Arenosols
Dystric
Floodplain grassland
Fluvisols
Table 4. 1: Vegetation Types and Soil Associations Aggregated by Land Syatem
Floodplain, river bank, old flood plain (terrace) and terrace slope comprise the main
sequences of landforms bordering the Kavango River. Grasslands with species such as
Vossia cuspidata, Cynodon dactylon and Setaria sphacelata dominate the floodplain, while
riverbanks originally supported riverine forests with Acacia nigrescens, Peltophorum africanum
and Diospyros mespiliformis as dominant tress and a dense shrub undergrowth of various
species. However, due to intense clearing and cultivation along the river, riverine forest has
disappeared almost entirely and only few, localised patches remain. Today's river banks and
terrace present an open parkland with few trees, cultivated land and many villages in between.
Remnants of shrubland with Combretum imberbe, Acacia erioloba, Terminalia sericea and
Bauhinia petersiana indicate the potential vegetation types of former terraces. However,
human impact has also resulted in an increase in shrubs, often on old farmland and may thus
give a false indication of what may have occurred naturally on these old floodplain terraces.
61
EFA Namibia Geomorphology
Figure 4. 1: Schematic diagram (not to scale) of Pterocarpus angolensis Schinziophyton
Rautanenii woodland and associated soil families on the floodplain and terrace
system of the Okavango River.
The terrace slopes support open stands of Schinziophyton rautanenii, possibly still prevalent
despite human impact because of their value as fruit trees.
The northern sandplain forms a sheet of several meters of sand cover with very few pans.
Although Pterocarpus angolensis and Schinziophyton rautanenii woodlands are prominent,
localised patches of Baikiaea plurijuga and Burkea africana woodlands occur throughout this
map unit. Combretum collinum forms another important tree component, while Combretum
zeyheri, Combretum psidioides, Bauhinia petersiana and Baphia massaiensis are prominent in
the shrub layer. Common grasses associated with these woodlands comprise Digitaria seriata,
Schmidtia pappophoroides and Urochloa brachyura.
62
EFA Namibia Geomorphology
4.3.2 Soils and agriculture
Mendelsohn (2006, p.24-29) provides a useful description of soils in Namibia including the
Kavango Region. He points out that the dry climate is not only a problem for agriculture due to
low rainfall, but also because over millions of years -it has provided little opportunity for soil
formation and nutrient retention.
The Kalahari sands are wind blown deposits that cover most of Kavango. They are called
arenosols and they are extremely poor in nutrients. Being highly permeable, they are also poor
in water retention. Quartz grains make up the bulk of the soil and there is little humus or
nutrients. These soils are particularly low in phosphorous, which results in deficiencies of
nitrogen.
Along the river and omuramba, Kalahari sand has been reworked and deposited by water.
Here small and patchy areas of better soils occur, with slightly higher clay and nutrient content.
But even these are rated has having "low" agricultural potential (p.29). The fertility of these
soils has to be maintained by the regular application of suitable fertilisers.
Mendelsohn and el Obeid (2007) provided a specialist socio-economic report for the Kavango
Biofuel EIA. Farming near the river consists largely of a mix of small-scale dryland crop
production and livestock farming. Most households practise this type of subsistence farming.
The dominant crop planted (95%) is mahangu because it is the only cereal that grows
relatively well on sandy, nutrient poor soils. Mahangu is also a bit more reliable in conditions of
unreliable rainfalls.
Yields are usually too low to support households. Average yields of mahangu amount to
between 100 and 300 kg/hectare. At a market value of N$3.00 N$4.00/kg, one hectare of
mahangu has a value of only N$300 to N$1,200 per year. Most fields cover less than 2ha. The
estimated maximum daily rate of return on labour devoted to mahangu amounts to only about
N$13.60 (Mendelsohn, 2006). Therefore 80% of households rely on cash incomes from other
sources to meet some or all of their cereal needs. Most rural Kavango households are not
self-sufficient in food production. There is also no need for them to be self-sufficient because
their food security comes largely from off-farm cash incomes.
Mendelsohn and el Obeid (2007) used satellite or aerial images to identify land within 10km of
the river to identify available land for the proposed Kavango Biofuel project. They found 65,000
ha of land that had been cleared for cultivation before 1990. They also found that only 75% of
cleared fields were actually utilised for crops. This fact is testimony to the poor yields, low
value of crops produced, and rapid decline of soil fertility under cultivation.
Most of the cultivation occurs outside the dry banks of the river, while some smal fields are
made on the old, higher floodplains.
Apart from dry-land cultivation, a number of irrigation projects exist or are being planned.
Figures supplied by the Directorate of Agriculture, indicated that 770 ha were already under
irrigation along the Namibian reach of the River, while a further 7,793 were being planned.
Thus a total of some 8,563 hectares may ultimately be irrigated. This would require a water
consumption of 128 million m3/year, or an abstraction rate from the River of 15.45 m3/second
(Eco.plan 2003).
The soils thus targeted are not good soils, by international standards, but Namibia is short of
arable lands. The Green Scheme, as this irrigation initiative is known is fairly contentious
critics of the scheme argue that:
63
EFA Namibia Geomorphology
a. The existing irrigation schemes are of dubious economic viability, requiring
subsidies,
b. A considerable input of artificial fertilisers is required,
c. Maize can be imported from South Africa more cheaply than it can be
produced in Kavango, and
d. The Green Scheme is being driven by a misguided government policy that
aims for food self-sufficiency, instead of aiming for food security through
developing cash economies, e.g. based on tourism and commerce.
Simmonds (1998) prepared an overview of soils for OKACOM which provided the following
exerpt.
Soils of the Okavango River floodplain and terraces
Soils of the Namibian and Botswana segments of the Okavango River floodplain and terraces
are composed predominantly of infertile aeolian sands of the Kalahari Formation with a low
organic matter content (Bethune, 1991; OCC, 1995).
Along the Okavango River channel on the Namibian side, the sandy soils of the floodplain and
river terraces are enriched with interspersed clay and silt layers deposited by seasonal flood
waters (FAO, 1984). Vegetated slacks between scroll bars on the floodplain trap deposits of
clay and fine silt, whilst layers of calcrete are exposed on many river terraces. Away from the
main river channel the soils are predominantly grey to yellow-orange sands (Schneider, 1987).
According to Bethune (1991) floods seldom deposit alluvial silt on the higher river terraces.
Both the floodplains, and the river terraces are cultivated by subsistence farmers (Bethune,
1991). Formal (irrigation) agriculture is confined to the higher river terraces, and is restricted to
a few locations with suitable soils, along the Okavango River between Rundu and Bagani
(Cashman et al., 1986).
Soil erosion is evident over much of the southern bank of the Okavango River between Rundu
and Bagani. Extensive soil erosion is particularly evident for a distance of some 50 kilometres
downstream of Rundu and between Andara and Popa Falls. Elsewhere along the Okavango
River, soil erosion occurs in scattered areas which mark the sites of existing and former
mahangu gardens, livestock kraals and villages. Most of this "scattered" erosion pattern is due
to trampling by livestock, indiscriminate clearing of riparian vegetation for agriculture and
collection of construction materials and fuel-wood.
Further notes on Okavango River terrace soils
Along the south and west banks of the Okavango River the terrace system constitutes a
distinct but discontinuous physiographic unit, lying some 7m above the river and separated
from it by a floodplain which varies in width between 2-6 km. The terrace itself is flat to even
and gently sloping, and incised by minor drainage lines.
Clovelly, Oakleaf and Hutton soil forms pre-dominate on the terrace system exhibiting classic
catenery associations with slope position (Schneider, 1987).
The three soils exhibit physical, chemical and mineralogical properties typical of semi-arid soils
whereby moderate to high base saturation due to slow leaching of the basic cations results in
64
EFA Namibia Geomorphology
pH values between 6.8 and 7.6 (slightly acid to slightly alkaline). The cation exchange capacity
is low to moderate and in all three soil forms kaolinite is the most abundant clay mineral.
Soil Series identified are mostly sandy and differ little in colour. Particle size analysis indicated
that the sand fraction dominates in all series by more than 50%. The clay content of all
identified series varies from 0-15%; locally more clay was recorded, as for example in the case
of an Oakleaf Limpopo.
All three soils have a relatively thin surface layers of non-calcareous loamy sand, overlying
thicker non-calcareous to loamy and clayey sand in B1 and B2 horizons. Calcareous subsoils
are found in the lower parts, as in the C-horizon of the Hutton Zwartfontein. The calcic C-
horizon (which has more carbonate than the parent material) appears to have formed by the
upward movement of carbonate-rich capillary water from the shallow groundwater table.
Landuse and soil quality on the Okavango river terraces
Soil formation processes on the river terrace system have ultimately been controlled by
topography and semi-arid climatic processes. These factors, together with sandy parent
materials and the accumulation of carbonates, soluble salts and silica, combine to yield soils
unfavourable for agricultural purposes. Even so, the terraces of the Okavango River and its
tributaries are cultivated by traditional subsistence farmers, and during the past 25 years, by
centre-pivot irrigation schemes.
Subsistence farming on the river terraces downstream from Rundu is dominated by dryland
cropping of maize and mahango along the silt-enriched river banks. Stock (primarily cattle) are
grazed on the floodplains.
A marked degree of environmental degradation is evident along the entire length of the
riparian environment. Landslip features, slope failures and moderate gully formations can be
observed on the discontinuous terrace slopes east of Rundu. In the area 0-4 km east of
Rundu, these appear to be associated with excavations and earth diggings. Both incipient and
extensively developed gully systems, observed from Rundu to Mukwe, radiate from numerous
footpaths and animal tracks linking villages situated on terrace ridges to the floodplain and
river banks. Accelerated and extensive riverbank erosion was observed on the northern
(Angolan) river banks and floodplain where land has recently been cleared for village and
garden development .
Parastatal irrigation schemes have been developed on the river terraces. No comment can be
made on the environmental status of these areas without further research, although detailed
soil studies (Schneider, 1987; Engels, pers.comm) indicate that the quality of the terrace soils
is expected to deteriorate rapidly under irrigation, given their high potential to allow upward
movement of carbonate-rich capillary water.
Production levels of the schemes appear to be lower than the development opportunity
suggests, as volumes of water abstracted from the Okavango River for irrigation have recently
been measured at levels substantially lower than the maximum permissible abstraction limit of
13.223 Mm3/annum, set by the Agreed Commitment for irrigation permits in Namibian territory
(Crerar,1997). It is possible that the reasons for under-utilization of available abstraction water
for irrigation purposes are rooted in decreased production levels under conditions of declining
soil fertility and the inability to fund increased production costs.
65
EFA Namibia Geomorphology
4.4 Summary & Information Gaps
There is almost no literature available that is applicable to geomorphology on the Namibian
section of the river. Hence nothing has been found of relevance to the Indicators here.
Several major gaps in knowledge have been identified for further research:
1.
Sediment transport studies on the Cuito and Kavango/Cubango Rivers -
separately. All the studies that have estimated sediment transport/discharge
have been done in the Delta (e.g.
2.
by McCarthy et al see reference list) and for the Popa Falls Hydro
Power EIA (Eco.plan, 2003). No studies were found on sediment dynamics
upstream of the confluence with the Cuito.
Geomorphological rates of change within the natural flow regime are not known. For example
at Kapako, we have no idea of the rates of meander migration / scroll bar formation. We need
to identify methodologies for dating parts of the floodplain and therefore estimating the natural
rates of change of various geomorphological features.
A detailed ground survey is needed for Kapako. I suggest using (a) digital laser aerial survey
in conjunction with (b) digital full colour images:
(a)
The laser survey has the advantage of penetrating vegetation and thatched roofs. It is
capable of generating 0.5m contours (or better). This was done by Water Transfer
Consultants for NamPower, for the Popa Falls Hydro Power project. It covers the area
from just below Popa Falls up to Andara (or Mukwe, somewhere). It should be possible
to get this from NamPower.
(b)
Full colour digital overlay is necessary to see vegetation, which gives a lot of
information about geomorphological processes e.g. tree lines.
Accurate mapping would help to better understanding of floodplain processes, areas
inundated at specific flow levels etc.
The river section between Divundu and Andara/Mukwe is fundamentally different from the rest
of the Namibian reach of the Okavango River. It should be included if the EFA is extended in
future. Some of the key differences are:
(a)
It is almost the only Riverine Forest left on the Namibian reach of the river,
(b)
The channel is largely rock controlled, with a lot of extensive islands made of sand on
top of a base of bedrock. The islands are forested. The islands would be particularly
vulnerable to rapid changes in water level (Eco.plan 2003), they would also be
66
EFA Namibia Geomorphology
vulnerable to people if the water levels are low enough to allow people easy access to
the islands.
(c)
There are many small backwaters, rock pools, riffles and rapids which are surely of
ecological importance being scarce habitats.
67
EFA Namibia Geomorphology
5 PREDICTED RESPONSES OF INDICATORS TO CHANGE IN FLOW REGIME
This chapter provides a summary of present understanding of the predicted responses of all
geomorphological indicators to potential changes in the flow regime. This is done first for
Kapako (Section 5.1) and then for Popa Falls (Section 5.2). In those two sections a table is
presented for each indicator that is relevant to that site. In each of those tables, an attempt is
made to answer the questions posed in Table 3.2. That is, to predict the response of the
indicator in question to potential changes in the flow regime.
The tables present the background understanding that was entered into the model at the
Knowledge Capture Workshop in Windhoek. However, one important parameter could not be
taken into account in the model, namely the effects of sediment-trapping dams or weirs that
may be built in future. This represents an important limitation on the Response Curves in the
model.
5.1 Site 4: Kapako Indicator responses to changes in flow regime
Six indicators are applicable to Kapako and its floodplains:
a. Cross sectional area of channel,
b. Extent of backwaters,
c. Percentage silts and clays in top 30cm on floodplain,
d. Extent of inundated floodplain,
e. Inundated pools and pans,
f. Extent of cut banks.
68
5.1.1 Cross Sectional Area of the Channel
Table 5. 1: Predicted response to possible changes in the flow regime of cross sectional area of the channel
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very low,
low, medium, high)
1
Onset is earlier or later
Nil
High
than natural
2
Dry Season Water levels are higher or
Nil
High
lower than natural
3
Extends longer than
Extended dry seasons without intervening floods would result in the
Low
natural
channel slowly narrowing as vegetation (reeds etc) encroaches
4
Transition 1 Duration is longer or
Nil
High
shorter than natural i.e.
hydrograph is steeper or
shallower
5
Flows are more or less
Nil
variable than natural
6
Flood
Onset is earlier or later
Nil
High
than natural
synchronisation with rain
may be changed
7
season
Natural proportion of
If floods are regularly larger than normal, one would expect channel
Med
different types of flood year cross section to increase. If floods are repeatedly lower than normal, a
changed
very slow decrease in channel cross section may occur because
stabilising vegetation on the sides of the channel would be taken out
less easily.
8
Onset is earlier or later
Nil
High
than natural
9
Transition 2 Duration is longer or
Steeper than normal falling hydrograph would tend to cause bank
Medium
shorter than natural i.e.
collapse.
hydrograph is steeper or
shallower
69
5.1.2 Extent of Backwaters
Table 5. 2: Predicted response to possible changes in the flow regime of slow/no flow backwaters
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very low,
low, medium, high)
1
Onset is earlier or later
Nil
High
than natural
2
Dry Season Water levels are higher or
Because backwaters are, buy our definition, connected to the main
High
lower than natural
channel, water level will be the same as in the main channel. The
backwaters are steep sided, therefore area does not change very
much in response to changes in river stage.
3
Extends longer than
Nil
Nil
natural
4
Transition 1 Duration is longer or
Nil
High
shorter than natural i.e.
hydrograph is steeper or
shallower
5
Flows are more or less
Nil
High
variable than natural
6
Flood
Onset is earlier or later
Nil
High
than natural
synchronisation with rain
may be changed
7
season
Natural proportion of
Nil
High
different types of flood
year changed
8
Onset is earlier or later
Nil
High
than natural
9
Transition 2 Duration is longer or
Nil
High
shorter than natural i.e.
hydrograph is steeper or
70
EFA Namibia Geomorphology
shallower
71
5.1.3 Percentage Silt and Clay on Floodplain
Table 5. 3: Predicted response to possible changes in the flow regime of percentage silt and clay in the top 30cm on floodplains
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low, medium,
high)
1
Onset is earlier or later
Nil
High
than natural
2
Dry Season Water levels are higher or
Nil
High
lower than natural
3
Extends longer than natural Nil
High
4
Transition 1 Duration is longer or
Nil
High
shorter than natural i.e.
hydrograph is steeper or
shallower
5
Flows are more or less
Nil
High
variable than natural
6
Flood
Onset is earlier or later
Nil
High
than natural
synchronisation with rain
may be changed
7
season
Natural proportion of
If substantial flooding does not occur, clays will not be added. Since
Med
different types of flood
clays that are present will be eroded by wind or mixed down by soil
year changed
organisms, failure of overbank flooding will result in silt and clay levels
declining over time.
8
Onset is earlier or later
Nil
High
than natural
9
Transition 2 Duration is longer or
Nil
High
shorter than natural i.e.
72
EFA Namibia Geomorphology
hydrograph is steeper or
shallower
73
5.1.4 Extent of Inundated Floodplain
Table 5. 4: Predicted response to possible changes in the flow regime of the extent of inundated floodplains
Question Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low, medium,
high)
1
Onset is earlier or later
Nil
High
than natural
2
Dry Season Water levels are higher or Nil
High
lower than natural
3
Extends longer than
Nil
High
natural
4
Transition 1 Duration is longer or
Nil
High
shorter than natural i.e.
hydrograph is steeper or
shallower
5
Flows are more or less
Nil
High
variable than natural
6
Flood
Onset is earlier or later
Nil
High
than natural
synchronisation with rain
may be changed
7
season
Natural proportion of
The river is contained by levees along its banks in most places. Once
Medium
different types of flood
the levees are overtopped, most of the floodplain will be inundated
year changed
quickly. However, as the flood continues to rise, it increases water
depth on the floodplain but the area inundated probably increases
relatively little in proportion to discharge.
8
Onset is earlier or later
Nil
High
than natural
9
Transition 2 Duration is longer or
Nil
High
74
EFA Namibia Geomorphology
shorter than natural i.e.
hydrograph is steeper or
shallower
75
5.1.5 Extent of Inundated Pools and Pans on Floodplains
Table 5. 5: Predicted response to possible changes in the flow regime of inundated pools and pans
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very low,
low, medium, high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry
Water levels are higher or lower
Assuming that pools are connected to the river channel by
Med (assumption not
Season
than natural
groundwater and that the floodplain is permeable, then higher flows in tested)
the dry season would increase the likelihood that pools will retain
water to the end of the dry season.
3
Extends longer than natural
Assuming that pools are connected to the river channel by
Med (assumption not
groundwater and that the floodplain is permeable, then higher flows in tested)
the dry season would increase the likelihood that pools will retain
water to the end of the dry season.
4
Transition Duration is longer or shorter
Nil
High
1
than natural i.e. hydrograph is
steeper or shallower
5
Flows are more or less variable
Nil
High
than natural
6
Flood
Onset is earlier or later than
Nil
High
natural synchronisation with
rain may be changed
7
season
Natural proportion of different
Greater floods would increase the likelihood of pools filling and
Medium
types of flood year changed
remaining full because the groundwater in the floodplains would be
recharged.
8
Onset is earlier or later than
Nil
High
natural
9
Transition Duration is longer or shorter
Nil
High
2
than natural i.e. hydrograph is
steeper or shallower
76
EFA Namibia Geomorphology
5.1.6 Extent of Cut Banks
Table 5. 6: Predicted response to possible changes in the flow regime of cut banks
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low, medium,
high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry Season
Water levels are higher or lower Nil
Med
than natural
3
Extends longer than natural
Nil
Med
4
Transition 1
Duration is longer or shorter
Nil
High
than natural i.e. hydrograph is
steeper or shallower
5
Flows are more or less variable
Nil
High
than natural
6
Flood
Onset is earlier or later than
Nil
High
natural synchronisation with
rain may be changed
7
season
Natural proportion of different
Increased flood levels would result in greater erosion of banks (assuming no change
High
types of flood year changed
in sediment supply)
8
Onset is earlier or later than
Nil
High
natural
9
Transition 2
Duration is longer or shorter
If the falling hydrograph is steepened, water seeping out of the saturated banks will
High
than natural i.e. hydrograph is
tend to cause collapse i.e. the likelihood of bank collapse will be increased
steeper or shallower
77
Site 5: Popa Falls Indicator Responses to Changes in Flow
Regime
Four indicators are applicable to Popa Falls:
a) Extent of exposed rocky habitat,
b) Cross sectional area of channel,
c) Sand bars at low flow,
d) Extent of vegetated islands.
78
5.2.1 Extent of exposed rocky habitat
Table 5. 7: Predicted response to possible changes in the flow regime of exposed rocky habitat
Question Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low, medium,
high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry Season
Water levels are higher or
Increased water levels in the dry season would reduce the area of rocky habitat
Medium
lower than natural
for Rock Pratincoles. These birds breed on the rocks in the rapids during the low
flow season.
3
Extends longer than natural
Nil
High
4
Transition 1
Duration is longer or shorter
Nil
High
than natural i.e. hydrograph is
steeper or shallower
5
Flows are more or less variable Nil
High
than natural
6
Flood
Onset is earlier or later than
During a low flood season there is still enough rocky habitat for Rock Pratincoles High
natural synchronisation with
to feed upon, but at above-average flood levels this habitat becomes drowned out.
rain may be changed
7
season
Natural proportion of different Nil
High
types of flood year changed
8
Onset is earlier or later than
Nil
High
natural
9
Transition 2
Duration is longer or shorter
Nil
High
than natural i.e. hydrograph is
steeper or shallower
79
EFA Namibia Geomorphology
5.2.2 Cross Sectional Area of the Channels
Table 5. 8: Predicted response to possible changes in the flow regime of cross sectional area of the channels
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low, medium,
high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry Season
Water levels are higher or
If low flow levels are consistently lower than normal without intervening floods, then High
lower than natural
encroachment by plants (reeds etc) would reduce the cross sectional area of the
channel. Scouring of the bed would also be decreased.
3
Extends longer than natural
Ditto as above
Ditto as above
4
Transition 1
Duration is longer or shorter
Prolonged high floods would tend to enlarge the channel a little, even though at
Medium
than natural i.e. hydrograph is
Popa Falls the banks are well stabilised by vegetation.
steeper or shallower
5
Flows are more or less variable Higher flood peaks would tend to enlarge the channel a little, even though at Popa
Medium
than natural
Falls the banks are well stabilised by vegetation. However, reduced flood peaks
would result in a much slower reduction in channel cross section.
6
Flood
Onset is earlier or later than
Nil
High
natural synchronisation with
rain may be changed
7
season
Natural proportion of different
Nil
High
types of flood year changed
8
Onset is earlier or later than
Nil
High
natural
9
Transition 2
Duration is longer or shorter
A rapid decline in the hydrograph (greater than natural) would tend to destabilise
Medium
than natural i.e. hydrograph is the banks as water seeps out of the saturated banks. This may result in bank
steeper or shallower
collapse, even though at Popa Falls the banks are quite well stabilised by
vegetation.
80
EFA Namibia Geomorphology
5.2.3 Extent of sandbars at low flow
Table 5. 9: Predicted response to possible changes in the flow regime of sandbars at low flow
Question Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very
low, low,
medium, high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry Season
Water levels are higher or
Higher flow levels in the dry season would tend to cover sand bars, which are used High
lower than natural
for breeding by African skimmers.
3
Extends longer than natural
Nil
High
4
Transition 1
Duration is longer or shorter
Nil
Medium
than natural i.e. hydrograph is
steeper or shallower
5
Flows are more or less variable Nil
Medium
than natural
6
Flood
Onset is earlier or later than
Nil
High
natural synchronisation with
rain may be changed
7
season
Natural proportion of different
Nil
High
types of flood year changed
8
Onset is earlier or later than
Nil
Medium
natural
9
Transition 2
Duration is longer or shorter
Nil
Medium
than natural i.e. hydrograph is
steeper or shallower
81
EFA Namibia Geomorphology
5.2.4 Extent of vegetated islands
Table 5. 10: Predicted response to possible changes in the flow regime of vegetated islands
Question
Season
Possible flow change
Predicted response of indicator
Confidence in
number
prediction (very low,
low, medium, high)
1
Onset is earlier or later than
Nil
High
natural
2
Dry Season
Water levels are higher or lower Nil
High
than natural
3
Extends longer than natural
Prolonged dry seasons without intervening floods would al ow papyrus and reeds to
High
encroach on the channel, so increasing the area of islands in the main channels
4
Transition 1
Duration is longer or shorter
Nil
Medium
than natural i.e. hydrograph is
steeper or shallower
5
Flows are more or less variable
Nil
Medium
than natural
6
Flood
Onset is earlier or later than
Nil
High
natural synchronisation with
rain may be changed
7
season
Natural proportion of different
Higher flood peaks would tend to erode the margins of islands where they are not
High
types of flood year changed
made of bedrock. Flood peaks of longer duration would also tend to increase erosion.
Consistently low peaks may allow islands to grow as vegetation encroaches on the
channels.
8
Onset is earlier or later than
Nil
Medium
natural
9
Transition 2
Duration is longer or shorter
Nil
Medium
than natural i.e. hydrograph is
steeper or shallower
82
5.3 CONCLUSION
No quantitative data, other than the hydrological data, were available as a basis for
predictions about the responses of indicators to changes in potential flow regime.
Thus the entries to the EF-DSS model were based on knowledge and experience of
a non-quantitative kind. The ages of features in the floodplains were also unknown.
Often it was possible to predict, within a reasonable level of confidence, the direction
of change but not the rate of change.
Some of the uncertainties could be overcome with further research, including the
following:
a. Detailed surveys that can produce contours accurate to 0,2m. This would
enable the hydrologists to predict the extent floodplain inundation with for
given discharge of the river with greater accuracy.
b. The discharge of sediment at the Kapako site needs to be measured and
calculated as it was at Divundu.
c. Groundwater studies on the floodplain would help to understand the degree
of permeability, and hence the likelihood of individual pools drying out at any
given water level.
d. The percentage of silt and clays in the top 30cm on the floodplains needs to
be measured for selected representative locations, on active and older
floodplains, as baseline data.
e. It would be useful to find some method by which to date the scroll bars on the
active floodplains to develop an understanding of the time frames involved.
That may even lead to the possibility of correlating specific scroll bars with
specific flood events in historical flow data.
f. Analysis of old aerial photos that can be geo-referenced in a GIS programme
would be very useful to determine the rates of change in extent of islands,
backwaters, and floodplains.
Most importantly, any future modelling needs to take into account the interaction of
water and sediment. These two components of the system are equally important in
the creation or destruction of habitats. The EF-DSS model in its present form does
not take into account the effects of sediment trapping dams or weirs, although it is
possible that the consequent environmental impacts may reach as far downstream
as the Okavango Delta (McCarthy, 2003).
FLOW-RESPONSE RELATIONSHIPS FOR USE IN THE OKAVANGO EF-DSS
Flow-Response Curves showing the relationship between every indicator and every
relevant flow category were generated at the Knowledge Capture Workshop in
Windhoek in April 2009. These were submitted to the EFA team leaders at that
workshop and the instruction was given to the workshop delegates to omit the details
from this report as the Response Curves will be made available on CD as part of the
final Project reporting.
However, there is one point that must be recorded here. The flow-response model
that was used at the workshop was unable to handle the effect of sediment trapping
83
EFA Namibia Geomorphology
by dams or weirs on the proposed geomorphological indicators that were under
discussion. We were therefore told to ignore the effect of sediment-trapping dams in
completing the response curves, and consider the response of indicators in relation
to water flows alone and in isolation of likely sediment load changes which may occur
under the various development scenarios. This represents an important limitation on
the EFA study. The importance of this issue is highlighted with reference to the
following examples:
a. If sediment was trapped in a weir or dam, the sandy river beds
downstream would be scoured for some distance downstream. This
would make the channel more efficient in transporting floods and
consequently reduce the extent of overbank flooding into the
floodplains with significant ecological consequences.
b. If sediment is trapped, including fine materials (clays and organic
particles) then there will be less deposition of fine material and
nutrients on the floodplains, with the consequence of deteriorating soil
quality there.
It is therefore recommended that the model needs to be amended in some way to
take into account the effect of sediment trapping by weirs and dams. It must be
stressed that, in determining the flow-response curves for geomorphology, or for the
ecological disciplines where physical habitat is an important criterion, the flow of
water cannot meaningfully be separated from the flow of sediment.
6 REFERENCES
BBC News / Science-Nature / African sands `set for upheaval' (30 June 2005)
CCA (2007) Kavango Biofuel Project: Plantations of Jatropha curcas
Environmental Impact Assessment, Volume 1: Report. Compiled for Prime
Investment (Pty) Ltd by Colin Christian & Associates CC, Windhoek.
Coles, S.K.P. (2003) Application of side-scan sonar and bathymetric survey
techniques to a determination of bedload sediment transport rates in the Okavango
river at Divundu, Caprivi, Namibia. Prepared for Eco.plan (2003).
Dollar, E.S. (2004) Fluvial Geomorphology. Progress in Physical Geography, Vol.28,
No.3. (1 September 2004) pp.405-450.
Eco.plan (2001) Rundu Floodplain Development Plan: Investigation Report
Prepared for Lux-Development, Project NAM 320 : 01 161. Windhoek.
August 2001
Eco.plan (2003) Pre-feasibility Study for the Popa Falls Hydro Power Project:
Preliminary Environmental Assessment, Volumes 1 & 2. Prepared for Water
Transfer Consultants and NamPower, November 2003.
Gumbricht, T. and McCarthy, T.S. and Merry, C.L. (2001) The topography of the
Okavango Delta, Botswana, and its tectonic and sedimentological implications.
South African Journal of Geology. Vol.104, pp 243-264. 2001.
84
EFA Namibia Geomorphology
Hamilton, S.K., Kellndorfer, J., Lehner, B., and Tobler, M. (2007) Remote sensing of
floodplain geomorphology as a surrogate for biodiversity in a tropical river system
(Madre de Dios, Peru). Geomorphology, Vol.89, No.1-2 (1 September 2007) pp.23-
38
Lane, S.N. (1998) Hydraulic modeling in hydrology and geomorphology: a review of
high resolution approaches. Hydrological Processes, Vol.12, No.8. (1998) pp.1131-
1150.
McCarthy, T.S. (2003) Sediment Movement in the Kavango River at Divundu. In
Eco.plan (2003) Volume 2. Appendix K.
McCarthy, T.S., Barry, B., Bloem, A., Ellery, W.N., Heister, C. Merry, L., Rüther, H.
and Sternberg, H. (1997) The gradient of the Okavango fan, Botswana, and its
sedimentological and tectonic implications. Journal of African Sciences, Vol. 24
No.1/2 pp.65-78.
McCarthy, T.S. and Ellery, W.N. (1998) The Okavango Delta. Transactions of the
Royal Society of South Africa, 53. pp157-182.
McCarthy, T.S., Ellery, W.N., and Dangerfield, J.M. (1998) The Role of Biota in the
Initiation and Growth of Islands on the Floodplain of the Okavango Alluvial Fan,
Botswana. Earth Surface Processes and Landforms, Vol 23. pp291-316.
McCarthy, T.S., Ellery, W.N., and Stanistreet, I.G. (1992) Avulsion mechanisms on
the Okavango fan, Botswana: the control of a fluvial system by vegetation.
Sedimentology (1992) 39, pp 779-795.
McCarthy, T.S., Green, R.W., and Franey, N.J. (1993) The influence of neo-
tectonics on water dispersal in the northeastern regions of the Okavango Swamps,
Botswana. Journal of African Sciences, Vol. 17 No.1 pp.23-32, 1993.
McCarthy, T.S., Smith, N.D., Ellery, W.N., and Gumbricht, T. (2002) The Okavango
Delta Semiarid Alluvial-fan Sedimentation Related to Incipient Rifting.
Sedimentation in Continental Rifts. SEPM Special Publication No.73 pp.179-193.
Society for Sedimentary Geology.
McCarthy, T.S., Stanistreet, I.G., and Cairncross, B. (1991) The sedimentary
dynamics of active fluvial channels on the Okavango fan, Botswana. Sedimentology
(1991) 38, pp 471
487.
Mendelsohn, J. (2006) Farming Systems in Namibia. Published by RAISON,
Windhoek for Namibia National Farmers Union.
Mendelsohn, J. and el Obeid, S. (2004) Okavango River: the Flow of a Lifeline.
Struik, Cape Town.
Mendelsohn, J. and el Obeid, S. (2007) Kavango Biofuel Project: Specialist
Component Report on Socio-economic Impacts. In CCA (2007) Kavango Biofuel
Project: Plantations of Jatropha curcas Environmental Impact Assessment, Volume
85
EFA Namibia Geomorphology
2: Appendices. Compiled for Prime Investment (Pty) Ltd by Colin Christian &
Associates CC, Windhoek.
Mendelsohn, J., Jarvis, A., Roberts, C., and Robertson, T. (2002) Atlas of Namibia: A
Portrait of the Land and its People. Published for the Ministry of Environment and
Tourism by David Philip.
ORG (1986 2005) Okavango Research Group, Volumes 1 7. A collection of
papers / reprints of papers that were published in various journals during the period.
University of the Witwatersrand, Johannesburg.
Osterkamp and Toy (1997) Geomorphic considerations for erosion prediction.
Environmental Geology, Vol.29, No.3. (16 February 1997) pp.152-157
http://www.citeulike.org/user/peterheng/article/3857130
Peckham, S.D. (2003) Fluvial landscape models and catchment-scale sediment
transport. Global and Planetary Change, Vol.39, No.1-2. (October 2003), pp.31-51.
Willing, P. (1998) A Review of hydrological Aspects of the Proposed Epupa
dam and Reservoir, Cunene River, Namibia.
http://internationalrivers.org/en/africa/epupa-dam namibia/review-
hydrological-aspects
WTC (2003) Pre-Feasibility Study for the Popa Falls Hydro Power Project:
Technical Report, Volume 1: Main Report. November 2003. Prepared for
NamPower, Windhoek.
Yuquian, L. (1989) Manual on Operational Methods for the Measurement of
Sediment Transport. Operational hydrology Report No.29. World Meteorological
Organisation. Geneva, Switzerland.
SOILS BIBLIOGRAPHY
Bethune, S. (1991): Kavango River Wetlands. Madoqua, 17(2): 77-112. Contact:
Library, Department of Water Affairs, MAWRD, Windhoek.
Cashman, A.M., Harris, M., Plettenberger. & Volkman, B. (1986): Preliminary
Reconnaissance Report on Irrigation Possibilities Along the Okavango River in the
Kavango. Internal Report, Planning Division, Department of Water Affairs, Windhoek.
Contact: Library, Department of Water Affairs, MAWRD, Windhoek.
FAO/UNDP (1984): Assessment of Potential Land Suitability -Namibia -Land
Regions and Land-Use Potential. AG: DP/NAM/78/004 Technical Report 2, Food and
Agriculture Organization of the United Nations, Rome, 1984. Contact: Soil Mapping
and Advisory Services, Ministry of Agriculture, Gaborone.
ISSS (1996): Proceedings of Soil Correlation Workshop. Institute of Soil, Water and
Climate, Pretoria. Contact ISSS organiser: Professor Laker (e-mail:
grondkl@scientia.up.ac.za)
Loxton, R.F., Hunting & Associates (1971): Consolidated Report on Reconnaissance
Surveys of the Soils of Northern and Central South West Africa in Terms of their
86
EFA Namibia Geomorphology
Potential for Irrigation. Consultancy report for the Department of Water Affairs,
Windhoek. Contact: Library, Department of Water Affairs, MAWRD, Windhoek.
Macvicar, C.N., Loxton, R.F., Lambrechts, J.J.N., Le Roux, J., De Villiers, J.M.,
Verster, E., Merryweather, F.R., Van Rooyen, T.H., and Von M. Harmse, H.J. (1977,
updated 1991) : Soil Classification -A Binomial System for South Africa. The Soil and
Irrigation Research Institute, Department of Agricultural Technical Services, Pretoria,
RSA. Contact: Ms M. Coetzee, Land Evaluation Unit, MAWRD, Windhoek.
MINADER (1996): Angola -Agricultural Recovery and Development Options Review.
Draft Report. Report No: 96/116 TCP-ANG. Ministry of Agriculture and Rural
Development (MINADER) assisted by FAO/IFAD/UNDP/WB/WFP. 17 December
1996.
Namibian Groundwater Development Consultants (1991): Groundwater Investigation
in Kavango and Bushmanland. Final report prepared for the Department of Fisheries
and Water, Government of the Republic of Namibia, Windhoek. Contact: Library,
Department of Water Affairs, MAWRD, Windhoek, or Interconsult Namibia (Pty) Ltd,
Windhoek.
Schneider, M.B. (1987): Notes on the Terrace Soils of the Okavango River, northern
S.W.A./Namibia. SWA Scientific Society Journal, Vol. XL/XLI -1985/86, 1986/7.
Simmonds, E.B. (1997): The Soils of the Okavango River in Namibia and the
Okavango Delta in Botswana. Specialist Report prepared for CSIR and Water
Transfer Consultants. Initial Environmental Evaluation of the Okavango River -
Grootfontein Pipeline Link to the Eastern National Water Carrier in Namibia, January
1997. Contact: Library, Department of Water Affairs, MAWRD, Windhoek, or Water
Transfer Consultants, c/o Parkman (Namibia), Windhoek.
Simmonds, E.B. (1998): Soils Overview. Report prepared for Okavango River Basin
Preparatory Assessment Project (OKACOM), Windhoek, February 1998.
Simmonds, E.B.and Burke, A. (2001): Natural Resource Mapping of Kavango
Region, Namibia. Joint consultancy report by Interconsult Namibia, EnviroScience
and GISL (UK) for the Directorate of Environmental Affairs, Ministry of Environment
and Tourism.
87
EFA Namibia Geomorphology
APPENDIX A: RAW DATA -SOILS
88
89
EFA Namibia Geomorphology
90
91
The Okavango River Basin Transboundary Diagnostic Analysis Technical Reports
I
Transboundary Diagnostic Analysis to establish a
n 1994, the three riparian countries of the Okavango
base of available scientific evidence to guide future
River Basin Angola, Botswana and Namibia
decision making. The study, created from inputs from
agreed to plan for collaborative management of the
multi-disciplinary teams in each country, with
natural resources of the Okavango, forming the
specialists in hydrology, hydraulics, channel form,
Permanent Okavango River Basin Water
water quality, vegetation, aquatic invertebrates, fish,
Commission (OKACOM). In 2003, with funding from
birds, river-dependent terrestrial wildlife, resource
the Global Environment Facility, OKACOM launched
economics and socio-cultural issues, was
the Environmental Protection and Sustainable
coordinated and managed by a group of specialists
Management of the Okavango River Basin (EPSMO)
from the southern African region in 2008 and 2009.
Project to coordinate development and to anticipate
and address threats to the river and the associated
The following specialist technical reports were
communities and environment. Implemented by the
produced as part of this process and form
United Nations Development Program and executed
substantive background content for the Okavango
by the United Nations Food and Agriculture
River Basin Transboundary Diagnostic Analysis
Organization, the project produced the
Final Study
Reports integrating findings from all country and background reports, and covering the entire
Reports
basin.
Aylward, B.
Economic Valuation of Basin Resources: Final Report to
EPSMO Project of the UN Food & Agriculture Organization as
an Input to the Okavango River Basin Transboundary
Diagnostic Analysis
Barnes, J. et al.
Okavango River Basin Transboundary Diagnostic Analysis:
Socio-Economic Assessment Final Report
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment Project
C.A.
Initiation Report (Report No: 01/2009)
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment EFA
C.A.
Process Report (Report No: 02/2009)
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment
C.A.
Guidelines for Data Collection, Analysis and Scenario Creation
(Report No: 03/2009)
Bethune,
S.
Mazvimavi,
Okavango River Basin Environmental Flow Assessment
D. and Quintino, M.
Delineation Report (Report No: 04/2009)
Beuster, H.
Okavango River Basin Environmental Flow Assessment
Hydrology Report: Data And Models(Report No: 05/2009)
Beuster,
H. Okavango River Basin Environmental Flow Assessment
Scenario Report : Hydrology (Report No: 06/2009)
Jones, M.J.
The Groundwater Hydrology of The Okavango Basin (FAO
Internal Report, April 2010)
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment
C.A.
Scenario Report: Ecological and Social Predictions (Volume 1
of 4)(Report No. 07/2009)
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment
C.A.
Scenario Report: Ecological and Social Predictions (Volume 2
of 4: Indicator results) (Report No. 07/2009)
King, J.M. and Brown,
Okavango River Basin Environmental Flow Assessment
C.A.
Scenario Report: Ecological and Social Predictions: Climate
Change Scenarios (Volume 3 of 4) (Report No. 07/2009)
King, J., Brown, C.A.,
Okavango River Basin Environmental Flow Assessment
Joubert, A.R. and
Scenario Report: Biophysical Predictions (Volume 4 of 4:
Barnes, J.
Climate Change Indicator Results) (Report No: 07/2009)
King, J., Brown, C.A.
Okavango River Basin Environmental Flow Assessment Project
and Barnes, J.
Final Report (Report No: 08/2009)
Malzbender, D.
Environmental Protection And Sustainable Management Of The
Okavango River Basin (EPSMO): Governance Review
Vanderpost, C. and
Database and GIS design for an expanded Okavango Basin
Dhliwayo, M.
Information System (OBIS)
Veríssimo, Luis
GIS Database for the Environment Protection and Sustainable
Management of the Okavango River Basin Project
Wolski,
P.
Assessment of hydrological effects of climate change in the
Okavango Basin
Country Reports
Angola
Andrade e Sousa,
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Biophysical Series
Helder André de
Okavango: Módulo do Caudal Ambiental: Relatório do
Especialista: País: Angola: Disciplina: Sedimentologia &
Geomorfologia
92
EFA Namibia Geomorphology
Gomes, Amândio
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Okavango: Módulo do Caudal Ambiental: Relatório do
Especialista: País: Angola: Disciplina: Vegetação
Gomes,
Amândio
Análise Técnica, Biofísica e Socio-Económica do Lado
Angolano da Bacia Hidrográfica do Rio Cubango: Relatório
Final:Vegetação da Parte Angolana da Bacia Hidrográfica Do
Rio Cubango
Livramento, Filomena
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Okavango: Módulo do Caudal Ambiental: Relatório do
Especialista: País: Angola: Disciplina:Macroinvertebrados
Miguel, Gabriel Luís
Análise Técnica, Biofísica E Sócio-Económica do Lado
Angolano da Bacia Hidrográfica do Rio Cubango:
Subsídio Para o Conhecimento Hidrogeológico
Relatório de Hidrogeologia
Morais, Miguel
Análise Diagnóstica Transfronteiriça da Bacia do Análise Rio
Cubango (Okavango): Módulo da Avaliação do Caudal
Ambiental: Relatório do Especialista País: Angola Disciplina:
Ictiofauna
Morais,
Miguel
Análise Técnica, Biófisica e Sócio-Económica do Lado
Angolano da Bacia Hidrográfica do Rio Cubango: Relatório
Final: Peixes e Pesca Fluvial da Bacia do Okavango em Angola
Pereira, Maria João
Qualidade da Água, no Lado Angolano da Bacia Hidrográfica
do Rio Cubango
Santos,
Carmen
Ivelize
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Van-Dúnem S. N.
Okavango: Módulo do Caudal Ambiental: Relatório de
Especialidade: Angola: Vida Selvagem
Santos, Carmen Ivelize
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Van-Dúnem S.N.
Okavango:Módulo Avaliação do Caudal Ambiental: Relatório de
Especialidade: Angola: Aves
Botswana Bonyongo, M.C.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report: Country:
Botswana: Discipline: Wildlife
Hancock, P.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module : Specialist Report: Country:
Botswana: Discipline: Birds
Mosepele,
K. Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report: Country:
Botswana: Discipline: Fish
Mosepele, B. and
Okavango River Basin Technical Diagnostic Analysis:
Dallas, Helen
Environmental Flow Module: Specialist Report: Country:
Botswana: Discipline: Aquatic Macro Invertebrates
Namibia
Collin Christian &
Okavango River Basin: Transboundary Diagnostic Analysis
Associates CC
Project: Environmental Flow Assessment Module:
Geomorphology
Curtis, B.A.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report Country:
Namibia Discipline: Vegetation
Bethune, S.
Environmental Protection and Sustainable Management of the
Okavango River Basin (EPSMO): Transboundary Diagnostic
Analysis: Basin Ecosystems Report
Nakanwe, S.N.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report: Country:
Namibia: Discipline: Aquatic Macro Invertebrates
Paxton,
M. Okavango River Basin Transboundary Diagnostic Analysis:
Environmental Flow Module: Specialist
Report:Country:Namibia: Discipline: Birds (Avifauna)
Roberts, K.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report: Country:
Namibia: Discipline: Wildlife
Waal,
B.V. Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module: Specialist Report: Country:
Namibia:Discipline: Fish Life
Country Reports
Angola
Gomes, Joaquim
Análise Técnica dos Aspectos Relacionados com o Potencial
Socioeconomic
Duarte
de Irrigação no Lado Angolano da Bacia Hidrográfica do Rio
Series
Cubango: Relatório Final
Mendelsohn,
.J.
Land use in Kavango: Past, Present and Future
Pereira, Maria João
Análise Diagnóstica Transfronteiriça da Bacia do Rio
Okavango: Módulo do Caudal Ambiental: Relatório do
Especialista: País: Angola: Disciplina: Qualidade da Água
Saraiva, Rute et al.
Diagnóstico Transfronteiriço Bacia do Okavango: Análise
Socioeconómica Angola
Botswana Chimbari, M. and
Okavango River Basin Trans-Boundary Diagnostic Assessment
Magole, Lapologang
(TDA): Botswana Component: Partial Report: Key Public Health
93
EFA Namibia Geomorphology
Issues in the Okavango Basin, Botswana
Magole,
Lapologang
Transboundary Diagnostic Analysis of the Botswana Portion of
the Okavango River Basin: Land Use Planning
Magole, Lapologang
Transboundary Diagnostic Analysis (TDA) of the Botswana p
Portion of the Okavango River Basin: Stakeholder Involvement
in the ODMP and its Relevance to the TDA Process
Masamba,
W.R.
Transboundary Diagnostic Analysis of the Botswana Portion of
the Okavango River Basin: Output 4: Water Supply and
Sanitation
Masamba,W.R.
Transboundary Diagnostic Analysis of the Botswana Portion of
the Okavango River Basin: Irrigation Development
Mbaiwa.J.E. Transboundary Diagnostic Analysis of the Okavango River
Basin: the Status of Tourism Development in the Okavango
Delta: Botswana
Mbaiwa.J.E. &
Assessing the Impact of Climate Change on Tourism Activities
Mmopelwa, G.
and their Economic Benefits in the Okavango Delta
Mmopelwa,
G.
Okavango River Basin Trans-boundary Diagnostic Assessment:
Botswana Component: Output 5: Socio-Economic Profile
Ngwenya, B.N.
Final Report: A Socio-Economic Profile of River Resources and
HIV and AIDS in the Okavango Basin: Botswana
Vanderpost,
C.
Assessment of Existing Social Services and Projected Growth
in the Context of the Transboundary Diagnostic Analysis of the
Botswana Portion of the Okavango River Basin
Namibia
Barnes, J and
Okavango River Basin Technical Diagnostic Analysis:
Wamunyima, D
Environmental Flow Module: Specialist Report:
Country: Namibia: Discipline: Socio-economics
Collin Christian &
Technical Report on Hydro-electric Power Development in the
Associates CC
Namibian Section of the Okavango River Basin
Liebenberg, J.P.
Technical Report on Irrigation Development in the Namibia
Section of the Okavango River Basin
Ortmann, Cynthia L.
Okavango River Basin Technical Diagnostic Analysis:
Environmental Flow Module : Specialist Report Country:
Namibia: discipline: Water Quality
Nashipili,
Okavango River Basin Technical Diagnostic Analysis: Specialist
Ndinomwaameni
Report: Country: Namibia: Discipline: Water Supply and
Sanitation
Paxton,
C.
Transboundary Diagnostic Analysis: Specialist Report:
Discipline: Water Quality Requirements For Human Health in
the Okavango River Basin: Country: Namibia
94
1