1124 Land Use

7/28/2019 1124 Land Use

1/82

Sustainable land use for the 21 st century

May 2012

Sustainable Development in the 21 st century (SD21)

7/28/2019 1124 Land Use

2/82

Design by ormatoverde.ptAll photographs by United Nations

7/28/2019 1124 Land Use

3/82 Ackowledgement | Sustainable land use for the 21 st century PAGE I

Acknowledgement

This study is part o the Sustainable Development in the21st century (SD21) project. The project is implementedby the Division or Sustainable Development o the UnitedNations Department o Economic and Social A airsand unded by the European Commission - Directorate-General or Environment - Thematic Programme orEnvironment and sustainable management o NaturalResources, including energy (ENRTP).

Acknowledgements: The study was carried out byEphraim Nkonya (IFPRI), who coordinated inputsrom a team o authors that comprised Alain Karsenty(CIRAD), Siwa Msangi (IFPRI), Carlos Souza Jr(IMAZON), Mahendra Shah (IIASA), Joachim von Braun(ZEF, University o Bonn), Gillian Gal ord (Woods HoleResearch Center), and SooJin Park (Seoul NationalUniversity). Supervision o the study was done by DavidLe Blanc (UN-DESA). The study relied on a team o peerreviewers that comprised Iddo Wernick (Rocke ellerUniversity), Ken Chomitz (World Bank), Arild Angelsen(University o Norway), Gerald Nelson (IFPRI), andTimothy Thomas (IFPRI).

Disclaimer: This publication was produced with theassistance o the European Union. The contents o thispublication are the sole responsibility o the UnitedNations Department o Economic and Social A airs andcan in no way be taken to re ect the views o the EuropeanUnion. Similarly, the views expressed in this publicationare those o the authors and do not necessarily re ectthose o the United Nations or its senior management.

7/28/2019 1124 Land Use

4/82Sustainable land use for the 21 st century | Executive SummaryPAGE IV

across regions are driven by economic development,government policies and other socio-economic actors.In the past two decades (1990-2010), orest densityhas increased globally while orest extent has slightlydecreased by 0.2% per year in 1990-2000 and by0.1% in 2000-10. Overall, orest density and extenthas increased in high income countries and generallydeclined in low income countries. Related to orest isbiodiversity, which is enhanced by establishment oprotected areas. Globally, protected area increased by38% in 2010 rom its level in 1992. Despite the impressiveincrease in protected area, loss o biodiversity remainsquite high since biodiversity is naturally developed overa long time and there ore increase in protected area isnot matched with immediate increase in biodiversity at least in the short-run.

An environmental Kuznets curve which shows adecline in orest extent as the economy grows andsubsequently an increase a ter reaching a threshold explains LUCC change trends in most countries. Forestextent and density and biodiversity also reveal anenvironmental Kuznets curve pattern. Many countriesin the tropics are in phase two increasing orestextent and density and biodiversity. However, the oresttransition the environmental Kuznets curve o orestextent has not been observed in some countries dueto a number o reasons including strong timber markets,civil wars, government policies, etc. Additionally, thepredictive power o the environmental Kuznets curvehas been reduced by globalization and the increased roleo international trade.

Rates o agricultural expansion are decreasingglobally but still expanding in sub-Saharan A rica(SSA) and Latin American countries. The decreasein expansion o agricultural land is largely due toincreasing agricultural productivity. From 1961-2005,crop yield accounted or 77% o the global increase inood production but in SSA, contribution o crop yieldto total production was only 38%, the lowest among allregions. Hence the yield gap the di erence betweenpotential and actual yield remains wide in SSA andother developing countries south and central Asia buthas narrowed in high income countries. It is in lowincome countries that there remains a large potentialor increasing ood production without increasingagricultural area. This requires investment to addressconstraints which contribute to low agriculturalproductivity, which include poor market in rastructureand generally low investment in agriculture.

Executive Summary

Introduction

It is estimated that the human footprint has affected 83% of theglobal terrestrial land surface and has degraded about 60% of theecosystems services in the past 50years alone. Land use and land cover (LUCC) change has been themost visible indicator of the human footprint and the most important driver of loss of biodiversity and other forms of land degradation.Recent trends on global demand or ood and bioenergychange which are closely linked to ood and energyprice spikes and volatility have raised concerns onthe impact o LUCC change on biodiversity and otherenvironmental impacts. Additionally, LUCC changecould lead to natural resource degradation whicha ect the poor the most since they heavily depend onnatural resources. Since the earth Summit in 1992,the international community, individual countries,communities, civil society and businesses haveincreasingly become aware o the environmental impacto LUCC change. This paper assesses the LUCC changeand explores actors which could be addressed toensure sustainable development. The paper is dividedinto fve sections and the frst one begins by exploringwhat science tells us about LUCC change. The secondsection uses three case study countries to discuss howLUCC changes happen in practice. This is ollowed byan analysis o the land management programs and thee ectiveness o market-based instruments. The ourthsection discusses LUCC modeling and the last sectionconcludes the paper by looking at the uture prospectso LUCC change.

What does science tell us about LUCCchange?

We explore what science tells us about LUCC usingthe major terrestrial land use types, namely orestsand agriculture. We also discuss biodiversity and waterresources, both o which are closely linked to eachother and to agriculture and orest. Forest trends

7/28/2019 1124 Land Use

5/82 Executive Summary | Sustainable land use for the 21 st century PAGE V

Demand or water is increasing ast. It estimated thatthe growing human population will require more ood,which will translate into doubling demand or water oragriculture rom the current level o 7,130 km 3 to 12,050to 13,500 km 3 in 2050. Additionally, water availability isexpected to be less reliable in arid and semi-arid areasdue to climate change, which is expected to increaseintra-annual variability in precipitation and increasedseverity and requency o droughts.

How does land use change happen inpractice, and how are competing demandson land managed?

Brazil, DRC and Indonesia are used as case studiesto illustrate the impact o country policies on LUCCchange. Brazil home to the largest part o the Amazon has implemented policies which successively led todegradation and later rehabilitation o the Amazon. Until2011, about 19% (762,000 km 2) o the original Amazonorest area was cleared under policies which encouragedcolonization o the Amazon. About 72% o the orestclearing took place rom 1980-2011. However, Federal,state, and municipal governments in Brazil realized thenegative impacts o the losses and took actions to stopde orestation. In collaboration with international donors,Brazil was able to reduce the de orestation rate by 74%in only fve years (2004-2009).

Similarly, Indonesia provided timber concessions whichled to rapid de orestation. Agricultural expansion especially palm oil production and decentralization oorest management also contributed to de orestation.Decentralization o orest management coupled withlimited local government budgets led local governmentsto use timber concessions to generate revenue. Asin the case o Brazil, the Indonesian government incollaboration with international donors embarked one orts to reduce de orestation. These e orts includedstrict en orcement o protected areas and incentives orprotecting orest areas. Community orest managementprograms were also implemented. Recent data showthat the annual de orestation rate in Indonesia ell rom1.7% in 1990-2000 to only 0.5% in 2000-10.

Forest trends in the Democratic Republic o Congo (DRC)show a signifcantly di erent pattern. De orestation inDRC has been limited by the poor in rastructure and byinsecurity, which deters commercial logging. However,there is a large in ormal logging activity run by chainsawloggers and small-scale enterprises or domestic

markets and illegal export to neighboring countries.Additionally, there has been a steady increase in legalconcessions. Consequently the rate o de orestation hasbeen increasing. With the help o the various internationale orts, the country has recently taken several steps toprotect the Congo orest and other natural resources.One o such e orts is the community orest. The decreeestablishing community orests is still pending and itwould probably give large powers to the customarychie s, who in DRC have o ten shown a limited sense oaccountability. Overall, the reach o the public authoritiesis very limited in a country that remains a ragile State,where corruption is still omnipresent and the judiciarysystem is down.

What do we know about the effectiveness ofland management systems at the sectorallevel?

There is increasing debate on the role o market-basedinstruments (MBIs) to reduce land use conversions andother environmental issues. MBIs have the potentialto serve as an e fcient alternative to administrativeregulations and prescriptive laws or addressingenvironmental issues. Forest certifcation and eco-labeling have been probably the most success ul MBIover the last two decades or enhancing sustainableorest harvesting and management. Forest productssold in high income countries with strict environmentalstandards are required to have a orest certifcateshowing that the products were not obtained romprotected areas or other ecologically important areas.However, e ectiveness o orest certifcation is largelyrestricted to a hand ul o companies exporting theirproducts to environmentally-concerned markets.

In high income countries, conservation easementprograms bilateral contracting with land owners orusers to not use land or certain development or use have also shown considerable success. Other MBIs havebeen used but they have been more success ul in highincome countries than in low income countries.

Programs which have also been airly success ul inmedium and low income countries are those aimed atenhancing Payments or ecosystem services (PES).The PES programs pay land owners/users to conductenvironmentally riendly initiatives or to give updestructive practices. Interest in PES has increasedrapidly over the past two decades. Today there aremore than 300 programs implemented worldwide,

7/28/2019 1124 Land Use

6/82Sustainable land use for the 21 st century | Executive SummaryPAGE VI PA

predominantly used to address biodiversity, watershedservices, carbon sequestration and landscape beauty.Empirical evidence tends to indicate that water-relatedPES have been more e ective than others, probablybecause the buyers are the direct benefciaries o theservice, unlike or biodiversity and carbon PES wherethe benefciaries are the global community and buyersare those serving as intermediaries or the present anduture world community.

A major problem o PES is the compensation based on theopportunity cost. This has been regarded as inequitableor the poorest populations. Freezing user rights suchas clearing, hunting or even the prospect o working in aorestry company deprives people o opportunities to li tthemselves out o poverty. Additionally, elite capture inPES has also been reported.

Additionality has also remained a major challengeo PES programs. So do the competing priorities oprotected areas. For example, the debate on the REDD+revolves around three distinct interest networks thosewho give priority to carbon, those who are concernedabout biodiversity, and those who de end the rights olocal and indigenous populations. As a result o thisand other challenges, most decisions and rules andregulations on the REDD+ unded by governments andinternational organizations are still pending. The weakprospects o climate change negotiations are posinganother challenge to PES programs. In sum, PESand other MBI programs in general continue to acedaunting challenges to reach agreement on a number ocontentious issues. However the signifcant progress ininternational cooperation on sustainable developmentmade in the past 20 years o ers some hope and lessonsor acing such challenges.

How are land use and land use modeled inscenario exercises?

Models or predicting uture LUCC change use theoryto link changes with its biophysical and socio-economicdrivers. Statistical approaches are then used toestablish historical relationship between LUCC andits drivers. Three main types o models have evolvedbased on di erent disciplines: geographic, economic andecological models.

Geographic models are ocused on land allocation basedon suitability o land use and the spatial location oecosystems and population. Hence geographic models

tend to better allocate land use to areas with minimale ect on the ecosystems. The models better capturethe potential productivity o di erent land uses and arebetter able to re ect land management than economicmodels. However, geographic models assume thatprices and other international eedback variables areexogenous. This makes them less able to re ect thein uence o international trade on market-driven agentbehavior. Economic models ocus on the demand andsupply o land-based goods and services. They moree ectively re ect the e ect o international trade andglobalization on LUCC change. Additionally, economicmodels use scenarios to capture the in uence o policiesand other socio-economic actors on LUCC. Ecologicalmodels link land allocation to species abundance andextinction, ecological ootprints and other environmentalconcerns. Ecological modeling methods also o tenassume that prices and other economic variables areexogenous actors, thus ailing to ully account or theirimpacts and associated trade-o s in land allocation.

Over time, LUCC modeling has become more integrated,breaking the disciplinary divide. In act, the predictiveaccuracy o integrated models is higher than those o thespecialized models. Such integrated approach fts wellwith the ecological interrelationships o di erent landuses and the integrated approach that characterizessustainable development. For example, solutions tosimultaneously achieve the ood security, biodiversityand bioenergy objectives o maximizing humanwel are require use o integrated models. Despite suchprogress however, prediction o uture LUCC remains achallenge. Un oreseeable shocks and events as well asincorporation o human behavior in LUCC models haveremained elusive and have contributed to poor prediction.

Prospects for the future

LUCC change is posing a grave danger to earthsecosystems. One estimate puts the sa e upper boundaryor global cropland area to 15% o the total terrestrialarea, a level that is only about three percentage pointhigher than current cropland area which account or12% o global land area. However, another estimate bythe UNFCCC Commission has concluded that currentglobal agricultural production has already steppedoutside the sa e boundary. Loss o biodiversity is alreadyoutside the upper boundary with the current rate oextinction being 100-1,000 times higher than the pre-industrial age level. Additionally, reshwater resourcesare also overwhelmed by the increasing population and

7/28/2019 1124 Land Use

7/82 Executive Summary | Sustainable land use for the 21 st century PAGE V

climate change, which have increased their variabilityand reduced their supply in dry areas. The increasingdemand or bioenergy has posed yet another challengeto land and water resources. Hal o the global cerealsconsumption in 2005/6-2007/8 was due to US ethanolproduction and projections by FAO/OECD show that 52%o maize and 32% o oilseeds demand to year 2020 willbe due to bioenergy. Estimates show that a large portiono the area or bioenergy production will be derived romclearing orests and grassland. These trends show thatbusiness as usual is not sustainable.

So what can be done to sustainably achieve food security,protect biodiversity and energy security? A recentorecasting study showed a decreasing yield growthat the global level. Food security is achievable but thiswill require increasing ood production by increasingagricultural productivity in low income countries wherethe yield gap is widest. This will require addressingconstraints which limit higher yield in such regions. Theseinclude increased investment in agricultural research aswell as addressing market conditions and rural services,which will provide technical support and incentives orincreasing productivity. Achieving ood security alsorequires reducing post-harvest losses, which are high inboth developing and developed countries. Post-harvestood losses could be reduced by investment in processingand storage investment in developing countries and bypublic awareness in developed countries to change oodconsumption habits which lead to ood losses. Greaterwater productivity is also required to increase yield inthe regions where water productivity is low.

On bioenergy, studies have cast doubt on the e fcacy obio uels as mechanisms or reducing GHG using currenttechnologies. E orts to use second generation eedstockprovide some potential or liquid bioenergy which doesnot compromise ood security and biodiversity.

Given the current high biodiversity losses, reducingbiodiversity loss to pre-industrial levels will be hard toachieve. However, signifcant reduction in biodiversityloss is possible. In this regard, the recent increase inprotected areas o ers some hope.

Finally, prospects or international instruments or landuse change management require synergistic programs,which provide several ecosystem services. This meansinternational cooperation on carbon and other ecosystemservice initiatives need to explore closer collaborationto achieve synergistic objectives. For example, closer

collaboration o UNCCD, CBD, UNFCCC and others cansimultaneously combat land degradation, conservation obiodiversity and carbon sequestration. This is in line withthe spirit o Agenda 21, which promotes cooperation andthe building o synergies among ecosystem initiatives.

GE VIII

Acronyms and abbreviations

CBD Convention on Biological DiversityGHG Greenhouse gasesHANPP Human appropriation o net primary productionLAC Latin America and the CaribbeanLUCC Land use and land cover changeMA Millennium Ecosystem AssessmentPES Payment or Ecosystem ServicesREDD Reduced emissions rom avoided de orestationSSA Sub-Saharan A ricaUNFCCC United Nations Framework Convention onClimate ChangeUNCCD United Nations Convention to CombatDeserti ication

7/28/2019 1124 Land Use

8/82Sustainable land use for the 21 st century | Contents PAGE VIII

Contents Executive summary .................................................................................................................................................. IV Acronyms and abbreviations ............................................................................................................................. VII

Part 1: What does science tell us about land use change? .............................................................................

1Introduction ............................................................................................................................................................... 2Science o land use and land cover change ................................................................................................................ 2

Forests .................................................................................................................................................................. 3Agriculture ............................................................................................................................................................ 5Water resources .................................................................................................................................................... 7Biodiversity ............................................................................................................................................................ 7Ecosystem services and ecological ootprints ...................................................................................................... 9

Main drivers o change and sources o pressure on land-use .................................................................................... 9 Part 2: How land use change happens in practice, and how competing demands on land are managed ....... 15

Introduction ............................................................................................................................................................... 16Land management in Brazil ....................................................................................................................................... 16

Main interests at play ............................................................................................................................................ 16Government control on LUCC ............................................................................................................................... 18Pattern o LUCC in the Amazon and government objectives ................................................................................ 20

Indonesia ................................................................................................................................................................... 20Interests at play ................................................................................................................................................... 20

Democratic Republic o Congo .................................................................................................................................... 22 Part 3: What do we know about the effectiveness of land management systems at the sectoral level? ........... 27

Market-based instruments: Promises not (yet?) ulflled ........................................................................................... 28O sets ................................................................................................................................................................... 28Conservation easements ...................................................................................................................................... 28Payments or environmental services ................................................................................................................... 29

Certifcation and eco-labeling .................................................................................................................................... 30The FLEGT initiative .............................................................................................................................................. 31Forestry within the climate regime: CDM and REDD+ .......................................................................................... 31From CDM to REDD ............................................................................................................................................... 32

REDD+: the end o a relative and ragile consensus ................................................................................................... 32The way orward: what prospects or change? ..................................................................................................... 33

Part 4: How are land use and land use change modeled in scenario exercises? Land use and land cover modeling and their assumptions .......................................................................... 36

Introduction ............................................................................................................................................................... 37Assumptions o land allocation & drivers .................................................................................................................. 38Factors contributing to disagreement o land allocation across LUCC models ......................................................... 39Accuracy o prediction o LUCC models ..................................................................................................................... 41

Comparison o selected model projection and actual LUCC ................................................................................. 41Reasons behind discrepancy between LUCC projections and actual land use change ......................................... 43

Part 5: Prospects for the future ............................................................................................................................. 45Given past trends, what can be expected rom a business as usual scenario? .......................................................... 46What is achievable? .................................................................................................................................................... 47

Food security ......................................................................................................................................................... 47A higher water use e fciency is required .............................................................................................................. 49Biodiversity ............................................................................................................................................................ 49

Choices or managing land use or multiple objectives and critical areas o global coordination .............................. 50Prospects or international instruments or land use change management .............................................................. 51

Annex 1: Major LUCC models ................................................................................................................................. 52 References ................................................................................................................................................................... 59

7/28/2019 1124 Land Use

9/82 Contents | Sustainable land use for the 21 st century PAGE I

List of tables

Table 1: A ricas in rastructure defcit and cost ............................................................................................................... 11

Table 2: Contribution o domestic ood policies (trade tax) on international price spikes o major crops, 2005-08........

12Table 3: Predicted change in global land use due to US & EU mandates ........................................................................ 14Table 4: Summary o LUCC models ................................................................................................................................ 38Table 5: Major assumptions o major LUCC models ........................................................................................................ 39Table 6: LUCC e ects in di erent models ....................................................................................................................... 40Table 7: Disagreement o global land cover data ............................................................................................................ 41Table 8: LUCC model projection versus actual ood production and consumption in year 2000 .................................... 42

List of gures

Figure 1: Change o agricultural and orest area, 1992-2009 .......................................................................................... 3Figure 2: Forest transitions stylized model ................................................................................................................... 4Figure 3: Source o growth o agriculture, 1961-2005 ..................................................................................................... 6Figure 4: Contribution o regions to global suitable land ................................................................................................. 7Figure 5: Change in protected area in global sub-regions rom 1992 level ..................................................................... 8Figure 6: Global biodiversity conditions ........................................................................................................................... 9Figure 7: Water requirement o crops and livestock products ......................................................................................... 10Figure 8: Global Food price Index trend, 1961-2010 ....................................................................................................... 12Figure 9: Relationship between land tenure security and living biomass carbon density ............................................... 13Figure 10: Trend o loss o arable land area per capita across regions, 1961-2009 ........................................................ 14Figure 11: Contribution o the orest sector to GDP and employment in case study countries ....................................... 16Figure 12: Annual de orestation rates and cumulative de orested areas in the Brazilian Amazon Biomes .................... 17Figure 13: Modern human occupation in the Brazilian Amazon and Protected Areas in the Region ............................... 19

Figure 14: Trend o protected area in Brazil, DRC and Indonesia since 1992 ................................................................. 19Figure 15: Concessions and protected areas in DRC ....................................................................................................... 23Figure 16: Changes in the sur ace allocated to timber concessions ............................................................................... 24Figure 17: Map o mining squares in orests .................................................................................................................... 26Figure 18: Global boundaries versus current status/use o biodiversity, CO2 emission and reshwater consumption ... 46Figure 19: Yield gaps o major ood crops across regions ............................................................................................... 48Figure 20: Trend o global yield change o major cereals ................................................................................................ 49Figure 21: Current and uture (2050) water use (km 3/year) ............................................................................................ 49Figure 22: Integrated LUCC models ............................................................................................................................... 58

7/28/2019 1124 Land Use

10/82

What does sciencetell us about land usechange? 1.

7/28/2019 1124 Land Use

11/82 Part 1 | Sustainable land use for the 21 st century PAGE2

Introduction

It is estimated that the human ootprint has a ected83% o the global terrestrial land sur ace (Sandersonet al 2002) and has degraded about 60% o the ecosystemsservices in the past 50 years alone (MA 2005). Therates o land use and land cover change (LUCC), whichhad already increased in the last century, acceleratedin the last three decades at an alarming level (Lambinand Geist 2006). These LUCCs mostly impacted humidand sub-humid areas (Baiet al 2008) that were largelyalong roads (Lambin and Geist 2006) and in agriculturalareas. Today, agriculture occupies 38% o the globesice- ree terrestrial sur ace and is the largest land covertype by area (FAOSTAT 2011a).

Since the 1992 Rio summit, the global communityhas become increasingly aware o the environmentand the need or sustainable development througha range o practices and policies, including reducedde orestation, increased environmental monitoring,agricultural intensi ication, restoration o degradedlandscapes, reduction o environmental pollution, andpayment or environmental services (PES). Followingthe publication o the Millennium EcosystemAssessment in 2005, sustainable development e ortshave increasingly become more inclusive to ocus on allcomponents o ecosystems. According to Costanzaet al (1997), ecosystems are goods and services providedby living organisms and their habitat with direct andindirect bene it to human populations. According toMA (2005), ecosystem goods and services include:

Pr v s n n s rv c s: Goods provided ood,ber, orage, uelwood, pharmaceutical products,biochemicals, resh water, etc.

S pp rt n s rv c s: Services that maintain theconditions o li e on Earth soil development (conservation/ormation), primary production, nutrient cycling

R at n s rv c s: Bene ts obtained rom theregulation o ecosystem processes water regulation,pollination/seeds, climate regulation (local and global)

C t ra s rv c s: Intangible bene ts obtained romnatural ecosystem including recreation, landscapes,heritage, aesthetic, etc.

The economic value o supporting and regulating these

services is not well-captured in the market and there orealways undervalued. Additionally, the bene t o the non-marketed ecosystem services is not yet well-known(Balm ord et al 2002). Such a lack o knowledge andapparent lack o economic value poses a challenge ordetermining land use allocation and modeling land usechange. However, what is known about the non-marketedecosystem services suggests that they are likely to havea larger value than marketed ecosystems. For example,Constanza et al (1997) estimated the annual economicvalue o 17 ecosystem services most o which werenot traded in the market and there ore not consideredin the traditional GDP and other economic statistics to be about US$ 38 trillion per year (adjusted or the2000 value). The equivalent GDP in the same year wasUS$32.216 trillion (IMF 2011).

This paper assesses LUCC since 1992 and exploresrelated actors which should be addressed to ensuresustainable development. It starts by analyzing whatscience tells about LUCC, and then examines howLUCC happens in practice and how countries andthe global community are managing the competingdemand or land. The third section o the paperexamines what we know about the e ectiveness odi erent land management systems. The ourthsection reviews the state o knowledge on LUCCmodels. The inal section discusses prospects or theuture and draws policy implications.

Science of land use and land cover changeWe explore what science tells us about LUCC using themajor terrestrial land use types, namely orests andagriculture. The two terrestrial ecosystems are closelylinked to each other and to water and biodiversity. Therelationships o the ecosystems are complex and thisdrives the increasing need to analyze and treat them asecosystem services, the source o provisioning, culturaland regulating services that are crucial to humanwel are (MA 2005; Rockstrmet al 2009; Constanza2011). In e ort to determine the trend o all ecosystemservices, we examine the ecological ootprint, a resourceaccounting tool that compares two opposing processes:1) the biological capacity o land and sea area toproduce ood, ber, timber, energy, absorb by products oconsumption and provide space or in rastructure usingthe prevailing technology and 2) the demand or theseecosystem services by a given population (Kitzes andWackernagel 2009).

7/28/2019 1124 Land Use

12/82Sustainable land use for the 21 st century | Part 1 PAGE3

Forest density, tree density per hectare, is the secondactor o interest in orest change. In the past twodecades (1990-2010), orest density has increasedglobally. The increase in orest density was mostpronounced in North America and Europe; the increase inA rica and South America was only modest (Rautiainenet al 2011). Overall, orest density increased in 68countries and accounted or 72% o the global orestarea and 68% o global carbon mass (Rautiainenet al 2011). In Asia, orest density increased in 1990-2000but decreased in 2000-2010 as orest area increasedsigni cantly, largely rom a orestation in China.Conversely, de orestation rates and net losses in Southand Southeast Asia increased (Rautiainenet al 2011).

F r st trans t ns

Forest transition theory o ers some explanation behindthe trend o orest extent and density across countriesand regions. As it will be seen below, this theory habeen tested empirically and shown to be valid withsome evidence showing di erent patterns (e.g. seeMey roidt and Lambin 2011). Forest transition ocuseon the orest stock change that has a predictablerelationship with economic development (Mathe1990). Forest transition has three phases representedby the environmental Kuznets curve. In the rstphase, increases in de orestation as the economy andpopulation grow prompt greater demand or agricultura

F r sts 1

In 2010, the orest area covered 31% o global land areaand was equivalent to 0.6 ha per capita (FAO 2010).Globally, de orestation, the permanent clearing o orests,decreased by almost 20% rom 16 million ha year-1 in1990-2000 to 13 million ha year-1 in 2000-2010 (FAO2010). Sub-Saharan A rica (SSA) and South Americacontributed the largest share o de orested area in the last

two decades (Figure 1). Recently, Brazil and Indonesihave signi cantly reduced de orestation rates. Australiasaw an increase in orest loss, largely due to drought andorest res (FAO 2010). The orest trends in Australiaunderscore the role played by biophysical actors in orestcover trends. As will be seen in the discussion o orestransitions and drivers o LUCC, orest trends acrossregions are driven by economic development, governmenpolicies and other socio-economic actors.

FiguRe 1 Change o agricultural and orest area, 1992-2009.

12

10

8

6

4

2

0

-2

-4

-6

-8

% C

h a n g e

Agriculture Forest

L A C

N o r t h A m

e r i c a E U

E a s t E

u r o p e S S

A

S o u t h e r n

A s i a

E a s t e r n

A s i a

C e n t r a

l A s i a

O c e a n

i a L D

C W o

r l d

Note: Change computed as ollows , where y1 = average area 1992-2000 and y2 = average area 2001-2009.

LAC = Latin American Countries; SSA = sub-Saharan A rica; EU = European Union.Source: FAOSTAT data.

y2 - y1y1y1 *100

7/28/2019 1124 Land Use


and orest products. In the second stage, migrationto urban areas, increasing rural labor wage rates, andintensi ying arming reduces labor demand. The value oorest products also increases. In the third phase, orestrecovery begins when a threshold period is reachedduring which the value o orest products has increased,prompting land owners to protect and/or plant trees(Cookeet al 2008) (Figure 2).

In most o the tropics today, countries are in phase two,when urban population growth and agricultural exportsare the primary drivers o de orestation (DeFrieset al.2010). For some countries (Ethiopia, Haiti and Togo),orest transition has not occurred largely due to a lacko alternative employment and/or institutions thatcould enhance tree planting (Ruddelet al 2005). Others(Burundi, El Salvador, Rwanda and Sierra Leone), havebeen embroiled in the insecurity o war, which has led tode orestation. In other countries (Brazil, Indonesia andCameroon), strong orest product markets throughoutthe early 21st century caused de orestation despitesigni cant economic development (Ruddelet al 2005).Recently, however, Brazil has reduced de orestation bymore than two thirds in only ve years due to aggressivepolicies and international cooperation. Increases in thecountrys orest plantations or paper, charcoal and chipboard production have reduced pressure on pristineorests. In the 1990s, 38% o all countries experiencedan increase in orest area a ter de orestation, suggesting

that they reached the threshold and moved towardsphase three (Ruddel et al 2005). Most Europeancountries and all o North America experienced orestrecovery in the 20th century largely due to generalindustrialization and economic development. Forestrecovery in Asia, including the recovery in China, Indiand Bangladesh, exhibits a di erent pattern. Ruralpoverty in these countries remains entrenched, but theincreasing value o orest products has spurred ruralcommunities and the government to plant trees (Fosterand Rosenzweig, 2003; Fanget al 2001). Governmenttree-planting programs, which were implementedto head o fooding, wind storms and other disastersattributable to de orestation, also helped to increaseorest area (Ibid).

However, the predictive power o the orest transitionmodel is being a ected by globalization and theincreased role o international trade. Similarly,government policies, such as the one implemented inBrazil to control de orestation at the municipal levelthrough the creation o a black list o municipalities thatmost contribute to de orestation, could also change theorest transition by astening the recovery process. Forexample, recent analysis suggests that the relationshipbetween rural populations and orest cover hasweakened as globalization has linked well-capitalizedranchers, armers and loggers, and their products withdistant markets (Rudelet al. 2009).

FiguRe 2 Forest transitions stylized model.

Source: Authors illustration.

P e r c e n t o

f f o r e s t c o v e r

12

10

8

6

4

2

0

-2

-4

Time

1 2 3 4 5 6 7 8 9 10

7/28/2019 1124 Land Use


Box 1 Regreening o the Sahel in Niger.

A r c t r

Agriculture is the leading orm o human appropriationo net primary production (HANPP), which is humansharvest o photosynthetic products or alterationo photosynthetic production. HANPP in luencesbiodiversity, reduces energy low to non-human speciesand in luences the provision o other ecosystems

(Harbelet al 2007). Harbelet al (2007) estimate thateven though cropland occupies only 12% o the globaland area, crop harvest accounted or 49.8% o theHANPP in 2000 and grazing accounted or 28.9o HANPP in the same year. Rates o agriculturaexpansion are decreasing globally but still expandinin SSA and Latin American countries (LAC) (Fig1). It is estimated that 70% o the grasslands, 50% o

A major conclusion rom the orest transition analysisdiscussed above is that while economic developmentand orest product scarcity could trigger an increase inorest area, other socio-economic characteristics mayinhibit orest recovery. As shown in Asia, governmentintervention could help orest recovery. In Niger orexample, the government passed a statute (rural code)giving land owners tenure security o any tree that theyplant or protect (Larwanou, Abdoulaye, and Reij 2006).It is estimated that at least 3 million hectares o landhave been rehabilitated through tree protection, whichallowed or natural regeneration (Adamet al 2006).

However, the rural code was not the only deciding actothat led to this remarkable success. The prolongeddrought that spanned the 1970s and 1980s led to loss otrees, increasing the price o tree products. This providedstrong incentive to armers to plant and protects trees.Planted orest area as a share o total orest area in Nigerwas 12% in 2010 and was among the highest in SSA (FA2010). As discussed in Box 1, this achievement was result o a combination o e orts by local communitieschange in government policies and statutes, support romNGOs, and religious organizations and environmentstress, which prompted communities or a solution.

Regreening the Sahel in Niger is a success story due to its remarkable progress in planting and protecting trees that resulted froma combination of initiatives by the government, local communities, donors, NGOs, and religious organizations. Starting in the 1970s,in response to extensive vegetation loss due to droughts that lasted until the 1980s, the Nigerien government aggressivelypromoted tree protection and planting. One measure was recasting Independence Day as National Tree Day. Additionally, sincethe 1980s, more than 50 government programs including the Special Program of the President and the Projet de Gestion desRessources Naturelles (Natural Resources Management Program) were promoted by the government, NGOs, and donors (WorldBank 2009). NGOs and religious organizations involved in these efforts mobilized communities to plant and protect trees. They alsobuilt the capacity of local communities to manage natural resources. For example, a religious organization initiated thefarmer-managed natural regeneration (FMNR)in which communities protect or plant new trees and in return harvest fuelwood,fodder, nitrogen xation from leguminous trees, windbreaks, and other ecosystem benets (Reij, Tappan, and Smale 2008).

The government also revised its institutions and passed the rural code in 1993. This legislation gave customary leaders moreland management power and encouraged them to plant and protect trees and to benet from such efforts without governmentintervention. The forest policy gave landholders tenure rights to trees that they planted or protected (Yatich et al . 2008; World

Bank 2009). The changes provided incentives for communities to plant and protect trees and helped them to cope with riskyagricultural production.

Additional policy changes and efforts by donors and NGOs also followed the 1970s1980s drought, creating a new value for trees.Firewood and water as well as livestock were in short supply following the drought. The loss of livestock wiped out the traditionalstrategy of using livestock as buffer stock against shocks (Fafchamps et al 1998) especially in northern Niger, where trees areused as fodder during the dry season. People responded to this challenge by protecting growing trees instead of cutting them, ashad been the case in the past. Hence, tree scarcity signicantly affected the livelihoods of rural communities, prompting them tochange from land clearing to tree protection.

Studies carried out to understand the drivers of regreening of the Sahel found that villages where tree planting and protectionprojects were operating were much greener than what could be explained by change in rainfall (Herrmann, Anyamba, and Tucker2005). It is estimated that villages with FMNR had 1020 times more trees than they had had before FMNR started. Contrary toexpectations, tree planting and natural regeneration in villages with higher population density were higher than in villages with

lower population density (Reij, Tappan, and Smale 2008).

7/28/2019 1124 Land Use


the savanna, 45% o the temperate deciduous orestsand 27% o the tropical orests have been clearedor agriculture (Foleyet al 2011). Between 1992 and

2009, agricultural land area increased by about 4% inSSA-- the largest increase in all regions considered(Figure 1).

FiguRe 3 Source o growth o agriculture, 1961-2005.

Notes: LDC = Least Developed Countries; SSA = Sub-Saharan A rica; NENA = Near East and North A ricaSource: Computed rom Bruinsma (2009).

-20

-10

0

10

20

30

40

50

60

70

80

90

L D C

S S A

N E N A L A

C

S o u t h

A s i a

E a s t

A s i a

W o r l d

Arable land expansion Increase in cropping intensity Yield increase

% C

o n t r i b u

t i o n o

f a g r i c u l t u r a l p r o d u c t i o n

About 80% o agricultural expansion in the tropicsreplaces orests, leading to serious consequencesor biodiversity, greenhouse gas emissions and otherenvironmental outcomes. Theory posits that populationgrowth induces intensi cation (Boserup 1965) butsuch intensi cation occurs only i prices, markets andother socio-economic conditions avor such decision(Mortimore and Harris 2005; Boyd and Slaymaker2000). I price and market conditions do not permitintensi cation, population density has been associatedwith land degradation (Grepperrud 1996; Scherr 2000).

Agricultural intensi cation has occurred in all regions including SSA. Figure 3 shows that rom 1961-2005,crop yield accounted or 77% o the global increasein agricultural production but in SSA, contribution ocrop yield to total production was only 38%, the lowestamong all regions. The small contribution o yieldincrease to total production was due to the poor marketconditions that provide incentives or armers to investmore in increasing productivity. The global yield ocrops increased by 47% between 1965-1985, but onlyincreased by 20% between 1985-2005 (Foleyet al2011).

Wh r has a r c t r b n pand n ?

Agriculture can only expand in an area that providesthe ecological requirements o crops or livestock.FAOs Global Agro-Ecological Zone (GAEZ) de nesuitable land as land with soil, terrain and climatecharacteristics which meet the crop productionrequirements with speci ed input levels (Fischeret al., 2002). On a regional scale, the countries o LatinAmerican and Caribbean (LAC) and SSA account othe largest share o arable land (Figure 4). However,the largest share o arable land has already been putunder use and available land or expansion is limited.About 90% o the remaining 1.8 billion ha o arable landin developing countries is in LAC and SSA (Bruinsm2009). Seven countries (Brazil, Democratic Republico the Congo, Angola, Sudan, Argentina, Colombiaand Bolivia) account or about 50% o the remainingsuitable land (Ibid). Regions that have virtually runout o suitable land or expansion include South Asiaand the NENA (Near East/North A rica). Expansion oagricultural land in such regions requires investment inirrigation or other soil amelioration measures.

7/28/2019 1124 Land Use


It is also in countries with large arable land area that thereare still large gaps between agricultural yield potentialand actual yield. Such a large gap provides the potentialor increasing agricultural production to cater to theincreasing demand or agricultural products. As will beseen below, closing the wide agricultural productivity gaprequires signi cant investment to address constraintswhich lead to the low agricultural productivity.

FiguRe 4 Contribution o regions to global suitable land.

Source: Bruinsma (2009).

25%

2%

25%5%

9%

21%

12%

Sub-Saharan Africa

Near East/North Africa

Latin America

South Asia

East Asia

Industrial countries

Transition countries

Wat r r s rc s

There is abundant supply o reshwater (47.97 millionkm3) per year but only a small share is available at theright time and place. For example, only a third o the110,000 km3 annual precipitation reaches rivers, lakesand the aqui ers, o which only 12,000 km3 is availableor irrigation, domestic and industrial use (SIWIet al 2005). Two thirds o precipitation is absorbed assoil moisture or evaporates (Ibid). Only 0.79% oreshwater is not rozen in ice or glaciers and o that, alarge share is groundwater.

Irrigation water use has tripled in the past 50 years,and irrigation accounts or 70% o global reshwaterwithdrawals (UN water 2009). One estimate amongmany estimates with di erent volumes o currentwater use or ood production is 6800 km3/year(Shiklomanov, 2000). The world population growth o

80 million per year translates to an additional annualdemand o reshwater o 64km3 (UN water 2009). Itis estimated that the growing human population wilrequire more ood, which will translate into doublindemand or water or agriculture rom the currentlevel o 7,130 km3 to 12,050 to 13,500 km3 in 2050(CA 2007). According to another estimate, additionwater withdrawal and use equivalent to 5,600km3/year would be required to eliminate hungerand undernourishment and to eed the additionalthree billion inhabitants in 2050 (Falkenmark anRockstrm, 2004), which is about three times thewater used or irrigation today (Shiklomanov, 2000)

Water scarcity is already evident in dry areas. Moldenet al(2007) estimate that about 25% o the earths river basinsrun dry be ore reaching the ocean due to water use. In2000 about 2.3 billion people lived in river basins with watstress, i.e. had access to less than 1,700 m3 per capita/year,below which, disruptive water shortages can requentlyoccur (Revenga,et al2000). By 2025, Revengaet al (2000)estimated that 48% o the global population will have watestress under business as usual.

The irrigation water supply reliability index (IWSR) measure o availability o water relative to ull waterdemand or irrigation is also projected to decline rom 0.7globally in 2000 to 0.66 by 2050 (Ringler and Nkonya 201Additionally, water availability is expected to be less reliabin arid and semi-arid areas due to climate change, which expected to increase intra-annual variability in precipitatioand increased severity and requency o droughts (Meehlet al. 2007). It is estimated that climate change will accountor 20% o the increase in global water scarcity (UN 2003)

The increasing water scarcity calls or strategies toaddress the water stress an aspect that could helpincrease agricultural productivity, which in turn will reducconversion o natural ecosystems to agriculture. We analyzthese strategies in the last chapter.

B d v rs ty

Biodiversity provides a variety o ecosystem services, whicor a long time, have been ignored or undervalued. Greatebiodiversity ensures more stable and resilient ecosystemsThe Chennai Declaration states that biodiversity musbe conserved because it is the raw material or ood andhealth nutrition and provides material or biotechnologindustry (Kooha kan and Altieri 2011). Hence changes the abundance and diversity o species may have serious

7/28/2019 1124 Land Use


impacts on human wel are. For example, up to 80% opeople rom developing countries rely on wild fora andauna or health care and wild meats provide 30-80% oprotein or many rural communities (Nasiet al2011).

Realizing the rapid loss o biodiversity and its potentialimpact on ecosystems and consequently human wel are,193 o the 194 countries in the world are signatories othe CBD and 170 countries have already prepared theirnational strategies and action plans (NBSAPs). The CBD,rati ed in 1992, set 11 goals2 with 21 speci c sub-targets tobe achieved by 2010 (CBD 2010). A 2010 evaluation o theachievement o the sub-targets done in 2010 showed thatwhile no single goal has been ully achieved; there has beensigni cant reduction in the rate o biodiversity loss or mosto the sub-targets (CBD 2010).

Since 1992, the protected area in all regions hasincreased signi cantly (Figure 5). Globally, theprotected area increased by 38% rom its level in1992. The increase was especially large in the MiddleEast and North A rica (MENA) region, due to theour old increase by Saudi Arabia, whose o protectedarea increased about 4 times rom about 148,000km2 in 1993 to 588,000 km2 in 1994. The Latin andCaribbean countries (LAC) countries, which accountor the largest share (25%) o global protected area,saw the second largest increase. This increase largelywas due to Brazil, where protected areas increasedalmost three old - rom 812,000 km2 in 1992 to2.242 million km2 in 2010. However ambitious plansto accelerate economic development in Brazil couldthreaten protected areas.

FiguRe 5 Change in protected area in global sub-regions rom 1992 level.

Notes: CA=Central Asia; EA=East Asia; EEU=East Europe; LAC=Latin America; MENA=Middle East and Northern A rica; NA=North AmericaAsia; SEA=Southeast Asia; SSA=sub-Saharan A rica.

Source: IUCN and UNEP-WCMC (2011).

0

20

40

60

80

100

120

CA

LAC

EA

MENA

EEU

NA

EU

OCEANIA

SA

SEA

SSA

World

1 9 9 5

1 9 9 6

1 9 9 7

1 9 9 8

1 9 9 9

1 9 9 4

2 0 0 1

2 0 0 2

2 0 0 3

2 0 0 4

2 0 0 5

2 0 0 0

2 0 0 7

2 0 0 8

2 0 0 9

2 0 1 0

2 0 0 6

1 9 9 3

% I

n c r e a s e

Despite the impressive increase in protected area, losso biodiversity remains quite high. The 2009 report othe Convention o Biological Diversity (CDB) showsthat 36% o the 47,677 species already assessed arethreatened by extinction (Figure 6), i.e., under thecurrent trends, the species will become extinct (CBD2010). Rockstrmet al (2009) also report that onaverage, more than 100 species extinction per millionspecies per year (E/MSY) are lost, a level which ismore than 100 times the planetary boundary (10 E/

MSY) deemed to be earth systems sa e operatingspace or human wel are. Current rate o extinctionis 100-1000 higher than the Holocene (pre-industrial)age level (0.1 1 E/MSY) (Ibid).

7/28/2019 1124 Land Use


FiguRe 6 Global biodiversity conditions.

Source: IUCN (cited by CBD 2010).

Even though recent e orts to increase extent o orestcover through re orestation and a orestation programshave helped to reduce net de orestation, they do not ullyrestore lost biodiversity since such natural biodiversitywas built over hundreds o years and is composed ocomplex and diverse biomes. Species that have shownthe most rapid decline include birds, mammals, andamphibians (hunted or ood) and medicinal plants(Ibid). Terrestrial biodiversity losses have been drivenby habitat loss and degradation through slash-and-burn clearing, orest res, land use conversions, over-exploitation o plant and animal species, climate change,pollution, and invasive and alien species (CBD 2010).

Projections show that the impact o climate changeon biodiversity loss is expected to increase in uture.For example, the loss o ice sheets and the meltingo perma rost in the arctic and Antarctic regions arethreatening polar biomes.

There are underlying causes o biodiversity loss. Forexample, a study covering 73 countries examined therelationship between income and the threat o extinctionor our major species (mammals, birds, plants andreptiles). It showed a U-shaped quadratic relationship(Perrings and Halkos 2010) i.e., the Kuznets curvecomparable with the orest transition pattern discussed

above. The turning points di er or each country but thresults were very robust (Perrings 2010).

ec syst m s rv c s and c ca f tpr ntsEcosystem services are the components o nature,directly enjoyed, consumed, or used to yield human welbeing (Boyd and Banzha 2006). The ecosystem servicetrend is analyzed using the concept o ecological ootprintThe ecological ootprint is a measure o the biologicallproductive land area and water required to provide oodeed, ber, timber, energy and to absorb CO2 waste undercurrent technology (WWF 2012). The ecological ootprrepresents the demand or ecological services. The supplyo ecological services is represented by biocapacity, which the area o productive land and water available to produceresources or absorb CO2, under current managementpractices. The net biocapacity is the di erence betweenbiocapacity and ecological ootprint. Estimates show thasince the mid-1980s, the world entered an ecological de citi.e., earths biological capacity to produce ecosystem goodand services and human demand or provisioning serviceand regulating services -- mainly absorption o CO2 - wassurpassed (GFN 2010). The ecological ootprint is noestimated to be 45% higher than the earths biocapacityi.e., it takes the earth about 1.5 years to produce what irequired by the global human population and to absorb thCO2 produced using the current technologies (Ibid).

The ecological de cit is highest in North America anEurope, where carbon emissions account or the largestdemand or land area to absorb CO2. Oceania and LAChave the largest surplus ecological balance, whileA rica has a delicate balance with only cropland havina slight negative balance o 0.06 global hectares (gha)Global population growth has been the largest actorcontributing to the ecological de cit (GFN 2010). Thpopulation dynamics in SSA poses the biggest challengsince it is only in this region that population is expecteto continue growing beyond 2100 and that the currengrowth rate is the highest (UNFPA 2011).

Main drivers of change and sources of pressure on land-useP p at n r wth

Population has been the major driver o agriculturaexpansion (Foleyet al., 2011; Bruinsma, 2009; Ramankuttet al., 2002). Even though the positive correlation betweepopulation growth and cropland is expected due to

Extinct in the wild

Critically endangered

Endangered

Vulnerable

Near threatened

Least vulnerable

No data

2%

14%

40%

8%

19%

10%

7%

7/28/2019 1124 Land Use


increasing demand or ood, it is also true that people tend tosettle in areas suitable or agriculture (Ibid). Technologicaldevelopment and international trade have weakened therelationship between population and expansion as wellas settlement in areas o arable land (Ibid). For example,the Green Revolution in Asia led to a much slower areaexpansion than would have been the case without theproductivity improvement. Borlaug (2000) estimates that icereal yields o 1950 had been unchanged through 2000,a total o 1.8 billion ha o land would have been convertedto cropland to meet the cereal demand. Instead, cropyield increases accounted or 77% growth in agriculturalproduction between 1961 and 2005. Expansion into arableland accounted or only 14% o growth (Figure 3).

incr as n nc m

Increases in income and changing ood pre erences areimpacting the quantity and type o oods that the worldconsumes. Over the past 60 years, the global averageannual per capita growth o ood and ber consumption dueto income growth was 0.27% (Buchananet al., 2010). Thegrowth in developing countries is much higher than is thecase in high-income countries (Ibid). Increased incomeshave particularly increased the demand or livestockproducts, ruits and vegetables. Delgadoet al. (1999)estimated that the demand or milk in developing countrieswill increase annually by 3.3% rom 1993 to 2020. SouthernAsia will account or 60% o the increase while SSA willaccount or only 17% o the increase (Ibid). Increasingdemand or livestock orces extensi cation since thedemand or land and water or livestock is much bigger thanor crops. For example, while 15 m3 o water are requiredto produce one kilogram o bee , only 0.6 m3 o water isrequired to produce a kilogram o cereals (Figure 7).

FiguRe 7 Water requirement o crops and livestock products.

Source: SIWI (2005).

0

2

4

6

8

10

12

14

16

C e r e a

l s

P o u l t

r y L a

m b B e e f

P u l s e

s , r o o

t s

& t u b

e r s

C u b i c m e t e r s p e r k g

15

10

3,5

10,6

urban zat n

Urban areas are growing ast and increasinglyoccupying larger land areas. The urban populationsurpassed the rural population in 2008 (UN 2008;Tolle son 2011) and it is expected that by 2050, theurban population will account or 70% o the totalglobal population (Seto and Shepherd 2009). Citiesoccupy less than 3% o the global land area but theyaccount or 78% o carbon emissions, 60% o potablewater use and 76% o industrial wood consumption(Grimmet al 2008). However, measurement o urbanarea is not well captured by current LUCC models(Olsonet al 2008).

infrastr ct r d v pm nt

Road development reduces transaction costs andincreases access to natural resources. Hence, holdingother actors constant, road development could leadto de orestation as observed by Nelson and Hellerstein(1997) in Central America. However, recent works haveshown that such a pattern holds only in countries withweak institutions. In countries with strong institutions,road development does not a ect de orestation. Forexample, between 1990- 2010, orest extent increasedin Europe and China and has remained almost constantin North America (CBD 2010). All these countriesand regions have good road in rastructure, stronginstitutions and high or middle incomes.

Studies have shown that access to road, electricity andcommunication in rastructure is strongly correlatedwith agricultural total actor productivity (Kamara2008; Foster and Briceno-Garmendia 2010). Thismeans that poor road in rastructure could lead to lowagricultural productivity, which in turn could lead tothe conversion o orest and other natural ecosystemsto agriculture. However, poor road in rastructure couldalso hamper the cutting o orests or timber. Forexample, in Central A rica, the relatively low rate ode orestation is correlated with weak in rastructures.The SSA region has poor road in rastructure, whichhave contributed to low agricultural productivity,which in turn has led to conversion o virgin land toagriculture. As shown in Table 1, SSA has a largein rastructure de cit as compared to other developingcountries. Such large de cit contributed to thesmallest contribution o yield increase to agriculturalproduction in 1961-2005 and to the astest decline oper capita arable land ( gure 3).

7/28/2019 1124 Land Use


TABle 1 A ricas in rastructure de cit and cost.

A ricaOther developingcountries

Paved road density (km/km2 o arable land)a 0.34 1.34

Population with access to electricity (%)a 14 41

Population with access to improved potable water (%)a 61 72

Power tari s ($/kwh) 0.02-0.46 0.05-0.1

Transportation cost ($/ton/km) 0.04-0.14 0.01-0.04

Tari s o urban potable water ($/cu m) 0.86-6.56 0.03-0.6

Note:aExcludes medium income A rican countries (South A rica, Kenya, Botswana, Gabon, Namibia, Cape Verde, etc.) and is compared to oincome countries. The rest o the statistics re ers to entire A rica and other developing countries.

Source: Foster and Briceno-Garmendia (2010).

Access to markets and in ormation helps land usersto make in ormed decisions. Farmers with access tomarket in ormation will respond to market signals andcould respond avorably when they have better access tomarket and in ormation. For example, the amous morepeople less erosion study in Kenya by Ti enet al (1994)was a result o better market access by land users inMachakos district, which is only 54 km away rom thecity o Nairobi (Boyd and Slaymaker 2000).

F d pr c s and pr c ast c ty f d mand

Food prices provide incentives or armers to convertland to ood production. The recent high ood prices andthe consequent land grabbing illustrate this pattern.Large international deals were made ollowing the oodprice spikes in 2007-08 and 2010-11 (Figure 8), whichwere driven by a variety o actors that vary across thedi erent agricultural commodities. Among these areinvestment-driven speculation, the government policyresponse and the news media (Andersen and Watson2011) as well as the rapid increase in bio uel productionrom maize in the U.S. (Rosegrantet al 2008; Ba esand Hanniotis 2010), particularly in the run up to the2007-08 price spike. A more detailed discussion othe actors underlying the 2007-08 ood price spikeis given by Headey and Fan (2011), who synthesize anumber o ndings that have emerged in recent yearsas more data has become available. The trend that hasbeen observed in which rich countries with arable and/or water de cits acquire lands in developing countries

with abundant arable land and/or water has causedconcern among policy analysts and researchers (vonBraun and Meinzen-Dick 2009; von Braun 2011) as the socio-economic and institutional implications osmall-holders (Bomuhangiet al. 2011) and women(Behrmanet al 2011). Additionally, several investmenunds have indicated their intention to invest more thanUS$ 2 billion in land or ood production in A rica (IbiGlobally, about 46.5 million ha were acquired betwee2004-2009 in 81 countries (Deiningeret al 2001;Toulmin et al 2011). Such large land deals changeland use. Many recent land deals, or example, havedisplaced community-managed lands which combinshi ting cultivation, livestock, and orest resources(Toulminet al 2011) with monocrop systems or otherlarge-scale production systems -- which in turn reducebiodiversity. A recent study observed that oreign lanacquisition was more likely to occur in countries witabundant land and weak land governance, supportingthe growing concern regarding the lack o protectiono vulnerable groups against oreign land acquisition(World Bank 2011a).

Price elasticity o demand or ood also drives LUCCFood demand is price inelastic in high-income anairly elastic in low-income countries (Hertel 2011).Thmeans high ood prices will lead to greater incentives oproducers to convert land.

7/28/2019 1124 Land Use


FiguRe 8 Global Food price Index trend, 1961-2010.

Source: FAOSTAT.

P c s at nat na and nt rnat na v

Policies both at national and global level have a largeinfuence on LUCC. Recent studies have shown thatincreasing ood prices have prompted importing countriesto change their trade policies to protect consumers whileexporting countries have changed trade policies to the

bene t o armers. For example, the Global Trade Alert(http://globaltradealert.org/) ound that 45 ood exportingmeasures and 85 import measures were changedbetween November 2008 and November 2011. Theimpact o such trade policy changes on the internationalrice, wheat and maize international prices were estimatedto be respectively 31%, 13% and 18% (Table 2).

TABle 2 Contribution o domestic ood policies (trade tax) on international price spikes o major crops, 2005-08.

Cr p Tta (P rc nt) of wh ch: of wh ch:

R c 31 1 30 13 18

Wh at 13 6 7 6 7

Ma z 18 8 10 7 11

High-incomecountries

Developingcountries

Importingcountries

Exportingcountries

Source: Anderson and Nelgen (2012).

The impact o the price change due to such policies couldbe elt through the price impact on LUCC and throughthe direct impact. Minimizing the negative impacts ocountry-level policies on global or regional communityrequires a global action through the World TradeOrganization (WTO) and other orms o internationalcooperation (Andersen and Nelgen 2012 and Martin andAnderson 2012).

Countries have also used policies that encouragearmers to use or not to use land or to improve or

degrade land. For example, the U.S. ConservationReserve Program (CRP) allowed 12.5 million hain 2005 (Wunderet al 2009; Claassen et al 2009).As will be seen below, systems o payment orecosystem services implemented by countries andthe international community have shown promisingresults o land improvement and LUCC in general.However, the bene its o some PES programs havebeen questioned. For example, Wnscheret al (2008)observed limited additionality o PES in Costa Ricadue to the countrys low de orestation rate.

F o o d P r i c e I n d e x , r e a l t e r m s

( 1 9 9 0 = 1 0

0 )

0

50

100

150

200

250

300

350

400

1 9 6 1

1 9 6 3

1 9 6 5

1 9 6 7

1 9 6 9

1 9 7 1

1 9 7

3

1 9 7 5

1 9 7 7

1 9 7

9

1 9 8 1

1 9 8 3

1 9 8 5

1 9 8 7

1 9 8 9

1 9 9 1

1 9 9 3

1 9 9 5

1 9 9 7

1 9 9 9

2 0 0 1

2 0 0 3

2 0 0 5

2 0 0 7

2 0 0 9

7/28/2019 1124 Land Use


land t n r and pr p rty r hts

Tenure security provides incentives or long-term landinvestments such as tree planting and soil and waterconservation structures (Feder 1987; Alstonet al 1995)and LUCC. However, it has also been established that landholders with insecure tenure may plant trees or engagein other long-term investments to enhance their securityor as a method o claiming ownership (Place and Otsuka2002; Braselle et al 2002). Other studies have also shownthat investments by armers with customary land tenurewere comparable or greater than investment by armersholding land with secure title deeds (leasehold or reehold)

(Toulmin and Quan 2000; Deininger 2003). Additionasecure land tenure is a necessary but insu cientcondition to determine investment or LUCC. Other actodriving LUCC and land investment incentives (e.g. thodiscussed above) play a key role.

Global studies have shown diverse systems o landownership, tenure, and land rights exist acrosscontinents, with di erent degrees o tenure security. Arecent study by Bruceet al (2010) showed that areaswith strong land tenure security have relatively loweliving biomass carbon density than areas with ricbiomass carbon density (Figure 9).

FiguRe 9 Relationship between land tenure security and living biomass carbon density.

Source: Bruceet al. (2010).

B n r y

Bioenergy places a new pressure on land demand. Theglobal growth in bio uel comes in the ace o increasingscarcity o energy resources and growing energy demandor transport uel and other productive uses. A numbero OECD countries have engaged in large-scale bio uelsproduction as a way o exploiting renewable resources tosupplement and diversi y their domestic energy port olio.North America has been the largest consumer o bio uels,worldwide, ollowed by Latin America and the EuropeanUnion (IEA, 2008; von Braun 2008). Together, Brazil andthe U.S. account or over 90% o the worlds ethanol

production; the U.S. overtook Brazil as the worlds leadiproducer o ethanol in 2004. Biodiesel, on the other handis mostly concentrated in the EU (IEA, 2008). Besides thdesire or enhanced energy security and diversi cation, amajor policy motivation or bio uels production has alsbeen to reduce greenhouse gas (GHG) emissions romossil uels, especially in the EU. The actual GHG emissiosavings, however, depends heavily on the productiopathway, and is a source o active debate and research.

The extent o land use changes that are caused by large-scale bio uels production has generated a great deal odebate within the energy and environmental policy an

7/28/2019 1124 Land Use


research communities. A recent study showed that othe 203.4 million ha o land acquired globally since 2000,66% was obtained rom A rica and that o the 71 millionha veri ed by the study, 40% were acquired or bio uelproduction while only 25% was or production o cropsor ood, 3% or livestock production and 5% or non- oodcrops such as cotton (Anseeuwet al 2012). This refectsthe land competition and potential or compromising oodsecurity e orts in A rica -- the worlds most ood insecureregion--and carbon sequestration i such acquisitions arelocated on orest land. Hertel (2008) estimated that U.S.and EU bio uel mandates will increase crop land cover atthe expense o orest and pasture cover (Table 3).

TABle 3 Predicted change in global land use due to US& EU mandates.

uS eu-27 Braz

2001-2006 (% chan )

2006-2015 (% chan )

0.3 0.7 1.1Cr p

-0.7 -2.1 -2.6F r st

-1.4 -2.3 -2.2Past r

0.8 1.9 2Cr p

-3.1 -8.3 -5.1F r st

-4.9 -9.7 -6.3Past r

Source: Hertelet al (2008).

Additionally, the mandates will lead to greater use oertilizer and other agricultural inputs, which in turn couldlead to environmental pollution. For example, Britz andHertel et al (2011) estimate that by year 2015, nitrogenertilizer use will increase by 0.14% in EU-27 due to EUbionergy mandates compared to its level in 2001.

land d radat n

Land degradation, de ned as loss o the capacity o landto provide ecosystem services, a ected about 24% o theglobal land area between 1981 and 2003 (Baiet al., 2008).This is equivalent to a degradation o about 1% o globalland area each year or about 12 Mha (UNCCD, 2011). Thisarea could produce 20 million tons o grain each year or 1%

o the global annual grain production o 2.241 billion tons(UNCCD 2011; USDA 2011). Globally, 1.5 billion peopleon degraded lands. It is also estimated that 42% o thevery poor live on degraded lands (UNCCD, 2011).

Land degradation reduces both land productivity andarable land area. Land area is reduced when landis degraded beyond productive level. A reduction inthe productive capacity o land leads to agriculturalexpansion into orests and other natural ecosystems.Land degradation could also change land use. Forexample, it is common or armers to turn highly degradedcropland into grazing land. Increase in population densityalso contributes to land degradation in developingcountries when armers continuously cultivate landwithout adequate replenishment o soil nutrients. Theper capita arable land area in SSA has decreased morethan in any other regions in the world (Figure 10). Oparticular importance is re, which has a large impact onland cover. Human-induced and natural res all changeland cover signi cantly. The ability to monitor res usinghigh requency satellite observations has improved overthe years (Giglioet al 2009). In 2000, human-inducedres accounted or 3.2% o HANPP (Harbel et al 2007).Naturally occurring res also alter land cover; it isestimated that there are 200 million ha o lands, mainly inthe ar northern boreal orests, that have been degradedby wild res (Minnemeyer, et al 2011). All this underlinethe importance o land degradation in LUCC and how itsprevention could help address the overall impact o landuse change.

FiguRe 10 Trend o loss o arable land area per capitaacross regions, 1961-2009.

Key: SSA = Sub-Saharan A rica; LAC = Latin American and Caribbecountries.

Source: Calculated rom FAOSTAT data using trend line regression.

0

10

2030

40

50

60

70

80

S S A

S o u t h

e r n A s i a

L A C

S o u t h

e a s t

A s i a

W o r l d

L o s s o

f a r a b l e l a n d ( s q

. m e t r e s / y e a r ) 76

24

384646

7/28/2019 1124 Land Use

24/82

How land use changehappens in practice, and how competing demandson land are managed 2.

7/28/2019 1124 Land Use


Introduction

Since LUCC patterns and trends and their drivers varysigni cantly across countries, we use three case studycountries to better understand their dynamics andinterrelationships. We use Brazil, Democratic Republico Congo (DRC) and Indonesia, three countries withlarge rain orests that account or more than 50% othe land area (Figure 11) and are at di erent stageso development. Brazil accounts or 13% o the 2010global orest extent o 4,033.06 million ha while DRCand Indonesia respectively account or 4% and 2% othe global orest extent (FAO 2011a). The Brazilianagricultural sector has been a unique example; itscontribution to the GDP has increased rom 5% in2006 to 6.1% in 2010 (World Bank 2011c) while itsde orestation rate has allen dramatically. DRC is hometo the largest rain orest in A rica. With 68% o its landarea under orest (FAO 2010), the country accounts or34.6% o the regions carbon stock (Bacciniet al 2008).However, orest contributed only 2.3% o DRCs GDP in2006 (FAO 2010). Forest and agriculture in Indonesiadrive land use change. Forest is the largest land usetype in Indonesia. The extent o orest in Indonesiacovers about 53% o land area and Indonesia has thethird largest tropical orest (FAO 2010)3. The sectorcontributes about 2.5% o Indonesias GDP (Ibid). Theagricultural sector, which contributes 16% o the GDP(World Bank 2011c) covers only 22% o the land area(FAO 2010).

FiguRe 11 Contribution o the orest sector to GDP andemployment in case study countries.

Note: Percent o labor employed in orest sector in DRC is not available.Source: FAO (2011a).

Land management in Brazil

Ma n nt r sts at p ay

Occupation o the Brazilian Amazon, herea ter calledthe Amazon, by humans dates several thousand years.For most o the time, human occupation did not a ectthe integrity o the Amazon ecosystems, but in the lastour decades anthropogenic change has proceeded atan unprecedented pace, putting at risk hydrological andbiogeochemical natural cycles. Land occupation o theAmazon has been driven by several players or agents ochanges operating at di erent intensities and in di erentareas and time periods. There are also multiple interestsby these agents o change that can be summarized inour groups.

The rst group is ormed by ederal and stategovernments, large construction rms and politicianswho have interest in accelerated development in theregion. Several governmental development programswere established in the past our decades to constructand pave roads (Fearnside, 2002; Peres, 2001), buildhydroelectric dams to supply the energy demand in thecountry and invest in the development o industrial andmining activities. The second group is crop armers andranchers who were supported by the government in theorm o subsidies and bankers, traders and politicianswho provided credit and other nancial services to theseactivities. The third group includes settlers in publicunoccupied territories who harvest timber and small-scale miners. The ourth group is the leaders o localenvironmental sustainable development programs,advocacy groups and the international community, all owhom aim to promote the environmental conservationand maintenance o ecosystem services and culturaldiversity (Foley, 2007). The national and local-levelsustainable development and advocacy groups includeNGOs, governmental agencies, indigenous and traditionalcommunities and progressive private companies.

The allocation o timber concessions to the nationalorest (Flonas) is also a major objective o the governmentor the sustainable management o orests and to avoidde acto open access. Up to 35 M ha are targeted ortimber concessions.

Due to the interaction o these groups, the Amazonecosystem has undergone rapid changes in the pastour decades. Until 2011, 762,000 km2 o natural orestswere converted to predominantly cattle ranching and

0

20

40

60

80

0

0,51

1,5

2

2,5

3

Brazil Indonesia DRC

Contribution to GDP (%)

% of labor employed

Forest extent as % of land area C o n t r i b u t i o n t o

G D P

, l a b o r e m p l o y e d

( p e r c e n t )

F o r e s t e x t e n t a s

% o

f l a n d a r e a

7/28/2019 1124 Land Use


crop production (Inpe, 2009). This implies that 19%o the original Amazon orests were cleared. The paceo orest conversion varies through years, but in thelast three decades (i.e. 1980-2009) o cial statisticso de orestation shows that 4-5% o the original orest

FiguRe 12 Annual de orestation rates and cumulative de orested areas in the Brazilian Amazon Biomes.

Source: Inpe, Prodes Project @ http://www.obt.inpe.br/prodes/.

was lost per decade. Most o the orest conversion (72%happened rom 1980 to 2011 (Figure 12). Additionallselective logging, which is mostly predatory and illegaand orest res a ect an area o the same magnitude ode orestation (Asner, 2005; Peres, 2006).

Favorable conditions to initiate de orestation at a largescale in the Amazon were set in the late 1960s. By thattime, the region had very low human population densityand this made the government to promote its occupationwith new road in rastructure and settlement inuninhabited territories. This was done in order to developand integrate the region to the other parts o the country.The government also aimed at consolidating Braziliansovereignty over its vast territory o 5 million km2. TheNational Integration and Land Redistribution plans werethe main policies implemented to achieve these goals.As a result o this, ast development and occupationprocess, de orestation rapidly increased in the 1970s,mainly within human settlement areas and along themain axial roads (Cuiaba-Porto Velho, Santarm-Cuiaba,Belm-Braslia and Transamazonica Roads).

In the 1980s, the government continued to play a majorrole in pushing the development rontier o the Amazonregion by providing subsidies to cattle ranching andsmall-scale agriculture, and maintaining in rastructure

investments. Long-term tax reductions provided incentiveto developers o large- scale mines. Hydroelectric damconstruction also played an important role in attractingmore people to low density occupation areas. So, too, dthe spontaneous occupation o large territories by goldminers in areas like the Tapajs basin and Marab SerraPelada in Par. Logging activities, especially or very higvalue tree species like mahogany, was also scaling updamaging vast areas o orests due to illegal tree harvestingand construction o large road networks (Verissimo, 1995It was when these activities were underway that the rstalarming satellite images showed the dangerous speed orain orest destruction. By the end o 1980s, 420,000 km2 had been de orested (Figure 12).

In the beginning o the 1990s, an NGO movemenemerged in Brazil as a result o growing socioenvironmental concerns. By this time, there was decline in government investments that encouragedthe occupation o the Amazon; nonetheless, the regionabundance o natural resources, such as timber and

7/28/2019 1124 Land Use


gold, and land availability, accelerated the regionsuno cial occupation (Uhl, 1997). Logging roads openedin the late 1980s, providing access to new settlementsand leading to the appropriation o large areas or mainlycattle ranching. The logging boom brought ephemeraleconomic prosperity to dozens o milling centers instates throughout the Amazon. Milling centers likeParagominas, in eastern Par had 240

Documents

1124 Land Use