This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 50.193.129.250

This content was downloaded on 05/11/2013 at 20:37

Please note that terms and conditions apply.

REDD+ readiness: early insights on monitoring, reporting and verification systems of project

developers

View the table of contents for this issue, or go to the journal homepage for more

2013 Environ. Res. Lett. 8 034038

(http://iopscience.iop.org/1748-9326/8/3/034038)

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ. Res. Lett. 8 (2013) 034038 (15pp) doi:10.1088/1748-9326/8/3/034038

REDD+ readiness: early insights onmonitoring, reporting and verificationsystems of project developers

Shijo Joseph1, Martin Herold2, William D Sunderlin1 andLouis V Verchot1

1 Center for International Forestry Research, Jalan CIFOR, Bogor 16115, Indonesia2 Center for Geoinformation, Department of Environmental Science, Wageningen University,The Netherlands

E-mail: s.joseph@cgiar.org and shijonrsa@gmail.com

Received 23 March 2013Accepted for publication 17 September 2013Published 30 September 2013Online at stacks.iop.org/ERL/8/034038

AbstractA functional measuring, monitoring, reporting and verification (MRV) system is essential toassess the additionality and impact on forest carbon in REDD+ (reducing emissions fromdeforestation and degradation) projects. This study assesses the MRV capacity and readinessof project developers at 20 REDD+ projects in Brazil, Peru, Cameroon, Tanzania, Indonesiaand Vietnam, using a questionnaire survey and field visits. Nineteen performance criteria with76 indicators were formulated in three categories, and capacity was measured with respect toeach category. Of the 20 projects, 11 were found to have very high or high overall MRVcapacity and readiness. At the regional level, capacity and readiness tended to be highest in theprojects in Brazil and Peru and somewhat lower in Cameroon, Tanzania, Indonesia andVietnam. Although the MRV capacities of half the projects are high, there are capacitydeficiencies in other projects that are a source of concern. These are not only due to limitationsin technical expertise, but can also be attributed to the slowness of internationalREDD+ policy formulation and the unclear path of development of the forest carbon market.Based on the study results, priorities for MRV development and increased investment inreadiness are proposed.

Keywords: deforestation, carbon pool, climate change mitigation, remote sensing, REDD+

1. Introduction

Reducing emissions from deforestation and degradation andenhancing the carbon stock (REDD+) is a key strategy formitigating climate change. The United Nations FrameworkConvention on Climate Change (UNFCCC) takes a leadrole in formulating the strategy through consultation witha range of agencies including other international bodies,

Content from this work may be used under the terms ofthe Creative Commons Attribution 3.0 licence. Any further

distribution of this work must maintain attribution to the author(s) and thetitle of the work, journal citation and DOI.

national and subnational governments, non-governmentalorganizations and the private sector. Formulated at the 13thConference of the Parties (COPs), known as the Bali ActionPlan, REDD+ comprises local, subnational, national andglobal actions whose primary aim is to reduce emissionsfrom deforestation and forest degradation and enhance forestcarbon stocks (increase removals) in developing countries(UNFCCC 2007, Angelsen et al 2012). UNFCCC (2011) hascalled for a three-phase approach to REDD+ implementationwhere ‘Phase 1 is the REDD+ readiness to establishREDD+ strategies, policies and accounting frameworks;Phase 2 is implementation of REDD+ policies and measures,and possibly payments based on their implementation; and

11748-9326/13/034038+15$33.00 c© 2013 IOP Publishing Ltd Printed in the UK

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Phase 3 is the results-based payments for REDD+ (i.e. foremissions and removals)’.

REDD+ adopts several funding streams and models forits implementation that operate simultaneously at variouslevels (van der Eynden 2012, Lotsch 2012). The firstmodel involves developed countries contributing to theGreen Climate Fund or Forest Carbon Partnership Facility(FCPF), which are administered by the UNFCCC and theWorld Bank, respectively. These funds are allocated tonational governments, which administer and distribute themto subnational governments and individual REDD+ projects.A second model involves international aid agencies support-ing national governments and individual REDD+ projectsthrough bilateral agreements. For example, Norway supportsthe Brazilian and Indonesian governments in the implemen-tation of REDD+, and also independent REDD+ projectsin Tanzania. A third model involves development of thevoluntary carbon market (which could be later changed intoa compliance market) for selling carbon credits generatedby the REDD+ projects. This funding stream is exclusivelyfor the REDD+ projects and therefore needs to follow strictguidelines issued by the certifiers such as the verified carbonstandard (VCS) and Plan Vivo.

As of 2009, about 100 local REDD+ projects hadbeen initiated in 40 developing countries (Cerbu et al 2011,Obersteiner et al 2009). A 2012 count of both REDD+ andnon-REDD+ forest carbon projects found 340 sites in52 countries (Kshatriya et al 2012). Unlike many otherconservation initiatives, the key goal of REDD+ projects is toimplement performance-based rewards for increase of forestcarbon stocks (additionality) against a baseline, while alsoachieving social and environmental co-benefits. REDD+ aimsto ensure that activities to generate additionality in oneplace should not cause emissions in neighbouring areas(leakage), and should also satisfy permanence criteria (i.e., noreversibility of sequestered carbon).

The economic and political dimensions of REDD+ andits performance-based criteria require highly precise mea-suring/monitoring, reporting and verification (MRV) ofemissions and removals from forests (Herold and Skutsch2011). Establishing an effective MRV system is a majorstep towards REDD+ readiness. UNFCCC has called forthe development of ‘modalities for measuring, reporting andverifying anthropogenic forest-related emissions by sourcesand removals by sinks, forest carbon stocks, and forestcarbon stock and forest-area changes resulting from theimplementation of the activities’ (UNFCCC 2011). Keyadditional references refer to activity data and emissionfactors as the essential building blocks of an MRV system(IPCC 2003, 2006, Shoch et al 2011, GOFC-GOLD 2012).Activity data reflect the extent of the area in which carbonchanges occur over time, while emission factors reflect theamount of carbon per unit area. The Intergovernmental Panelon Climate Change (IPCC) proposed a tiered approach formeasurements: Tier 1, the simplest to use, is based on globallyavailable activity data and default emissions values, Tier 2is based on country- or region-specific data, and Tier 3 useshigher-order, spatially explicit data to obtain greater certainty(IPCC 2006).

Several studies have analysed MRV systems at thenational scale using data from the Food and AgricultureOrganization’s Forest Resources Assessment, national com-munications to UNFCCC and other online compilations toreview existing capacity and identify gap areas (Herold 2009,Romijn et al 2012, Hosonuma et al 2012). A number ofconcept papers on MRV requirements at different scaleshave primarily looked at data availability (Gibbs et al 2007,Joseph et al 2011, Goetz and Dubayah 2011, De Sy et al2012) and the associated cost (Bottcher et al 2009). Fry(2011), Larrazabal et al (2012) and Pratihast et al (2013)highlighted the need to develop community MRV protocolsto maximize community participation and reduce costs. Noneof the above publications focused on MRV systems perse or the operational design of MRV systems required forREDD+ projects. The only exception to this trend wasbaseline modelling and monitoring readiness with respect tointernationally recognized carbon certification standards andmethods (Estrada and Joseph 2012).

2. Data and methods

2.1. Data

Assessing MRV capacity and readiness at project sitesrequires a complete understanding of project developers’MRV systems, the drivers of deforestation at the sites, andthe planned mitigation activity. After careful examination ofthe national context and subnational initiatives (for the fullselection criteria, see Sunderlin et al 2010, pp 17–20), 20REDD+ project developers were selected for this assessment:seven in Latin America (five in Brazil and two in Peru),seven in Africa (five in Tanzania and two in Cameroon)and six in Southeast Asia (five in Indonesia and one inVietnam). Table 1 summarizes key project characteristics, andfigure 1 shows their geographical distribution. The projectdevelopers have varied characteristics ranging from local andnational NGOs (e.g.: MCDI—Tanzania, CED—Cameroon,IPAM—Brazil), international conservation organizations (FFIand TNC), government organizations (States of Acre—Braziland Aceh—Indonesia), and private organizations (MazarsStarling Resources—Indonesia and Bosques Amazonicos—Peru). One key criterion for the selection of these projects inour sample was that they have sufficient funding to operate forseveral years. This was important because we are conducting alongitudinal survey, and we need to be assured we can collecta second round of data.

The assessment was carried out from October 2011 toSeptember 2012. The first step was to carry out an MRVquestionnaire survey to assess the status and progress ofMRV systems of the principal project developers. The surveywas divided into three sections, namely: remote sensingand GIS; ground measurements; and reference level andmonitoring plans. The survey was done by communicatingdirectly with the key MRV expert in the respective projectdevelopment organizations. Furthermore, a field-based surveywas conducted at each project site that collected informationon the overall condition of the project area, the drivers of

2

Environ. Res. Lett. 8 (2013) 034038 S Joseph et alTa

ble

1.D

etai

lsof

part

icip

atin

gR

ED

D+

proj

ects

ites

and

proj

ectd

evel

oper

s.

Cou

ntry

Stat

e/pr

ovin

cePr

ojec

tnam

ePr

inci

palp

roje

ctde

velo

per

Maj

orde

fore

stat

ion

and

degr

adat

ion

driv

ers

Miti

gatio

nac

tivity

plan

neda

12

34

5

Bra

zil

Mat

oG

ross

oN

orth

wes

tMat

oG

ross

oPi

lotR

ED

D+

Proj

ect

Inst

ituto

Cen

tro

deV

ida

(IC

V)

Cat

tlera

nchi

ngan

dbe

efpr

oduc

tion,

subs

iste

nce

farm

ing,

illeg

allo

ggin

g

YY

YY

N

Para

Avo

ided

defo

rest

atio

nin

smal

lrur

alpr

oper

ties

inth

ere

gion

ofth

eT

rans

amaz

onH

ighw

ay

Am

azon

Env

iron

men

tal

Res

earc

hIn

stitu

te(I

PAM

)C

attle

ranc

hing

and

beef

prod

uctio

n,su

bsis

tenc

efa

rmin

g,ill

egal

logg

ing

YY

YY

Y

Para

RE

DD+

Pilo

tPro

gram

inSa

oFe

lixdo

Xin

guT

heN

atur

eC

onse

rvan

cy(T

NC

)B

razi

lC

attle

ranc

hing

and

beef

prod

uctio

n,su

bsis

tenc

efa

rmin

g,ill

egal

logg

ing

YY

YY

N

Am

azon

asB

losa

Flor

esta

Prog

ram

Fund

acao

Am

azon

asSu

sten

tave

l(FA

S)Il

lega

llog

ging

,min

ing,

infr

astr

uctu

rede

velo

pmen

t,su

bsis

tenc

efa

rmin

g

YY

YY

N

Am

apa

Bio

fılic

aIn

vest

imen

tos

Am

bien

tais

and

Fund

acao

Ors

a,A

map

a

Bio

filic

aSu

bsis

tenc

efa

rmin

g,ill

egal

logg

ing,

infr

astr

uctu

rede

velo

pmen

t

YY

YY

N

Peru

Mad

rede

Dio

sB

razi

lNut

Con

cess

ions

RE

DD+

Proj

ect

Bos

ques

Am

azon

icos

(BA

M)

Cat

tlera

nchi

ngan

dbe

efpr

oduc

tion,

subs

iste

nce

farm

ing,

min

ing,

illeg

allo

ggin

g

YY

YY

N

San

Mar

tinA

ltoM

ayo

RE

DD+

Initi

ativ

eC

onse

rvat

ion

Inte

rnat

iona

lC

omm

erci

alpl

anta

tions

(cof

fee)

,cat

tlera

nchi

ngan

dbe

efpr

oduc

tion,

subs

iste

nce

farm

ing

YN

NN

N

Cam

eroo

nSo

uth

and

east

regi

onPa

ymen

tfor

Eco

syst

emSe

rvic

es(P

ES)

proj

ect

Cen

tre

pour

l’E

nvir

onne

men

tetl

eD

evel

oppe

men

t(C

ED

)

Subs

iste

nce

farm

ing,

logg

ing,

extr

actio

nof

non-

timbe

rfo

rest

prod

ucts

YY

YY

Y

Sout

hw

estr

egio

nM

t.C

amer

oon

RE

DD

Proj

ect

GFA

-Env

est

Com

mer

cial

plan

tatio

ns(o

ilpa

lm,r

ubbe

ran

dte

a),

subs

iste

nce

farm

ing,

illeg

allo

ggin

g

YY

YY

N

Tanz

ania

Shin

yang

aC

omm

unity

Bas

edR

ED

DM

echa

nism

sfo

rSu

stai

nabl

eFo

rest

Man

agem

enti

nSe

mi-

Ari

dA

reas

Tanz

ania

Tra

ditio

nal

Ene

rgy

Dev

elop

men

tand

Env

iron

men

tal

Org

aniz

atio

n(T

aTE

DO

)

Subs

iste

nce

farm

ing,

fuel

woo

dan

dch

arco

alpr

oduc

tion,

over

graz

ing

YY

YY

N

Mor

ogor

oM

akin

gR

ED

Dw

ork

for

Com

mun

ities

and

Fore

stC

onse

rvat

ion

inTa

nzan

ia–K

ilosa

Tanz

ania

Fore

stC

onse

rvat

ion

Gro

up(T

FCG

)

Subs

iste

nce

farm

ing,

fuel

woo

dan

dch

arco

alpr

oduc

tion,

fire

YY

YY

N

3

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Tabl

e1.

(Con

tinue

d.)

Cou

ntry

Stat

e/pr

ovin

cePr

ojec

tnam

ePr

inci

palp

roje

ctde

velo

per

Maj

orde

fore

stat

ion

and

degr

adat

ion

driv

ers

Miti

gatio

nac

tivity

plan

neda

12

34

5

Lin

diM

akin

gR

ED

Dw

ork

for

Com

mun

ities

and

Fore

stC

onse

rvat

ion

inTa

nzan

ia–L

indi

Tanz

ania

Fore

stC

onse

rvat

ion

Gro

up(T

FCG

)

Subs

iste

nce

farm

ing,

fuel

woo

dan

dch

arco

alpr

oduc

tion,

fire

YY

YY

N

Ung

uja/

Zan

ziba

rH

IMA

—Pi

lotin

gR

ED

Din

Zan

ziba

rth

roug

hC

omm

unity

Fore

stM

anag

emen

t

CA

RE

Inte

rnat

iona

lin

Tanz

ania

Fuel

woo

dan

dch

arco

alpr

oduc

tion,

subs

iste

nce

farm

ing,

infr

astr

uctu

rede

velo

pmen

t

YY

NN

N

Lin

diM

ping

oC

onse

rvat

ion

Proj

ect

Mpi

ngo

Con

serv

atio

nan

dD

evel

opm

entI

nitia

tive

(MC

DI)

Fire

,sub

sist

ence

farm

ing,

fuel

woo

dan

dch

arco

alpr

oduc

tion

YY

NN

N

Indo

nesi

aA

ceh

Red

ucin

gC

arbo

nE

mis

sion

sfr

omD

efor

esta

tion

inth

eU

luM

asen

Eco

syst

em

Gov

ernm

ento

fA

ceh

(Tas

kFo

rce

RE

DD

Ace

h)Su

bsis

tenc

efa

rmin

g,lo

ggin

gY

YY

YN

Wes

tKal

iman

tan

RE

DD

Pilo

tPro

ject

Dev

elop

men

t,C

omm

unity

Car

bon

Pool

s

Faun

aan

dFl

ora

Inte

rnat

iona

lInd

ones

ia(F

FI-I

ndon

esia

)

Com

mer

cial

plan

tatio

ns(o

ilpa

lm),

subs

iste

nce

farm

ing,

illeg

allo

ggin

g,m

inin

g

YY

YY

Y

Cen

tral

Kal

iman

tan

The

Rim

baR

aya

Bio

dive

rsity

Res

erve

Proj

ect

Infin

iteE

arth

PT.R

imba

Ray

aC

onse

rvat

ion

Com

mer

cial

plan

tatio

ns(o

ilpa

lm),

subs

iste

nce

farm

ing,

illeg

allo

ggin

g,pe

atfir

es

YN

NN

N

Cen

tral

Kal

iman

tan

Kat

inga

nC

onse

rvat

ion

Are

a:A

Glo

balP

eatla

ndC

apst

one

Proj

ect

Star

ling

Res

ourc

es/P

T.R

imba

Mak

mur

Uta

ma

(RM

U)

Ille

gall

oggi

ng,c

omm

erci

alpl

anta

tions

(oil

palm

and

rubb

er)

YN

NN

N

Eas

tKal

iman

tan

Ber

auFo

rest

Car

bon

Prog

ram

The

Nat

ure

Con

serv

ancy

(TN

C)

Com

mer

cial

plan

tatio

ns(o

ilpa

lm),

logg

ing,

subs

iste

nce

farm

ing,

min

ing

YN

NN

N

Vie

tnam

Lam

Don

gSN

VSi

te,C

atT

ien

The

Net

herl

ands

Dev

elop

men

tOrg

aniz

atio

n(S

NV

)

Subs

iste

nce

farm

ing,

logg

ing,

infr

astr

uctu

rede

velo

pmen

t

YY

YY

N

aM

itiga

tion

activ

ities

:1=

redu

cing

emis

sion

sfr

omde

fore

stat

ion,

2=

redu

cing

emis

sion

sfr

omfo

rest

degr

adat

ion,

3=

cons

erva

tion

offo

rest

carb

onst

ocks

,4=

sust

aina

ble

man

agem

ent

offo

rest

s,5=

enha

ncem

ento

ffo

rest

carb

onst

ocks

.

4

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Figure 1. Location of participating project sites.

deforestation and degradation, and the context of REDD+ atthe project site. Subsequently, three regional workshops werecarried out to discuss the operationalization of MRV atthe project sites (a workshop was held in Dar es Salaam,Tanzania, for the Tanzania and Cameroon project developers;in Petropolis, Brazil, for the Brazil and Peru projectdevelopers; and in Jakarta, Indonesia, for the Indonesia andVietnam project developers). About 20–25 people attendedeach workshop, including one to two participants from eachproject development organization, and also international MRVexperts.

2.2. Methods

Three broad categories were developed to assess the capacityand readiness of the project developers. Each category wasdivided into three subcategories, and criteria were formulatedto measure each after careful examination of the internationalstandards of carbon certification (such as VCS and Plan Vivo),IPCC and UNFCCC guidelines, and other reference guidessuch as GOFC-GOLD REDD Sourcebook (GOFC-GOLD2012).

(1) Remote sensing and GIS capacity and readiness—thisincludes data coverage at the project sites and the capacityof the developer to produce derivatives from the rawimage data. It is important because the spatial datainfrastructure and availability of remote sensing dataare fundamental to producing estimates of the baseline,and to developing meaningful monitoring plans forlong-term performance assessment of REDD+ projects.This category was subdivided into three subcategories:data, tools and methods, and implementation capacity.Data criteria were availability of medium- and high-resolution datasets; access to other higher-end monitoring

tools such as LiDAR, aerial, SAR and hyperspectraldatasets; and availability of ancillary spatial data layers atdifferent jurisdictional scales. Tools and methods criteriawere availability of data processing software, and thealgorithms used for image classification and changedetection. Implementation capacity indicates the projectdevelopers’ in-house expertise to analyse data.

(2) Carbon pool inventorying capacity and readiness—thisanalyses developers’ plans to measure various carbonpools with respect to IPCC tiers (IPCC 2006), andthe availability of existing emission factor data at theproject sites. The same three subcategories were used.Data criteria were the availability of allometric equations,carbon fraction coefficients and other emission factordatasets. Tools and methods criteria were the samplingscheme and the methods used for the measurement ofvarious carbon pools with respect to IPCC’s AFOLU(agriculture, forestry and other land use) guidelines.Implementation capacity was defined as the professionalexpertise available within the organization to carry outfield inventorying.

(3) Baseline, intervention and monitoring readiness—thiscategory assesses developers’ ability to understandthe drivers of deforestation and incorporate them inbaseline modelling, develop effective intervention plans,and develop monitoring strategies. The criteria weremethods used for baseline modelling, understanding ofthe proximate and underlying drivers of deforestation,intervention plans, and long-term monitoring strategies toevaluate the success of REDD+ interventions.

Figure 2 summarizes the three main categories and theirsubdivisions.

Performance indicators were assigned to each criterion,and an indicator value and score were given to each

5

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Figure 2. Categories used to assess monitoring, reporting and verification readiness.

indicator. For example, the first criterion, availability ofmedium-resolution remote sensing data, was divided intofour performance indicators. If 100% spatial data coveragewas available for more than three points in time (which isimportant for reliable estimation of the deforestation rate andlinear projection of the baseline), that indicator received thehighest score, 3. A score of 2 was awarded for 100% datacoverage in two to three time points, and 1 for those with onetime point.

Similar guidelines were formulated for each indicator,and scores were assigned (see table 2). A total of 19 criteriawith 76 performance indicators under nine subcategories wereassessed. Each subcategory received a score of 0–3, for 9possible points in each major category and 27 possible pointsoverall. Capacity was assessed as very high for a score greaterthan 20, high for a score of 16–20, medium for 11–15, and lowfor 10 or less. This arbitrary scale was defined after analysingthe relative distribution of scores and the performance ofdeveloper organizations with respect to individual criteria.

Further analyses were carried out to check whether theslowness in readiness could be attributed to limitations in thepersonnel expertise and technical capacity available with theproject developer organizations. Firstly, the scores of in-housecapacity (a measure of expertise) for remote sensing and GISwork was compared with the scores of in-house capacity forthe field inventorying (i.e., remote sensing and GIS in-houseexpertise versus carbon pool inventorying in-house expertise)across projects. Second comparison was carried out with theaggregated scores for remote sensing and GIS with respect toaggregated scores for baseline, intervention and monitoring.The former comparison intend for attributing expertise whilethe latter for technical capacity.

3. Results

3.1. Deforestation drivers and mitigation activities

Deforestation and degradation drivers at the REDD+ demon-stration sites involved a complex interplay of commercial

and subsistence activities (table 1). The primary threats inBrazil and Peru were commercial activities such as cattleranching and beef production and mechanized illegal logging;in Tanzania and Cameroon, they were subsistence activitiessuch as farming and charcoal production (which is partiallycommercial); and in Indonesia and Vietnam, they were a mixof commercial and subsistence activities such as oil palmplantations and subsistence farming. Reducing emission fromdeforestation was the primary mitigation activity planned bythe project developers in all three regions. Forest degradationwas reported as a major problem at sixteen project sites.Conservation of carbon stock and sustainable managementof forests have been taken up by 14 project developers.Enhancement of carbon stock was not a major objective formost of the project developers, and only 3 out of 20 wereengaged in this activity.

3.2. Remote sensing and GIS capacity and readiness

Eleven of 20 project developers (55%) showed high orvery high remote sensing and GIS capacity and readiness(figure 3(a)), seven (35%) were ranked medium, and two(10%) were ranked low. Most project developers wereequipped with medium-resolution satellite data, primarilyLandsat series data available free of cost. The other majorsensors included SPOT series, IRS-LISS series, and CBERS.About half of the developers had high-resolution (>10 m)satellite data, in most cases with spatial coverage of morethan 50% of the study area. Eleven used at least one type ofhigher-end remote sensing technology (such as LiDAR, aerialphotos and SAR). SAR data were primarily PALSAR, whichwere freely available. All these organizations have manyyears of experience at their respective sites, and therefore theavailability of ancillary GIS layers at the project, provincialand national scales was not a problem.

Five organizations were well equipped with processingsoftware. Nine had some limitations with respect tocommercial software, but relied on open-source software

6

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Table 2. Detailed breakdown of assessment categories, criteria and indicators.

Category Subcategory Criterion Performance indicatorIndicatorvalue Score

Remote sensing andGIS capacity andreadiness

Data Availability ofmedium-resolution remotesensing data with 100% spatialcoverage of project site at aspecific number of time points

>3 time points Very high 3

2–3 time points High 21 time point Medium 10 time points Low 0

Availability of high-resolutionremote sensing data (<10 m) inat least one time point withpercentage cover

>50% Very high 320–50% High 2<20% Medium 1None Low 0

Access to other higher-endmonitoring tools such as LiDAR,aerial, hyperspectral and SAR

3 or more tools Very high 32 tools High 21 tool Medium 1None Low 0

Availability of GIS data atnational, provincial and projectscales

All 3 scales Very high 32 scales High 21 scale Medium 1None Low 0

Tools andmethods

Availability of remote sensingand GIS processing software(ERDAS, ENVI, ArcGIS,IDRISI and open source)

3 or more programs Very high 32 programs High 21 program Medium 1None Low 0

Use of higher-end techniques forclassification and changeanalysis

Multivariate iterative algorithms,decision tree and otherhigher-end techniques.

Very high 3

Supervised/unsupervisedtechniques andpost-classification changedetection

High 2

Visual Medium 1None/yet to start Low 0

Implementationcapacity

Internal capacity for remotesensing and GIS work

In-house expertise within themain project implementationoffice

Very high 3

In-house expertise atheadquarters with assistancefrom the field office

High 2

such as R and QGIS for spatial analysis. Fourteenorganizations used decision tree or multivariate iterativealgorithms for image classification and change analysis.Eleven organizations showed very good in-house expertisewith the main project implementation office, five partiallyoutsourced their work, and the rest carried out the work atheadquarters with inputs from the regional project office.

Remote sensing and GIS capacity and readiness variedby region (figure 4(a)). Six out of seven projects in Brazil andPeru showed high or very high capacity and readiness, andone was in the medium category. The projects in Tanzaniaand Cameroon showed more variability: two of seven (29%)were rated very high, and the remaining five (71%) were ratedmedium to low. None of the projects in Indonesia and Vietnamshowed very high capacity and readiness in this category;three were rated high, two medium, and one low. Withrespect to high-resolution satellite data and other higher-endmonitoring tools such as LiDAR, aerial photos and SAR, allthree regions experienced limitations; these limitations weresevere in Tanzania and Cameroon. African project capacity

was further weakened by insufficient expertise and lack ofsoftware and processing infrastructure. In contrast, Indonesiaexperienced rapid progress in the acquisition of LiDAR inrecent years together with improvement in data processingalgorithms and facilities.

3.3. Carbon pools inventorying capacity and readiness

Eight (40%) project developers showed high or very highcapacity and readiness, while seven (35%) showed low(figure 3(b)) in this area. Eleven of the sites had site-specificallometric equations, while only five had specific carbonfraction coefficients. The rest of the sites followed globaldefault values available from the IPCC’s AFOLU guidelines.The other emission factor databases such as wood density,biomass expansion factor and root–shoot ratio were alsolimitedly available for most of the sites. Only two projectdevelopers had plans for monitoring all five carbon pools;eight focused on three carbon pools; five measured onlyabove-ground biomass, and the rest had not started carbon

7

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Table 2. (Continued.)

Category Subcategory Criterion Performance indicatorIndicatorvalue Score

Partial outsourcing Medium 1Complete outsourcing Low 0

Ground carbon poolinventoryingcapacity andreadiness

Data Availability of allometricequations

Site specific for most of the landcover types in the project area

Very high 3

Site specific for certain landcover types, global defaultequations for the rest

High 2

Only global default equationsfrom literature

Medium 1

None Low 0Availability of carbon fractioncoefficients

Site specific for most of the landcover types in the project area

Very high 3

Site specific for certain landcover types, global defaultequations for the rest

High 2

Only global default equationsfrom literature

Medium 1

None Low 0Availability of other emissionfactor databases such asroot–shoot ratio, biomassexpansion factor and wooddensity

Site specific for most of the landcover types in the project area

Very high 3

Site specific for certain landcover types, global defaultvalues for the rest

High 2

Only global default values fromliterature

Medium 1

None Low 0Tools andmethods

Monitoring of all five carbonpools (above-ground,below-ground, deadwood, litterand soil organic carbon)

All carbon pools Very high 3

At least 3 carbon pools High 2At least above-ground biomass Medium 1No plans/yet to start Low 0

Sampling scheme and methods Well defined with configurationsof permanent and temporarysampling units

Very high 3

Well defined with onlypermanent sampling unit

High 2

Loosely defined temporarysampling unit

Medium 1

No sampling scheme/yet to start Low 0

Implementationcapacity

Internal capacity for fieldinventorying

In-house expertise within themain project implementationoffice

Very high 3

In-house expertise atheadquarters with assistancefrom the field office

High 2

pool measurements. The AFOLU recommendation to usea mix of permanent and temporary sampling units formonitoring additionality was not followed by any projectdevelopers. Eleven focused only on permanent sampling units,while six had a vague sampling plan that was just enough todevelop a rough estimate of biomass. However, in terms ofin-house expertise, 90% of the project developers showed highor very high capacity to take up the inventorying program withthe main project implementation office.

Region-wise, three projects in Indonesia and Vietnamshowed high or very high capacity and readiness, and theremaining three showed low capacity (figure 4(b)). Theprojects in Brazil and Peru were more or less evenly

distributed among classification categories, while the projectsin Tanzania and Cameroon were predominantly in the mediumcategory. Distribution of allometric equations was similar inall regions: at least 50% of the project areas had site-specificallometric equations. However, the availability of otheremission factor coefficients such as carbon fraction and wooddensity was higher in Latin America, followed by SoutheastAsia and Africa. The number of carbon pools monitored byTanzania project developers was higher than in the othertwo regions. This is probably due to the fact that manyproject sites are less carbon rich, as they are located indry forest areas, and thus more pools must be counted tomake the project economically viable. Projects in Indonesia

8

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Table 2. (Continued.)

Category Subcategory Criterion Performance indicatorIndicatorvalue Score

Partial outsourcing Medium 1Complete outsourcing Low 0

Baseline,intervention andmonitoringreadiness

Baselinemodelling

Methods used for baselineestimate

GIS-based modelling withcovariates of deforestationagents

Very high 3

Simple historic average and/orlinear projection

High 2

Assumption based on expertknowledge

Medium 1

No specific plans/yet to start Low 0Deforestationdrivers andinterventionplans

Understanding of the proximatedrivers of deforestation

Very high degree ofunderstanding established overseveral years

Very high 3

Good degree of understanding High 2Limited experience with the site Medium 1No experience with the site Low 0

REDD+ intervention plans forproximate drivers

Well-defined one-to-oneintervention plans

Very high 3

Reasonably defined interventionplans

High 2

Loosely defined interventionplans

Medium 1

No definite plans Low 0Understanding of the underlyingdrivers of deforestation

Very high degree ofunderstanding established overseveral years

Very high 3

Good degree of understanding High 2Limited experience with the site Medium 1No experience with the site Low 0

REDD+ intervention plans forunderlying drivers

Well-defined one-to-oneintervention plans

Very high 3

Reasonably defined interventionplans

High 2

Loosely defined interventionplans

Medium 1

No definite plans Low 0Monitoringplans

Long-term monitoring plans toevaluate theREDD+ interventions

Well-defined monitoring plans Very high 3

Reasonably defined monitoringplans

High 2

Loosely defined monitoringplans

Medium 1

No definite plans Low 0

for which the assessment is already in the advanced stagesincluded soil organic carbon, since peat land carbon is amajor contributor to the total carbon dynamics of the area. Incontrast, project developers in Brazil and Peru mostly focusedon above-ground biomass alone, as this itself contributes alarge share of the carbon pool and can generate adequatecredits.

3.4. Baseline, intervention and monitoring readiness

None of the projects showed very high readiness for baselinemodelling, intervention plans and monitoring. Nine projects(45%) showed high, five (25%) medium, and six (30%)low readiness (figure 3(c)). Seven project developers used

historical averages and linear projection to develop baselineestimates, while six used GIS-based modelling incorporatingdeforestation agents. One relied on expert knowledge forestimating a baseline.

All project developers had very good understanding ofthe proximate drivers of deforestation at the site, since theyhad worked in the same area for multiple years. However,their intervention strategies varied. Only three organizationshad a well-defined strategy to slow and halt deforestationagents. About half of the organizations defined their strategyreasonably well, while two organizations defined theirstrategy loosely, and five did not have specific strategies oractivities at the time of the survey. Project developers seemedto have a reasonable understanding of underlying drivers of

9

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Figure 3. The MRV capacity and readiness of project developers at a sample of REDD+ projects in Brazil, Peru, Cameroon, Tanzania,Indonesia and Vietnam. (a) Remote sensing and GIS; (b) ground carbon pool inventorying; (c) baseline, intervention and monitoring; and(d) overall MRV readiness. The total number of projects in the sample is 20, and the values in the diagram represent number of projectdevelopers with percentage in parentheses.

deforestation, but they expressed limitations in their abilityto address these drivers, since many originate outside theproject boundary. In such cases, project developers outlinedtheir plans to work with local or regional government bodiesto develop intervention plans.

With respect to monitoring, only 20% of the projectdevelopers had reasonable monitoring plans, while the resthad loosely defined monitoring plans or no plan at all. Thiswas probably due to the fact that the projects are still in theplanning or verification stage and are not yet at the stage ofestablishing a fully functioning MRV system.

Although no project in any region was classified ashaving very high readiness, the projects in Brazil and Perutended to have better in this category than the other fourcountries (figure 4(c)). The lowest readiness was observedamong Southeast Asian projects, while most African projectsranked either high or low with a negligible percentagein the medium category. A higher-end future deforestationmodelling capacity (five out of seven) was clearly visibleamong Latin American projects for developing baselinemodelling; whereas the modelling capacity was largelylimited in the other regions (none in Africa, and one outof six in Southeast Asia). Progress with intervention planswas greater in Brazil and Peru, followed by Tanzania andCameroon and then Indonesia and Vietnam. This implies thatAfrica project developers have good understanding of theground conditions and can make well-defined intervention

plans although they have limited technical capacity. Withrespect to monitoring plans, projects in Brazil and Peru didnot show a clear difference although these countries were inthe forefront for many of the other activities.

3.5. Overall MRV capacity and readiness

In an aggregation of the three main categories, onlytwo projects (10%) had very high capacity and readiness(figure 3(d)). Nine (45%) had high, six (30%) medium, andthree (15%) low capacity. Considered by region, six out ofseven Latin American projects showed high or very highcapacity; no project in Africa or Southeast Asia had veryhigh capacity (figure 4(d)). Two out of seven African projectshad low capacity; three out of six Southeast Asian projectshad medium capacity. In general, the projects in Braziland Peru showed the highest capacity and readiness. Thosein Cameroon, Tanzania, Indonesia, and Vietnam showedcapacity that is on average lower than the two Latin Americancountries.

The comparative analysis of in-house capacities forremote sensing, GIS and field inventorying, and technicalcapacities for baseline, intervention and monitoring acrossprojects indicate that expertise and technical limitationswere not the primary reasons for the slow progress towardsreadiness. In-house capacities to carry out both remote sensingand GIS works, and carbon pools inventorying ranked high

10

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Figure 4. Region-wise capacity and readiness of project developers at a sample of REDD+ projects in Brazil, Peru, Cameroon, Tanzania,Indonesia and Vietnam. (a) Remote sensing and GIS; (b) ground carbon pool inventorying; (c) baseline, intervention and monitoring; and(d) overall MRV readiness. (The values within the column represent number of projects.)

or very high in 14 of the 20 projects (70%) (table 3).Four projects (20%) (two each in Africa and SoutheastAsia) had very high field inventorying expertise but partiallyoutsourced their remote sensing and GIS work. One projecthad high remote sensing and GIS capacity but low carbonpool inventorying capacity, and one was ranked mediumfor both categories. With respect to technical capacities forbaseline, intervention and monitoring readiness, five projectsshowed very high remote sensing and GIS capacity whilenone showed very high baseline, intervention and monitoringreadiness, suggesting that slow progress in the latter categorywas not solely due to limitations in the former category(table 4).

4. Discussion and recommendations

Performance indicators need to be identified at each phaseof REDD+ implementation to measure its success (Wertz-Kanounnikoff and McNeill 2012). The present study, whichaddresses the accounting framework of REDD+ projects inthe readiness phase, developed 19 performance criteria and 76indicators to measure MRV capacity and readiness. The studyfound that many projects are still working towards variouscategories of MRV capacity and need additional momentumto move forward to the next phases. The critical issues inthe accounting framework could be sorted into the followingcategories.

Table 3. Comparison of in-house project personnel expertise tocarry out remote sensing, GIS and carbon pool inventorying worksat a sample of REDD+ projects in Brazil, Peru, Cameroon,Tanzania, Indonesia and Vietnam. The upper left columns indicatehigh degree of expertise (70%) within the organizations.

Carbon pool inventoryingin-house expertise

Veryhigh High Medium Low Sum

Remote sensing andGIS in-houseexpertise

Veryhigh

6 5 0 0 11

High 2 1 0 1 4Medium 4 0 1 0 5Low 0 0 0 0 0

Sum 12 6 1 1 20

Data availability and methodological guidelines—theREDD+ projects are located in the tropical belt wherea number of prevalent atmospheric and ground conditionssuch as cloud cover, haze and uneven topography oftendisrupts a satellite sensor’s capacity to image good qualityground data (Joseph et al 2013). The access to time seriesremote sensing data was serious problem in most of theprojects in our study. This data limitation problem havebeen reported widely; including the world’s first verifiedREDD project, Noel Kempff Mercado Climate Action Project

11

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Table 4. Comparison of remote sensing and GIS capacity withbaseline, intervention and monitoring readiness at a sample ofREDD+ projects in Brazil, Peru, Cameroon, Tanzania, Indonesiaand Vietnam.

Baseline, intervention andmonitoring readiness

Veryhigh High Medium Low Sum

Remote sensingand GIScapacity

Veryhigh

0 4 1 0 5

High 0 2 3 1 6Medium 0 2 2 3 7Low 0 0 0 2 2

Sum 0 8 6 6 20

in Bolivia, where lack of time series data at the rightscale was a major constraint in generating the accurate andprecise estimates of deforestation and spatial modelling ofbaseline (Brown 2005). Although operational monitoring ofdeforestation is reasonably possible with the coarse resolutionremote sensing data as evident from INPE’s PRODESprogram (INPE 2011), establishing such a scheme usingmedium-resolution data still underway. Recent initiatives suchas mapping of annual deforestation rate using Landsat datain the Congo basin (Potapov et al 2012) and Sumatra Island(Margono et al 2012) are notable in this context. Moreimportantly, such system should make use of the pixel-basedtime series data compositing to remove the noises fromcloud, haze and other atmospheric impurities. This involvesthe implementation of number of higher-end computationalgorithms; for example: LEDAPS (Landsat EcosystemDisturbance Adaptive Processing System) for atmosphericcorrection (Masek et al 2006), Fmask (Function of mask)for cloud and cloud shadow masking (Zhu and Woodcock2012), and change break detection using BFAST (breaks foradditive seasonal and trend) (Verbesselt et al 2010a, 2010b).The spatial infrastructure and technical expertise are criticalcomponents in the implementation of such algorithms (Josephet al 2013).

Similarly ambiguity exists in methodological guidelineson how to set reference emission level in REDD+ projects(Angelsen and Verchot 2009, Dutschke 2013). The commonand easiest method is simple linear regression based onhistorical deforestation rate. Verified Carbon Standard alsorecommend the use logistical regression to overcome theproblems of linear regression (Shoch et al 2011). Other thanthese two, complex GIS-based predictive land cover modelsare also suggested for estimating the REL. Cortez et al(2010) proposed a double set of RELs at the project level;an absolute REL and a proportional REL to the respectivereference region; both could be independently validated. Thisis important for upscaling the REL to the subnational andnational scales through a nested approach (Pedroni et al2009), ensuring no displacement in the emission, and alsoto avoid the potential double counting issues. The overallchallenge is to make sure that the sum of all projects andsubnational emission reductions does not exceed the totalclaimed at national scale (Dutschke 2013).

Monitoring cost and choices—the earlier studies that focusedat the national scale observed critical gaps forest monitoringcapacities (Romijn et al 2012), largely attributed to the lackof technical capacity and expertise. The results of this study,which was focused at the project rather than national scale,indicate that technical and expertise limitations are not thesole reasons for the slow progress towards readiness. Of the20 projects, 18 had high or very high in-house expertise forcarbon pool inventorying, but only 15 reached these levelsfor remote sensing and GIS (table 3). The difference isinteresting, although not large, because at the country scalethese values are commonly reversed (Herold 2009, Romijnet al 2012). This implies that projects are better at producinglocal measurements of carbon stocks but rely more on externaltechnological expertise for remote sensing and GIS.

A national nesting protocol may be needed to addressthe monitoring choices (Cortez et al 2010). Monitoringchoices should be decided based on a cost-benefit analysiswhich is dependent on the size of the project and theemission reductions and removals expected. Bottcher et al(2009) compile the monitoring cost of different techniqueswhich ranges from $0.5 to $550 km−2 depending on therequired precision of carbon stock and area change detection.The cost of medium-resolution satellite data ranges from$0.001 to $0.5 km−2 (except for Landsat and CBERS)while high-resolution data are in the range of $2–30 km−2

(GOFC-GOLD 2012). Bottcher et al (2009) also argue thatsingle country or project solutions will face relatively highermonitoring costs and recommend an international cooperationto overcome initialization costs and unequal access tomonitoring technologies. Hardcastle and Baird (2008) studiedthe monitoring cost in 25 tropical countries and recommendedthe establishment of shared regional capacity, as mostnon-data costs are largely independent of the geographicalextent of the monitored area. Other advantages of such sharedsystem would include standardization of methodologies forcarbon accounting and reporting, facilitating a regional ratherthan national integration with carbon markets, and perhapsreducing transaction costs (Hardcastle and Baird 2008).

International policy dialogues—perhaps the single mostprominent factor determining the progress of the projects isthe pace of the international REDD+ negotiation progressingat the UNFCCC. Although REDD+ started with ambitiousplans and a fast pace in 2007, the momentum has slowednoticeably after the Copenhagen Summit. The economicdownturn in 2008 led many countries to withdraw fromtheir original ambitious funding plans (Angelsen and McNeill2012). The announcement of a Green Climate Fund inDurban stimulated enthusiasm, but the time frame put forthfor operationalizing market mechanisms (by 2020) was notpromising. This created a vacuum and an unclear path ofdevelopment in the forest carbon market. The projectedgrowth of the forest carbon market has shown a considerabledecrease in the last three years (Peters-Stanley and Hamilton2012). In addition, market fluctuations and the price collapseof clean development mechanism (CDM) credits—anotherUNFCCC initiative for emission reduction—made investorsdoubtful about the voluntary carbon market. Perhaps, these

12

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

uncertainties together with the deficiency in the human andtechnical expertise might have resulted a general slowness inthe prospects of MRV development in the REDD+ projects.Although this was the case in private and multilateral funding,bilateral investment has increased in recent years, especiallyfrom the development assistance fund (Wertz-Kanounnikoffand McNeill 2012, Streck 2012).

RecommendationsFurther research and investment are needed to enable

REDD+ to move forward to the operational (‘results-basedaction’) phase. Although the sample is too small to allowdefinitive conclusions, some tentative recommendations canbe made.

• Improve remote sensing data access. Landsat seriesdata are the primary source of information for theproject developers. However, the availability of timeseries data was a limitation for many of the projectsites (especially African sites). This limitation is moresevere for high-resolution satellite data and other high-endmonitoring tools like LiDAR and aerial photos. Possibleways to address this would be the establishment of anEarth Observation Data Directory for REDD+ projectsby international donors, and other key stakeholderorganizations.

• Develop emission factor data. Although allometricequations are well developed in the Amazon, there is littleunderstanding of the carbon content of living biomass inthese areas. Data on wood density, the biomass expansionfactor and the root–shoot ratio were also limited at theAmazon project sites. Indonesia is a unique case inwhich peatland carbon plays an important role in the totalcarbon dynamics (Hergoualc’h and Verchot 2011). Datadeficiency, unless addressed with a substantial investment,will make it difficult for Indonesia to progress to theoperational phase of REDD+ (Verchot et al 2012). Thisdata deficiency problem is particularly severe in theinitial stages of REDD+ implementation. Once allometricequations are developed for each bioclimatic region andhave precise biomass values for each stratum, this problemwould be less important for future REDD+ projects.

• Account all relevant carbon pools. The dry forestproject sites in Tanzania are less carbon rich comparedto tropical wet forest sites and need to combine allsignificant carbon pools to make the project attractive forinvestors. Methodologically, projects focusing on avoideddeforestation might not attract large investment becauserates of deforestation is low in these landscapes. In suchsituations, mitigation activity focused reducing emissionsand increasing removals might be more suitable (Skutschet al 2009, Skutsch and Ba 2010). The forest enhancementactivities could also generate additional job opportunitiesand income for local communities and ensure more activeparticipation in the participatory MRV process. It is alsoimportant to combine bilateral, multilateral and otherinternational funds in addition to the carbon market tomake projects economically viable.

• Advance method development. Although degradation wasreported as an issue at 16 of the sites, its monitoringreceived less attention, largely attributed to methodologicaland technical limitations. The spatial and temporal scaleof various data products and their integration was anotherchallenging task in larger projects.

• Collaborate to build capacity. The software and dataprocessing capacity of project developers varies con-siderably. The level of technical expertise is a sourceof concern in many projects. Capacity developmentthrough collaboration should be encouraged within regionsand complemented by south–south cooperation technicalexchange. Also, it is important to organize more capacitybuilding training for development of expertise in dataprocessing.

• Coordinate efforts to slow, halt and reverse deforestationat the jurisdictional scale. Focusing on REDD+ creditingwithin project boundaries alone may shift deforestationactivities to neighbouring areas outside the project, makingnet change negligible. Project developers said some ofthe underlying drivers originate from outside their projectboundaries and are therefore under the purview of local,provincial or national governments. Therefore, the nextstep in the investment process should focus on nestingREDD+ projects and other efforts to slow and haltdeforestation within broader efforts to stop deforestationand degradation at higher jurisdictional scales.

• Ensure continued funding. Many project developersexpressed concern about continued availability of fundingafter the present project period. Assured project funding,or the establishment of a fund to support the efforts ofREDD+ projects might be necessary.

5. Conclusion

This study found that capacity and readiness tended to behighest in the projects in Brazil and Peru and somewhat lowerin Cameroon, Tanzania, Indonesia and Vietnam. Capacityand readiness also varied with respect to different MRVcategories. Remote sensing and GIS capacity was highestin Brazilian and Peruvian projects. With respect to carbonpool inventorying, the projects in Brazil and Peru focusedmostly on above-ground biomass; Tanzania combined asmany carbon pools as possible; and Indonesia focusedprimarily on above-ground, below-ground and peat carbon.On baseline modelling, intervention and monitoring, none ofthe projects showed very high capacity. Although the overallMRV capacity and readiness of half the projects is high,there are capacity deficiencies in other projects. This couldbe attributed not only to technical and expertise limitationsbut also to the general slowness of REDD+ progress ininternational negotiations and the unclear path of developmentof the forest carbon market. Substantial investment is requiredto make REDD+ operational at the project scale, and morechallenges are likely for monitoring, reporting and verificationat larger scales including jurisdictional counting.

13

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Acknowledgments

This study is part of a larger initiative, the Global ComparativeStudy on REDD+, undertaken by the Center for InternationalForestry Research with funding support from NORAD(Grant Agreement no. QZA-10/0468) and AusAID (GrantAgreement no. 46167). The authors would like to thankthe national coordinators of this study, Abdon Awono, AmyDuchelle, Daju Pradnja Resosudarmo, Demetrius Kweka,and Thu Ba Huynh, and other field researchers who helpedwith the survey and fieldwork. The study also drew oninsights developed during three regional workshops, and thecontribution of each participant is duly acknowledged. Theauthors also wish to thank Manuel Estrada, who played anindispensable role in organizing the workshops. Commentsby two anonymous reviewers and the editorial board memberhave substantially enhanced the quality of the paper.

References

Angelsen A, Brockhaus M, Sunderlin W D and Verchot L V 2012Analysing REDD+: Challenges and Choices (Bogor: Centerfor International Forestry Research)

Angelsen A and McNeill D 2012 The evolution of REDD+Analysing REDD+: Challenges and Choices ed A Angelsen,M Brockhaus, W D Sunderlin and L V Verchot (Bogor: Centerfor International Forestry Research) pp 31–48

Angelsen A and Verchot L V 2009 Monitoring and establishingreference levels Realising REDD+: National Strategies andPolicy Options ed A Angelsen (Bogor: Center for InternationalForestry Research) pp 90–1

Bottcher H, Eisbrenner K, Fritz S, Kindermann G, Kraxner F,McCallum I and Obersteiner M 2009 An assessment ofmonitoring requirements and costs of ‘reduced emissions fromdeforestation and degradation’ Carbon Balance Manag. 4 7

Brown S 2005 Spatial Modeling of the Averted DeforestationBaseline for the Noel Kempff Mercado Climate Action ProjectBolivia (Arlington, VA: Winrock International)

Cerbu G A, Swallow B M and Thompson D Y 2011 LocatingREDD: a global survey and analysis of REDD readiness anddemonstration activities Environ. Sci. Policy 14 168–80

Cortez R, Saines R, Griscom B, Martin M, De Deo D, Fishbein G,Kerkering J and Marsh D 2010 A Nested Approach to REDD+:Structuring Effective and Transparent Incentive Mechanismsfor REDD+ Implementation at Multiple Scales (Arlington,VA: The Nature Conservancy)

De Sy V, Herold M, Achard F, Asner G P, Held A, Kellndorfer Jand Verbesselt J 2012 Synergies of multiple remote sensingdata sources for REDD+ monitoring Curr. Opin. Environ.Sustain. 4 696–706

Dutschke M 2013 Key Issues in REDD+ Verification (Bogor:Center for International Forestry Research)

Estrada M and Joseph S 2012 Baselines and monitoring in localREDD+ projects Analysing REDD+: Challenges and Choicesed A Angelsen, M Brockhaus, W D Sunderlin andL V Verchot (Bogor: Center for International ForestryResearch) pp 247–60

Fry B P 2011 Community forest monitoring in REDD+: the ‘M’ inMRV? Environ. Sci. Policy 14 181–7

Gibbs H K, Brown S, Niles John O and Foley Jonathan A 2007Monitoring and estimating tropical forest carbon stocks:making REDD a reality Environ. Res. Lett. 2 045023

Goetz S and Dubayah R 2011 Advances in remote sensingtechnology and implications for measuring and monitoringforest carbon stocks and change Carbon Manag. 2 231–44

GOFC-GOLD 2012 A sourcebook of methods and procedures formonitoring and reporting anthropogenic greenhouse gasemissions and removals caused by deforestation, gains andlosses of carbon stocks in forests remaining forests, andforestation GOFC-GOLD Report Version COP18-1(Wageningen: GOFC-GOLD Land Cover Project Office,Wageningen University) (www.gofcgold.wur.nl/redd/sourcebook/GOFC-GOLD Sourcebook.pdf, last accessed 21December 2012)

Hardcastle P D and Baird D 2008 Capability and cost assessment ofthe major forest nations to measure and monitor their forestcarbon Report Prepared for the Office of Climate Change(Penicuik: LTS International Ltd)

Hergoualc’h K and Verchot L V 2011 Stocks and fluxes of carbonassociated with land use change in Southeast Asian tropicalpeatlands: a review Glob. Biogeochem. Cycles 25 GB2001

Herold M 2009 An assessment of national forest monitoringcapabilities in tropical non-Annex I countries:recommendations for capacity building Report for the Prince’sRainforests Project and the Government of Norway (Jena:Friedrich–Schiller-Universitat Jena and Global Observation ofForest and Land Cover Dynamics Land Cover Project Office)

Herold M and Skutsch M 2011 Monitoring, reporting andverification for national REDD+ programmes: two proposalsEnviron. Res. Lett. 6 014002

Hosonuma N, Herold M, De Sy V, De Fries R S, Brockhaus M,Verchot L, Angelsen A and Romijn E 2012 An assessment ofdeforestation and forest degradation drivers in developingcountries Environ. Res. Lett. 7 044009

INPE 2011 Monitoramento da Floresta Amazonica Brasileira porSatelite (Sao Jose dos Campos: Instituto Nacional de PesquisasEspaciais) (www.obt.inpe.br/prodes/prodes 1988 2010.htm,10 August 2013)

IPCC 2003 Good Practice Guidance for Land Use, Land-UseChange and Forestry (GPG-LULUCF) ed J Penman,M Gytarsky, T Krug, D Kruger, R Pipatti, L Buendia, K Miwa,T Ngara, K Tanabe and F Wagner (Kanagawa: IPCC-IGES)(www.ipcc-nggip.iges.or.jp/public/gpglulucf/gpglulucf.html,last accessed 8 September 2012)

IPCC 2006 IPCC Guidelines for National Greenhouse GasInventories vol 4 (www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html, last accessed 8 September 2012)

Joseph S, Herold M, Sunderlin W D and Verchot L V 2013Challenges in operationalizing remote sensing in climatechange mitigation projects in developing countries IGARSS:2013 IEEE Int. Geoscience and Remote Sensing Symp.pp 2752–5

Joseph S, Murthy M S R and Thomas A P 2011 The progress onremote sensing technology in identifying tropical forestdegradation: a synthesis of the present knowledge and futureperspectives Environ. Earth Sci. 64 731–41

Kshatriya M, Sills E O and Lin L 2012 Global Database ofREDD+ and Other Forest Carbon Projects: Interactive Map(www.forestsclimatechange.org/redd-map/, last accessed 15January 2013)

Larrazabal A, McCall M K, Mwampamba T H and Skutsch M 2012The role of community carbon monitoring for REDD+: areview of experiences Curr. Opin. Environ. Sustainabil.4 707–16

Lotsch A 2012 REDD+ readiness preparation under the forestcarbon partnership facility Capacity Development in NationalForest Monitoring: Experiences and Progress for REDD+ed B Mora, M Herold, V De Sy, A Wijaya, L V Verchot andJ Penman (Bogor: Center for International Forestry Research)

Margono B A, Turubanova S, Zhuravleva I, Potapov P,Tyukavina A, Baccini A, Goetz S and Hansen M C 2012Mapping and monitoring deforestation and forest degradationin Sumatra (Indonesia) using Landsat time series data setsfrom 1990 to 2010 Environ. Res. Lett. 7 034010

14

Environ. Res. Lett. 8 (2013) 034038 S Joseph et al

Masek J G, Vermote E F, Saleous N E, Wolfe R, Hall F G,Huemmrich K F, Feng G, Kutler J and Teng-Kui L 2006 ALandsat surface reflectance dataset for North America,1990–2000 IEEE Geosci. Remote Sens. Lett. 3 68–72

Obersteiner M et al 2009 On fair, effective and efficient REDDmechanism design Carbon Balance Manag. 4 1–11

Pedroni L, Dutschke M, Streck C and Porrua M E 2009 Creatingincentives for avoiding further deforestation: the nestedapproach Clim. Policy 9 207–20

Peters-Stanley M and Hamilton K 2012 Developing dimension:state of the voluntary carbon markets 2012 Report byEcosystem Marketplace and Bloomberg New Energy Finance(www.forest-trends.org/publication details.php?publicationID=3164, last accessed 15 September 2012)

Potapov P V, Turubanova S A, Hansen M C, Adusei B, Broich M,Altstatt A, Mane L and Justice C O 2012 Quantifying forestcover loss in Democratic Republic of the Congo, 2000–2010,with Landsat ETM+ data Remote Sens. Environ. 122 106–16

Pratihast A K, Herold M, De Sy V, Murdiyarso D and Skutsch M2013 Linking community-based and national REDD+monitoring: a review of the potential Carbon Manag. 4 91–104

Romijn E, Herold M, Kooistra L, Murdiyarso D and Verchot L V2012 Assessing capacities of non-Annex I countries fornational forest monitoring in the context of REDD+ Environ.Sci. Policy 19/20 33–48

Shoch D, Eaton J and Settelmyer S 2011 Project Developer’sGuidebook to VCS REDD Methodologies. Version1.0/November 2011 (Arlington, VA: ConservationInternational) (www.conservation.org/about/centers programs/carbon fund/Documents/project developers guide to vcsredd 11282011FINAL.pdf, last accessed 30 November 2012)

Skutsch M and Ba L 2010 Crediting carbon in dry forests: thepotential for community forest management in West AfricaForest Policy Econ. 12 264–70

Skutsch M, McCall M K and Lovett J C 2009 Carbon emissions:dry forests may be easier to manage Nature 462 567

Streck C 2012 Financing REDD+: matching needs and ends Curr.Opin. Environ. Sustainabil. 4 628–37

Sunderlin W D, Larson A M, Duchelle A, Sills E O, Luttrell C,Jagger P, Pattanayak S, Cronkleton P and Ekaputri A D 2010Technical Guidelines for Research on REDD+ Project Sites

with Survey Instruments and Code Book (Bogor: Center forInternational Forestry Research)

van der Eynden M 2012 Measuring, reporting and verifyingREDD+: experiences from the Government of Norway’sInternational climate and forest initiative CapacityDevelopment in National Forest Monitoring: Experiences andProgress for REDD+ ed B Mora, M Herold, V De Sy,A Wijaya, L V Verchot and J Penman (Bogor: Center forInternational Forestry Research)

Verbesselt J, Hyndman R, Newnham G and Culvenor D 2010aDetecting trend and seasonal changes in satellite image timeseries Remote Sens. Environ. 114 106–15

Verbesselt J, Hyndman R, Zeileis A and Culvenor D 2010bPhenological change detection while accounting for abrupt andgradual trends in satellite image time series Remote Sens.Environ. 114 2970–80

Verchot L V, Anitha K, Romijn E, Herold M and Hergoualc’h K2012 Emissions factors: converting land use change to CO2estimates Analysing REDD+: Challenges and Choicesed A Angelsen, M Brockhaus, W D Sunderlin andL V Verchot (Bogor: Center for International ForestryResearch) pp 261–78

Wertz-Kanounnikoff S and McNeill D 2012 Performance indicatorsand REDD+ implementation Analysing REDD+: Challengesand Choices ed A Angelsen, M Brockhaus, W D Sunderlin andL V Verchot (Bogor: Center for International ForestryResearch) pp 247–60

UNFCCC 2007 Decision 2/CP.13: Reducing Emissions fromDeforestation in Developing Countries: Approaches toStimulate Action (Bonn: United Nations FrameworkConvention on Climate Change) (http://unfccc.int/resource/docs/2007/cop13/eng/06a01.pdf, last accessed 16 June 2013)

UNFCCC 2011 Report of the Conf. of the Parties on its 16thSession, Held in Cancun from 29 November to 10 December2010, Addendum: Part Two: Action Taken by the Conf. of theParties at its 16th Session (Bonn: United Nations FrameworkConvention on Climate Change) (http://unfccc.int/resource/docs/2010/cop16/eng/07a01.pdf, last accessed 19 November2012)

Zhu Z and Woodcock C E 2012 Object-based cloud and cloudshadow detection in Landsat imagery Remote Sens. Environ.118 83–94

15