38°E

40°E

40°E

10°S10°S

36°E

8°S

36°E

6°S

4°S

38°E

2°S

4°S

40°E

40°E

2°S

42°E

42°E

6°S

8°S

6000

3000

2000

1500

1000

500

100

Elevation (in meters)

scale: 1/6,250,000data: SRTM

Topography40 0 40 80 120 160

kilometers

Kilimanjaro

SOMALIA

K E N Y A

T A N Z A N I A

M O Z A M B I Q U E

Taita Hillsscale: 1/190,000source image resolution: 20 meters2 0 2 4 6 8

kilometers

CEPF funding area

Alliance for Zero Extinction (AZE) site

Key Biodiversity Area (extent, geographic extent unavailable)

international political border

district border

land cover

25 0 25 50 75

kilometers

1990—2000 Forest Cover and Change inEastern Arc Mountains and Coastal Forests

scale: 1/1,000,000projection: Lamberts Equal Area Azimuthal 38° east longitude 7° south latitudesource image resolution: 28.5 meters

town (national capitals underlined) (extent, location only)

mangroves

forest lost by 2000

mangrove lost by 2000

woodland lost by 2000

sea (outside of analysis)

water

cloud datesboth

cloud 1990s nonforest 2000

nonforest datesboth

forest and woodland 1990s cloud 2000

woodland both dates

forest both dates

wetland

CEPF funding area

1990—2000 Forest Cover and Change in

Eastern Arc Mountains and Coastal Forests of Tanzania andKenya1 / 1,000,000

Ea

stern

Arc

Mou

nta

ins

an

d C

oa

sta

l Fore

sts

· 1

990—

2000

Fo

rest

Co

ver

an

d C

ha

ng

e i

nS O M A L I A

K E N Y A

T A N Z A N I A

M O Z A M B I Q U E

MethodologyEastern Arc Mountains and Coastal Forests Taita HillsForest cover and change was mapped by analyzing Landsat A multi-scale segmentation/object relationship modeling satellite imagery from circa 1990 and circa 2000. Most of the (MSS/ORM) approach was applied to map land cover at a images were obtained for free from NASA's Geocover project, landscape level in the Taita Hills. The software tool used was which stores data at the University of Maryland's Global Land eCognition. Various segmented image object spectral, contextual Cover Facility ( ). Additional images for and hierarchical properties were utilized in the classification cloudy areas were purchased from USGS and SPOT to have process. The output map was subject to final visual inspection more complete coverage. and manual editing of any noted errors, relative to the SPOT

imagery. Overall map accuracy of 89% was calculated using The analysis was conducted at a spatial resolution of 28.5 meters. ground reference test data collected during field visits to the Taita The Landsat images from circa 1990 and circa 2000 were Hills in January 2005 and 2006 and from 0.5m resolution true-combined into one file, and the classification of forest cover and color digital aerial photography flown in January 2004 change was conducted in a single process with the multi- (3 months after the SPOT acquisition date).temporal data. The classification algorithm applied was a supervised maximum likelihood classifier. In this process, analysts delineate training sites for each land cover or change class, based on visual interpretation, and referring to ground reference data and high-resolution QuickBird imagery available on Google Earth. The entire Landsat images are classified based on the statistics of the image data in each class. The final classification was filtered to remove patches of less than 2

2hectares. This method follows that reported in Harper et al.

The Eastern Arc forest and woodland cover and change analysis incorporated ground truthing data from the following mountain blocks: Uluguru, Ukaguru, Rubeho, Mahenge, Image and Udzungwa. Twenty-meter SPOT imagery was also used to assist image interpretation for the East Usambara, Nguru, Nguu, Ukaguru, Rubeho, and Uluguru mountain blocks.

Accuracy of the land cover classification of the Coastal Forest region was estimated using high-resolution QuickBird imagery acquired between 2000 and 2001. Over 3,000 randomly-selected points from available high-resolution images were used in the accuracy assessment. These points were visually interpreted and labeled according to their true class, and then intersected with the final map to determine areas of agreement and disagreement. The overall map accuracy is 88% for forest cover in circa 2000.

2. Harper, G., Steininger, M.K., Tucker, C.J., Juhn, D. and Hawkins, 2007. Fifty Years of Deforestation and Forest Fragmentation in Madagascar. Environmental Conservation 34:1-9.

www.landcover.org

F.

Forest Analysis DataCoastal ForestsBaseline forest cover analysis for the Coastal Forests of Tanzania and Kenya was performed as part of a CEPF-funded BirdLife International project, “Instituting a standardized, sustainable biodiversity monitoring system in the Eastern Arc Mountains and Coastal Forests Hotspot.”

Analysis conducted by:B.P. Mbilinyi and J. Kashaigili Sokoine University of Agriculture. Morogoro, TZ K. Tabor and M. Steininger Center for Applied Biodiversity Science. Arlington, VA USAAcknowledgment: Neil Burgess, John Watkin

Eastern Arc MountainsThe forest area baseline for the Eastern Arc Mountains was commissioned by the Forestry and Beekeeping Division (FBD) of the Ministry of Natural Resources and Tourism in Tanzania, through the project Conservation and Management of Eastern Arc Mountain Forests (CMEAMF) financed by the Global Environment Facility (GEF) through the United Nations Development Programme (UNDP). Analysis conducted by:B.P. Mbilinyi, R.E.Malimbwi, D.T.K. Shemwetta, A. Songorwa, E. Zahabu, J.Z. Katani and J. Kashaigili. Sokoine University of Agriculture. Morogoro, TZAcknowledgment: Neil Burgess, Felician Kilahama, and Ministry of Natural Resources and Tourism

Taita HillsForest cover analysis of the Taita Hills carried out as part of the

1TAITA project funded by the Academy of Finland.

Analysis conducted by:B.J.F. Clark and P.K.E. Pellikka University of Helsinki. Helsinki, FI

1. B.J.F. segmentation and object orientated classification. In: Aplin, P. (ed.) Recent Advances in Remote Sensing and Geoinformation Processing for Land Degradation Assessment. Taylor & Francis.

Clark and P.K.E. Pellikka, 2008. Landscape analysis using multiscale

data:BirdLife InternationalCenter for Applied Biodiversity Science at Conservation International, Arlington, VA USAForestry and Beekeeping Division Ministry of Natural Resources and Tourism, TanzaniaGlobal Shoreline Database, January 2001, Veridian (GDAIS)Institute of Resource Assessment at the University of Dar es SalaamInternational Livestock Research InstituteKenya Forest ServiceNature KenyaVMap0, National Geospatial-Intelligence AgencyWildlife Conservation Society of TanzaniaWorld Wildlife Fund - TanzaniaWorld Wildlife Fund - United States

this map was produced by the Conservation Mapping Program Mark Denil - Chief CartographerKellee Koenig - CartographerCenter for Applied Biodiversity ScienceConservation International 2011 Crystal Drive Arlington, VA 22202 USA© June 2008 Conservation International

The Critical Ecosystem Partnership Fund is a joint initiative of l'Agence Française de Développement, Conservation International, the Global Environment Facility, the Government of Japan, the John D. and Catherine T. MacArthur Foundation, and the World Bank. A fundamental goal is to engage nongovernmental organizations, community groups, and other sectors of civil society in conserving Earth's biodiversity hotspots.

The political and geographic designations shown on this map do not imply the expression of any opinion on the part of Conservation International or any of its partners concerning the legal status or delineation of the frontiers of any country, territory or area.

Voi

Mombasa

Tanga

Moshi

Nairobi

Stonetown

Dar es SalaamMorogoro

Dodoma

Iringa

Lindi

MorogoroMorogoro

Lindi

Mombasa

Voi

MoshiMoshi

Iringa

Dar es Salaam

Dodoma

Stonetown

Tanga

0°

30°E

30°N

30°S

equator

The East Africa Natural History Society

NatureKenya

Pemba

Unguja

Mafia

Roster of Key Biodiversity Areas† Denotes Alliance for Zero Extinction (AZE) site

†

†

†

†

†††

†

1 Arabuko-Sokoke Forest 69 Marafa 2 Bagamoyo 70 Marenji Forest 3 Bagamoyo (Kikoka Forest Reserve) 71 Masasi 4 Bagamoyo District Coastal Forests 72 Masasi (Nyengedi) 5 Boni Forest 73 Mida Creek, Whale Island and the 6 Buda Forest Reserve Malindi-Watamu Coast 7 Bungu 74 Mikindani (Mnima) 8 Cha Shimba 75 Mikumi National Park 9 Chale Island 76 Mkomazi Game Reserve10 Chivara 77 Mnazi Bay11 Chuna Forest 78 Mount Kasigau12 Dakatcha Woodland 79 Mrima Hill Forest13 Dar es Salaam Coast 80 Mtwara District Coastal Forests14 Diani Forest 81 Muheza District Coastal Forests15 Dodori 82 Mwache Forest Reserve16 Dzitzoni 83 Newala (Kitama)17 Dzombo Hill Forest 84 Newala (Kitangari)18 East Usambara Mountains 85 Newala (Mahuta)19 Gede Ruins National Monument 86 Newala District Coastal Forests20 Gongoni Forest Reserve 87 Nguru Mountains21 Handeni District Coastal Forests 88 Nguu Mountains22 Jozani National Park, Zanzibar 89 North Pare Mountains23 Kambe Rocks 90 Nyumba ya Mungu Reservoir24 Kaya Bombo 91 Nzovuni River25 Kaya Chonyi 92 Pande Game Reserve and26 Kaya Dzombo Dondwe Coastal Forests 27 Kaya Fungo 93 Pangani28 Kaya Gandini 94 Pangani (Bushiri)29 Kaya Gonja 95 Pangani (Mwera)30 Kaya Jibana 96 Pangani Dam31 Kaya Kambe 97 Pangani District Coastal Forests32 Kaya Kauma 98 Pangani River (Hale-Makinjumbe)33 Kaya Kinondo 99 Pangani River (Mauri)34 Kaya Lunguma 100 Pemba Island35 Kaya Msambweni 101 Ras Kituani36 Kaya Mtswakara 102 Rubeho Mountains37 Kaya Muhaka 103 Rufiji Delta38 Kaya Mwarakaya 104 Rufiji District Coastal Forests39 Kaya Puma 105 Sabaki River Mouth40 Kaya Rabai 106 Selous Game Reserve41 Kaya Ribe 107 Shimba Hills42 Kaya Sega 108 Shimoni Forests43 Kaya Teleza 109 South Pare Mountains44 Kaya Tiwi 110 Taita Hills Forests45 Kaya Ukunda 111 Tana River Delta46 Kaya Waa 112 Tana River Forests47 Kilombero Valley 113 Tanga (Duga)48 Kilwa District Coastal Forests 114 Tanga (Gombero Forest Reserve)49 Kisarawe District Coastal Forests 115 Tanga (Morongo)50 Kisiju 116 Tanga (Nyamaku)51 Kisimani wa Ngoa 117 Tanga (Pangani)52 Kisite Island 118 Tanga North - Kibo Saltpans53 Kiunga Marine National Reserve 119 Tanga South54 Korogwe (Kwashemshi Sisal Estate) 120 Tsavo West National Park55 Latham Island 121 Tumbatu Island56 Lindi 122 Udzungwa Mountains57 Lindi (Mkindani) 123 Ukaguru Mountains58 Lindi (Ngongo) 124 Uluguru Mountains59 Lindi (Nyangao River) 125 Uvidunda Mountains60 Lindi (Ras Rungi) 126 Uzaramo (Dar to Morogoro)61 Lindi (Tendaguru) 127 Uzaramo (Msua)62 Lindi Creek 128 West Usambara Mountains63 Lindi District Coastal Forests 129 Witu Forest Reserve64 Lunghi 130 Zanzibar (Kituani)65 Mafia Island 131 Zanzibar (Muyuni)66 Magombera Forest Reserve 132 Zanzibar Island-East Coast67 Mahenge Mountains 133 Zanzibar Island-South Coast68 Mangea Hill

Phryn

obatra

chus kreffti, male / Elizabeth B. Harper

Tana

Ga l ana

Pangan i

Wa mi

Ruvu

Ru f i j i

Ruvuna

I N D I A N

O C E A N

Ki l omb e r o

K i l omb e r o

maparea

Zanzibar

he Eastern Arc Mountains and Coastal Forests of Tanzania and Kenya has among the highest density of Tendemic species of any biodiversity hotspot in the world.

The combination of extraordinary species endemism and an extremely high degree of threat led to the designation of this region as a biodiversity hotspot. This is one of the regions of the world that is likely to witness significant species extinctions in the coming decades. In 2004, a refinement of the original hotspots analysis resulted in this region being included as important parts of two larger, separate hotspots: the Eastern Afromontane – the montane regions of eastern and north-eastern Africa, and the Coastal Forests of Eastern Africa – which include a lowland forest mosaic along the coast of East Africa. However,

2this area still covers over 30,000 km , thus further prioritization is required to direct investment towards conservation sites on the ground. The Critical Ecosystem Partnership Fund (CEPF) investment in the region is focused on conserving the region's 333 globally threatened species, which are primarily found within 133 Key Biodiversity Areas (KBAs), which are sites of global significance for biodiversity conservation. The highest priority KBAs are called Alliance for Zero Extinction (AZE) sites; these sites contain one or more highly threatened species found nowhere else on Earth, thus, if an AZE site is lost one or more species will be extinct. CEPF's strategy in the region is to select sites for conservation intervention that deliver maximum conservation impact in terms of preventing extinctions, restore and increase connectivity among fragmented forest patches, increase protected area coverage, and assist communities to develop alternative livelihoods that prevent further forest destruction. Mapping the baseline of forest cover is key to carrying out this strategy. Finally, monitoring KBAs and AZE sites provides an important measure of both overall threat to biodiversity and success of current conservation efforts.

Habitat loss is the primary factor threatening species with extinction. Thus, the single most effective means of conserving biodiversity worldwide is by protecting Key Biodiversity Areas—sites where threatened and restricted-range species occur. Remote sensing data helps us understand the human and natural modifications to habitats that support important biodiversity. Fine-resolution, low-cost satellite data can be analyzed to track changes over time in habitat cover. In this case we mapped change in primary forest and woodland cover within KBAs between 1990 and 2000. In East Africa, primary forest and woodland are important habitats for many globally threatened and range-restricted species, with the forests being of particular importance.

Change in primary forest and woodland cover from 1990 to 2000 within Key Biodiversity Areas (KBAs) and Alliance for Zero Extinction (AZE) sites

The chart above shows change in the proportion of natural habitat cover for all KBAs and AZE sites. Findings across the extent of analysis show that overall habitat extent is greater for AZE sites compared to all KBAs. However, a higher rate of habitat decline is seen within AZE sites compared with all KBAs during the 10-year period. Similar overall habitat extent and rate of loss is seen for Tanzanian KBAs and AZE sites. For Kenya however, results show a lower rate of habitat loss in KBAs and AZE sites.

ConclusionThere is a great need to continue directing conservation investments within KBAs; in particular AZE sites where habitat loss threatens species more susceptible to extinction. Such conservation efforts need to be based upon sound science and address the needs of the communities reliant on the products and services provided by the forests.

Monitoring and reporting of biodiversity status, pressures and conservation responses must also continue over the long-term. In a world of limited resources and increased accountability, it is essential that such biodiversity monitoring efforts serve as a central means for guiding government policy and conservation planning and investment in East Africa. By systematically monitoring national and regional biodiversity status and trends we can support both reporting obligations and natural resource decision-making processes; specifically where, how and why future biodiversity conservation investment and strategies should be dedicated.

The chart above shows change in the proportion of natural habitat cover for all KBAs and AZE sites. Findings across the extent of analysis show that overall habitat extent is greater for AZE sites compared to all KBAs. However, a higher rate of habitat decline is seen within AZE sites compared with all KBAs during the 10-year period. Similar overall habitat extent and rate of loss is seen for Tanzanian KBAs and AZE sites. For Kenya however, results show a lower rate of habitat loss in KBAs and AZE sites.

ConclusionThere is a great need to continue directing conservation investments within KBAs; in particular AZE sites where habitat loss threatens species more susceptible to extinction. Such conservation efforts need to be based upon sound science and address the needs of the communities reliant on the products and services provided by the forests.

Monitoring and reporting of biodiversity status, pressures and conservation responses must also continue over the long-term. In a world of limited resources and increased accountability, it is essential that such biodiversity monitoring efforts serve as a central means for guiding government policy and conservation planning and investment in East Africa. By systematically monitoring national and regional biodiversity status and trends we can support both reporting obligations and natural resource decision-making processes; specifically where, how and why future biodiversity conservation investment and strategies should be dedicated.

Mtwara

Mtwara

Micheweni

Wete

Chake chake

Mkoani

Tanga

Muheza

Pangani

Handeni

Korogwe

Lushoto

Same

Mwanga

Moshi Rural

Simanjiro

Bagamoyo

MorogoroRural

Kilosa

North A

North B

West Central

South

Urban

Kibaha

Kinondoni

Ilala Temeke

Mkulanga

Kisarawe

Rufiji

Mafia

Liwale

Kilwa

Ruangwa

Lindi Rural

nabrU arawt

M

TandahimbaNewala

Masasi

Ulanga

i l oK m b e r o

Mufindi

MpwapwaDodomaRural

Iringa Rural

Lamu

Malindi

Kilifi

Tana River

Taita Taveta

Garissa

Kwale

Mombasa

Moshi UrbanMoshi Urban

Kilindi

Kiteto

MorogoroUrban

IringaUrban

Kilolo

Njombe

Mtwara Rural

Lindi Urban

A F R I C A

102

3

100

718

54

4

21

81

97

94

95

96

98

99

119

113115

116

117

93

121

132130

2

13

22

49

87

92

133

127

131

126

66

75

124

125

123

47

67

106

122

48

50

65

104

103

101

61

62

59

56

57

5860

63

71

8586

77

80

83

84

74

111

1

19

51

68

69

73

105

12

10

8

16

27

32

36

39

17

31

11

6

9

14

26

20

23

24

25

28

29

30

33

34

35

37

38

40 41

42

43

4445

46

52

70

79

8291

107

108

118

112

129

5

15

53

64

120 110

78

76

8990

109

128

114

88

72

55

extent of Taita analysis / see insetextent of Taita analysis / see inset

extent of Coastal Forest analysis

extent of Coastal Forest analysis

extent of Coastal Forest analysis

extent of Coastal Forest analysis

ext

ent of C

oast

al F

ore

st a

naly

sis

ext

ent of C

oast

al F

ore

st a

naly

sis

ext

ent of C

oast

al F

ore

st a

naly

sis

ext

ent of E

ast

ern

Arc

analy

sis

ext

ent of E

ast

ern

Arc

analy

sisextent of Eastern Arc analysis

extent of Eastern Arc analysis

extent of Eastern Arc analysis

extent of Eastern Arc analysis

ext

ent of C

oast

al F

ore

st a

naly

sis

ext

ent of C

oast

al F

ore

st a

naly

sis

extent of Eastern Arc analysis

extent of Eastern Arc analysis

P r i m a r y F o r e s t a n d W o o d l a n d C o v e r C h a n g e 1 9 9 0 — 2 0 0 0

Primary forest and woodland loss statistics by district within extent of analysis

Path165165165165166166166166166166167167167167168168

Row61626667616263646566636465666566

Date c199009/01/199518/03/198812/03/199307/06/199126/01/198725/06/199216/06/198626/01/198709/06/199508/06/198901/01/198702/07/199505/06/199105/06/199111/07/199011/07/1990

Date c200001/03/200522/03/200122/05/200022/05/200019/07/200122/01/200030/01/200330/01/200330/06/200030/06/200025/10/199917/08/200307/07/200025/10/199901/11/199901/11/1999

AnalysisCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestCoastal ForestEastern ArcEastern ArcEastern ArcEastern ArcEastern ArcEastern Arc

Landsat imagery used in Eastern Arc andCoastal Forest analyses

SPOT imagery used in Taita Hills analysisDate19902000

Path143357

Row143357

SensorSPOT 2SPOT 4

by KBA by district

Primary forest and woodland loss statistics in Key Biodiversity Areas with greater than 2

1 km primary forest and woodland cover in c1990

† Denotes Alliance for Zero Extinction (AZE) site

CF = Coastal Forests FR = Forest Reserve GR = Game Reserve

in extent of analysis

square kilometersTanzania KBABagamoyo (Kikoka FR)Bagamoyo District CFEast Usambara MountainsHandeni District CFKilombero ValleyKilwa District CFKisarawe District CFLindi (Nyangao River)Lindi CreekLindi District CFMagombera FRMahenge MountainsMikindani (Mnima)Mikumi National ParkMnazi BayMuheza District CFNewala (Kitama)Newala District CFNguru MountainsNguu MountainsNorth Pare MountainsPande GR and Dondwe CFPangani District CFRubeho MountainsRufiji DeltaRufiji District CFSelous GRSouth Pare MountainsUdzungwa MountainsUkaguru MountainsUluguru MountainsUvidunda MountainsWest Usambara Mountains

c1990

7.87113.42268.4147.30448.35994.61202.9022.894.03311.403.6023.017.09306.6142.3315.1619.2535.18280.39226.3810.918.9120.92303.4879.6974.764276.10143.082419.55151.93308.0689.99243.45

c2000

7.84106.30259.9444.19

167.91990.53201.7022.253.99

304.880.92

21.857.09

289.5239.3014.5118.6829.53

265.02189.2410.427.41

20.67277.3977.3874.10

4015.32126.00

2012.48143.56280.4726.62

224.47

loss

0.376.283.166.57

62.550.410.592.771.012.09

74.545.00

05.577.174.282.95

16.045.48

16.404.48

16.881.178.602.900.876.10

11.9416.825.518.96

70.427.80

†

†

†††

†

34

182147484959626366677475778183868788899297

102103104106109122123124125128

percent

Kenya KBAArabuko-Sokoke ForestBoni ForestBuda FRDakatcha WoodlandDodoriDzombo Hill ForestGongoni FRKaya DzomboKaya GonjaLunghiMangea HillMarafaMarenji ForestMount KasigauMrima Hill ForestShimba HillsShimoni ForestsTaita Hills ForestsTana River DeltaTana River ForestsWitu FR

c1990

325.60224.10

4.0137.6120.571.294.482.201.07

125.978.803.72

12.292.152.75

124.427.094.13

10.5552.4226.26

c2000

325.26224.10

4.0032.9420.571.294.482.201.07

125.956.703.38

12.292.142.75

124.367.043.15

10.5449.8226.26

loss

0.100

0.1212.42

000

0.260

0.0223.899.24

00.42

00.050.60

23.730.124.96

0

†

†

156

121517202629646869707879

107108110111112129

square kilometers percent Kenya District GarissaKilifiKwaleLamuMalindiMombasaTana River

total area

18,4864,6468,4606,1407,748

21321,270

c1990

1,215243232963659

less than 1357

c2000

1,211232228962620

less than 1348

loss

0.354.641.650.165.905.542.32

square kilometers percent

Tanzania DistrictBagamoyoHandeniIlalaIringa UrbanKibahaKilindiKiloloKilomberoKilosaKilwaKinondoniKisaraweKitetoKorogweLindi RuralLindi UrbanLiwaleLushotoMasasiMkurangaMorogoro RuralMorogoro UrbanMpwapwaMtwara RuralMtwara UrbanMufindiMuhezaMvomeroMwangaNewalaNjombePanganiRuangwaRufijiSameSongea RuralTandahimbaTangaTemekeUlanga

total area

8,5358,045

335147

1,8368,983

13,05413,03311,754

5474,8561,8946,2943,1376,315

24020,3824,0811,0052,722

12,977535

4,9283,860

1956,7614,2096,2401,7501,4594,1631,7851,382

13,1776,338

5011,882

594749

17,066

c1990

1,4212,810

130

4382,2821,6831,2554,738

351,771

9966296

3,47448

4,640577188952

1,40329

1602,142

7053752460636

353155263633

3,76723526

7716

332,102

c2000

1,2401,676

110

4141,812

883864

4,64127

1,7204

604235

3,37846

4,540386183888

1,14622

1221,947

6436049546331

33470

237628

3,665172

3691

432

1,377

loss

12.7540.3410.32

05.56

20.6147.5331.182.04

23.532.92

54.0537.4620.472.773.362.17

33.152.286.76

18.2921.8223.969.117.67

32.855.42

23.5914.765.46

55.1710.020.772.71

26.8889.3210.4320.113.69

34.50

square kilometers percent

1990 2000

AZE sites

Year

Perc

enta

ge o

f Fore

st a

nd W

oodla

nd C

ove

r (%

)

all KBAs

70

60

50

40

30

20

10

0

Tanzania

1990 2000

AZE sites

Year

all KBAs

Kenya

1990 2000

AZE sites

Year

all KBAs

extent of analysis

Change in primary forest and woodland cover within Key Biodiversity Areas and AZE sites 2with greater than 1 km forest and woodland cover in c1990.