Transcript The Role of Anthropogenic Fire and Livestock Grazing in

The Role of Anthropogenic Fire and Livestock Grazing in
Sustaining Avian Diversity in East African Savannas
Nathan Gregory1 and Ryan Sensenig2
1. Department of Ecology and Evolutionary Biology, Princeton University. Princeton, New Jersey 08544, USA.
Email: [email protected]
2. Plant Sciences Department, University of California at Davis. One Shields Avenue, Davis, California 95616, USA
Control
10
Burned
8
6
4
2
0
•What are the mechanisms driving these changes?
As land use change has been identified as the most important factor
threatening global biodiversity, particularly in savanna ecosystems (Sala
et al. 2000), it is essential to develop an understanding of how savannas
function and how human impacts can be used to good advantage.
Determining the manner in which anthropogenic disturbance influences
community dynamics has important management implications and can
shed light on the mechanisms underlying the maintenance of diversity.
9
81
Plot Size (Ha)
Figure 4: 9 and 81 hectare sites host greater densities of
birds relative to controls
Number of Individuals per 100
•How do the effects of these disturbances compare?
1
Burned plots also host a slightly more diverse bird community than
controls. While species-area curves indicate that richness in controls and
burned plots do not differ significantly, abundance influences diversity
measures. The abundance-based coverage estimator indicates that
burned plots are more diverse for both 81 hectare plots (Figure 5) and 9
hectare sites (Figure 9). A similar pattern is observed in the winter.
Species Accumulation Curve: 81 Ha Burn vs. Control
70
60
Number of Species
•In a system where normal disturbance regimes have been altered,
how do the practices of controlled burning and livestock grazing
(bomas) affect avian diversity, community composition, and
abundance?
Figure 3: An example
of a controlled burn
plot with transects
appearing in red.
50
81 Burn
40
81 Control
81 Burn ACE
30
81 Control ACE
20
10
0
0
200
400
600
800
1000
1200
Number of Individuals
Boma vs Postburn vs Control
Figure 5: Species accumulation curves for 81 hectare burned and control sites
indicate no significant difference in species richness. ACE values (large
diamonds) show that burns are more diverse.
Number of Species
9 Control
9 Burn
40
Boma ACE
30
Control ACE
Burn ACE
20
0
Percent Occurence of Bird Species Detected in All Burned vs.
Control Areas
100
200
300
400
500
600
700
800
900
Number of Individuals
Figure 9: Species accumulation curves for bomas, 9 hectare burns, and control sites.
Large diamonds are ACE values. Bomas are the most diverse of the three
treatments.
100
90
80
70
60
50
40
30
20
10
0
Summer: Percent Occurence of Bird Species in Burns, Bomas, and Controls
100
90
80
Species
Figure 6: Percent occurrence of species observed at least 10 times on burned vs. control
sites during the summer.
70
Boma
Burn
Control
60
50
40
30
20
10
0
Control
Burn
Boma
Species
Figure 11: Preferences of species observed at least 10 times during the winter
for all three habitats. Bars outlined in yellow indicate palearctic migrants.
Mechanisms
Arthropod Biomass by Treatment and Collection Method
2.5
2
1.5
Pitfall
1
Sweepnet
0.5
0
Boma
Burn
Unburn
-0.5
Treatment
Figure 12: Arthropod biomass by treatment.
Disturbance appears to
negatively impact food
resources such as
arthropod abundance
(Figure 12) which suggests
that habitat characteristics
may be key to structuring
avian communities.
However, further
experiments are underway
to disentangle the
confounding effects of
predation by birds on
arthropod communities.
Conclusions
•Avian abundance and diversity increase on burned plots, and these
areas attract a unique assemblage of birds. These effects vary with the
size of the disturbance.
•Bird species richness in bomas is higher than in burns of similar size,
and some species show strong preferences for bomas over other
habitats.
•In general, most of the species observed show strong preferences for
disturbed habitats, and only some Ploceus weaver species, queleas,
rattling cisticolas, and some shrub-loving species are specifically
attracted to undisturbed sites.
•The prevalence of undisturbed habitat may be a relatively recent
phenomenon resulting from post-colonial human impacts on fire and
grazing regimes. The lack of birds specializing in these areas points to
the key role of disturbance in maintaining avian diversity in East African
savannas.
•Conservation of bird diversity will require the incorporation of both types
of disturbance into management strategies.
Acknowledgements
Burn
Control
70
60
Boma
50
Burn
40
Control
30
20
10
0
AFGL
WAST
NWCS
GRPI
BEGS
YNSP
PAFL
RNDO
CSBA
PBPI
BNMO
CAWO
SPWE
SUST
CHSP
COBU
CODR
YRSE
RBQU
RACI
RUSP
CRPL
WESP
PPCI
ETSW
Methodology
This part of the study is a component of the Scale and Fire Ecology
Project (S.A.F.E.), a collaboration that includes Ryan Sensenig of the
University of California at Davis and Isaac Kimathe of the University of
Nairobi. Experimental plots are organized into treatments that combine
burns of three sizes (81 hectares, 9 hectares, and 1 hectare) with
control plots of equal area. This design is replicated four times.
To measure bird diversity and abundance, we employ the fixed-width
transect method (Bibby 2000). Each plot has two transects that begin in
the center, run to the corners, and extend 300 meters into the matrix
(Figure 3). Observers record the locations of every bird detected within
50 meters on either side of the transect.
Boma
50
Percent Occurence
Fire: Effects on Avian Abundance, Diversity, and
Community Composition
60
0
ch W
er att
Bl
' l
ue Pla s S ed
-e C in- pa Sta
Ye ar he ba rro rl
llo ed st ck w- ing
w G nu ed la
-n lo t P rk
ec ss W ip
k y e it
C Pal ed Staave
he e S r r
C
F p l
ar S stn lyc urf ing
di u ut a ow
p
Bl nal er Sp tch l
Af ue G Wob S arr er
ric -na ra o ta ow
an p ss dp rlin
e l e
Sl C Gr d Mand ck g
at r ey o P er
e- es F u ip
c t l s i
R olo ed ycaebir t
in ur F tc d
g e ra h
R R -ne d B nc er
Ye ed ufo ck o oli
llo -fac us ed ubon
G w-ru Ched Sp Dovu
m in- Croarr e
N rey
o
p s
or
th bac C ed po mb w
er
ke om Se t B ec
n
W Sd C mo ed atis
hi p a n ea
te ek m B te
-c e ar ul r
R
C
ed N o row 's Wop bul
-c ub mm ne e ter
he ia o d av a
ek n W n S er
h
R ed oo Dro rik
ed C d n e
R -bi ord pec go
at lle o k
tli d n- er
ng Q b
l
W Ci ue eu
ea st le
ve ico a
r s la
p_
This study is taking place on three ranches in the Laikipia District of
central Kenya (Figures 1 and 2). Since at least the 1950s, fire has been
actively suppressed, and in its absence, bomas serve as the major
landscape modifier. The experimental plots are located in Acacia
drepanolobium-dominated habitat of homogeneous structure and soil
type.
70
10
Fi
s
Study Area
Community composition differs between burns and undisturbed
areas. For those species encountered at least 10 times, most displayed
a strong preference for burned sites (Figure 6). Only red-billed queleas,
rattling cisticolas, and some Ploceus weavers show a clear preference
for the undisturbed habitat.
Percent Occurence
Figure 1: Map of Kenya
with Laikipia District shown
in red
Figure 2: Land use map of the Laikipia
District. White areas are large, private
commercial ranches. The approximate
study area is shown in red.
80
Percent Occurence
12
80
W
pp att
ed led
St
So
a
ci
N
al rlin
or
g
th
er We
Bl
av
n
ue
W
er
P
he
-e
i
e
ar
at
ed d W
e
he ar
G
lo
ss ate
ar
y
C
St
ro
ar
wn
lin
e
g
d
Su
pe Plo
ve
rb
r
St
Ba
ar
lin
rn
Sw g
al
Sw
lo
R
a
w
in
l
l
o
gw
ne
sp
ck
Et
_
hi
ed
op
D
ia
Is
ov
n
ab
e
S
el
w
lin
al
lo
e
w
W
R
h
uf
ea
ou
te
G
s
ar
R
re
S
a
p
yttl
ar
ba
in
ro
g
ck
w
C
ed
is
tic
Ca
ol
m
a
a
C
r
o
om
pt
er
m
a
on
Bu
lb
ul
14
90
yca
16
100
G
re
18
Winter: Percent Occurence of Bird Species in Burns, Bomas, and
Controls
Weight (g)
20
Fire suppression has made grazing one of the dominant forces
structuring the landscape specifically in the form of bomas, thorn-scrub
corrals, built by pastoralists. These structures house livestock at night, and
once abandoned, these areas, denuded and enriched with dung, become
productive grazing lawns with distinct plant communities that persist for
decades (Stelfox 1986, Augustine 2003). Recently abandoned bomas have
a very distinctive structure and are roughly 9 hectares in size (Figure 7).
Methodology
Survey methods are the same as
for the fire component of the study.
Transects begin in the center of the
boma and proceed 300 meters into
Interior
the matrix. 8 bomas are being
surveyed.
Results
Scrub fence
Bomas host densities of birds
Edge of
similar to burned sites (Figure 8) but
impacted
area
are more diverse than both burned
sites and controls (Figure 9). This
Figure 7: Diagram of a typical boma site
pattern holds during the winter.
As is the case with burned
Bird Density by Treatment
areas, some species of birds
20
show strong preferences for
18
bomas (Figure 10). There are
16
14
some size-related differences as
12
well. For instance, northern
10
8
white-crowned shrikes show no
6
preference between 81 hectare
4
2
burned or control sites but show
0
a clear preference for 9 hectare
Control
Burn
Boma
Treatment
burns relative to controls. In the
winter, migrants show strong
Figure 8: Comparison of bird density between
preferences for disturbance
controls, 9 hectare burns, and bomas
(Figure 11).
2
Bird Density by Plot Size
Bomas and Fire: Comparative Effects on Avian
Abundance, Diversity, and Community Composition
m +/- Std Dev.
In Africa, savanna ecosystems are sustained through the interacting
effects of fire and grazing, and these disturbances have long been
influenced by humans (Werner 1990, Roques et al. 2001). Land use
changes since European settlement have resulted in the suppression of
large-scale wildfires, but there is growing interest among land managers
in using controlled burning as a tool to improve grazing conditions for
livestock and native ungulates. The increased use of prescribed fire is
likely to change the savanna landscape and, by extension, its flora and
fauna. We are investigating the effects of a modified disturbance regime
on avian abundance, diversity, and community composition by asking the
following questions:
Results
Figures summarize results from data obtained from June through
August. Significantly higher densities of birds occur on 9 and 81 hectare
burned sites (Figure 4).
Number of Individuals per 100 m2 +/Std Dev.
Introduction and Research Questions
Co
BBoumrnnt rol
a
Species
Figure 10: Preferences of species observed at least 10 times during the
summer for all three habitats.
Funding:
Princeton University EEB
Princeton PEI/ STEP
Wildlife Conservation Society
Teresa Heinz Scholars for Environmental Research
David A. Gardner Magic Project
American Museum of Natural History
Cooper Ornithological Society
Thanks to:
David Wilcove
Ryan Sensenig
Hussein Mohamed Dan Rubenstein
Claire Kremen
Michaela Hau
Andy Dobson
Nick Georgiadis
Mpala Research Center and staff
References
•Augustine, D. J. 2003. Long-term, livestock-mediated redistribution of nitrogen and phosphorus in an East
African savanna. Journal of Applied Ecology 40:137-149.
•Bibby, C. J. 2000. Bird census techniques, 2nd edition. Academic Press, London ; San Diego, CA
•Roques, K. G., T. G. O'Connor, and A. R. Watkinson. 2001. Dynamics of shrub encroachment in an African
savanna: relative influences of fire, herbivory, rainfall and density dependence. Journal of Applied Ecology
38:268-280.
•Sala, O. E., F. S. Chapin, J. J. Armesto, E. Berlow, J. Bloomfield, R. Dirzo, E. Huber-Sanwald, L. F.
Huenneke, R. B. Jackson, A. Kinzig, R. Leemans, D. M. Lodge, H. A. Mooney, M. Oesterheld, N. L. Poff, M.
T. Sykes, B. H. Walker, M. Walker, and D. H. Wall. 2000. Biodiversity - Global biodiversity scenarios for the
year 2100. Science 287:1770-1774.
•Stelfox, J. B. 1986. Effects of Livestock Enclosures (Bomas) on the Vegetation of the Athi Plains, Kenya.
African Journal of Ecology 24:41-45.
•Werner, P. A. 1990. Ecological Determinants of Savannas - Abiotic and Biotic. Journal of Biogeography
17:401-402.