Lepnwh wi C S AlN 24996 L g ry Q United Nations Environment Programme bo Neha td World Conservation Monitoring Centre UNEP WCMC Global Ecological Forest Classification and Forest Protected Area Gap Analysis Analyses and recommendations in view of the 10% target for forest protection under the Convention on Biological Diversity (CBD) ® 2 - os) WORLD ~ ee # 8” ALBERT-LUDWIGS- Ly my : ! UNIVERSITAT FREIBURG 7) Re eee WWF INSTITUTE OF FOREST AND ENVIRONMENTAL POLI td id INSTITUTE ana 1 Lhe q Xv : <_ | ym J - we ' i ‘ | ; qi i a ic a ; , mi ; : O, '= q ‘ , ~— D | ih : f : . 7 R * | ; he we le : ‘ | i ; aa eu , - + / ‘ el aS i Tene hie a. : . Global Ecological Forest Classification and Forest Protected Area Gap Analysis Analyses and recommendations in view of the 10% target for forest protection under the Convention on Biological Diversity (CBD) Report prepared by: United Nations Environmental Programme World Conservation Monitoring Centre (UNEP-WCMC) World Wide Fund for Nature (WWF) Network World Resources Institute (WRI) Institute of Forest and Environmental Policy (IFP) University of Freiburg Freiburg University Press, May 2008 The United Nations Environment Programme World Conservation Monitoring Centre (UNEP- WCMC) is the biodiversity assessment and policy implementation arm of the United Nations Environment Programme (UNEP), the world's foremost intergovernmental environmental organization. The Centre has been in operation since 1989, combining scientific research with practical policy advice. UNEP-WCMC provides objective, scientifically rigorous products and services to help decision makers recognize the value of biodiversity and apply this knowledge to all that they do. Its core business is managing data about ecosystems and biodiversity, interpreting and analysing that data to provide assessments and policy analysis, and making the results available to international decision-makers and businesses. N N Ny, yY | \0O/ Cl UNEP WCMC Disclaimer: The contents of this report do not necessarily reflect the views or policies of UNEP-WCMC, other contributory organisations and the supporting institutions. The designations employed and_ the presentations do not imply the expressions of any opinion whatsoever on the part of UNEP-WCMC, other contributory organisations and the supporting institutions concerning the legal status of any country, territory, city or area or its authority, or concerning the delimitation of its frontiers or boundaries. Citation: Schmitt C.B., Belokurov A., Besancon C., Boisrobert L., Burgess N.D., Campbell A., Coad L., Fish L., Gliddon D., Humphries K., Kapos V., Loucks C., Lysenko |., Miles L., Mills C., Minnemeyer S., Pistorius T., Ravilious C., Steininger M. and Winkel G. 2008. Global Ecological Forest Classification and Forest Protected Area Gap Analysis. Analyses and recommendations in view of the 10% target for forest protection under the Convention on Biological Diversity (CBD). Freiburg University Press, Freiburg, Germany. ISBN: 978-3-922139-99-7 Contact: Lauren Coad, UNEP-WCMC, Cambridge, UK: Lauren.Coad@unep-wcemc.org Christine Schmitt, IFP, University of Freiburg, Germany: christine.schmitt@ifp.uni-freiburg.de This Project is supported by the German Federal Agency for Nature Conservation (BfN) with funds from the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) Federal Agency for Nature Conservation Preface Deforestation and forest degradation continue at an alarming rate worldwide and jeopardize the tremendous diversity of species and habitat types present in forests around the globe. They also put at risk the large variety of ecosystem services forests provide to humankind. In view of this global problem, the Parties of the Convention on Biological Diversity (CBD) adopted a specific Programme of Work on Forest Biological Diversity that sets as one of its targets the conservation of “at least 10% of the world’s forest types” by 2010. The present study illustrates the major issues related to this target, such as the geographic distribution of the world’s remaining forest areas, difficulties related to the world’s forest types and WWF ecoregions, and an up-to-date global gap analysis for forest protected areas. The objectives of this study were developed within a larger project carried out by the Institute of Forest and Environmental Policy (IFP), University of Freiburg, called “Conservation of forest biodiversity under the CBD: Options for a global forest protected area network”. Its overall aim is to provide scientific analyses and policy advice with regard to the feasibility, financing and targets for the establishment of a global network of forest protected areas. The project is supported by the German Federal Agency for Nature Conservation (BfN) with funds from the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU). The present study is a result of the joint efforts of a consortium comprising the United Nations Environmental Programme World Conservation Monitoring Centre (UNEP-WCMC), World Wide Fund for Nature (WWF) Network and World Resources Institute (WRI) in cooperation with IFP. Staff from the WWF network, in collaboration with those from UNEP- WCMC and WRI, prepared the assessment of the feasibility of WWF ecoregions for use as a global guidance for forest types; UNEP-WCMC took the lead on conducting the global gap analyses for forest protected areas. On behalf of the consortium, IFP is thanked for funding the majority of this work through support provided by BfN / BMU; WWF International also provided additional funding. All partners of the consortium are thanked for contributing their staff time, as well as data sets and analyses for use in this study. Conservation International provided permission to use their GIS layers for High Biodiversity Wilderness Areas and Biodiversity Hotspots. Last but not least, many thanks go to Neil Burgess, Alison Campbell, Lauren Coad and Christine Schmitt for their tireless efforts to finalize this study on behalf of the broader group. Digitized by the Internet Archive in 2010 with funding from UNEP-WCMC, Cambridge http://www.archive.org/details/globalecological08schm Contents FXO YES TUT NITE AY, ccoocostaceoconecesoece cercncisocuazhodacadcédadacoaeeeeeeq5050qz36boc.90ed 1n6dcodad0abacan6onaceRTaCHoOASCONEAC i [UTES Gt 7 el Ke TRIM SS so coccoeesoseeace6caacecesoc0 s8c0sdco0c6cee eecbo¢F5005000060050030023060003003550300 063 GOnFED HEAGOHBSCESORODOOE iv Thee HIRI LTC Ton) eassstecotecercescdecooctondacsoocedGhosseed condenaseeesdascsanqcqans hobs sossoasbouedodedbecdocctranSoddencatadaas 1 ile [PYRO OLS Tirn) SEV TOIT Yi) pcace ep deococoeconosdceceueceooe8esecocosansacpootesbendaceadue en anbeasccooSéncacorastaheodsn 1 1.2 @bjectivestamGloutlime seco seeecesceseeseee eee ce seen eereete rece nee tereee eee sacs seeosetennscases 2 ® (NE tinterelo) tole [1233 scecocccsoscsocascodoccooses ck fanenconcoceadacn68-anovacs0 Ja senbuooadockodesoucdeccardanancegeonccacassenene06 3 s| Updated Global Forest Map ............:::::c.cceccsscsssecescecsesseseresscenceeccesceereerecossceseerssees 3 2.2 WWE ecoregion analysis ...-....0.cecccc-ctsctecsscssocessccsocncoveescnenencecuectceeserceurensartesncse== 3 2.3 Forest protected area gap AMalySis ............::cccccscssecseeteeneeeeetseeteteeetecseeteeeseeessaes 4 Pxey a Eidol (stent =1o] 1. ¥2) (CFE U2 baocacoovoceoqdeoacoahocanéccanaécedebsaossaoa5oee6oocobesudsasaccesdocosodechcanasosaxidc500 4 2.3.2 Forest Protection GAP ANALYSES ...........:ccccsccescccesecseenteceeeetteeeeeenseeeeecseeesenaesens 5 6) FRIEKSUTTS late DIEU ATOM cescecsenacecesdrocte ccadeandnanos2chasaeascaosooetieessosodsodacscoodcecsSBcoosanaaaaocoaoccioaRtt 6 Sal Updated Global Forest Map..........-....:::::cssseccccssesssseessecsneecereneesenrecencencsreocersssessesecs 6 3.2 WWE ecoregion analySis ............c:cccccesccsseeseesecseeneececeeeeereraeeneeseceneeeseeseesesseressaseas 8 3.2.1 Forest cover across ‘forest’ and ‘Non forest’ CCOFEGIONS.............cscceeeceeeeseenees 8 3.2.2 How well do ecoregioiis capture fOreSt tyPOS? ........c..ccccceceneneeneenretnentetteeees 8 3.3 Forest protection Gap ANALYSES ............eccesccseeesecseceneeeeeeneeeecteetseetseeseetseesesenseeaeees 10 3.3.1 Protection of global fOreSt tyVPOS ............:cccccccsccscesecneeseeeeeenecteteeeneeteeteeeensenssens 10 3.3.2 Protection of forest types by DIOgeOGraPHIC FOAIM.........cccccccccereceeseeeeetetteettaes 12 3.3.3 Protection of forest DY WWF @COrEQIONSG .............1cccccccecceeteceesteetenteettneettneeetenses 15 3.3.4 Protection of global PriOrity AFCAS.........cccccccccccceeceseeteeneetetteetetteeteecseeeseeneseeens 20 4 Summary and Conclusions. .............ccccccccceccseecceceeeecneeseeeeeeecaetecsesssescseassesessecseseeeeneceees 22 4A Methodology for assessing progress towards forest protection targets ................. 22 4.2 Current level of protected area coverage for the world’s forests..........--:::sseeeeee 24 4.3 Utility Of the 10% target... ecceesecseseeeneecceseeeeecneeseeetaeserseesessesseesesssceseasnesseneaees 25 4.4 (Lian 1 OlAES Ge | CEIYEYE 11S) ecerecocdsonnsuacasnotoodno-pooccagddeannecendcode saescann00Dpans0ce=cececcoou0BDoGE. 26 I) [RETO OE LITO o-corsennonncoan coc ooccnacdeadssceders0s70% of their natural habitat, whereas wilderness areas have >70% of their remaining habitat, and include the most intact forest areas such as the Amazon and Congo Basin rainforests. The results for each are presented in terms of percentage forest protection under IUCN PA management categories I-IV and I-VI (Section 2.3.1). Areas that have less than 10% protected area coverage are highlighted across each scale of analysis in order to identify gaps in the protected area network. 3 Results and Discussion 3 Results and Discussion 3.1 Updated Global Forest Map The updated GFM is a 500m resolution raster GIS dataset using the latest available satellite imagery to provide an accurate global estimate of forest cover relative to current remote sensing capabilities. By excluding areas of non-natural tree cover, it identifies predominantly natural forest cover. A separation was made between forest cover of 10 to 30% and greater than 30% (Figure 1). This differentiation improves the analysis in that in biomes such as tropical moist forest, where natural forests have a closed canopy, a tree cover of less than 30% indicates severely degraded forests or mosaic of forest and agriculture - in either case a forest significantly altered from a natural state. In addition, at low levels of canopy cover there may be some confusion of shrubs or agroforestry with forests. Meanwhile in drier biomes, the 10% to 30% forest cover class may indicate natural forests. Keeping these two classes of forest cover separate allows flexibility to interpret results at the biome or ecoregion level, and avoids the over-estimation of forest cover where natural forests have a canopy cover much greater than 10%. Figure 1: Global forest area as defined by MODISO5 VCF at >10% and > 30% forest cover; the >10% cover definition of forest was used in this report. The analyses in this study are based on the extent of forest area with > 10% canopy cover, the threshold level used by the UN Food and Agriculture Organization (FAO) for their Forest Resource Assessment (FRA) products, in order to ensure that the results are representative of the most conservative and widely used global forest definition. The estimate of global forest cover provided by the updated GFM is 38.9 million km? (28.8% of the land area 3 Results and Discussion excluding lakes, rocks and ice). This is larger than the latest FAO estimate from 2005 of 34.4 million km?, which is derived mainly from national scale inventories (FAO 2005). The updated GFM still contains large tracts of unresolved forest type because the forest areas newly identified by MODISO5 VCF could not be integrated with the existing GFM forest types in the timeframe available and were grouped according to the GLC 2000 forest classes. The updated GFM thus distinguishes 28 forest types, in which the GFM 2000 forest types and the GLC 2000 forest classes are treated separately (Table 3). Whilst this was necessary for consistency within the timeframe of the study, the forest types of these two different systems should ideally be harmonized to generate a single set of statistics for forest area of different forest types to be applied globally. The development of such a globally consistent forest map should be considered a priority and this finding is reiterated as one of the major conclusions of this study. Table 3: Forest types as identified by the updated Global Forest Map (GFM). Unresolved forest types indicate areas of forest that could not be identified to be consistent with the existing GFM forest types in the timeframe available. They are grouped according to Global Land Cover (GLC) classes. Forest Type Forest area ( ‘000 km?) Forest types identified in the GFM Temperate evergreen needleleaf forest 8,168 Temperate deciduous needleleaf forest 3,234 Temperate deciduous broadleaf forest 2,689 Temperate sparse trees/parkland 1,939 Tropical deciduous/semi-deciduous broadleaf forest 1,729 Temperate mixed broadleaf / needleleaf forest 1,435 Tropical sparse trees/parkland 1,007 Tropical semi-evergreen moist broadleaf forest 843 Tropical freshwater swamp forest 536 Tropical upper montane forest 476 Tropical lower montane forest 448 Temperate sclerophyllous dry forest 393 Tropical sclerophyllous dry forest 241 Temperate broadleaf evergreen forest 179 Tropical mangrove 137 Tropical needleleaf forest 95 Temperate freshwater swamp forest 88 Tropical thorn forest 10 GLC 2000 Unresolved forest classes Unresolved tree cover, broadleaved, evergreen 2,942 Unresolved tree cover, broadleaved, deciduous, closed 2,606 Unresolved tree cover, broadleaved, deciduous, open 1,791 Unresolved mosaic: tree cover / other natural vegetation 846 Unresolved tree cover, burnt 31 Unresolved forest unknown type 10 Total forest cover 38,998 3 Results and Discussion 3.2 WWF ecoregion analysis 3.2.1 Forest cover across ‘forest’ and ‘non forest’ ecoregions An analysis of forest cover across all ecoregions showed that forest is distributed across various ecoregions, including those which are not officially defined as ‘forest’ (Table 1). 732 ecoregions have some degree of forest cover according to the updated GFM. Of the 525 ‘forest’ ecoregions, only 489 (or 93%) contain forest (Table 4). This means that 36 of the ‘forest’ ecoregions no longer contain forest. The discrepancies in forest cover are partly related to ecoregions found on small tropical islands, where forest cover could not be measured due to the lack of satellite data. In addition, it is likely that some other ecoregions have experienced extreme deforestation over the past 500 years (for instance on islands such as the Mascarenes, Madagascar, and Comoros), and do therefore no longer contain forest according to the updated GFM. In contrast, 81% of ‘non-forest’ ecoregions did in fact contain areas of forest cover. Many of these ecoregions are within montane grassland-shrubland mosaics and the savanna woodland biome, which covers huge areas of the drier regions, predominantly in the tropics. Although the average percentage forest cover in ‘non forest’ ecoregions was low (11% cover compared to 42% in ‘forest’ ecoregions), this is still a significant amount of forest. The forest cover can be due to forests along riparian areas or at the limits of ecoregions; or a result of the broad WWF biome classification. Table 4: Assessment of the forest cover of the ecoregions falling within ‘forest’ and ‘non-forest’ biomes (see Table 1). Number of ecoregions Number of ecoregions % ecoregions with with forest cover forest cover ‘Forest’ ecoregions 525 489 93.1 ‘Non forest’ ecoregions 81.3 All ecoregions 825 732 88.8 Our results clearly indicate that the set of ecoregions within ‘forest’ biomes does not capture all the world’s forested areas. It is therefore apparent that the further separation of ecoregions into ‘forest’ and ‘non forest’ is not an accurate indicator of forest coverage. Consequently, all further analyses focus on the 732 ecoregions with forest cover, irrespective of whether they are officially defined as ‘forest’ ecoregions by WWF. Furthermore, it can be concluded that we need to look at actual forest cover, rather than total area of the ecoregion, for analyses of forest protection. 3.2.2 How well do ecoregions capture forest types? The overlay of the updated GFM with the WWF terrestrial ecoregions highlighted that most ecoregions contain four to six forest types (Figure 2). Although this analysis was hampered by the fact that some of these forest types refer to unresolved forest cover (Table 3), it can be concluded that ecoregions do not adequately capture the variety of global forest types identified by the updated GFM. This holds true especially for tropical ecoregions that may contain a large number of forest types, up to fourteen in some cases. Although tropical ecoregions have a finer resolution than than the temperate ones, they are still too coarse to represent the large number of forest types that result from the complex vegetation patterns in tropical forests. “sadA} jSsa10) uaamjeq ysinBursip Ajayenbape you op yoiym suoibe109a ayeoipul sease sayieg ‘Ajjeqo|5 uoiBas009 4AM YORd UUM (| BIGQe) Bas) dey Jsas0y jeqo}y payepdn ayy Aq paynuap! se sad} }sa10} yo saquiny :z anbi4 wy 000'y 000'2 0 000'C00'0SL} 31B9S Aplemjjoyy ‘UaNIa!O14 LO0Z ‘UNeR UO 9417 Jo dey Man Y PHOM au) Jo SUdIBaI0D= 4A 1002 ‘QWN “JeAgD e911 jUBDGY ‘SPj9i-J SNONUNUOD UONE\eBeA SIGOW SO0Z - 0002 ‘OSWOM-d3NN ‘dey ysas04 Jeqo}D - Eo JONOD 1SAJO4 JUBA . BSeQUNOS ” e-1 | yunos Ajayien sadfy ysas04 aes uolssnosiq pue s}insey ¢ 3 Results and Discussion In addition to this, certain forest types of exceptional biological and hydrological importance, such as montane forests and riparian forest are not identified in the ecoregion framework. Riparian vegetation was explicitly not incorporated into the WWF framework due to its small spatial extent; despite usually being comprised of relatively unique vegetation compared to the surrounding ecoregional vegetation. It is clear, therefore, that ecoregions cannot be taken to be accurate measures of forest cover in terms of representation of forest types. This finding was to be expected, as the fine scale coverage of the various forest types are hard to separate into the relatively coarse framework of ecoregions. In fact, ecoregions were not intended for use in vegetation Classification, but rather to identify similar biogeographic assemblages. This is not only a drawback but also an advantage because the ecoregions can distinguish differences in species assemblages not recognized by the GFM forest types. Africa, for example, has been divided into four separate mangrove ecoregions. As a consequence, global forest cover should be classified by using a combination of forest types and ecoregions. Further discussion of this topic can be found in Section 4.1. 3.3. Forest protection gap analyses This section aims at assessing the global progress towards achieving the CBD target of conserving “at least 10% of the world’s forest types’ until 2010. There is not yet an ideal way of assessing this target, because the updated GFM contains a large number of unresolved forest types, whilst the ecoregions do not accurately represent the variety of forest types that can be present within a biogeographic region. We therefore decided to assess the protected area coverage for a variety of different spatial units, i.e., for forest types, realms, forest types by realms, ecoregions and global conservation priorities. Although neither methodology is perfectly suited for forest gap analyses, they are the best methods available and together can provide valuable insights into the gaps in global forest conservation. The purpose of these analyses is to identify forest areas in which the 10% protected area coverage target has been met, and those for which there are gaps. However, it should be noted that there is some debate as to the relevance of the 10% target (Section 4.3), and the results should therefore be viewed with this in mind. Surpassing the 10% targets does not necessarily mean that there are no gaps in the protected area network, and although the 10% target provides a representative and global measure of progress, the results presented here show that the level of protection differs depending on which scales / units are used for the analyses. It should also be noted that protected areas that have not been assigned an IUCN category have not been reported in this analysis. This provides a level of consistency in that only the designated protected areas that have been assigned a category are included, but does mean that the figures presented here may underestimate the level of forest protection. 3.3.1. Protection of global forest types Globally, forests are protected at levels below tne 10% target when strictly protected areas (categories I-IV) are considered, with total protection of 7.7% of forest area (Table 5). When all management categories are considered (including areas designated for sustainable use), this value nearly doubles to 13.5%. It is interesting to note that approximately half of the world’s forest protected areas fall under categories V-VI, where sustainable use of the forest is permitted, and invites discussion as to what is meant by ‘protection’ when considering biodiversity targets. 10 3 Results and Discussion Table 5: The area, and percentage coverage of global forest types (described in the updated Global Forest Map), at IUCN categories I-IV and I-VI. Forest types failing to meet the 10% target for protection (at IUCN I- IV) are highlighted in grey. Forest Type Total Forest Area % Protected % Protected (‘000 km?) (IUCN I-IV) (IUCN I-VI) GEM Forest Types: Temperate freshwater swamp forest 88 Temperate mixed broadleaf / needleleaf forest 1,435 Tropical mixed needleleaf / broadleaf forest 9 Temperate deciduous needleleaf forest 3,234 Tropical needleleaf forest 95 Temperate deciduous broadleaf forest 2,689 Temperate sparse trees/parkland 1,939 Temperate evergreen needleleaf forest 8,168 Tropical freshwater swamp forest 536 Tropical sparse trees/parkland 1,007 Tropical deciduous/semi-deciduous broadleaf forest 1,729 Tropical thorn forest 10 Tropical lowland evergreen broadleaf rain forest 6,489 Tropical lower montane forest 448 Tropical mangrove 137 Temperate sclerophyllous dry forest 393 Tropical sclerophyllous dry forest 241 Tropical semi-evergreen moist broadleaf forest 843 Tropical upper montane forest 476 Temperate broadleaf evergreen forest 179 GLC 2000 Forest Types: Unresolved tree cover, regularly flooded, saline water 1 Unresolved tree cover, broadleaved, deciduous, 2,606 closed Unresolved tree cover, broadleaved, deciduous, open 1,791 Unresolved tree cover, burnt 31 Unresolved tree cover, mixed leaf type 625 Unresolved tree cover, broadleaved, evergreen 2,942 Unresolved mosaic: tree cover / other natural 846 vegetation Unresolved forest unknown type 10 Total forest cover 38,998 Coarse global averages are useful for assessing international achievement of targets, however they often mask discrepancies in terms of representativeness of biodiversity. As forest types differ in terms of their species richness and provision of ecosystem services, the protection coverage of forests needs to be assessed by forest type rather than simple global forest cover. Whilst this study makes no attempt to assign ‘value’ to these different forest types, it is clearly necessary to distinguish between different types of forest if ‘representativeness’ of forest protection is to be established. There is a wide range in the global protection of various forest types by strictly protected areas, with coverage of Temperate freshwater swamp forest protected at only 3.2%, whereas Temperate broadleaf evergreen forest is protected at 28% (Table 5). Nineteen of the 28 forest types (close to 70%) identified by the updated GFM are under the 10% target when measured according to strict protection, and 10 (35%) are still below the target when all protected area categories are considered. This means that a significant proportion of global 11 3 Results and Discussion forest types have not gained an acceptable level of protection according to the CBD targets, including large tracts of both tropical and temperate forest. There are clearly some gaps to be addressed in the protection of forest types. Only two temperate forest types meet the 10% target, whilst needleleaf forests and thorn forests in the tropics are also strongly underrepresented. The range of protection again invites discussion as to the relevance of the 10% target in terms of the species diversity that will be protected at 10%. It could be considered, for example, that species within some temperate forest types could be well conserved even at levels below 10%, whereas the protection of some tropical forest types at levels over the 10% target would not be enough to achieve the same purpose. Regardless, these results highlight areas in which protection needs to be strengthened in order to achieve the 10% targets of forest protection to ensure full representation. They also highlight the necessity for the production of a global forest map, in which all global forest cover is assigned a resolved forest type, if forest type gap analyses are to fulfil their potential in assessing progress towards forest targets. 3.3.2 Protection of forest types by biogeographic realm In addition to forest type, there is a need to assess levels of forest protection across a number of regions, as the same forest type can differ in terms of species composition and biophysical conditions according to its geographical location. The WWF realms split the globe into 8 distinct biogeographic realms (7 containing forest), which can be used to investigate the protection of forest cover in different regions of the world (Figure 4 provides geographic representation of realms). Table 6 describes the protection of forest cover within each of the realms. Five of the realms fail to meet the 10% targets under strict protection; three when all categories are considered. Table 6: Percentage protection of forest cover within each WWF realm (Antarctic not included due to no forest coverage within realm). Realms failing to meet the 10% targets for protection (at IUCN L-IV) are highlighted in grey. Realm Total Forest % Protected % Protected Area (‘000 km?) (LUCN I-IV) (IUCN I-VI) Palearctic 11,792 5.5 8.8 Bulk of Eurasia and North Africa Afrotropics 6.4 (Sub-Saharan Africa) Nearctic 6.6 (most of North America) Oceania 75 (Polynesia, Fiji and Micronesia) Indo-Malay (South Asian subcontinent and Southeast Asia) Neotropics South America and the Caribbean Australasia Total forest cover 38,992 7.7 13.5 The lowest level of forest protection is found in the Palearctic (including Eurasia and North Africa), with a protected area forest coverage of 5.5% for categories I-IV, significantly lower than the 10% target forest cover. This is particularly relevant considering that this realm has the largest area of forest coverage, and is still below the 10% target when all categories are considered. Indeed, only two realms (Australasia, Neotropics) achieve the 10% target for both strictly protected and all protected area categories (although the Indo-Malay realm could 12 3 Results and Discussion be considered to have achieved the target. at 9.9%). Australasia has the highest percentage of strictly protected forest at 13.4%. It is interesting to note that over half of the protection in the Nearctic, mostly comprising North America and some parts of Central America, is protected under categories V-VI. These realms are extremely large geographic areas, and although this is interesting for identifying regions in which protection efforts should be focused, such as in the Afrotropics and the Palearctic, it gives no indication of the level of protection of forest types within each geographic region. Although the Neotropics and Australasia have achieved the 10% target for forest coverage, for example, there are still a number of forest types that are under represented (Table 7). Indeed, 65% of forest types occurring in the Neotropics do not meet the 10% target (Annex). Table 7: Protection of forest types by realm at IUCN categories I-IV (unresolved forest types included). Realm Number of forest Percentage below 1% Percentage below types protection 10% protection Nearctic 33.3 90.5 Oceania Neotropics 11.5 65.4 Indo-Mala 8.7 47.8 Australasia The majority of the geographic realms have a number of forest types protected at levels even below 1%. Even Australasia, which has the best protection coverage of its forest types by realms, fails to adequately protect the three forest types most commonly underrepresented across all realms; Tropical deciduous/semi-deciduous broadleaf forest, Temperate deciduous broadleaf forest, and Tropical lowland evergreen broadleaf rain forest (Table 8). Conversely, there are a number of forest types within the less well protected realms that are protected at levels above the 10% target, such as the Tropical semi-evergreen moist broadleaf forest in the Indo-Malay realm, which is protected at 30% (Annex). This information is important for informing the designation of new protected areas of forest, as they should be focused on representative coverage of forest types within realms, rather than achieving the blanket 10% targets for the realm as a whole. Identifying areas in which new protected areas should be established based on forest type would be a large step forward towards creating a representative network of forest protection. As mentioned previously, a larger number of forest types is yet unresolved and this analysis only represents an approximate way of assessing progress towards the 10% target. In addition, as with the global forest protection figures, however, these realms can still be considered coarse scale in biogeographic terms (Dudley and Parish 2006) and further detail can be gained by comparing these results with those from determining forest protection levels at the ecoregion level. 13 3 Results and Discussion Table 8. Forest types below the 10% protection target within each realm (for unresolved forest types and forest types above the 10% target see Annex). Forest types highlighted in grey are below the target (at IUCN I-IV) across a number of realms. Realm Forest Type Total Area % Protected % Protected 2 (‘000 ie) (IUCN I-IV) (IUCN I-VI) Australasia | Tropical deciduous/semi-deciduous broadleaf forest 17 Si Afrotropics | Tropical thorn forest 1 0 0 Tropical lower montane forest 29 0.7 1.2 Tropical freshwater swamp forest 191 3 Sil Tropical mangrove 36 3.8 4.2 Temperate deciduous broadleaf forest 34 6 6.1 Temperate sparse trees/parkland 35 6 6.1 Tropical semi-evergreen moist broadleaf forest 18 6.6 Wel Tropical deciduous/semi-deciduous broadleaf forest 1,299 6.8 10 Tropical lowland evergreen broadleaf rain forest 1,526 8 9.6 Tropical sparse trees/parkland 476 10 12.7 Indo-Malay | Tropical sclerophyllous dry forest 0) 0 34.5 Temperate sparse trees/parkland Temperate deciduous broadleaf forest Temperate evergreen needleleaf forest Temperate sclerophyllous dry forest Tropical freshwater swamp forest Nearctic Tropical lowland evergreen broadleaf rain forest Tropical sparse trees/parklana Tropical mixed needleleaf/broadleaf forest Tropical deciduous/semi-deciduous broadleaf forest Tropical needleleaf forest | Tropical upper montane forest Tropical lower montane forest : Temperate freshwater swamp forest 88 3.2 Temperate deciduous broadleaf forest 957 4.4 Temperate deciduous needleleaf forest 10 4.4 Temperate mixed broadleaf/needleleaf forest 1,113 5 Temperate sparse trees/parkland 1,036 5.5 Temperate sclerophyllous dry forest 91 6.8 Temperate evergreen needleleaf forest 3,431 7.8 Tropical mangrove 0 8.6 Worse Temperate sclerophyllous dry forest 35 0 Temperate mixed broadleaf/needleleaf forest 6 0.1 Temperate sparse trees/parkland Temperate deciduous broadleaf forest Temperate freshwater swamp forest Temperate evergreen needleleaf forest : Tropical needleleaf forest 72 45 8.9| Tropical mixed needleleaf/broadleaf forest 9 45 6.9 Tropical sparse trees/parkland 409 5.3 8.8 | Tropical freshwater swamp forest Tropical deciduous/semi-deciduous broadleaf forest Oceania Tropical lowland evergreen broadleaf rain forest Palearctic | Temperate mixed broadleaf/needleleaf forest Temperate sclerophyllous dry forest Temperate deciduous needleleaf forest 3,223 Temperate sparse trees/parkland 652 Temperate evergreen needleleaf forest 4,356 Temperate deciduous broadleaf forest 1,522 14 3 Results and Discussion 3.3.3. Protection of forest by WWF ecoregions Although it has already been established that forest types are not adequately captured within ecoregions, ecoregions still provide important forest biodiversity information related to biogeographic units. As forest types are not accurately identified on a global forest map (with many forest types still unresolved), it is still relevant to analyse protection of forest on an ecoregion scale. This is particularly true given their aforementioned use in a number of biodiversity analyses and the fact that they are widely accepted biogeographic units with which to undertake this kind of analysis. Drawing from our previous results, the protected area coverage was calculated for the forest area contained within each ecoregion. The distinction in this report between protection of the forest area within each ecoregion, rather than the percentage protection of the forest ecoregion as a whole, is an important step in utilising ecoregions in forest analyses. 732 ecoregions contain some level of forest cover and have been used in the following analyses. The average level of protection of forest cover per ecoregion is 10% under strict protection categories and 18% when all categories are considered. On an aggregated scale, the protection of ecoregions reaches the 10% target. However, aggregated statistics rarely show the true picture and can mask significant variation in the range of forest protection within ecoregions. The majority of ecoregions actually fall below 10% protection (Figure 3), and it is likely that a small number of ecoregions with high levels of protection (e.g., the Fiordland temperate forests with 98% protection), raise this average to a level that is not truly representative. Similarly, some ecoregions with either high or low forest cover protection may actually contain only small forest areas, an issue that has not been controlled for in this analysis. @ Category LIV Category FVI Number of ecoregions under10 10-25 25-50 50-75 75-100 Percentage area protected Figure 3: The number of forested ecoregions at different levels of protection, at IUCN categories I-IV and I-VI. It is clear that there is a wide range in percentage protection of forest within ecoregions; only 35% of ecoregions reach the 10% target under strict protection, whereas the Fiordland temperate forests are protected at 98%. This can be put in context through analysis of the regional distribution of forest protection within ecoregions (Figure 4), from which it is clear 15 3 Results and Discussion that high levels of ecoregion forest protection (relative to the 10% targets) are concentrated in a few regions, such as Australia, the Amazon, the Andes, and South East Asia. It is again noteworthy that there is very little protection for northern latitude forest, outside of Alaska. Ecoregions can also be split according to geographic realm, and it can be seen that, as with forest types, a significant proportion of ecoregions within these realms are protected at levels of less than 1% (Table 9). Indeed, there are 110 ecoregions in which the forest cover has no protection at all. On a general scale, the Afrotropics and Palearctic again appear to require increased attention with large numbers of ecoregions protected at levels below the 10% targets; an issue also identified by the global ecoregion protection map (Figure 4). A partial reason for this is that in Africa and some parts of Asia there are thousands of forest reserves managed by Forestry Departments that are of uncertain conservation status and have not been assigned IUCN protected area categories. These reserves were not considered for this analysis; however, their conservation status merits further investigation. Table 9: Protection of ecoregions by realm at IUCN categories I-IV. Ecoregions with forest Percentage below 1% Percentage below 10% cover protection protection Palearctic 172 38 82 Afrotropics 96 28.1 75 Nearctic 114 24.6 70.2 Neotropics 168 25 61.9 Indo-Malay 101 11 51.5 Oceania 4 25 50 Australasia 77 19.4 47 A number of ecoregions with low levels of forest protection and large forest area have been selected and presented in Table 10 for comment. This highlights the large areas of biogeographically distinct forest areas within ecoregions that have little or no protection. Although Australasia is in general the best protected realm, significant areas of rainforest have no protection at all. Subtropical ecoregions of the Indo-Malay; as well as the Borneo lowland rainforests are also protected at levels below 5%, and the lack of protection for moist forests in the Neotropics is also worthy of comment, along with that of the Cerrado and Caatinga ecoregions. In the Afrotropics, already noted to be in need of increased protected area coverage, the low levels of protection for the large tracts of forest within the various Congo ecoregions is striking, as is that of the Miombo woodlands. The fact that the Congo has been identified as a Cl high biodiversity wilderness area makes these gaps appear even more relevant. A large number of taiga ecoregicns within the Nearctic and Palearctic are also protected at extremely low levels, including forest stretching over areas up to 3 million km’. Although these areas are not classified as ‘high biodiversity’ they contain large amounts of carbon. It should again be noted, therefore, that there are many different values for forest ecosystems not accounted for in this report. 16 *J9A0D }S91O} OU BJESIPU! SBae ayIUAA “SUeas D1ydesBHoaboIq aJeOIPU! Saul] YOeIg “S}SOJ0} |ea10g UBYWON pue eoyy jesjUaD ul UISeg OHU0D ay} apNjouU! UO!}9a}01d MoO] JO SBale aIqeJON “eySe|\y pue elsy 3S ‘uozewy ay) ‘elejsny jo syed ul uaes aq ued UONDa}01d Jo sjang] SaYyBiy ay “SUOIBa109a 4AM UIYYM Base JSa10} pa}de}01d jo abeyuadiad |y} Jo UONNqUISIG :p aunbi4 WSS - WOlL 396121 %0} au) Jepun %001 - %0S FRR %05 - %Sz | eisejesjsny a dIdosjoly e1uess9 _ =<" _s>youraejed uolssnosiq pue s}insay ¢ 17 3 Results and Discussion The complete list of ecoregions that either meet or fall below the 10% targets is too detailed to present here, but is available upon request. It should again be noted, however, that those ecoregions currently protected above 10% should not be considered to be adequately protected without further analysis of the utility of the 10% target for that region, and an analysis of how well effectively conserved the forest types are within that particular ecoregion; a task beyond the scope of this report. Indeed, comparison of the ecoregion protection map (Figure 4) with that of forest types contained within ecoregions (Figure 2) highlights the deficiencies in this analysis. It is clear that many of the ecoregions which appear to be exceeding the requirements for forest protection set by the CBD are also ecoregions which contain a large number of forest types; the levels of protection of which cannot be captured through this ecoregion assessment. It is clear that there are a large number of ecoregions with low percentage protection which require further analysis of the forest types in need of protection in order to inform protected area designation. Conversely, the ecoregions of adequate protection relative to CBD targets also need proper examination to ensure that these levels of protection are representative of all forest types. This analysis could be facilitated by the improvement of the GFM, which would allow further regional assessment of the forest types requiring protection within separate ecoregions; a task too complex to carry out on a global scale. 18 3 Results and Discussion Table 10: Selected ecoregions of low protection and with large forest area, separated by realm. % Protected (IUCN I-VI) % Protected (IUCN I-IV) Ecoregion Forest Area (‘000 km?) Australasia Solomon Islands rain forests Southeastern Papuan rain forests Halmahera rain forests Vogelkop-Aru lowland rain forests Southern New Guinea freshwater swamp forests Southern Miombo woodlands Guinean forest-savanna mosaic Central Zambezian Miombo woodlands Indo-Malay Jian Nan subtropical evergreen forests 369 0 6.2 Northern Indochina subtropical forests 162 2.2 9.9 Borneo lowland rain forests 257 4.4 6.1 Nearctic Eastern Canadian Shield taiga 422 Southeastern mixed forests 249 Appalachian mixed mesophytic forests 165 Sierra Madre Occidental pine-oak forests 148 Piney Woods forests 114 Northwest Territories taiga 191 Northern Canadian Shield taiga 234 : Central U.S. hardwood forests 139 1.8 5 Southeastern conifer forests 130 2.7 8.7 New England-Acadian forests 211 Appalachian-Blue Ridge forests 125 Midwestern Canadian Shield forests 406 Eastern Canadian forests 367 Neotropics Iquitos varzeB 103 0.4 19.5 JuruB-Purus moist forests 240 0.6 15.8 Humid Chaco 120 Ucayali moist forests 102 Purus-Madeira moist forests 167 Caatinga 169 Cerrado 334 Mato Grosso seasonal forests 263 Purus varzeB 166 Southwest Amazon moist forests 701 Palearctic Da Hinggan-Dzhagdy Mountains conifer forests 204 Manchurian mixed forests 313 Western European broadleaf forests 176 Trans-Baikal Bald Mountain tundra 111 East Siberian taiga 3,019 Scandinavian and Russian taiga 1,679 West Siberian taiga 1,091 3 Results and Discussion 3.3.4 Protection of global priority areas Analyses thus far have focused on the classification of regions according to factors such as forest types and biogeographic units, but no assumptions of the ‘biodiversity value’ of the area have been made. However, global priority areas for conservation do exist, and the results presented in this section illustrate the degree to which forest is protected within Cl biodiversity hotspots and Cl high biodiversity wilderness areas. Both these priority systems follow the ecoregion classification evaluated in this study, which allows direct comparison with the previous ecoregional results in this study. Conservation International Biodiversity Hotspots Conservation International have defined a number of biodiversity hotspots, i.e., areas where at least 0.5% of the total global plant species are strictly endemic, but less than 30% of the natural habitat remains (Mittermeier et a/. 2004). Each of these hotspots contains at least some forest cover, with the total protection across all hotspots covering only 9.6% of the forest area for strictly protected areas, but surpassing the 10% target at 15.5% when all categories are considered (Table 11). Table 11: Protection of forest area within Conservation International’s biodiversity hotspots. Hotspots failing to meet the 10% targets for protection (at IUCN I-IV) are highlighted in grey. Cl Biodiversity Hotspot Total Forest Area % Protected % Protected (‘000 km’) (IUCN I-IV) (IUCN I-VI) East Melanesian Islands 7 0.7 0 Mountains of Southwest China 12 0 13.8 Guinean Forests of West Africa Mesoamerica Polynesia-Micronesia Wallacea Madagascar and the Indian Ocean Islands Atlantic Forest Sundaland Tumbes-Choco-Magdalena Eastern Afromontane Himalaya Cape Floristic Region California Floristic Province Caucasus Philippines Indo-Burma Horn of Africa Caribbean Islands Chilean Winter Rainfall and Valdivian Forests Mountains of Central Asia Western Ghats and Sri Lanka Tropical Andes Southwest Australia New Zealand Total forest area 635 9.6 15.5 20 3 Results and Discussion 20 out of 34 hotspots fail to meet the 10% target under strict protection, including the hotspot with the largest area of forest cover, the Sundaland. The entire East Melanesian Islands hotspot has no protection at all, and it should also be noted that the majority of Africa’s hotspots do not reach the 10% target for protection. In addition, the hotspot forest areas of the Cerrado, already highlighted as an ecoregion with low protection, are protected at just over 5%, and the Cerrado would appear to be a priority for protection in Latin America. Hotspots of Australasia again appear comparatively well protected, with the Tropical Andes also achieving levels of protection over 18%. Conservation International High Biodiversity Wilderness Areas Conservation International have also identified high biodiversity wilderness areas, i.e., areas of over 750,000 km? that are rich in biodiversity, with levels of endemism comparable to the most diverse biodiversity hotspots (Mittermeier et a/. 2003). These differ from hotspots in that over 70% over the natural land cover is intact. Combining these high biodiversity areas with hotspots accounts for 61.5% of all vascular plants and 43.2% of non-fish vertebrates. Five of the Cl high biodiversity wilderness areas contain forest cover, and three fail to meet the 10% targets; the North American Deserts, the Congo Forests, and New Guinea (Table 12). It is interesting to note that although the Miombo-Mopane wilderness area meets the 10% target, none of the individual Miombo ecoregions do (Table 11). This discrepancy is probably related to the fact that the wilderness areas are aggregates of ecoregions and this further underlines that averages at the broader level might blur the reality on the ground. Table 12: Protection of forest area within Conservation International’s high biodiversity wilderness areas. Areas failing to meet the 10% targets for protection (at IUCN I-IV) are highlighted in grey. Cl High Biodiversity Wilderness Area Total Forest % Protected % Protected Area (‘000 km’) (IUCN I-IV) (IUCN I-VI) North American Deserts 41 4.2 18.7 Amazonia 5617 Vile! 25 Miombo-Mopane Woodlands and Savannas The low levels of protection within the Congo forest have been highlighted throughout this report, and it is clear that forest protection should be made a priority in this area, as it is a significant gap in the global protected area network. As these areas have been labelled ‘high biodiversity’ the Taiga forests have not been included in this analysis. However the results shown previously for forest type, realm, and ecoregion protection have also highlighted that these wilderness areas do not even come close to achieving the protection targets. Although the Amazonia is protected at levels above the 10% target, the 11% protection could still be considered low given the importance of the area for biodiversity. This again raises the question as to whether the 10% target should be applied to all situations. Indeed, none of the wilderness areas are protected at levels above 15%, and only one of the hotspots above 20%. This does not seem adequate in relation to the high biodiversity values of these areas. In addition, over half of Amazonia is protected under categories V-VI; a pattern also seen for some of the hotspots. Whether or not these high biodiversity areas should have such a skewed proportion of ‘sustainable use’ protected areas is another topic for discussion. In summary, a large proportion of globally accepted high biodiversity areas are protected at levels below the 10% target under strict protection; a situation that could be considered even bleaker with reference to the fact that 10% protection is not likely to be enough in these areas that harbour such a significant proportion of global biodiversity. This is particularly the case for hotspots, which have already lost a significant proportion of their forested area. 21 4 Summary and Conclusions 4 Summary and Conclusions 4.1 Methodology for assessing progress towards forest protection targets Two potential ways to assess progress towards the CBD 10% target for forest protection were investigated here, as follows: e Analysis of the percentage protected area cover of forest types in the updated GFM. e Analysis of the percentage protected area cover of the forested area within WWF ecoregions (forest cover as identified by the updated GFM). In the following, we discuss the advantages and disadvantages regarding the use of these two methodologies for assessing the 10% target for forest protection. Global Forest Map and Forest Types The work undertaken to complete an updated GFM has resuited in a map where 78% of the world’s forest cover according to MODIS 2005 imagery has been assigned to a forest type. However this still means that 8.8 million km? of forest is not assigned (unresolved forest cover), which reduces the utility of the map as a tool for tracking protection of forest types within the framework of the CBD. The updated GFM is also not a bicgeographic classification of the world’s forests, but rather identifies 28 broader forest types at the global scale. This means, for example, that although the GFM may map tropical lowland forest cover across the globe, it does not differentiate the variation in species composition of forests in SE Asia, from those of Africa, or those of South America. This again makes the basic map of limited use for tracking the protected area coverage of biogeographically different forest ecosystems. However, it is possible to partly solve this issue by breaking the map of forest types up according to biogeographic realms. This increases the number of separate forest units from 28 to 86 but that is still not a perfect representation of forest type coverage. For example, it is still far fewer than the 732 ecoregions with forest cover that are available if the WWF ecoregions are used as a biogeographic classification of the world. The updated GFM is based on forest classifications that have been accepted by FAO and are widely used and understood by governments, which could potentially facilitate the acceptance of the updated GFM as a part of the monitoring system for the CBD forest targets. Another positive feature of the updated GFM is that it captures some forest types that are not considered by the ecoregion framework. Particularly important is the fact that the GFM recognises forests on mountains as separate types, which better reflects the distribution of some of the world’s important biodiversity features. Conclusions with regard to the GFM: e The existence of large areas of unresolved forest types within the updated GFM currently reduces its utility as a tool for tracking levels of protection of different forest types. This would need to be solved before this new map can be properly used for CBD reporting. e Completing a fully updated GFM, where all ferest areas are assigned to a forest type is a significant task, beyond the scope of the current project. This task would involve reviewing each area of ‘unresolved’ forest cover against regional and national forest datasets, and hence individually assigning these areas to the different forest types. Whilst this is a significant new piece of work, it is fundamental to the work of the CBD and other global conservation processes. 22 4 Summary and Conclusions WWF Ecoregions This report had the specific aim of identifying the degree to which WWF ecoregions are useful in forest analyses and for tracking progress with achieving the CBD target for forest conservation. The first point to be emphasised is that ‘forest’ and ‘non-forest’ ecoregions do not reflect the current situation with regard to forest cover, and analyses cannot rely on this delineation but should rather focus on the area of forest cover within all ecoregions. In addition, the ecoregions do not fully track the distribution of major forest types on the updated GFM. For example, several forest types can be present within a single ecoregion, particularly in the tropics, and some forest types such as montane forest areas and riparian forests are not accounted for in the WWF ecoregions framework. Another potential problem with the WWF ecoregions framework is that the work was undertaken by a large NGO, and is perhaps not as widely accepted by governments and UN agencies as the GFM forest types. However, the ecoregions map has the advantage of being at a finer biogeographic resolution than the updated GFM, and thus tracks species distribution patterns more accurately. As examples, there are between 77 and 172 ecoregions per realm (not including Oceania, data available upon request), which compares with 17 to 26 forest types when split according to realm (Annex). The high resolution of the ecoregions is therefore an advantage for an assessment of the representativeness of the global protected areas in terms of forest biodiversity distribution. Another advantage of the ecoregion approach is that many organizations have used the ecoregion dataset as a basis for a wide variety of regional and global biodiversity analyses. It has thus become an accepted biogeographic framework for the world, across all biomes (forest and non-forest). Conclusions with regard to the WWF ecoregions: e lf ecoregions were to be used for analyses of the CBD forest protection target, all forested ecoregions need to be considered, not only those in ‘forest’ biomes. Moreover, the level of protection needs to be calculated as the percentage of forest area within the ecoregions that is protected. A simple measure of percentage protection of the ecoregion as a whole is not regarded as useful for tracking the CBD forest target. e Ecoregions are less than perfect as a framework for measuring forest protection globally, because they fail to map mountain and riparian forests and a single ecoregion may also contain a number of forest types according to the updated GFM. e At the global scale, the assessment of forest protection by ecoregion currently provides finer biogeographic representation than forest types by realm. This is a distinct advantage of the ecoregions system for monitoring the CBD target to develop a globally representative network of forest protected areas. e At the regional scale, e.g. within ecoregions, it will be necessary to use forest classification systems with higher resolution in order to distinguish between forest types. They could be national forest classification systems or, where these do not yet exist, the GFM forest types. Finally, we conclude: e A fully updated GFM would be useful as a framework for tracking progress towards CBD forest protection targets. However, this map would be of limited use for differentiating the species diversity of forest types across the globe. To further divide the forest types according to biological patterns, existing biogeographic classifications might also be overlain on the forest map; realms for coarse scale analyses (global level), and ecoregions for fine scale analyses. This approach would provide a systematic and flexible global framework based on forest cover, forest type and biogeographic pattern, against which targets of forest protection and representativeness could be assessed. 23 4 Summary and Conclusions 4.2 Current level of protected area coverage for the world’s forests This report aimed to assess the degree to which the CBD target “at least 10% of the world’s forest types effectively conserved’ (decision VIII/15) has been achieved. In summary, the results for IUCN protected area management categories I-IV, if not mentioned otherwise,are as follows: Global forest area protected. When considering only those protected areas with an IUCN category of I-IV (strict protection in terms of biodiversity conservation), only 7.7% of the worlds forest cover is protected. If all protected areas with IUCN protected area management categories are included (IUCN I-VI), then 13.5% of the world’s forest is protected. However protected areas under IUCN categories V and VI allow some level of sustainable use, which may modify the original species assemblages of natural forests. In addition, several thousand reserves are not regarded as protected areas in this study, because they have not been assigned an IUCN category or information on IUCN category is not available in the WDPA. If we are to measure the level of global forest conservation in a more accurate way, there is a strong need to assign IUCN categories to all existing reserves in the future. Forest types protected. 19 of the 28 GFM forest types are protected below the 10% target. Only 2 temperate forest types are protected at levels above 10%, although the highest level of protection globally is afforded to Temperate broadleaf evergreen forest (28%). Needleleaf forest types lack adequate protection in both temperate and tropical regions. Forest protected by realm. Of the seven realms with forest cover, only the forests of Australasia and the Neotropics are protected above the 10% threshold. The Realms with the lowest protection are Palearctic (5.5%), Afrotropics (6.4%), and Nearctic (6.6%). Forest types protected within realm. Each realm, including Australasia and the Neotropics, has a number of forest types that do not meet the 10% targets. In addition, a number of forest types are underrepresented across several realms. These include Tropical deciduous / semi-deciduous broadleaf forest, Temperate deciduous broadleaf forest, and Tropical lowland evergreen broadleaf rain forest. Forest protected by WWF ecoregions. Within the 732 ecoregions with forest cover, only 35% achieve the 10% target for forest protection. Globally the target is best met in parts of Australasia and Latin America (Neotropics), and is being least well met in Africa and northern latitude boreal forests. Within each realm however, there is a significant proportion of ecoregions that are protected at levels of less than 1%. Forest protected in Conservation International's biodiversity hotspots. Biodiversity hotspots have already lost 70% of their habitat cover. Of the 34 hotspots, all contain forested regions, and of these 20 have less than 10% of their remaining forest habitat protected. However, even if 10% of the remaining forest is conserved, this is only a small portion of the original habitat area, especially given the high importance of hotspots in terms of narrowly endemic and threatened species. Forests protected in Conservation International's high biodiversity wilderness areas. Of the five forested high biodiversity wilderness areas, three fall below the 10% targets for forest protection. These areas of forest are largely intact and, together with hotspots, harbour the majority of vascular plant species. The forests oi the Congo have the most obvious gap regarding protected areas. Forests in New Guinea also fail to meet protection targets. 24 4 Summary and Conclusions In conclusion, the most urgent gaps in the global forest protection are: e Many temperate forest types in general fail to meet the 10% target; the Palearctic is the least protected realm and the Nearctic also falls below target protection levels. Considerable areas of Northern Taiga forest are also badly protected as identified through ecoregion analysis. e Many forests of the Afrotropics also fail to meet the 10% protection target. In this realm of high biodiversity and complex vegetation types, almost 90% of forest types fall below the 10% protection levels. The majority of hotspots in this region fail to meet protection targets, as does the Congo Basin wilderness area. The protected area gaps within the Afrotropical region are known to leave species of birds, mammals and plants unprotected (Burgess et a/. 2005; De Klerk et a/. 2004; Fjeldsa et al. 2004; Rodrigues et a/. 2004a; Rodrigues et a/. 2004b). e Across the world many WWF ecoregions have been identified as not meeting the 10% protection target, including in biogeographic realms of generally good protection, but are too numerous to mention individually here. One important example is the Cerrado forest area of Latin America, which is also a biodiversity hotspot. e Detailed gap analyses for each of these ecoregions is beyond the scope of this report, but further regional scale analysis by forest type would inform the designation of additional forest protected areas. Similarly, for those ecoregions that do appear to meet the target, analysis by forest type would identify whether the forests within these ecoregions are adequately represented by protected areas. 4.3 Utility of the 10% target The 10% target for forest types provides a representative approach to conservation planning that is easy to understand and can be measured using available data. Achieving the 10% target, however, does not automatically mean that a representative proportion of the world’s forest biodiversity will be adequately conserved. This is important in view of other political goals endorsed by the CBD - such as “a significant reduction of the current rate of biodiversity loss by 2010’. The 10% target and biodiversity distribution. As a simplistic indicator, the arbitrary 10% target does not account for the actual distribution of biodiversity within forests, including area requirements of particular species and small scale habitat variations (Rodrigues and Gaston 2001; Svancara et al. 2005). For instance, although 10% of an ecoregion might be protected, the forest protected areas might not adequately represent the ecological character, because they are too small, have the wrong shape or lack key species (Dudley and Parish 2006; Langhammer et al. 2007). Studies on species based gap analysis (Rodrigues et al. 2004a; Rodrigues et al. 2004b) and regional work on designing representative protected area networks (Cowling et al. 1999; Cowling et a/. 2003) suggest that 10% of the remaining forest protected will not be adequate to conserve biodiversity, or continue to provide ecosystem services for people. This holds true especially for the ecoregions with high species richness and high numbers of endemic species, such as Cl conservation priority areas. The 10% target and Conservation International's conservation priority areas. Our analysis shows that many hotspots have less than 10% of their forest area protected. As they have already lost most of their original habitat, this protection is a tiny proportion of the former habitat extent, and probably not sufficient to prevent further loss of endemic or threatened species. These areas require detailed regional scale plans that design a protected area network to incorporate the species distribution patterns and to capture as many of the species in viable populations as possible. This can probably only be achieved if more than 10% of the remaining forest area is protected. Similarly, the high biodiversity wilderness 25 4 Summary and Conclusions areas have intrinsic value not only for the maintenance of biodiversity pattern, which often operates across large scales, but also for the mitigation of global climate change (Saatchi et al. 2007). Thus, conservation efforts should also aim to protect more than 10% of these areas. The 10% target_and_ systematic conservation planning. Our analysis has shown that assessments of progress towards the 10% targets differ depending upon the scale of analysis, and it is important that these targets are not assessed at too coarse a scale, as this masks gaps in forest conservation. Systematic conservation planning considers biodiversity attributes of the area, the existing network of protected lands, conservation effectiveness and the costs of conservation (including social costs). It provides options for designing a fully representative network of protected areas that cover as much biological value as possible at the least possible cost. An appropriate scale for these analyses is within a single WWF ecoregion or Cl hotspot — and several such initiatives are either completed, or are underway, for example in the Congo Basin (WWF unpublished), Madagascar (WWF unpublished) and South Africa (Cowling et a/. 2003). As the Congo Basin has been highlighted for its lack of adequate protection, this is a positive step for forest conservation. Conclusions regarding the CBD 10% targets for forest protection: e The 10% protected area target has managed to draw political attention to the protection of forests and meeting this target for all ecoregions and forest types can help consolidating forest conservation worldwide. e Whilst the 10% targets for protection should be considered a baseline for protection, some areas and forest types will require greater levels of protection, whereas some will have adequate protection at less than 10%, depending on what ‘value’ is being measured (e.g., carbon storage or species richness). Such analysis was beyond the scope of the present study, but will be important for prioritising areas for protection in the future. e The generic 10% target needs reconsideration, especially for areas rich in endemic species, such as biodiversity hotspots, and for vast intact forest landscapes. In these areas the approach of ‘systematic conservation planning’ might provide a better way forward for nations under the CBD. 4.4 Limitations and caveats During the course of this project a number of limitations and caveats to the results have become evident that should be noted by the readers of the report. Global Forest Map. Although the map is a step forward in terms of global mapping of forest cover and forest types, a number of forest areas could not be assigned to a forest type within the time available. As such there are a number of unresolved forest areas that are presented as separate forest classes — when in reality they should probably be subsumed within one of the existing forest types. This makes the calculation of amounts of forest types protected problematic and the results presented here cannot be regarded as definitive. The further improvement of the updated GFM should therefore be considered as high priority action. World Database of Protected Areas. The WDPA is the only global protected area database and has been sourced from multiple places, such as national authorities, regional authorities (e.g., the Common Database on Designated Areas from the European Union), conservation NGOs, projects and individuals. For some countries the information is up-to-date, including accurate maps with IUCN protected area management categories assigned and other important attribute data for every protected area. However, many other countries do not provide accurate boundary maps and attribute data can be missing (Such as IUCN category). 26 4 Summary and Conclusions The data presented here are considered as the best available estimate of the reality on the ground, although they cannot be 100% accurate. As a new process of protected area verification and expert review is rolled out globally, we expect the WDPA to improve greatly in coming years. Areas of biodiversity importance. We have selected Cl’s biodiversity hotspots and high biodiversity wilderness areas as proxies for biologically important regions of the world. However, there are a number of other global prioritisation schemes in existence and a comprehensive assessment of the issue of protection of important areas for conservation would require a more thorough assessment than has been undertaken here. Countries versus forest types and biogeographic regions. It is important to consider that none of the biogeographic units assessed in this report conform to national boundaries. The CBD is a nationally driven process that relies upon its Parties to achieve its objectives. For this analysis we have by necessity taken a global view of the world’s forests and their biogeographic division. This makes sense from an analytical perspective and in terms of measuring overall progress towards achieving CBD targets. However, it is the nations of the world that have signed up to the CBD who will need to act upon the analyses and recommendations contained within this report. As such this document can only provide broad guidance to the CBD process and the achievement of national targets. 27 5 Recommendations 5 Recommendations 5.1 How should we measure the 10% target for forest protection? An_interim solution to measuring progress towards the CBD forest targets. Until a comprehensive GFM has been developed, we recommend that progress towards the 10% forest protection targets is measured through analysis of the level of protection for forest cover within WWF ecoregions. A proposed long term solution to measuring progress. Once a comprehensively updated GFM is available we believe this would ultimately provide a better template to measure the CBD protected area target for forests. Completing this GFM is therefore of the highest priority in order to accurately measure global progress with this CBD target. Once the GFM has been completed we recommend that: e At global level, the analysis of forest types protected per ecoregion is complemented by an analysis of forest types split by realms to account for the forest types missed out by the ecoregions system. This exercise could be taken on by a consortium of international organizations. e At regional level, i.e., within each ecoregion, the analysis of protected area coverage should be broken down by global forest types or finer-scale national classification systems. This work could form part of a process of ‘systematic conservation planning’, which is being employed to great effect in some countries. Such a process is essential if the designation of new protected areas is to contribute to a representative coverage of forest types. This exercise could be carried out by NGO and individual countries in a collaborative way. 5.2 What are the priorities for forest conservation? Close the global gaps in forest conservation. From a global perspective, the current level of forest protection is inadequate. Further protection of forest area, with a focus on certain forest types and ecoregions as presented in this study, is therefore required. Evaluate existing forest reserves without IUCN categories. |n view of the large number of existing forest reserves without IUCN protected area management categories, governments are encouraged to assess whether these areas can be assigned an IUCN category. Doing so essentially requires evaluating if an existing area conforms to the definition of a protected area first and if yes, then assigning an IUCN category based upon the management objective. Secondly, governments may wish to consider whether stricter forms of protected area management are appropriate in order to meet the CBD targets for forest biodiversity conservation. This second suggestion should be undertaken by governments in full recognition of the rights and livelihoods needs of local and indigenous peoples. Reconsider the 10% target. The CBD 10% target can be regarded as a minimum political target for forest protection. Especially for forest areas with globally significant biodiversity concentrations and for large wilderness areas, expansion of the protected area coverage above the 10% threshold is recommended (see below). Increase conservation in the most important areas for forest biodiversity. |In forest areas with globally significant concentrations of endemic species protecting 10% of the remaining 28 5 Recommendations habitat, is not likely to be sufficient to conserve the existing species and habitat values, especially if this habitat is already highly deforested as in biodiversity hotspots. Governments should therefore consider increasing their forest protection beyond 10% coverage. Large protected areas are also required in high biodiversity wilderness areas in order to encompass the nature of the biodiversity in such areas, e.g., megafauna and large scale ecosystem processes. Further update the Global Forest Map. Resolving the currently unresolved GFM forest types could greatly assist the tracking of progress in forest protection globally and regionally. Conduct systematic conservation planning. At regional level, e.g., in individual ecoregions or hotspots, detailed spatial planning might resolve the best design of a representative protected area networks for forests. Systematic planning can ensure that protected areas are located in a way that they adequately capture variations in forest species and habitats. In these plans full use should be made of all IUCN categories (IUCN I-VI) in order to design conservation landscapes that are beneficial for people and for conservation. Regional planning processes can be facilitated by GFM forest types, the ecoregion framework and national forest classification systems. Consider protected area management effectiveness. Evaluating and monitoring protected area management effectiveness is needed to ensure that existing protected areas of all IUCN categories (IUCN I-VI) are managed well enough to conserve the forest habitat and biodiversity values which they were established to protect. Enhance _ sustainable forest management outside protected areas. Protected areas are a major tool for global biodiversity conservation; however, considering the pressure on the world’s forests due to increasing demands for food, biofuels and timber there is also a strong need for sustainable forest management outside protected areas. Various forms of sustainable forest utilisation can have an important role to play in achieving conservation objectives, and forest protected areas should be integrated into the wider landscape in accordance with the ecosystem approach. 29 6 References 6 References ACHARD F., EVA H.D., STIBIG H.J., MAYAUX P., GALLEGO J., RICHARDS T. and MALINGREAU J.P. 2002. Determination of deforestation rates in the world's humid tropical forests. Science 297, 999-1002. BROOKS T.M., MITTERMEIER R.A., DA FONSECA G.A.B., GERLACH J., HOFFMANN M., LAMOREUX J.F., MITTERMEIER C.G., PILGRIM J.D. and RODRIGUES A.S.L. 2006. Global Biodiversity Conservation Priorities. Science 313, 58-61. BURGESS N.D., KUPER W., MUTKE J., WESTAWAY S., BROWN J., TURPIE S., MESHACK C., TAPLIN J., MCCLEAN C. and LOVETT J. 2005. Major gaps in the distribution of protected areas for threatened and narrow range Afrotropical plants. Biodiversity and Conservation 14, 1877-1894. BURGESS N.D., LOUCKS C., STOLTON S. and DUDLEY N. 2007. The potential of forest reserves for augmenting the protected area network in Africa. Oryx, doi: 10.1017/S0030605307001895. COWLING R.M., PRESSEY R.L., LOMBARD A.T., DESMET P.G. and ELLIS A.G. 1999. From representation to persistence: requirements for a sustainable system of conservation areas in the species-rich mediterranean-climate desert of Southern Africa. Diversity and Distributions 5, 51-71. COWLING R.M., PRESSEY R.L., ROUGET M. and LOMBARD A.T. 2003. A conservation plan for a global biodiversity hotspot: the Cape Floristic Region, South Africa. Biological Conservation 112, 191-216. DE KLERK H.M., FJELDSA J., BLYTH S. and BURGESS N.D. 2004. Gaps in the protected area network for threatened Afrotropical birds. Biological Conservation 117, 529-537. DUDLEY N. and PARISH J. 2006. Closing the Gap. Creating Ecologically Representative Protected Area Systems: A Guide to Conducting the Gap Assessments of Protected Area Systems for the Convention on Biological Diversity. Technical Series No. 24. Secretariat of the Convention on Biological Diversity, Montreal, Canada. 108 p. EUROPEAN COMMISSION, JOINT RESEARCH CENTRE 2006. Global Land Cover 2000 database. http://www-gem jrc.it/glc2000/. FAO 2005. State of the World’s Forests 2005. Food and Agriculture Organization of the United Nations, Rome. 166 p. http://www.fao.org/docrep/007/y5574e/y5574e00.htm. FJELDSA J., DE KLERK H.M., BLYTH S. and BURGESS N.D. 2004. Where are the major gaps in the reserve network for Africa's mammals? Oryx 38, 17-25. GULLISON R.E., FRUMHOFF P., CANADELL J., FIELD C.B., NEPSTAD D.C., HAYHOE K., AVISSAR R., CURRAN L.M., FRIEDLINGSTEIN P., JONES C.D. and NOBRE C. 2007. Tropical Forests and Climate Policy. Science 316, 985-986. HANSEN M., DEFRIES R., TOWNSHEND J.R., CARROLL M., DIMICELI C. and SOHLBERG R. 2006. Vegetation Continuous Fields MOD44B, 2005 Percent Tree Cover, Collection 4. University of Maryland, College Park, Maryland. HOEKSTRA J.M., BOUCHER T.M., RICKETTS T.H. and ROBERTS C. 2005. Confronting a biome crisis: global disparities of habitat loss and protection. Ecology Letters 8, 23-29. IUCN 1994. Guidelines for Protected Area Management Categories. IUCN and World Conservation Monitoring Centre (WCMC), Gland, Switzerland and Cambridge, UK. 94 p. LANGHAMMER P.F., BAKARR M.I., BENNUN L.A., BROOKS T.M., CLAY R.P., DARWALL W., DE SILVA N., EDGAR GiJ., EKEN G., FISHPOOL L.D., DA FONSECA G.A., FOSTER M.N., KNOX D.H., MATIKU P., RADFORD E.A., RODRIGUES A‘S., SALAMAN P., SECHREST W. and TORDOFF A.W. 2007. Identification and Gap Analysis of Key Biodiversity Areas: Targets for Comprehensive Protected Area Systems. IUCN, Gland, Switzerland. 132 p. 30 6 References MAGIN C. and CHAPE S. 2004. Review’ of the World Heritage Network: Biogeography, Habitats and Biodiversity. Final Draft. A Contribution to the Global Strategy for World Heritage Natural Sites. IUCN, UNESCO, UNEP-WCMC. viii+178 p. MITTERMEIER R.A., MITTERMEIER C.G., BROOKS T.M., PILGRIM J.D., KONSTANT W.R., DA FONSECA G.A.B. and KORMOS C. 2003. Wilderness and biodiversity convention. PNAS 100, 10309-10313. MITTERMEIER R.A., ROBLES GIL P., HOFFMANN M., PILGRIM J., BROOKS T., MITTERMEIER C.G., LAMOREUX J. and DA FONSECA G.A. 2004. Hotspots Revisited: Earth's Biologically Richest and Most Endangered Terrestrial Ecoregions. CEMEX, Mexico City. 392 p. OLSON D.M., DINERSTEIN E., WIKRAMANAYAKE E.D., BURGESS N.D., POWELL G.V.N., UNDERWOOD E.C., D'AMICO J.A., ITOUA |., STRAND H.E., MORRISON J.C., LOUCKS C.J., ALLNUTT T.F., RICKETTS T.H., KURA Y., LAMOREUX J.F., WETTENGEL W.W., HEDAO P. and KASSEM K.R. 2001. Terrestrial Ecoregions of the World: A New Map of Life on Earth. BioScience 51, 933-938. PATRY M. and RIPLEY S. 2007. World Heritage Forests: Leveraging Conservation at the Landscape Level. Proceedings of the 2nd World Heritage Forest Meeting, Nancy, March 9-11, 2005. World Heritage reports 21. UNESCO World Heritage Centre, Paris. 173 p. RODRIGUES A.S.L., AKCAKAYA H.R., ANDELMAN S.J., BAKARR M.I., BOITANI L., BROOKS T.M., CHANSON J.S., FISHPOOL L.D.C., DA FONSECA G.A.B., GASTON K.J., HOFFMANN M., MARQUET P.A., PILGRIM J.D., PRESSEY R.L., SCHIPPER J., SECHREST W., STUART S.N., UNDERHILL L.G., WALLER R.W., WATTS M.E.J. and YAN X. 2004a. Global Gap Analysis: Priority Regions for Expanding the Global Protected-Area Network. BioScience 54, 1092-1100. RODRIGUES A.S.L., ANDELMAN SwJ., BAKARR M.|., BOITANI L., BROOKS T.M., COWLING R.M., FISHPOOL L.D.C., DA FONSECA G.A.B., GASTON K.J., HOFFMANN M., LONG J.S., MARQUET P.A., PILGRIM J.D., PRESSEY R.L., SCHIPPER J., SECHREST W., STUART S.N., UNDERHILL L.G., WALLER R.W., WATTS M.E.J. and YAN X. 2004b. Effectiveness of the global protected area network in representing species diversity. Nature 428, 640-643. RODRIGUES A.S.L. and GASTON K.J. 2001. How large do reserve networks need to be? Ecology Letters 4, 602-609. SAATCHI S.S., HOUGHTON R.A., DOS SANTOS ALVALA, R.C., SOARES J.V. and YU Y. 2007. Distribution of aboveground live biomass in the amazon basin. Global Change Biology, 13(4), 816-837. SANDERSON E.W., JAITEH M., LEVY M.A., REDFORD K.H., WANNEBO A.V. and WOOLMER G. 2002. The Human Footprint and the Last of the Wild. BioScience 52, 891-904. SCHMITT C.B. 2007. Approaches for setting global conservation priorities. /n A global network of forest protected areas under the CBD: Opportunities and challenges. Proceedings of an International Expert Workshop, Freiburg, May 9-11, 2007. Freiburger Schriftenreihe zur Forst- und Umweltpolitik, Eds Schmitt C.B., Pistorius T. and Winkel G. pp 9-37. Verlag Kessel, Remagen. SVANCARA L.K., BRANNON R., SCOTT J.M., GROVES C.R., NOSS R.F. and PRESSEY R.L. 2005. Policy-driven versus Evidence-based Conservation: A Review of Political Targets and Biological Needs. BioScience 55, 989-995. UNEP/CBD/SBSTTA/11/INF/2 Report of the Inter-Sessional (Second) Meeting of the Ad Hoc Technical Expert Group on the Review of Implementation of the Programme of Work on Forest Biological Diversity. 28 April 2005. 24 p. UNEP-WCMC 2000. Global Distribution of Current Forests, United Nations Environment Programme - World Conservation Monitoring Centre (UNEP-WCMC). http://www.unep- wemc.org/forest/global_map.htm. 31 Annex: Protection of forest types within each biogeographical realm Annex: Protection of forest types within each biogeographic realm Total forest area and percentage protected area coverage at IUCN categories I-IV and I-VI for forest types as identified by the updated Global Forest Map (GFM). Unresolved forest types indicate areas of forest that could not be identified to be consistent with the existing GFM forest types in the timeframe available. They are grouped according to Global Land Cover (GLC) classes. Forest types highlighted in grey are below the 10% target for protection at IUCN I-IV. Forest Type Total Area % Protected % Protected (‘000 km?) (IUCN I-IV) (IUCN I-VI) Australasia Tropical deciduous/semi-deciduous broadleaf forest 17 3.7 5.8 Temperate evergreen needleleaf forest 18 3.8 4.4 Unresolved mosaic: tree cover / other natural vegetation 20 4.2 7 Tropical lowland evergreen broadleaf rain forest 417 6.4 11.9 Tropical semi-evergreen moist broadleaf forest 62 7.3 12.2 Unresolved forest unknown type 3 9.9 13.9 Tropical sparse trees/parkland 119 10.2 11.2 Tropical sclerophyllous dry forest 58 10.3 11.6 Tropical lower montane forest 52 ee 18.6 Unresolved tree cover, broadleaved, evergreen 240 12.3 16.3 Tropical freshwater swamp forest 79 12.8 19.2 Temperate sparse trees/parkland 155 13.4 14.1 Tropical upper montane forest 117 13.9 30.3 Tropical mangrove 36 16.5 19.6 Unresolved tree cover, broadleaved, deciduous, open 135 1723 18.2 Temperate sclerophyllous dry forest 182 23.6 24.4 Unresolved tree cover, regularly flooded, saline water 0.05 26.2 26.7 Unresolved tree cover, mixed leaf type 34 31.1 51 Temperate broadleaf evergreen forest 38 59.3 72.1 Afrotropics Tropical thorn forest 1 0 0 Unresolved tree cover, regularly flooded, saline water 6.001 0 0 Tropical lower montane forest 29 0.7 41 Tropical freshwater swamp forest 191 3 6.9 Tropical mangrove 36 3.8 12.9 Unresolved tree cover, broadleaved, deciduous, closed 14.2 Unresolved forest unknown type 14.7 Unresolved tree cover, broadleaved, evergreen 13.6 Unresolved mosaic: tree cover / other natural vegetation 13.3 Temperate deciduous broadleaf forest 724 Temperate sparse trees/parkland 10 Unresolved tree cover, broadleaved, deciduous, open 16.5 Tropical semi-evergreen moist broadleaf forest 18 6.6 28 Tropical lowland evergreen broadleaf rain forest 1,526 8 18.2 Tropical sparse trees/parkland 19.6 Temperate mixed broadleaf/needleleaf forest 47.2 Tropical upper montane forest 26.3 Indo-Malay Temperate freshwater swamp forest 0 Tropical sclerophyllous dry forest 34.5 Temperate sparse trees/parkland 69 Temperate deciduous broadleaf forest 12.9 Unresolved mosaic: tree cover / other natural vegetation 9 Temperate evergreen needleleaf forest 7.2 Annex: Protection of forest types within each biogeographical realm Forest Type Total Area % Protected % Protected (‘000 km’) (IUCN I-IV) (IUCN I-VI) Indo-Malay (continued Temperate sclerophylious dry forest 5 3.8 11.6 Unresolved tree cover, broadleaved, evergreen 533 5.7 13 Unresolved tree cover, broadleaved, deciduous, closed 277 6.1 6.9 Tropical freshwater swamp forest 62 6.6 16.5 Unresolved tree cover, broadleaved, deciduous, open 1 7 7 Tropical mangrove 30 10.5 18.4 Unresolved forest unknown type 1 Dales) Us) Tropical thorn forest 5 12 18.3 Tropical lowland evergreen broadleaf rain forest 383 12.2 23.3 Temperate mixed broadleaf/needleleaf forest 38 13.4 18.3 Tropical lower montane forest aly 15.4 32.4 Tropical deciduous/semi-deciduous broadleaf forest 283 18.5 20 Tropical needleleaf forest 6 21.4 35.6 Tropical sparse trees/parkland 3 26.5 32.3 Tropical semi-evergreen moist broadleaf forest 128 30 36.8 Tropical upper montane forest 41 34.8 56 Unresolved tree cover, regularly flooded, saline water 0.003 71.5 71.5 Nearctic Tropical mixed needleleaf/broadleaf forest 0.31 0.2 2.2 Tropical deciduous/semi-deciduous broadleaf forest 1 0.3 2.3 Unresolved tree cover, broadleaved, evergreen 0.14 0.5 6.7 Tropical lower montane forest 16 1.3 10 Unresolved tree cover, burnt 9 1.8 3.7 Unresolved tree cover, mixed leaf type 110 3.1 10.6 Temperate deciduous broadleaf forest 957 A 23.7 Temperate deciduous needleleaf forest 10 76.1 Temperate mixed broadleaf/needleleaf forest 1,113 8.8 Temperate sparse trees/parkland 1,036 : 7.8 Temperate evergreen needleleaf forest 3,431 A 21.1 Tropical mangrove 0.15 9.1 Unresolved forest unknown type 1 j 26.5 Unresolved mosaic: tree cover / other natural vegetation 166 25:5 27.8 Neotropics Temperate sclerophyllous dry forest 35 0 2.7 Unresolved tree cover, regularly flooded, saline water 1 0.4 0.6 Temperate deciduous broadleaf forest 41 1.1 2.6 Temperate sparse trees/parkland 47 1.1 5.4 Temperate freshwater swamp forest 0.37 1.5 1.9 Unresolved tree cover, broadleaved, deciduous, closed 468 3.1 8.9 Unresolved tree cover, broadleaved, deciduous, open 142 4.3 9.8 Temperate evergreen needieleaf forest 8 4.4 17.2 Tropical mixed needleleaf/broadleaf forest 9 45 6.9 Tropical needleleaf forest 72 45 10.1 Tropical sparse trees/parkland 409 5.3 17.3 Unresolved tree cover, broadleaved, evergreen 1,545 8 29.1 Tropical thorn forest 4 8.6 30.8 Tropical freshwater swamp forest 204 9.3 36.8 33 Annex: Protection of forest types within each biogeographical realm Forest Type Total Area % Protected % Protected (‘000 km?) (IUCN I-IV) (IUCN I-VI) Neotropics (continued) Tropical deciduous/semi-deciduous broadleaf forest 129 9.7 32.4 Tropical lowland evergreen broadleaf rain forest 4,161 11.4 41.9 Tropical lower montane forest 236 13.9 19.4 Tropical semi-evergreen moist broadleaf forest 635 16.6 50.8 Tropical sclerophyllous dry forest 183 17.9 18.3 Temperate broadleaf evergreen forest Tropical mangrove er montane forest Unresolved mosaic: tree cover / other natural vegetation 0.21 eon 30.9 Unresolved forest unknown type Oceania Unresolved forest unknown type = ; 8. 2 Tropical lowland evergreen broadleaf rain forest Tropical mangrove 0.01 94. 3 Tropical thorn forest 0.01 Unresolved tree cover, broadileaved, evergreen 106 0. : 13. : Temperate mixed broadleaf/needleleaf forest 3 2 y 8.5 Temperate sclerophylious dry forest Unresolved tree cover, broadleaved, deciduous, closed 1 = : = 10. 5 L * Temperate deciduous needleleaf forest BE 2 Unresolved forest unknown type Unresolved mosaic: tree cover / one natural vegetation = 6. : Unresolved tree cover, mixed leaf Unresolved tree cover, regularly flooded, saline water 0.01 6.8 6.8 Temperate deciduous broadleaf forest 1,522 Unresolved tree cover, burnt 22 Unresolved tree cover, broadleaved, deciduous, open 21 Temperate freshwater swamp forest hee 34