Content uploaded by Zacharia Malley
Author content
All content in this area was uploaded by Zacharia Malley on Aug 16, 2017
Content may be subject to copyright.
Environmental change and vulnerability
in the Usangu plain, southwestern
Tanzania: Implications for sustainable
development
Zacharia J. U. Malley1, Mohammed Taeb2, Tetsuo Matsumoto3and Hiroyuki Takeya3
1Ministry of Agriculture and Food Security, Agricultural Research Institute-Uyole, Tanzania
2United Nations University/Institute of Advance Studies, International Organizations Center,
Yokohama, Japan
3Nagoya University, International Cooperation Center for Agricultural Education, Nagoya,
Japan
Key words: Environmental insecurity, human activities, hydropower generation; livelihood insecurity;
wildlife biodiversity
SUMMARY
Natural resources are important assets for sustainability of rural livelihoods and economic
development. The objectives of the study were to assess linkage of environmental change
to emerging water scarcity, livelihoods insecurity and decline in wildlife biodiversity and
hydropower generation in the Great Ruaha ecosystem, and to discuss policy lessons for
sustainable development. Literature survey, participatory assessments, collection of
records from stakeholders, participatory observations and a questionnaire survey were
used. Participatory assessment found that increasing water scarcity is associated with
degradation of land, vegetation cover and change in rainfall intensity and duration. Rain-
fall change is supported by 83% of respondents and rainfall trends analysis. About 42% of
variation in cereal production is described by the rainfall amount variability. Analysis of
trends for per capita cereals production from district records showed a significant
(p< 0.05) decline over the years. Changes in water flows in the Great Ruaha River have
been evident since 1993, and have affected wildlife diversity in the Ruaha National Park
and power generation from the Kidatu hydroelectric plant downstream. Human liveli-
hood activities, exclusiveness, weak coordination and collaboration between sectors and
institutions in policy formulation and implementation have played a role. The pro-poor
policy formulation through integrated planning, strong coordination and collaboration
are recommended for sustainable development.
International Journal of Sustainable Development & World Ecology 14 (2007) 1–15
Correspondence: Zacharia JU Malley, Agricultural Research Institute-Uyole, PO Box 400, Mbeya, Tanzania. Email:
zjmalley@yahoo.co.uk
1
INTRODUCTION
Environmental insecurity and subsequent climatic
changes, including increased temperatures, de-
creased rainfall and rainfall variability, are now
occurring in different parts of the world, but
people living in dry and semi-arid areas in Africa
are particularly vulnerable. The Intergovern-
mental Panel on Climate Change (IPCC) warned
that average temperatures are increasing and that
associated higher rates of evaporation cause
drier conditions in the interior areas of the
major continents (IPCC 2001). Karl et al. (1995)
observed that total rainfall may increase in some
regions, but variability is likely to increase, such that
droughts will become more frequent and intense,
while rainfall will be concentrated in short and
severe storms.
In Africa, land and water degradation are critical
environmental changes caused by human activities.
Stocking (2002) considered land degradation as
the reduction or loss of biological and economic
productivity of major land uses (rainfed cropland,
irrigated cropland, rangeland and forests) result-
ing from processes arising from human activity,
such as soil erosion, changes in physical, chemical
or biological soil properties, and long-term loss of
natural vegetation. He further noted that, globally,
land degradation has a number of complex direct
and indirect interactive effects on climatic change
through emissions of greenhouse gases, reduction
of forests that acts as terrestrial carbon sinks, dimin-
ished effectiveness of carbon cycling, reduction in
biodiversity, and pollution and sedimentation of
international waters. Vegetation cover is a critical
service, providing replenishment of underground
water, sustenance of water levels in rivers, and
habitat for biodiversity. Impacts of the degradation
of vegetation cover are drying up of rivers and lakes,
loss of biodiversity, increased floods, accelerated
soil erosion, declining land productivity, intensifi-
cation of desertification, and declining rainfall and
crop production. These degradation processes lead
to vulnerability of the livelihoods of the people.
Johnson (1999) reported that about one billion
people live in poverty worldwide as a result of land
degradation, and the number is growing in Africa.
People become vulnerable through reduced access
to clean and safe drinking water that poses threats
to health, increased drought that results into food
insecurity, and low incomes.
In East Africa, media reports show that there
have been notable declines in access to clean and
safe drinking water and water for other production
and socio-economic services (Mbaria 2005; Mpinga
1999; Simbeye 2004; Wangari 2005). Increasing
drought frequencies and drying up of flows in
rivers, streams and springs, threats to food security
and poverty have been reported in East Africa
(Mbaria 2005; Mpinga 1999; Simbeye 2004;
Wangari 2005). These problems are attributed to
increasing degradation of land and forest resources
in the watersheds. For example, Wangari (2005)
indicated reduced flows in the tributaries of the
Tana River have been due to continuous abuse of
the environment in Mount Kenya and the Aberdare
Mountains on the eastern edge of the rift valley.
Mbaria (2005) reported that the Ewaso Nyiro River
has dwindled in volume as a consequence of des-
truction of its catchments by Kipsigis community
activities, thereby affecting livelihoods of the Maasai
community downstream. This has heightened con-
flicts between the Kipsigis and Maasai communities.
In Tanzania, drying up of the Great Ruaha River
since 1993 has been of great concern to local peo-
ple who use it for water to sustain their livelihood
activities and whose water is used nationally for
electricity generation to the national grid and for
biodiversity conservation in Ruaha National Park
downstream. The link between vulnerability con-
cerns and environmental change is not adequately
understood. The objectives of this work were, there-
fore, to assess environmental change and its link to
vulnerability of rural livelihoods in the Usangu
plain; examine availability of water for hydropower
generation and wildlife biodiversity of the Great
Ruaha River ecosystem downstream; and to discuss
policy lessons for sustainable development.
METHODS
Study area
The field study was conducted in the Usangu plain,
Mbarali District, southwestern Tanzania from April
2004 to March 2005. The Usangu plain is a
bowl-shaped area with flat to undulating topo-
graphy, surrounded by rolling watershed hills. The
plain is estimated to cover about 16,500 km2, about
the same size as Mbarali District. The plain and
surrounding uplands form part of the catchment
of the Rufiji River basin, which include the
Environment change in Usangu plain Malley et al.
2 International Journal of Sustainable Development & World Ecology
southwestern Kipengere and Livingstone Moun-
tain ranges. One of the largest tributaries of the
Rufiji River is the Great Ruaha River, which receives
water from a number of small streams originating in
the Kipengere and Livingstone Mountains. The
tributaries which flow into the Usangu plain form
the Great Ruaha River, which then flows across the
plain to feed the Mtera and Kidatu dams down-
stream for electricity generation, and finally enters
the Rufiji River which drains into the Indian Ocean.
The Great Ruaha River contributes about 22% of
the total Rufiji River flows (WWF 2002). The
Usangu plain is made up of wetlands and upper
drier fan ecosystems. The uses and significance of
these ecosystems for local people and national
economic development are detailed in Malley et al.
(2005).
Data collection and derivation
Literature survey, participatory assessments and
household survey
Research methods employed were: survey of
published and unpublished literature, discussions
with key stakeholders, collection of quantitative
records from stakeholders, village participatory
assessments, participatory observations and a field
questionnaire survey of 266 households in six -
villages, Mhwela, Matebete, Mabadaga, Sololwambo,
Ikoga and Yala, in the Usangu plain. Major sources
of literature were: previous project reports pro-
vided by stakeholders, newspapers, internet
documents and journal articles. Participatory
assessments employed identification of locally per-
ceived environmental change and vulnerability
attributes and participatory trends analysis by scor-
ing the identified attributes over 10-year periods for
30 years. Closed and open-ended questionnaires
were prepared, pre-tested, improved for clarity,
translated into a common language (Swahili) and
finally administered to the selected households.
Record collection from stakeholders
Available records on crop production and livestock
number, wildlife populations, water level in Mtera
reservoir and monthly and annual rainfall were
collected from the Mbarali District Council, Ruaha
National Parks, the Tanzania Electric Supply
Company (TANESCO) and the Directorate of the
Meteorological Service. Crop production informa-
tion for four cereals grown by smallholder farmers
in the district, maize, rice, sorghum and finger mil-
let, were used. Meteorological information included
total monthly and annual rainfall from two stations,
Lugerere at 1100 m asl was typical of the Usangu
plain and the Agricultural Research Institute-Uyole
at 1850 m asl, for a typical upland watershed. Live-
stock number records were collected from the
Mbarali district census and the Sustainable Manage-
ment of Usangu Wetlands and its Catchments
(SMUWC) project (2002) aerial and ward surveys
reports. The livestock number records that were
consistent were those of 1998 to 2002 and were used
for livestock biomass trends analysis in Usangu plain.
Monthly records of Mtera reservoir water levels were
collected from TANESCO. Records on wildlife
counts made in 1993 and repeated in 1996 by the
Tanzania Wildlife Conservation Monitoring Unit
(TWCM) were collected from the Ruaha National
Park (RNP) reports (TWCM 2000).
Derivation of data from collected records
In order to assess changes in cereal crop yields,
recorded production between 1990 and 2003 from
Mbarali District Council’s agriculture and livestock
development department was used. Each cereal
crop production record for a given year was divided
by the cultivated area of the crop in the same year to
determine yield in a given year. The individual crop
yield for each year was then multiplied by the aver-
age area cultivated for that crop over 14 years to
obtain production figures that are free of area
expansion influence. The total cereal production
was then determined as the sum of maize, paddy,
sorghum and finger millet production, free from
area changes over the years. The per capita produc-
tion figures for cereals were derived by dividing
total cereal production by projected district popu-
lation from the 1988 census. Average population
annual growth rate of 2.7% was used, which is the
figure reported in the 1988 and 2002 population
census results for Mbarali District. Livestock bio-
mass data were derived from the average numbers
of cattle, donkeys, goats and sheep for the Usangu
plain from 1998 to 2002. The livestock biomass in
Tropical Livestock Units (TLUs) were calculated
according to Bekure et al. (1991) (i.e. 1 cow = 0.7
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 3
TLUs; 1 donkey = 0.5 TLUs; and 1 goat, sheep or pig
= 0.11 TLUs).
Data analysis
Participatory assessment data were analyzed with
the participants to ensure that trends reflect the
situation of the villagers. Quantitative data were
analyzed using the Statistical Package for Social
Science (SPSS). Simple correlation analyses were
performed for total annual rainfall and total
monthly rainfall effects on cereal crop yields, total
and per capita cereal production. The hypothesis
tested is that change in monthly and annual rainfall
in the plain and upland catchments is not associ-
ated with changes in cereal yields, total and per
capita production in the Usangu plain. The correla-
tion analyses of Mtera reservoir water levels and
variances in the amount of annual rainfall in the
upper catchments and plain were calculated to
assess the extent of their associations. These analy-
ses tested the hypotheses that total rainfall amount
in the plain and upland watersheds of the Great
Ruaha River are associated with the water level and
water level variations in the Mtera reservoir down-
stream. The alternative hypothesis states that water
level and variations in Mtera reservoir are inde-
pendent of rainfall in the plain and watershed
catchments.
The least square regression model, Y = a + bX,
was used to estimate linear change in per capita
cereal production, variability of water levels in
Mtera reservoir and livestock biomass over the years
and the relationships between total and per capita
cereal production and total annual rainfall in the
Usangu plain, and mean annual water level with
rainfall in the watershed catchments. Y = calculated
total cereal production, per capita cereal pro-
duction, water level variance, mean annual water
level in Mtera reservoir and livestock biomass, and
X = years over which the data on per capita cereal
production, water level variance and livestock
biomass were derived, and the total rainfall. The
hypotheses tested were that there were no con-
sistent changes in per capita cereal production,
water level variation and livestock biomass over
the years in the Usangu plain, and that the mean
annual water level in Mtera reservoir, total and
per capita cereal production trends are not
significantly described by total annual rainfall
variations.
RESULTS
Environmental and demographic change
Changes in demographics and natural resource quality
and quantity
Participatory trends analyses with villagers in the
study areas show that the human population, land
and water degradation, wind and crop area are
increasing (Figure 1). In contrast, vegetation cover
and grazing areas decreased over the past 30 years.
The villagers’ perceptions in Figure 1 are in agree-
ment with quantitative results from Mwalukasa
(2002) on assessment of land-use change. His
results showed a decline in bushlands, grasslands
and dense woodland and an increase in cropping
area in the upland catchments and irrigated crop-
ping area on the plain. There were inconsistent
perceptions in different villages on changes in tem-
perature. For example, in Mabadaga, Mhwela and
Ikoga villages they thought that temperature has
decreased over time and attributed these changes
to a decline in humidity and increasing wind. In the
other three villages the temperature was reported
to be increasing and there had not been a notable
change in wind. Furthermore, participatory obser-
vations of vegetation cover degradation revealed
the presence of desert-like pockets in the study
villages (Figure 2). None of the households have
access to electricity and depend on fuelwood and
fossil fuel for heat and light.
Change in rainfall patterns in the Usangu plain
The general perceptions of the stakeholders visited
in the Usangu plain show that rainfall duration and
quantity have declined over time. The participa-
tory appraisal results confirmed this observation
(Figure 1). SMWUC (2002) found that rainfall
amount has not declined significantly, while
Sosovele and Ngwale (2002) argued that both
pattern and duration of rainfall has changed in the
Usangu plain. They showed declining trends in
rainfall between 1980 and 2000, using data from
Mbarali station in the Usangu plain. In this study,
analysis of total annual rainfall trends for 15 years
from the Lugerere rainfall station in the Usangu
plain supported this perception (Figure 3). Figure
3 shows that the highest and below average annual
rainfall were received in the mid-1990s to early
2000s (data from 1990–2004). Furthermore, total
Environment change in Usangu plain Malley et al.
4 International Journal of Sustainable Development & World Ecology
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 5
Time interval scored
2000s1990s1980s1970s
Villagers' scores of trends
26
24
22
20
18
16
14
12
10
8
6
4
2
0
Environmental aspect
Wind
W at er degra dat ion
Temperature
Rainfall
La n d degr ada t ion
Land cover
Human population
Grazing area
Cultivat ed area
Figure 1 Villagers’ perception of changes in environmental attributes, resource uses and demography over 30 years
(1970s–2000s) in six villages in the Usangu plain, SW Tanzania
Figure 2 One of an advancing pockets of desert like conditions on the land in Sololwambo village, the Usangu plain,
SW Tanzania
rainfall of > 5 mm in 1998 and 2002 amounted to
5 months, as compared to average rainfall received
in the early 1990s, when it rained for 6 months.
Vulnerability trends in the Usangu plain
and Great Ruaha River sub-basin
Rural conflicts and availability of water and land
resources
Participatory analysis of rural land and water
conflict trends showed that these have increased
steadily since the 1970s (Figure 4). This was further
verified in the household survey which found 69%
of the respondents had conflicts over land and/or
water resources and another 64% reported that
competition for land and water resources for pro-
duction activities are linked to conflicts.
Villagers’ scores for resource availability change
patterns showed that the land and water shortages
have increased over time (Figure 4), particularly
since the mid-1980s. These rapid increases were
attributed to population growth due to immigra-
tion of farmers and pastoralists from different areas
into the Usangu plain. This is verified by a fact
that 61% of the interviewed 266 households are
immigrants from other areas and 80% of them
arrived in the Usangu plain between the late
1970s and early 1990s. Villagers perceive that
decline in land productivity caused by increasing
frequency of droughts and depletion of soil fertility
over time is creating vulnerability in the plain.
Mwalukasa (2002) reported 63% and 75% of land
and water scarcities problems along the Matamba-
Chimala ‘basin’ in the Usangu plain, showing that
water shortage is more critical than land shortage.
Interviews with 266 households showed that 83%,
47% and 43% of respondents attribute increased
water shortage to a decline in rainfall, increased
water abstraction by upstream users and upland
catchment degradation, respectively.
Decline in water flows from upland catch-
ment tributaries into the Great Ruaha River have
been noted since the early 1990s. The river
completely dried up for a few weeks in 1993 and for
111 days in 1999 (WWF 2002). Many streams and
springs have become seasonal, forcing seasonal
movement of pastoralists with their herds to
water and grazing resources in irrigated farmland
up-stream.
Environment change in Usangu plain Malley et al.
6 International Journal of Sustainable Development & World Ecology
Yea r s
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
Total annual rainfall (mm)
1000
900
800
700
600
500
400
300
200
Figure 3 Trend in rainfall pattern over 15 years at Lugerere station in the Usangu plain, SW Tanzania
Food production
Household food security in the Usangu plain
depends on agricultural production. Crop and live-
stock production generates 62% of rural household
cash income in the Usangu plain. On average, 26%
of the household cash income is used for food items
not produced locally or when food production
deficits emerge. Scores in all villages of the study
consistently showed that there has been a decline in
crop yield over the years (Figure 4). The villagers’
perception is supported by an average linear trend
estimate derived from analysis of district per capita
production of cereals (R2= 0.33, p< 0.05), despite
marked fluctuations from year to year. This fluctua-
tion is correlated with the amount of rainfall
received in the Usangu plain, as shown by the
pattern of relatively high rainfall in 1998 and 2002
associated with high per capita production of
cereals. The argument is further supported by the
fact that 42% of total cereal production variations
are accounted for by rainfall variability. Cereals are
major food crops and income source for the
Usangu plain crop growers.
Pastoral societies use livestock sales to purchase
cereals and use milk and meat from livestock. The
assessments of livestock numbers showed
increasing and declining trends over a 30-year
period (Figure 4). On average, there was an increas-
ing trend from the 1970s to 1980s, which decreased
from the 1990s to 2000s. Livestock biomass de-
clined sharply from 1998/1999 values and were
constant in 2000 to 2002 (R2= 0.785, p< 0.05). This
sharp decline may be explained by livestock keep-
ers’ out-migration from the plain in 1998 and 1999
as a result of a government decision to convert dry
season grazing areas into wetlands. This was linked
with a period of eastern wetlands conversion into
the Usangu game reserve in 1998, causing emigra-
tion of pastoralists and their livestock from the
Usangu plain. According to villagers, prohibition
of grazing in the wetlands area resulted into high
mortality of livestock and declining birth rates due
to shortages of water and pastures. The wetlands
area was important to pastoralists’ adaptive mecha-
nisms during the dry season and drought years. In
the household survey, 81% and 77% of respondents
complained about declining income and increased
food shortage frequencies, respectively. The food
vulnerability trends are associated with current
declining trends in crop and livestock production.
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 7
Time intervals scored
2000s1990s1980s1970s
Villagers' scores of changes
16
14
12
10
8
6
4
2
0
Vulne r ability f a c tor
Water shortage
Reso urc es con flic t s
Livestock numbers
Land shortage
Crops yields
Figure 4 Local perception of human vulnerability trends over 30 years (1970s–2000s) attributed to environmental
change in six villages in the Usangu plain, SW Tanzania
Social services and health
Most areas of the Usangu plain are flooded during
the rainy seasons, which make the roads impass-
able. This has implications for rural livelihoods
security, development and social services access-
ibility. The negative impacts have been observed
in agricultural extension, education and health
services accessibility and attendance. Discussions
with professionals revealed that many workers in
agriculture, community development, health and
education either abandoned their jobs or never
reported for duty in villages in rural areas due to the
transport problems. Of the six villages surveyed,
only Mhwela, which is close to the main road, had
an agricultural extension worker. The rest of the
villages depend on ward and/or district agricul-
tural extension workers. In primary schools, pupil
to teacher ratio is 12 pupils higher than the national
average of 45 pupils per teacher. District records
show that only 1% of pupils who enrolled in pri-
mary schools enter secondary education. Discus-
sions with villagers revealed concern over long
distances to medical services, carrying the sick and
expectant mothers on bicycles or their shoulders
to the hospital or dispensaries, and lack of skilled
medical staff who can deal with complications
in expectant mothers and clinical services for
children. The district has only one hospital, located
at the district headquarters. There were 41 dis-
pensaries in the district in 2002, covering an area
of about 16500 km2and 86 villages, or one dis-
pensary for 5200 people, with each village cover-
ing about 192 km2. Furthermore, most of the
dispensaries are ill-equipped and under-staffed,
and more than two villages share a dispensary for
medical services, causing inadequate attendance.
This is supported by the fact that, among the 266
respondents interviewed, 43% have lost a house-
hold member due to malaria. Low infiltration of
runoff in the watershed areas and increasing
floods due to short intense rainfall in the Usangu
plain intensify malaria transmission and outbreaks
of the other waterborne diseases, and inaccessibility
of social services tends to increase health insecu-
rity. In addition, the 266 households survey
result show that 11.3% of households have lost
household members to HIV/AIDS. Overall, the
results suggest that in the Usangu plain there is
higher health risk and declining decent liveli-
hood opportunities.
Wildlife biodiversity
Aerial surveys conducted by SMWUC (2002) in the
wetlands of the Usangu game reserve showed sig-
nificant numbers of topi, impala, ostrich, zebra,
reedbuck, warthog, sable, hartebeest and giraffe.
These animals migrate between the Ruaha National
Park and Usangu game reserve. The wet season
wildlife counts in 1993 and 1996, undertaken by the
Tanzania wildlife conservation and monitoring
unit, indicate that there has been a significant
decline in number of bushbuck, giraffe, hippo-
potami, ostrich, reedbuck, warthog, impala, duiker
and zebra in the Great Ruaha ecosystem (Figure 5).
Water availability for hydroelectricity generation
The variation in water level in Mtera reservoir
increased from 1990 to 2003 (Figure 6). This indi-
cates declining environmental stability in the Great
Ruaha River ecosystem that gradually recharges the
river to feed the reservoir downstream throughout
the year. In 2003 water inflow to Mtera reservoir
was 40% of the 60-year average (Simbeye The East
African, July 12–18, 2004). According to TANESCO,
power generation from the Kidatu plant has
declined, forcing the company to use diesel gen-
eration, purchase power from other sources, and
implement a load-shedding strategy.
Relationship of environmental change to
vulnerabilities
The simple correlations of the annual and the
monthly rainfall to cereal crop yields and total pro-
duction of cereals are shown in Table 1. The total
annual rainfall in the Usangu plain is significantly
(p< 0.01) and positively associated with cereal
production, especially paddy yields. Rainfall for
December, January and February are positively and
significantly (p< 0.05) associated with paddyyields.
February rainfall is significantly (p< 0.05) associ-
ated with maize and paddy yields and showed a
highly significant (p< 0.01) association with total
annual cereal production. These significant posi-
tive associations suggest that yields and total pro-
duction of cereals are influenced by variations in
rainfall amount and period in which it falls on the
Usangu plain. February rainfall in the watershed is
positively and significantly (p< 0.01) correlated to
yields of maize, paddy (Table 1) and total cereal
Environment change in Usangu plain Malley et al.
8 International Journal of Sustainable Development & World Ecology
production. From this analysis, it seems that paddy
performance is associated with watershed rainfall
amount, and the paddy growers noted that good
rainfall in the upland watershed is important to
paddy production in the plain. In Usangu, paddy
crop irrigation depends on water from recharged
rivers and streams or direct flooding of fields by
run-on from watersheds.
The mean annual water level of Mtera reservoir
and its variations are not significantly (p= 0.05)
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 9
Years
19961993
Wildlife popul ation
70000
60000
50000
40000
30000
20000
10000
0
Wildlife type
Zeb ra
Warthog
Reedbuck
Ostrich
Impal a
Hippo
G i r a ffe
Figure 5 Decline in numbers of some types of wildlife in the Great Ruaha ecosystem
Yea r s
200420022000199819961994199219901988
Variance of waterlevels
12
10
8
6
4
2
0
Observed
Linear
Figure 6 Change in variations of water levels in Mtera reservoir in the Usangu plain, SW Tanzania
associated with annual rainfall variability at
Lugerere station in the plain (Table 2). The total
annual rainfall variability in the watershed is signifi-
cantly (p< 0.01) correlated to mean annual water
level and water level variations in Mtera reservoir
(Table 2). These results indicate that there is a rela-
tionship between total annual rainfall variations in
the upland catchments and mean annual water
level and water level variations in Mtera reservoir
downstream.
The total annual rainfall variations in the plain at
Lugerere significantly (p< 0.05) accounted for
42% and 27.4% of variations in total production
and in per capita production, respectively, of
cereals in the Usangu plain (R2= 0.42, p< 0.05).
The watershed total annual rainfall variations
significantly (p< 0.05) accounted for 36.4% of
variations in mean water level in Mtera reservoir.
These variations in rainfall show the extent of the
links between rainfall quantity variations and water
insecurity experienced in the Usangu plain and
downstream in the Great Ruaha sub-basin of the
Rufiji basin.
DISCUSSION
Environmental change and
vulnerabilities: causes and processes
Over-dependence on natural resources for livelihoods
Results of the survey showed that over-dependence
on natural forests for domestic energy is a cause of
Environment change in Usangu plain Malley et al.
10 International Journal of Sustainable Development & World Ecology
Cereal yield
Annual/monthly rainfall Maize Paddy Sorghum Finger millet Total production
Uyole station (upland)
Annual
November
December
January
February
March
0.46
−0.41
−0.41
0.32
*−0.57*
−0.23
0.24
−0.27
−0.39
−0.12
** 0.79**
0.03
0.08
−0.19
−0.34
−0.13
0.06
0.22
0.44
−0.17
−0.38
0.03
0.21
0.22
0.38
−0.34
−0.44
0.02
** 0.84**
0.22
Lugerere station (plain)
Annual
November
December
January
February
March
0.41
−0.45
0.22
0.18
*0.58*
−0.07
** 0.79**
−0.35
*0.55*
*0.68*
*0.68*
−0.14
0.09
0.53
−0.40
0.21
−0.05
0.53
0.38
0.34
0.25
0.26
0.33
*0.56*
*0.65*
−0.45
0.39
0.41
** 0.74**
−0.12
*Significant at p< 0.05, ** significant at p< 0.01
Table 1 Linear associations of the watershed and plain annual and monthly rainfall to cereal crop yields and total
production in the Usangu plain, SW Tanzania (1990–2003)
Rainfall parameters Annual mean water level Water level standard deviation
Uyole station (upland)
Total annual rainfall
Rainfall standard deviations
0.603*
0.580*
0.753**
0.561*
Lugerere station (plain)
Total annual rainfall
Rainfall standard deviations
0.287 ns
−0.203 ns
0.380 ns
0.148 ns
*Significant at p< 0.05, ** significant at p< 0.01, ns, not significant at p= 0.05
Table 2 Associations of the rainfall and rainfall standard deviations to the mean annual water level and water levels
standard deviations in the Mtera reservoir, downstream of the Usangu plain in SW Tanzania (1990–2003)
environmental degradation and subsequent vul-
nerabilities. In this study area, as in many places in
Africa, forests are the main source of domestic
energy. The trends show that the increase in human
population has had a negative impact on environ-
mental resources, mainly via expansion of cultiva-
tion and fuelwood harvesting activities, leading to
degradation of vegetation cover. The importance
of natural vegetation cover for land cover is a source
of environmental services. Vegetation cover acts as
a sponge by slowing the velocity of the pounding
rain and allowing greater infiltration of runoff into
the soil and ground water. This water is then stored
in underground aquifers, and gradually released to
replenish streams and springs when rains recede.
When land cover is lost to degradation, the conse-
quences are accelerated soil erosion, water losses in
the form of short duration floods which destroy
properties, siltation of water channels, land and
water infrastructure downstream, and biodiversity
loss (Lowdermilk 1986). The major effect of vegeta-
tion loss is that many streams and springs dry up or
become seasonal. As captured by this study in the
Usangu plain and other studies (SMWUC 2002;
Sosovele and Ngwale 2002), many rivers are now
seasonal or completely dried up, which is linked to
land degradation in upper catchments. Forest and
other forms of land cover are a natural means for
reducing greenhouse gases in the atmosphere that
are responsible for climate change.
Rainfall variability versus ecosystem and economic
insecurity
Rainfall variability has increased in the Usangu
plain (Figure 3). The variability pattern is in agree-
ment with Karl et al. (1995) and is likely to increase,
such that droughts will become more frequent and
intense, while rainfall will be concentrated in short
and severe storms. The decline in rainfall is asso-
ciated with decline in quality of land, water sources
and vegetation cover (Figure 1). According to
Mbaria (2005), loss of vegetation is strongly linked
to an increase in rainfall variability and decreased
quantity. Furthermore, Mbaria argues that initial
signs of the impact of vegetation degradation are
long periods of drought and short intensive rainfall
that destroys farming activities. The absence of
vegetation creates dry air by disrupting forest vege-
tation function of forming tiny water droplets from
evapo-transpired water which later fall as rain.
The pattern of increasing drought frequency and
short intensive rainfall, and links to environmental
resource degradation is evident from this case study
of the Usangu plain. Furthermore, rainfall variability
explains 42% of variations in cereal production,
29% of per capita cereal production, and 36% of the
water level changes in Mtera reservoir. The decline
in per capita production of cereals is therefore partly
explained by climatic variations over the years. Local
perceptions are supported by scientific analysis that
rainfall is an important factor for rural livelihoods in
terms of food production and cash income to access
food and other livelihoods items.
Drought and reduced water flow in the rivers
and streams from the Usangu catchments that
enter the Great Ruaha River are linked to drying of
the river, which, in turn, had severe impacts on
wildlife, including aquatic fauna (WWF 2002).
According to a Ruaha National Park informant, in
dry seasons (October to December) since 1994,
when the river dried up, mortality of wildlife that
depends on the Great Ruaha River water system was
observed in amphibians, reptiles, fish and hippo-
potami. Some aquatic flora was also lost as a result of
drying of the river and turbid and shallow waters.
Similarly, the water level in Mtera reservoir was
seriously affected by the decreased water flows of
the river from the Usangu plain and its upland
catchments. As a result of power generation failure
in Kidatu dam, because water in Mtera reservoir
was close to its dead level (Yawson et al. 2003), the
government opted for alternative energy genera-
tion. Investment in alternative electric powers
sources by Independent Power Tanzania Limited
(IPTL) and Songas Limited has created a national
debt of US$5.5 million in monthly fixed fees to
these two power firms (Kenge, The East African, June
7–13, 2004): US$2.5 million to Songas for 4 years
and US$3.0 million to IPTL for 25 years. In addi-
tion, the government signed a loan with the World
Bank for US$33.6 million to purchase electric
energy from IPTL for distribution to consumers by
TANESCO, and another US$10.2 million for fossil
fuel imports to avert the load shedding in power
stations (Simbeye, The East African, July 12–18,
2004). These strategies have increased electric
tariffs which, in turn, increased pressure on the
forests for fuelwood as a major source of domestic
energy. Increasing use of charcoal for domestic
heating and intensification of woodland deforesta-
tion are leading to environmental degradation.
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 11
Resource quality change and scarcity
The trends in land and water resource quality
decline are correlated with demographic growth
and expansion of cultivated areas (Figure 1). This
suggests that, apart from direct shortages caused by
an increasing population in need of land and water
resources, activities of the growing population
increase degradation of land and water resources,
which, in turn, intensify scarcity of these resources.
Continuous cultivation without fertility improve-
ment efforts causes depletion of soil productivity
in the study area. Shifting cultivation results into
degradation of the natural vegetation needed for
land protection and climatic change mitigation.
The protective vegetation cover that prevents soil
erosion fails when vegetation cover is removed. Soil
erosion leads to silt deposition in rivers and denu-
dation of the land, resulting into less infiltration of
rain water and increased floods in the low-lying
plains. The recharging of underground aquifers is
reduced by reduced infiltration in the catchments
and floods increase. The floods make roads inacces-
sible for social services during the rainy season.
Impassable roads present difficulties for provisions
of agricultural extension, health and educations
services. In addition, temporary ponds and swamps
resulting from floods provide breeding grounds for
malaria-transmitting mosquitoes.
Change in social-cultural livelihoods
Degradation of the environmental resources leads
to vulnerability of rural people through reduced
livelihood opportunities offered by agricultural
activities. There are trends of social ethics dis-
integration caused by livelihood difficulties in
Tanzania, in the form of prostitution and involve-
ment in the drug trade. Lihawa (1995) reported
that desperate families from interior dryland areas
of Tanzania have exported young girls to provide
cheap labour as barmaids in urban areas, and these
girls become prostitutes to enhance their income.
Lovett et al. (2001) found that large groups of young
men have moved to cities to become itinerant and
occasional participants in the informal urban econ-
omy. A recent UN report cited by the correspon-
dent of The East African (July 11–17, 2005) noted
that some 330 young Tanzanians are held in foreign
jails on drug trafficking charges. In sub-Saharan
Africa poverty is linked to vulnerability of rural
people to contract HIV/AIDS (IFAD 2002). In the
study area, 11.3% of households have lost a relative
to HIV/AIDS, which suggests the presence of such
health risks in the Usangu plain. Those who died of
HIV/AIDS had formerly immigrated to the cities
and either returned when they became sick, or had
a migratory livelihood system between villages and
urban centres. Food shortage and low income lead
to low quality and inadequate nutrition as a result of
declining per capita agricultural production. These
livelihood insecurities, in turn, force people to be
involved in risky livelihood activities such as prosti-
tution and drugs trafficking.
Development policy situation and lessons
Conflicts and contradictions in development actions
Different institutions formulate policies for man-
dated development in Tanzania. At the national
level, there are no mechanisms of coordinating and
harmonizing formulation of policies between the
ministries and, in some cases, even within ministries
and between departments of the same ministry
dealing with the different development sectors.
Each ministry and/or sector tends to implement its
policies irrespective of effects on other sectors’
development objectives. This has led to contra-
dictions and conflicts among development actions
for rural development. The evident conflicts and
contradictions found in the Usangu plain are:
implementation of a policy objective for wildlife
biodiversity conservation and tourism that envis-
ages expansion of protected areas, with the objec-
tives of increased agricultural production, right to
land use for the local people’s livelihoods, water
use and management. The policies that empha-
size increased production through expansion of
cultivation in irrigated areas for food security is not
in harmony with availability of water resources,
hydroenergy objectives and environmental poli-
cies. Continuous upward adjustments of the
electric energy tariffs are in conflict with the envi-
ronmental policy of protecting and conserving the
environment. In turn, pressure on natural forests
increases as a result of harvesting trees for the
fuelwood energy, which causes more environmen-
tal degradation. The agriculture policy that pro-
moted irrigated agricultural expansion affects
water levels in Mtera reservoir, leading to closure of
the cheap hydroelectric energy generation plant in
Environment change in Usangu plain Malley et al.
12 International Journal of Sustainable Development & World Ecology
Kidatu dam at certain times of year. Expansion of
irrigated areas increases water demand for agricul-
tural production activities, placing pressure on
water resources. Deforestation and over-use of land
resources lead to reduced capacity of the natural
functions of the land for environmental services
and in effect removes the sinks for carbon dioxide
sequestration for mitigating climatic changes.
Processes of policy formulation
The processes of policies formulation in Tanzania
are dominated at higher levels by ministries, with
minimum participation of those affected and
expected to implement the policies at lower levels.
This top-down policy formulation process excludes
important stakeholders: local people whose survival
is affected by the policies that regulate use of
natural resources for their livelihoods; civil organi-
zations (NGOs, CBOs, etc.) and research and
extension institutions responsible in the advisory
services; and implementation of development poli-
cies. This process of policy formulation does not
incorporate the practical concerns of sustainable
development. The laws enforcing the policies are
not fully grasped by the implementers and users of
the natural resources.
Development attitudes
Sustainable development requires bottom-up
approaches in order to communicate the needs
and interests of local people to national authorities
in an effective manner. These include promotion
of opportunities for sustainable rural livelihoods
and facilitation of empowerment. This means
strengthening participation of local people in the
political process and in decision making for
enhanced human security, reduced vulnerability to
shocks and increasing their capacity to recover.
In this study it has been noted that the greatest
challenge in the Usangu plain is to change attitudes
in the traditional roles of authorities, professionals
and local people themselves. SMWUC (2002)
observed that new approaches that are based on
participation in decisions, formulation of plans and
division of responsibilities are not necessarily
embraced and supported by officials. This attitude
of excluding local people from the development
process was explicitly evident among natural
resources workers in the district council and
officials of the wildlife and biodiversity conserva-
tion division of the Ministry of Natural Resources
and Tourism. The current policy emphasizes par-
ticipatory approaches in the development process
and seems to be perceived differently by different
people. As a result, participation is mostly inter-
preted as a short consultative process as opposed to
active and continuous participation of the people
in all stages of the development process. Natural
resources workers and wildlife officials tend to
maintain their traditional responsibilities and
authority in order to safeguard their traditional
professional roles as the most knowledgeable and
responsible people to bring development to the
nation. However, local people do not understand
their new roles and are not confident to take up
challenges presented by the new powers and
responsibilities given to them by the on-going
decentralization processes.
RECOMMENDATIONS
It is evident that there is need to establish coordina-
tion mechanisms and integrated planning at
national level between development sectors to
achieve progress towards sustainable development
in Tanzania. Currently, the most urgent need is for
government to prevent increasing crisis from: (1)
escalating electricity tariffs caused by losing opera-
tional capacity of cheap hydroelectric power in
Kidatu dam; (2) decline in wildlife biodiversity in
the Ruaha National Park and Usangu game reserve;
(3) deteriorating environment in the Usangu plain
and Great Ruaha River ecosystem; and (4) increas-
ing human insecurity in rural areas of the Usangu
plain through declining incomes, food insecurity
and loss of peace through increased inaccessibility
of resource use and competition over resources.
TANESCO should invest in the environment
through buying half of the water used for irrigation
from paddy farmers. This would be a cheaper
option and more sustainable than continually pur-
chasing electricity from expensive sources such as
IPTL and Songas. Compensation for farmers for
production lost by not using water is worth consid-
eration and would allow adequate river flows to
Mtera reservoir and Kidatu dam. Advantages of this
option are: (1) water would be available for live-
stock and domestic uses downstream, this would
contribute to human security and peace; (2) pre-
vent complete drying of the Great Ruaha River and
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 13
mitigate the impact on wildlife in the Ruaha
National Park; (3) withdrawn cultivated irrigable
land could be available and utilized as dry season
grazing for pastorals; (4) environmental conserva-
tion and management would be enhanced; and (5)
investment in Great Ruaha power projects (Mtera
reservoir and Kidatu dam) would continue to be a
source of cheap power for TANESCO. Ruaha
National Parks and TANESCO, assisted by the
government, may sustain their business, livelihoods
of rural natural resource users in the Usangu plain
and its catchments, and contribute to environ-
mental sustainability of the Great Ruaha ecosystem.
There should be collaboration between sectors and
stakeholders in formulation and implementation
of development policies.
The process of formulating development policy
should start at the village, ward, district, region and
then national level. This could be part of capacity
building and empowerment of local people in
democratization and decentralization of develop-
ment processes. The role of professionals and
district authorities should change from an authori-
tative approach towards development to that of
facilitator and partner in pursuance of sustainable
development goals. This needs continuous capacity
building in facilitation skills of the professionals
and authorities in order that local people can
freely participate in decision-making processes and
planning of development activities. Jointly agreed
plans should be implemented through division of
responsibilities between local people, professionals
and local authorities. This approach would capture
needs and interests of local people at grassroots
level and would incorporate in-depth understand-
ing of practical constraints and needs. Active parti-
cipation in policy formulation could contribute
to harmony, enhanced understanding of policy-
enforcing laws, practical realistic plans and imple-
mentation strategies for rural development.
Poor socio-economic services and physical infra-
structure affect sustainability of rural economic
activities and sustainable development in many
ways. To stimulate rural production and ensure
sustainability of the environment and development
processes, research and extension services must be
provided on a continuous basis. Poor infrastructure
greatly affects research and extension services in
the Usangu plain. In addition, poor infrastructure
leads to waste of time as a result of low accessibility
to quality health services. Improved physical infra-
structure would increase access to inputs and mar-
kets for profitable production and access to health
services. Poor access to health and education facili-
ties at village level increases human insecurity, since
education and health are fundamental factors for
building social capital for sustainable development.
Development of education and health facilities at
village level should be given highest priority by the
government as an important means of removing
risks to livelihoods. In addition, education would
promote innovation in people to adopt sustainable
development practices through exposure to new
ideas and knowledge.
People-centred environmental management is
needed in the southwestern Tanzania highland
mountain ranges, the major catchment of the Rufiji
River basin. For rapid adoption of resource conser-
vation measures and successful management of
resources, a policy of environmental services pay-
ments to poor local people should be in place. This
provides incentives for local people to forego their
resource use and ensures commitment to environ-
mental conservation in the highlands watersheds.
ACKNOWLEDGEMENTS
The International Foundation for Science (IFS) in
Sweden and the United Nations University Institute
of Advanced Studies (UNU/IAS) in Japan finan-
cially supported this study. The Ministry of Agricul-
ture and Food Security in Tanzania made its
research facilities available for the work and admin-
istered the funds at the Agricultural Research
Institute-Uyole (ARI-Uyole). We would like to
thank Mbarali District council, Ruaha National
Parks, TANESCO, the Regional Water Engineer-
Mbeya and the Directorate of Meteorological
Services, Tanzania for use of their records, needed
to undertake the analysis. The authors also wish to
thank C. Kabungo, S. Swai, J. Kaijage of ARI-Uyole
and the Mbarali District Council agricultural
workers, from , district to , village level, for partici-
pation in field data collection. We thank K.L. Haule
of ARI- Uyole for reading the manuscript and help-
ful comments.
Environment change in Usangu plain Malley et al.
14 International Journal of Sustainable Development & World Ecology
REFERENCES
Bekure S, De Leeuw PN, Grandin BE and Neate PJH.
Maasai herding: An analysis of livestock production
system of Maasai pastoralists in Eastern Kajiado District,
Kenya. International Livestock Center for Africa
(ILCA) system Study 4. Addis Ababa: ILCA;
1991:154 pp.
Correspondent. Tanzania now world’s fourth largest
producer of Marijuana. The East African 2005; July
11–17
Intergovernmental Panel on Climate Change (IPCC).
Climatic Change 2001. Impacts, Adaptation and
Vulnerability. Adaptation to Climate Change in the
Context of Sustainable Development and Equity.
Cambridge: Cambridge University Press; 2001:
chapter 18
International Fund for Agricultural Development
(IFAD). Assessment of Rural Poverty: Eastern and
Southern Africa. Report for Eastern and Southern
Africa Division, Programme Management Depart-
ment. Italy: IFAD; 2002
Johnson I. The Comprehensive Development Framework
(CDF) and Country Assistance Strategies (CASs):
Integrated Approach to Land and Water Issues.
Washington DC: World Bank; 1999
Karl TR, Knight RW and Plummer N. Trends in High
Frequency of Climatic Variability in the Twentieth
Century. Nature 1995;377:217–20
Lihawa HA. Rural-urban female migration in Tanzania:
The case of barmaids in Dar-es-Salaam. MA thesis,
University of Dar-es-Salaam; 1995
Lovett J, Quinn C, Kiwasila H, Stevenson S, Pallangyo
N and Muganga C. Overview of Common Pool Resource
Management in Semi-Arid Tanzania: A review of
common pool resource management in Tanzania. York:
DFID; 2001
Lowdermilk WC. Man-made deserts. In Morgan RPC
(ed.), Soil Erosion and its Control. New York:
Hutchison Ross; 1986
Malley ZJU, Taeb M, Matsumoto T and Takeya H.
Rural livelihoods insecurity and degradation of
land and water resources in southwestern Tanza-
nia 1: implications for sustainable agriculture
development. The International Journal of
Agricultural Sustainability, in press
Mbaria J. Politics behind settlement of Maasai Mau
forest. The East African 2005; March 7–13
Mpinga J. (1999). Usangu: Where the Gods are Not
Just Crazy, But Angry, Too. The East African 1999;
July 7–13
Mwalukasa EH. Effects of Land Degradations in the
Uplands on Land Use Changes in the Plains: The Case
Study of Chimala River Catchment in the Usangu
Plains. MSc Dissertation, Sokoine University of
Agriculture; 2002
Simbeye F. Songas to Replace Imported Furnace Oil.
The East African 2004; July 12–18
SMUWC. Baseline 2001. http://www.usangu.org/
baseline2001/part1–5.shtml; 2002
Sosovele H and Ngwale JJ. Socio-economic Root Cause of
the Loss of Biodiversity in the Ruaha Catchment Area.
Dar-es-Salaam: WWF-Tanzania; 2002
Stocking M. Agro-diversity, Environmental protection
and sustaining rural livelihoods: the global view. In
Brookfield H, Padoch C, Parsons H and Stocking
M (eds), Cultivating Biodiversity: Understanding,
Analyzing & Using Agricultural Diversity. London:
ITDG Publishing; 2002
Tanzania Wildlife Conservation Monitoring. Aerial
Census in the Greater Ruaha Ecosystem. TWCM/FZS
Wildlife Survey Report. Arusha: TWCM; 2000
Wangari M. Environment should be at the centre of
policy in Africa. The East African 2005; June 6–12
World Wildlife Fund (WWF). African Rivers Initiative:
Concept Paper. Living Water Programme. Iringa:
WWF; 2002
Yawson DK, Kashaigili JJ, Kachroo J and Mtalo FW.
Modelling the Mtera-Kidatu reservoir system to
improve integrated water resources management.
International Conference on Hydropower. Arusha:
Hydro Africa; 2003
Environment change in Usangu plain Malley et al.
International Journal of Sustainable Development & World Ecology 15
