Received: 24May 2023 Revised: 23October 2023 Accepted: 29October 2023

DOI: 10.1002/aff2.140

R E V I EW ART I C L E

State of knowledge of aquatic ecosystem and fisheries of the
Lake Edward System, East Africa

LabanMusinguzi1,2 Nathan Vranken2,3 ViannyNatugonza4 WilliamOkello1

Maarten van Steenberge2,5,6 Jos Snoeks2,3

1National Fisheries, Resources Research

Institute (NaFIRRI), Jinja, Uganda

2Department of Biology, Fish Diversity and

Conservation, KU Leuven, Leuven, Belgium

3Vertebrates Section, Ichthyology, Royal

Museum for Central Africa, Tervuren, Belgium

4Maritime Institute, BusitemaUniversity,

Tororo, Uganda

5OD Taxonomy and Phylogeny, Royal Belgian

Institute of Natural Sciences, Brussels, Belgium

6Centre for Environmental Studies, University

of Hasselt, Hasselt, Belgium

Correspondence

LabanMusinguzi, National Fisheries Resources

Research Institute (NaFIRRI), Jinja, Uganda.

Email: musinguzilaban@gmail.com;

labanmusinguzi@firi.go.ug

Funding information

Belgian Development Cooperation and

Humanitarian Aid (DGD); Belgian Science

Policy Office (BELSPO), Grant/Award

Numbers: B2/202/P1/KeaFish,

BR/154/A1/HIPE

Abstract

Poor and unreliable knowledge of the status of freshwater fisheries limits their inclu-

sion in governance processes, thereby impeding effectivemanagementmeasures. This

threatens the livelihoods of people, particularly in developing countries. Improved

knowledge is required to draw the attention of policymakers and stimulate effective

management measures to accelerate the sustainability of the freshwater fisheries. In

line with this requirement, this paper provides the state of knowledge of the aquatic

ecosystem and fisheries of the Lake Edward system, East Africa, focusing on lakes

Edward,George and theKazinga channel. The state of knowledgewas accomplishedby

reviewing existing data and information on aspects of primary productivity and water

quality, invertebrates, fish fauna, fish life history and ecology, and fisheries. The water-

bodies have been monitored since the 1930s, albeit sporadically, providing data on all

the above aspects but with substantial temporal gaps. Adequate and updated data and

information exist on the water quality status of the water bodies, extant aquatic taxa

(including fishes) and fish catches but with uncertainties in the latter. Data and infor-

mation gaps exist on the abundance of biotic communities, fish life history, quantitative

trophic ecology and fisheries management reference points. The aggregated data and

information can directly support decisions for fisheries management. We recommend

regular monitoring to fill the data and information gaps, more comprehensive stock

assessments and the development of aquatic ecosystemmodels.

KEYWORDS

Democratic Republic of the Congo, fisheries management, freshwater, inland fisheries, Lake
George, small-scale fisheries, stock assessment, Uganda

1 INTRODUCTION

Inland fisheries require effective management approaches to achieve

the targets of the sustainable development goals (SDGs). Such

approaches are urgent in Africa because inland fisheries are intrin-

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided

the original work is properly cited.

© 2023 The Authors. Aquaculture, Fish and Fisheries published by JohnWiley & Sons Ltd.

sically linked to food security and income. Inland fisheries on the

continent employ ∼ca. 5 million people and contribute 0.33% to the

continent’s GDP, with a gross added value of ∼US$6.3 billion (de

Graaf & Garibaldi, 2014). The need for effective management of inland

fisheries in Africa is consistent with global, regional and national

Aqua. Fish & Fisheries. 2024;4:1–25. wileyonlinelibrary.com/journal/aff2 1

https://orcid.org/0000-0001-7915-3218
mailto:musinguzilaban@gmail.com
mailto:labanmusinguzi@firi.go.ug
http://creativecommons.org/licenses/by/4.0/
https://wileyonlinelibrary.com/journal/aff2
http://crossmark.crossref.org/dialog/?doi=10.1002%2Faff2.140&domain=pdf&date_stamp=2023-11-22


2 MUSINGUZI ET AL.

development strategies. Although the SDGs do not directly cover

inland fisheries (Cooke et al., 2016), some targets inmost SDGs, such as

14 (life below water), 2 (zero hunger), 15 (life on land) and 12 (respon-

sible consumption and production), are applicable. For instance, SDG

14 has targets to end pollution, eutrophication and overfishing, which

are all relevant to inland fisheries (UnitedNations, 2017). In Africa, rel-

evant strategies include the African Union’s 2063 agenda, the Africa

blue economy strategy and the Pan-African fisheries and aquaculture

policy frameworkand reformstrategy (AUC, 2015;NEPAD&AU-IBAR,

2016). Numerous other applicable strategies exist at the national level

to operationalize their global and regional equivalents. All these strate-

gies aspire to use water resources sustainably for food security and

wealth creation.

Achieving the stipulated targets in all the development strategies

requires tools to set priorities, allocate resources, stimulate action and

measure progress (SDSN, 2015). For inland fisheries, priority require-

ments are improved knowledge and effective management measures

(Cooke et al., 2016; FAO & MSU, 2016). Knowledge is particularly an

integral part for achieving the targets because it is the basis not only

for inland fisheries to feature in governance processes, but also for

effective managementmeasures (Cooke et al., 2016).

Unlike most marine resources, the knowledge of aspects of inland

fisheries, such as trophic interactions, the status of fish stocks and

the magnitude and impact of threats, is scanty. Ultimately, inland fish-

eries are often forgotten in critical governance processes, and when

management occurs, it is based on unreliable information (Cooke

et al., 2016). This state hinders effective management and undermines

progress on the SDGs and associated policies.

Given that knowledge is vital for the sustainability of inland fish-

eries, we conducted a literature review to aggregate available data and

information to establish the state of knowledge of the aquatic ecosys-

tem and fisheries, and identify knowledge gaps in the Lake Edward

system focusing on lakes Edward, George and the Kazinga channel.

Located in East Africa, these waterbodies are among the most pro-

ductive freshwater systems (Beadle, 1981). Fisheries on these water

bodies are vital to the riparian rural communities (Bassa et al., 2014;

Lubala et al., 2018). The state of knowledge is useful for guiding

decisions for fisheries management and development. The data and

information aggregated could support aquatic ecosystem modelling

and comprehensive fish stock assessments, strongly improving our

understanding of the aquatic ecosystem and fisheries, thus stimulating

more effectivemanagementmeasures.

2 METHODS

2.1 Lakes Edward and George

The Lake Edward system is a watershed encompassing lakes Edward

and George as the main waterbodies, numerous crater lakes, rivers

and streams (Figure 1). The two lakes are connected by the 40 km

long Kazinga Channel. The system (∼29,000 km2) is transboundary:

Lake George (250 km2) and the Kazinga Channel are entirely situated

within Uganda, whereas Lake Edward (2325 km2) is shared between

Uganda (29%) and the Democratic Republic of the Congo (DRC) (71%).

The whole of Lake Edward and the Kazinga Channel and a larger

part of Lake George are surrounded by protected areas (Queen Eliza-

beth National Park in Uganda and Virunga National Park in the DRC).

The Lake Edward system drains into Lake Albert through the Sem-

liki River. However, the exchange of fish species between the system

and Lake Albert is effectively limited by the Semliki rapids on the river

(Greenwood, 1976a).

The Lake Edward system is important for freshwater biodiversity

and fisheries. The system is the fourth largest among the African great

lakes in terms of fish species richness (Snoeks, 2000). Lakes Edward,

George and theKazinga channel, which do not depend on fish stocking,

support about 23,000 fishers in the two riparian countries. In Uganda,

the waterbodies are the fourth most important producers of fish after

lakes Victoria, Kyoga and Albert. In the DRC, Lake Edward is a major

contributor to inland fish production, with its annual catches placing it

among the top 5major fish-producing inland water bodies in the coun-

try (Breuil & Grima, 2014). Other key features of the system and the

waterbodies within the systemwere described in detail by Decru et al.

(2020) and Stoyneva-Gärtner et al. (2020), Rumes et al. (2011).

2.2 Approach and scope

This reviewwas based on literature to aggregate data and information

on biophysical aquatic ecosystem indicators, invertebrate commu-

nities, fish fauna, fish life history and ecology, and fisheries. The

reviewwas conducted to guide fishery development andmanagement.

Aspects of the biophysical aquatic ecosystem included physical and

chemical indicators of water quality and primary production. Aspects

of fish life history and ecology focused mainly on length–weight

relationships, reproductive biology, growth parameters and trophic

ecology. For fisheries, the focus was on catches, species composition in

the catches, fishing effort and fisheries management reference points.

A literature search was conducted for published papers in the web

of science (https://login.webofknowledge.com), AquaDocs (https://

aquadocs.org) and the published resources of the Food Agricultural

Organization of theUnitedNations (FAO) (http://www.fao.org/fishery/

publications/en). These were searched using terms including Lake

Edward, Lake George, Lake George AND Lake Edward and the Lake

Edward system. However, these terms would be refined if necessary

to narrow the search, for instance, to one of the aspects covered in

this review. All relevant literature was retained. Relevance was based

on whether the resources retrieved covered the aspects of interest,

that is biophysical aquatic ecosystem indicators, invertebrate commu-

nities, fish fauna, fish life history and ecology, and fisheries. AquaDocs

is a global repository of published and unpublished research, con-

tributed by members of the International Association of Aquatic and

Marine Science Libraries and Information Centres (IAMSLIC) and the

International Oceanographic Data and Information Exchange (IODE).

It is a source of literature from, for example research project reports

and annual reports of fisheries departments, which do not exist in

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense

https://login.webofknowledge.com
https://aquadocs.org
https://aquadocs.org
http://www.fao.org/fishery/publications/en
http://www.fao.org/fishery/publications/en


MUSINGUZI ET AL. 3

F IGURE 1 Lake Edward system indicating the location of lakes Edward, George and the interconnecting Kazinga channel, fishing villages and
protected areas. Shapefiles of protected areas were obtained fromUNEP-WCMC and IUCN (2021) and those of waterbodies from Lehner and
Grill (2013).

academic journals. AquaDocs provided access to reports from the past

Game and Fisheries Department in Uganda and the National Fish-

eries Resources Research Institute (NaFIRRI), Uganda, which conducts

research on water bodies in the system. The publications retained for

review (Musinguzi, 2023) were 95 including 47 peer reviewed publica-

tions equivalent to 49.5% of the total (43 journal papers and 4 books

or book chapters), and 48 publications of grey literature (50.5%). Most

of the publications of grey literature (64.6%) were from AquaDocs,

followedby29.2% fromother internet sources, and6.3% fromtheFAO.

From the NaFIRRI, we also obtained data from catch assessment

surveys (CAS), and other fishery-dependent and fishery-independent

surveys that are conducted to examine the status of fish stocks. The

data fromCASwas analysed todetermine annual catches, fishing effort

and species composition in the catch. The data was available from

2000 to 2019, but with gaps within years for lakes Edward (2006–

2008, 2011–2013, 2017, 2019), George (2000–2001, 2011–2013,

2017, 2019) and theKazingaChannel (2000, 2011–2013, 2017, 2019).

Details on the design of the CAS are available in Bassa et al. (2014).We

used weight for each species or species group in catches to generate

daily catch rates for vessel gear combinations (kg/boat/day). The aver-

age number of fishing days in a week, available from the CAS, was used

to determine the number of fishing days in a year to estimate annual

catch rates from the daily catch rates. The number of boats obtained

from frame surveys (surveys that provide data on fishers, fishing gear

and boats, and landing site facilities) was used to raise the annual catch

rates to annual catches.

3 RESULTS

3.1 Primary production and water quality as
indicators of aquatic ecosystem productivity and
health

Physical and chemicalwater quality indicators are indicators of aquatic

ecosystem productivity and health (Carlson & Simpson, 1996). Table 1

presents values for availablewater quality indicators from past studies

on lakes Edward, George and the Kazinga Channel.

The earliest studies on water quality occurred in 1921 and 1931

(Worthington, 1932). Although most of the indicators used at the time

are not in use anymore (Binding et al., 2007; Carlson & Simpson, 1996),

qualitative observations and measurements of Secchi depth (water

transparency) provide insights into how productive the water bodies

were at the time. Water in Lake George and the Kazinga Channel was

depicted as greendue to the presence ofmore phytoplankton than that

of Lake Edward. The water transparency was 0.4 m in Lake George,

and the Kazinga Channel, 1.4 m in Katwe bay (Lake Edward) and 2.2–

2.8 m in offshore sites of Lake Edward (Figure 1) (Worthington, 1932).

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



4 MUSINGUZI ET AL.

T
A
B
L
E
1

P
hy
si
ca
la
n
d
ch
em

ic
al
p
ar
am

et
er
s
o
fl
ak
es

E
d
w
ar
d
,G

eo
rg
e
an
d
th
e
K
az
in
ga

C
h
an
n
el
w
it
h
:c
o
n
d
u
ct
iv
it
y
(σ
),
te
m
p
er
at
u
re

(T
),
Se
cc
h
id
ep

th
(S
D
),
to
ta
ld
is
so
lv
ed

so
lid

s
(T
D
S)
,d
is
so
lv
ed

ox
yg
en

(D
O
),
ch
lo
ro
p
hy
ll
a
(c
h
l-
a)
,t
o
ta
lp
h
o
sp
h
o
ru
s
(T
P
),
n
it
ra
te

(N
O

3
−
)a
n
d
so
lu
b
le
re
ac
ti
ve

p
h
o
sp
h
o
ru
s
(S
R
P
).

W
at
er

b
o
d
y

σ
(μ
S/
cm

)
T
(◦
C
)

SD
(c
m
)

T
D
S
(m

g/
L)

D
O
(m

g/
L)

C
h
la

(μ
g/
L)

T
P
(μ
g/
L)

N
O

3
−
(μ
g/
L)

SR
P
(μ
g/
L)

Y
ea
r
o
f

sa
m
p
lin

g
R
ef
er
en

ce

La
ke

E
d
w
ar
d
/p
ar
ts
o
fL
ak
e
E
d
w
ar
d

–
–

1
8
0
–
2
8
0

–
–

–
–

–
–

–
W
o
rt
h
in
gt
o
n

(1
9
3
2
)

–
1
8
.7
–
3
0
.4

1
9
0
.0
–
3
0
0
.0

–
6
.0
–
9
.0

–
–

–
–

1
9
5
2
/1
9
5
3

V
er
b
ek
e

(1
9
5
7
)

K
at
w
e
b
ay

2
5
0
–
3
5
0

1
9
5
2
/5
3

V
er
b
ek
e

(1
9
5
7
)

–
–

–
–

–
–

9
9
.2

9
5
.1

1
9
5
4

Ta
lli
n
g
(1
9
6
3
)

–
–

–
–

–
1
2
7
.0

1
0
5
.4

5
7
.0

1
9
6
1

Ta
lli
n
g
an

d

Ta
lli
n
g

(I
9
6
5
)

8
3
0

2
4
–
2
6

1
3
0

7
0
0
–
8
0
0

5
.0
–
7
.5

–
–

–
2
4
0
.0
±
4
0
.0

1
9
7
6
/7
7

B
u
ge
ny
i

(1
9
7
9
,

1
9
8
2
)

–
2
7
.5

–
–

6
.2

–
–

>
6
.2

5
7
.0

1
9
9
5

Le
h
m
an

et
al
.

(1
9
9
8
)

–
2
6
.6
±
0
.5

1
0
5
±
2
7

–
2
1
.3
±
2
2
.8

5
8
.9
±
9
.2

–
1
0
.6
±
5
.2

2
0
0
8
/2
0
0
9

P
o
st
e
et

al
.

(2
0
1
3
)

9
2
0
.5
–
9
3
9
.4

2
5
.5
–
2
6
.9

–
–

7
.6

8
.2

3
4
.2
–
5
2
.8

3
4
.6

1
6
.4

2
0
0
8

N
aF

IR
R
I

(2
0
0
8
)

7
1
8
.3
±
4
.2

2
7
.5
5
±
0
.1
9

1
0
8
.5
6
±
1
0
.7

3
6
1
.1
2
±
1
.4
7

5
.1
6
±
0
.3
6

-
-

-
-

2
0
1
1
/2
0
1
2

M
b
al
as
sa

et
al
.

(2
0
1
4
)

6
8
5
.8
5
±
1
5
0
.7

2
3
.7
6
±
1
.7
1

1
3
0

–
4
.6
3
±
0
.5
5

–
–

–
–

2
0
1
3

B
ag
al
w
a
et

al
.

(2
0
1
4
)

La
ke

E
d
w
ar
d
,

ex
cl
u
d
in
g

K
at
w
e
b
ay

8
6
2
(8
1
9
–
8
8
4
)

2
6
.1 (2
5
.5
–
2
7
.8
)

1
5
0 (6
8
–
2
3
2
)

-
6
.1
(0
.8
–
9
.9
)

6
.8
±
2
.7
–

1
0
.7
±
5
.6

-
9
9
.2 (0
–
5
3
9
.4
)

1
2
3
.5

(0
–
3
7
9
.9
)

2
0
1
6
/2
0
1
8

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

P
el
ag
ic

(1
0
–
8
5
m
)

–
–

1
0
7
–
2
3
2

–
–

8
.0
±
3
.9

6
1
.9
±
2
7
.9

8
6
.8
±
9
9
.2

–
–

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

(C
o
n
ti
n
u
es
)

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 5

T
A
B
L
E
1

(C
o
n
ti
n
u
ed

)

W
at
er

b
o
d
y

σ
(μ
S/
cm

)
T
(◦
C
)

SD
(c
m
)

T
D
S
(m

g/
L)

D
O
(m

g/
L)

C
h
la

(μ
g/
L)

T
P
(μ
g/
L)

N
O

3
−
(μ
g/
L)

SR
P
(μ
g/
L)

Y
ea
r
o
f

sa
m
p
lin

g

R
ef
er
en

ce

Li
tt
o
ra
ls
it
es

(0
–
6
.1
m
)

–
–

6
8
–
1
4
9

–
–

8
.4
±
3
.7

–
–

–
–

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

6
9
2
(5
8
9
–
7
9
1
)

2
6
.5 (2
5
.7
–
2
8
.8
)

4
5
(1
7
–
6
9
)

–
9
.1 (8

.1
–
1
1
.6
)

5
0
.4
±
3
1
.8

–
3
1
.0 (0
.0
–
7
4
.2
)

4
1
.0
(1
2
.1
–

1
4
5
.9
)

2
0
1
6
/2
0
1
8

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

La
ke

G
eo

rg
e

–
–

4
0

–
–

–
–

–
–

–
W
o
rt
h
in
gt
o
n

(1
9
3
2
)

–
–

–
–

–
4
1
2
.0

–
<
5
7
.0

1
9
6
1

Ta
lli
n
g
an

d

Ta
lli
n
g

(1
9
6
5
)

2
2
7
.5
±
1
8
.9
*

–
–

–
–

–
–

n
.d

1
7
1
.0
–

3
9
8
.9

1
9
6
7
/6
8

V
in
er

(1
9
6
9
)

2
0
0

2
5
–
3
5

–
–

–
–

–
–

1
9
6
7
/6
8

D
u
n
n
et

al
.

(1
9
6
9
)

2
3
0

2
3
–
3
0

2
0

1
6
0
–
2
0
0

4
.0
–
8
.0

–
–

–
5
2
0
.0
±
1
3
0

1
9
7
6
/7
7

B
u
ge
ny
i

(1
9
7
9
,

1
9
8
2
)

–
–

–
–

–
–

–
–

1
2
3
.5

1
9
9
5

Le
h
m
an

et
al
.

(1
9
9
8
)

–
–

–
–

7
±
3
.8

–
–

–
–

2
0
0
1
/2
0
0
3

O
w
o
r
et

al
.

(2
0
0
7
)

–
2
6
.4
±
1
.0

3
7
±
8

–
–

1
3
8
.0
±
3
9
.1

1
8
6
.5
±
2
6
.2

–
9
.9
±
4
.9

2
0
0
8
/2
0
0
9

P
o
st
e
et

al
.

(2
0
1
3
)

2
5
0
(2
3
7
–
2
7
6
)

2
5
.7 (2
4
.5
–
2
9
.5
)

2
8
(2
4
–
3
2
)

–
9
.0 (5

.0
–
2
2
.5
)

1
9
0
.7
±
1
1
4
.3

–
1
5
5
.0
(5
5
.8
–

3
0
3
.8
)

6
3
.2
(5
.8
–

1
6
2
.2
)

2
0
1
6
/2
0
1
8

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

K
az
in
ga

ch
an

n
el

4
0

–
–

–
–

–
–

–
W
o
rt
h
in
gt
o
n

(1
9
3
2
)

(C
o
n
ti
n
u
es
)

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



6 MUSINGUZI ET AL.

T
A
B
L
E
1

(C
o
n
ti
n
u
ed

)

W
at
er

b
o
d
y

σ
(μ
S/
cm

)
T
(◦
C
)

SD
(c
m
)

T
D
S
(m

g/
L)

D
O
(m

g/
L)

C
h
la

(μ
g/
L)

T
P
(μ
g/
L)

N
O

3
−
(μ
g/
L)

SR
P
(μ
g/
L)

Y
ea
r
o
f

sa
m
p
lin

g

R
ef
er
en

ce

5
0
0

2
4
–
2
7

–
1
8
0
–
3
0
0

6
.5
–
7
.0

–
–

4
0
0
.0
±
8
0
.0

1
9
7
6
/7
7

B
u
ge
ny
i

(1
9
7
9
,

1
9
8
2
)

2
4
1

–
–

–
7
.4

6
1
.5

1
5
9
.8

4
6
.5

1
3
.2

2
0
0
8

N
aF

IR
R
I

(2
0
0
8
)

–
2
5
.5
±
2
.1

5
0
±
2
5

–
–

6
6
.3
±
4
6
.2

1
2
9
.1
±
5
4
.7

–
1
0
.3
±
6
.1

2
0
0
8
/2
0
0
9

P
o
st
e
et

al
.

(2
0
1
3
)

2
6
6
(2
0
2
–
7
2
6
)

2
6
.4 (2
5
.2
–
2
8
.6
)

3
0
(1
9
–
4
4
)

–
8
.4 (5

.9
–
1
5
.4
)

1
2
8
.9
±
1
1
4
.3

–
6
2
.0
(0
.0
–

2
2
3
.2
)

6
4
.0
(1
2
.1
–

1
3
5
.7
)

2
0
1
6
/2
0
1
8

St
oy
n
ev
a-

G
är
tn
er

et
al
.

(2
0
2
0
)

N
ot
e:
*V
al
u
e
w
as

ag
gr
eg
at
ed

fr
o
m

va
lu
es

fo
r
w
et

an
d
d
ry

se
as
o
n
s
in

1
9
6
7
an

d
1
9
6
8
.V

al
u
es

in
μm

o
l/
L
fo
r
T
P,
SR

P
an

d
N
O

3
−
w
er
e
co
nv
er
te
d
to
μg

/L
u
si
n
g
u
n
it
co
nv
er
si
o
n
s
o
f
th
e
In
te
rn
at
io
n
al
C
o
u
n
ci
lf
o
r
th
e

E
xp
lo
ra
ti
o
n
o
ft
h
e
Se
a
(I
C
E
S)
(h
tt
p
s:
//
w
w
w
.ic
es
.d
k/
d
at
a/
to
o
ls
/P
ag
es
/U

n
it
-c
o
nv
er
si
o
n
s.
as
px
).

A
b
b
re
vi
at
io
n
:n
.d
,n
o
n
-d
et
ec
ta
b
le
.

Using data collected in 1952 and 1953 in Lake Edward, Verbeke (1957)

reported water transparency as 1.9–3 m in offshore sites, 0.5 m in Vit-

shumbi Bay, 0.25–0.35 m in Kamande and Katwe bays (see location of

these sites in Figure 1) and 0.25–0.5 m in the deltas of rivers. Compar-

ing these values to the trophic categories, where values >4 m denote

oligotrophic, 2–4 m mesotrophic, 0.5–1.99 m eutrophic and <0.5 m

hypertrophic states (Carlson, 2007; Forsberg&Ryding, 1980), suggests

that Lake George, Kazinga Channel and most bays and river mouths in

Lake Edwardwere hypertrophic.

Stoyneva-Gärtner et al. (2020) conducted a comprehensive study of

water quality and primary production in these water bodies, offering

the most recent observations. Values of lake-wide water transparency

from Stoyneva-Gärtner et al. (2020), excluding Katwe bay, indicated

that Lake Edward is eutrophic, with an average water transparency

of 1.5 m. However, the range was 0.68–2.32 m, suggesting that some

areas are mesotrophic. On the other hand, the mean water trans-

parency in Katwe bay was 0.45 m, indicating that the bay was hyper-

trophic and as productive as Lake George (0.28 m) and the Kazinga

Channel (0.30 m), in agreement with past studies (Table 1). The hyper-

trophic state of Katwe bay may be attributed to the influence of the

Kazinga Channel (Verbeke, 1957;Worthington, 1932). Measurements

of chlorophyll a (chl-a) also demonstrated high productivity in Katwe

bay, Lake George and the Kazinga Channel, with recent values indi-

cating increased productivity compared to historical values (Table 1).

Based on the values of chl-a, the Kazinga Channel and Lake George

are hypertrophic, whereas Katwe bay is eutrophic with some hyper-

trophic parts. The rest of Lake Edward is mainly mesotrophic, but with

eutrophic littoral zones.

Bugenyi (1982) attributed the higher productivity in Lake George

and the Kazinga Channel to a higher concentration of phosphates

compared to Lake Edward. Phosphorous concentration in thesewater-

bodies is vital for primary production because they are limited in

Nitrogen (Ganf & Viner, 1973; Stoyneva-Gärtner et al., 2020). In the

past, Lake George and the Kazinga Channel appeared to have more

soluble reactive phosphorous (SRP) than Lake Edward (Bugenyi, 1982;

Lehman et al., 1998; Table 1). This trend has changedwith SRP concen-

tration being highest in Lake Edward compared to LakeGeorge and the

Kazinga Channel, where the concentration is reduced by demand by

the higher biomass of phytoplankton (Stoyneva-Gärtner et al., 2020).

Primary production in the waterbodies supports a rich phyto-

plankton community. Stoyneva-Gärtner et al. (2020) identified 248

taxa of the phytoplankton community belonging to Cyanoprokaryota,

Euglenophyta, Pyrrhophyta, Cryptophyta, Ochrophyta, Tribophyceae,

Chrysophyceae, Synurophyceae, Bacillariophyceae, Chlorophyta and

Streptophyta. Confirming the trophic status described above, mea-

surements of absolute primary production showed that Lake George

and the Kazinga Channel have higher phytoplankton biomass than

Lake Edward. Burgis et al. (1973) estimated the mean phytoplankton

biomass of LakeGeorge as 46.8 gm−2. Ganf andViner (1973) estimated

a mean of 30 g C m−2 and a range of 20–40 g C m−2 for the same lake.

The phytoplankton biomass of the Kazinga Channel was probably sim-

ilar. In contrast, primary production in Lake Edward was lower, ranging

from1.5 to 12 gCm−2 (Lehman et al., 1998). Unlike the data in Lehman

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense

https://www.ices.dk/data/tools/Pages/Unit-conversions.aspx


MUSINGUZI ET AL. 7

(1998) that was collected within only 1 month, the data in Burgis et al.

(1973) andGanf andViner (1973)was obtained over a period of 1 year,

and thus spans seasons.

Cyanobacteria (Cyanoprokaryota) are the most dominant group of

phytoplankton (Stoyneva-Gärtner et al., 2020). In Lake Edward, this

group is responsible for approximately 90% of the primary produc-

tion (measured as chl-a concentration) inKatwebay and approximately

60% in the rest of the lake. Ochrophyta is the second most domi-

nant group responsible for 24.7%–27.7% of the primary production

in the lake excluding Katwe bay, where the group contributes 7.7%.

In Lake George and the Kazinga Channel, nearly all primary produc-

tion is by Cyanobacteria, comprising 98.6% and 96.1%, respectively.

The dominance of Cyanobacteria corresponds with earlier studies.

In the late 1960s, Cyanobacteria were responsible for 80% of the

mean phytoplankton biomass of Lake George biomass (Burgis et al.,

1973). Thesewater bodies are limited in nitrogen (Ganf &Viner, 1973).

Cyanobacteria are dominant because of their ability to fix atmospheric

nitrogen, tolerance to low dissolved oxygen, their higher efficiency in

light absorption and nutrient assimilation, and their tendency to limit

the availability of light to other phytoplankton (Burgus et al., 1973;

Ganf & Viner, 1973; Stoyneva-Gärtner et al., 2020).

Water quality and primary production are relatively stable in Lakes

Edward and George, and the Kazinga Channel (Ganf & Viner, 1973).

However, some temporal and spatial differences may occur. Stoyneva-

Gärtner et al. (2020) demonstrated differences in water quality and

primary production among littoral sites, pelagic sites andwithin Katwe

bay in Lake Edward. Dissolved nutrients (SRP and Dissolved Inorganic

Nitrogen [DIN]) and chl-a exhibited significant differences between

rainy and dry seasons. Unlike Lake George and the Kazinga Channel,

which are shallow, thermal stratification in Lake Edward is eminent and

affects water quality.Worthington (1932) showed that temperature in

Lake Edward was uniform within 10 m, dropping slightly and remain-

ing uniform between 10 and 40 m, and then dropping by 1◦C beyond

60 m. This stratification had substantial effects on other water qual-

ity conditions and biotic communities. For instance, below 50–60 m,

water was anoxic. No zooplankton was found below 60 m, where only

Chaoborus larvae (macroinvertebrates) were found. A few fish were

found to enter the hypolimnion. Stoyneva-Gärtner et al. (2020) found

uniformwater quality within 15–20m throughout the year and expan-

sion of the mixed layer to 55 m during the dry season, suggesting that

stratification in the lake has weakened since the 1920s.

3.2 Aquatic invertebrates

3.2.1 Macroinvertebrates

For Lake George, information onmacroinvertebrates wasmainly avail-

able from the International Biological Programme for the period 1966

and 1971 (Green, 2009; Greenwood, 1976b). The lake has a ben-

thic macroinvertebrate community composed of Gastropoda, Bivalvia,

Chaoborus spp., Oligochaeta, Chironomidae, Hydracarina, Ostracoda,

Ephemeroptera, Nematoda and Trichoptera (Burgis et al., 1973; Dar-

lington, 1977; Greenwood, 1976a). In the open waters of the lake,

Chaoborus spp., Chironomidae (Chironomus spp. and Procladius sp.),

Oligochaeta and Ostracoda are the main groups (Burgis et al., 1973).

These groups are species-poor due to the unstable, soft and deoxy-

genated mud in the lake (Burgis et al., 1973). The abundance and

species richness of groups of macroinvertebrates, apart from Ostra-

coda, increase from the middle of the lake, where mud is dominant,

towards the inshore habitats that have firmer, less disturbed andmore

diverse benthic habitats with sand, clay and gravel substrates (Bur-

gis et al., 1973; Darlington, 1977; Greenwood, 1976b). Some taxa like

Gastropoda, Bivalvia, Ephemeroptera, Nematoda, Oligochaeta and Tri-

choptera, and some taxa of Chironomidae, such as Chironomus imicola

and Tanytarsus sp., are infrequent in the open lake (Burgis et al., 1973;

Darlington, 1977).

Burgis et al. (1973) reported the total absolute biomass of macroin-

vertebrates as 0.519 gm−2 based on mid-lake samples. Chaoborus spp.

comprised 41.2% of the total biomass (0.214 gm−2). Darlington (1977)

also estimated the total biomass of macroinvertebrates in the open

lake, agreeing with Burgis et al. (1973) on the dominance of Chaoborus

spp. However, the estimate of Chaoborus spp. by the former was higher

at 0.348 gm−2, comprising 35.7% of the biomass. Other taxa were

recorded by Darlington (1977) as follows: Oligochaeta (0.186 gm−2),

C. imicola (0.178 gm−2), Procladius brevipetiolatus (0.171 gm−2) and

Ostracoda (0.92 gm−2). The dominance of the inshore habitats was by

oligochaeta, with a total biomass of 0.533 gm−2, equivalent to 41.6%of

the total biomass of macroinvertebrates (Darlington, 1977).

In Lake Edward, the community of macroinvertebrates was

examined comprehensively in studies undertaken between 1930

and 1960. Synthesized in Green (2009), the studies grouped the

macroinvertebrates in the lake into Turbellaria, Ostracoda, Decapoda,

Hemiptera, Trichoptera, Coleoptera, Diptera, Mollusca, Oligochaeta,

Nematoda, Hirudinea, Acarina, Collembola, Ephemeroptera, and

Odonata. Coleoptera (69 species), Hemiptera (30 species), Ostracoda

(20 species) and Trichoptera (15 species) were the most species-rich

groups.

A single sampling event conducted in January 2008 as part of

Environmental Impact Assessments for oil exploration projects pro-

vided the most recent information on macroinvertebrates of the lake

(NaFIRRI, 2008). Macroinvertebrates were recorded in seven broad

groups: Gastropoda, Bivalvia, Diptera, Ephemeroptera, Odonata, Tri-

choptera and Oligochaeta, differing from past observations due to the

absence of some groups like Caridina spp. and Ostracoda. Ostracoda

are everywhere in the lake and can be observed in the stomachs of

some fish species (Vranken, Steenberge, Kayenbergh et al., 2020). This

groupwas probablymissed in these samples because its individuals are

smaller than the size spectrum of groups that are considered macroin-

vertebrates in these samples. Gastropoda, Diptera, and Oligochaeta

were the most widely distributed, with at least one representative

taxon in all six sampled sites for the first two groups, and five sites for

Oligochaeta. Gastropoda and Diptera were the most taxa-rich groups

with four and seven representative taxa, respectively. Other groups

were each represented by one taxon. Ephemeroptera, Odonata, and

Trichoptera were each recorded in only one of the six sampled sites,

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



8 MUSINGUZI ET AL.

suggesting limiteddistributionandabundance in the lake. Taxa richness

did not differ remarkably among sites, and between offshore and near-

shore regions. Density (individuals per square meter) suggested that

Diptera, comprising Chaoborus spp. and Chironomidae, was the most

dominant group.

3.2.2 Zooplankton

Zooplankton in Lakes Edward and George have been studied from

the perspective of three major groups: Copepoda, Rotifera and Clado-

cera. Dunn et al. (1969) reported Copepoda as the most abundant in

LakeGeorge (72% relative abundance), followed by Rotifera (25%) and

Cladocera (3%). Burgis et al. (1973) observed that, generally, the abun-

dance of zooplankton peaked during the wet season and was higher in

open water than in inshore habitats, except for Rotifera, whose abun-

dancewas higher in the inshore parts of the lake. The lower abundance

of zooplankton in inshore habitats was attributed to intense grazing by

fish, whose biomass is higher in the inshore habitats (Gwahaba, 1975).

The absolute biomass of zooplankton in the lake was estimated as

0.488 gm−2, with copepods comprising more than 80% of the biomass

(Burgis et al., 1973).

In Lake Edward, copepods also dominate the zooplankton commu-

nity, comprising 40%–60% of the density (number of individuals per

square meter) at sites of varying depths (8.5–29.5 m). Unlike Lake

George, Cladocera follows in abundance, comprising 6%–36% of the

density. Rotifera is the least abundant with 0.2%–2% of the density

(Green, 2009). These observations are corroborated by observations

from the most recent study, which indicated that Cladocera and Cope-

poda exhibited a lake-wide distribution, whereas Rotifera were rare

(NaFIRRI, 2008). Copepoda comprised 76%–97% of the abundance

at inshore and offshore sites, followed by Cladocera (1%–17%) and

Rotifera (1%–12%).However, Rotiferawas themost diverse groupwith

11 taxa (nine species and two genera), followed by Cladocera with six

(five species and one genus) andCopepodawith one genus, one species

in addition to nauplius larvae and copepodites. Estimates of total zoo-

plankton biomass for Lake Edward are available from Lehman et al.

(1998), as 0.66, 2.21, and 1.28 g C m−2 at sites at 4, 18 and 25 m from

the shoreline, respectively.

3.3 Fish species

3.3.1 Species extant in Lakes Edward and George

Knowledge of fish species in an ecosystem is important for effec-

tive fisheries management. Lakes Edward and George are rich in fish

species, including a large assemblage in the genus: Haplochromis. The

list of the species in the water bodies was based on recent reviews and

descriptions of fish species in the Lake Edward system (Table 2). Decru

et al. (2020) reviewed literature, FishBase andmuseumcollections, list-

ing34 fish species in21genera (excludingHaplochromis) and10 families

in the system. A recent review of Enteromius re-identified specimens

of Enteromius perince and Enteromius stigmatopygus as Enteromius cf.

mimus and Enteromius alberti, respectively (Maetens et al., 2020). Based

on these studies, lakes Edward and George have 19 non-Haplochromis

species in 8 families and 15 genera occurring in the lakes. Although

the lakes share most of these fish species, Laciris pelagica (endemic

to the open water of Lake Edward), Labeo forskalii and Heterobranchus

longifilis are not known to be in Lake George. H. longifilis has not been

reported in Lake Edward since 1956 (Hulot, 1956). For this reason,

the presence of the species in the lake can be classified as possibly

extant.

More than 60 species of Haplochromis spp. are estimated to occur

in lakes Edward, George and the Kazinga Channel (Greenwood, 1991;

Snoeks, 2000; Vranken et al., 2019). However, only 40 are described

(Table 2), presenting a substantial knowledge gap in the ichthyofauna

of the system.However, efforts are underway to describemore species

(Vranken, Steenberge, Kayenbergh et al., 2020, Vranken, Steenberge,

Snoeks, 2020, Vranken, Steenberge, Balagizi et al., 2020, Vranken

2022).

3.3.2 Habitat use, distribution, and relative
abundance of the fish species in Lakes Edward and
George

Notes on habitats and the distribution of the fish species in the two

lakes are presented in Table 2.Worthington (1932) and Poll andDamas

(1935) provided the earliest insights into the habitat use, distribution

and abundance of the fish species. Fish species, including Clarias lio-

cephalus, Mormyrus kannume, and those belonging to Cyprinidae, other

than Labeobarbus altianalis, were depicted as being less abundant or

rare. Only a few individuals were recorded for these species at the

time. Apart from L. forskalii, these species predominantly use inshore

areas, vegetated fringes and river mouths as habitats. L. forskalii was

restricted to rocky deep open waters in Lake Edward, such as those

close to the western shores in the DRC. Other species recorded at

the time (L. altianalis, Bagrus docmak, Oreochromis niloticus, Oreochromis

leucostictus, Clarias gariepinus, Protopterus aethiopicus, L. pelagica, and

Lacustricola vitschumbaensis)weredepicted as abundant. These species,

apart from L. pelagica, which is restricted to the open deep waters

of Lake Edward, and L. vitschumbaensis, which is restricted to the

inshore areas, especially those associated with river mouths, were

found throughout the lakes. The abundance of these species, in gen-

eral, decreased from the inshore to offshore areas and was highest

in the following habitats: shallow areas, river mouths, sheltered bays

and vegetated and swampy fringes (Worthington, 1932). However, the

western shoreline of Lake Edward has some deep nearshore areas,

lacking thepreferredhabitats of these species. As a result,P. aethiopicus

is absent in these areas and the abundance of others is remarkably low

(Poll & Damas, 1935). Worthington (1932) demonstrated the general

decrease of abundance from the inshore habitats towards the offshore

habitats using O. niloticus in Lake Edward. Catch rates of the species

from experimental gillnets diminished from 120 fish per net at the

mouth of Kazinga Channel (Figure 1) to 2–3 fish per net at a site about

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 9

TABLE 2 Fish species in lakes Edward and George.

Family

Trophic group (for

Haplochromis spp.) Species George Edward

Distribution and habitat

use in Lakes Edward and

George Main diet

Anabantidae Ctenopomamuriei
(Boulenger, 1906)

X X Vegetated fringes and river

mouths

Insect larvae and

crustacea

Bagridae Bagrus docmak (Fabricius,
1775)

X X Abundant in Lake George

and shallowwaters of

Lake Edward; mainly in

closed bays and river

mouths

Fish, detritus and

insects

Cichlidae Astatoreochromis alluaudi
(Pellegrin, 1904)

X X Inshore areas and river

mouths

Insect larvae

**Coptodon zillii (Gervais,
1848)

X X Vegetated fringes Higher plant

materials

Detrivores **Haplochromis aeneocolor
(Greenwood, 1973)

X X Papyrus swamp edges Detritus

*Haplochromis akika
(Lippitsch, 2003)

X X Papyrus shores Detritus (based on

morphology)

*Haplochromis eduardii
(Regan, 1921)

– X Insufficient data Detritus (based on

morphology)

Insectivores *Haplochromis elegans
(Trewavas, 1933)

X X Sandy shoals and papyrus

shores

Insect larvae and

adults

*Haplochromis engystoma
(Trewavas, 1933)

– X Insufficient data Insect larvae (based

onmorphology)

*Haplochromis labiatus
(Trewavas, 1933)

– X Inshore habitats Insect larvae

*Haplochromis lobatus
(Vranken et al., 2020)

– X Inshore habitats Insect larvae

*Haplochromis angustifrons
(Boulenger, 1914)

X X Offshore habitats Insect larvae

*Haplochromis
macropsoides
(Greenwood, 1973)

X X Sublittoral habitats Insect larvae and

adults

*Haplochromis oregosoma
(Greenwood, 1973)

X X Sublittoral habitats Possibly

phytoplankton,

morphology

suggests insects

Haplochromis schubotzi
(Boulenger, 1914)

X X Sublittoral and offshore

habitats

Insect larvae

*Haplochromis
schubotziellus
(Greenwood, 1973)

X X Muddy bays and near

papyrus fringes

Insects

Phytoplanktivores *Haplochromis nigripinnis
(Regan, 1921)

X X Shallow offshore habitats Phytoplankton

*Haplochromis vicarius
(Trewavas, 1933)

– X Insufficient data Phytoplankton

(based on

morphology)

Phytoplanktivores (Algae

scrapers-epilithic)

*Haplochromis serridens
(Regan, 1925)

– X Insufficient data Aufwuchs on rocks

(based on

morphology)

*Haplochromis fuscus
(Regan, 1925)

– X Insufficient data Aufwuchs on rocks

(based on

morphology)

(Continues)

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



10 MUSINGUZI ET AL.

TABLE 2 (Continued)

Family

Trophic group (for

Haplochromis spp.) Species George Edward

Distribution and habitat

use in Lakes Edward and

George Main diet

Phytoplanktivores (Algae

scrapers-epiphytic)

*Haplochromis limax
(Trewavas, 1933)

X X Vegetated shores Aufwuchs on plants

and lake substrate

Molluscivores

(Pharyngeal crushers)

*Haplochromis mylodon
(Greenwood, 1973)

X X Inshore and offshore

habitats

Gastropods and

insects

*Haplochromis pharyngalis
(Poll and Damas, 1939)

X X Rocky shores Gastropods and

some insect larvae)

Molluscivores (Oral

shellers)

*Haplochromis concilians
(Vranken et al., 2020)

– X Inshore habitats over sand Gastropods

*Haplochromis erutus
(Vranken et al., 2020)

– X Inshore and offshore

habitats

Gastropods

*Haplochromis planus
(Vranken et al., 2020)

– X Inshore and offshore

habitats

Ostracods

Zooplanktivores *Haplochromis pappenheimi
(Boulenger, 1914)

X X Upper water layers in

offshore habitats

Zooplankton

Piscivores *Haplochromis mentatus
(Regan, 1925)

X X Mostly in shallowwaters

offshore

Fish (based on

morphology)

*Haplochromis latifrons
(Vranken et al., 2022)

– X Offshore habitats Fish (based on

morphology)

*Haplochromis rex (Vranken
et al., 2022)

X Over sandy substrates Fish (based on

morphology)

*Haplochromis simba
(Vranken et al., 2022)

– X Inshore areas over hard

substrates

Fish (based on

morphology)

*Haplochromis glaucus
(Vranken et al., 2022)

– X Over sandy substrates Fish (based on

morphology)

*Haplochromis aquila
(Vranken et al., 2022)

– X Inshore areas over muddy

substrates

Fish (based on

morphology)

Piscivores (Microdontic) Haplochromis squamipinnis
(Regan, 1921)

X X Offshore, muddy shore and

papyrus fringes

Fish

*Haplochromis kimondo
(Vranken et al., 2022)

– X Over sandy substrates Fish (based on

morphology)

*Haplochromis falcatus
(Vranken et al., 2022)

– X Over sandy substrates Fish (based on

morphology)

*Haplochromis curvidens
(Vranken et al., 2022)

– X Inshore areas Fish (based on

morphology)

*Haplochromis pardus
(Vranken et al., 2022)

– X Inshore areas Fish (based on

morphology)

*Haplochromis quasimodo
(Vranken et al., 2022)

X X Offshore, benthic areas in

shallow and deepwaters

Fish (based on

morphology)

Piscivores (Paedophages) *Haplochromis gracilifur
(Vranken et al., 2019)

– X Inshore waters Fish eggs and larvae

*Haplochromis molossus
(Vranken et al., 2019)

X X Inshore habitats Fish eggs and larvae

*Haplochromis paradoxus
(Lippitsch and Kaufman,

2003)

X X Inshore and offshore areas Fish larvae

*Haplochromis relictidens
(Vranken et al., 2019)

X X Inshore habitats Fish eggs and larvae

*Haplochromis taurinus
(Trewavas, 1933)

X X Inshore habitats Fish eggs and larvae

(Continues)

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 11

TABLE 2 (Continued)

Family

Trophic group (for

Haplochromis spp.) Species George Edward

Distribution and habitat

use in Lakes Edward and

George Main diet

Parasite eaters *Haplochromis eduardianus
(Boulenger, 1914)

X X Inshore and sublittoral

habitats

Presumably parasites

Oreochromis leucostictus
(Trewavas, 1933)

X X Common everywhere in

Lake George and inshore

waters of Lake Edward

Higher plant material

and detritus

Oreochromis niloticus
(Linnaeus, 1758)

X X Found throughout Lakes

Edward andGeorge. In

Lake Edward, the typical

habitat is shallow inshore

waters, found only

occasionally in open

waters and deep steep

western shores

Detritus, higher plant

material, diatoms

and insects

Clariidae Clarias gariepinus
(Burchell, 1822)

X X Found throughout the

lakes, especially in river

mouths and papyrus

fringes

Fish, insects, higher

plant material and

detritus

Clarias liocephalus
(Boulenger, 1898)

X X Papyrus fringes of the

Kazinga Channel and

Lake George

Dipteran larvae and

plant material

Heterobranchus longifilis
(Valenciennes, 1840)

– X Insufficient data Insufficient data

Cyprinidae Enteromius kerstenii
(Peters, 1868)

X X Inshore areas and river

mouths

Diptera larvae

Enteromius cf. mimus
(Boulenger, 1912)

X X Inshore areas and river

mouths

Diptera larvae

Enteromius alberti (Poll,
1939)

X X Inshore areas and river

mouths or sources

Diptera larvae

Labeo forskalii (Rüppell,
1835)

– X Only in Lake Edward. Deep

clear waters close to the

western shore and in

rocky shorelines

Insufficient data

Labeobarbus altianalis
(Boulenger, 1900)

X X Common in river mouths

particularly that of the

Semliki river

Fish, detritus and

insects

Mormyridae Mormyrus kannume
(Forsskål, 1775)

X X Papyrus fringes and river

mouths

Insect larvae

Pollimyrus nigricans
(Boulenger, 1906)

X X Inshore and offshore

waters

Dipteran larvae

Procatopodidae *Laciris pelagica
(Worthington, 1932)

– X Endemic in openwaters Zooplankton

Lacustricola
vitschumbaensis (Ahl,
1924)

X X Inshore and river mouths Dipteran larvae and

emergents

Protopteridae Protopterus aethiopicus
(Heckel, 1851)

X X Vegetated/swampy fringes

and shallow areas

Fish, mollusks,

insects, Ostracoda,

higher plant

material and

detritus

Note: X denotes presence. Notes on habitat and trophic ecology are fromWorthington (1932), Poll and Damas (1935), Greenwood (1973), Gwahaba (1975),

Dunn (1975), Yatuha et al. (2013), Cox (2018), Kusters (2019), Vranken et al. (2019) andVranken, Steenberge, Kayenbergh et al. (2020), Vranken, Steenberge,

Snoeks (2020), Vranken, Steenberge, Balagizi et al. (2020). Species with an asterisk are endemic to the Lake Edward system. Species with two asterisks are

introduced.

Source: The list was adopted fromDecru et al. (2020) andMaetens et al. (2020) for all species apart fromHaplochromis spp. (Cichlidae).Haplochromis spp. were
basedonGreenwood (1973), Vrankenet al. (2019), Vranken, Steenberge,Kayenberghet al. (2020), Vranken, Steenberge, Snoeks (2020), Vranken, Steenberge,

Balagizi et al. (2020) and Vranken et al. (2022).

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



12 MUSINGUZI ET AL.

F IGURE 2 Annual catches of exploited fish species or species groups in lakes Edward, George, and the Kazinga Channel. Each graph shows
catches for a specified species or species group in different water bodies: Lake Edward in Uganda (LE_UG), Lake George (LG), the Kazinga Channel
(KC), Lake Edward (Uganda) and the Kazinga Channel (LE_UG_KC), Lake Edward in the Democratic Republic of the Congo (LE_DRC), lakes Edward
(Uganda), George and the Kazinga Channel (LE_System_UG). Two ormore water bodies are combinedwhere available data was not segregated by
waterbody. Lake Edward is shared between Uganda and the Democratic Republic of Congo (DRC). For more clarity, separate graphs weremade for
each species or species groups (Online Resource 1 Figures S1–S12). Source: Data obtained fromGame and Fisheries Department (1935, 1938,
1947, 1948, 1949, 1950, 1951, 1952, 1953, 1955, 1957, 1958, 1959, 1967, 1968, 1969, 1994), Okaranon and Kamanyi (1989), Fisheries
Department (1971, 1972, 1973), Lubala et al. (2018), NBI (2020, 2021) and National Fisheries Resources Research Institute (NaFIRRI).

8 mil offshore.

Worthington (1932) and Poll and Damas (1935) did not provide

notes on distribution and abundance for species ofHaplochromis. How-

ever, the distribution of Haplochromis spp., in general, was like that of

the non-Haplochromis species that were not restricted to inshore or

offshore habitats (Worthington, 1932). Subsequent studies provided

more information on the habitat use, distribution and abundance of all

described fish species, reaffirming earlier observations and providing

more information on Haplochromis spp. For instance, in Lake George,

Gwahaba (1975) found that most of the species (15 species), includ-

ing most of the Haplochromis spp. known at the time, Astatoreochromis

alluaudi and Enteromius kerstenii, were more abundant within 100 m

from the shoreline and only found beyond that distance, occasionally.

Gwahaba (1994) showed that species found in all regions of the lake

moved freely between inshore and offshore regions, and in addition,

some Haplochromis spp. moved to deeper parts of the lake during the

day, reflected in lower catch rates near the surface. The preferred habi-

tats for Haplochromis spp. in Table 2 are based on areas where they

were found to be more abundant in Lake George or recorded in Lake

Edward.

Recent studies have showed that the distribution of the fish species

in the water bodies mirrors the pattern observed in the past. For

instance, NaFIRRI (2008) recorded six species of non-Haplochromis:

B. docmak, L. altianalis, C. gariepinus, O. leucostictus, O. niloticus and P.

aethiopicus. All these species were recorded in the inshore sites, but

only B. docmak and P. aethiopicus were also recorded in the offshore

sites. The study recorded 14 taxa ofHaplochromis spp. in offshore sites.

In each site, three fleets of gillnets of mesh sizes 1–8 in were set

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 13

F IGURE 3 Total annual catches of exploited fish species or species groups in lakes Edward, George and the Kazinga Channel. Two ormore
water bodies are combinedwhere available data was not segregated bywaterbody. Lake Edward is shared between Uganda (UG) and the
Democratic Republic of Congo (DRC). Source: Data obtained fromGame and Fisheries Department (1935, 1938, 1947, 1948, 1949, 1950, 1951,
1952, 1953, 1955, 1957, 1958, 1959, 1967, 1968, 1969, 1994), Okaranon and Kamanyi (1989), Fisheries Department (1971, 1972, 1973), Lubala
et al. (2018), NBI (2020, 2021) andNational Fisheries Resources Research Institute (NaFIRRI).

at varying distances from the shoreline. The study showed that, gen-

erally, species diversity in the lake decreased from the shoreline, in

conformity with earlier studies (Worthington, 1932).

The habitat use of fish species is best studied by tracking the move-

ment of tagged individuals. Mbalassa et al. (2015) attempted this

approach on C. gariepinus in Lake Edward. Observations indicated that

the species predominantly used littoral areas, river channels, and wet-

lands as general habitats and spawning areas, and to a less extent,

pelagic areas, consistent with earlier observations.

Attempts have been made using more quantitative fishing experi-

ments to estimate the abundance of the fish species in Lakes Edward

andGeorge. These experiments involve the capture of fishmainly using

gillnets in sites selected to represent diverse habitats in the waterbod-

ies (Ogutu-Ohwayo et al., 1997; NaFIRRI, 2008). Due to the selective

nature of the gillnets, multiple mesh-sizes are used to capture fish

in many size classes. The capture of fish is followed by systematic

identification andenumeration to acquire dataon the abundance.Mea-

sures of abundance from the experiments showadominance of cichlids

(Cichlidae). In Lake George, Ogutu-Ohwayo et al. (1997) showed that

the cichlids comprised 91.4% of the lake’s fish biomass, based on

the relative weight from experimental catches. Haplochromines (Hap-

lochromis spp.) formed 54.1% of the relative biomass followed by O.

leucostictus (30.7%) and O. niloticus (6.7%). Earlier, Gwahaba (1975)

determined the biomass of fish in Lake George, reporting total fish

biomass as 29 gm−2. Haplochromis nigripinnis comprised most of the

fish biomass (40%) in the lake, followed by O. niloticus (18%), H. angus-

tifrons (15%), P. aethiopicus (9%), C. gariepinus (7%), O. leucostictus (4%),

B. docmac (4%), H. squamipinnis (3%), Aplocheilichthys (now Lacustricola)

(2%), and H. pappenheimi (0.1%). These observations suggested that

other fish species belonging to Cyprinidae, Mormyridae, Anabantidae

and some cichlids such as Coptodon zillii were less abundant in the

lake.

In Lake Edward, NaFIRRI (2008) showed that haplochromines

comprised 95.9% of the fish community by number. The six non-

haplochromine species captured accounted for only 4.1% by num-

ber. By weight, the haplochromines accounted for 76.4% and non-

haplochromines 23.6%. Furthermore, the most dominant species,

among the haplochromines, wasH. nigripinnis, whereas the most domi-

nant non-haplochromine species was B. docmak. The dominance of the

haplochromines in Lake Edward was also demonstrated by a recent

experimental fishing expedition conducted in 2019 in which the hap-

lochromines comprised ∼90% of the fish community by number (LEAF,

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



14 MUSINGUZI ET AL.

2019). It is important to note that gillnets, the fishing gear on which

these observations are best, is highly selective for species and size,

although multiple mesh sizes are used to target all species and size

classes.

3.4 Life history characteristics of the fish species

3.4.1 Reproductive biology

Life history characteristics of fish species determine how vulnera-

ble or resilient the species and the ecosystem services they support

are to fishing pressure and environmental change (McKinney, 1997;

Pitcher et al., 2013). For data-poor fisheries, these characteristics may

be the only information available to define fish stock status and sup-

port decision-making. The characteristics are also direct or indirect

inputs into ecosystem models and comprehensive stock assessments.

We established that not many studies have examined the life history

and biological characteristics of the populations of the fish species

occurring in lakes Edward and George.

The reproductive biology of fishes is examined through aspects

such as sex ratio, size at maturity, fecundity and timing and location

of spawning. Sex ratios were available for six species of commercial

importance (Table 3). Sex ratios reported for species in Lake Edward

by NaFIRRI (2008) seemingly differed from the expected male:female

ratio of 1:1, probably because few specimens (4–26) were examined.

Estimates of size at first maturity (LM50) were available for only 4 of

the 8 fish species of commercial importance (Table 3).

Estimates of fecundity (absolute) in the lakeswere only retrieved for

O. leucostictus at a range of 230–718 eggs from 5 specimens (Ogutu-

Ohwayo et al., 1997). Using the distribution of the young, Gwahaba

(1973) provided information on the spawning behaviour and habitats

of selected species in Lake George and the Kazinga Channel. O. niloti-

cus spawns throughout the year, with peak spawning in the wet season

associated with a higher proportion of fish with active gonads and

young fish. B. docmak and L. altianalis predominantly use sandy bot-

toms for spawning. In Lake Edward, spawning locations forC. gariepinus

were determined as marginal wetlands, river channels, littoral zones,

and rivermouths (Mbalassa et al., 2015). However, generally, species in

the two lakes spawn in shallow nearshore habitats (Gwahaba, 1975).

3.4.2 Growth and mortality rates

Fish growth parameters include von Bertalanffy growth parameters

(Von Bertalanffy, 1938), age at a given size and longevity, whereas

mortality rates include total, natural, and fishing mortality rates. This

information was found to be limited for the populations of the fish

species in the water bodies. No estimates existed for mortality rates

of the exploited fish species in the water bodies and only O. niloticus in

Lake Edward had estimates of length and weight at infinity based on

observed maximum length (49 cm) and weight (2.0 kg), that is 51.6 cm

and 2.7 kg, respectively (Vakily, 1989). Substantial information on the

growth of the fishes was available only as Fulton’s condition factor (K),

and coefficients (b) of length–weight relationships (Table 4.

3.5 Trophic ecology

Table 2 synthesizes available information on the trophic ecology of

the fish species in Lakes Edward and George. Worthington (1932)

and Poll and Damas (1935) provided insights into the food of some

of the species in the two lakes. Labeobarbus altianalis was suggested

to be omnivorous, with fish remains, mollusks, chironomids, macro-

phytes and detritus present in its stomachs. Clarias gariepinus was

defined as a predatory species, feeding mainly on fish and to a lesser

extent, macroinvertebrates, macrophytes, algae, detritus and, proba-

bly, zooplankton. The species fed on a variety of fish species including

Oreochromis spp., Haplochromis spp., Barbus spp. (now Enteromius spp.

or L. altianalis) and L. pelagica. Protopterus aethiopicus was omnivorous.

In nine stomachs with food examined for the species, six were found

with fish remains, andeachof the three remainingoneswitheithermol-

lusks, chironomid larvae or plant materials. This data suggested that

fishwas themaindiet of the species.All stomachsofMormyridaeexam-

ined contained chironomid larvae. The food of L. pelagicawas reported

as zooplankton.

In Lake George and the Kazinga Channel, the diet ofO. niloticuswas

found tobepredominantly phytoplankton and zooplankton (Worthing-

ton, 1932). In the Kazinga channel, near Lake Edward, the composition

in the diet of the species shifted, with the zooplankton and phyto-

plankton taxamore abundant in Lake George, and the Kazinga channel

becoming replaced by those more abundant in Lake Edward. In the

open water of the lake, the species fed on macrophytes and Chao-

boridae, but to a lesser extent than zooplankton and phytoplankton.

Worthington (1932) remarked that the food of Haplochromis spp. was

diverse and observed the presence of molluscivores, planktivores and

piscivores, based on the morphology of mouth parts, jaws and teeth.

Dunn (1975) provided more information on the main food organ-

isms of the fish species in Lake George, including Haplochromis spp.

(Table 2). Trewavas (1983) suggested that the diet ofO. leucostictuswas

dominated by phytoplankton.

Until 2008, information on the diet of the species was qualitative

without the quantification of the relative importance of diet items.

Although NaFIRRI (2008) also listed the main food of C. gariepinus and

P. aethiopicus as insects and mollusks, respectively, the relative impor-

tance of food organisms of B. docmak and O. niloticus were quantified.

Odonata at 47.6% dominated the diet of B. docmak followed by fish

(mainly haplochromines) at 35.2%,with the rest comprisedChironomi-

dae, Chaoborus spp. and Ephemeroptera. The food items of O. niloticus

weredominatedbyalgae (76.7%), followedbyzooplankton (12.1%) and

detritus (11.3%).

Between 2015 and 2020,more studies on the trophic ecology of the

fish species were conducted under HIPE (human impacts on ecosys-

tem health and resources of Lake Edward), a project implemented to

study the aquatic ecosystems in the Lake Edward system (Borges et al.,

2021). The studies integrated stomach contents analysis with stable

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 15

TABLE 3 Reproductive biology of the commercial fish species in lakes Edward, Gorge, and the Kazinga channel.

Species Sex ratio (Male: Female)

LM50 (cm)

Total length

Sampling event

year Water body Reference

Oreochromis niloticus – 20 1997 Lake George Ogutu-Ohwayo et al. (1997)

10:4 (12) – 2007/2008 Lake Edward NaFIRRI (2008)

– 21 2011/2013 Bassa et al. (2015)

1:0.88 (942) 20.5 1972 Lake George Gwahaba (1973)

– 25.2 1957–1959 Lake George Fry and Kimsey (1960)

1:1.6 (751) – 1930/31 lakes Edward and George Worthington (1932)

– 20 (24) Lakes Edward, George and

the Kazinga Channel

Kamanyi (1996)

Bagrus docmak ∼1:2 (26) – 2007/2008 Lake Edward NaFIRRI (2008)

– 35–39 FL 1997 Lake George Ogutu-Ohwayo et al. (1997)

– 34.5 2011/2013 Bassa et al. (2015)

1:1.3 (150) – 1930–31 lakes Edward and George Worthington (1932)

– 35–39 FL (50–54) Lakes Edward, George and

the Kazinga Channel

Kamanyi (1996)

Clarias gariepinus 10:3 (8) – 2007/2008 Lake Edward NaFIRRI (2008)

1:0.7 (104) – 1930–31 lakes Edward and George Worthington (1932)

Oreochromis leucostictus 10:3 (9) – 2007/2008 Lake Edward NaFIRRI (2008)

– 15 1997 Lake George Ogutu-Ohwayo et al. (1997)

Protopterus aethiopicus 1:1 (4) – 2007/2008 Lake Edward NaFIRRI (2008)

– 56 2011–2013 Bassa et al. (2015)

– 55–59 1997 Lake George Ogutu-Ohwayo et al. (1997)

– 55–59 (75–79) Lakes Edward, George and

the Kazinga Channel

Kamanyi (1996)

1:1 (17) – 1930/31 lakes Edward and George Worthington (1932)

Labeobarbus altianalis 21.1 FL (male) 2015 Lake Edward Aruho et al. (2018)

35.4 FL (Female 2015 Aruho et al. (2018)

10:3 (23) – 2007/2008 NaFIRRI (2008)

1:2.5 (228) – – Worthington (1932)

Note: Values alongside sex ratios in parenthesis are number of fish examined. LM50 stands for size at first maturity, the length at which 50% of individuals in a

fish population aremature.

isotopes. Excluding synthetic materials, 21, 19, 13, 6, 16 and 14 items

were identified in the stomachs of C. gariepinus, P. aethiopicus, O. niloti-

cus, O. leucostictus, B. docmak, and L. altianalis, respectively, indicating

that the diet of the species was diverse (Cox, 2018; Kusters, 2019). In

broader terms, the diet items of these species were found to lie in six

major groups: detritus, phytoplankton, higher plantmaterial, zooplank-

ton, macroinvertebrates, and fish. Using prey specific abundance and

frequency of occurrence (Amundsen et al., 1996), the most frequent

and dominant of these diet items or broad groups were determined for

each of the species (Table 2).

Stable isotope analyses provided more information on the trophic

ecology of the fish species. The fish species community in Lake Edward

were found to have higher nitrogen isotope ratios and lower carbon

isotope ratios than in Lake George and the Kazinga Channel (Cox,

2018). Higher nitrogen ratios in fish are linked with a higher concen-

tration of DIN in waterbodies (Qu et al., 2021). However, this may not

explain the higher ratios in Lake Edward compared to Lake George and

theKazingaChannel because the former has the least concentration of

DIN (Stoyneva-Gärtner et al., 2020). Therefore, the differences in iso-

tope ratios could be attributed to the incorporation of higher trophic

level organisms into diet of the fish species in Lake Edward.

The variety of diet items for the different species or species groups

were combined to form a diet matrix useful in ecosystem models

based on Ecopath with Ecosim (EwE), the most common modelling

platform for aquatic ecosystems (Christensen et al., 2008). The matrix

(Online Resource 1; Table S1) comprises predators (columns in Online

Resource 1 Table 1) and their diet composition. The diet composition

shows the proportion each prey (rows in Online Resource 1; Table S1)

contributes to theoverall diet of a predator, relative to the contribution

of others (Christensen et al., 2008). The sum of the diet composi-

tion should be equivalent to one which was true for only five species

or species groups. For the others, the sum was indicated as NA for

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



16 MUSINGUZI ET AL.

TABLE 4 Estimates of some growth parameters for selected fish species in the lake Edward system.

Species

Length–weight

coefficients

Fulton’s condition

factor (K) Sampling year Waterbody Reference

Oreochromis niloticus a= 0.015; b= 3.09 – 1997 Lake George Ogutu-Ohwayo et al. (1997)

– 2.2 2008 Lake Edward NaFIRRI (2008)

– 2.0–2.3 1930–1931 lakes Edward and

George

Worthington (1932)

a= 0.023; b= 2.954 – 1989 Lake Edward, DRC Vakily (1989)

Oreochromis leucostictus a= 0.013, b= 3.13 – 1997 Lake George Ogutu-Ohwayo et al. (1997)

– 1.9 2008 Lake Edward NaFIRRI (2008)

Labeobarbus altianalis a= 0.0000021; b= 3.27 – 2013 Lake Edward and the

Kazinga channel

Ondhoro et al. (2017)

– 1.5 2008 Lake Edward NaFIRRI (2008)

– 1.0–1.1 1930–1931 lakes Edward and

George

Worthington (1932)

Clarias gariepinus – 0.8 2008 Lake Edward NaFIRRI (2008)

– 0.6–0.8 1930–1931 lakes Edward and

George

Worthington (1932)

Bagrus docmak – 1.2 2008 Lake Edward NaFIRRI (2008)

Protopterus aethiopicus – 0.4 2008 Lake Edward NaFIRRI (2008)

Note: Length–weight coefficients: a represents the intercept, whereas b represents the slope of the length–weight regression.

predators whose diet composition was only qualitative, and less than

one where some diet items could not be attributed to specific species

or species groups in thematrix. These two issues indicatedaknowledge

gap in trophic ecology in the waterbodies.

3.6 Fisheries

Lakes Edward,George, and theKazingaChannel support fisheries from

which riparian communities in Uganda and the DRC derive livelihoods.

By the 1930s, fisheries exploitation was occurring although fishing

effort and corresponding catches were small (Poll & Damas, 1935;

Worthington, 1932) (Figures 2–3). According to Worthington (1932),

Oreochromis spp. were the main targeted group, although L. altianalis,

C. gariepinus, B. docmak, P. aethiopicus,M. kannume, and L. forskaliiwere

also caught. The study ofWorthington (1932) also noted that by 1931,

fishing effortwas small and restricted to inshore areas anda few fishing

villages because of inadequate fishing craft to access the open waters,

poor road network and area closures. As a result, the catches at the

most active landing sites on lakes Edward and George in Uganda were

small, ∼100 fishes per day. Themost active site on Lake George had 16

fishermenoperating three fishing craft,whereas themost active site on

LakeEdwardwas inDRC (Kamande; Figure1)with several fishingboats

including onewith an outboard engine.

Since 1935, the initial targeted species or groups of species (B. doc-

mak, C. gariepinus, L. forskalii, L. altianalis, M. kannume, P. aethiopicus,

O. niloticus, and O. leucostictus) persisted in catches (Figure 2). Others

appeared in catches later (Haplochromis spp. from the 1980s andC. zillii

from2000) (Figure2). Species of commercial importance areB. docmak,

L. altianalis,P. aethiopicus,C. gariepinus,O. leucostictus,O. niloticus, andM.

kannume (Decru et al., 2020; Lubala et al., 2018; NaFIRRI, 2019; Petit,

2006). Generally, species extant in thewater bodies (Table 2) but not in

catches are less abundant.Haplochromis spp., which are themost domi-

nant in thewaterbodies bybiomass (Gwahaba, 1975), are not dominant

in catches because they are not a preferred target species. Described

as rare (Poll & Damas, 1935; Worthington, 1932), the presence and

persistence of M. kannume and L. forskalii in catches are noteworthy

(Figure 2).

Fisheries of Lakes George and Edward have developed steadily

since the 1930s, associated with an increase in the number of fishers

and fishing efficiency, utilizing gillnets and longlines (Dunn, 1972)

(Figure 3). In these water bodies, Oreochromis spp. initially comprised

most of the catches (Figure 4;Online Resource 2), contributing 38.4%–

82.2%. The development of the fisheries, however, was accompanied

by changes in species composition in the catches. In the Ugandan part

of Lake Edward, the contribution of Oreochromis spp. reduced from

87.7% in 1967 to 14.6% in 2019 (Figure 4;OnlineResource 2), whereas

in Lake George, their contribution decreased from 91.8% in 1950 to

11.0% in2019 (Figure4;OnlineResource2), and from79.1% in1969 to

6.85% in 2019 in the Kazinga Channel (Figure 4; Online Resource 2). In

the DRC (Lake Edward), the contribution of Oreochromis spp. reduced

from 78% in 1970 to 26% in 2016 (Figure 4; Online Resource 2). The

decline in the contribution of Oreochromis spp. suggests that species

in this group, particularly O. niloticus which is of higher abundance

and importance in the fisheries than O. leucostictus, became heavily

exploited around the 1970s. In Uganda, this decline in Oreochromis

spp. coincided with a decline in total catches (Figure 3). In response,

fishers redistributed fishing effort to B. docmak, P. aethiopicus,

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 17

F IGURE 4 Catch composition (%) by water body or a combination of water bodies. Two ormore waterbodies were combined for some periods
when data was not segregated bywaterbody. Lake Edward is shared between Uganda and the Democratic Republic of the Congo (DRC). Online
Resource 2 provides data onwhich this figure was based. Source: Data adopted fromGame and Fisheries Department (1935, 1938, 1947, 1948,
1949, 1950, 1951, 1952, 1953, 1955, 1957, 1958, 1959, 1967, 1968, 1969, 1994), Okaranon and Kamanyi (1989), Fisheries Department (1971,
1972, 1973), Lubala et al. (2018), NBI (2020, 2021) andNational Fisheries Resources Research Institute (NaFIRRI).

and C. gariepinus. This shift in target fisheries was followed by an

increase in catches of these species in lakes Edward and George

(Figure 2), especially for C. gariepinus (Online Resource 1; Figure S2)

and P. aethiopicus (Online Resource 1; Figure S12).

Where substantial long-term data was available, total catches for

fish species or water bodies illustrated a general increase, followed by

a decrease to a relatively stable level (Figures 2 and 3;Online Resource

1; Figures S1–S12). The trend in catches of Lake Edward in the DRC is

an exception that is discussed in the following.

The CAS conducted in 2019 and 2020 provided catch estimates for

the part of Lake Edward in the DRC that were enormously inconsis-

tent with previous catches in the country, and the known estimates

of maximum sustainable yield (MSY) (NBI, 2020, 2021). The estimates

of the catches, as depicted in Figures 2 and 3, were much higher

than expected. Total catches were estimated at 29,347 t for 2019

and 39,411 t for 2020. Dominated by B. docmak and Oreochromis

spp. (Figure 4; Online Resource 2), these estimates showed that the

actual catches in the DRC could be more than twice the previous

catches (Figures 2 and 3) and known estimates of MSY. Estimates of

MSY are a range of 14,000–16,000 t per year for the whole lake,

11,000–12,000 t for the DRC and 3000–4000 t for Uganda (Vakily,

1989).

Despite the inconsistency, the estimates of catches for 2019 and

2020 (NBI, 2020, 2021) could be believed, given that past authors

acknowledged gross underestimation of catches (Petit, 2006; Vakily,

1989) and these estimates were derived from the first comprehen-

sive CAS in the DRC. Several issues hinder the proper monitoring of

the fisheries resources of Lake Edward in the DRC. Fishing activities

in the part of the lake are managed by multiple stakeholders (Lubala

et al., 2018; Vakily, 1989). Due to lack of coordination, these stake-

holders, if they do, only report catches from fish landing sites within

their jurisdictions and selected fishing gears (Petit, 2006). A substan-

tial part of the fishery is controlled by rebel groups and for safety

reasons, this part is rarely, if at all, monitored (Petit, 2006). Research

studies also restrict data collection to safe landing sites and have sub-

stantial methodological limitations. For instance, Lubala et al. (2018)

sampled only four safe landing sites to determine annual catches, and

the approach used to derive lake-wide estimates from the observa-

tions made at the sampled sites was not indicated. NBI (2020, 2021),

on the other hand, engagedmultiple stakeholders in the DRC, sampled

many (10) fish landing sites and used standardized approaches for CAS

(LVFO, 2005). These estimateswere corroboratedwith estimates from

an independent monitoring exercise by the Congolese Institute for

Nature Conservation-Virunga National Park (ICCN-PNVI) that began

in 2019 to record daily catches in four landing sites. Using data from

this monitoring exercise, Omombo et al. (unpublished) estimated total

catches in theDRCas28,427 t for 2019, 29,338 t for 2020and22,868 t

for 2021.

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



18 MUSINGUZI ET AL.

The inconsistency of the estimates of 2019 and 2020 with the

known estimates of MSY could be explained by the origin of the MSY

estimate. Vakily (1989) derived the MSY from a model that relates

the morphoedaphic index (MEI) of a water body with its fish yield

(Schlesinger&Regier, 1982). In themodel, theMEI, derived froma ratio

of total dissolved solids (TDS) and mean depth, is directly proportional

to MSY. Environmental changes such as eutrophication that increase

TDS and decrease mean depth should increaseMEI, and consequently,

MSY (Ryder, 1965).Given that LakeEdwardhasbecomemorenutrient-

enriched (Stoyneva-Gärtner et al., 2020), its currentMSY based on the

model should be higher. However, this is not the case. Recalculating the

MSY using recent estimates of TDS (derived from mean conductivity

from Stoyneva-Gärtner et al. (2020) using a function by Rusydi (2018)

that correlates TDS and conductivity) andmean depth (Hamilton et al.,

2022) generated a value (16,550 t), close to that of Vakily (1989). This

finding suggests that TDS may not be a good predictor of fish yield in

the lake. Therefore, the MSY value by Vakily (1989), compared to the

catches of 2019 and 2020, was probably an underestimate. In addition,

research efforts before 2019 as suggested by Vakily (1989) and Petit

(2006) underestimated catches in the DRC. For this reason, the pat-

tern in catches differs from that in the Ugandan part of Lake Edward

and Lake George (Figure 3). Actual catches in the past could have been

more than the values determined for 2019 and 2020. These catch lev-

els account for the high fishing effort which has increased remarkably

on the lake as indicated by the number of fishers and fishing boats

(Figure 5).

Lakes Edward and George have had a long history of fisheries man-

agement. In Uganda, the leading role of fisheries management has

always been a responsibility of a designated government agency. In

the DRC, management responsibilities are mainly shared among dif-

ferent stakeholders with different spatial or functional jurisdictions

(Lubala et al., 2018; Petit, 2006; Vakily, 1989). Some fishing areas are

controlled by rebel groups that encourage illegal fishing, discourage

monitoring and contest management measures by the government

(Marijnen, 2022; Petit, 2006). The earliest report we retrieved for

the agency responsible for fisheries management in Uganda was from

1935 (Game and Fisheries Department, 1935), the time the fisheries

on thewater bodies were developing. The department practiced active

management, taking precautions to prevent overfishing. Management

practices included recording catches, licensing, enforcement of fishing

regulations and setting minimum mesh size. These practices showed

that management aimed at controlling fishing effort directly through

different measures, an approach that is still followed on these water

bodies today. Over time, thesemanagementmeasures became ineffec-

tive, especially in the DRC, as reflected in the proliferation of fishing

effort (Figure 5), the use of illegal fishing gear and the capture of imma-

ture fish (Bassa et al., 2014; Dunn, 1975; Lubala et al., 2018; Petit,

2006). These unsustainable fishing practices are depicted in declining

catches and reduced average weight of individuals in the catches for

most species (Figure 6).

In response to unsustainable fishing practices, Uganda intensified

the enforcement of fisheries regulations since 2018 to end illegal

fishing practices and methods (NPA, 2019). Since then, illegal activi-

ties have significantly reduced, and catches have improved on lakes

Edward and George (NBI, 2020). Fishers from the DRC are increas-

ingly crossing intoUganda to exploit the opportunities of the improved

enforcement on Lake Edward, resulting in fatal crashes and fre-

quent arrests byUganda’smilitary, which coordinates the enforcement

(Kyalwahi, 2021;Marijnen, 2022).

4 DISCUSSION

Reliable data is an important requirement for sustaining inland fish-

eries and the ecosystem services they support (Cooke et al., 2016). The

first of the 10 steps of the Rome Declaration for responsible inland

fisheries calls for the enhanced acquisition of accurate and complete

data on inland fisheries, including at local scales (FAO andMSU, 2016).

The availability of data is envisaged to spur global assessments of the

inland fisheries, akin to those of marine fisheries and stimulate inclu-

sion into global governance processes. Required actions at local scales

according to the RomeDeclaration include data collection, monitoring,

and assessment of fisheries. In all these actions, standard method-

ologies are recommended and all forms of inland fisheries, including

subsistence, recreational and illegal and unregulated fisheries should

be covered.

This review showed that Lakes Edward and George have been sub-

jected to research surveys, providing data and information on aspects

of water quality, extant taxa, the abundance of biotic communities, life

history of fish species, trophic ecology, fishing effort and fish catches.

However, substantial gaps exist. First, all the aspects examined lacked

adequate time series data. Unlike aspects such as water quality, extant

fish species, fishing effort and fish catches which had recent data,

data available on abundance of all biotic communities and life history

parameters (Tables 3 and 4) require updating. In addition, the commer-

cial fish species in the waterbodies lacked values on fish mortalities

and von Bertalanffy growth parameters which are important life his-

tory characteristics (King & McFarlane, 2003). Information available

on trophic ecology was mainly qualitative (Table 2; Online Resource 1;

Table S1), indicating the need for quantifying diet composition of all the

fish species or species groups in the water bodies. In fisheries, apart

from the uncertainty in the catches of DRC, there is absence of fish-

eries management reference points, benchmarks used to measure the

status of exploited fisheries (ICES, 2017).

Lack of adequate time series and presence of aspects that require

updating or consideration in research can be explained by lack of

establishedmonitoring programmes for regular data collection, a char-

acteristic of most inland fisheries (Cooke et al., 2016; Plisnier et al.,

2022). Routine monitoring programmes with a wide scope covering

all aspects of aquatic ecosystem health and fisheries have been rec-

ommended on water bodies in the region, including lakes Edward and

George, to sustain data collection efforts and avoid the data gaps

(Plisnier et al., 2022). In these water bodies, monitoring programmes

could be designed to fill all the established data gaps above. Studies

of substantial magnitude on these water bodies such as Stoyneva-

Gärtner et al. (2020) and NaFIRRI (2008) considered spatial variations

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 19

F IGURE 5 Number of fishers and boats on lakes Edward, George and the Kazinga Channel. Source: (Fishers) Data adopted from: Lubala et al.
(2018) and National Fisheries Resources Research Institute); (Boats) Data adopted from: Vakily (1989), Petit (2006), NaFIRRI (2015), Lubala et al.
(2018) and NBI (2019).

in parameters studied. This should be maintained in the routine moni-

toring programmes. In Lake Edward, these studies were biased to the

Ugandan part of Uganda (Online Resource 1; Figure S13). The absence

of the part of the lake in the DRC in these studies is conspicuous

probably due to unrest in the region that makes sampling difficult

(Marijnen, 2022). This issue seems to have been the reason why for

instance, Stoyneva-Gärtner et al. (2020) sampled only one site in DRC

compared to 16 sites in Uganda. Where possible, the monitoring pro-

grammes should consider this part of the lake to ensure a complete

understanding of the spatial variations.

Routine data collection is required most on the aspects of fish-

eries using fishery-dependent surveys to obtain data on catch and

fishing effort preferably on an annual basis. In DRC, these could help

ascertain the catches observed in 2019 and 2020. Data collection

using fishery-dependent surveys could be improved by using person-

nel placed at fish landing sites, observers and fishers’ records (FAO,

1999). High costs of observation programmes and low literacy levels

among fishers in the area make personnel stationed at landing sites

the most viable option. In the past, such personnel recorded catches

on the water bodies in Uganda (Game and Fisheries Department,

1935), whereas, in the DRC, the personnel exist at some landing sites,

although they do not measure catches directly but record fishers’ dec-

larations which is problematic (Petit, 2006). Actualizing this approach

could be preceded by selecting representative sites whose estimates

could be extrapolated to generate lake-wide estimates. This is because

landing sites on the waterbodies are many and dispersed (Figure 1),

making it difficult to place personnel at each site to collect data. To

improve the completeness of data, data collection efforts should incor-

porate measurements of individual fish size (length and weight) and

always aim at disintegrating the catches data by species. Catches by

illegal segmentsof the fisheries arenotproperly recorded (Dunn, 1972;

Ogutu-Ohwayo, 1997; Petit, 2006). Modalities should be devised to

reflect these catches in records.

Fishery-independent surveys are required to supplement the data

from fishery-dependent surveys and cover aspects of the abundance

of biotic communities, the life history of fish, fish trophic ecology

and water quality. Data on abundance, preferably absolute biomass, is

required for biotic communities including fish, invertebrates, macro-

phytes, and phytoplankton. Fish biomass is best derived using swept

area (trawling) and hydroacoustic methods (Silliman and Gutsell,

1958). In the region, these methods are not common and may occur

only in larger lakes such as Lake Victoria (Hydro-acoustics Regional

Working Group, 2019). Using these methods in lakes Edward and

George needs the acquisition of infrastructure in terms of, for exam-

ple research vessels which are costly. For this reason, we are certain

that it will take time to apply these methods in these waterbodies.

Fishery-independent surveys should occur annually with standardized

methods to provide adequate time series data to assess ecosystem

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



20 MUSINGUZI ET AL.

F IGURE 6 Changes in average weight of individuals in catches. Values are aggregations for lakes Edward, George and the Kazinga Channel.
Source: Data adopted fromReports by the Game and Fisheries Department (1935, 1938, 1947, 1948, 1949, 1950, 1951, 1952, 1953, 1955, 1957,
1958, 1959, 1967, 1968, 1969, 1994), Fisheries Department (1971, 1972, 1973) and Ssentongo (1992).

and fisheries status (Froese et al., 2020). In these fishery-independent

surveys, efforts should be made to design the methods in a way that

facilitates the comparisonof resultswith those fromstudies in the past.

For water quality, methods and sensitivity of instruments for some

parameters may have changed in response to technological advances

(Zainurin et al., 2022). As result, temporal comparisons should bemade

cautiously.

Life history parameters of fish are key for monitoring and manage-

ment of exploited fish species yet in thesewaterbodies, the parameters

are eithermissing or require updating. Data from the proposed fishery-

dependent and fishery-independent surveys should be integrated to

obtain more reliable estimates of these parameters. In the short term,

gillnet surveys (fishery-independent) could be integrated with fishery-

dependent surveys to update existing estimates of growth, size at

maturity and coefficients of length–weight relationships (Tables 3

and 4). Only data-poor approaches could be applicable for determin-

ing von Bertalanffy growth parameters and mortalities using this data.

The most used of these methods is the electronic length frequency

analysis (ELEFAN) which generates growth parameters from length–

frequency data (Pauly & David, 1981). Data needs for this method

are monthly data collections, preferably for 1 year, covering all size

spectrums and populations of species of interest. Sampling commercial

catches to supplement fishery-dependent surveys could contribute to

data completeness for this approach.

The exploited fish species in the waterbodies also lacked fisheries

management reference points meaning that stock assessment in the

past neglected a critical stage where the fishery-dependent and -

independent data is used to define the status of fish stocks using

fisheries management reference points (Caddy & Mahon, 1995). The

reference points are used to establish rules for harvest control and

to evaluate management measures. We found this stage lacking, with

the stock assessment process ending at the estimation of catches, life

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense



MUSINGUZI ET AL. 21

history parameters and fishing effort, defined as either the number

of boats or fishers. The absence of this stage suggests that manage-

ment occurs with inadequate guidance. In addition, uncertainty exists

on the actual status of exploited fish species and the limits or targets

adequate to sustain and rebuild the fisheries. In response, avail-

able data could be subjected to stock assessments using appropriate

methods for data-poor fisheries to generate the fisheries manage-

ment reference points. A variety of data-poor methods that for

example use only length–frequency data, and catches exist (Froese

et al., 2017, 2018, 2020; Newman et al., 2015). In future, the ref-

erence points should be generated annually, simultaneous with the

fishery-dependent and -independent surveys, which we have also

recommended to occur annually.

With intensifying fishing pressure and emerging issues such as oil

exploitation (Verheyen et al., 2016), Lakes Edward and George could

benefit from Ecosystem-based Fishery Management (EBFM) which is

promoted as the best option for fisheries management (Pikitch et al.,

2004). Positive outcomes have been reported for inland fisheries, for

example in Indonesia, Brazil and Laos, where elements of EBFM have

been incorporated in management (Butorac et al., 2020; Ditya et al.,

2022; Koning et al., 2020). Supporting EBFM requires ecosystemmod-

elling to describe andquantify interactions among ecosystemelements

(functional groups), assess the impacts of environmental change and

fishing, and evaluate management options (Essington & Punt, 2011;

Plagányi, 2007). Acknowledging the significance of climate change to

fisheries (Allison et al., 2009), the EBFM is also the bestmeans to adapt

the fisheries to impacts of climate change (Holsman et al., 2020). The

development of operational ecosystem models of lakes Edward and

George to support decision-making should be considered in research.

Modelling efforts could focus on using EwE, the most common

framework for modelling aquatic ecosystems (Colléter et al., 2015).

Many EwEmodels exist formost of the AfricanGreat Lakes (Musinguzi

et al., 2017), but Lake Edward has never been considered for ecosys-

tem modelling using EwE, whereas one model of the 1970s exists

for Lake George (Moreau et al., 1993). Despite the gaps in data, this

review showed that minimum information required to define func-

tional groups for the models is available. The data available on water

quality and primary production (Table 1), fish life history and biologi-

cal characteristics (Section 3.2), biomass of invertebrates (Section 3.1),

trophic ecology (Table 2; Table S1) and catches (Section 3.5) could be

sufficient to derive model parameters or as direct inputs into the mod-

els. However, updating the data on these aspects through the data

collection efforts recommended above could improve the outputs of

the models. The available data could be supplemented with data from

nearbywater bodies, especially Lake Victoria, which is a common prac-

tice in ecosystem modelling globally (Christensen et al., 2008). Lake

Victoria is appropriate in this case because it is themost studied lake in

the region and sharesmany fish specieswith lakes Edward andGeorge.

AUTHOR CONTRIBUTIONS

Laban Musinguzi: Conceptualization; data curation; formal anal-

ysis; methodology; project administration; visualization; writing—

original draft; writing—review and editing. Nathan Vranken: Data

curation; writing—review and editing. Vianny Natugonza: Visualiza-

tion; writing—review and editing. William Okello: Conceptualization;

funding acquisition; project administration; supervision; writing—

reviewand editing.Maarten van Steenberge: Conceptualization; fund-

ing acquisition; methodology; supervision; writing—review and edit-

ing. Jos Snoeks: Conceptualization; funding acquisition; methodology;

project administration; supervision; writing—review and editing.

ACKNOWLEDGEMENTS

The PhD research of LM is part of the FishBase for Africa programme

funded through a framework agreement between the Royal Museum

for Central Africa (RMCA) and the Belgian Development Cooperation

and Humanitarian Aid (DGD). This study builds upon the HIPE project

(human impacts on ecosystem health and resources of Lake Edward;

BR/154/A1/HIPE) and was executed partly within the KEAFish project

(the biodiversity, biogeography and evolutionary history of the north-

ernbasins of theGreatAfrican Lakes: the enigmatic fish faunasof Lakes

Kivu, Edward and Albert revisited; B2/202/P1/KeaFish). Both HIPE

and KEAfish projects were funded by the Belgian Science Policy Office

(BELSPO)

CONFLICT OF INTEREST STATEMENT

No, I declare that the authors have no conflicts of interest.

FUNDING INFORMATION

The work was part of the FishBase for Africa programme funded

through a framework agreement between the Royal Museum for Cen-

tral Africa (RMCA) and the Belgian Development Cooperation and

Humanitarian Aid (DGD). This study builds upon the HIPE project

(human impacts on ecosystem health and resources of Lake Edward;

BR/154/A1/HIPE) and was executed partly within the KEAFish project

(the biodiversity, biogeography and evolutionary history of the north-

ernbasins of theGreatAfrican Lakes: the enigmatic fish faunasof Lakes

Kivu, Edward and Albert revisited; B2/202/P1/KeaFish). Both HIPE

and KEAfish projects were funded by the Belgian Science Policy Office

(BELSPO)

ETHICS STATEMENT

The research conducted in thiswork did not involve direct contactwith

animals whichmay require approval.

DATA AVAILABILITY STATEMENT

Data used in this article is provided in tables, Online Resource 2, and

on figshare at https://doi.org/10.6084/m9.figshare.22013312.v3 and

https://doi.org/10.6084/m9.figshare.24480898.v1.

ORCID

LabanMusinguzi https://orcid.org/0000-0001-7915-3218

PEER REVIEW

The peer review history for this article is available at https://publons.

com/publon/10.1002/aff2.140.

 26938847, 2024, 1, D
ow

nloaded from
 https://onlinelibrary.w

iley.com
/doi/10.1002/aff2.140 by M

akerere U
niversity, W

iley O
nline L

ibrary on [07/02/2024]. See the T
erm

s and C
onditions (https://onlinelibrary.w

iley.com
/term

s-and-conditions) on W
iley O

nline L
ibrary for rules of use; O

A
 articles are governed by the applicable C

reative C
om

m
ons L

icense

https://doi.org/10.6084/m9.figshare.22013312.v3
https://doi.org/10.6084/m9.figshare.24480898.v1
https://orcid.org/0000-0001-7915-3218
https://orcid.org/0000-0001-7915-3218
https://publons.com/publon/10.1002/aff2.140
https://publons.com/publon/10.1002/aff2.140


22 MUSINGUZI ET AL.

REFERENCES

Allison, E.H., Perry, A.L., Badjeck, M.C., Adger, N.W., Brown, K., Conway, D.

et al. (2009)Vulnerability of national economies to the impacts of climate

change on fisheries. Fish and Fisheries, 10(2), 173–196
Amundsen, P.A., Gabler, H.M. & Staldvik, F.J. (1996) A new approach to

graphical analysis of feeding strategy from stomach contents data—

modification of the Costello (1900) method. Journal of Fish Biology, 48,
607–614

Aruho, C., Ddungu, R., Nkalubo, W., Ondhoro, C.C., Bugenyi, F. & Rutaisire,

J. (2018) Optimizing selection of sexually mature Barbus Altianalis for
induced spawning: determination of size at sexual maturity of popula-

tions from Lake Edward and Upper Victoria Nile in Uganda. Fisheries and
Aquatic Sciences, 21(1), 1–13.

AUC (2015)Agenda 2063: the Africa wewant. Addis Ababa, Ethiopia: African
Union. https://doi.org/10.18356/8cdc8224-en

Bagalwa, M., Yalire, M., Balole, E. & Karume, K. (2014) A preliminary assess-

ment of physico-chemical and bacteriological characteristics of Lake

Edward and majors tributaries rivers, democratic Republic of Congo.

Scholars Academic Journal of Biosciences, 2(3), 236–245
Bassa, S., Mbabazi, D., Taabu-Munyaho, A., Nakiyende, H., Muhumuza, E.,

Nsega, M. et al. (2014) Can protected areas work in artisanal fisheries

of Uganda? The case of Lakes Edward-George and Kazinga Channel.

Uganda Journal of Agricultural Sciences, 15(1), 27–39
Bassa, S., Taabu-Munyaho, A., Mbabazi, D., Nakiyende, H., Muhumuza, E.,

Amiina, R. et al. (2015) Technical report of frame and catch assess-

ment surveys of lakes Edward and George; and the Kazinga Channel

conducted in January and July 2011–2013.In: Technical document. Jinja,
Uganda: National Fisheries Resources Research Institute.

Beadle, L.C. (1981) The inland waters of tropical Africa: an introduction to
tropical limnology. London: LongmanGroup Limited.

Binding, C.E., Jerome, J.H., Bukata, R.P. &Booty,W.G. (2007) Trends inwater

clarity of the lower Great Lakes from remotely sensed aquatic color.

Journal of Great Lakes Research, 33(4), 828–841
Borges, A., Lambert, T., Descy, J.P., Darchambeau, F., Deirmendjian, L.,

Roland, F. et al. (2021) Human impacts on ecosystem health and resources
of Lake Edward. Belgium: Belgian Science Policy Office.

Breuil, C. & Grima, D. (2014) Baseline report DR Congo. SmartFish programme
of the Indian Ocean Commission, Fisheries Management FAO component.
Ebene, Mauritius: FAO.

Bugenyi, F.W.B. (1979) Copper ion distribution in the surfacewaters of Lake

George and Idi Amin.Hydrobiologia, 64(1), 9–15
Bugenyi, F.W.B. (1982) Copper pollution studies in Lakes George and

Edward, Uganda: the distribution of Cu, Cd and Fe in the water and

sediments. Environmental Pollution Series B, Chemical and Physical, 3(2),
129–138.

Burgis, M.J., Darlington, J.P., Dunn, I.G., Ganf, G.G., Gwahaba, J.J. &

McGowan, L.M. (1973) The biomass and distribution of organisms

in Lake George, Uganda. Proceedings of the Royal Society, 184(1076),
271–298

Butorac, D., Santos, P., Phouvin, P. & Guegan, F. (2020) Freshwater fisheries

conservation can increase biodiversity. PLoS ONE, 15(5), e0233775
Caddy, J