Many countries bordering the African Great Lakes are involved in the current conflict, particularly the DRC, Rwanda, Burundi, and Uganda. Although many more countries are actually implicated, notably the Sudan, Zimbabwe, Tanzania, Angola, and Namibia, the immediate development of the war had most to do with the Great Lakes states.
Historical setting: refugees and internally displaced persons (IDPs)
The Congo has long been a destination for refugees. During the colonial era, the Congo borders allowed through peasants in flight from famine (as was the case for refugees from Rwanda in 1905, 1928, 1940–43 and 1950–53) and those fleeing ethnic conflicts in Rwanda, Burundi, Sudan, and Angola (Matthieu and Tshongo 1999; Ndaywel è Nziem 1998). Thousands of others were encouraged to migrate by the Belgian government, which was in need of laborers for its colonial plantations of coffee, tea, and quinine, as well as in its mines and on its roads (Matthieu and Tshongo 1999). Since the Belgian Congo's independence a central reason for mass movements into the Congo has been armed conflict resulting from ethnic tensions (as in Rwanda and Burundi), religion (Sudan), and politics (Angola and Uganda).
Along with refugees migrating over national borders, DRC has also experienced successive internal movements of internally displaced persons, IDPs. At the time of independence, there was a massive migration of Luba people out of the city of Kananga in Kasai Occidental (Fig. 1), following an abortive secessionist movement (Legum 1961). There were numerous groups of displaced people from the muleliste movement, a rebellion started by one of Patrice Lumumba's ministers following Lumumba's assassination. By 1964, the mulelistes controlled the eastern part of the current DRC, an area similar in extent to the current rebel-held territory. Then, as now, local militias were important. During the rebellion of the mulelistes or simbas, IDPs fled the northwestern rim of Lake Tanganyika toward Maniema and south Kivu (Ndaywel è Nziem 1998); throughout the forest zone of what is now Province Orientale, entire villages disbanded, and their inhabitants made temporary gardens far from the road and could only return when peace was restored in the latter half of the 1960s.
The immediate buildup to the current conflict
The long period of relative peace that Congo experienced under Mobutu can perhaps be attributed to a national reaction to the bloody muleliste rebellion that claimed about 500,000 victims (Ndaywel è Nziem 1998). By the early 1990s, however, with a failed national economy and numerous local ethnic conflicts, tension began to mount. At first, these were expressed through peaceful dialogue, optimistically called the Conférence Nationale, as a process of "democratic transition". The dialogues lasted several years but made little progress.
Under a slogan of regionalism, ethnic tension started to mount again in the early 1990s, causing thousands of Luba people to be expelled from Katanga and forced to the Kasais. Likewise, south of Bunia, bloody local wars broke out between the Hima and the Geti. But local bloodshed did not escalate to a national scale until after the Rwandan genocide and subsequent migration of Rwandan refugees into the DRC.
Preceding the Rwandan genocide, troops primarily from the Rwandan refugee population in Uganda (largely Tutsi), started a movement in 1990 to regain power in Kigali, the capital of Rwanda. They had Ugandan support as they had assisted Ugandan President Museveni to power in Kampala in 1986. The assassination of Rwandan president Juvenal Habyarimana set off a cycle of violence that resulted in the death of 500,000 to 1 million Rwandans (mainly Tutsi and moderate Hutu). Following a military operation, Operation Turquoise, carried out by the French between June and August 1994, 2 million Rwandan refugees crossed into DRC.
While the Rwandans were trying to reassemble their country following the trauma of genocide, Zaïre, in its far western capital, was in the fourth year of its long "democratic transition". In apparent acquiescence to both internal and international pressure, Mobutu had declared the end of his single-party rule in April 1990 and started a "transition", which was to lead to the "third republic" (Kabamba and Lanotte 1999). This transition failed, largely because of a blocking of the political process, the doubling of political institutions, and the rampant multiplication of personalized political parties. It was into this political stalemate that the massive migration of Rwandan refugees took place in 1994. Just as the political process lost credibility with the national population, Mobutu managed to re-enter the international political scene. According to Kabamba and Lanotte (1999), the transfer of Rwandan populations into the eastern province of Kivu also marked the transfer of the Rwandan war into then-Zaïre.
The Rwandan refugees flooded into a country nearing complete economic collapse. Zaïre's economic decline had taken a vertiginous slide with Mobutu's program of "Zaïrianization" in the mid-1970s. Soaring fuel costs and a collapse in the price of copper compounded the negative blow from this nationalization of foreign investment. According to some estimates, Zaïre's national economic growth plummeted from 7 percent at the end of the 1960s to 2 percent at the end of the 1970s (Tshombe 1999). High level corruption and an obsession with personal empire rather than social and economic development led to national asphyxia by the mid-1990s: infrastructures crumbled, national inflation reached four digits, and what had been incipient industry was paralyzed.
It was into this national context that the massive numbers of Rwandan peasants, ex-Rwandan military, and anti-Tutsi guerrillas poured across the Rwanda/Zaïre border and established refugee camps in North and South Kivu. Tension between Hutus, Tutsis, local populations and Mobutu's forces multiplied throughout the Kivu countryside. In 1996, the ethnic violence reached the Itombwe Mountains west of Lake Tanganyika where there is a significant population of Tutsi known as "Banyamulenge". Some Banyamulenge had received military training in Rwanda since 1994; they were joined in Zaïre by remnant guerrilla groups (from the muleliste rebellion) and locally formed militias. In late 1996, this coalition swept through the Zaïrian refugee camps on the Rwandan border dislodging massive numbers of Rwandan peasants along with Interahamwe and ex-FAR. Close to 1 million Hutu refugees fled "home" to Rwanda, but still tens of thousands of other refugees fled west into the vast Congolese forests.
Less than a year later, with support from Rwanda, Uganda, and Angola, the disparate rebels that had emptied the refugee camps, a group known as the new Alliance of Democratic Forces for the Liberation of Congo-Zaïre (ADFL), had marched across the country. They overthrew Mobutu and established a new regime.
This is not, however, an analysis of the impact of a one-year war.
The protected areas and the forests are not just recovering from the fleeing and pillaging troops of Mobutu and the subsequent war-like occupation of the advancing ADFL forces and their allies. Although Laurent Kabila took power in 1997, although governors were assigned and incipient administrations assembled, and although Kabila clearly stated that the protected areas of Mobutu's era were still acknowledged and protected during his regime, Kabila himself was in power for just one year before the country was resubmerged in civil war.
The ethnic tensions between Hutus and Tutsis and between people of Rwandan extraction and traditional Bantu had been exacerbated by the presence of Rwandan refugees in the provinces of North and South Kivu, but over the following years, ethnic strife continued to erupt in local massacres and cause large internal population movements. This has included both strong anti-Rwanda sentiments in South Kivu and strife between Hima and Lendu peoples in the north. The latter, alone, resulting in more than 100,000 displaced and 7,000 deaths from 1999 to 2001.
Neither Laurent Kabila, nor his son and successor, Joseph Kabila, has been recognized as the legitimate leader of the DRC in the eastern part of the country (as of June 2001), an area that includes four out of five of DRC's World Heritage Sites (WHSs), as well as the MNP and several significant reserves (Fig. 1). Nor is the eastern DRC itself united. There were three clear eastern factions, although two of them under the "patronage" of Uganda, Rassemblement Congolais pour la Democratie–Mouvement de Libération (RCD-ML) or RCD-Kisangani resulting from the split of the RCD, and the Mouvement pour la Libération de Congo (MLC), united in January 2001 as the Front pour la Libération Congolaise (FLC). The third group Rassemblement Congolais pour la Démocratie (RCD-Goma) operates under Rwandan patronage (Fig.1). The Rwandan and Ugandan backed administrations claim international recognition as the principal opposition forces in the DRC, those that have locked the country, as of the year 2001, into a multi-year stalemate (ICG report, 2000). There are also some significant rebel factions such as the Interahamwe and ex-FAR that operate within the Kivus and are aligned with Kinshasa. Other rebel factions (such as the Mayi-Mayi) are only weakly aligned with Kinshasa. They have strong local roots and may actually change sides depending on the immediate implications for traditional authority. But all factions contain elements principally concerned with personal profit. Furthermore, these feuding groups, whether they have an administrative "capital" or not, often operate from forest villages, and their troops' subsistence resources are frequently what is in or on the ground and often what is in the parks and reserves.
Impacts on Biodiversity
There is essentially no knowledge of the state of nature conservation, or the extent and conditions of exploitation of resources except where there is an observer who records and transmits the information. Furthermore, information on the diversity and distribution of plants and animals requires expertise as well as scientific sampling. In the Congo, even before the war, such information was lacking for many important areas. As a result, details of the war's many impacts are sketchy at best, and sometimes completely absent. The reporting is most thorough in areas, near refugee camps from 1994 to 1996, where the presence of the United Nations High Commission for Refugees (UNHCR), humanitarian organizations, and numerous news agencies maintained communication with the outside world. The role of the international conservation organizations and the DRC national parks agency (ICCN), is crucial, but these conservation NGOs are only present in certain parks, and the ICCN only has the means to communicate and even to operate where supported by one of these NGOs. Therefore, we report information on these impacts in the following order:
Early impact of the war on biodiversity
Alarm at the biodiversity destruction associated with the ongoing conflict in the DRC reached international levels in 1994, after the establishment of refugee camps on the Rwandan and Burundian borders. These refugee camps were associated with immediate and large-scale destruction. Other important changes that mobilized Congolese anger over the environmental consequences of the camps included the fact that environmental destruction in the DRC was now being carried out by Rwandans rather than Zaïrians, and that the international community was seen as a partner to the destruction.
Before the refugee crisis, conservation of biodiversity clearly had international priority and Zaïrian authorities were taking appropriate postures to gain support, but in 1994 the international priority was the welfare of refugees, all the more so because there had been almost no international response to the preceding genocide that had raged through Rwanda. A full-scale humanitarian response to the refugee crisis was a way of making amends that fit within the framework of accepted international response. Mobutu's government easily switched to humanitarian rhetoric and thereby gained international approval and recognition. As a result, environmental destruction was not considered, at least at first, as a priority problem during the Great Lakes crisis.
Wildlife destruction
During the refugee crisis, large-scale destruction occurred in and around the Virungas National Park, the oldest National Park in Africa (created in 1925) and certainly a crown jewel for Zaïre with the snow-capped Ruwenzori Mountains, active volcanoes, Lake Edouard, savanna, lowland and high-altitude forests, alpine habitats and an array of charismatic megafauna, including mountain gorilla (Gorilla beringei beringei), elephant (Loxodonta africana), and hippopotamus (Hippopotamus amphibius) (see José Kalpers, Biodiversity Support Program [BSP]). A total of 600,000 refugees were hustled into camps set up in and around the park (Biswas and Tortajada-Quiroz 1996), with predictably dire consequences for this natural World Heritage Site.
One of these large refugee camps, Mugunga, had a strictly military section that lodged deposed military of the old Rwandan regime, which was estimated at about 50,000 persons. These military refugees, whose camp was situated within the park proper, maintained their arms and ammunition and soon after their arrival started full-scale economic activities, which included intensive poaching (Biswas and Tortajada-Quiroz 1996). The poaching continued after the war of liberation when, during November and December of 1996, the Mayi-Mayi rebels decimated essentially all of the remaining hippos on the rivers Rutshuru and Rwindi (Mushenzi 1997, memorandum). The purportedly more disciplined army of liberation (ADFL) continued the slaughter, including that of antelope and buffalo. A traveler on the road from Butembo to Goma at the time would be offered by the ADFL military at the barrier where they entered Virunga park hippo meat ($50 an animal), antelope meat ($20 for a kob [a savanna antelope]) or other game (Mushenzi 1997, memorandum).
The most recent war of liberation (1998) has offered no relief for the wildlife. The greatest threat is still from "men in uniform", which now includes Ugandan (in the north) and Rwandan (in the south) troops. In 1999, in the sector of VNP controlled by RCD-Goma (Rwandan-aligned rebel group), 330 buffalo and 450 antelope are known to have been killed, and these numbers must be evaluated knowing that the park staff cannot access many areas for a thorough assessment (d'Huart and Hart 2000).
What is perhaps amazing is that despite losses, a large proportion of the mountain gorilla population continued to survive into the year 2000 (José Kalpers, BSP). Also, despite destruction of parts of their forest and despite human terrorism and loss in nearby villages, the habituated chimpanzees of Virunga National Park's Tongo forest also survived at the advent of this millennium (Annette Lanjouw, personal communication).
Deforestation
In the mid-1990s, the most visibly striking environmental consequence of the refugee camps was deforestation. More than 1 million refugees needed wood for cooking fires. Around Goma, much of this deforestation happened in the park, VNP, itself. The continuous collection of firewood over two years caused serious ecological transformation over vast areas of the VNP (José Kalpers, BSP). The deforestation also elicited more of a response from the international community than did the poaching, perhaps because it was felt that a constructive response through reforestation programs was possible and perhaps because the refugee camps which were sponsored by international humanitarian organizations were seen as directly responsible. In order to cook, fires were needed. For fires, fuel was needed. Although the high level of poaching was arguably a consequence of having so many people and so many guns on the border of the park, it was still not so obviously an inevitable consequence of the refugee camps. The humanitarian organizations did after all provide alternative protein.
Impact of continuing conflict on biodiversity in the better-protected parks
The impact of war on the natural environment as on the human populations has been chaotic. Although its environmental devastation has been unevenly distributed during the latter part of the 20th and early 21st century, all of the parks and reserves of the eastern DRC were eventually affected, and certainly, the unprotected forests are being severely looted and trampled. The greatest damage was caused by various armed groups and even unarmed groups during alternating periods of crises in different areas. For some of the protected areas critically important for their contribution to the country's biodiversity, there is still no way to evaluate what has been destroyed because of continued insecurity and difficulty of access, such as in the lowland areas of KBNP, MNP, and parts of VNP).
Garamba National Park (GNP)
Garamba National Park (4,900 km2), located in the northern DRC along its border with Sudan (Fig. 1), was one of the first protected areas affected by armed conflict, when 80,000 refugees fleeing the civil war in Sudan were set up in camps on the park's borders (starting in 1991). Arms and ammunition were abundant and military units (Sudanese People's Liberation Army, SPLA) were stationed adjacent to the park. As a result, meat poaching escalated, first in the northern sector of the park, but eventually moving toward the southern part of the park, which is the habitat of the only remaining wild population of northern white rhinoceros (K. Hillman Smith et al. 1998). During the first of DRC's "wars of liberation" (1996–97) the park and its headquarters were occupied by mercenaries working for Mobutu until they were ousted by the ADFL. All of the park's logistical equipment was looted, including fuel, radios, and vehicles. The guards were disarmed, and the park's law enforcement capacity was severely crippled. Two rhinos were poached in 1997 (K. Hillman Smith et al. 1998).
Following this period, there was a slow build-up in law enforcement capability. As the Park was partially re-equipped in 1998, staff was re-organized and paid, and motivation was high. (K. Hillman Smith et al. 1998; F. Smith and K. Hillman Smith 1998) During the second "war of liberation", guards were not disarmed and were able to continue antipoaching activities. Poaching has since decreased and pulled back from the critical southern sector (F. Smith and K. Hillman Smith 1999; Réunion de programmation 2000).
Okapi Faunal Reserve (OFR)
The Okapi Faunal Reserve (13,400 km2) is not on an international border and, being farther from the origins of the conflict, was among the most recently affected protected areas (Fig. 1). It was only in the year 2000 that elephant poaching swept through the entire reserve. Recently, elephant meat has been sold in the village of the park administrative center, and the park guards were essentially mocked by an imperious order from passing military to "guard" the ivory they had poached and were carrying with them. Four years after the first war, there is a significant increase in the number of illegal "coltan" (colombo-tantalite) and gold mines, which, in the year 2000, reached to within just a few kilometers of one of the reserve's study areas. The mining camps are foci for poaching as well as local deforestation and serious disturbance to watercourses. (personal communication, Coordination Committee of the OFR or CoCoSi, August 2000).
The identities of the poachers are diverse. Some are local people with their own recently acquired arms. More often, local people are helping military from outside of the Reserve (RCD-ML, Ugandan, and recently FLC soldiers) or ex-military (FAZ [Forces Armées Zairoises], ADFL). Military deserters have large camps to the southeast, southwest and west of the reserve, as well as within the reserve itself in the northeast sector. Active military, both Congolese and Ugandan, are involved in poaching from the military base of Mambasa just to the east of the reserve and coming from Bafwasende to the west. (CoCoSi report, August 2000).
In the months of October 2000 through January 2001, the guards of the OFR, with the aid of the Ugandan military, the Ugandan People's Defense Force (UPDF), brought elephant poaching under control in the central sector of the reserve. In February of 2001, coltan mining was reduced in the central part of the reserve, and the UPDF undertook a military training of the OFR guards. However, in April 2001, after the termination of the antipoaching campaign, elephant poaching recommenced.
Kahuzi-Biega National Park (KBNP)
Kahuzi-Biega National Park (6,000 km2), like VNP, is located on the DRC's eastern border (Fig. 1) and has suffered consistently since the period of Rwandan refugee camps (1994 to the present). Like VNP, the root causes for much of the damages in KBNP are associated with the social problems arising from the Rwandan war itself. A high population stresses resources within the Park. In certain sections of the park, internal displacements have sent population density rising to 300 persons per km2. In mid-2000, there were 35,000 displaced people around the park as a result of continuous raids and battles involving Mayi-Mayi, Interahamwe, and Rwandan military (Bishikwabo 2000b).
In 2000–01, the ICCN only controlled the smaller highland portion of KNP (covering approximately 10 percent). The entire low-altitude sector (about 90 percent) occupied by at least 2,500 people, was purportedly in the hands of the Interahamwe and Mayi-Mayi, who are involved, along with others, in mining gold, castorite, and "coltan". These ores, along with ivory, are exported from airstrips, five of which are just outside the borders of the low-altitude sector of the park. These airstrips have been unmonitored by the presence of any outside U.N. observers or news agencies. In the meantime, the belligerents include various groups (e.g., Mayi-Mayi and Interahamwe) who, although aligned with the major sides of the war, are not signatories to any peace agreements and continue looting the Park, while others negotiate (d'Huart and Hart 2000; Bishikwabo 2000a, 2000b).
The ability to rehabilitate KBNP is also greatly challenged by the fact that the forest corridor connecting the highland and lowland sections has been illegitimately acquired by an individual in a strong political position in the local government. The current RCD administration in Bukavu took steps to rescind these land claims, but the mixed team of park staff, local population, and military who set out in August 2000 to mark park boundaries was attacked by a "guerilla group". Ten members of the team were killed (Guy Debonnet, GTZ [German Development Aid], personal communication).
Virungas National Park (VNP)
Settlement within the park is also a major problem for Virungas National Park, particularly in the park's northern sector where the park staff has little local legitimacy. The ICCN is unable to control this part of VNP with respect to limiting park infiltration, which includes increasing populations in fishing villages around Lake Edouard, an influx of agriculturalists and agriculture, and an immigration of Hima pastoralists with more than 3,500 cattle. The ineffectiveness of the ICCN has several causes, all of which are related to the continuing state of armed anarchy in the region. The northern Virungas is cut off from the rest of the park, including the higher-level ICCN administrators in Goma, as it is in the "Ugandan territory" of the RCD-ML, whereas the other sectors are in the Rwandan RCD-Sud. As with KBNP and the southern VNP, a high level of insecurity predominates in the northern part of the VNP, in this case, caused by the Nalu and the ADF, which are guerrilla factions fighting against Museveni's government in Kampala. A critical factor underlying the general inability of the ICCN to make any stand in the northern VNP is the lack of a supporting conservation organization (such as the international conservation NGOs functioning in other parts of the VNP). Even before the war (pre-1996), most investment in the park was given to the southern or central sectors, and most of the ICCN staff assigned to northern VNP has received no remuneration (d'Huart and Hart 2000).
Salonga National Park (SNP)
In western DRC, farther from the original causes of conflict in the east, but more importantly for biodiversity, farther from outside observation and control, is Salonga National Park (36,560 km2), the largest of all DRC's Parks. On a map (Fig. 1) the SNP seems remote, and for a foreign visitor its path of entry is indeed long and difficult. Nevertheless, the SNP, with its navigable rivers, has been a poacher's paradise for more than two decades. A recent survey (Krunkelsven et al. 2000) verified that some elephants did indeed remain despite enormous hunting pressure. They also verified the continued presence of bonobo (Pan paniscus) and Congo peacock (Afropavo congensis). Poachers invaded the park soon after its creation in 1970, but their numbers now include deserters from Mobutu's army who are heavily armed with automatic weapons. The poachers admitted that it has become increasingly difficult to find elephants (Krunkelsven et al. 2000).
Areas that are unprotected and/or little known during the ongoing conflict
Outside of protected areas, along the eastern border with Uganda at least, there has been an uncontrolled incremental increase in logging. This is evident by the wood stacked by the roadside between Beni and Mambasa (a town about 100 km north of Beni) in areas where there was previously no logging. There is also a constant movement of logging trucks carrying the wood toward the Kasindi border crossing (personal observation). Other border crossings with Uganda (Mahagi and Aru, north of Bunia, and Kasenyi just south of Bunia) are also regularly transferring timber. A trucker who, before the war, was transporting people, food, and cows between Bunia and Niania (a town west of the OFR) is now transporting wood across the border north of Bunia at Mahagi (personal observation). The extent of such logging will not be easily determined by satellite imagery as clear-cutting is generally associated with deforestation for gardens, whereas logging is mainly high-grading where particular trees are felled, cut into planks and hand carried through the forest to the road. This type of logging, particularly when rough roadways or even broad footpaths are cut into the forest, greatly facilitates settlement and bushmeat hunting.
Although our best information comes from protected areas where a foreign presence has been maintained, most protected areas in the DRC have not had any foreign aid during the conflict. Many of these were poorly known even before the war, as there had never been baseline wildlife or plant inventories.
In the Okapi Faunal Reserve, we have been able to observe the consequences of the prolonged unrest in the reserve because of our continued presence on site. Likewise, in the four other WHSs of the DRC, ICCN staff and staff from other conservation NGOs have kept each other informed regarding the changing situation. These sites are now all working together through a project of the World Heritage Center of the United Nations Educational, Scientific and Cultural Organization (UNESCO) and the United Nations Foundation, which began early in 2001 (World Heritage Center 2000).
Most protected areas of the DRC, however, are neither WHSs, nor have they received long-term and continuing investment from an international conservation organization. Of particular concern is the Maiko National Park (Fig. 1), which has been a battleground before and during the ongoing conflict, where military and other armed groups are involved in poaching and where mining camps have become widespread (Fofolo 2000; Masasu 2000). Like most other protected areas in the DRC, MNP lacks funding not only for the park's operating expenses, but also for remunerating the guards that remain on site.
We can only speculate about what impact the current conflict has had on the many unprotected areas critical to the maintenance of DRC's biodiversity. One such area is the Itombwe Mountains, west of Uvira. The latest census, the most thorough in more than 30 years, was carried out in 1996 (Ilambu et al. 1999) just before the outbreak of war. Two new populations of Grauer's gorillas were discovered to the west of an advancing pastoral front, but five populations found in 1959 (Emlen and Schaller 1960) had been eliminated. According to this survey, an estimated population of 900 gorillas still inhabited the mountains. Elephant, buffalo, and chimps were also found in the mountains, and a rich population of prosimians (among which are the nocturnal primates), including Albertine Rift endemics. Particularly spectacular was the diversity of birds, a total of 588 known species with the highest total of Albertine Rift endemics from any of the rift's remnant montane forests (Ilambu et al. 1999). The survey team found that intensive bushmeat hunting, extension of pastures and expanding agriculture are all major threats within the Itombwe region. The Itombwe is also the home of the Banyamulenge pastoralists, the group that led the attack on refugee camps, instigating the 1996 outbreak of warfare in the DRC.
With its long history of human occupation, the best conservation option for the Itombwe highlands will probably be one based on agreements with local authorities that establish different levels of protection for different areas. It is essential that a well-defined extent of the highlands also be given official protected status by higher-level state authorities and that a staff of national conservationists become associated with and dedicated to the protected area.
In the OFR, there is now a zoning project underway which, it is hoped, will allow a limited human use of the reserve and, particularly, will allow the continued practice of Mbuti hunting without putting wildlife species at risk (Mwinyihali and Tshombe 2000). If such a project is successful during this period of prolonged anarchy and strife, it should serve as a model that can be modified for other protected areas or areas in need of protection, such as the Itombwe highlands.
A quantitative assessment of the impact of war on wildlife
Although it is easy to describe the widespread poaching and other exploitation associated with the war and ongoing anarchy, it is more difficult to evaluate its actual impact. Such an evaluation requires a prewar database with which to compare current inventories. Such prewar assessments do exist for several of the protected areas or parts of them.
For example, Garamba, a primarily savanna park, developed an aerial method for counting animals and assessing their distribution in 1993 (Hillman Smith et al 1995; Watkin and Hillman Smith1999). A comparison of aerial counts between 1995, just before the civil war crisis, and 1998, just after, show precipitous declines in populations of large mammals (Smith and Hillman Smith 1998). The elephant population was halved (from >11,000 to < 5,500), and the number of buffalo were down to almost one-third of the prior population (from >25,000 to <8,000). There were also notable decreases in the populations of hippopotamus and hartebeeste (Alcelaphus buselaphus).
KNP, MNP (10,830 km2), and the OFR are all forest parks, and thus aerial surveys are not suitable for assessing animal populations. However, mixed teams of park guards, with foreign and Congolese researchers carried out prewar ground surveys of varying duration and scope in all three areas, and some of the surrounding forests as well (Hall et al. 1998, 1997; Hart and Bengana 1997; Hart and Sikubwabo 1994). The results from all three areas are summarized and compared by Hart and Hall (1996). During the decade preceding the 1996 outbreak of war, there were probably more than 5,000 elephants in both MNP and in the OFR and probably more than 1,500 in the lowland sector of the KBNP. Surveying a part of Kahuzi-Biega lowland sector and surrounding forest, Hall et al. (1997) estimated over 3,700 elephants in a survey area of 15,570 km2, an average of about one elephant per 4 km2. Hart and Hall (1996) also found that the protected areas had higher populations than surrounding areas with more than one elephant per 2 km2 in both MNP and OFR. As to Grauer's gorilla, the greatest single concentration was in the lowland extension of KBNP and adjacent forest, forming a core population of greater than 14,500 individuals (Hall 1998). The chimpanzee population was greatest in Okapi with >7500 individuals, followed by >4,000 individuals in MNP and >1,300 in KBNP. Of all the protected areas, the okapi had its greatest continuous population in Okapi Wildlife Reserve, with probably more than 4,000 individuals.
Since the onset of this latest war, it was only during 2000 that attempts began to carry out systematic surveys in any of these areas. During the year 2000 and early 2001, a recensus of large mammals was undertaken in the OFR. The same transects were evaluated as during the first census. So far, only the results for the elephant population have been analyzed. Remarkably, there was no appreciable difference between the prewar and postwar census.
The results were not so positive for the KBNP. The ICCN park guards are only able to patrol the highland sector of the KBNP. A ground census of this sector was carried out in July–August 2000, which suggested that the gorilla population had been halved during the ongoing war, reduced from 245 to only 130 gorillas (Ilambu, press release). This survey also indicated that the gorillas had suffered more in the prolonged anarchy following the war than from the actual battles that swept through the area in 1996–97 (Ilambu 1998). The entire population of elephants, approximately 350 individuals, had been decimated. Furthermore, illegal snares that kill or maim smaller mammals, including gorillas, had increased fourfold (Omari Ilambu, press release, Wildlife Conservation Society, 2000).
Why the difference in the apparent impact of war on the wildlife of KBNP and OFR? Three important elements are likely to affect the speed and severity of poaching outcome:
+ Poaching pressure—the period of time and the number of guns per square Kilometer;
+ Protection effort—the ability of park personnel to arrest, disarm, and otherwise dissuade poachers; and
+ Size and connectedness of the protected area.
Poaching had a relatively recent and brief insurgence in the OFR, whereas in the highland area of KBNP, the pressure was consistently high from the beginning of the conflict. Most importantly, the area protected is much larger in the OFR than in the highland KBNP (1,373,000 hectares versus 75,000 hectares), the latter having been effectively cut off from the lowland KBNP. Recovery of the highland will necessitate the rejoining of the two sectors and, this is assuming that the lowland, which has been essentially without protection, will nevertheless have maintained populations of key animals because of its size and connectedness to other forest areas.
Before the war, Barnes et al. (1993) calculated that one-third of the forest elephant population of central Africa was found in what is now the DRC. At that time, he estimated that Zaïre accounted for 36 percent of the raw ivory exported from Africa during the 1980s, most of which was illegally obtained. The rate of offtake, all of it illegal, took a sharp upturn in the mid-1990s with the onset of war, and this rate has not yet slackened. This rate of offtake increases the risk that it will be the elephants' rarity rather than the implementation of effective control measures that will curb the slaughter.
Without aerial support, Mubalama (2000) carried out ground surveys of the savanna elephant in VNP in April–July 1998 after the first civil war. He estimated the population of the central sector of Virungas to be between 485 and 535 individuals, down by less than 100 from the estimate of 1983. This small decrease could be due to elephant movements from other areas of the park (or from across the Ugandan border) into the central sector or because, at least at that time, poaching concentrated on other species. Mubalama reported the confiscation of some poached tusks, but also reported high numbers of buffalo, antelope, and hippo carcasses within the park.
Poaching for bushmeat has clearly escalated alarmingly everywhere. Even the use of traditional methods, such as the hunting nets woven from liana fibers used by the Mbuti (Pygmies) in the OFR, can lead to sharp depletion of game animals. Before 1996, hunting camps that were between 10 and 20 km from villages were considered remote camps in the central Ituri Forest (Hart 19992000). Although these camps provided higher yield of duikers and included a higher percentage of larger more vulnerable species than camps nearer to villages, they were only used infrequently. During the period of anarchy in 1999, however, most hunting camps were greater than 20 km from the road and the larger hunting catchments were used for months at a time (Tshikaya and Mapilanga 1999).
During this study, the majority of meat was consumed locally, indicating the importance of local villages as a market; and even the 30 percent that was exported out of the immediate locality went to towns close to the reserve borders. The low price of bushmeat in Mbuti camps and the fact that it was bought through barter for farm produce kept it affordable for local villagers. For example, a blue duiker (Cephalophus monticola) cost the equivalent of about U.S. $0.75 in the hunting camps. (Tshikaya and Mapilanga 1999).
The sustainability of bushmeat hunting was considered problematic even before the war (Wilkie et al. 1998; Hart 19992000), and the rapid deterioration of the situation with the war only emphasizes the importance of protection of large forest areas from settlement and fragmentation (Hart 2000). Increased local settlement increases demand for bushmeat while simultaneously decreasing available habitat.
The impact of armed conflict on civilians
Perhaps the most noticeable change to a visitor who is familiar with an area during a prewar time returning now several years into this drawn-out, war-related anarchy is the enforced isolation and loss of options endured by the local people. This has been obvious in the OFR, where the authors have returned regularly since the outbreak of war and where CEFRECOF and OFR staff have traveled the roads encircling the reserve. The pattern in the OFR is very similar to that in other protected areas.
As was the case with poaching, the war alone cannot be blamed for the deterioration of social conditions. For example, there has been no road maintenance in the Ituri Forest since the late 1980s, and even then it was minimal. During rainy seasons, truck traffic was already rare and caravans of bicycles were becoming the common form of transportation used by merchants, as well as individuals by the mid-1990s. With the onset of war most local vehicles were looted, including those of the OFR, other World Heritage Sites and all Catholic and Protestant missions within the Kivus and Province Orientale. The continued state of insecurity has not allowed them to be replaced in the Ituri, and they have only been partially replaced in some of the other protected areas.
One of the most wearing forms of insecurity for local villagers has been the presence of armed personnel. It did not matter who they represented: Mobutu's old troops, Mayi-Mayi, ADFL, or RCD. In all cases, they had to be fed and their presence in or near any town was a burden. Any villager who had had the audacity to plant a surplus of beans, rice, or peanuts was likely to lose a significant portion to the military.
Although there remained a public administration at most stages of the insecurity, the administration was underpaid and, as a result, often corrupt. Taxes were collected from bicyclists transporting goods. Taxes were collected from conservation NGOs trying to maintain a presence. Taxes were collected from all workers that received any kind of wage. And permits bought for a fee were distributed for anything that could conceivably "need" a permit. That was the case with forest concessions and mines (personal observation). Although mines were illegal in the reserve because, officially the laws of the old regime were still valid, still permits for prospecting and mining were sold. Approximately 50 gold and coltan mines were established with permits from the local district administration by 1999 (Enckoto June 2000).
With a bankrupt administration and almost all commerce occurring through the black market, local communities have had almost no basis on which to bargain. Local prices remain low. The Mbuti continue to expand their hunting area but become poorer and poorer, as indicated by their lack of purchasing power and inability to procure basic health care. The same can be said for other villagers in the area. Without outside markets for selling farm produce, individuals carry beans and rice into the mining camps to sell, and many become miners. But the local value for the mined minerals is very low. For example, the highest quality of coltan found in the OFR ranged from $25–$30 a kilo in late 2000, while a lower-quality coltan costs less than $10 a kilo. Although the price subsequently rose, by mid–2001, the price per kilo had sunk even lower. In May–June 2001, it was at $5–$10 a kilo. To provide a sense of the labor involved, three people working together all day would be considered fortunate to extract a kilo from the streamside sediments. A similar situation exists for ivory for which prices range from $5 to $10 a kilo in Congo and are negotiated at $30–-$40 a kilo in Kampala (personal observation).
The overall effect of war has been not only to keep the local economy resource-based, but also to keep profits low. The resources themselves are undervalued, as competition occurs distant from the source. Local stores are scarcer than before the war, many have closed, and the products they sell are more basic. Medical services and medications are prohibitively expensive relative to the area's low incomes. And the local school is nothing but a parody of learning, with severely underpaid teachers and grossly undersupplied classes. Certainly, there is a large profit made off of the resources being mined out of the protected areas and shaved from the forests, but these profits are, for the most part, being made outside of the DRC's borders.
Disaster preparedness and Resilience of Nongovernmental organizations
The usual case for a nation embroiled in armed conflict is that bilateral- and multilateral-funded conservation projects pack up at the first shots and leave until the peace accords are signed and enacted (Hart and Hart 1997). Even NGOs, feeling unable to make conservation progress or work in their usual manner under the circumstances, are likely to become discouraged, reduce their commitment, and eventually leave. In the DRC however, enough NGOs have remained in key conservation sites to make a positive difference, potentially the critical difference necessary to maintain these sites' biodiversity through the conflict. This "experiment in conservation" has grown from the perseverance of a handful of international conservation groups in allegiance with trained Congolese conservationists, and in many cases, they have made measurable positive impacts (World Heritage Centre 2000).
In all cases it is the combined strength of numerous dedicated individuals that manages to hold up the dikes, but it is still too soon to say whether the united effort of the "conservation team" (i.e., employees of international conservation groups and their national collaborators) will actually be able to turn back the destructive tide. In the GNP the team has rearmed and resupplied guards, run overflight censuses, and engaged with the SPLA to disarm poachers. In OFR the team has built and occupied a new guard post, undertaken a zoning project and a large mammal census, negotiated with the Ugandan military for a mixed military/guard operation to discourage poachers and miners, trained new park guards, and undertaken inventories of the reserve's plants. In the VNP the team has braved patrols despite increasing dangers and minimal support, and it has also organized a joint census of gorillas with collaboration between park staff and military from neighboring Rwanda. In the KBNP the team has organized training with Rwandan military, allowing the park guards to be partially rearmed and to reestablish control over the upland sector. This collaborative team has also worked with the RCD government to try and reassert park status in the KBNP's critical corridor zone, undertaken construction of park headquarters, and carried out a preliminary recensus of large mammals.
An important benefit of the perseverance of the organizations that had previous long-term presence on the ground in these sites is that additional organizations, NGO or bilateral, have been able to make short-term contributions through them to specific sites. But at sites where no international organization had a long-term prewar commitment to conservation, the situation is worse than precarious. Invaluable wildlife and unique habitat are threatened with no on-site advocates. In parks such as Salonga and critical unprotected areas such as the Itombwe highlands, it is not at all certain if waiting for the end of the conflict won't mean waiting too long for critical populations of elephants, gorillas, and bonobos. Both local people and foreign researchers are convinced that even under the current circumstances, it is possible to make significant progress, perhaps decisive moves for conservation (van Krunkelsven et al. 2000). The critical moves include training and strengthening the ICCN, ensuring that they are involved in law enforcement monitoring throughout the park or pushing to increase the area over which they can operate, and initiating dialogue with all the competent authorities to encourage collaboration. It also means assessing and reporting the condition of the protected biodiversity, the fauna and the flora, such that the state of resources that have global value is known and followed globally.
One new project that may greatly improve the capacity of the park staff and their conservation collaborators to reassert control over some protected areas is the UNESCO/United Nations Foundation initiative through the World Heritage Center, "Biodiversity Conservation in Regions of Armed Conflict: Protecting World Natural Heritage in the Democratic Republic of Congo". The project will assure remuneration to the guards in all the World Heritage Sites of the DRC (World Heritage Centre 2000) over a period of four years (2001–2004). Furthermore, this project will assure the purchase of certain fundamental equipment and provide seed-money for certain critical initiatives such as wildlife monitoring and community relations. The facilitation of communications between NGOs and park staff that has occurred through the World Heritage project has already proven to be a significant encouragement to conservation efforts in the DRC. Although the various conservation organizations previously worked independently of one another, they realize that it is only through a joint effort on their part and in close collaboration with DRC park staff that they will have the information, the competence, and the international voice to make a meaningful, lasting difference. The World Heritage project may provide the forum through which site staff can develop international-breadth professionalism based on long-term in-country commitment.
Conclusions—before the end of the war
The effort to conserve nature during a period of extreme insecurity and instability and the effort to do so as a united international and national front, together carve graspable contours to a future conservation ethic. In most African countries, strong conservation leadership is not provided by a government organization. Conservation cannot be a top priority in a bankrupt nation. Therefore, the progress made to even uphold the conservation laws on the books depends on outside donors. The weaknesses of outside donors have been their short-term perspective and their autonomous modes of operation. What conservation under conditions of armed anarchy requires is strong unity of all its proponents, long-term goals, and a fundamental investment in the development of a diverse and competent local leadership.