Technical ReportPDF Available

Modelling forest carbon dynamics for REDD+ using the Generic Carbon Budget Model (GCBM) Pilot Project: Los Rios Region - Chile

Authors:
  • SilvaCarbon program
  • Natural Resources Canada, Victoria, BC, Canada

Abstract and Figures

This technical report describes the progress to date of the implementation of the Generic Carbon Budget Model (GCBM) in Los Rios Region, Chile for REDD+ MRV purposes, that was conducted with the collaboration of the National Forest Corporation of Chile (CONAF), the Canadian Forest Service and the Mullion Group. This report provides an overview of the steps conducted in a GCBM proof of concept in a pilot area, considering a subset of the REDD+ activities and carbon pools included in Chile’s FREL. This document also reflect on the next steps required for the full implementation of the GCBM model to account for all of the activities and regions eligible for REDD+, along with the required steps to improve GHG emissions estimation, reporting and analysis capabilities in Chile using open-source modelling tools, reflecting on the potential role a non-profit consortium, such as moja global, may have in supporting countries to achieve multiple objectives for sustainable land-use planning and policy development. This technical document was prepared as part of the deliverables of the “Expanding Market Opportunities” contribution agreement between Her Majesty the Queen in Right of Canada, represented by the Minister of Natural Resources and the non-for-profit organization “Climate Advisers Trust” (Project: "moja global - International collaboration on sustainable forest management tools”).
Content may be subject to copyright.
Modelling Forest Carbon for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
Technical document
Modelling forest carbon dynamics for REDD+
using the Generic Carbon Budget Model (GCBM)
Pilot Project
Los Rios Region - Chile
Progress to date
31 March 2020
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
This technical document was prepared as part of the deliverables of the “Expanding Market
Opportunities” contribution agreement between Her Majesty the Queen in Right of Canada,
represented by the Minister of Natural Resources and the non-for-profit organization “Climate
Advisers Trust” (Project: "moja global - International collaboration on sustainable forest
management tools”).
The views expressed in this report do not necessarily reflect the positions of the Government of
Chile or the Government of Canada.
Authors:
Julián Cabezas, Marcela Olguín, Werner A. Kurz, Max Fellows, Robert de Ligt and Daniel
Montaner
Suggested Citation:
Cabezas, J., Olguín, M., Kurz, W.A., Fellows, M., de Ligt, R., Montaner, D. (2020). Modelling
forest carbon dynamics for REDD+ using the Generic Carbon Budget Model (GCGB). Pilot
Project: Los Rios Region, Chile. Progress to date. Technical Document. 31p.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
Table of Contents
Executive Summary ............................................................................................................... 1
1 Introduction ................................................................................................................... 4
2 Methods and input data ................................................................................................. 7
2.1 Pilot area ............................................................................................................................7
2.2 General approach ................................................................................................................8
2.3 Modelling approach ............................................................................................................9
2.4 Sources of data ................................................................................................................. 10
2.5 Simulation runs ................................................................................................................. 11
2.5.1 Inventory data at the start of the simulation (Classifiers) ..................................................................... 12
2.5.2 Biomass dynamics .................................................................................................................................. 14
2.5.3 Disturbances .......................................................................................................................................... 17
3 Results and Discussion .................................................................................................. 19
3.1 Areas of Disturbance ......................................................................................................... 19
3.2 Results for the deforestation emissions ............................................................................. 20
3.3 Results for the substitution emissions ................................................................................ 22
3.4 Afforestation removals ...................................................................................................... 24
3.5 Spatially-Explicit Results .................................................................................................... 26
4 Lessons learned ............................................................................................................ 27
5 Next Steps .................................................................................................................... 29
6 Conclusions .................................................................................................................. 31
References .......................................................................................................................... 33
Modelling Forest Carbon for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
1
Executive Summary
REDD+ (Reducing emissions from deforestation and forest degradation, conservation of forest
carbon stocks, sustainable management of forests and enhancement of carbon stocks) is one of
the main mechanisms of the United Nations Framework Convention on Climate Change
(UNFCCC) for driving emission reductions and enhanced removals of greenhouse gases (GHG)
in the land sector (UNFCCC, 2016). The REDD+ approach incentivizes countries to voluntarily
report annual estimates of GHG emissions and removals from forests. For this purpose, countries
typically establish national MRV (measurement, reporting, and verification) systems (Birdsey et
al., 2013; GOFC-GOLD (2014).
One of the challenges in developing MRV systems is the integration of national and global data
to obtain GHG fluxes to fulfill the principles of transparency, accuracy, comparability, consistency
and completeness (TACCC) (IPCC 2003). To face these challenges, the moja global project, of
the Linux Foundation, is developing the Full Lands Integration Tool (FLINT) which is an open-
source modelling framework that integrates different data types and modules to calculate GHG
fluxes more efficiently for the land sector, while adhering to the TACCC principles (moja global,
2017). One of the implementations of the FLINT is the Generic Carbon Budget Model (GCBM),
developed by the Canadian Forest Service, and based on the same science modules used in the
Carbon Budget Model of the Canadian Forest Sector (CBM-CFS3) (Kurz et al., 2009) for
UNFCCC GHG inventory reporting and mitigation scenarios analysis.
Chile is one of the countries that already consigned their Forest Reference Emissions Levels
(FREL) (2001-2013), along with a Technical Annex on REDD+ results (2014-2016), to the
UNFCCC. This makes the country potentially eligible for results-based payments through funds
such as those established by the Green Climate Fund (GCF) and the Forest Carbon Partnership
Facility (FCPF). Although Chile has a relatively mature MRV system in place, the calculations are
being made using tabular data, so new tools are being tested to analyze and report spatially-
explicit results in an integrated framework, allowing the country to gradually implement more
complex forest carbon models. The GCBM is one of the tools evaluated.
The purpose of this report is threefold:
1) Provide an overview of the steps conducted in a GCBM proof of concept in a pilot area,
considering a subset of the REDD+ activities and carbon pools included in Chile’s FREL;
2) Reflect on the next steps required for the full implementation of the GCBM model to account
for all of the activities and regions eligible for REDD+, along with the required steps to improve
GHG emissions estimation, reporting and analysis capabilities in Chile using open-source
modelling tools, and
3) Reflect on the potential role a non-profit consortium, such as moja global, may have in
supporting countries to achieve multiple objectives for sustainable land-use planning and policy
development.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
2
Chile’s MRV professionals, located in the National Forestry Corporation (CONAF), designed and
conducted the proof of concept implementation and simulation runs in collaboration with the
Canadian Forest Service, moja global and the Mullion Group. The pilot project was undertaken in
the Los Rios Region in southern Chile, and included estimations of GHG emissions and removals
from changes in the total biomass component resulting from the following activities: deforestation
(conversion of forest to non-forest land use), substitution (conversion of native forest to exotic
trees plantations), and afforestation (conversion of non-forest to forest land use).
National data included in the land-use cadastre maps and the national forest inventory were
preprocessed and used as input to the model, while specific parameters were left as default when
not available from the national data. The process to customize the GCBM included the adjustment
of the volume to biomass parameters, the root parameters, the inclusion of a local spatial data
layer of temperature and the creation of custom disturbance matrices, among others. To obtain
annual disturbance layers for the area from discrete land-use maps covering multi-year periods,
the year of the disturbance was assigned at random within the periods between the land use
maps.
Although this exercise was only a rapid proof of concept and a demonstration of how the available
spatially-explicit data can be integrated in a transparent and consistent way, it is also possible to
examine the numerical results as long as it is understood that these will be revised with improved
implementation of data that reflect the Chilean forest conditions. The estimates of GHG emissions
from total biomass in the deforestation and substitution activities were very similar to the ones
presented in Chile’s FREL, with differences of less than 4% between both approaches (the GCBM
producing slightly higher emissions). The afforestation activity estimates from the GCBM showed
a difference of +25.4% relative to the FREL, which was attributed to model assumptions before
the start of the FREL period (e.g., annual GHG removals from afforestation activities include the
year 1997 and onwards; the GCBM pilot assumes zero GHG emissions from afforestation
activities, while Chile’s FREL accounts for biomass losses due to the conversion of non-forest
land to forest land use). These assumptions can be revised to be consistent with the assumptions
previously used by Chile.
This initial phase showed that the GCBM was able to produce similar estimates for the main
source of GHG emissions reported in the FREL, which corresponds to forest biomass losses due
to substitution and deforestation (62.2% of the net emissions reported in the FREL of Los Rios
Region). Furthermore, the GCBM implementation on the Chilean case is capable of producing
spatially-explicit results of forest carbon dynamics that can potentially inform national policy
making and be used to monitor actions to address the risks of reversals and to reduce
displacement of emissions as referred in the Cancun safeguards (UNFCCC, 2010).
This exercise proved that the use of complex models such as the GCBM can be easier when
performed by the national professionals that work with the data on a day-to-day basis, together
with the support of a global community of users of and contributors to open-source tools such as
moja global. This can enable the country and the domestic professionals to implement
sophisticated platforms, that have the potential of improving analytical power and reducing
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
3
dependencies on rotating consultant on short term contracts, showing the importance of national
staff consolidation and training.
This implementation of the GCBM is one of the first attempts to adapt a spatially-explicit model
for forest carbon accounting with the Chilean data at a regional level. Examples of the next steps
include continuing to improve model calibration and assumptions of biomass, dead organic matter
and soil organic carbon components, the inclusion of the REDD+ activities occurring in forest land
remaining as forest land, expanding the project to all the Chilean accountability zone for REDD+
activities, and supporting the development of open-source tools that can enhance MRV-AFOLU
systems globally.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
4
1 Introduction
The Paris Agreement of the United Nations Framework Convention on Climate Change
(UNFCCC) sets ambitious goals for reducing greenhouse gas (GHG) emissions to limit global
warming to less than 2 above pre-industrial levels before the end of this century (UNFCCC,
2015; IPCC, 2018). According to the Intergovernmental Panel on Climate Change (IPCC), the
Agriculture, Forest and Other Land Use (AFOLU) sector generates about 23% of the total net
anthropogenic emissions of GHG (IPCC, 2019). However, with enough economic investment and
capacity building support, particularly in developing countries, the AFOLU sector can help achieve
net zero GHG emissions, while providing other important social and natural co-benefits (e.g.,
increased adaptation, biodiversity conservation, improved water and soil health; IPBES, 2019;
FABLE, 2019).
REDD+ is one mechanism for driving reductions and enhanced removals from the land sector,
through results-based payments. REDD+ stands for reducing emissions from deforestation and
forest degradation plus conservation, sustainable management of forests and enhancement of
forest carbon stocks (UNFCCC, 2016). Under the UNFCCC’s Cancun Agreements, REDD+ is
divided into three phases (UNFCCC, 2010, 2011): Readiness, Demonstration Activities, and
Activities for Results-based Payments. The REDD+ mechanism requires that countries
undertaking REDD+ activities for results-based payments provide annual estimates of changes
in forest carbon stocks, and their corresponding GHG emissions and removals, by establishing a
system for national measurement, reporting and verification (MRV). In order to be credible, these
systems must be able to generate results that adhere to the TACCC principles: transparency,
accuracy, comparability, consistency and completeness, as described in the IPCC’s 2003
Guidance for National Greenhouse Gas Inventories (IPCC 2003).
In general, to design and implement MRV systems for REDD+, countries rely on a combination
of remote-sensing imagery and data collected through ground-level forest inventories to help
monitor carbon stock and stock changes (Birdsey et al., 2013; GOFC-GOLD, 2014). These
approaches include monitoring changes from the past, improving the level of understanding of
the drivers of these changes, and estimating emissions in the future (e.g., for both reference
emissions levels and mitigation pathways). Although substantial progress has been made on the
quality of the available information, particularly in the remote-sensing arena (Hansen et al., 2013;
Wulder et al., 2019; Dubayah et al., 2020), there is still a need to enhance the data integration
processes to reflect realities on the ground in a scalable, operational context. For example,
integrating multiple sources of data (both spatial and aspatial) under the same analytical
framework creates consistency, and can help countries to fulfill their UNFCCC reporting
requirements while informing operational decision-making. In addition, many developing countries
face the challenge that REDD+ activities and other land sector data are occurring at a variety of
scales (national, sub-national, and project level), and therefore require a system that can estimate
emissions at all scales.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
5
In 2017, the Linux Foundation launched the moja global project, an open-source initiative to
support sustainable land management efforts through the continuous development of software
for the land sector (moja global, 2017). The first major project for moja global software is the Full
Lands Integration Tool (FLINT) which is a modelling framework that integrates different data types
(spatial and aspatial) and modules to calculate GHG emissions and removals for the AFOLU
sector. FLINT builds on 20 years of experience with integration frameworks developed and
operated in the land sector by countries like Australia and Canada. Because modules, data, and
output processes can be attached to the FLINT framework in a unique configuration (known as
an “implementation”), the system can be tailored to meet country-specific needs and capacity in
a progressive and efficient manner. To date, there are three examples of implementations of
FLINT and available modules: the System for Land Based Emissions Estimation in Kenya
(SLEEK), a software-as-a-service version of FLINT called FLINTpro, and the Generic Carbon
Budget Model (GCBM) developed by the Canadian Forest Service. The latter is based on the
same science modules used in the Carbon Budget Model of the Canadian Forest Sector (CBM-
CFS3; Kurz et al., 2009), which is a modelling framework focused on forest carbon dynamics,
with several applications outside of Canada (Kim et al., 2015; Pilli et al., 2016; Dugan et al., 2018),
including examples for the analysis of REDD+ scenarios (Olguin et al., 2011, 2018).
Currently, Chile is one of a growing number of countries that have published a Forest Reference
Emissions Level / Forest Reference Level (FREL) and a Technical Annex of REDD+ Results
(UNFCCC 2020)
1
, making it eligible to apply for results-based payments under the Green Climate
Fund of the UNFCCC (i.e., US$ 63.3m in 2019). However, many of the estimates of GHG
emissions and removals are conducted mainly using Excel spreadsheets, limiting the country’s
ability to obtain spatially-explicit results. The possibility to model forest carbon dynamics and
report forest carbon results in a spatially-explicit manner is one of the key aspects of the planned
steps Chile identified to improve their MRV system. This would also allow the country to enhance
the information that is available to policy makers and to monitor Cancun safeguards such as
emissions reversals and to reduce emissions displacements (decision 1/CP.16; UNFCCC, 2010).
The purpose of this report is to document progress to date on the use of the GCBM tool by Chilean
government experts in Chile, located in the National Forestry Corporation (CONAF), the national
focal point for REDD+. This report is part of Chile’s stepwise approach to compare results
between the methods used in the compilation of the FREL, against those of a forest carbon
dynamics model with a spatially-explicit approach (GCBM). Specifically, the study provides an
overview of the steps conducted in a proof of concept in the Los Ríos Region, considering a
subset of the REDD+ activities and pools included in Chile’s FREL. These results will help define
the next steps for the full implementation of the GCBM model in the complete accountability zone
for REDD+, along with the required steps to improve GHG emissions estimation, reporting and
analysis capabilities in Chile using open-source carbon modelling tools. The hope is that this kind
of tools will improve the capacity of the country to obtain accurate, spatially-explicit, and
streamlined estimates of GHG fluxes associated with REDD+ activities. Finally, the report also
1
Documents available at https://redd.unfccc.int/submissions.html?country=chl
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
6
reflects on the potential role a non-profit consortium, such as moja global, may have in supporting
countries to achieve multiple objectives for sustainable land-use planning and policy
development.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
7
2 Methods and input data
As part of its REDD+ Program, Chile submitted its subnational FREL to the UNFCCC, which
comprised four REDD+ activity types: reducing emissions from deforestation, reducing emissions
from forest degradation, conservation of forest carbon stocks, and enhancement of forest carbon
stocks. The reference level was constructed at a subnational scale, covering five regions from
central-south Chile: Maule, Biobío (including the Ñuble region), La Araucanía, Los Rios and Los
Lagos, comprising 11 out of 12 Chilean forest types and 41% of the forest area of the country.
After the successful publication of the FREL on March 3, 2017, Chile started developing the
Technical Annex of REDD+ Results, which was submitted on 3 December 2018. Measured
against the FREL, the results reported a reduction of emissions of 6,454,090 tCO2e per year for
the years 2014-2016. One of the purposes of this proof of concept is to test if the assumptions
and calculations made by Chile can be replicated using the alternative, model-based approach of
the GCBM.
2.1 Pilot area
In order to test the capability of the model to work with Chilean data, a representative pilot region
was chosen from the regions described in the subnational FREL. Los Rios Region was selected
as it contains activity records for the four REDD+ activities that Chile reports, and it contains 10
of the 11 forest types of the reporting area. Additionally, the data were well-organized and
available to use for the integration framework.
Located in southern Chile (Figure 1), Los Ríos Region is 18,429.5 km2 (2.4% of the national
territory) in size, and comprises 11.1% of the area reported in the FREL. It is characterized by the
presence of temperate rainforest, a high amount of rainfall, that ranges from 1,487 mm per annum
in San Jose de la Mariquina (Central Valley) to 3,903 mm per annum in the Valdivian Coastal
Reserve. The mean annual temperature is 11 °C (Pichoy Meteorological Station) (Center for
Climate and Resilience Research, 2019a).
The Capital of Los Rios Region is Valdivia and according to the 2017 census, the total population
of the region is 384,837 inhabitants meaning a density of 20.88 inhabitants per square kilometer.
A large part of the regional economy is based on the forestry industry (Biblioteca del Congreso
Nacional, 2019), with a high presence of exotic planted forest, mainly Pinus radiata and
Eucalyptus globulus species.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
8
Figure 1- Location of Los Rios Region
2.2 General approach
As part of Chile’s stepwise procedures to improve their MRV processes, new tools are being
tested to integrate the internal processes for estimation and reporting. As Chile already consigned
their FREL to the UNFCCC, which was consequently reviewed and approved, it is necessary that
tools such as the GCBM are, in the first place, able to replicate the procedures that are already in
place. Thus, this proof of concept focused on the replication of the assumptions and
methodologies presented in the Chilean FREL using the GCBM. At the same time, tools such as
the GCBM can allow for continuous improvements in methods also required by mechanisms such
as the GCF and FCPF.
In this pilot, a subset of REDD+ activities was selected as part of this first test run, with the
eventual aim of implementing all the REDD+ activities in Chile. The pilot project had the objective
of quantifying the CO2 emissions and removals derived from REDD+ activities or sub-activities
that involve a land-use change, meaning deforestation, substitution (land-use change from native
forest to forest plantation) and afforestation (planting of new native forests) (Table 1). It is
important to point out that for REDD+ accounting purposes, the forest plantations with exotic
species (hereafter forest plantations) are not considered in the context of the national definition.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
9
Table 1. REDD+ activities included in Chile’s FREL
REDD+ activity
Land Use Change
Permanent forest (Forest
remaining forest)
Reducing emissions from
Deforestation
Conversion of native forest to
other non forest land uses*
Reducing Emissions from Forest
degradation
Conversion of native forest to
forest plantations (substitution)*
Permanent forest degradation
Forest fires
Enhancement of forest carbon
stocks
Transformation of other land
uses to native forests*
Recovery of degraded forests
Conservation of forest carbon
stocks
Net flux of emissions in
permanent forest by degradation
of permanent forest and
recovery of degraded forest in
conservation areas
* Included in this Pilot project.
2.3 Modelling approach
The Generic Carbon Budget Model (GCBM) is an open-source modelling framework focused on
forest carbon dynamics, consistent with the Intergovernmental Panel on Climate Change (IPCC)
Guidelines for National GHG Inventories reporting (IPCC, 2006) and the Good Practice Guidance
for Land Use, Land-Use Change and Forestry (IPCC, 2003), that combines empirical data on
biomass dynamics with a process modelling approach for the simulation of dead organic matter
(DOM) and soil organic carbon (SOC) pools. The GCBM is built on the platform of the Full Lands
Integration Tool (FLINT) using Tier 3 modules, that are based on the well documented science of
the CBM-CFS3 (Kurz et al., 2009). The GCBM allows the integration of inputs using a spatially-
explicit approach (Approach 3 of IPCC 2003). In general, the GCBM integrates information from
forest inventories, growth and yield curves, and natural and/or anthropogenic disturbance events,
to simulate carbon stocks for the IPCC´s five forest carbon pools (above and belowground
biomass, dead wood, litter and soil), and the fluxes associated with changes in these pools, both
at the stand (e.g. pixel) and landscape levels.
The GCBM implements the IPCC gain-loss method to estimate stocks and stock changes. The
GCBM models the overall ecosystem carbon balance from all sources and sinks at each annual
time step, considering carbon gains from gross forest growth minus carbon losses due to
decomposition of DOM at the pixel level, plus carbon stock changes due to the effect of
disturbances over larger regions and time spans. The model tracks carbon mass transfers through
known pathways in and out of the forest ecosystem to ensure carbon mass balance.
In order to be able to obtain similar results between the FREL report and the GCBM runs, several
of the model parameters were customized using local data, while others were left as defaults for
this proof of concept. In future applications of the integration framework, some of the parameters
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
10
that were left as defaults, such as decomposition or litterfall rates, can be changed to national
variables taken from the literature or other research programs.
In this first pilot project, the calibration of the parameters was performed focusing on the GHG
emissions and removals associated with carbon stock changes in the forest biomass (both below
and aboveground), while the DOM and SOC values were left “as is” and would have to be
improved in future applications of the model to better reflect Chile’s national ecosystem
conditions.
2.4 Sources of data
To feed the model with the necessary information, two main sources of data were used: The
Forest Cadastre maps, and the National Forest Inventory.
The Forest Cadastre maps are the product of an ongoing project called “Cadastre and Evaluation
of Vegetation Resources in Chile”, developed by the Government of Chile from 1993 to the
present time. The main objective of this government project is to create a national cadastre of
land use and vegetation classification, with a focus on the native and planted forest. This
information is periodically updated via regional projects that monitor the changes in land use and
vegetation, providing public data that are used for decision making and land management.
The methodology used to build the Forest Cadastre was developed by Etienne & Prado (1982)
and consists of the manual interpretation of land use data that is supported by fieldwork and
interpretation of high-resolution satellite and aerial photos. This allows Chile to have a rich set of
information that includes the forest type (as classified by Donoso, 1981), forest structure and
detailed land use, among others.
In the case of Los Rios Region, the baseline of the Cadastre was developed in 1997, with updates
in 2006 and 2013. These land use maps were developed using manual interpretation along with
a mixture of high- and medium- resolution satellite data, supported by extensive fieldwork. In
2017, another update of the land use maps was developed using semi-automated remote sensing
techniques, allowing the country to have data in shorter periods of time.
The four above mentioned land-use maps are merged together into an integrated file called
“Trazabilidad”, that consists of polygons of land use and an associated attribute table containing
the historical variations of the polygons in the four periods of time, allowing the user to quickly
derive the land-use changes, and accordingly, the REDD+ activities which are reported to have
occurred in the different periods of time.
The second set of input data used in this project is the National Forest Inventory. This project,
called “Continuous Inventory of Forest Ecosystems”, is managed by the National Forest Institute
(INFOR) since 2000. The purpose of the inventory is to generate detailed information about the
native forests of Chile, supporting the decision-making process. The Forest Inventory uses a two-
stage statistical sampling design, distributing three circular plots in an area of 500 m2, and using
a 7x5 km systematic grid. The first cycle of measurements was developed between 2001 and
2010. A portion of the sample is re-measured each year in order to obtain information about the
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
11
growth and functionality of the forests. The Forest Inventory is used to obtain the Annual
Periodical Increment (growth) of each forest type, the regional emission factors for deforestation,
and the factors used for volume to biomass conversions.
2.5 Simulation runs
To conduct simulation runs, the GCBM uses a stepwise iterative approach which allows users to
change data sources, if and when new or improved data become available. Using this stepwise
approach, the different aspects of the GCBM were filled in order to transition from using the GCBM
with default parameters to running the simulations with customized Chilean parameters. To
accomplish this, the different components of the GCBM were customized using Python (Version
2.7) and R (Version 2.6.3) codes, allowing the user to input the data of the “Trazabilidad” layer
and the factors derived from the National Forest Inventory into the GCBM. A general workflow of
the input data used and the processing steps involved in the GCBM implementation, that will be
explained in detail in the next sections, can be found in Figure 2.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
12
Figure 2. Workflow of Chile’s GCBM implementation
The resolution used for the conversion of the polygon data provided by the cadastre into the raster
data used by the GCBM was set to 0.0005 degrees, being approximately equivalent to a pixel
size of 55 m by 55 m. This resolution was chosen because the more detailed maps of the
“Trazabilidad file (2017) were produced with a minimum mapping unit of 0.36 ha.
2.5.1 Inventory data at the start of the simulation (Classifiers)
The GCBM, as in the CBM-CFS3, requires an initial dataset called “inventory”, comprising the
forest stands that are present at the beginning of the simulation. This dataset, that in the GCBM
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
13
has to be spatially explicit, contains the age of the stand and a set of characteristics of the stand
called “classifiers,” which are defined by the user and can be related to site productivity, leading
species and ownership, among others. Each set of unique classifiers is related to a specific yield
curve (Kurz et al., 2009)
In order to start the modelling, the inventory dataset to run the GCBM was derived from the first
year of the Forest Cadaster (1997). To create the inventory layer, three classifiers were used to
associate the stands with the different land uses and growth curves that were included in the
FREL (Table 2). The first classifier corresponds to the forest type classification of Donoso (1981),
that is determined by the dominant species of the ecosystem and has a strong influence on the
growth rate of the forests. The second classifier is the Structure of the forest, as it also influences
the growth rates. The latter classifier is measured in the forest cadaster and reflects the vertical
and horizontal layer distribution and is comprised by four categories: (a) Renoval: Secondary
grown forest, generally presenting a high number of trees per hectare. (b) Adulto: Adult forest,
primary grown, with a high presence of old trees with high DBH. (c) Adulto-Renoval. A stand
containing a mixture of secondary and primary grown forests (d) Achaparrado: Low height forest
growing in climate and soil restrictive conditions, as in the high altitudes of the Andes Ranges.
Additional to these categories, two more classes were created: Bosque Mixto, to reflect native
forest with a high presence of exotic invasive species, and non-forests, for land uses different to
native forest. Apart from these two classifiers that are used in the FREL to determine the annual
growth of the forest, a new classifier was created (Origin) to track the emission reductions derived
from non-forest to forest land use changes (afforestation or reforestation).
The initial age of the forest stands in the inventory layer was set to 100 years old, to reflect a
forest reaching the regional average of merchantable volume. Thus, the native forest coming from
the initial inventory have 375.29 m3/ha of merchantable volume (average in Los Rios Region).
This assumption is used in the FREL, as the assumed maximum carbon stock for intact forests,
and thus a complete reduction of that stock would be classified as deforestation
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
14
Table 2. Classifiers used in Chile’s GCBM implementation
Classifier
Forest type
Structure
Origin
Alerce
Araucaria
Bosque Mixto
Cipres de la Cordillera
Ciprés de las Guaitecas
Coihue - Rauli - Tepa
Coihue de Magallanes
Esclerofilo
Lenga
Roble - Rauli - Coihue
Siempreverde
Non-forest (includes forest
plantations)
Renoval (secondary grown)
Adulto (Adult, primary
Grown)
Adulto Renoval (Mixture of
primary and secondary
grown forests)
Achaparrado (low-height
Forest growing in climate
and soil restrictive
conditions)
Non-forest (includes forest
plantations)
Bosque Mixto (Native forest
with high presence of exotic
species)
Initial (Native Forest coming
from the 1997 Forest
Cadaster)
New Forests (new detected
native forests in the forest
cadasters of 2006, 2013 and
2017)
Non-forests (includes forest
plantations)
2.5.2 Biomass dynamics
2.5.2.1 Growth curves
In order to model forest biomass accumulation through time, the GCBM requires yield or growth
curves that are linked to different sets of classifiers. These curves are represented as
merchantable volume of stem wood per year, for each set of classifiers. The GCBM growth curve
library can accommodate a wide range of growth curve types defined by the user. These could
be simple “constant growth” rates similar to those used in Tier 1 or Tier 2 emission factors (growth
per hectare and year by forest type). More sophisticated sigmoidal growth curves (e.g. Chapman-
Richards growth equations) or derived from growth and yield models can also be used, where
these are available.
In the Chilean case, to reflect the dynamics of the native forest of Los Rios Region, customized
growth curves were constructed and used as input data for the model. To incorporate the different
assumptions that were used in the FREL, two groups of growth curves were developed, using the
Origin classifier to differentiate them.
Origin: Initial Forest
In order to estimate the CO2 emissions from deforestation, Chile’s FREL uses regionally-specific
emission factors (Tier 2), derived from the National Forest Inventory. These emission factors do
not change with time, and thus, a single growth curve was used (Figure 3) to reflect this “static,
homogeneous condition. This growth curve reaches its maximum at the age of 100 years, a
number that was used as an approximation of the age in which the native forest reaches dynamic
equilibrium, representing 375.29 m3/ha of merchantable volume (average in Los Rios Region).
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
15
Figure 3. Growth curves for initial forests
Origin: New Forest
For the afforestation activities, Chile’s FREL uses Annual Periodic Increment values to calculate
the growth of each new forest, according to forest type and structure. To reflect this decision,
linear growth curves were constructed for each combination of forest type and structure (see
Table 2 for structure classes). To be consistent with the initial forest condition, the growth curves
stop their increments at the age of 100 years (Figure 4), when they reach the maximum volume
value (Annual Periodic Increment * 100). The Bosque Mixto structure, that is not included in Figure
4, presents an average annual periodic increment weighted by area of the different forest types
in the region (the same assumption as in the FREL)
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
16
Figure 4. Simplified growth curves by forest type and structure (new forests) used in the pilot.
With additional data and time, the growth curves can be improved.
2.5.2.2 Volume to biomass expansion
The GCBM model growth curves use values of merchantable stem wood volume per hectare
(m3/ha), that are converted to aboveground merchantable stem biomass per hectare in the model
using parameters derived from sample plot measurements. After that, this variable is used to get
the biomass values of the different biomass components of the forest, including sapling trees,
branches and foliage of merchantable and non-merchantable wood. In order to do that, the
equations of Boudewyn et al. (2007) were fitted for a set of Canadian forests in the original CBM-
CFS3 model.
The approach of Boudewyn et al. (2007), includes several equations to derive the biomass of the
different tree components from the volumes of merchantable total biomass. However, Chile’s
approach in the FREL is to use a single expansion factor (1.75) to derive the total volume of the
forest and then to use the basic wood density (0.5) to get the oven-dry biomass and a factor of
0.5 to estimate the carbon content in that biomass. In order to reconcile these two different
approaches, only the merchantable total biomass was calculated for the forests in this GCBM
pilot. Equation 1 of Boudewyn et al. (2007) was used to calculate the total merchantable stem
biomass:
b_m = a * volumeb
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
17
where,
volume = gross merchantable volume/ha of all live trees (volume does not include stumps,
tops, or trees< merchantable DBH), in m3/ha.
b_m = total stem wood biomass of merchantable-sized live trees (biomass includes
stumps and tops), in metric tonnes per ha.
a,b = non-linear model parameters fit separately by Canadian jurisdictions, ecozones, and
lead tree species
In the case of Los Rios Region, awas set to 0.875 (expansion factor = 1.75 multiplied by the
basic wood density = 0.5) and “b” was set to 1.
The other equations included in Boudewyn et al. (2007) were parameterized to give a value of
zero for non-merchantable-sized trees, saplings, and a proportion of zero for bark, branches and
foliage, making the stem wood the only component of the tree that was considered in this
simulation.
The root parameter in this case was set to 0.2829 for hardwood and softwood values (same value
as in the FREL). The proportions for the coarse and fine root components were left as default.
2.5.2.3 Dead Organic Matter and Soil Carbon
In accordance with the mass balance concept in the 2006 IPCC guidelines, the GCBM explicitly
links dead organic matter (DOM) and soil organic carbon (SOC) dynamics with the dynamics in
the live biomass carbon pools. Similar to the CBM-CFS3 model (Kurz et al. 2009), the initialization
of the DOM and SOC pools in GCBM uses a spin-up procedure to track carbon transfers from
live to dead biomass, as a function of tree productivity, decay rates, climate (mean annual
temperature), historic disturbances, as well as the type of, and time since, the last stand-replacing
disturbance.
In this study, all the ecological parameters required to calibrate DOM dynamics (e.g., biomass
turnover rates, litterfall transfer rates, decomposition rates) were obtained from default data
available within the GCBM. Only the information on climate was specific to Chile. These data
were derived from the raster products from the Center for Climate and Resilience Research
(CR2). To obtain the mean annual temperature for the period, the CR2MET monthly observations
(Center for Climate and Resilience Research, 2019b) were averaged for the period of 1997 and
2016. These were used to create a raster layer with 5 km spatial resolution, that was used as
input for the GCBM. In the future finer resolution climate data can also be used, where available.
2.5.3 Disturbances
To simulate the effect of natural and human disturbances (e.g. land-use changes, selective
harvesting, wildfire, hurricanes, pests, etc.) on ecosystem carbon dynamics, the GCBM uses
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
18
disturbance matrices. These matrices contain information about 22 carbon pools that can be
easily grouped into the five IPCC carbon pools, to represent carbon transfers among carbon pools
in the forest ecosystem and among these pools and the atmosphere due to land cover change
events (Kurz et al., 2009). Where harvesting occurs, the disturbance matrix also defined the
amount of carbon transferred to the wood product sector. To recreate deforestation activities as
in the FREL, a customized disturbance matrix was created in which 100 per cent of the biomass
in the aboveground and belowground components of the trees are immediately converted to CO2
(instant oxidation). This assumption matches the assumption used in the FREL. In the future,
biomass can be transferred to dead organic matter pools and then either be burned or decompose
over time to generate a more realistic representation of the impacts of land-use changes.
The abovementioned disturbance matrix was also used for the substitution activity, as the
conversion of native forest to forest plantations is calculated as a deforestation event for Chile’s
REDD+ purposes (consistent with the policy assumptions made by Chile, the growth of the forest
plantations are not considered in the emission/removal estimations). In the future, the contribution
of forest plantations to the GHG balance can also be evaluated for other reporting purposes.
The third disturbance type that was considered is afforestation. This disturbance occurs when a
non-forest polygon (including forest plantations) is converted to native forest. In each of the new
forest polygons, the forest type and structure were specified, and the origin classifier was set to
“New forest” to link it with the corresponding growth curve.
As mentioned above, the land use maps that are being used to derive the disturbances were
produced in the years 1997, 2006, 2013 and 2017. Thus, in order to create a more realistic
scenario, a randomization of the disturbance year was implemented. As an example, if a
disturbance in a polygon was detected in the 2006 land use map, a random year between 1997
and 2006 was assigned to the disturbance event. In the future, auxiliary data on land cover
changes could be used to detect and assign the year of disturbance.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
19
3 Results and Discussion
3.1 Areas of Disturbance
The results of the randomization of the disturbance year are shown in Figure 5. The results show
a large proportion of the substitution events were detected between 1997 and 2006, with a large
decrease in the 2014-2016 period, while the deforestation regime tends to have a more
homogeneous behavior over time. On the other hand, the afforestation events had a much larger
contribution in the results-based payment period (2014-2016) than in the reference level period
(2001-2013). Note that the sum of the disturbances in any of the periods is based on data, while
the allocation of the sum to individual years in the period is based on random allocation. With
more time, Monte-Carlo simulation with alternative allocations could be conducted to quantify the
contribution to the uncertainties resulting from annual allocation of disturbance events within each
of the periods.
Figure 5. Annual areas affected by afforestation, deforestation and substitution disturbances
To determine the annualized area of disturbance in the FREL, a linear interpolation was used. To
compare the approach of the FREL with the one used in this pilot project, the mean disturbance
rates were calculated and compared with the ones reported by Chile (Table 3). It is possible to
see that the interpolation and randomization approaches produced relatively small differences.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
20
Table 3. Comparison between the disturbance areas in the FREL and this pilot project.
Area (ha/year)
GCBM Pilot
Project
Area (ha/year)
FREL
Difference (%)
756.19
750.77
0.72%
908.43
919.21
-1.17%
912.84
911.93
0.1 %
3.2 Results for the deforestation emissions
The deforestation activity was modeled in the GCBM, and calculations of the CO2 emissions from
losses of live biomass (both above and belowground biomass) and DOM were extracted (although
the latter was not calibrated for this pilot and should therefore only be interpreted as examples).
As seen in Figure 6, the emissions from biomass are far larger than the ones coming from the
transfer of DOM to CO2 (atmosphere).
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
21
Figure 6. CO2 emissions from deforestation
When comparing these results with the ones presented in the FREL for the period between 2001
and 2013, it is clear that the emissions from biomass were relatively similar, with the emissions
calculated by the GCBM only 3.92% larger than the ones calculated in the FREL. This difference
can be attributed to the fact that, in the FREL, the biomass of the post-disturbance land use
classes is subtracted from the CO2 emissions from deforestation. In the case of the GCBM, all
the non-forest land use classes have default biomass of zero, which contributes to these small
differences.
In the case of the emissions from DOM, the estimations of the GCBM are significantly less than
the ones reported in the FREL (-75.02%). This difference is due to the fact that the volume to
biomass factors that were used in the model were set to only consider the stem biomass, omitting
the inclusion of foliage, branches and other elements that contribute to the accumulation of DOM.
This caused the spin-up procedure of the GCBM to accumulate low quantities of carbon in the
DOM reservoir, and thus making the emissions significantly less. This behavior of the figures
demonstrates that the model was designed with an integrated approach as the main focus,
respecting mass balance at all times, and showing that it is important to take into account the
interaction between carbon pools to obtain consistent estimates.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
22
Although not the focus of this pilot the large difference in the emissions from the DOM pool makes
the total emissions from deforestation lower than reported in the FREL (-15.86%), as indicated in
Table 4.
Table 4. Comparison of CO2 emissions from deforestation between the FREL and the GCBM. Note
that DOM emission estimates are based on a preliminary and incomplete implementation of the
model.
GCBM estimated
emissions
(tCO2/year)
FREL estimated
emissions
(tCO2/year)
Differences
between FREL
and GCBM
estimations (%)
40,368
161,591
-75.02%
502,065
483,1050
3.92%
542,434
644,696
-15.86%
3.3 Results for the substitution emissions
Regarding the emissions from the conversion of native forest to forest plantations (substitution),
the results of the model show, as in the previous section, that the emissions from biomass are
significantly larger than those from the DOM (Figure 7).
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
23
z
Figure 7. CO2 emissions from substitution
When comparing the GCBM estimated emissions with the ones reported in the FREL, as
expected, the behavior is similar to that observed in the deforestation emissions, presenting a
much lower amount of emissions from DOM (-77.43%) due to the assumptions described above
(e.g., volume to biomass conversion). In the case of the GCBM estimates from biomass, the
emissions are almost equal to the ones reported in the FREL (-0.5%). This can be attributed to
the differences between the randomization of the disturbance year and the interpolation of the
disturbed area used in the FREL, as in the FREL the forest plantations are considered to have
zero biomass, hence not affecting the emission estimations. The total CO2 in this case shows a
difference of 156,102 tCO2, being much lower (-20.15%) in this GCBM pilot project (Table 5).
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
24
Table 5. Comparison of CO2 emissions from substitution between the FREL and the GCBM. Note
that DOM emission estimates are based on a preliminary implementation of the model.
GCBM estimated
emissions
(tCO2/year)
FREL estimated
emissions
(tCO2/year)
Differences
between FREL
and GCBM
estimations (%)
44,6519
197,845
-77.43%
573,869
576,777
-0.50%
618,520
774,621.5
-20.15%
3.4 Afforestation removals
The results for the afforestation activities were measured with the “Delta total biomass” indicator
(Figure 8), with the knowledge that in the FREL, the removals due to DOM accumulation were
not considered. This indicator shows a steady increment with the inclusion of new forest areas,
as new forests are planted, increasing the removals in time.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
25
Figure 8. Delta total biomass indicator in new forests (afforestation)
The comparison of the estimations of the GCBM and the FREL gives greater results for the former
of +24.41% (Table 6). This result can be explained by two main differences between the
approaches. In the first place, in this pilot all the new native forest areas planted from 1997 to
2016 were considered to calculate the mean yearly removal in 2001-2013, while in the FREL a
mean annual afforestation rate was calculated and applied in 2001 onwards, producing much
lower removal rates in the first years of the reference level period. In the second place, the GCBM
model does not consider a biomass value for the pre-afforestation land uses, while the FREL
considers the emissions caused by the land use change, subtracting them from the afforestation
removals.
Table 6. Comparison of CO2 removals from afforestation between the FREL and the GCBM
GCBM estimated
removals
(tCO2/year)
FREL estimated
removals
(tCO2/year)
Differences
between FREL
and GCBM
estimations (%)
66,383
53,356
24.41%
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
26
3.5 Spatially-Explicit Results
One of the key aspects of the GCBM model is its capability to generate spatially-explicit results.
As an example, in Figure 9, the annual change in total biomass carbon density (tCO2/ha/year)
indicator results are shown in a zone south of Valdivia, where several deforestation and
substitution events occurred between 2002 and 2005 (considering the year of disturbance
randomization approach), while on the other hand new forests are planted, accumulating
removals year by year.
Figure 9. Example of the spatially-explicit outputs produced by the GCBM
Moreover, results like the ones shown in Figure 9 could be compared against other remote
sensing analyses that can demonstrate that areas protected are indeed still present. Also, these
results can be used to monitor safeguards such as removals reversal or emissions leakage, when
implemented across a broader area.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
27
4 Lessons learned
The implementation of this first step in Chile’s pilot project on the use of the GCBM integration
framework has generated several lessons. These lessons can be valuable for the next steps in
the implementation of this kind of model or for other similar initiatives in the region.
Firstly, this model was implemented in a short amount of time, because a great portion of the
spatially explicit data in Chile was already preprocessed and integrated into consolidated land
use change files. The professionals that programmed the preprocessing steps were already
familiar with the characteristics of the Chilean data and the assumptions made in the calculation
of the FREL, making the comparative nature of this pilot process much more cost-effective.
When comparing the GHG emission from biomass values, both results (GCBM and FREL) were
quite similar, showing that the GCBM can potentially provide almost the same results as in the
FREL. Despite the understandable differences in the DOM emissions, this initial phase showed
that the GCBM was able to produce similar estimates for the main source of GHG emissions
reported in the FREL, which corresponds to forest biomass losses due to substitution and
deforestation (62.2% of the net emissions reported in the FREL of Los Rios Region). Moreover,
as in this pilot project the focus was centered on the calibration of biomass values, the national
team was able to assess the effects of the biomass parameters in the DOM estimations in the
GCBM model, demonstrating the effects of certain assumptions on the accumulation rates and
transfer of carbon between the different forest carbon pools, showing the importance of
considering an integrated approach for forest carbon modelling.
The GCBM has the big advantage of providing spatially-explicit results that could significantly
improve the way the MRV system of Chile operates. The GCBM also provides an opportunity for
continuous improvement beyond the limitations of spreadsheet-based approaches (e.g., difficult
to integrate large quantities of data, more time consuming and prone to human errors), enhancing
the quality of the estimates (e.g., ensuring carbon mass balance) and progressively transitioning
to more complex analysis of forest carbon dynamics, that includes transfers and interaction
between the different carbon pools. In addition, it enhances the operational capabilities that MRV
systems need to have to comply with a range of reporting and scenario analysis commitments
(e.g., FCTF, UNFCCC, Nationally Determined Contributions, etc.), without compromising any of
the TACCC principles. GCBM data processing can all be scripted (e.g. using Python) and
because of that the scripts can be re-used and audited by other team members or outside experts,
This increases both the transparency and the reproducibility of the results important criteria
when requesting results-based payments.
The internal capacity of Chile’s team was complemented by the help of the professionals from the
Canadian Forest Service Carbon Accounting Team (CFS-CAT), professionals from the moja
global project, and the Mullion Group, which collaborated with their experience of the model
functioning and implementation. This synergy between local knowledge and the broader
experience in forest carbon modelling was crucial to the success of an early stage implementation
of the model in Chile, that started with a training workshop on the use of the CBM-CFS3 organized
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
28
in Santiago, Chile, with participation of professionals from seven countries of Latin America (moja
global, 2019). The following communication between the local community and the global experts
was performed via e-mail or by verbal communication and as such should be codified, via the
inclusion of user manuals, public forums and technical documents like this one.
For the implementation and customization of the GCBM for this pilot project, intermediate skills in
the SQL and Python programming languages were necessary, making the inclusion of a
professional with intermediate knowledge of computer programming necessary in the first phase.
However, in future iterations, because the GCBM is already customized, changing simple
variables or the inclusion of updated data is straightforward and can be done with the computer
skills of a professional from the environmental sciences. This fact makes the GCBM suitable for
countries that want to have a semi-automated framework to obtain spatially-explicit results and to
test their estimations, producing different scenarios for REDD+ or other forest carbon accounting
applications. The establishment of these models can also provide a cost-effective tool to help the
country test different scenarios of public policy, such as reforestation initiatives helping the country
to make policy decisions based on national data.
To implement MRV systems on these more sophisticated platforms has the potential to greatly
improve the analytical power, reduce dependencies on rotating consultants on short-term
contracts, and ultimately will lead to increased national capacity to support both reporting and
projections of GHG emissions and removals in the land sector. To be successful, however,
countries need to invest into staff training, and commit to sustaining the teams that are the basis
for national reporting and projection capacities. Recurring tasks such as GHG reporting, Biannual
Reporting, submission of Nationally Determined Contributions and collection of results-based
payments, all require that capacity in domestic MRV systems is developed and maintained. Moja
global was created to facilitate such capacity building through the use of open-source software
such as the models tested in this pilot project.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
29
5 Next Steps
The potential next steps in the development of the implementation of the GCBM in the Chilean
case study include:
A. Include the results period (2014-2016) in the analysis: So far, the analysis was done
using the reference period (2001-2013) for the comparison of results. The results period,
2014-2016, was not considered in the pilot project.
B. Improve the estimates of emissions from Biomass and DOM components: As seen
in the methods section, the calibration of the model took into account the biomass, but not
the DOM, causing the emissions from DOM to be considerably underestimated. In order
to improve the DOM estimations, the way the volume to biomass factors are designed has
to be changed to be able to input variables in a more generic and simple approach,
improving both the biomass and DOM estimations. These changes should be
complemented by an intensive, specific literature review and expert opinions, that takes
into account the most recent scientific findings on rates of litterfall, decomposition rates,
dead organic matter to soil organic carbon transfer rates, and many other parameters that
are not included in the FREL. In the case of the biomass components, new more detailed
growth curves can be included, together with the inclusion of snags, branches and foliage
data in the volume to biomass conversions. For example, depending on time and
resources available, future simulations could include testing and selecting those species
from the GCBM database with the most similar allometry to the Chilean species, or,
developing a database of volume to biomass parameters within the GCBM structure that
is specific to Chile.
C. Inclusion of the activities that occur in forest remaining forest: The activities that
were included in this pilot project have one characteristic that makes them easy to
implement: all of them follow a gain-loss approach in the Chilean FREL. On the other
hand, the estimations for the activities that occur in permanent forest (forest remaining
forest) were made using a stock-difference approach based on mosaics of Landsat
images, that were used to interpolate ground data and calculate the difference in forest
stocks in two different years.
The implementation of the permanent forest related REDD+ activities will take longer than
the implementation of the land-use change based activities, due to the fact that the input
data would have to be adapted in order to meet the requirements of the GCBM gain-loss
based approach.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
30
Overcoming this barrier, it will be possible to evaluate the use of this type of tool in new
managed forest areas as a result of the implementation of results-based-payments (e.g.,
the green climate fund between the years 2020-2025 or others as FCPF CF).
D. Expand the pilot program to the rest of the regions of the FREL: This pilot project
considered only one out of the five regions included in Chile’s sub-national REDD+
accounting area. As the format and characteristics of the information in Los Rios region is
relatively similar to the other regions, the implementation of the GCBM in all the regions
between Maule and Los Lagos should be straightforward.
E. Continuing to support global initiatives that enhances the operational capabilities
of MRV-AFOLU systems: The moja global project of the Linux Foundation builds on an
international community of users and contributors that supports open-source tools for
MRV-AFOLU systems. Applying the best available science, technical skills and software,
moja global promotes the creation of operational systems that can be tailored to meet
country-specific needs and capacity in a progressive and efficient manner.
This report describes progress to date in replicable steps of the use of the GCBM-FLINT
tool in MRV-Chile for FREL/REDD+ analysis. At the same time, it exemplifies how this
type of global initiative can support countries to speed up the analytical and reporting
capabilities of their MRV systems, paving the way for new collaboration opportunities with
other countries, that will then strengthen regional capacity-building processes (e.g. South-
South collaboration) and the consistency of international reporting of GHG emissions in
the AFOLU sector.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
31
6 Conclusions
The implementation of the GCBM in the Chilean case allowed Chile to test, compare, and learn
about the most recent practices in forest carbon modelling, allowing the country to have more
options and approaches to update their FREL and REDD+ Technical Annex in the future. It also
allowed them to obtain spatially-explicit results, improving the capabilities of the National Forest
Measurement and Monitoring System. Thus, the GCBM could be one possible solution to
integrate and automate the estimation and reporting workflows for the UNFCCC and other
reporting requirements, allowing the country to have one core, vertically-integrated, accurate, and
spatially-explicit system for forest carbon accounting.
The implementation of the model was developed in a short time due to the availability of global
networks, that allowed Chile’s MRV professionals to test and receive feedback about the different
problems they faced when implementing it. This has demonstrated that developing the internal
capacity of individual countries, complemented by horizontal international collaboration is a more
effective way to improve countries' MRV processes.
This project corresponds to one of the first attempts of the country to obtain spatially-explicit and
integrated results on the REDD+ activities, generating coherent results when comparing them
with the established FREL, providing an effective proof of concept of what can be done in the
future in terms of the development of integrated carbon models for Chile’s MRV system.
Furthermore, with the use of these technologies, in the near future Chile could undertake
accounting with an integrated and semi-automatic platform to test scenarios, build reports for
different initiatives and manage REDD+ policies, among others, in a cost-effective manner.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
32
Acknowledgments
This work was made possible by the collaboration between the professionals of the MRV team of
the Climate Change and Environmental Services Unit (UCCSA) of the National Forest
Corporation (CONAF) Julian Cabezas, Daniel Montaner, Georgina Trujillo, Cesar Mattar and José
Antonio Prado, a group of advisors from the Canadian Forest Service Carbon Accounting Team,
including Werner Kurz, Max Fellows and Stephen Kull and the Mullion Group professionals
Marcela Olguin, Robert de Ligt, Bryan Adkins and Robert Waterworth. The international training
workshop on the use of the CFS-CBM3 in June 2019, that enabled Chile to start the
implementation on the model, was financed and supported by the Swiss Agency for Development
and Cooperation, UNFCCC, FAO and SilvaCarbon (an interagency technical cooperation
program of the US Government). The work-hours to implement the model were provided by the
MRV professionals from CONAF, while the funding to write and consolidate this document came
from the “Expanding Market Opportunities” contribution agreement between Her Majesty the
Queen in Right of Canada, represented by the Minister of Natural Resources and the non-for-
profit organization “Climate Advisers Trust.
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
33
References
Biblioteca del Congreso Nacional, (2019). Región de Los Ríos. Retrieved from:
https://www.bcn.cl/siit/nuestropais/region14.
Birdsey R., Angeles G., Kurz W. A., Lister A., Olguin M., Pan Y., Wayson C. A., Wilson B. Johnson
K., (2013) Approaches to monitoring changes in carbon stocks for REDD+ Carbon
Management, 4, 51937.
Center for Climate and Resilience Research, (2019a). Explorador climático CR2. Retrieved from:
http://explorador.cr2.cl/
Center for Climate and Resilience Research, (2019b). Producto
CR2MET_v1.3_t2m_month_1979_2016. Retrieved from http://www.cr2.cl/datos-productos-
grillados/
Donoso, C., (1981). Tipos forestales de los bosques nativos de Chile. Documento de trabajo n°
38. Corporación Nacional Forestal. 70 pp.
Dubayah, R., Blair, J.B., Goetz, S., Fatoyinbo,L.,... Silva C. (2020) The Global Ecosystem
Dynamics Investigation: High-resolution laser ranging of the Earth’s forests and topography,
Science of Remote Sensing, 1, 1000002 DOI: https://doi.org/10.1016/j.srs.2020.100002
Dugan, A.J., Birdsey, R., Mascorro, V.S., Magnan, M., Smyth, C.E., Olguin, M., Kurz, W.A.,
(2018). A systems approach to assess climate change mitigation options in landscapes of
the United States forest sector. Carbon Balance Manag. 13. https://doi.org/10.1186/s13021-
018-0100-x
Etienne, M.G; Prado, C.C., (1982) Descripción de la vegetación mediante la cartografía de
ocupación de tierras: conceptos y manual de uso práctico. Ciencias Agrícolas N°10,
Universidad de Chile, 120 pp.
FABLE, (2019). Pathways to Sustainable Land-Use and Food Systems. 2019. Report of the
FABLE Consortium. Laxenburg and Paris: International Institute for Applied Systems
Analysis (IIASA) and Sustainable Development Solutions Network (SDSN).
GOFC-GOLD, (2014) A sourcebook of methods and procedures for monitoring and reporting
anthropogenic greenhouse gas emissions and removals associated with deforestation, gains
and losses of carbon stocks in forests remaining forests, and forestation. GOFC-GOLD
Report version COP20-1, (GOFC-GOLD Land Cover Project Office, Wageningen University,
The Netherlands).
Government of Kenya (2015). Kenya: Second national communication to the United Nations
Framework Convention on Climate Change. Retrieved from:
https://unfccc.int/documents/109628.
Hansen, M.C., Potapov, P.V. Moore, R Hancher, M … Townshend, J.R.G. (2013) High-resolution
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
34
global maps of 21st-century forest cover change, sICENCE, 342 (6160): 850-853. DOI:
10.1126/science.1244693
Intergovernmental Panel on Climate Change (IPCC), (2003). Good Practice Guidance for Land
Use Change, Land Use Change and Forestry (Hayama: Institute for Environmental
Strategies). Retrieved from: www.ipcc-nggip.iges.or.jp/public/gpglulucf/gpglulucf.html
Intergovernmental Panel on Climate Change (IPCC), (2006). Guidelines for National Greenhouse
Gas Inventories Agriculture Forestry and Other Land Use (AFOLU) vol 4 (Hayama: Institute
for Environmental Strategies) (also available at: www.ipcc-
nggip.iges.or.jp/public/2006gl/index.html).
Intergovernmental Panel on Climate Change (IPCC), (2019). IPCC Special Report on Climate
Change, Desertification, Land Degradation, Sustainable Land Management, Food Security,
and Greenhouse gas fluxes in Terrestrial Ecosystems Summary for Policymakers Approved
Draft. Retrieved from: https://www.ipcc.ch/site/assets/uploads/2019/08/4.-
SPM_Approved_Microsite_FINAL.pdf
IPBES, (2019). Global assessment report on biodiversity and ecosystem services of the
Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. E. S.
Brondizio, J. Settele, S. Díaz, and H. T. Ngo (editors). IPBES Secretariat, Bonn, Germany.
39 pages.
Kurz W. A., Dymond C. C., White T. M., Stinson G., Shaw C. H, Rampley G J, Smyth C, Simpson
B N, Neilson E T, Trofymow J A, Metsaranta J and Apps M. J., (2009). CBM-CFS3: A model
of carbon-dynamics in forestry and land use change implementing IPCC standards Ecol.
Model. 220 480504
Kim, M., Lee, W.K., Kurz, W.A., Kwak, D.A., Morken, S., Smyth, C.E, Ryu. D., (2016). Estimating
carbon dynamics in forest carbon pools under IPCC standards in South Korea using CBM-
CFS3. iForest - Biogeosciences and Forestry, 10 (1), 83-92, https://doi.org/10.3832/ifor2040-
009
moja global, (2017). Project Charter. Retrieved from: https://github.com/moja-global/About-moja-
global
moja global (2019), Moja global news: A summary of Chile workshop. Retrieved from:
https://moja.global/2019/07/a-summary-of-the-chile-workshop-2/
Olguín M., Kurz W., De Jong B., Paz F., Ángeles G., Zermeño C., Flores R., (2011). Hacia el uso
del modelo CBM-CFS3 a escala nacional en México: proyecto piloto Chiapas. En: Paz F.,
Cuevas R.M. (Eds.). Estado Actual del Conocimiento del Ciclo del Carbono y sus
Interacciones en México: Síntesis a 2011. Programa Mexicano del Carbono, Universidad
Autónoma del Estado de México e INE. ISBN 978-607-715-085-5. pp. 90-97.
Olguin, M., Wayson, C., Fellows, M., Birdsey, R., Smyth, C.E., Magnan, M., Dugan, A.J.,
Mascorro, V.S., Alanís, A., Serrano, E., Kurz, W.A., (2018). Applying a systems approach to
Modelling forest carbon dynamics for REDD+ using the GCBM
Pilot Project: Los Rios Region, Chile
35
assess carbon emission reductions from climate change mitigation in Mexico’s forest sector.
Environ. Res. Lett. 13, 035003. https://doi.org/10.1088/1748-9326/aaaa03
Pilli, R., Grassi, G., Kurz, W., Fiorese, G., Cescatti, A., (2016) The European forest sector: past
and future carbon budget and fluxes under different management scenarios, Biogeosciences
DOI: 10.5194/bg-2016-525
United Nations Framework Convention on Climate Change (UNFCCC), (2010). Report of the
Conference of the Parties on its sixteenth session, held in Cancun from 29 November to 10
December 2010. Retrieved from:
https://unfccc.int/resource/docs/2010/cop16/eng/07a01.pdf#page=12
United Nations Framework Convention on Climate Change (UNFCCC), (2011). Guidance on
systems for providing information on how safeguards are addressed and respected and
modalities relating to forest reference emission levels as referred to in decision 1/CP.16
Decision CP.17 FCCC/CP/2011/9/Add.2
United Nations Framework Convention on Climate Change (UNFCCC), (2015). United Nations
Framework Convention on Climate Change Adoption of the Paris Agreement Report No.
FCCC/CP/2015/L.9/Rev.1.
United Nations Framework Convention on Climate Change (UNFCCC), (2016). Key decisions
relevant for reducing emissions from deforestation and forest degradation in developing
countries (REDD+). Retrieved from:
https://unfccc.int/files/land_use_and_climate_change/redd/application/pdf/compilation_redd
_decision_booklet_v1.2.pdf
United Nations Framework Convention on Climate Change (UNFCCC), (2020). REDD+ MRV
and results-based payments. REDD+ web platform. Retrieved from:
https://redd.unfccc.int/fact-sheets/redd-mrv-and-results-based-payments.html
Wulder M.A., Loveland T.R., Roy, D.P., Crawford, C.J., Zhu, Z., (2019) Current status of Landsat
program, science, and applications. Remote Sensing of Environment, 225, 127-147
https://doi.org/10.1016/j.rse.2019.02.015
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Obtaining accurate and widespread measurements of the vertical structure of the Earth’s forests has been a long-sought goal for the ecological community. Such observations are critical for accurately assessing the existing biomass of forests, and how changes in this biomass caused by human activities or variations in climate may impact atmospheric CO2 concentrations. Additionally, the three-dimensional structure of forests is a key component of habitat quality and biodiversity at local to regional scales. The Global Ecosystem Dynamics Investigation (GEDI) was launched to the International Space Station in late 2018 to provide high-quality measurements of forest vertical structure in temperate and tropical forests between 51.6° N & S latitude. The GEDI instrument is a geodetic-class laser altimeter/waveform lidar comprised of 3 lasers that produce 8 transects of structural information. Over its two-year nominal lifetime GEDI is anticipated to provide over 10 billion waveforms at a footprint resolution of 25 m. These data will be used to derive a variety of footprint and gridded products, including canopy height, canopy foliar profiles, Leaf Area Index (LAI), sub-canopy topography and biomass. Additionally, data from GEDI are used to demonstrate the efficacy of its measurements for prognostic ecosystem modeling, habit and biodiversity studies, and for fusion using radar and other remote sensing instruments. GEDI science and technology are unique: no other space-based mission has been created that is specifically optimized for retrieving vegetation vertical structure. As such, GEDI promises to advance our understanding of the importance of canopy vertical variations within an ecological paradigm based on structure, composition and function.
Article
Full-text available
Formal planning and development of what became the first Landsat satellite commenced over 50 years ago in 1967. Now, having collected earth observation data for well over four decades since the 1972 launch of Landsat-1, the Landsat program is increasingly complex and vibrant. Critical programmatic elements are ensuring the continuity of high quality measurements for scientific and operational investigations, including ground systems, acquisition planning, data archiving and management, and provision of analysis ready data products. Free and open access to archival and new imagery has resulted in a myriad of innovative applications and novel scientific insights. The planning of future compatible satellites in the Landsat series, which maintain continuity while incorporating technological advancements, has resulted in an increased operational use of Landsat data. Governments and international agencies, among others, can now build an expectation of Landsat data into a given operational data stream. International programs and conventions (e.g., deforestation monitoring, climate change mitigation) are empowered by access to systematically collected and calibrated data with expected future continuity further contributing to the existing multi-decadal record. The increased breadth and depth of Landsat science and applications have accelerated following the launch of Landsat-8, with significant improvements in data quality. Herein, we describe the programmatic developments and institutional context for the Landsat program and the unique ability of Landsat to meet the needs of national and international programs. We then present the key trends in Landsat science that underpin many of the recent scientific and application developments and follow-up with more detailed thematically organized summaries. The historical context offered by archival imagery combined with new imagery allows for the development of time series algorithms that can produce information on trends and dynamics. Landsat-8 has figured prominently in these recent developments, as has the improved understanding and calibration of historical data. Following the communication of the state of Landsat science, an outlook for future launches and envisioned programmatic developments are presented. Increased linkages between satellite programs are also made possible through an expectation of future mission continuity, such as developing a virtual constellation with Sentinel-2. Successful science and applications developments create a positive feedback loop-justifying and encouraging current and future programmatic support for Landsat.
Article
Full-text available
Background: United States forests can contribute to national strategies for greenhouse gas reductions. The objective of this work was to evaluate forest sector climate change mitigation scenarios from 2018 to 2050 by applying a systems-based approach that accounts for net emissions across four interdependent components: (1) forest ecosystem, (2) land-use change, (3) harvested wood products, and (4) substitution benefits from using wood products and bioenergy. We assessed a range of land management and harvested wood product scenarios for two case studies in the U.S: coastal South Carolina and Northern Wisconsin. We integrated forest inventory and remotely-sensed disturbance data within a modelling framework consisting of a growth-and-yield driven ecosystem carbon model; a harvested wood products model that estimates emissions from commodity production, use and post-consumer treatment; and displacement factors to estimate avoided fossil fuel emissions. We estimated biophysical mitigation potential by comparing net emissions from land management and harvested wood products scenarios with a baseline ('business as usual') scenario. Results: Baseline scenario results showed that the strength of the ecosystem carbon sink has been decreasing in the two sites due to age-related productivity declines and deforestation. Mitigation activities have the potential to lessen or delay the further reduction in the carbon sink. Results of the mitigation analysis indicated that scenarios reducing net forest area loss were most effective in South Carolina, while extending harvest rotations and increasing longer-lived wood products were most effective in Wisconsin. Scenarios aimed at increasing bioenergy use either increased or reduced net emissions within the 32-year analysis timeframe. Conclusions: It is critical to apply a systems approach to comprehensively assess net emissions from forest sector climate change mitigation scenarios. Although some scenarios produced a benefit by displacing emissions from fossil fuel energy or by substituting wood products for other materials, these benefits can be outweighed by increased carbon emissions in the forest or product systems. Maintaining forests as forests, extending rotations, and shifting commodities to longer-lived products had the strongest mitigation benefits over several decades. Carbon cycle impacts of bioenergy depend on timeframe, feedstocks, and alternative uses of biomass, and cannot be assumed carbon neutral.
Article
Full-text available
The Paris Agreement of the United Nation Framework Convention on Climate Change calls for a balance of anthropogenic greenhouse emissions and removals in the latter part of this century. Mexico indicated in its Intended Nationally Determined Contribution and its Climate Change Mid-Century Strategy that the land sector will contribute to meeting GHG emission reduction goals. Since 2012, the Mexican government through its National Forestry Commission, with international financial and technical support, has been developing carbon dynamics models to explore climate change mitigation options in the forest sector. Following a systems approach, here we assess the biophysical mitigation potential of forest ecosystems, harvested wood products and their substitution benefits (i.e. the change in emissions resulting from substitution of wood for more emissions-intensive products and fossil fuels), for policy alternatives considered by the Mexican government, such as a net zero deforestation rate and sustainable forest management. We used available analytical frameworks (Carbon Budget Model of the Canadian Forest Sector and a harvested wood products model), parameterized with local input data in two contrasting Mexican states. Using information from the National Forest Monitoring System (e.g. forest inventories, remote sensing, disturbance data), we demonstrate that activities aimed at reaching a net-zero deforestation rate can yield significant CO2e mitigation benefits by 2030 and 2050 relative to a baseline scenario ('business as usual'), but if combined with increasing forest harvest to produce long-lived products and substitute more energy-intensive materials, emissions reductions could also provide other co-benefits (e.g. jobs, illegal logging reduction). We concluded that the relative impact of mitigation activities is locally dependent, suggesting that mitigation strategies should be designed and implemented at sub-national scales. We were also encouraged about the ability of the modeling framework to effectively use Mexico's data, and showed the need to include multiple sectors and types of collaborators (scientific and policy-maker communities) to design more comprehensive portfolios for climate change mitigation.
Article
Full-text available
The comprehensive analysis of carbon stocks and fluxes of managed European forests is a prerequisite to quantify their role in biomass production and climate change mitigation. We applied the Carbon Budget Model (CBM) to 26 European (EU) countries, parameterized with country information on the historical forest age structure, management practices, harvest regimes and the main natural disturbances. We quantified C stocks for the five forest pools plus Harvested Wood Products (HWP), and the fluxes among these pools, from 2000 to 2030. The aim is to quantify the main C fluxes as affected by land-use changes, natural disturbances and forest management and to asses the impact of specific harvest and afforestation scenarios after 2012 on the mitigation potential of the EU forest sector. Substitution effects and the possible impacts of climate are not included in this analysis. Results show that for the historical period (2000–2012) the net primary productivity (NPP) of the forest pools at the EU level is on average equal to 639 Tg C yr−1, the losses are dominated by heterotrophic respiration (409 Tg C yr−1) and fellings (110 Tg C yr−1 due to removals), with direct fire emissions being only 1 Tg C yr−1, leading to a net carbon stock change (i.e. sink) of 110 Tg C yr−1. Fellings also transferred 28 Tg C yr−1 of harvest residues from biomass dead organic matter pools. The average annual forest Net Sector Exchange (NSE), i.e. the carbon stock changes in the forest pools plus HWP, equals 122 Tg C yr−1 (i.e., about 19 % of the NPP) for the historical period and in 2030 reaches 126 Tg C yr−1, 101 Tg C yr−1 and 151 Tg C yr−1, assuming respectively a constant, increasing (+20 %) and decreasing (−20 %) scenario of both harvest and afforestation rates compared to the historical period. Under the constant harvest rate scenario, our findings show an incipient aging process for the forests existing in 1990: despite the NPP is still increasing (+7 %) the greater increase of heterotrophic respiration (+13 %) leads to a decrease of the sink in the forest pools (−6 %) in 2030 compared to the historical period. The detailed picture of the C fluxes within the EU forest sector condensed in this study, and their evolution under different harvest scenarios, provide a useful framework for more detailed future analyses (spatially or temporally) on the mitigation potential of the forest sector. Overall, our study confirms that, in the majority of European countries, the build-up of biomass stocks results from woody NPP exceeding losses by timber harvest and natural disturbances. However, at the country level, we highlighted some statistical differences, suggesting that this relationship cannot be assumed as constant for all European countries. Specific forest conditions, such as the harvest rate, the current age structure and forest composition, may have a different impact on the country-specific evolution of biomass stocks.
Article
Full-text available
Reducing emissions from deforestation and forest degradation plus improving forest-management (REDD+) is a mechanism to facilitate tropical countries' participation in climate change mitigation. In this review we focus on the current state of monitoring systems to support implementing REDD+. The main elements of current monitoring systems - Landsat satellites and traditional forest inventories - will continue to be the backbone of many forest-monitoring systems around the world, but new remote-sensing and analytical approaches are addressing monitoring problems specific to the tropics and implementing REDD+. There is increasing recognition of the utility of combining remote sensing with field data using models that integrate information from many sources, which will continue to evolve as new sensors are deployed and as the availability of field data increases.
Article
Full-text available
The scientific community, forest managers, environmental organizations, carbon-offset trading systems and policy-makers require tools to account for forest carbon stocks and carbon stock changes. In this paper we describe updates to the Carbon Budget Model of the Canadian Forest Sector (CBM-CFS3) implemented over the past years. This model of carbon-dynamics implements a Tier 3 approach of the Intergovernmental Panel on Climate Change (IPCC) Good Practice Guidance for reporting on carbon stocks and carbon stock changes resulting from Land Use, Land-use Change and Forestry (LULUCF). The CBM-CFS3 is a generic modelling framework that can be applied at the stand, landscape and national levels. The model provides a spatially referenced, hierarchical system for integrating datasets originating from different forest inventory and monitoring programs and includes a structure that allows for tracking of land areas by different land-use and land-use change classes. Ecosystem pools in CBM-CFS3 can be easily mapped to IPCC-defined pools and validated against field measurements. The model uses sophisticated algorithms for converting volume to biomass and explicitly simulates individual annual disturbance events (natural and anthropogenic). Several important scientific updates have been made to improve the representation of ecosystem structure and processes from previous versions of CBM-CFS. These include: (1) an expanded representation of dead organic matter and soil carbon, particularly standing dead trees, and a new algorithm for initializing these pools prior to simulation, (2) a change in the input data requirement for simulating growth from biomass to readily available merchantable volume curves, and new algorithms for converting volume to biomass, (3) improved prediction of belowground biomass, and (4) improved parameters for soil organic matter decay, fire, insect disturbances, and forest management. In addition, an operational-scale version of CBM-CFS3 is freely available and includes tools to import data in standard formats, including the output of several timber supply models that are commonly used in Canada. Although developed for Canadian forests, the flexible nature of the model has enabled it to be adapted for use in several other countries.
Article
The main objective of this study was to estimate past and future dynamics of forest carbon pools in South Korea, as classified by the Intergovernmental Panel on Climate Change (IPCC) Good Practice Guidance (GPG) and to test the Carbon Budget Model of the Canadian Forest Sector 3 (CBM-CFS3). Inventory data required in CBM-CFS3 were extracted from Korea’s 3rd Forest Type Map. The site index in the inventory data was estimated through regression models developed using c National Forest Inventory (NFI) data with 4000 plots. Necessary yield tables to a stand age of over 100 years, stratified by species, were prepared through the estimation of volume per ha with tree species, site index and stand age. We considered thinning and cutting regulations for sustainable forest management announced by the Korean Forest Service and that the forest area was constant over from 1992 to 2092. We estimated the carbon stocks and their changes of aboveground (AGB) and belowground (BGB) biomass, litter (L), dead wood (DW), and soil organic matter (SOM). Validation was conducted based on the 5th NFI and statistical data. Our results showed that the carbon content of the total forest area increased from 509.7 Tg C to 1007.3 Tg C at a rate of 11.8 Tg C yr⁻¹ during 1992-2034. Results also showed that AGB, BGB, DW, L and SOM changed from 54.6, 30.2, 15.9, 43.6, and 373.9 Tg C in 1992 to 455.6, 100.77, 32.89, 65.46, and 369.2 Tg C in 2034, respectively. However, the amount of forest carbon was projected to decrease due to large increases in harvest rates as most of the forest reaches the legislated cutting age during 2035-2045. Our simulation estimated that by 2045 AGB, BGB, and SOM decreased to 347.8, 78.2, and 368.9 Tg C, respectively, while DW and L increased to 46.8 and 89.1 Tg C, respectively. The carbon content of the total forested area starts to stabilize after 2045 as the annual stand growth gradually decreases. Finally in 2092, the carbon content of AGB, BGB, DW, L, and SOM was estimated as 422.3, 93.9, 31.2, 63.0, and 365.1 Tg C, respectively. Through harvesting, over the 100 year period more than 200 Tg C were transferred from South Korea’s forests to meet demands for timber, fiber and energy. Good agreement between model results, NFI data and independent studies, demonstrates the applicability of CBM-CFS3 for estimating past and future forest carbon budgets in South Korean forests and for exploring forest management activity impacts in managed forests.
Tipos forestales de los bosques nativos de Chile. Documento de trabajo n° 38. Corporación Nacional Forestal
  • C Donoso
Donoso, C., (1981). Tipos forestales de los bosques nativos de Chile. Documento de trabajo n° 38. Corporación Nacional Forestal. 70 pp.