Transcript Eastern Arc stability? - Nelson Institute for

Photo © info: www.easternarc.or.tz
Monitoring human impacts and ecosystem function
in the Eastern Arc biodiversity hotspot of Tanzania
and Kenya.
Rob Marchant, Antje Ahrends, Andrew Balmford, Neil Burgess, Jemma Finch,
Alistair Jump, Jon Lovett, Colin McClean, Amos Majule, Cassian Mumbi; Phil Platts,
Carsten Rahbek, Stephen Rucina, Pius Yanda
Environment Department University of York
Managing Dynamic Ecosystems understanding ecosystem dynamics
•Palaeoecology, Biogeography, Phylogenetics
•Modelling: developing bioclimatic approach
•What are controls on Eastern Arc ecosystems
•How are biodiverse areas formed
•How will they change in the future
•Use information to aid in valuation of services
•Scenario and capacity development
The Eastern Arc Mountains: ideal for studying
environmental change impacts
Ancient: In excess of 30 MY old
(Kilimanjaro 1-2 MY) Forests
‘are’ remnants of Miocene panAfrican tropical forest
Diverse geography: Upland
areas vary in size and
connectivity
Biodiversity hotspot: High
proportion of endemic species
Socially important: 80%
reduction in forest cover,
important watershed, HEP,
produce, tourism.
Palaeoecological ‘desert’:
a single site to understand
long term ecology
IOD: El Niño-like coupled
ocean-atmosphere system –
only differentiated from El
Niño in 1997. Character and
periodicity still being
researched.
The IOD: the unsung driver of
climate change in Eastern
Africa. Marchant et al., 2007
Climate controls
•ITCZ
•Trade winds (NE/SW)
•Altitude – lapse rate
•Atlantic Ocean
•Pacific Ocean
•Indian Ocean
Palaeoecological Indicators
• Pollen
• Stable isotope
• Phytoliths
• Charcoal
• Plant & Insect Macrofossils
Tropical areas are
highly responsive to
environmental
changes – indeed
numerous times the
tropical records have
been precursors of
environmental shifts
‘recorded’ at highlatitudes
Stager et al., 2007;
Vershuren et al., 2005
Depth
D
es
at
37726
0.00
350.00
yr
s
B
P)
20
40
20
40
60
20
40
20
40
60
80
nc
e
20
20
20
ro
te
R au b ty
p
A i ac e
m ea
a e
A ran un
na th d
c a i
B ar d ce ff.
or i a
a a e
C gin cea un
ar a e di
ff
y c
A oph eae und .
ni
s yl u iff.
C eia lac ndif
r o s ea f.
t im e
E on- .
un
up ty
di
h p
ff.
S orb e
hi
i
ra a
k
Fa i a
ba
H ce
ym a e
A eno un
ca c di
c a ff
Ti i a rdi .
a
l ia
ac
c
C ea
id
el e
a
tis
Upper montane forest
P
A
p
ca ha
l la
A yph r to
lc a sh
ty
pe
M orn
ac ea
a
t
N r a yp
eo ng e
b a
A ou
po to
d n
La yte iam s typ
d
A i ace im e
nt
h ae idi
Lo ocl un ata
ra eis di
Ly nth ta- ff.
co ac typ
po ea e
di e u
um n
Fi
di
cu
ff.
s
M -ty
or p
el e
la
sa
S
li c
yz
ifo
y
li a
P gi u
od m
oc c
ar om
pu p
s .
E
C
l if
fo
H rtia
ag n
e
C nia itid
u
us
s a la
P on bys
ol ia s
ys -t in
V cia yp ica
er s e
n
C oni ful v
ya e a
th a e
ea u
- ty nd
pe iff.
co
t
N ea
ux u
i s
O a/F am
le ic ba
a al r
ho en
a- se
ty
pe
O
A
ca
nt
st hac
er e
ac ae
ea u
e nd
un iff
di
ff.
Ile
x
m
C iti
or s
n
E us
ric v
ac olk
ea en
e sii
un
di
ff.
A
al
(c
Eastern Arc stability?
Dama Swamp
Udzungwa Mtns
vs.
Deva-Deva
Uluguru Mtns
Montane forest
Lowland forest
50.00
100.00
200.00
300.00
>42606
40
60
Zone
200
Deva 3-b
3870
Deva 3-a
9008
150.00
Deva 2
23177
Deva 1-b
250.00
Deva 1-a
Biogeographical Research
Using plant distribution data
(2500 plots) to extrapolate forest
types and character. Initially need
to check for artefacts within the
data
Are we stating conservation
priority areas in a circular fashion
while potentially overlooking
other important areas and basing
theories on distribution of species
richness on a biased pattern?
Nectarinia
loveridgei
(Finch)
Impatiens sp. (Menegon)
Data
 Species data:
 W3Tropicos from Missouri Botanical Gardens > 26,000 specimens
Records in KITE database
 Funding data:
 ODI Tropical Forests Information System, Critical Ecosystem
Partnership Fund, Other multi- and bi-lateral donors and
indiv.
 Environmental data:
 Forest area and altitudinal range, 8 climate variables (CRESS)
Other
Funding
.92
.69
Species richness
-.23
PC productivity
-.05
PC precipitation
.80
PC heterogeneity
& area
Climate Surfaces – for
regional modeling
•0.05 dd CRES African surfaces.
(Hutchinson)
•8 Climatic Variables:
-Mean temp warmest month
-Mean temp coldest month
-Monthly temp range
-Mean total annual precipitation
-Moisture index
-Mean total precipitation wettest month
-Mean total precipitation driest month
-Mean total precipitation warmest
month
Elevation
Aspect
Species presence/absence
Slope angle
Identify poorly represented
response curves
Fieldwork in Tz
Clean
data
Climate model
(USGS)
(CRES)
Validate site locations
Species
presence/absence
Weight
absences
Elevation model
Climatic
predictors
Topographical
predictors
Test for
correlations
Determine complexity of species-environment response curves
Detect spatial
autocorrelation
Refine model
Guide fieldwork /
ground truth
Calibrate generalised additive model
Crossvalidation
Crossvalidation
Predict habitat suitability
1.0
0.5
0.0
Probability
distribution: Myrsine
melanophloeos
Ground-truthing
North Pare Mountains
Newtonia buchananii
Modelling the Eastern Arc
September 17, 2007
Podocarpus milianjanus (Habitat Suitability / Probability of occurrence)
Present
2025
2055
2085
Species richness (habitat suitability scores summed over 120 tree species)
Present
Platts, 2007
2025
2055
2085
Linking science with
stakeholders to sustain
natural capital
Our vision: building a robust, scientifically
credible and practical framework which
captures the true value of natural capital in
development decisions for the Eastern Arc
Valuing the Arc concept
1. Inventory services, people &
landscapes
2. Model & map service
production & use
7. Explore
plausible
scenarios
3. Model & map service values
4. Measure & map
conservation costs
•
•
•
•
•
•
•
biodiversity
hydrology
carbon
timber
non-timber
pollination
tourism
5. Map governance
structures
6. Map winners & losers
8. Design mechanisms to
capture service values
Preliminary map of water provision
Climate
Topography
Landuse
Water abstraction / use
SWAT InVest
Preliminary map of carbon storage
Climate
Topography
Landuse
Permanent plots
ID and measurement
Re-measure in 2010
REED
Needs and gaps
Data needs
•Inventories from under researched areas (BREAM)
•Land use map and land use change map – ‘complicated’
•Climate / environmental data
•Ground-truthing of generated data
Methodological needs
•Accounting for potentially bias in existing data
•Model development (dispersal, animal interactions, land use)
•Climate model development – inapprporiate climate models
•Incorporation of socioeconomic trends within scenarios
•Linking of results to policy (international to village)
•Move through space and time scaling issues
Capacity needs
•Teaching and direct dissemination of information (resource) within
Governmental, Intergovernmental and NGO organizations
•Portrayal of results in an appropriate manner
•Linking exploratory tools such a models, INVEST, MARXAN
•Generation of mechanisms (e.g. REDD) and appropriate
governance to maximize opportunities