Productivity in Cloud Forests in the Andes
Carbon cycle measurements and model results from JULES
Toby Marthews
http://marthews.tripod.com
Leeds, 10th June 2010
at the Oxford University Centre for the Environment (OUCE)
in association with the Andes Biodiversity and Ecosystem
Research Group (ABERG) and the RAINFOR network
Cloud Forests
Cloud forests (=upper montane forest =ceja de la
montaña) only represent 0.26% of the world’s land
surface, but are a high conservation priority: If the
Amazon warms by 2-4°C then will the cloud forest
biome go extinct or migrate to higher elevations?
Worldwide Cloud Forest distribution
Most of my work concerns a transect of forest census plots in the
Kosñipata Valley in SE Peru (Wayqecha - San Pedro - Tono).
I am also using a ‘pseudo-transect’ Tambopata - Manaus - Caxiuanã to
compare with the lowland Amazon.
We are using the RAINFOR network coordinated by University of Leeds
Kosñipata Ridge
200m
2000m
1250m
1000m
2250m
1750m
1500m
2500m
2750m
3450m
3250m
3000m
Kosñipata Ridge: a natural ecosystem laboratory
26.4oC
15.3oC
21oC
22.1oC
16.7oC
19oC
14.8oC
13.3oC
11.9oC
8.5oC
9.2oC
10.5oC
1 ha plots
13°3’ 52.55’’ S, 71°39’ 23.11’’ W
Fire
PUNA
Plot 1 (3391m)
Plot 2 (3205m)
Plot 3 (3033m)
Plot 4 (2745m)
Plot 5 (2540m)
Plot 6 (2290m)
Plot 7 (2038m)
Plot 2 (3205m)
Plot 8 (1855m)
Plot 1 (3391m)
Plot 3 (3033m)
Plot 10 (194m)
Plot 4 (2745m) Plot 5 (2540m)
Plot 6 (2290m)
PUNA
Plot 7 (2038m)
Plot 9 (210m)
LOWLAND RAINFORESTS
Plot 7 (1855m)
CLOUD MONTANE RAINFORESTS
CLOUD FORESTS
LOW MONTANE FORESTS
SUBMONTANE FORESTS
N
0
1km
13°10’ 22.70’’ S, 71°31’ 19.02’’ W
Kosñipata Valley, SE Peru
Source: Dr. Roman-Cuesta (2006)
Forest-grasslands tree line, 3600 m elevation
Upper montane cloud forest, 3000 m elevation
Upper montane cloud forest, 2750 m elevation
Submontane forest, 1500 m elevation
Presumably, half a
mountainside &
several tree snags
are busy heading
along the Amazon
from San Pedro to
the Atlantic ...
The Carbon Cycle of a Forest
Source: Dr. Cécile Girardin
Above-ground biomass growth
Canopy production
0.1 m m
nylon mesh
0.5 m
0.5 m
~1m
Root
production
Manaus lowland forest: 75 m asl
GPP = 30.4
R Total = 29.3±4.7
NPPTotal = 10.1±1.4
R Aut = 19.8±4.6
NPPAg = 7.3±1.3
R Het = 9.6±1.2
NPPBg = 2.8±0.7
NPP VOC = 0.13±0.06
NPP leaves,flowers,fruit = 3.6±0.7
R leaf = 10.0±4.0
NPP branch turnover = 1.0±1.0
R stem = 4.2±1.0
DFine litterfall = 3.6±0.7
NPP stem = 2.58±0.06
Predicted R soil = 12.6±2.3
Measured R soil = 12.1±1.7
R CWD = 2.5±0.3
DCWD= 3.58±1.0
R roots = 5.6±2.0
NPP coarse roots = 0.8±0.2
R soil het. = 7.1±1.1
DRoot = 2.8±0.7
NPP fine roots = 2.1±0.7
ALL UNITS ARE
t C / ha per year
Fdoc = 0.19±0.07
Wayqecha cloud forest: 3025 m asl
GPP estimated =16.2±2.5
GPP(canopy physiology) = 14
R Total = 15.7±2.0
NPPTotal = 5.2±0.7
R Aut = 9.8±1.7
NPPAg = 3.6±0.5
R Het = 5.9±1.0
NPPBg = 1.6±0.4
NPP leaves,flowers,fruit =1.9±0.1
R leaf = 6.0±1.2
NPP branch turnover = 0.5±0.5
R stem = 1.1±0.2
Predicted R soil = 9.2±
±1.9
Measured R soil = 11.6±0.1
DFine litterfall =1.9±0.1
NPP stem = 1.2±0.1
R CWD = 0.6±0.2
DCWD= 2.4±0.6
R roots = 3.9±1.6
NPP coarse roots = 0.4±0.1
R soil het.=5.3±1.1
ALL UNITS ARE
t C / ha per year
NPP fine roots = 1.2±0.4
DRoot = 1.6±0.4
JULES v2.1.1 simulations with
can_rad_mod=4:
Solid line = 24 hr mean
Broken lines = daytime &
nighttime means
Dotted lines = 24 hr min & max
Points = Field measurements
NPP:
ON THE Y-AXIS
30 t C / ha per
year is approx.
7.9 µmol C/m2
per sec flux.
MAN
TON
SP
15
TAM
CAX
20
WAY
Net Primary Productivity (NPP) in µmol CO2/m2s
25
10
5
Grey band is an estimate
from Clark et al. (2001, Ecol
Appl)
0
-5
15
20
25
Mean annual temperature in ºC
Girardin et al. (in press GCB). Net primary productivity and its allocation along a tropical forest elevational
transect in the Peruvian Andes.
Marthews et al. (in prep). Carbon fluxes in six Amazonian and Andean forests: ecosystem productivity and
carbon use efficiency.
JULES v2.1.1 simulations with
can_rad_mod=4:
Solid line = 24 hr mean
Broken lines = daytime &
nighttime means
Dotted lines = 24 hr min & max
Points = Field measurements
ON THE Y-AXIS
30 t C / ha per
year is approx.
7.9 µmol C/m2
per sec flux.
20
MAN
SP
TON
TAM
CAX
30
WAY
Gross Primary Productivity (GPP) in µmol CO2/m2s
GPP:
Grey band is an estimate
from Clark et al. (2001, Ecol
Appl) and Piao et al. (2010,
Ecology).
10
0
15
20
25
Mean annual temperature in ºC
Girardin et al. (in press GCB). Net primary productivity and its allocation along a tropical forest elevational
transect in the Peruvian Andes.
Marthews et al. (in prep). Carbon fluxes in six Amazonian and Andean forests: ecosystem productivity and
carbon use efficiency.
JULES v2.1.1 simulations with
can_rad_mod=4:
Solid line = 24 hr mean
Broken lines = daytime &
nighttime means
Dotted lines = 24 hr min & max
Points = Field measurements
Carbon use efficiency (CUE)
CUE = NPP/GPP
(undefined at night0.8
when GPP=0):
0.6
0.4
MAN
TAM
CAX
TON
SP
WAY
0.2
Grey band is an estimate
from Piao et al. (2010,
Ecology).
0.0
15
20
25
Mean annual temperature in ºC
Girardin et al. (in press GCB). Net primary productivity and its allocation along a tropical forest elevational
transect in the Peruvian Andes.
Marthews et al. (in prep). Carbon fluxes in six Amazonian and Andean forests: ecosystem productivity and
carbon use efficiency.
The Future: Scaling up to Regional Simulations
(last slide!)
THANKS FOR LISTENING