Status of the protected areas of the Central African Republic

BIOLOGICAL CONSERVATION

Biological Conservation 118 (2004) 479–487 www.elsevier.com/locate/biocon

Status of the protected areas of the Central African Republic Allard Blom a

a,b,*

, Jean Yamindou

c,z

, Herbert H.T. Prins

a

Department of Terrestrial Ecology and Nature Conservation, Wageningen University, Bornsesteeg 69, 6708 PD Wageningen, The Netherlands b Department of Anthropology, State University of New York at Stony Brook, Stony Brook, NY 11794, USA c World Wildlife Fund, B.P. 1053, Bangui, Central African Republic Received 25 May 2002; received in revised form 23 September 2003; accepted 26 September 2003

Abstract The Central African Republic (CAR) has made an impressive commitment to biodiversity conservation, with a total of 15 protected areas covering about 10.9% of the country. This study critically examines the status of these protected areas in light of their potential for long-term protection of biodiversity. First of all, the protected area system does not protect a representative sample of the ecoregions of the CAR. Even more important is the fact that only 32% of the protected areas are adequately managed. Law-enforcement is inadequate, and poaching poses the largest threat to biodiversity conservation. Given the dim economic prospects, the CAR will need continued financial as well as technical assistance to deal with protected area management. Experiences have shown that such intervention can make a difference. However more attention needs to be given to capacity building, long-term funding mechanisms, regional collaboration and political commitment to make such interventions sustainable in the long-term. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Central African Republic; Protected areas; Ecoregions

1. Introduction The Central African Republic (CAR), landlocked and without mountains, is still one of the most diverse countries on the African continent because of its climatic range from the desert north to the rainforest in the south. With a land area of about 623,000 km2 and a population of about 3 million, the country has a low population density (World Bank, 2000). The CAR is among the least developed countries in the world. It ranked 149 among 174 countries on the Indicator of Human Development (UNDP, 1995). The economic situation has clearly deteriorated in recent years due to severe political disturbances combined with violence and looting. The altitude varies from 325 m in the Oubangui valley to 1410 m at Mount Ngaoui, on the border with Cameroon (Boulvert, 1983). The country overall shows little

* Corresponding author. Present address: ECOFAC, B.P. 15115, Libreville, Gabon. z Deceased. E-mail address: [email protected] (A. Blom).

0006-3207/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2003.09.023

relief and is almost entirely situated below 1000 m. Two thirds of the country form part of the Congo Basin drainage system, with the remainder largely draining into the Chadian system (Boulvert, 1983). The CAR has four main climatic zones (Boulvert, 1983; Carroll, 1997), with which the vegetation zones are strongly correlated (Boulvert, 1986). From north to south, the following can be distinguished (adapted from Boulvert, 1986 and Carroll, 1997) (Fig. 1): 1. Sahelian savanna: open savannas, 2. Sudanian savanna: vast grasslands with small groups of trees, 3. Congolian forest-savanna mosaic: wooded savanna and dry deciduous forest, 4. Congolian dense forest: from deciduous forest through transition forest to evergreen forest. These vegetation zones correspond fairly well with the ecoregions described by Underwood et al. (1998). These authors distinguish from north to south the Sahelian Acacia Savanna, the East Sudanian Savanna and the Northern Congolian Forest-Savanna Mosaic. The dense forest zone they split in two: the

480

A. Blom et al. / Biological Conservation 118 (2004) 479–487

Fig. 1. Protected areas of the Central African Republic with vegetation zones.

Northwestern Congolian Lowland Forests and the Northeastern Congolian Lowland Forests. This paper provides an updated database of existing protected areas of the CAR, as the present databases of the World Conservation Monitoring Center (WCMC), United Nations Environmental Program (UNEP) and the Conservation Union (IUCN) are incomplete or nonexistent (e.g. WCMC, 1994). It also provides a baseline against which future conservation initiatives can be measured, and provides an outline of a methodology for assessing the status of protected areas as well as their potential to provide biodiversity protection in the longterm that could be expanded to other countries. 2. Methods This study used existing data from the CAR Government as well as international organizations. These Table 1 Central African Republic categories of protected areas and their equivalent IUCN categories CAR category

Description

IUCN equivalent

RI PN PP RF RS RB

Reserve Integrale Parc National Parc Presidentiel Reserve de Faune Reserve Speciale Reserve de Biosphere

1a II IV IV VI VI

data were, whenever possible, corroborated and combined with our own experiences, government sources and interviews with people familiar with the different regions of the country. We used the definition of a protected area as adopted by IUCN (1994): An area of land and/or sea especially dedicated to the protection and maintenance of biological diversity, and of natural and associated cultural resources, and managed through legal or other effective means.

In the case of the CAR, this describes an ideal rather than an actual situation, because many of the protected areas are certainly not managed effectively. In Table 1 we give the corresponding CAR categories for the equivalent IUCN categories that are in use to define protected areas. The CAR has made an important effort in the field of conservation and has over the years gazetted 15 protected areas (Fig. 1, Table 2, and Table 6). All of these still exist on paper, although some have undergone name changes, upgrades or extensions. For an historical overview we refer to IUCN (1992). The total area currently protected is 68,185 km2 , which represents 10.9% of the national territory. UNESCO designated the Basse–Lobaye Forest and the Bamingui–Bangoran Conservation Area as Man and Biosphere (MAB) Reserves, the former in 1977 and the latter in 1979 (UNESCO, 2000). Both designations include areas outside the actual protected area, following

A. Blom et al. / Biological Conservation 118 (2004) 479–487

the MAB model. The Manovo–Gounda–St. Floris National Park was inscribed on the list of World Heritage Sites in 1988 (World Heritage Committee, 1988; UNESCO, 1999) and was added to the list of World Heritage in Danger in 1997 (World Heritage Committee, 1998) following reports of illegal grazing and poaching. In practice these designations have contributed little to the actual functioning of these areas.

2.1. Assessment factors In order to obtain an indication of the conservation potential of protected areas in CAR, we used several factors in an overall assessment of their present status. We assessed conservation potential as the overall potential for conservation of biodiversity in the mid (10 years) to long-term (50 years). All the assessments were based on our own extensive experience in CAR, and were verified by Government staff involved in protected area management (Doungoube, pers. comm.). [It should be noted here that the responsibility for protected areas in CAR was not consistently the responsibility of the same ministry and sometimes even shared between different ministries though, most of the time, the same government personnel remained involved.] The data were comparable with information obtained during informal interviews with people familiar with the areas. We also visited each of the protected areas at least once in the last 5 years, with the exception of Nana–Barya, Zemongo, Andre–Felix and Yata–Ngaya. The following

481

4 factors were assessed and then used to quantify the overall conservation potential. 2.1.1. Threat Seven factors – logging, mining, hunting, grazing, farming, villages and roads – were assessed and given scores from 1 to 4, where 1 represents an area with few threatening or disturbing activities present, and 4 indicates an area being overrun by activities incompatible with protected area management. These were then combined to quantify the overall threat to conservation (Table 3). In the case of logging activities, we defined the type of logging as traditional (logs sawn in the forest), or commercial (logs transported to mills or exported). We also considered the way the logging was regulated (certified, supervised or unregulated) in assessing the overall ranking. Human presence in the form of permanent settlements is clearly a threat, but active involvement of the local population in the conservation management of the area can diminish this threat. This mitigating aspect is assessed below under management. Before averaging, the ranking of hunting, which is by far the most threatening activity, was multiplied by two to take account of its importance. We took into consideration the stated objective of the protected area. If, for example, sustainable logging is allowed in a reserve and such logging was indeed carried out in a sustainable way, then this would not be noted as a threatening or disturbing activity. Also, the size of the disturbance or threat was taken into consideration, as a localized disturbance, even a seri-

Table 2 Protected areas of the Central African Republic Name

IUCNa

CARb

Year gazetted

Ecoregionc

Location

Vassako–Bolo Andre Felix Bamingui–Bangoran Dzanga–Ndoki (Ndoki sector) Dzanga-Ndoki (Dzanga sector) Manovo–Gounda–St. Floris Avakaba Aouk–Aoukale Gribingui–Bamingui Koukourou–Bamingui Nana–Barya Ouandjia–Vakaga Yata–Ngaya Zemongo Dzanga–Sangha Basse Lobaye

Ia II II II II II IV IV IV IV IV IV IV IV VI VI

RI PN PN PN PN PN PP RF RF RF RF RF RF RF RS RB

1960 1960 1933 1990 1990 1933 1980 1939 1940 1940 1960 1925 1960 1925 1990 1977

41 41 41 14 14 41 41 39/41 41 41 41 41 41 44 14 14

08°060 N; 09°290 N; 08°350 N; 02°270 N; 02°570 N; 09°310 N; 08°400 N; 09°520 N; 07°410 N; 07°240 N; 07°400 N; 09°020 N; 09°150 N; 06°540 N; 02°530 N; 03°400 N;

Total

Size (in km2 ) 19°470 E 23°180 E 19°430 E 16°150 E 16°220 E 21°210 E 20°400 E 21°250 E 19°170 E 19°570 E 17°290 E 22°180 E 23°250 E 26°040 E 16°130 E 17°500 E

860 1700 10700 725 495 17400 2500 3300 4500 1100 2300 4800 4200 10100 3359 146 68185

Updated and corrected from IUCN, 1998 with: a IUCN: IUCN categories of protected areas (IUCN, 1994). b CAR: CAR categories of protected areas as in Table 1. c Ecoregions, with 14 ¼ Northwestern Congolian Lowland Forest; 39 ¼ Sahelian Acacia Savanna; 41 ¼ East Sudanian Savanna; 44 ¼ Northern Congolian Forest-Savanna Mosaic (Underwood et al., 1998).

482

A. Blom et al. / Biological Conservation 118 (2004) 479–487

Table 3 Scoring of factors affecting threat of protected areas in the Central African Republic Code

Factor

Scoring

Log

Presence/absence of logging activities

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

See methods

Mine

Presence/absence of mining activities

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No mining Affecting <5% of area 5–10% >10%

Hunt

Presence/absence of hunting in area

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No hunting Affecting <5% of area 5–10% >10%

Graz

Presence/absence of grazing in area

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No grazing Affecting <5% of area 5–10% >10%

Farm

Presence/absence of farming activities

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No farming Affecting <5% of area 5–10% >10%

Vil

Presence/absence of villages

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No villages Affecting <5% of area 5–10% >10%

Road

Presence/absence of roads in area

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

No roads or paths Only foot paths Logging/secondary roads Commercial/main roads

Threat

Overall threat assessment

1 2 3 4

¼ ¼ ¼ ¼

None Low Moderate High

ous one, can be negligible in relation to the overall size of the protected area. In the CAR this is often the case with agriculture. 2.1.2. Biodiversity significance Biodiversity significance was assessed by examining the representation of each ecoregion within the protected areas network in the CAR (Table 4). Some areas might be significant at the national level (e.g. forest), and not on the larger regional scale. However, this assessment focuses on a national protected area system and it was considered that representation at the national level should take priority. Protected areas that were part of a larger complex were all given the same ranking as if they were one. 2.1.3. Integrity Integrity is a snapshot assessment of the present status of the protected area taking into consider-

Description of value

ation destruction, degradation and fragmentation (Table 4). 2.1.4. Management The present status of the management of the protected area was assessed taking into consideration presence or absence of law enforcement staff, level of financial support and level of local (village) participation (Table 4). These three factors were all considered equally important and were averaged to give an overall assessment of the present management, with 1 indicating that management is excellent and 4 indicating no management at present. 2.2. Overall assessment of the conservation potential of protected areas We averaged the scores for biodiversity threat, significance, integrity and management to reach an overall

A. Blom et al. / Biological Conservation 118 (2004) 479–487

483

Table 4 Scoring of factors affecting significance, integrity and management of protected areas in the Central African Republic Code

Factor

Scoring

Sign

Biodiversity significance

1 2 3 4

Int

Assessment of the present status

1 ¼ High

¼ ¼ ¼ ¼

Description of value

High Moderate Low None

Covering >5% of the vegetation zone 2–5% 0.1–2% <0.1% <5% of area is destroyed or severally degraded and/or <10% is moderately degraded or fragmented <5% destroyed and/or <25% fragmented <25% destroyed and/or <50% fragmented >25% destroyed and/or >50% fragmented

2 ¼ Moderate 3 ¼ Low 4 ¼ Very low Guard

Presence/absence of guards

1 2 3 4

¼ ¼ ¼ ¼

Well Staffed Moderately Staffed Inadequately staffed No staff

>0.03 guard/km2 0.01–0.03 <0.01 No guards

Supp

Level of external support

1 2 3 4

¼ ¼ ¼ ¼

High Moderate Low None

Well funded Moderately funded Inadequately funded No funding

Part

Level of local (village) participation

1 2 3 4

¼ ¼ ¼ ¼

High Moderate Low None

Local people involved in all aspects Involved in some activities and consulted on others No active participation, some consultation No involvement

Manage

Assessment of the present management

1 2 3 4

¼ ¼ ¼ ¼

High Moderate Low None

See methods

Pot

Conservation potential

1 2 3 4

¼ ¼ ¼ ¼

High Moderate Low None

See methods

assessment of the conservation potential of the protected areas, i.e. the potential to protect biodiversity in the long-term. The lower the final figure, the more conser-

vation potential an area is expected to have. This result was used to rank the protected areas in priorities and opportunities for conservation in the CAR.

Table 5 Mean values of the factors used in the assessment of the conservation potential of the protected areas of the Central African Republic, ranked by their potential (with 1 ¼ highest, 4 ¼ lowest) Name

Threat

Significance

Integrity

Management

Potential

Dzanga–Ndoki (Ndoki sector) Dzanga–Ndoki (Dzanga sector) Dzanga–Sangha Manovo–Gounda–St. Floris Aouk–Aoukale Bamingui–Bangoran Vassako–Bolo Ouandjia–Vakaga Gribingui–Bamingui Koukourou–Bamingui Nana–Barya Avakaba Zemongo Andre Felix Basse Lobaye Yata–Ngaya

1.5 1.8 2.4 2.6 2.3 2.3 2.1 2.3 2.3 2.4 1.9 2.3 2.3 2.3 2.6 2.3

1 1 1 1 2 2 3 3 3 3 3 3 2 3 4 3

1 1 2 3 3 3 3 3 3 3 3 3 4 4 3 4

2 1.7 2 2 2.7 2.7 2.7 2.7 2.7 2.7 4 3.7 4 3.7 3.7 4

1.4 1.4 1.8 2.1 2.5 2.5 2.7 2.7 2.7 2.7 3.0 3.0 3.1 3.2 3.3 3.3

Average

2.2

2.4

2.9

2.9

2.6

2.6 2.9 2.4 2.9 3.1 2.4 3.1 2.2 3.1 1.3 1.1 3 3 1.2 1.2

see Table 3 for description of these field attributes, with Log ¼ logging; Mine ¼ mining; Hunt ¼ hunting; Graz ¼ grazing; Farm ¼ farming; Vil ¼ Village; Supp ¼ external support; Part ¼ local participation; Sign ¼ significance; Int ¼ integrity; Pot ¼ potential.

Average

a

Pot. Int

3 4 3 1 1 3 3 3 3 3 3 3 4 4 2 3 3 3 2 1 1 1 3 2 3 3 3 3 3 2 1 4

Sign Manage

2.7 3.7 2.7 2 1.7 2 3.7 2.7 2.7 2.7 4 2.7 4 4 2 3.7 3 4 3 2 2 3 4 3 3 3 4 3 4 4 2 3

Part Supp

2 4 2 1 1 1 4 2 2 2 4 2 4 4 1 4 3 3 3 3 2 2 3 3 3 3 4 3 4 4 3 4

Guard Threat

2.1 2.3 2.3 1.5 1.8 2.4 2.1 2.1 2.1 2.3 1.9 2.3 2.3 2.3 2.4 2.6 3 3 3 2 3 4 3 3 3 3 2 4 3 4 4 2

Road Vil

1 1 2 1 1 2 1 1 1 2 1 2 1 1 2 3 1 1 1 2 1 1 1 1 1 1 1 1 1 1 2 3

Farm Graz Hunt

1 1 1 1 2 1 1 1 1 1 1 1 1 1 3 3 Vassako–Bolo Andre Felix Bamingui–Bangoran Dzanga–Ndoki (Ndoki sector) Dzanga–Ndoki (Dzanga sector) Manovo–Gounda–St. Floris Avakaba Aouk–Aoukale Gribingui–Bamingui Koukourou–Bamingui Nana–Barya Ouandjia–Vakaga Yata–Ngaya Zemongo Dzanga–Sangha Basse Lobaye

Mine Loga

The 1997 United Nations list of protected areas in the CAR mentions 14 protected areas, with a total area of 54,456 km2 (IUCN, 1998) although one protected area has been omitted, namely the Basse–Lobaye Reserve of 146 km2 . However, this UN list does not provide any further information other than category, location, size and year of establishment. The present paper, then, provides more detailed information on the protected area network in the CAR, including an assessment of the potential of each individual area in comparison with the other areas in the national protected area network. Our assessments and weightings given to each factor and category had to be somewhat subjective as very little concrete information is available but, with our extensive experience in CAR, we feel confident that the end results do present a reasonable assessment of the potential of each individual area in comparison with the other areas in the national system. The raw scores presented in Table 6 allow other workers to re-assess our estimates using different weightings, if so desired. In any event our data provide a baseline for monitoring and evaluating the effectiveness of management. With well over 10% of the country designated as protected areas, the CAR has made an exceptional commitment to biodiversity conservation. Given the fact that not all the areas pre-date independence, this clearly demonstrates a continuing commitment in economic terms. Even though many of these areas are on marginal

Table 6 Individual scores for all variables assessed for the protected areas of the Central African Republic

4. Discussion

4 3 4 1 1 4 4 4 4 4 3 3 3 2 1 1

The factors used to calculate the conservation potential of each protected area in the CAR are summarized in Table 5. The resulting final column assesses the overall potential for conservation of biodiversity in the long-term (50 years). The two Dzanga–Ndoki sectors have much the highest potential mainly because the biodiversity significance and integrity are high and there is relatively modest threat, while Yata–Ngaya has very low potential because of its low significance and minimum integrity and management. A full breakdown of the details is provided in Table 6. The average management assessment is positively correlated with both the relative significance and the integrity assessment (Pearson correlation two-tailed, respectively 0.838 and 0.768, n ¼ 16, P < 0:01), but not with the average threat assessment (Pearson correlation two-tailed, 0.240, n ¼ 16, NS).

3 4 3 2 2 3 3 3 3 3 3 3 4 4 2 4

3. Results

1 1 1 1 2 1 1 1 1 1 1 1 1 1 3 1

We used Pearson correlation coefficient of significance to examine if certain assessment variables co-varied or were interdependent (SPSS, 1997).

2.7 3.2 2.5 1.4 1.4 2.1 3.0 2.5 2.7 2.7 3.0 2.7 3.3 3.1 1.8 3.3

A. Blom et al. / Biological Conservation 118 (2004) 479–487

Name

484

A. Blom et al. / Biological Conservation 118 (2004) 479–487

lands, some, like the Dzanga–Ndoki National Park, can generate important revenue from logging and/or safari hunting. For example a cost–benefit analysis of a similar conservation project in Korup National Park in Cameroon suggests that about 1000 euro per km2 per year would be required to compensate Cameroon for the area protected in the park (Ruitenbeek, 1992). 4.1. Ecoregion representation Most of the human population of the CAR is concentrated in the cities in the south and west of the country, leaving the north and east almost completely barren of human settlements. This fact, combined with the extensive system of protected areas, would seem to provide an almost ideal situation for effective biodiversity conservation. However, the reality is different. First of all the protected area system in the CAR does not include a representative sample of the vegetation zones or ecoregions of the CAR. Both the Sahelian Acacia Savanna (except for a part of the Aouk–Aoukale Reserve) and Northeastern Congolian Lowland Forest are not represented at all in the protected area system of the CAR. Even though it could be argued that national representation is not that important for biodiversity protection overall, protected areas in nationally underrepresented vegetation zones such as Dzanga–Sangha, which is part of the much larger Congo basin forest, are considered important even on a regional scale (Blom et al., 2001). 4.2. Management and law enforcement More importantly, the management of the protected area system is generally inadequate. In fact three protected areas (Zemongo, Yata–Ngaya and Nana–Barya), representing 24% of the area under protection by law, have no management at all. The isolation of these areas has not provided protection against severe poaching pressure (with the possible exception of Nana–Barya, an area that needs to be surveyed urgently). Three additional protected areas (Andre–Felix, Avakaba and Basse–Lobaye) or 6% of the total land area protected, have only superficial management, and six other areas (37% of area under protection) are managed inadequately. This leaves only the Dzanga–Sangha protected area complex and the Manovo–Gounda–St. Floris National Park. Even in these moderately well managed areas problems remain. Owing to the lack of management and particularly law-enforcement, poachers have overrun most of the protected areas in the CAR (e.g. Moussa, 1992). In the north and east the problem is especially serious, resulting in the death of four park rangers in 1997 in Manovo–Gounda–St. Floris and disruption of tourism to this park (WCMC, 1997). Here the park rangers con-

485

front local poachers as well as regular infiltration from Sudan and Chad. The main impetus for placing the Manovo–Gounda–St. Floris on the list of World Heritage in Danger list was that poaching had decimated >80% of the ParkÕs wildlife populations (UNESCO, 1999). These poachers are well organized and equipped, and concentrate on bushmeat as well as ivory. In the past they also collected rhinoceros horn, but they have long since eliminated the rhinoceros in CAR (Mokombo, in Oglethorpe and Ham, 1999). We were unable to visit the three eastern most areas (Zemongo, Andre–Felix and Yata–Ngaya), and in recent years few others have, as this area has become dangerous for travel. However a recent river rafting expedition on the Chinko river, close to Zemongo, indicated that little wildlife was left in that area. Gunther Klemm (pers. com.) of the GTZ Dzanga–Sangha Project, who visited the area in 1997 as well as the safari hunters (Haut– Chinko Safaris, pers. com.), who had to abandon the region due to intense poaching, confirmed this situation. A World Wildlife Fund team surveyed the forests of Bangassou, south west of Zemongo, in 1995–1996. Even that far from the northeastern border with Sudan they found evidence of intense poaching by the Sudanese. Only an estimated 1600 elephants remain in this forest (Kpanou, J.-B., pers. comm.). The Bangassou forest is at present the subject of a new conservation initiative, which may lead to the establishment of CARÕs first protected area in the Northeastern Congolian Lowland Forest ecoregion. Overall, poaching is by far the most important threat in the short and medium term to the survival of the entire protected area system in the CAR. In the south the situation is somewhat better, but even here poachers armed with automatic rifles, as up north, have been reported (Blom, 1999). Illegal grazing is also a major problem in some of the areas in the north (e.g. WCMC, 1997). 4.3. Paper parks Only three areas, which together make up the Dzanga–Sangha protected area complex, seem to have escaped major degradation and fragmentation. We consider that any area with an overall score of 3 or more on the potential scale would probably be difficult or impossible to restore in the present situation. In effect, we consider such areas ‘‘paper parks’’, especially because none of these areas receive any serious effort in management. One national park (Andre–Felix) and five reserves fall in this category, representing 31% of the area officially under protection. Thus almost one third of the protected areas of the CAR are so called ‘‘paper parks’’, with little or no management. If no action is taken in the near future it is likely that we can add to this list another national park and five more reserves, which each have scores of 2.5 or more.

486

A. Blom et al. / Biological Conservation 118 (2004) 479–487

That would leave only two national parks (Manovo– Gounda–St. Floris and Dzanga–Ndoki) and the Dzanga–Sangha Special Reserve, representing only 32% of the present protected area system, relatively intact. These areas are also being degraded at present, but at a reduced rate. 4.4. Initiatives and potential solutions First and foremost, the responsibility for finding solutions lies with the CAR government and specifically with the ministry responsible for protected areas. Although the government has made some remarkable commitments to conservation, these have clearly been insufficient. Faced with the survival of the protected area system, the government has to make some hard choices. We propose that all those areas with a score of >2.5 on the potential scale are surveyed to confirm that our assessments are accurate, and if so that these areas are simply declassified. This would allow the government and its partners to concentrate their limited resources on the four areas that have a reasonable longterm conservation potential. Fortunately, our analyses already show that there is a positive correlation between the average management assessment and the relative significance of the protected area, indicating that the ministry responsible for protected areas and its partners have made the most of their limited resources by concentrating management efforts in the most significant areas. These areas also show the highest integrity in our snapshot assessment. This could indicate that either the management has been effective and does make a difference or that effort has gone into less damaged areas to start with. Some fairly recent initiatives by Northern Region Development Project (PDRN) and the Dzanga–Sangha project (e.g. Blom, 1999) confirm that management has been effective, because in contrast to surrounding areas, the areas that are managed, staffed and funded are indeed largely capable of maintaining wildlife populations at least in the short to medium term. For example most of the areas outside of management control have seen a complete collapse of elephant populations (Tello, pers. comm.; pers. obs.). Law enforcement at all levels is a prerequisite for the survival of the protected area system in the CAR. The experiences with PDRN and Dzanga–Sangha also showed that when villagers are involved in revenue sharing there are opportunities to develop collaboration and reduce poaching. For example, some members of the local development committee in Bayanga, the committee that receives some of the parkÕs revenue, carried out a citizenÕs arrest of an elephant poacher. Revenue sharing initiatives should be extended to cover all the protected areas in CAR. Recent meetings between the wildlife authorities of Sudan, Chad and CAR to address the serious issues of

cross-border poaching and regional stability seem to provide some optimism for reducing the high poaching pressure in northern CAR (Aveling, pers. comm.). A similar cross-border regional collaboration was previously implemented successfully in southern CAR, where a tri-national Sangha park agreement (of which the Dzanga–Sangha complex is part) was signed in 1998. This agreement allowed for example for joint patrols between Cameroonian, Congolese and Centralafrican rangers and exchange of intelligence, resulting in some successful anti-poaching missions. These initiatives address some of the underlying issues driving the degradation of the protected area system, and thus are promoted by the donor community. For example the European Union through its ECOFAC program sponsored the above-mentioned meetings between CAR, Chad and Sudan. These meetings and agreements do not only benefit conservation, but also contribute to regional stability and cooperation. Given the meagre financial resources of the country and its dim economic prospects, it is clear that the CAR will still need both financial as well as technical assistance to improve protected area management, even if it declassifies those areas that are no longer viable. Experiences with conservation projects such as PDRN, ECOFAC–Ngotto and Dzanga–Sangha (e.g. Blom, 1999) have shown that such intervention can make a difference in the short to medium-term. However, as these programmes depend largely on donor funding and foreign technical assistance they are at present not longterm solutions. More attention needs to be given to capacity building at the national level. Recent initiatives such as creation of a trust fund for Dzanga–Sangha (e.g. Blom, 1996) will hopefully start addressing the longterm financing of the remaining protected areas in the CAR.

Acknowledgements We would like to thank the World Wildlife Fund, Inc. and the different government agencies in the CAR for supporting this research. We are particularly grateful to Gustave Doungoube, Mike Harrison, John Oates, Natasha Shah, Caroline Tutin, David Wilkie and an anonymous reviewer for constructive discussions and comments.

References Boulvert, Y., 1983. Carte pedologique de la Republique Centrafricaine, a 1:1,000,000. Notice explicative #100. ORSTOM, Paris, p. 126. Boulvert, Y., 1986. Republique Centrafricaine. Carte phytogeographique a 1:1,000,000. Notice explicative # 104. ORSTOM, Paris, p. 131.

A. Blom et al. / Biological Conservation 118 (2004) 479–487 Blom, A., 1996. Proposal for the creation of an environment trust fund for Dzanga–Sangha. World Wildlife Fund, Bangui, CAR. Blom, A., 1999. Ten years Dzanga–Sangha Project: 1988–1999. World Wildlife Fund, Bangui, CAR. Blom, A., Kamdem Toham, A., DÕAmico, J., OÕHara, D., Abell, R., Olson, D., 2001. Assessment of biological priorities for conservation in the Guinean–Congolian Forest Region. World Wildlife Fund, Washington, DC. Carroll, R.W., 1997. Feeding ecology of the western lowland gorilla (Gorilla gorilla gorilla) in the Central African Republic. Ph.D. Dissertation, Yale University, New Haven, CT. IUCN, 1992. Protected areas of world: a review of national systems. Volume 3: Afrotropical. Prepared by the World Conservation Monitoring Center (WCMC). IUCN, Gland, Switzerland and Cambridge, UK. xxii+360 pp. IUCN, 1994. Guidelines for protected areas management categories. IUCN, Cambridge, UK and Gland, Switzerland. IUCN, 1998. 1997 United Nations list of protected areas. IUCN, Cambridge, UK and Gland, Switzerland. Moussa, M., 1992. Le grand bracconage en Centrafrique. In: Cleaver, K., Munasinghe, M., Dysan, M., Egli, N., Peuker, A., Wencelius, F. (Eds.), Conservation of West and Central African rainforests. World Bank, Washington, DC, pp. 182–188. Oglethorpe, J., Ham, R., 1999. Armed conflict and protected areas in Africa. Available from . Ruitenbeek, H.J., 1992. Rainforest supply price: a tool for evaluating rainforest conservation expenditures. Ecological Economics 6, 57–78.

487

SPSS, 1997. SPSS Base 8.0, SPSS, Inc. Chicago. Underwood, E., Itoua, I., Olson, D., Dinerstein, E., Loucks, C., Wettengel, W., 1998. Terrestrial ecoregions of Africa. WWF-US Conservation Science Program, WWF-US, Washington. UNDP, 1995. Rapport Mondial sur le Developpement Humain. Economica, Paris. X+251 pp. UNESCO, 1999. World Heritage list in danger. Available from (site last updated: 30 November 1999). UNESCO, 2000. UNESCO MAB Biosphere Reserve Directory. Available from (site last updated: 14/02/2000). WCMC, 1994 Biodiversity data sourcebook. World Conservation Press, Cambridge, UK. WCMC, 1997. Description of Natural World Heritage Properties. Available from (Last updated: August 1997, Access date: 9 March 2000). World Bank, 2000. World Development Indicators. The International Bank for Reconstruction and Development/The World Bank, Washington, DC, USA. p. 389. World Heritage Committee, UNESCO, 1988. Report of the World Heritage Committee. Convention concerning the protection of the world cultural and natural heritage. Twelfth Session, Brasilia, 5–9 December 1988. World Heritage Committee, UNESCO, 1998. Report of the World Heritage Committee. Convention concerning the protection of the world cultural and natural heritage. Twenty-first Session, Naples, Italy, 1–6 December 1997.