PES and Food Security. Agroforestry Rubber

8/3/2019 PES and Food Security. Agroforestry Rubber

1/26

P A Y N I N T SI C O $ 1 , 5 1 1 . 1 U 1 Y K f l . a dS I C U I I T Ypaymentsfor

ecosystem services andood security


2/26


3/26

1

F o o d a n d A g r i c u l t u r e O r g a n i z a t i o n o t h e U n i t e d N a t i o n s | 2 0 1 1

paymentsforecosystem services and

ood security


4/26

The designations emploed and the presentation o material in this inormation product do not impl the expression o

an opinion whatsoever on the part o the Food and Agriculture Organization o the United Nations (FAO) concerning the

legal or development status o an countr, territor, cit or area or o its authorities, or concerning the delimitation o

its rontiers or boundaries. The mention o specic companies or products o manuacturers, whether or not these have

been patented, does not impl that these have been endorsed or recommended b FAO in preerence to others o a similar

nature that are not mentioned.

ISBN 978-92-5-106796-3

All rights reserved. FAO encourages reproduction and dissemination o material in this inormation product. Non-commercial

uses will be authorized ree o charge, upon request. Reproduction or resale or other commercial purposes, including

educational purposes, ma incur ees.

Applications or permission to reproduce or disseminate FAO copright materials, and all queries concerning rights and

licences, should be addressed

b e-mail to

[email protected]

or to

the Chie, Publishing Polic and Support Branch,

Oce o Knowledge Exchange, Research and Extension, FAO, Viale delle Terme di Caracalla,

00153 Rome, Ital.

ao 2011

e n

This publication was edited b Daniela Ottaviani and Nadia El-Hage Scialabba, Natural Resources

Management and Environment Department, the Food and Agriculture Organization (FAO) o the United

Nations. The editors wish to thank all the authors who prepared the dierent chapters and sections

and those who provided cartographic and photographic materials, as well as Patricia Balvanera,

Rebecca Goldman, Tara Moreau, Nomi Nemes, Am Rosenthal or their valuable suggestions to specic

sections o this publication. Special thanks go to Bran Hugill or language editing and to Pietro

Bartoleschi and Arianna Guida or the design and desktop publishing o this publication. The FAO is

grateul or the generous support o the Swiss Federal Oce or Agriculture (FOAG) to this project.


5/26

Case Study4

ruBBer aGroorestry and pes or preservation o

Biodiversity in BunGo district, sumatraLaxman Joshi

1, Rachman Pasha

2, Elok Mulyoutami

2and Hendrien Beukema

3

1International Centre or Integrated Mountain Development (ICIMOD), Kathmandu, Nepal

2World Agroorestr Centre (ICRAF), Nairobi, Kena

3Universit o Groningen, Groningen, The Netherlands


6/26

1 1 51 1 5

RUBBER AGROFORESTRy AND PES FORPRESERVATION OF BIODIVERSITy IN

BUNGO DISTRICT, SUMATRA

The introduction o the rubber tree (Hevea brasiliensis), naturall ound in the foodplains orests

along the Amazon River, began in Indonesia in the second hal o the 19th

centur. In Sumatra

and Borneo, rubber cultivation, initiall restricted along rivers with good accessibilit, rapidl

spread to even relativel remote areas in the countr. Currentl, Indonesia is the worlds second

largest gum exporter with an overall rubber area o 3.5 million hectares. More than one million

households depend on rubber-generating income in Indonesia, as 83 percent o the rubber

cultivation area is constituted b smallholder rubber agroorestr sstems (Wibawa et al., 2005).

Bungo district, located in the western area o the Jambi Province, the third most important

Indonesian province or rubber production, is surrounded b three national parks: Kerinci

Seblat, Bukit Dua Belas and Bukit Tiga Puluh. The district has been severel deorested (60

percent orest loss) and orests have been replaced b rubber and oil palm plantations, as well

as other agricultural land uses. In particular, rom the late 1980s, an increased spread in oil

plantation cultivation has led to the additional loss o native trees and simplication o the

agro-ecological landscape (Fentreine et al., 2010). A remote sensing stud showed that in 1998

the remaining orests, mostl located on the Barisan range, covered onl 28 percent o Bungo

district, while in the area occupied b jungle rubber has decreased rom 17 percent (1988) to

11 percent (2008) due to a parallel increase in monoculture covering rom 23 percent (1988)

to 49 percent (2008) o the district area (Ekadinata et al., 2010) (Figure 12 and 13).

In Bungo district, rubber is cultivated in monoculture sstems, as well as in more complex

rubber agroorestr sstems. A rubber agroorest usuall starts rom slashing a orest plot (eitherprimar or secondar orest) or an old rubber garden, ollowed b burning the elled trees during

the dr season. For the rst one to two ears, rubber seedlings are grown with rice and other

annual crops. When the rubber trees begin to shade annual crops, the plots are let allow and

the native vegetation regenerates. Non-rubber trees are regularl removed or kept below the

level o rubber trees and periodic weeding is done around the rubber saplings. The rubber trees

reach maturit in seven to ten ears, at which time the armers begin tapping (Joshi et al.,

a S i a

i N d o N E S i a

South chaSa

philippinesmalaYsia

australia

Jakartapj

indiAn OceAn

PAcific

O c e A n

indiAn OceAn

PAcific

O c e A n


7/26

1 1 61 1 6

Case Study4

A.

Ayat/ICRAF

A.

Ayat/ICRAF

A.

Ayat/ICRAF

Adapted rom original map b Andree Ekadinata (ICRAF)

L E G E N D

Oilpalm plantation

Rice padd

Shrub

Settlement

Water bod

Forest

Rubber orest

Rubber plantation

Figure 12

l Bg 1988

120 0 S

140 0 S

10140 0 E 102200 E10200 0 E

K i l o m e t r e s

30 450 6015

n


8/26

t

1 1 71 1 7



Current pages (from left to right):

>Surroundings o Lubuk Beringin, the frst village granted with the legalright (hutan desa) by the Indonesian Government to manage state orests

or their own prosperity.>View o the orested area designated or community orestry permits,which could help meet orest management targets and livelihood interests olocal villages.

>Rubber jungle, a traditional agroorestry practice that mixes jungleplants among rubber trees.

>Example o jungle rubber bordering a rice paddy.

L.

Joshi

Adapted rom original map b Andree Ekadinata (ICRAF)

Figure 13

l Bg 2008

120 0 S

140 0 S

10140 0 E 102200 E10200 0 E

KILOMETRES

L E G E N D

Oilpalm plantation

Rice padd

Shrub

Settlement

Water bod

Forest

Rubber orest

Rubber plantation

K i l o m e t r e s

300 604515

n


9/26

1 1 81 1 8

2003; Wibawa et al., 2005). These traditional rubber gardens are complex in structure. Graduall

over time rubber trees die due to natural causes and other native species begin to become more

dominant. The latex productivit in these gardens, thus, graduall declines. About 25-40 ears

ater planting, when tapping is no longer economical, all the trees are elled and the plot is

cleared or replanting. However, some armers plant rubber seedlings in the gaps caused b

the death o rubber and non-rubber trees; this gap-planting, locall known assisipan, leads to

unevenl aged rubber trees when carried out over multiple ears. The rubber productivit period

can be prolonged using thesisipan technique, but thesisipan plots are never as productive as

normal rubber gardens. Compared to slash-and-burn, however, thesisipan practice is less labour

intensive and does not require much capital investment. It also allows a reduced but continuous

income rom the plot (Joshi et al., 2002; Wibawa et al., 2005); hence, it is practised mostl

b poor armers in less accessible areas. The biodiversit inside suchsisipan plots is normall

ver high, comparable to surrounding orests both in structure and unction as large trees and

naturall regenerating vegetation is retained in the plots. These plots become ver complex

rubber agroorests that are oten reerred to as jungle rubber.

In 2004, ICRAF initiated a PES pilot project in Bungo district (Jambi province) to develop a

reward mechanism in order to conserve the rich biodiversit inside the complex rubber agroorests.

In general terms, quantiing biodiversit in jungle rubber is methodologicall quite challenging

as the potential occurrence o man conounding variables and the high variabilit ound amongst

jungle rubber gardens would require a large number o sampling units. In act, in the Jambiregion, rubber cultivation is composed o a mosaic o small jungle rubber gardens at dierent

development stages, rubber densities and management practices. Potential actors that infuence

the species number ( diversit) and the rate o change in species composition ( diversit)

are the plot size, the histor and management o the plot and the surrounding landscape, the

geographic location o the jungle rubber garden, the elevation, and the adjacenc to orest

remnants, to other rubber jungles or the infuence o an agricultural matrix (Beukema et al.,

Case Study4

A.

Ayat/ICRAF

A.

Ayat/ICRAF

A.

Ayat/ICRAF


10/26

1 1 91 1 9

2007; Wibawa et al., 2005). In addition, extensive biodiversit surves in tropical ecosstems

are ver challenging due to the high densit o species (e.g. 100 vascular plant species in

0.02 ha o jungle rubber) and the dicult and time-consuming task o species identication

(Gillison et al., 2000b).

A stud o the available published and unpublished investigations conducted in the 1990s on

anddiversit recorded in primar orest, jungle rubber and rubber monoculture plantationsrevealed that jungle rubber had a much lower number o epiphtic pteridophte and tree species,

a similar number o bird species, and a higher number o terrestrial pteridophte species than

primar orest (Beukema et al., 2007). The lower number o epiphtic pteridophte species ma

be due to the act that man epiphtes depend on later successional stages o orest and ma

not have had enough time to establish and reproduce. Thus, or some species, even a 40-ear-

old jungle rubber garden might be too oung to serve as a suitable habitat.

The lower richness o tree species recorded in jungle rubber (Figure 14) ma also be explained

b the act that jungle rubber is a tpe o secondar orest, where late-successional tree species

ma not have established et. Selective species removal b the armer is another important actor.

Although the total number o bird species in jungle rubber and primar orest (Figure 15) was

similar, the number o orest-specialist birds was much lower in jungle rubber.

The same was true or terrestrial pteridophtes (Figure 16): or a subset o orest species,

the number o species ound was much lower in jungle rubber than in primar orest (Beukema

et al. 2007).RUPES also carried out rapid biodiversit assessments in Bungo district and ound that o

a total o 971 tree species recorded inside jungle rubber gardens (77 analsed plots), 376 tree

species were ound both in jungle rubber gardens and natural orest patches (31 analsed plots).

Complex rubber agroorests also harbour a air number o mammals species (n=37) compared

to the number ound in the surrounding national parks (n=85). O these 37 mammals species,

nine are endangered species under CITES criteria (ICRAF, n.d.).



Current pages (from left to right):

>The economic boom in palm oil since the 1980s has seen millionso hectares o community orests in Sumatra converted into oil palm

plantations.>Oil palm is much more proftable or smallholders than rice productionand is highly competitive with rubber.

> In Bungo, rubber cultivation is done in a mosaic o small rubber jungleplots interspersed with other crop felds, such as rice paddies.

>Rice paddies near Lubuk Beringin village are an important livelihoodsource or villagers in Bungo.

A.

Ayat/ICRAF


11/26

o

. 4 . . f f i lI I 1 0

, .

1 2 01 2 0

Figure 14

s- dBH 10 , 3.2 h (l, 1997, ) 3.2 h jg bb (H et al., 1999; ).

Open diamonds: all trees including rubber trees. Filled diamonds: rubber trees excluded rom the jungle rubber data.

Case Study4

The biodiversit assessments indicated that complex rubber agroorests in Bungo not onl

represents secondar habitats/reuges or orest species, but the are also important connectors

amongst remaining ragmented orest patches. According to the landscape conguration, complex

rubber agroorests can constitute a series o stepping stones or more continuous corridors (van

Noordwijk, 2005).At the communit level, the RUPES project initiated a number o activities aimed to assess

the strengths, weaknesses, threats and opportunities o traditional rubber cultivation that can

maintain rich biodiversit. Local perception and needs were assessed through consultations and

research. Activities to enhance the awareness o the local communities about the value o their

traditional sstem or biodiversit conservation were implemented. Communities o Letung, Sangi,

Mengkuang Besar, Mengkuang Kecil and Lubuk Beringin villages agreed to retain their complex

A.

Ayat/ICRAF

A.

Ayat/ICRAF

A.

Ayat/ICRAF

2 4 6 8 4010 20

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

Total time (manhours)

Num

ber

of

species

(cumulative)

L E G E N D

Forest,all species

Jungle rubber,all species

Rubberplantations,all species

Forest,orest species

Jungle rubber,orest species

Rubberplantation,orest species

Adapted rom Figure 6 in Beukema et al., 2007: 227


12/26

N

1 2 11 2 1

rubber agroorests (total o about 2 500 ha) i incentives are provided. The incentives local people

requested include support to establish micro-hdro power plants, setting up o rubber nurseries

and demonstration plots o improved rubber agroorests, and clonal plants o high ielding rubber

trees or intensivel managed rubber gardens elsewhere. Conservation agreements were signed bthese our villages in 2006 (ICRAF, n.d.; Leimona and Joshi, 2010). The incentives provided then

were seen onl as an interim reward while a more permanent reward mechanism is being sought.

RUPES is currentl considering an eco-certication scheme or these complex rubber agroorests

that will etch a price premium or the natural rubber rom the jungle to be used in niche markets,

such as green cars and biccle tres. There is also a possibilit o bundling biodiversit services

together with other services, such as carbon or water qualit (Leimona and Joshi, 2010).



Current pages

(from left to right):

>Natural rubber comes rom the milky latexound in the bark o rubber trees.

>Tapping involves extracting latex rom arubber tree by shearing o a thin layer o barkin downward hal spiral on the tree trunk.

>Rubber slab containing a high percentage(about 45 percent) o dry rubber content.

>Microhydropower as nonfnancial reward orLubuk Beringin village or conserving biodiversejungle rubber systems.

L.

Joshi

Figure 15

s- h b d Hg, 1995.Open symbols: all birds identifed to species level. Filled symbols: subset o orest species classifed in habitat

group 1: species mostly associated with the primary and old secondary orest interior.

L E G E N D

Forest,all species

Jungle rubber,all species

Rubberplantations,all species

Forest,orest species

Jungle rubber,orest species

Rubberplantation,orest species

Num

ber

of

species

(cumulative)

Total area (ha)

Number o plots (cumulative)

50

45

40

35

30

25

20

15

10

5

00.16 1 2 3 4

1 2 3 4 5 10 15 20 25



13/26

1 2 21 2 2

Case Study4

The Bungo case stud is a clear example on how biodiversit assessments are comprised

o multiple laers o inormation. In this case, the generic relationship between rubber

agroorestr and biodiversit has to be decomposed in at least our dierent levels,distinguishing between (a) plant and (b) animal levels o biodiversit, while considering

biodiversit conservation at both the (c) plot and (d) landscape levels. Moreover, jungle rubber

gardens also show the crucial relationship between biodiversit and land management over

time because not onl dierent management regimes infuenced the recorded biodiversit

level, but under the same management regime jungle rubber gardens o dierent ages host

dierent levels o biodiversit.

Figure 16

s- h (),jg bb () bb (g).

Open symbols: all terrestrial pteridophyte species; flled symbols: orest species subset.Plots were 0.16 ha each, non-adjacent and spread over a large area in Jambi province.

L E G E N D

Forest, 3.2 ha

Jungle rubber,3.2 harubbertreesexcluded

Jungle rubber,3.2 harubbertrees

included

Numbe

rofspecies

(cumulative)

Number o individuals (cumulative)

35 0

30 0

25 0

20 0

15 0

10 0

50

050 100 200 400 800 1600 3200


A.

Ayat/ICRAF

A.

Ayat/ICRAF

A.

Ayat/ICRAF


14/26

i

1 2 31 2 3

reerences

Bk, H., d, ., v, G., H, s. & a, J. 2007. Plant and bird diversit inrubber agroorests in the lowlands o Sumatra, Indonesia.Agroorestry Systems, 70: 217242.

d, . & Hg, m. 1995. Impact o logging and plantation development on species diversita casestud rom Sumatra.In . Sandbukt, ed. Management o tropical orests: Towards an integrated perspective, pp.7392. Oslo, Centre or Development and the Environment, Universit o Oslo.

ek, a., Zk, m.t. & w, a. 2010. Agroorestr area under threats: Dnamics and trajectorieso rubber agroorest in Bungo district, Jambi. In B. Leimona & L. Joshi, eds. Eco-certifed natural rubber

rom sustainable rubber agroorestry in Sumatra, Indonesia Final Report. Bogor, World Agroorestr Centre(ICRAF)-Southeast Asia Regional Oce.

ek, a. & v, G. In press. Rubber agroorest in a changing landscape: Analsis o land use/covertrajectories in Bungo district, Indonesia. Forest, Trees and Livelihoods.

, l., chg, w.K. & lg, p. 2010. Wh do armers preer oil palm? Lessons learnt rom Bungodistrict, Indonesia. Small-scale Forestry, 9: 379396.

G, a.n. 2000.Above ground biodiversity assessment working group summary report 199699: Impact o dierentland uses on biodiversity and social indicators. Nairobi, ASB Working Group Report, World Agroorestr Centre(ICRAF) (available at http://www.asb.cgiar.org/PDFwebdocs/ASB Biodiversit Report.pd).

H, s., a, d.t., s, H.B., a, d., wh, m. & sb, s.m. 1999. Drat report othe research: Stand structure and species composition o rubber agroorests in tropical ecosystems o Jambi,Sumatra. yogakarta, Facult o Forestr, Gadjah Mada Universit and Bogor, World Agroorestr Centre(ICRAF)-Southeast Asia Regional Oce.

icra. n.d. Site profle: RUPES Bungo. Bogor, World Agroorestr Centre (ICRAF) (available at http://www.worldagroorestr.org/sea/networks/rupes/download/SiteProles/RUPES-Bungo_FINAL.pd).

Jh, l. , wb, G., v, G., B, d., ak, r., mg, G., njk, m. & w,s.e. 2002.Jungle rubber: A traditional agroorestry system under pressure. Bogor, World Agroorestr Centre(ICRAF)-Southeast Asia Regional Oce.

Jh, l., wb, G., Bk, H., w, s. & njk, m. 2003. Technological change andbiodiversit in the rubber agroecosstem o Sumatra. In J. Vandermeer, ed. Tropical Agroecosystems, pp.

133157. Florida, CRC Press.l, y. 1997. The vegetation and physiography o Sumatra. Geobotany 22 . Dordrecht, Kluwer.

l, B. & Jh, l. 2010. Eco-certifed natural rubber rom sustainable rubber agroorestry in Sumatra,Indonesia. Project Final Report. Bogor, World Agroorestr Centre (ICRAF).

r, s. 2006. Ekologi regenerasi tumbuhan berkayu pada sistem agroorest karet (Regeneration ecology o woodytrees in rubber agroorest systems). Sekolah Pasca Sarjana, Institut Pertanian Bogor, Bogor. (Doctoral dissertation).

njk, m. 2005. Rupes typology o environmental service worthy o reward. Bogor, World AgroorestrCentre (ICRAF).

wb, G., H, s. & njk, m. 2005. Permanent smallholder rubber agroorestr sstems inSumatra, Indonesia.In Cherl A. Palm, Stephen A. Vosti, Pedro A. Sanchez & Poll J. Ericksen, eds. Slash-and-burn agriculture: The search or alternatives. New york, Columbia Universit Press.



Examples of animal biodiversity found in the

forest and forest-edge habitat of Bungo district,

where jungle rubber gardens often constitute a

corridor between remaining forest patches(from left to right):

>Collared kingfsher (Halcyon chloris).

>Painted bronzeback snake(Dendrelaphis pictus).

>Crabeating macaque (Macaca fascicularis).

> Indian momtjac (Muntiacus muntiak).

A.

Ayat/ICRAF


15/26

1 2 4


16/26

1 2 5

C H A P T E R

cost-eectivetarGetinG o pes

Reproduced with permission rom:

OECD, 2010. Cost-eective targeting o paments or ecosstem services,

in Paing or Biodiversit: Enhancing the Cost-Eectiveness o Paments or Ecosstem Services,

OECD Publishing. Also available at http://dx.doi.org/10.1787/9789264090279-8-en

Katia Karousakis and Quiller BrookeOrganisation or Economic Co-operation and Development (OECD), Paris, France

contents

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126Targeting ecosstem services with high benets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

Spatial mapping tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

Targeting ecosstems services at risk o loss or degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Targeting providers with low opportunit costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133


17/26

PAyMENTS FORECOSySTEM SERVICES AND

FOOD SECURITy

1 2 6

aBstract

Individuals or communities with the potential to infuence the suppl o ecosstem services

will oten dier in the magnitude o benets the can provide, the risk that these services will

otherwise be lost or the extent to which their management activities can enhance biodiversit

and ecosstems, as well as the costs o service provision. This chapter discusses how PES

programmes can be designed to address these issues and presents the tools and methods

through which paments can be targeted to increase PES cost-eectiveness.

How paments or biodiversit and ecosstem services are targeted is critical in determining

the cost-eectiveness o a PES programme. In most cases, the available budget or biodiversit

and associated ecosstem services will be limited and competing with dierent demands.

Cost-eective targeting o paments enables greater total benets to be achieved with a given

PES budget and can thereore also contribute to the long-term success o the programme.

Man PES programmes allocate uniorm paments on a per hectare basis. This is cost eective

i ecosstem service benets and the costs o their provision are constant across space. In man

cases however, this is unlikel. The more heterogeneous the costs and benets are, the greater

the cost-eectiveness gains that can be realized via targeted and dierentiated paments.

Indeed, more and more PES programmes are incorporating design elements to address this. This

chapter examines the methods and tools that are available to target spatial heterogeneit in

biodiversit and ecosstem service benets, the threat o loss and the costs o their provision.

tarGetinG ecosystem services witH HiGH Beneits

Identiing areas with high biodiversit and ecosstem service benets requires metrics and

indicators to quanti them. Selecting an appropriate metric or indicator or PES that aims to

enhance biodiversit conservation and sustainable use is not necessaril

straightorward however. Unlike carbon, or example, which is measured in

tonnes o carbon dioxide equivalents (tCO2e), there is no single standardised

metric to quanti biodiversit. The multidimensionalit and the inherent

complexit o biodiversit require trade-os between the accurac o a metricand the costs o development. The appropriate biodiversit metric or indicator

selected or a PES programme ma also depend on the specic objectives o

the programme. Indeed, methodologies or constructing metrics and indicators

tend to be tailored to specic local, regional and national programmes and their objectives.

Examples o metrics and indicators used across two biodiversit PES programmes, namel the

Victorian BushTender programme in Australia and the PES scheme implemented in the Assiniboine

River watershed o east-central Saskatchewan province in Canada are presented in Box 1.

The inherent complexity

o biodiversity

requires trade-os

between measurementaccuracy and the cost o

biodiversity assessments


18/26

ENHANCING THECOST EFFECTIVENESS

OF PES

1 2 7

Box 1

m g b bf h v

Bht c pes

th Hb H hg h v Bht g

The aim o Victorian BushTender programme in Australia is to improve the management

o native vegetation on private land. To quanti biodiversit benets, the BushTender

programme uses the Habitat Hectare (HH) methodolog. The HH is comprised o an

assessment o the local benets via the Biodiversit Benets Index (BBI). The BBI

is based on the proposed management practices; the conservation signicance in

terms o regional priorities through the Biodiversit Signicance Score (BSS), the

cost o conserving the land (b) and the size o the proposed land (ha). Potential plots

are compared through an inverse auction, where landholders submit bids including

inormation on the proposed area, the BBI and the required pament. The BSS is

calculated separatel to improve competition (DSE, 2009).

HH = BBI x ha

BBI = (BSS x HSS) b

where

HH = Habitat Hectare;

BBi = Biodiversit Benets Index;

ha = area in hectares;

BSS = Biodiversit Signicance Score;

HSS = Habitat Service Score; b = cost o bid

tgg w c pes g

In Canada, a pilot PES programme initiated in 2008 to restore drained wetlands was

undertaken in the Assiniboine River watershed o east-central Saskatchewan. The

Environmental Benets Index (EBI) was based on the incremental increase in predictedhatched waterowl nests relative to the bid price. The EBI was based on the Ducks

Unlimited Canada Waterowl Productivit Model (DUC) which evaluated the potential

o wetland restoration on each plot to increase the number o hatched waterowl nests

in the Assiniboine watershed. The EBI was based on wetland area restored, waterowl

densit, existing wetland densit and the percentage o cropland in a 4x4 mile block

around the plot (Hillet al., 2011).


19/26


FOOD SECURITy

1 2 8

The use o such metrics to better target ecosstem service paments can substantiall

enhance PES cost-eectiveness. In the Tasmanian Forest Conservation Fund programme, or

example, a comparison o using the AUD/CVI1 metric with a simpler AUD/ha2 metric indicated

an 18.6 percent gain in conservation outcomes. Comparing the additional conservation gains

(valued at approximatel AUD 3.3 million) with the costs o achieving those benets (AUD

0.5 million), illustrate that the ratio o benets to costs rom investing in the CVI is 6.9:1.

Similarl, Wunscher et al. (2006) simulated dierent targeting approaches or the Costa Rican

PES and estimated that a scenario selecting highest scoring sites with the given budget would

have resulted in 14 percent higher benets than the current sstem o selecting sites (see Case

Stud 5 PES in Costa Rica).

spatial mappinG tools

Spatial mapping tools are increasingl being used to discern the spatial heterogeneit in

ecosstem costs and benets. Several o these tools are emerging to help design PES sstems at

the regional and national level; however, there are increasingl initiatives

o spatial mapping tools that are being developed at the international

scale, including the UNEP-WCMC Carbon and Biodiversit Demonstration

Atlas, ARticial Intelligence or Ecosstem Services (ARIES),3 the Integrated

Valuation o Ecosstem Services and Trade-os (InVEST)4 and SENSOR.

To target ecosstem service paments in Madagascar, Wendlandet al. (2010)

examined the spatial distribution o biodiversit (proxied b vector data

on species ranges o mammals, birds and amphibians), carbon and water

qualit. The let panel o Figure 17 depicts the degree o overlap between these three ecosstem

services. The right panel urther incorporates inormation on the probabilit o deorestation

and the opportunit cost o the land to identi where paments could be most cost-eectivel

targeted. One example o a spatial mapping tool developed at the international level is the Carbon

and Biodiversit Demonstration Atlas, produced b the UNEP World Conservation Monitoring

Centre (UNEP-WCMC) (Kapos et al.,2008). The Atlas includes regional maps as well as national

maps or six tropical countries showing where areas o high biodiversit importance coincidewith areas o high carbon storage. Figure 18 illustrates the national map or Panama, indicating

that 20 percent o carbon is stored in high carbon, high biodiversit areas.

1 AUD/CVI: ratio o Australian Dollars (AUD) to the Conservation Values Index (CVI)

2 AUD/ha: ratio o Australian Dollars (AUD) per hectare o land

3 http://esd.uvm.edu/

4 http://www.naturalcapitalproject.org/

Spatial mapping tools

are increasingly being

used to discern the

spatial heterogeneity

in ecosystem costs

and benefts


20/26

=ae

o

A1

11

z 4 a / d 1 1 4I f \ . I I 4 . 1 0 0 4 . 1 1 1 1 4 1 1 1 1 1

1 / { 4 , .

T o r g e i n d t o t P E Sa : A m b i t a r e aP h a rL a t oC A P I L Yi l a p

I A t I . 1 1 1 1 N W =I i 4 1 - = 1 . 2 4 1

C A . C . 1 . . . . . o . .. a s e , P l ih i m M o i r . e R i 1 1 1 4. 1 . 1 0 " I a r mL

' p - r 9 1 4 I / N o t% h e w n a s w i l l i i O M


OF PES

1 2 9

Figure 17

tgg pes mg

Figure 18

ex unep-wcmc : p

Source: OECD, 2010

Source:OECD, 2010


21/26


FOOD SECURITy

1 3 0

To identi areas o high biodiversit importance or the regional maps, UNEP-WCMC uses

six indicators or biodiversit, namel Conservation Internationals Hotspots, WWFs Global 200

ecoregions, Birdlie Internationals Endemic Bird Areas, Amphibian Diversit Areas, Centers o Plant

Diversit and the Alliance or Zero Extinction Sites. Areas o high biodiversit, as determined b

UNEP-WCMC, are areas where at least our o the above-listed biodiversit conservation priorit

areas overlap, with areas in dark green indicating a greater degree o overlap.

The maps identi the dierent areas with high biodiversit importance. The maps do not

necessaril identi areas with high biodiversit benets in economic terms. Ideall, spatial

maps on biodiversit benets would incorporate the total economic value o these sites, with

an assessment o both direct and indirect use values.

A number o spatial mapping initiatives are currentl underwa and are in dierent

stages o development. These include ARticial Intelligence or Ecosstem Services (ARIES)

(Villa et al.,2009); InVest (Tallis et al.,2010); the United States Geological Surve (USGS) Global

Ecosstems initiative;5 and SENSOR (Sustainabilit Impact Assessment:

Tools or Environmental, Social and Economic Eects o Multiunctional

Land Use in European Regions).6

As suggested in the Madagascar example above (Figure 17), PES

programmes can simultaneousl target multiple ecosstem service benets.

Bundling or laering (Figure 19) can allow a broader range o ecosstem

service benets to be obtained in a cost-eective manner, avoiding the

need or multiple programmes, reducing transaction costs and programme

overlap. Multiple ecosstem service provisions can help ensure that all

aspects o an ecosstem on enrolled land are properl managed, increasing the asset value

o the ecosstem. PES targeting multiple ecosstem services can enable the landholder to

maximise potential paments received, such that conservation becomes more economicall

easible, enabling greater ecosstem service provision.

The easibilit o targeting multiple ecosstem services simultaneousl depends on the

degree o spatial correlation between dierent tpes o ecosstem services. Spatial mapping

tools help to identi where multiple service benets coincide. Though there ma oten be

snergies in service provision (e.g.avoided deorestation results in both biodiversit and carbonbenets), there are cases when trade-os can also arise (Nelson et al.,2008). For example,

whereas native and mixed crops provide biodiversit benets, monocultures o ast-growing tree

species such as Eucalyptus ma provide more rapid carbon sequestration benets. Farle et al.

5 http://rmgsc.cr.usgs.gov/ecosstems/

6 http://www.ip-sensor.org

PES objectives must

be clear, potential

trade-os recognised

and saeguards

developed to prevent

adverse collateral

impacts


22/26


OF PES

1 3 1

(2005) highlighted this problem in West Arica, where carbon sequestration (i.e.aorestation/

reorestation) projects can negativel aect water regimes and biodiversit. The ultimate

objective o the PES programme must thereore be clear, potential trade-os recognised and

saeguards ma be needed to prevent adverse impacts on other ecosstem services (Karousakis,

2009). In this context, environmental benet indices and scoring approaches become not

onl a wa o evaluating the qualit o potential contract benets, but are also mechanisms

through which discrete ecosstem service priorities are traded o against each other. An

weights associated with an Environmental Benets Index (EBI) or scoring mechanism can

also be modied in sequential PES sign-up rounds to reconcile trade-os. This has been done,or example, in the Mexican PEHS7 programme (Figure 20) where weights have been adjusted

over time to better address the polic priorities. Similar targeting methods have been used to

allocate paments in the Socio Bosque programme in Ecuador. Based on a sstem o scores,

7 Paments or Environmental Hdrological Services (Pago de Services Ambientales Hydrologicas - Mexico)

Figure 19

mkg b j

Source:OECD, 2010

BundlinG: A PACKAGE OF SERVICES FROM THE SAME LAND AREA IS SOLD TO THE SAMESINGLE BUyER.

layerinG: A BUNDLE OF SERVICES FROM THE SAME LAND AREA IS SOLD TO DIFFERENTBUyERS.

piGGy BacKinG: ONE SERVICE IS SOLD AS AN UMBRELLA SERVICE AND BIODIVERSITy IS A FREE-RIDEROR ONLy TEMPORARILy REMUNERATED.

Package o services

(bird & watershed conservation)

Bird conservation services

Watershed protection services

Watershed protection services

No pament or start-up cost-sharing

b biodiversit beneiciaries

3

2

1


23/26


FOOD SECURITy

1 3 2

Figure 20

tgg peHs mx

Source:OECD, 2010

land area has been classied into three categories o priorit: priorit 1 (scoring rom 12.1 to

25); priorit 2 (7.1 to 12) and priorit 3 (0 to 7). The scores are based on high deorestation

pressure, storage o carbon in biomass, water suppl and povert alleviation.

Though these tpes o targeting approaches entail higher transaction costs, experience

with their use suggests that the resulting cost-eectiveness gains are improved. There are also

other tpes o PES design characteristics that can be introduced into the programme to reduce

transaction costs. In the Costa Rican PES, or example, private orest landholders are required

to have a minimum o one hectare to receive paments or reorestation and two hectares in the

case o orest protection. The maximum area or which paments can be received is 300 hectares

(and 600 hectares or indigenous peoples reserves) (Grieg-Gran et al.,2005). Aggregating

small projects is also possible to help reduce the transaction costs associated with a pament

contract. These tpes o PES design elements can help to ensure more equitable participation

in the PES programme and help to reduce administrative costs.

tarGetinG ecosystems services atrisK o loss or deGradation

In addition to targeting paments to ecosstem services with the highest benets, it is essential

to ensure that an pament leads to additional benets relative to the business-as-usual scenario.

For example, paments or habitat protection are onl additional i in their absence the habitat

Overexploited Aquiers

High waterscarcit inwatershed

High & Ver High Deorestation Risk

Ver highmarginalit

communities

50

40

30

20

10

0PEHS 2005

PEHS 2006

PEHS 2007

L E G E N D

PEHS - Paments orEnvironmentalHdrologicalServices


24/26


OF PES

1 3 3

would be degraded or lost. Inormation on the business-as-usual or baseline scenario is critical

in ensuring PES additionalit. Clear understanding o whether or not ecosstem services are

at risk o loss or degradation is thereore needed. Historical and current

trend data on biodiversit and ecosstem service loss are a starting point

and are needed to develop uture reerence projections. Though this can

be a complex task, there are dierent was this can be undertaken. For

example, to target PES in Madagascar, Wendland et al. (2010) estimate the

probabilit o deorestation (via a multivariate probit model) b examining

distance to roads and ootpaths, elevation, slope, population densit,

mean annual per capita expenditure and other characteristics. A similar approach is used to

assess deorestation risk in the Mexican PEHS programme. In this case, the variables used to

estimate deorestation risk include distance to the nearest town and cit, slope, whether it is

an agricultural rontier and i it is located in a natural protected area.

tarGetinG providers witH low opportunity costs

Finall, PES programmes can increase their cost-eectiveness i, given sites with identical

ecosstem service benets and risk o degradation or loss, paments are dierentiated and

prioritised to those sites where landholders have lower opportunit costs o alternative land

uses. In the Costa Rican PES, or example, Wunscher et al. (2006) illustrate that dierentiating

paments according to opportunit costs could allow the enrolment o almost twice the area

o land, representing more than double the environmental benets per cost (see Case Stud 5

PES in Costa Rica).

Obtaining accurate inormation on ecosstem providers opportunit costs is not straightorward

as the have an incentive to overstate these costs in an eort to extract inormation rents via

higher paments. Programme administrators have a number o options to assist revelation o

the landholders true opportunit costs. Specicall, the can gather additional inormation in

the orm o costl-to-ake signals or the can use inverse auctions.8

Inormation on ecosstem supplier attributes and activities which are correlated with their

opportunit costs can be used to iner the correct price. The inormation should be basedon costl-to-ake signals, or example, distance to markets, current land use, assessed value,

or labour and production inputs. Readil available market inormation can also be used and

incorporated into a model to estimate opportunit costs. In the USA Conservation Reserve

8 Screening contracts can be used in theor, but this is complicated in practice; see Ferraro (2008)

Inormation on thebaseline scenario is

critical to ensuring

the additionality o

PES projects


25/26


FOOD SECURITy

1 3 4

Program, or example, local land rental rates are combined with inormation on eld soil tpes,

a prox or productivit, to give a reasonable indication o the opportunit costs o retiring

agricultural land. This is then used as a maximum acceptable price, removing the landholders

abilit to claim unreasonabl high paments. To prox or opportunit costs in Madagascar,

Wendland et al. (2010) use data on the opportunit costs o agriculture and livestock produced

b Naidoo and Iwamura (2007). Naidoo and Iwamura compiled inormation on crop productivit

and distribution or 42 crop tpes, livestock densit and estimates o meat produced rom a

carcass and producer prices to measure the gross economic rents o agricultural land across

the globe. Wendland et al. (2010) clipped this global data to Madagascars boundaries. Gross

economic rents ranged rom USD 0 to 529 per hectare or Madagascar, with a mean value o

USD 45 per ha, per ear. The value o USD 91 per ha, per ear (one standard deviation) was

used as the cut-o to exclude areas o high opportunit costs.

However, obtaining inormation on costl-to-ake signals still incurs research costs. The

ecienc o the pament will directl depend on the qualit o this research and the strength

o the correlation between the signal and the opportunit costs, which must be assessed on

a case-b-case basis.

Exploiting competition between ecosstem service suppliers or conservation contracts

through inverse auctions can provide an eective cost-revelation mechanism. Where suppliers

are heterogeneous in their opportunit costs and demand or contracts exceeds suppl (i.e.the

conservation budget), competitive procurement auctions are possible.

The recognition o the potential gains rom the use o inverse auctions as a pament

allocation mechanism has stimulated heightened interest rom polic-makers. Though their use

in PES programmes is not et common, the are becoming more widespread in developed and

developing countries. Inverse auctions have been used to allocate PES contracts in Australia,

Canada, Finland, German, Indonesia, Tanzania, the United Kingdom and the USA (Claassen,

2009; DSE, 2009; EAMCEF, 2007; Hillet al., 2011; Jack, 2009; Juutinen and Ollikainen, 2010;

Latacz-Lohmann and Schilizzi, 2005).


26/26


OF PES

reerences

c, r. 2009. USDA briefng room - Conservation policy: Background (available athttp://www.ers.usda.gov/brieng/conservationpolic/background.htm).

dse (d sb e). 2009. ecoMarkets, Virginia, Australia, Vol. 11,pp. 15651576 (available at http://www.dse.vic.gov.au).

eamce (e ah m c e ). 2007. Welcome to the Eastern Archomepage (available at http://www.easternarc.or.tz/).

, K., Jbbg, e. & Jk, r. 2005. Eects o aorestation on water ield: A global snthesiswith implications or orestr. Global Change Biology, 11.

, p. 2008. Asmmetric inormation and contract design or paments or environmental services.Ecological Economics, 65(4): 810821.

Gg-G, m., p, i. & w, s. 2005. How can market mechanisms or orest environmentalservices help the poor? Preliminar lessons rom Latin America. World Development, 33(9): 15111527.

H, m., mm, d., H, t., Hh, a. & p, a. 2011. A reverse auction or wetlandrestoration in the Assiniboine River watershed. Canadian Journal o Agricultural Economics,1: 114.

Jk, B.K. 2009. Auctioning conservation contracts in Indonesia: Participant learning in multiple trialrounds. CID Graduate Student and Research Fellow Working Paper, No. 35. Center or InternationalDevelopment at Harvard Universit.

J, a. & ok, m. 2010. Conservation contracts or orest biodiversit: Theor andexperience rom Finland. Forest Science, 56: 201211.

K, v., r, c., cb, a., dk, B., Gbb, H., H, m., lk, i., m, l.,p, J., sh, J.p.w. & t, K. 2008. Carbon and biodiversity: A demonstration atlas.Cambridge, UK, UNEP-WCMC.

Kk, K. 2009. Promoting biodiversity co-benefts in REDD. OECD Environment Working PaperSeries, No. 11. Paris, OECD.

lz-lh, u. & shzz, s. 2005.Auctions or conservation contracts: A review o the theoreticaland empirical literature. Report to the Scottish Executive Environment and Rural Aairs Department.

mz p, c., G, a., t, J. & B, J. 2008. Paing or the hdrological services oMexicos orests: Analsis, negotiations and results. Ecological Economics, 65(4): 725736.

n, r. & i, t. 2007. Global-scale mapping o economic benets rom agricultural lands:Implications or conservation priorities. Biological Conservation, 140: 4049.

n, e., pk, s., l, d., pg, a., l, e., wh, d., B, d. & l, J.2008. Ecienc o incentives to jointl increase carbon sequestration and species conservation on alandscape. Proceedings o the National Academy o Sciences, 105(28): 94719476.

pg, s. 2006. Paments or environmental services in Costa Rica. Munich Personal RePEc Archive.

t, H.t., rk, t., n, e., e, d., w, s., o, n., vg, K., pg,d., mz, G., ak, J., , J., , J., c, d., ak, K., l, e. &K, c. 2010.InVEST 1.004 beta users guide. Stanord, The Natural Capital Project.

v, ., c, m., Bg, K., Jh, G. & K, s. 2009. ARIES (Artifcial Intelligence or

Ecosystem Services): A new tool or ecosystem services assessment, planning and valuation. Proceedingso the 11th Annual BIOECON Conerence on Economic Instruments to Enhance the Conservation andSustainable Use o Biodiversit.

w, K. 2008. Rewards or ecosstem services and collective land tenure: Lessons rom Ecuadorand Indonesia. Tenure Brie, 9: 110.

w, K.J., Hzk, m., p, r., v, B., rb, s. & r, J. 2010. Targeting andimplementing paments or ecosstem services: Opportunities or bundling biodiversit conservationwith carbon and water services in Madagascar. Ecological Economics, 69(11): 20932107.

wh, t., eg, s. & w, s. 2006. Paments or environmental services in Costa Rica:Increasing ecienc through spatial dierentiation. Quarterly Journal o International Agriculture,45(4): 317335.

Documents

PES and Food Security. Agroforestry Rubber