Content uploaded by Gilberto Tiepolo
Author content
All content in this area was uploaded by Gilberto Tiepolo on Jun 20, 2016
Content may be subject to copyright.
Measuring and Monitoring Carbon Stocks at the
Guaraqueçaba Climate Action Project, Paraná, Brazil
Nov, 2002 Extension Serie No. 153 (p 98-115) Taiwan Forestry Research Institute.
International Symposium on Forest Carbon Sequestration and Monitoring.
Gilberto Tiepolo1, Miguel Calmon2, André Rocha Feretti1
Abstract
Any climate action project that seeks to obtain recognition, and
potentially carbon benefits, for the GHG emissions reductions it achieves must
accurately calculate those reductions over time using scientifically rigorous
methods that will stand up to external review. Carbon inventory and
monitoring plans are designed to quantify the changes in key carbon (C) pools
in and around the project area and to project local land-use changes by
monitoring patterns of land-use in proxy regions, trend modeling, and analyzing
socio-economic and other data. Data from the inventory and monitoring
activities are used to calculate the difference between the with- and without-
project scenarios.
The Guaraqueçaba Climate Action Project is being implemented by
SPVS (Sociedade de Pesquisa em Vida Selvagem e Educação Ambiental), in
partnership with TNC (The Nature Conservancy) and AEP (American Electric
Power). It has an area of approximately 7,000 hectares and is located in
Paraná State, Brazil, within the Environmental Protection Area of
Guaraqueçaba in the Atlantic Rain Forest. This ecosystem is recognized by the
United Nations Economic and Social Organization (UNESCO) as one of the
planet’s highest priorities for conservation and has designated it World
Biosphere Reserve.
The main goals of the project are biodiversity conservation, restoration of
degraded pasture, sustainable development of local communities, and
generation of carbon offsets that are real, measurable, and verifiable.
The requirements for a good monitoring program should include the
following: (i) use the most appropriate and cost-effective methodology for the
region; (ii) follow quality assurance/control plan (QA/QC) and standard
operating procedures (SOPs); (iii) train local NGOs and field personnel on SOPs;
(iv) elaboration of a vegetation map and stratification of the project area; (v)
1 Sociedade de Pesquisa em Vida Selvagem e Educação Ambiental – SPVS
Rua Gutemberg, 296, Batel. 80420-030 Curitiba - PR, Brazil. http://www.spvs.org.br, E-Mail:
carbono@spvs.org.br
2 The Nature Conservancy – TNC
Alameda Júlia da Costa, 1240, Bigorrilho. 80730-070 Curitiba – PR, Brazil. http://www.nature.org –
www.tnc.org.br, E-Mail: mcalmon@tnc.org.br
post-monitoring independent verification; (vi) record-keeping and data entry,
analysis, interpretation, maintenance, and archiving. The Winrock international
methodology (MacDicken, 1997), adapted to the local conditions, was
selected to measure and monitor carbon at the project. A QA/QC and SOP
documents were developed for the project to guarantee that carbon
measurements done during the lifetime of the project are consistent and
accurate. Researchers and technicians from SPVS were trained on the SOP
before the beginning of the carbon inventory and monitoring field work. A
total of 12 strata were identified, but only 6 forest strata were under threat of
deforestation and therefore used to estimate carbon stocks and offsets. A total
of 188 permanent plots were established with 68 in the submontane forest
(1162,5 ha), 11 in the lowland forest (427 ha), 10 in the floodplain forest (173 ha),
63 in advanced/medium forest (1,783 ha), 24 in medium secondary forest (545
ha), and 12 in young secondary forest (279 ha). Twenty-eight clip plots were
established on the pasture (409 ha) and shrublands (279 ha).
The preliminary average carbon stock (aboveground woody biomass)
estimated for the 6 forest strata were the following: submontane forest: 135.9 t C
ha-1; lowland forest: 106.8 t C ha-1; floodplain forest: 64.12 t C ha-1;
advanced/medium forest: 106.1 t C ha-1; medium secondary forest: 101.96 t C
ha-1; young secondary forest: 42.89 t C ha-1. The above ground carbon for the
pasture strata was 2.4 t C ha-1 and for the shrublands 7.4 C ha-1. The general
wet biomass equation that is currently being used to estimate the carbon stock
is being verified and adjusted from the destructive sampling effort that is being
conducted by SPVS. A new biomass equation for tree fern was also
developed, which showed a strong correlation between biomass and height.
The results of this effort will help to improve and develop models to
measure and monitor carbon stock in very complex and heterogeneous
landscapes, such as the ones found in the Atlantic Forest Biome, and to
promote projects that are designed to generate multiple benefits such as
biodiversity, soil and water conservation, restoration of degraded lands, and
sustainable development of local communities.
Introduction
The Guaraqueçaba Climate Action Project is an innovative effort to
combine reforestation and forest stewardship strategies to help manage levels
of carbon dioxide in the atmosphere. Over a period of forty years (the project
term), the project will restore and protect approximately 7,000 hectares (17,000
acres) of partially degraded and/or deforested tropical forest within the
Guaraqueçaba Environmental Protection Area (EPA) of Paraná State, in
southern Brazil. The land is titled to SPVS, which assumed responsibility for its
long-term protection and stewardship, and will be registered as a private
reserve (Serra do Itaqui Natural Reserve). By protecting and restoring
threatened tracts of Atlantic Forest, the project will conserve biodiversity while
contributing to the mitigation of global climate change.
The project – a collaborative effort between Central and South West
Corporation, a Texas-based electric utility (now American Electric Power), The
Nature Conservancy (TNC), a US-based conservation organization, and
Sociedade de Pesquisa em Vida Selvagem e Educação Ambiental (SPVS), a
Brazilian conservation organization. The project is promoting assisted natural
forest regeneration and regrowth on pastures and degraded forests on the
acquired lands. It is also protecting standing forest that still exists within the
project area, but It is under threat of deforestation.
The main goals of the project are to protect and restore the ecological
health and biodiversity of the area in the project site and establish models for
the adequate use of resources in the Guaraqueçaba EPA, and generate
carbon offsets that are real, measurable, and verifiable. Through a rigorous
monitoring and verification program, the carbon benefits generated by the
project will be quantified and validated in such a way as to maximize the
possibility that they will be accepted under any future international carbon
trading regime(s) and to serve as a scientifically-based pilot project in
ecosystem restoration.
In addition to these primary objectives, the project also seeks to improve
local environmental quality, support sustainable economic development by
creating opportunities for local people, and promote environmental awareness
of the Guaraqueçaba region.
While the project is designed as a stand-alone effort, it will benefit from a
variety of existing programs and activities currently being undertaken by the
project partners, including other two climate action projects (The Antonina Pilot
Reforestation Project, a collaborative effort between Chevron-Texaco, TNC
and SPVS; and The Atlantic Rainforest Restoration Project, a joint effort between
General Motors, TNC and SPVS). This project leveraging will enable it to have a
broad impact and to contribute to a regional strategy for protecting the
Guaraqueçaba EPA.
Materials and methods
Project Location
The present study is based on the carbon inventory and monitoring
activities of the Guaraqueçaba Climate Action Project, located in the Atlantic
Forest biome in Paraná State between latitudes 25o 26’ and 25o 21’ South, on
the coastal plain. Lying about 45 kilometers from the seat of the municipality
and 140 Kilometers from Curitiba, the capital of the state. Its geographic
location is strategic for conservation purposes, as it connects 500 meters
mountains and 1,200 hectares of mangroves, borders Guaraqueçaba Bay to
the South, PR-405 highway to the North, Itaqui hills to the West and Tagaçaba
River to the East. It is a region of great environmental fragility composed of
freshwater and marine environments, in close proximity to Guaraqueçaba Bay
(Figure 1).
The Guaraqueçaba EPA is in the middle of the largest continuous piece
of Atlantic forest remaining today. Brazil’s Atlantic Forest is an internationally
recognized world biosphere reserve and home to one of the planet’s most
diverse and endangered ecosystems. Today, only seven percent of the original
vegetation cover remains, making the Atlantic Forest one of the most
threatened tropical forest in the world.
In the project site (Serra do Itaqui Natural Reserve), as well as in most of
the region, the original vegetation has been submitted to intense exploitation
that has stripped it of its original characteristics, particularly in the plains. Easy
access encouraged timber exploitation of mature forests, which implied either
complete removal or drastic changes to their structure and composition. Such
conditions resulted in a mosaic of secondary forests with varied floristic,
structural and physiognomic characteristics, where initial formations are more
prevalent in flat areas and intermediate stages on the hillsides.
The changes imposed on the vegetation, particularly in areas that
underwent total removal of the original cover, have altered original drainage
processes and caused severe soil degradation. Such conditions will require
considerable efforts to fully restore the biogeochemical processes needed to
support the original ecological systems.
Figure 1 Location of the Project (1- The Antonina Pilot Reforestation Project; 2 – The
Atlantic Rainforest Restoration Project; 3 – Guaraqueçaba Climate Action
Project.
Vegetation Map
The vegetation map of the project area was based on color aerial
photography (scale 1:30,000) and Ikonos satellite imagery. An orthophoto was
also used as the base map. After photointerpretation and field checking
several type of forests and other land uses were identified and classified. The
level of human intervention and successional stages for the different forest
types were also classified during this effort. A total of 12 strata, based on the
classification scheme developed by IBGE (1992), were identified within the
project area (Figure 2).
Figure 2 Vegetation Map of Guaraqueçaba Climate Action Project with the
localization of permanent plots.
Stratification
For the carbon inventory it was used a stratified sampling, which helped
to make the estimates more precise and cost-effective. From the 12
vegetation classes, 6 forest classes (Submontane forest, Lowland forest,
Floodplain forest, advanced / medium forest, medium secondary forest, and
Young secondary forest) were assumed to be under threat and therefore were
used during the carbon inventory work to estimate the carbon stock and
benefits to be generated at the project area.
In addition to those forest strata, other non-forest classes such as pasture,
herbaceous vegetation and shrus were also included as part of the carbon
inventory effort, but temporary plots were used on those strata.
Carbon Inventory
The methodology used for the carbon inventory was the one developed
by Winrock International (MacDicken, 1997) and adapted to the project
conditions. A Standard Operating Procedure (SOP) was developed for the
project and SPVS personnel was trained on every step (Brown and Delaney,
2000 A). Some of the main activities described on the SOP are listed below.
Before de installation of the permanent plots, 6 preliminary plots were installed
on each strata to determine the number of plots necessary to represent the
carbon stock within each stratum. The number of plots was calculated by using
a software (plot calculator) developed by Winrock International, which takes
into account the maximum allowable error, desired level of precision, total area
of the project and strata, variance, and approximate cost per plot. After that a
Carbon Inventory and Monitoring Plan was developed for the project (Brown et
al., 1999) in order to start the plot installation in 2000.
Establishment of Transects
Transects are lines cut through the forest or pastures to allow access to
permanent or clip plot locations. The selection of the places where the
transects were opened was based on the vegetation map in areas that most
represented each stratum composition and on the availability of existing trails
to facilitate the access and reduce suppression of vegetation within the project
area. In general the length of the 35 transects vary from 200 to 700 meters.
Establishment of Plots
A total of 188 nested circular plots were installed on the different forest
strata (Table 1). Four nested plots were used to measure aboveground biomass.
The 1-m radius plot was used to measure saplings with dbh <5 cm; the 4-m
radius (0.005 ha) was used for trees between 5-19.9 cm dbh; the 14-m radius
(0.06 ha) was used for trees between 20-69.9 cm dbh; and the 20-m radius was
used for trees with ≥ 70 cm dbh (Figure 3). The actual size of each plots was
adjusted depending on the percentage slope of the plot. At the center of the
plot a PVC pipe was installed and painted with bright color and plastic tapes.
A GPS was used to collect the coordinates of the center.
All trees were measured at dbh (1.3 m above ground) unless buttressed
or with defects at that height. Aluminum numbered tags were placed and
nailed on each tree measured within the four plots. More details can be found
on Brown and Delaney (2000 A).
Table 1 Description of the each stratum, area, and number of plots established.
Strata code Vegetation type Area (ha) Number of sample
plots established
DM submontane forest 1162,55 68
LL lowland forest 427,3 11
FP floodplain forest 172,9 10
M advanced/medium forest 1782,9 63
Y medium secondary forest 544,92 24
VY Young secondary forest 278,58 12
P Pure pasture 386 12
PS Pasture/shrub 30 10
S Shrubs 297 6
TOTALS 4,694 216
Small Plot-S
Code all trees S, M,
or L
Medium Plot-M
Code all trees M
or L
Large Plot-L
Code all trees L
4 m
14 m
20 m
Figure 3 Nested plot layout. Small plot radius is 4m; medium plot radius is 14 m and
large plot radius is 20 m; the 1.0-m radius plot for saplings is not shown.
Measurement of Biomass
Two main carbon pools were measured during the carbon inventory: a)
live biomass which included live trees, understory, and roots and b)
aboveground dead wood which included litter and standing and lying dead
trees;
Live trees
The criteria for including trees within each plot was based on the
diametric density of the different forest strata. A diameter class range was used
for each different plot size (see above), with the exception of trees with less
than 5 cm at dbh, lianas, heart-of-palm, and tree fern. In the last two cases the
height and dbh were measured.
For trees > 1.3 m high and less than 5 cm dbh, the number of trees within
the 1-m plot was counted and then multiplied by the average biomass that
was estimated by destructive sampling (10 plots per strata) on the opposite
side of the permanent plot. The average of sampling biomass for each strata
are:
submontane forest - 0,2 kg
lowland forest – 0,2 kg
floodplain forest – 0,2 kg
advanced/medium forest – 0,3 kg
medium secondary forest – 0,6 kg
young secondary forest – 1,6 kg
The young secondary forest got the highest coefficient because in this
formation the understory is less shaded than formations more developed,
allowing good condition to the establishment of natural regeneration.
The general wet biomass equation was used to estimate aboveground
biomass carbon for trees because it matches most closely to the climatic
conditions (rainfall amount and distribution; and near the southern extreme of
the tropical belt) in the project area (Brown, 1997, 2001). However, the
suitability of this equation is being verified and adjusted by the destructive
sampling program that was initiated in 2001. The goal is to cut 15-20 large trees
(dbh > 50 cm) and adjust the equation according to the results. At the same
time other smaller trees will be cut and included in the adjustment. Other
equations were also used to estimate the biomass of mature Cecropia, palms,
and lianas (Table 2), which were developed from work in Noel Kempff Climate
Action Project in Bolivia (Brown et al., 2000).
A new biomass equation for tree fern (Cyathea spp.) was also
developed for the project by destructively sampling 22 trees and developing a
regression equation between tree biomass and height and tree biomass and
dbh.
To characterize the understory, within each plot were established 4 Clip
plots (aluminum sample frames – 60 cm in diameter). Species with dbh < 5 cm
and less than 1,3 height , were cut, weighed and collected sub-sample for
determination of humidity percentage.
Table 2 Regression equations used for estimating biomass carbon (Y) in the 2000
analysis of plots in the GCAP area.
Equation Species R2
Dbh and
height
Range
Y=21.297-6.953(dbh)+0.74(dbh^2) General 0.91 4-116 cm
Y=0.3999+7.907*height Palms 0.75 1-33 m
Y =(-.48367+1.13488*(Sqr(dbh))*Log(dbh))^2 Cecropia 0.62 1-11 m
Y=563.56*(dbh)^2.6277 Lianas 0.89 0.3-2.5 cm
Y=-4266348/1-2792284e-0.313677 Fern tree 0.88 1 – 8 m
Roots
A recent review of the literature on root biomass for the world’s forests
including 39 studies for the tropics suggests that root:shoot ratios vary from 0.1 to
0.38 with root biomass ranging from 1 to over 130 t ha-1 (Cairns et al. 1997). We
assumed a root:shoot ratio of 0.20 that represents the lower 95% CI for tropical
forests and it is more conservative (Brown and Delaney, 2000 B)
Pasture
For sampling pasture biomass, it was used clip plots that are placed on
the ground at regular intervals along transects. The procedures for this were the
same way used for understory estimation.
Dead aboveground biomass
For litter it was used the same methodology that understory. Standing
dead trees were measured according to the same criteria as live trees; i.e.
classified either in the small, medium, or large nested plot. If the standing dead
tree contained branches and twigs and resembles a live tree (except for
leaves) the dbh was measured and its biomass was estimated using the
appropriate biomass regression equation as for live trees. If there were
branches, but no twigs remaining on the standing dead tree, the proportion of
the biomass was subtracted from the total for the tree. If the top of the standing
dead tree was missing, the height of the remaining stem was measured with a
clinometer and the top diameter was estimated – this can be done by
estimating the ratio of the top diameter to the basal diameter.
Lying dead wood were measured using the line intersect method
outlined in Harmon and Sexton (1986). Fallen coarse dead wood were defined
as all woody material on the ground with a diameter >10 cm. Several discs of
dead wood in each of the three density classes were collected and their
volume and dry mass were determined. The estimated densities are given in
Table 3.
Table 3 Density of wood disc samples used in dead wood calculations.
Type/Decomposition class of wood sample Density (t/m3)
General/Sound 0.47
General/Intermediate 0.34
General/Rotten 0.17
Palm/Intermediate 0.14
Palm/Rotten 0.09
(Source: Brown and Delaney, 2000 B)
Results and discussion
Biomass Regressions Tree Ferns
For the Cyathea spp., height is more strongly correlated to biomass (R2 =
0.88) than is DBH (R2 = 0.1). The equation for height provides an acceptably
robust model for estimating carbon storage without destructively harvesting
more individuals (Figure 4, 5 6. ). The biomass result can be converted to carbon
by multiplying the biomass by the carbon concentration found in the species
(about 0.5).
Figure 4 Relationship between height and biomass for fern trees.
Figure 5 Relationship between dbh and biomass for fern trees.
Figure 6 Allometric regression equations for fern trees.
Carbon in Forest
The total carbon in the forest strata (excluding soil) was 471547,89 t C
with a 95% confidence interval of ± 6,7% of the mean (Table 4). As expected,
the highest amount of carbon was in the submontane forest stratum (135,89 t
C/ha) that represents the oldest forest in the project site. It is an altered primary
forest that is located in the slopes of hills in continental soils, usually more deep.
The lowest amount occurs in the very young secondary forest stratum (42,89 t
C/ha), that is characterized by small trees (5 m height) with sparse crown.
Coefficients of variation for total carbon content by strata were relatively low
(29 -51 %), particularly for the advanced /medium stratum.
Table 4 Total, mean, and statistical measures for the carbon content (excluding soil)
of the four forest strata of the GCAP. See Table 6 for carbon content in
different components of the forests.
Strata submontane
forest
lowland
forest
floodplain
forest
advanced/
medium
forest
medium
secondary
forest
young
secondary
forest
n= 68 11 10 63 24 12
Area (ha) 1162,55 427,3 172,9 1782,9 544,92 278,58
Mean(t C/ha) 135,89 106,81 64,12 106,19 101,96 42,89
Min 61,1 27,5 18,5 44,8 38,3 9,8
Max 373,1 211,5 95,8 189 189,4 57
Variance 2314,1 2940,6 702,9 951 1565,7 267,6
Standard
Deviation 48,1 54,2 26,5 30,8 39,6 16,4
Standard Error 5,8 16,4 8,4 3,9 8,1 4,7
C.V. (%) 35,4 50,8 41,3 29 38,8 38,1
Mean(t C/ha) 114,36 ± 7,7
Total (tons C) 471547,89 ± 31593,64
CI % (+/-)
(tons C) 6,7
CV = coefficient of variation; CI = 95% confidence internal 21,954,21
Although the total carbon had relatively low variation, individual
components were more variable (Table 5). The most variable component was
standing dead biomass, with coefficients of variation of 128-287% (Appendix 1).
Lying dead wood was also variable. On a plot by plot basis, there is generally
some relationship between live and dead biomass, so that when combined the
overall variation decreases.
The overall weighted mean of the total carbon content of forests is 108 t
C/ha, 74% of which is in the live aboveground woody biomass (Table 5). Dead
wood carbon represented about 5% and litter and understory combined
represented about 3,6% of the total carbon stock. For the very young forests,
the litter and understory represented 19% of the aboveground biomass, and
thus is a more significant component.
Table 5 Mean carbon content by forest component and by forest strata for the 2000
inventory in the GCAP (details of the carbon content of each component
are given in Appendix 1 to 7.
Area
Above-
ground
biomass
Below-
ground
biomass
Standing
dead
biomass
Lying
dead
biomass
Trees
biomass
< 5 cm
dbh
Understory
vegetation Litter Total
Strata (ha) T C ha-1 T C ha-1 T C ha-1 T C ha-1 T C ha-1 T C ha-1 T C ha-1 T C ha-1
submontane 1162,55 109,30 21,86 2,86 1,43 0,44 nm nm 135,89
lowland 427,30 83,69 16,74 1,84 0,92 0,09 1,83 1,70 106,81
floodplain 172,94 43,96 8,79 5,48 2,74 0,32 0,52 2,32 64,12
Advanced
medium 1782,90 76,42 15,28 4,76 2,38 2,43 0,52 4,40 106,19
medium
secondary 544,92 70,83 14,17 4,41 2,20 3,62 1,19 5,54 101,96
young
secondary 278,58 21,24 4,25 1,25 0,62 5,94 0,91 8,67 42,89
Total 4369,20
DESPAD 40,1 8,0 5,0 2,5 3,4 8,0 2,7 46,9
Weighted
mean 80,4 16,1 3,7 1,9 20,6 3,3 107,9
CI 5,7 1,1 0,7 0,4 0,5 1,4 0,5 8,4
%74,5 14,9 3,5 1,7 1,8 0,6 3,1
*CI is the 95% confidence interval expressed as a percent of the mean
‡ nm= not measured
Carbon in pastures
The mean aboveground carbon content of pastures ranged from 0.7 to
3.5 t C/ha, with the highest carbon content found in pastures dominated by
shrub vegetation and the lowest in pure pastures (Table 6). The results were
variable with 95% confidence interval of 25.6% of the mean. Pasture grasses
generally have as much biomass below ground as above, thus the total range
of biomass carbon for pure pastures (7P) and pastures with shrubs (7PS) ranges
from 1.4 to 2.4 t C/ha. For the shrub formation (7S), belowground biomass is
about 30% of aboveground. Thus the shrub formation has a range of total
biomass carbon of 3.0 to 7.4 t C/ha. The maximum values were used in the
carbon benefit calculations as described above.
Table 6 Statistics for aboveground carbon in the pasture strata of the GCAP.
Strata Pasture
(7P)
Pasture/Shrubs
(7PS)
Shrubs
(7S)
n= 12 10 6
Area (ha) 386 30.4 296.7
Mean (t C/ha) 0.7 0.8 3.5
Min 0.2 0.4 2.3
Max 1.1 1.2 5.7
Variance 0.1 0.1 1.9
Standard Deviation 0.3 0.3 1.4
Standard Error 0.1 0.1 0.6
C.V. (%) 50.8 37.5 39.9
Mean (t C/ha) 1.8 ± 0.5
Total (tons C) 1305.7 ± 334.8
CI % (+/-) (tons C) 25.6
CV= coefficient of variation; CI= 95% confidence interval
Source preliminary carbo-offset report 2000
Conclusions
With the Carbon inventory conducted in the Guaraqueçaba Climate
Action Project it was possible to quantify the amount of carbon stored with a
reasonable level of precision. This inventory was used to estimate the
differences between the with- and without-project carbon pools and is the
primary basis for determination of project GHG benefits. Through ongoing
carbon inventory work, several aspects of the carbon inventories that could be
improved or significantly strengthened were identified.
In the Allometric regression equations for fern trees it was observed that
the relationship between height and biomass showed a strong correlation,
where the dbh x biomass showed a poor correlation. Therefore, we
recommend the use of the relationship height x biomass when estimating
aboveground biomass of ferns trees in the Atlantic Forest Biome.
The results of this effort will help to improve and develop models to
measure and monitor carbon stock in very complex and heterogeneous
landscapes, such as the ones found in the Atlantic Forest Biome, and to
promote projects that are designed to generate multiple benefits such as
biodiversity, soil and water conservation, restoration of degraded lands, and
sustainable development of local communities.
Also brought a lot of challenges during the implementation of the
carbon inventory. Many lessons were learned during this phase, but much more
are still to be learned in order to improve and adjust the methodology for future
measurements.
Acknowledgments
We thank Dr. Sandra Brown and Matt Delaney for making Winrock
International’s methodology available to the Guaraqueçaba Climate Action
Project and making it a very important component of the project. Their
participation during the planning and implementation phases of the carbon
inventory and their willingness to train and transfer the methodology to SPVS’s
team were very important for the success of the carbon inventory program.
We also would like to thank Bill Stanley for their valuable work and input as a
Forester during all phases of the carbon inventory work. This work could not be
done without the help and assistance of Flavio , Guilherme, Elielson, Arildo, Luis,
Nelson, Jair, Walkiria, that spent several weeks in the field, and also to Alia
Ghandour Warwick Manfrinato, Patricia Garffer, Joe Keenan , Alexandra
Andrade, Clóvis Borges, Franco Amato, Ricardo, Laércio, and many others that
could share with us the beauties of Atlantic forest.
References
Harmon, M. E. and J. Sexton. 1996. Guidelines for Measurements of Woody
Detritus in Forest Ecosystems. US LTER Publication No. 20. US LTER Network
Office, University of Washington, Seattle, WA, USA.
MacDicken, K. 1997. A Guide to Monitoring Carbon Storage in Forestry and
Agroforestry Projects. Winrock International, 1611 N. Kent St., Suite 600,
Arlington, VA 22209, USA.
Brown, S., M. Calmon, and M. Delaney. 1999. Carbon Inventory and Monitoring
Plan for the Guaraqueçaba Climate Action Project, Brazil. . Winrock
International, Arlington, VA.
Brown, S. and M. Delaney. 2000 A. Standard Operating Procedures for the
Guaraqueçaba Climate Action Project, GCAP-SOP, Version: 2.00. Winrock
International, Arlington, VA.
Brown, S. and M. Delaney. 2000 B. Preliminary Carbon-Offset Report for the
Guaraqueçaba Climate Action Project. Winrock International, Arlington, VA.
Brown, S., Burnham, M., Delaney, M., Vacca, R., Powell, M. and Moreno, A.
2000. Issues and challenges for forest-based carbon-offset projects: a case
study of the Noel Kempff Climate Action Project in Bolivia. Mitigat. Adapt.
Strategies Global Change 5, 99-121.
Brown, S. and M. Delaney. 2001. Preliminary carbon-offset report for the
Guaraqueçaba Climate Action Project.
Brown, P.: 1998. Climate, Biodiversity, and Forests: Issues and Opportunities
Emerging from the Kyoto Protocol, Washington, D. C., World Resources
Institute, 35p.
Brown, S. 1997. Estimating Biomass and Biomass Change of Tropical Forests: a
Primer. FAO Forestry Paper 134, Rome, Italy.
Cairns, M. A., S. Brown, E. H. Helmer, and G. A. Baumgardner. 1997. Root
biomass allocation in the world’s upland forests. Oecologia 111:1-11.
Appendix 01: Stats for aboveground biomass
Strata Y M M/A SM LL FP
N12 24 63 68 11 10
Mean 21,24167 70,83333 76,42111 109,2975 83,68818 43,95556
Variance 92,47667 995,0112 638,4199 1512,872 1905,853 357,3587
Standard Desviation 9,616479 31,5438 25,26697 38,89566 43,65608 18,90393
Standard Error 2,776038 6,438851 3,183339 4,716791 13,1628 5,977948
CV % 45,27177 44,53242 33,06282 35,58696 52,16516 43,00692
Mean 85,85332 4,893639
Total 351193,4 20018,02
CI % 5,7
Appendix 02: Stats for belowground biomass
Strata Y M M/A SM LL FP
N12 24 63 68 11 10
Mean 4,248333 14,16667 15,28422 21,8595 16,73764 8,791111
Variance 3,699067 39,80045 25,5368 60,51489 76,23412 14,29435
Standard Desviation 1,923296 6,30876 5,053394 7,779132 8,731216 3,780786
Standard Error 0,555208 1,28777 0,636668 0,943358 2,632561 1,19559
CV % 45,27177 44,53242 33,06282 35,58696 52,16516 43,00692
Mean 17,17066 0,197004
Total 70238,68 805,8677
CI % 1,147328
Appendix 03: Stats for Standing dead
Strata Y M M/A SM LL FP
N12 24 63 68 11 10
Mean 1,249167 4,4075 4,76 2,859853 1,841818 5,475556
Variance 12,86866 31,7084 32,24936 13,67173 7,873636 59,649
Standard Desviation 3,587292 5,631021 5,678852 3,69753 2,806 7,723277
Standard Error 1,035562 1,149427 0,715468 0,448391 0,846041 2,442315
CV % 287,1748 127,76 119,3036 129,2909 152,3495 141,0501
Mean 3,983516 0,028352
Total 16295,05 115,9778
CI % 0,711736
Appendix 04: Stats for Lying dead
Strata Y M M/A SM LL FP
N12 24 63 68 11 10
Mean 0,624583 2,20375 2,38 1,429926 0,920909 2,737778
Variance 3,217166 7,927101 8,062341 3,417932 1,968409 14,91225
Standard Desviation 1,793646 2,815511 2,839426 1,848765 1,403 3,861638
Standard Error 0,517781 0,574714 0,357734 0,224196 0,42302 1,221157
CV % 287,1748 127,76 119,3036 129,2909 152,3495 141,0501
Mean 1,991758 0,007088
Total 8147,527 28,99445
CI % 0,355868
Appendix 05: Stats for one meter
Strata Y M M/A SM LL FP
n12 24 63 68 11 10
Mean 5,94179 3,620778 2,42522 0,440017 0,086812 0,31831
Variance 98,25101 9,11727 5,083811 0,81066 0,042371 0,177312
Standard Desviation 9,912165 3,019482 2,254731 0,900366 0,205842 0,421085
Standard Error 2,861396 0,616349 0,284069 0,109185 0,062064 0,133159
CV % 166,8212 83,39317 92,97014 204,6208 237,1123 132,2876
Mean 2,091599 0,010163
Total 8555,939 41,57481
CI % 0,485918
Appendix 06: Stats for understory
Strata Y M M/A SM LL FP
N12 24 63 011 10
Mean 0,910739 1,189191 0,521873 1,834093 0,523628
Variance 0,438179 5,049935 0,103712 7,172959 0,188951
Standard Desviation 0,661951 2,247206 0,322043 2,678238 0,434686
Standard Error 0,191089 0,458709 0,040574 0,807519 0,13746
CV % 72,68284 188,9693 61,70911 146,0252 83,0141
Mean 0,661624 0,009501
Total 2706,454 38,86659
CI % 1,43607
Appendix 07: Stats for litter
Strata Y M M/A SM LL FP
n12 24 63 011 10
Mean 8,674554 5,541915 4,40001 1,703824 2,321543
Variance 16,26518 4,651904 3,263792 1,612673 3,274465
Standard Desviation 4,033011 2,156827 1,806597 1,269911 1,809548
Standard Error 1,16423 0,440261 0,22761 0,382892 0,572229
CV % 46,49243 38,91845 41,05893 74,53296 77,94594
Mean 3,522895 0,017332
Total 14410,83 70,90053
CI % 0,491995
