4.4.4.1. From Forests to Slash-and-Burn to Agroforests after Deforestation
This major land conversion practice takes place mainly at the margins of humid
tropical forests. Transformation of the original forest into various types of
agroforests results in a smaller decrease in carbon stocks than transformation
of forests into cropland, pastures, or degraded grasslands (Palm et al.,
2000).
Much of the uncertainty in the values of CO2 fluxes from the tropics is the
result of inadequate estimates of the biomass that is cleared, the fate of the
carbon lost, the type of biomass and time course of subsequent land-use systems,
and the regrowth rates of vegetation (Houghton, 1997). A project that used standardized
methods to compare several such systems in Brazil, Cameroon, Indonesia, and
Peru provided data on the foregoing parameters and the carbon sequestration
potential of many land-use systems at the margins of the humid tropics (Woomer
et al., 1999; Palm et al., 2000; Sanchez, 2000). The time course
of the land-use changes is described in the following subsections. An example
of one specific practice, complex agroforests, is described in Fact
Sheet 4.10.
4.4.4.1.1. Rates
Carbon sequestration rates are highly negative on forest clearance: -92 t C
ha-1 yr-1 during the first 2 years after slash-and-burn (Table
4-7)-a period that is normally under annual cropping or pasture establishment
(Neill et al., 1997). Table 4-7 shows that carbon
sequestration rates become positive with secondary forest fallows (5-9 t C ha-1
yr-1); complex agroforests (2-4 t C ha-1 yr-1); and simple agroforests with
one dominant species such as oil palm, rubber, or Albizia falcataria (7-9 t
C ha-1 yr-1). The lower carbon sequestration rate of some agroforestry systems
in relation to natural secondary succession found by Palm et al. (2000) is partly
because agroforestry products are removed from the system for family use or
for sale. This finding underscores the important tradeoffs between a global
public good (carbon) and a private good (economic gain) (Tomich et al.,
1998). Croplands, pastures, and degraded grasslands lost carbon at a slow rate
or show modest positive rates (-0.4 to +3 t C ha-1 yr-1). Land-use systems that
include trees, therefore, produce higher carbon sequestration rates than those
that are limited to annual crops, pastures, or grasslands (Palm et al.,
2000).
|
Table 4-7: Carbon uptake rates and time-averaged
system carbon stocks and differences in carbon stocks from land transformation
at margins of humid tropics. Summary of 116 sites with different land uses before
and after slash-and-burn located in Pedro Peixoto (Acre) and Theobroma (Rondônia),
Brazil; Ebolowa, M'Balmayo, and Yaounde, Cameroon; Jambi and Lampung, Sumatra,
Indonesia; and Yurimaguas and Pucallpa, Peru.a
|
|
| Land-Use Practice |
Carbon Uptake Rates
(t C ha-1 yr-1)
|
Duration
(yr)
|
Carbon Stocks
(time-averaged)
(t C ha-1)
|
Differences in Modal
Carbon Stocks
(time-averaged)
(t C ha-1)
|
|
Low
|
Modal
|
High
|
Low
|
Modal
|
High
|
Forest
|
Pasture/Grasslands
|
|
| Primary and logged forest |
n/ab
|
n/ab
|
n/ab
|
?
|
192
|
230
|
276
|
-
|
-201
|
| Cropping after slash-and-burn |
-76
|
-92
|
-112
|
2
|
39
|
46
|
52
|
-184
|
+17
|
| Crops/bush fallow |
2
|
3
|
4
|
4
|
32
|
34
|
36
|
-196
|
+5
|
| Tall secondary forest fallow |
5
|
7
|
9
|
23
|
95
|
112
|
142
|
-118
|
+83
|
| Complex agroforest |
2
|
3
|
4
|
25-40
|
65
|
85
|
118
|
-145
|
+56
|
| Simple agroforest |
5
|
7
|
9
|
15
|
65
|
74
|
92
|
-156
|
+61
|
| Pasture, Imperata grassland |
-0.2
|
-0.2
|
-0.6
|
4-12
|
27
|
29
|
31
|
-201
|
-
|
|
a Sanchez (2000). Calculated from data of Woomer et al. (1999) and
Palm et al. (2000), assuming the following time-averaged soil carbon
stocks (in Mg C ha-1): 40 for primary/logged forest and crops after slash-and-burn,
35 for tall secondary forest fallow and complex agroforest, 30 for bush fallow
and simple agroforest, and 25 for pasture and imperata grassland.
b Not available; likely close to zero.
|
|
|