1. Kyuma R.*, Wahome R., Kinama J. and Wasonga V. O.
*Corresponding Author: Kyuma R., Email: kavilakyuma@yahoo.com or kavilakyuma@gmail.com
The effects of climate variability on Prosopis juliflora
spread, vegetation trends and livestock dynamics in
the drylands of Kenya

2. Introduction
• Prosopis was introduced in Africa in the 1820s and
in Kenya in the 1970’s and 1980’s
• Prosopis juliflora (Sw.) DC has been aggressively
taking over grazing lands in pastoral areas
• Efforts to eradicate it have not succeeded
anywhere in the world.
• Due to its very hardiness and versatility, it grows
fast in dry degraded lands and wastelands

3. Prosopis Juliflora – an aggressive invader
It forms impenetrable thickets if not well managed

4. 4
It is evergreen
It is a prolific producer

5. Problem statement
• The relationships between climate variability and
Prosopis invasion on pastoral livestock production
systems and livelihoods is not well understood.
Justification
• There is no evidence that Prosopis spread patterns
are related to climate variability and how that in
turn affects vegetation trends and livestock
population dynamics in the ASALs.
Objective
• To determine the effect of climate variability on
Prosopis spread patterns, other vegetation trends
and livestock population dynamics

6. Materials and methods

7. Magadi
SHOMPOLE
OLKIRAMATIAN
MAGADI
OLDONYO-NYOIKE
36°30'0"E
36°30'0"E
36°20'0"E
36°20'0"E
36°10'0"E
36°10'0"E
36°0'0"E
36°0'0"E
S 1°40'0"S
S 1°50'0"S
S 2°0'0"S
S 2°10'0"S
Legend
Magadi Locations
L. Magadi
0 10 205 KM
WAJIR
TURKANA
MARSABIT
KITUI
GARISSA
ISIOLO
TANA RIVER
MANDERA
NAROK
KAJIADO
KILIFI
SAMBURU
TAITA TAVETA
BARINGO
KWALE
LAMU
WEST POKOT
Legend
L. Magadi
Magadi Locations
Dryland Counties with Prosopis
Kenya Boundary
0 140 28070 KM
Study site – Magadi Division of
Kajiado County - Kenya
Represents about 8% of the drylands
of Kenya affected by Prosopis.
Located in the southern rangelands;
lowland agroecological zone
Two landscape types were delineated
(plains and hillslopes)

8. Materials and methods Cont’d
Data types and sources
• Time series rainfall and temperature data were extracted
from the climate data collected from meteorological
stations climate data for 20 years
• Vegetation and Prosopis productivity data was derived
from the Terra MODIS (250m) series vegetation indices
Normalized Difference Vegetation Index (NDVI) satellite
data
• NDVI data was downloaded from the ENDELEO website
(http://endeleo.vgt.vito.be/),
• Livestock population data was collected and collated from
livestock offices – Magadi and Kajiado and KNBS

9. GIS and remote sensing methods of
vegetation biomass estimations
• MODIS (250m) satellite derived NDVI images were
used to establish the spread patterns of Prosopis in
the area.
• The 250m NDVI images were used to identify the
vegetation types which were photosynthetically
active during the drought periods.
• These vegetation types were most likely Prosopis
plants.

10. GIS and remote sensing methods
cont’d
• Participatory mapping of Prosopis clusters using community
opinion leaders was done and focused group discussions
were used in the plains and the hillslopes landscapes.
• Land cover and land use data, soil data, GPS data, GIS
databases were used to identify areas with the suitability
characteristics for Prosopis to thrive.
• The disappearance of other plant species was tracked using
NDVI from MODIS (250m) satellite images, land use, land
cover soil data and GPS data.

11. Data Analysis
• Monthly total rainfall and average temperature
trends for 20 years were established using Ms.
Excel
• Vegetation and Prosopis productivity (NDVI) for
the plains and hillslopes (2000 – 2014) for the
dry seasons’ trends were established
• Livestock population trends (20 years) was
established

12. Data analysis
• Comparisons and relationships of the trends was
done in the plains and the hillslopes landscapes
• Correlation coefficients were determined for the
relationships between rainfall and temperature,
Prosopis spread patterns, other vegetation cover
trends and livestock population.

13. Results
Annual rainfall totals trend in Magadi, Kajiado
Monthly rainfall totals in Magadi, Kajiado

14. Annual average temperatures in Magadi, Kajiado

15. Land cover change trends (1987-2008).
Prosopis invasion in 2014

16. Olkiramatian
plains short dry
season (January
to February)
Prosopis NDVI
trends
Olkiramatian
plains long dry
season (June
to September)
Prosopis NDVI
trends

17. Ngurumani hill
slopes short dry
season (January
to February)
Prosopis NDVI
trends
Ngurumani hill
slopes long dry
season (June
to September)
Prosopis NDVI
trends

18. Olkiramatian all
vegetation
short dry
season
(January to
February) NDVI
trends
Olkiramatian all
vegetation long
dry season (June
to September)
NDVI trends

19. Ngurumani all
vegetation short
dry season
(January to
February) NDVI
trends
Ngurumani all
vegetation long
dry season
(June to
September)
NDVI trends

20. Short dry
season
maximum
NDVI
comparisons
Long dry
season
maximum
NDVI
comparisons

21. Olkiramatian location livestock population
trends

22. Conclusions
• The study revealed decreasing and variable rainfall
amounts and patterns; and an increase in mean
annual temperatures in the study area
• vegetation cover was noted to decline especially
during the long dry seasons when livestock feed
supply is limited
• Prosopis cover was increasing during the same
period. The cattle populations were also on the
decline over a 20 year period while the sheep and
goats populations were on the increase.

23. Recommendations
• It is recommended that viable Prosopis utilization
options be explored to take advantage of the highly
adaptable Prosopis to climate variability.
• The economics of Prosopis pods and Prosopis
carbon stocks as an alternative source of animal
feeds and Prosopis based carbon trade need further
studies.

24. References
• Wahome et al., 2008;
• Choge and Pasiecznik, 2006
• Tewari et al., 2000;
• Pasiecznik et al., 2001
• Silva, 1986
• UNEP/CBD, 2010
• GOK- Kenya PDNA, 2012
• Kazmi et al., 2010
• Galvin et al., 2004; IPCC 2007
• Resilience Alliance, 2010;
• Tennigkeit T. and Wilkes A., 2008;
• WISP Policy Note No. 04, 2007