Tram Chim National Park (TCNP), Vietnam, is one of the last remaining remnants of freshwater wetland habitat in the Mekong Delta. The park receives international recognition as seasonal habitat for endangered wildlife species, including the Sarus crane, and as an extant portion of the Plain of Reeds. Objectives of restoration and biodiversity conservation conflict with local people’s need for economic subsistence and development. Arson is a major cause of fires and will require an economic incentive based fire prevention program. Management objectives of fire exclusion drive decisions relating to water levels, thus impacting all ecological processes within the park. Maintaining high water levels as a tool for fire exclusion conflicts with other management objectives integral to the park. There is opportunity for development of a fire use program within the park, in conjunction with fire effects research to expand the knowledge base. Although fire fighters are aggressive and professional, suppression capability is constrained due to limited equipment.
Upon request of the International Crane Foundation in collaboration with Dong Thap Province, Vietnam, the United States Forest Service conducted a fire management assessment in Tram Chim National Park, Vietnam. A United States Forest Service fire management specialist visited TCNP during the period 1-12 March 2000, to conduct the assessment. The information presented in this report is based on a review of available literature; site visits within the park; observations; and interviews with park staff, scientists, and provincial officials. The information presented may reflect slight distortions based on nuances of translation.
The Mekong River seasonally inundates 3.9 million hectares of land in the delta, including the area originally known as the Plain of Reeds. The Plain of Reeds was a vast wetland depression covered with a thick mat of vegetation. Over the last half century, the majority of the Mekong Delta seasonally flooded grassland and forests have been fragmented by canals, drained, and converted to agriculture production. In the Plain of Reeds area, much of this activity is relatively recent, with approximately 700,000 hectares converted to agriculture during the period 1975-1995.
In 1975, the People’s Committee of Dong Thap Province chose Tram Chim as a site for restoring the original Plain of Reeds ecosystem. Since that time, the area has gradually elevated in status as a conservation reserve to reach its current, designated status at the national level. Tram Chim National Park was established on 29 December 1998, with an objective to protect and restore the natural landscape of the Plain of Reeds and conserve biodiversity. The 7,588 hectare park area is one of the last remaining remnants of freshwater wetland habitat in the Mekong Delta. Over the last couple of decades, several thousand hectares of Melaleuca forests (Melaleuca cajuputi) have been planted within the park. TCNP receives national and international attention as a seasonal feeding and resting area for endangered bird species including the Sarus Crane, the tallest flying bird in the world.
Hydrologic Regime and Fire Issues
Tram Chim National Park is surrounded by 53 kilometres of dikes built in the mid-1980s. Water gates (four constructed in 1991 and two in 1995) control the outflow of water from canals within the park. Canals fragment the interior of the park into five separate management zones, identified as A1- A5. Water levels within each zone can be controlled as a separate unit. Thus, the water level during dry season within each zone of the park is regulated to a large extent by management decision. The hydrologic regime within the park has a fundamental relationship with fire management issues. Moreover, the hydrologic regime impacts all functional aspects of the ecosystem. The annual determination of water level within each zone is a critical management decision, with impacts on all aspects of the park.
The question of appropriate water level at TCNP is a complex and challenging issue. The historic hydrologic regime of the Mekong delta was characterized by a seasonal inundation of water followed by a dry season (December – May), with a gradual recession of water levels. Natural freshwater wetlands store water during times of flood and release it slowly as floodwaters recede during the dry season. The shallow depression morphology of the Plain of Reeds trapped floodwaters, further slowing the release of water. It is reported the surface of the soil was still moist even at the end of the dry season (Safford et al. 1997).
However, restoration of the “natural hydrologic cycle” within TCNP is difficult, as the larger Mekong Delta wetland system has been irretrievably altered by channelization and conversion of land to agriculture production. There is limited data available in Vietnam to re-establish water levels to the historic natural wetlands regime. The issue becomes more complex with the lack of definitive knowledge of the original vegetation occupying the site. Without this information, it becomes difficult to model rates of evapotranspiration and the effect of root systems and ground litter on floodwater retention that would have played a part in the historic hydrologic regime. Annual variability in the amount and timing of precipitation further complicates the principles of water management in TCNP.
In spite of these obstacles, there has been significant progress towards defining the original hydrologic regime. Aerial maps and interviews with long-term inhabitants have yielded valuable information on pre-canal vegetative distribution and abundance. Available hydrologic data has helped in modelling the system. The current water management plan for the park was developed by an international team of hydrologists and is a best guess approximation of historic levels. The plan is described by park staff as highly technical and difficult to implement. In practical terms, the gates are shut sometime shortly after the end of the rainy season. The general principle is that at the end of dry season, evaporation will have left a majority of dry area (moisture just below the surface) interspersed with areas of shallow water.
Implementation of this general water management principle is altered significantly in areas of the park, where water levels are used as a management tool for fire exclusion. This tool has been used extensively since 1995, when a total of 225 hectares of Melaleuca and 306 hectares of grassland burned. For the Year 2000, the fire prevention plan shows 6,758 hectares (zones A1, A2, and A5) of the total 7,588 hectares of land in the park maintained at high water levels for the purpose of fire prevention. Thus in the Year 2000, 89% of the park area is deliberately maintained at high water levels, in response to a fire exclusion management objective. Over the next few years, it is the intent of park managers to gradually lower water levels during the dry season to better simulate the historic hydrologic cycle. However, this intention may be dissuaded with continued pressure for fire exclusion within the park.
The long-term ecological consequences of perennial high water levels on plant and associated wildlife communities are unknown. However, some short-term impacts on crane habitat and Melaleuca stands, two key features of the park, can be evaluated.
Water level is one factor in Sarus crane selection of feeding and resting sites. Sarus crane feed in very shallow water or dry (moisture below the surface) Eleocharis sites with nearby forests with open patches for resting sites (Hoang Van Thang, personal communication). The seasonal presence of Sarus crane at TCNP is the primary tourist appeal, and contributes to the increasingly higher profile of the park, both nationally and internationally.
Of the total 7,588 hectares of park area, approximately 3,000 hectares is currently Melaleuca forest. It is envisioned that Melaleuca forests will constitute approximately 50% of the park area with future planting efforts focused in zone A5 and the strips of land along Phu Hiep and Dong Tien canal. A study of Melaleuca growth rates inside and outside TCNP conducted by the Forest Inventory and Planning Sub-Institute No 2, Ho Chi Minh City concluded the following. “Growth rates of Melaleuca forest differed strongly inside and outside the core and this is assumed to reflect difference in hydrology. In all age classes, growth of Melaleuca forest was considerably faster outside than inside the core. Therefore, water level and flooding duration requires detailed consideration in the management plan of Melaleuca forest.”(Safford et al. 1997). Other evidence of high water stress on Melaleuca trees is the number of trees that have fallen in locations inside zone A1 (Jeb Barzen, personal communication).
With such high stake consequences to key park features, it is critical to identify and address the underlying issues driving the fire exclusion motivation for water level management decisions.
Fire Prevention And Local Communities
All fires in TCNP are human caused. Park staff estimates that ninety percent of fires in the park result from arson, motivated by revenge of local people apprehended for exploiting resources within the park. Five percent is caused by carelessness and the remaining five percent is caused by fire used to smoke out bees or rats. Laws prohibit local people’s access to the park’s natural resources, unless operating under a specific utilization permit compatible with park objectives (removal of Mimosa pigra, collection of dead and downed woody material, etc.). As a result, relations between the park and local communities have been tense since the park’s establishment. Objectives of restoration of the original landscape of the Plain of Reeds and biodiversity conservation conflict with local people’s need for economic subsistence and development. The conflicts created by these two seemingly incompatible land use objectives have been a dominant community theme over the last two decades.
Five impoverished villages surround TCNP. Many of the people from these villages depend on natural resources for survival. Vietnamese local custom allows free use of the forest and wild land resources. Despite laws prohibiting trespass and use of resources within TCNP, impoverished local people routinely exploit accessible resources in the park. This exploitation includes: land development for agriculture (primarily rice); cutting Melaleuca for cooking fuel and lumber; extracting snails, snakes, fish, frogs, and water birds for food; cutting grass for vegetable garden mulch; and utilizing grazing land for livestock. Evidence of poaching activities abound (fishing nets, batteries for electro-shocking, cut stumps, fishing poles, cut grass, etc.) and local people were routinely observed exploiting resources inside TCNP. Although there is a fish harvest plan for the park, it is not implemented because estimates of illegal fishing harvest far exceed the amount authorized within the plan. Resentment against TCNP guards, who are allowed to utilize park resources, further fuels the tension. In a recent act of violence, apprehended local electroshock fishermen threw battery acid into the faces of two park guards when directed to leave the park.
A number of efforts are planned or in progress to improve relations with the community and lessen pressure on the natural resources of the park. A donor funded small loan program has been initiated to improve economic conditions in the impoverished communities around the park. The fund is coordinated by the district Women’s Union and used to support small-scale economic enhancement activities. A pilot project will be implemented next year to encourage community involvement in an anti-poaching patrol. Designated family units would have limited and specified utilization rights to five hectares of park land, in exchange for protection of that land. As part of a poverty alleviation program by the central government, the district has proposed to use the border strip along the edge of one side of the park as a utilization zone for people who currently live there. Although these people were officially “displaced” when the park was established, they remained in place and the strip is currently a de facto utilization zone. It is hoped the change in occupancy status of these people from illegal to legal will create goodwill and cooperation between the villagers and the park.
A commonality of successful fire prevention programs in developing countries is the linkage of fire prevention with some type of economic incentive. Communities must have a vested interest in fire prevention. In the case of TCNP, options for economic incentive of limited utilization of park resources must be balanced with minimal disturbance to park wildlife. In discussions with park staff and managers from commercial Melaleuca forests the following options seemed feasible.
Establish a community-monitored scheme to distribute benefits (extra rice, labour assistance, etc.) to local communities for successful prevention of fires during a dry season. The proposed scheme should be evaluated for efficiency, equitability, and sustainability before implementation.
Establish a small loan program (seed money of US$25,000) with eligibility requirements linked with successful performance of fire prevention.
Allow limited utilization rights within the park contingent upon successful fire prevention.
In evaluating program options, consideration must be given to local people’s confidence in the validity of the proposed long-term outcome, and local capacity for project management and oversight. If an effective community fire prevention program can be established with linkages to vested economic interests, the pressure to maintain high water levels as a means of fire exclusion will be lessened.
A TCNP fire prevention plan is prepared annually by the Coordinating Committee, which consists of the four TCNP department heads and other key staff. The Year 2000 fire prevention plan for the park contains four main elements.
Maintenance of high water levels in specified areas of the park.
Prescribed burning in high grassland to establish fuelbreaks around the Melaleuca stands.
Education to the surrounding communities.
Establishment of fire brigades at each hamlet with assistance from the district People’s Committee.
A section of the fire prevention plan lists each park zone and identifies the high-risk fuel complexes. Proposed fuels management activities for the Year 2000 include construction of 400 hectares of firebreak in A1, 200 hectares of firebreak in A2, and prescribed burning of 200 hectares in A4.
Community fire prevention education is accomplished through social structures. The Women’s Union, Farmer’s Union, and Youth Union give public service messages containing fire prevention messages and advise concerning forest protection laws.
There is a national standard of fire danger rating using a scale of 1-5. The fire danger rating is determined based on climate and fuel load. Every 15 days during the dry season, province level officials work with personnel at TCNP to determine the fire danger rating. Signs indicating the fire danger rating are posted at many of the access points to the park. A high fire danger rating is not linked with commensurate levels of prevention activities, as there are not enough resources for such a program.
Fire Ecology and Use of Fire
Of the total 7,588 hectares within the park, approximately 3,000 hectares are Melaleuca cajuputi forest. The remaining hectares are grasslands composed of a mixture of species that can be divided into five main plant communities.
Although there is little scientific research of fire effects on Melaleuca cajuputi, long-term park employees and scientists have considerable knowledge based on observations. In 1997, a dense mat of Melaleuca propagules was observed in a Melaleuca forest area that had burned in 1995. The seeds had germinated following the fire (Tran Triet, personal communication). On that same site, trees that had been salvage logged after the fire re-sprouted naturally from the stumps. As of March 2000, the stand was approximately 2 meters tall. In another Melaleuca site, there was little evidence of regeneration from a 1994, six-hectare, high intensity fire. At a third site, young Melaleuca trees regenerated from seed were plentiful in the area of a 1998 grassland fire. Park staff conjectured that scattered, large Melaleuca trees had seeded following fire stress. The seeds had then been transported by water and germinated.
These observations are consistent with fire effects research on Melaleuca quinquenervia found in Eastern Australia, New Caledonia, Papua New Guinea, and Irian Jaya. A summary of studies indicates the following (Coladonator 1992):
“Melaleuca is well adapted to fire. It has a thick, spongy bark that insulates the cambium. The outer layers of bark are flakey and burn vigorously, which conducts the fire into the canopy, igniting the oil-laden foliage. The leaves and small branches are killed, but dormant lateral buds on the trunk germinate within weeks after the burn. This prolific sprouting increases the surface area of small branches and therefore the tree’s reproductive potential. Melaleuca can flower within weeks after a fire. Each serotinous capsule contains about 250 tiny seeds which are released after a burn, frost, or any other event that severs the vascular connections to the fruit. A burned Melaleuca tree can release millions of seeds, which are dispersed a short distance by wind and water. Seedlings that are less than 3 to 6 months old, or only 4 to 8 inches (10-20 cm) high are often killed by hot surface fires. Death may occur as a result of lethal temperatures, insufficient food reserves to produce new shoots, or drying of the sediments due to increased exposure. Older seedlings are top-killed by most fires but recover quickly often with multiple shoots sprouting from the root collar. A mature Melaleuca tree will respond to fire by releasing millions of seed that are held in the capsules on the twigs and branches. Adults damaged by fire may also sprout vigorously from the roots, bole, or branches. The proliferation of branches on a burned adult tree often results in more twigs, and thus more capsules and more seed production than was the case prior to the burn.”
Studies of Melaleuca cajuputi in Thailand indicate a fast growing shade tolerant species with high tolerance to fire and flooding (Safford et al. 1997).
Park staff has expressed interest in thinning as a silvicultural treatment in Melaleuca, followed by low intensity prescribed burning. However, park regulations at the national level prohibit cutting of live standing trees. A variance for a small pilot project would be useful to determine applicability towards meeting management objectives.
Experiments using prescribed fire to create firebreaks in the park were initially conducted in 1992. During the dry season of 1999, a prescribed burn experiment was conducted in five grassland communities (Khanh et al. 1999). The objectives of this preliminary research were to: 1) determine the range of fire intensities and appropriate techniques for cost-effective fuelbreak construction, and 2) determine fire effects on species composition and abundance. This experiment lays the groundwork for future prescribed burning experimentation and program implementation. Anecdotal observations of Sarus crane observers indicate that feeding habitat is enhanced after a burn, where cranes readily feed in recently burned areas. These observations support the notion of fire effects research linked with an objective of wildlife habitat enhancement. Experiment design should incorporate variables such as burn seasonality (early/late dry season), burn severity (low moderate, high), burn frequency interval (3 months, one year, two years, five years, etc.), and vegetative communities (five different grassland communities and Melaleuca stands of varying age class/densities). Vegetative response characteristics should include such measurements as shoot height, density, weight, growth rate, and nutritional composition; aboveground standing crop; residual organic materials; mortality; and reproductive effort.
Given the pilot projects using fire to establish fuelbreaks, there exists ample foundation to establish a larger scale program of fire use and research experimentation at TCNP.
Fire Suppression Organization and Capabilities
The TCNP Forest Protection Unit is a special unit within the province Forest Protection Department. This unit has 34 employees (potential for 40), staffing twelve guard stations and the headquarters office. There is one 5-person mobile patrol, with motorcycle or boat transport. The province Forest Protection Department provides two days of fire suppression training annually. The training covers laws and regulations relating to forest fire, and new equipment and methodologies in fire suppression. In addition to fire suppression, the Forest Protection Unit is responsible for anti-poaching actions and planting contract administration.
The TCNP Forest Protection Unit has the following fire equipment:
1 Vanguard Briggs and Stratton 6 horsepower portable pump *
1 Industrial Plus Briggs and Stratton 5 horsepower portable pump*
3 larger pumps
300 meters of hose, 7.6 cm (3 inch) diameter in 20 meter lengths*
2 Wajax drip torches
7.6 cm (3 inch) quick-lock, haemophrodite couplings (possibly Storz or French variation)
The unit has no radio equipment. There are telephones at the headquarters and three of the guard stations. The other guard stations have no means of communication with headquarters. This is a critical safety concern not only in a fire response situation, but also during other protection activities. The lack of personal protective equipment, particularly hardhats with some type of face shield, was also noted as a safety concern.
Mechanized equipment is used in fire suppression efforts. Tractors, rented from local farmers, are used to construct firelines around grass fires. Park staff expressed an interest in a water tank trailer with spigot that would hitch to the back of a locally constructed all-terrain vehicle. The water trailer would be strategically placed along a fire perimeter to eliminate the need for labour intensive, hand-carried water.
The district government is responsible for organizing and mobilizing resources from the local villages to assist in fire suppression efforts within the park. By law, the village head is responsible for deployment of locals to fight fire. Each village designates one family as a focal point to assist with fire prevention and suppression efforts. One day of training is provided annually to these families. It was noted that local farmers usually bring their own tools or agriculture sprayer when responding to fires.
When a fire in TCNP exceeds the capability of the Forest Protection Unit, the province level Forest Protection Department assists the park. The province-level Forest Protection Department is structured as follows: Director, Vice-Director, a 5-person Fire Protection Unit, 5 mobiles units, Policy and Regulation Unit, Administrative Unit, and the special Fire Protection Unit at TCNP. There are district and province level command centres. Smaller fires are handled locally, but if a situation exceeds the capability of the district, requests for assistance is sent to the province central command and resources deployed from that level.
Figure 1. Freshly burned Melaleuca leucodendron L.stand in Long An, Viet Nam (Mekong delta region). Fires in Melaleuca are often set to obtain permission to salvage the stems and to bring them to the local markets. Photo: GFMC.
Figure 2. Local transport and trading of freshly salvage-logged Melaleuca poles. Photo: GFMC.
There is no formal firefighter training. Although fire suppression tactics safety is taught, they have no personal protective equipment. In 1999, the province Forest Protection Department responded to a total of 88 fire incidences in the province.
During the dry season of 1998, there were a total of 40 fires in Tram Chim National Park. Three of those were large fires requiring significant deployment of fire suppression resources. The largest fire, 600 hectares of grassland, required mobilizing 300 firefighters including police and army personnel. On March 8, 2000, the TCNP Forest Protection Unit responded to the first fire of the dry season. The fire was burning in grassland (0.7 meter high Penicum) and scattered, 2-year-old Melaleuca. Nine firefighters responded on initial attack using boats for transportation. Equipment included: 1 Vanguard Briggs and Stratton 6 horsepower portable pump, 1 Industrial Plus Briggs and Stratton 5 horsepower portable pump (Model 133432), fifteen 20-meter lengths of 7.6cm (3 inch) diameter hose with quick-lock haemophrodite coupling, 1 Kubota pressurized agricultural sprayer, and plastic buckets. A wetline anchored from the canal failed to contain the fire. All available hose (300 meters) had been deployed and the fire burned around the end of the hoselay. With winds around 16 kph, the fire burned back towards the canal, and soon spotted across. The firefighters attempted a wet line along one flank but the fire was too intense. After retreating to the canal dike for a short period, they attempted a hose lay attack from the black side of the flaming front. The wind had slowed and the wet line was successful this time in stopping the fire. The firefighters deployed all available hose and then resorted to beating the edge with Melaleuca boughs and carrying water in plastic buckets from the end of the hose lay. The firefighters wore sandals or were barefoot and had no personal protective equipment of any kind. Local people from the communities were conspicuously absent in the fire suppression efforts, although the fire clearly exceeded the suppression capabilities of the limited park personnel.
The firefighters were very aggressive, hard-working, and professional. Although they lacked personal protective equipment, safety was clearly a factor in selecting tactics. The limited water handling equipment was utilized as effectively as possible. Additional pumps, smaller diameter hose, adapters, couplings, nozzles, training in hose packs and hose lays with laterals, fold-a-tanks, and personal protective equipment would maximize the effectiveness and safety of this firefighting unit.
Analysis – Management Dilemmas and Tradeoffs
Fire exclusion as a management objective is a high priority with district, province, and national officials. Fire exclusion is the primary driving factor in determining water levels at TCNP. Succinctly displayed, the tradeoffs associated with the decision to utilize high water levels as a means of fire exclusion, are summarized as follows.
I.High water levels maintained for the purpose of fire exclusion
Reduced area of bird species habitat
Slow Melaleuca growth
Unknown other ecological consequences
II. Water levels allowed to recede “normally” during the dry season as per the water management plan:
Higher risk of human-caused fire
Larger proportion of park area suitable for bird species habitat
Increased Melaleuca growth
For the year 2000, high water levels were maintained in 89% of the park area, substantially limiting potential crane habitat. Paradoxically, the park logo, tourist brochures, and public relations material, all feature the crane as a centrepiece. From a political, managerial, and accountability perspective, hectares of Melaleuca forest and grasslands protected from fire is an easily measured target. In contrast, the temporal presence of Sarus cranes in TCNP is dependent on a number of variables, many of which are not within the control of the park (availability of suitable habitat along the migration route, poaching outside of TCNP, reproductive success, etc.).
Another recent management decision generated by wildfire concerns was the construction of new canals through some areas of the park. This action was justified by the need for access by fire suppression resources. Although canals do allow access by fire suppression crews, they also increase ease of access by locals exploiting resources within the park. Since all fires are human caused, this increased access by locals will likely result in more fire ignitions. Construction of additional canals has adverse impacts on soil acidity and the hydrologic regime, and is also contrary to objectives of the park to protect and restore the natural landscapes of the Plain of Reeds. Although canal construction was halted before completion, this management decision is another indicator of the high value placed on fire exclusion within the park.
Fire prevention in an atmosphere of tense and occasionally hostile community relations presents another daunting management challenge. In commercial Melaleuca plantations in Dong Thap province, utilization permits for locals and community benefits (rice distribution) purchased with profits from the plantation ensures positive relations with the community. In TCNP, unrestricted utilization rights in reward for fire prevention is not an option due to park objectives. The park does not generate revenues nor currently receive a budget for improving community relations, thus it is difficult to find incentives that could serve as a bridge for improved community relations. Yet improved relations with local communities is a critical step in reducing the risk of unwanted fire.
The issues faced by managers at Tram Chim National Park are complex and challenging. Outcomes of potential decisions should be evaluated carefully, within the continuum of single species management and biodiversity conservation. The concepts of disturbance factors and the role of fire in the ecosystem are poorly understood by officials at all levels and there is little research on fire effects in the ecosystems within the park. The current paradigm of fire exclusion as a management objective should be re-evaluated within the context of natural disturbance factors.
In summary, moving from a strict fire exclusion policy towards recognition of fire’s role in the ecosystem, augmented by a program of fire use in the park would allow for greater flexibility in management options. A deliberate strategy of fire use, in combination with an effective community fire prevention program, based on economic incentive extraneous to park resources, would result in a decrease in unplanned ignitions. Improving the capacity of fire fighting forces to suppress unwanted fires would be the final step in enhancing the overall fire management program.
Birdlife International. Wetlands in the balance. A strategy for balance and harmony in wetland resource management in the Mekong Delta. Unpublished report.
Coladonator, Milo. 1992. Melaleuca quinquenervia. In: The Fire Effects Information System (Data base) (W.C. Fischer, comp.). Missoula, MT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Intermountain Fire Sciences Laboratory. Magnetic tape reels; 9 track; 1600 bpi, ASCII with Common LISP present.
Khanh, Le Trong, and Le Thanh Cu. 1999. Report on the Research. Using prescribed burning as a mean to prevent fire in melaleuca forest and the management of plant communities in the frequently burned conditions. Unpublished report.
Kiet, Le Cong. 1993. Dong Thap Muoi; restoring the mystery forest of the Plain of Reeds. Restoration and Management Notes 11, 102-105
Safford, R.J., D.V. Ni, E. Maltby, and V.-T. Xuan (eds.). 1997. Towards sustainable management of Tram Chim National Reserve, Vietnam. Proceedings of a workshop on balancing economic development with environmental conservation. London: Royal Holloway Institute for Environmental Research.
IFFN/GFMC contribution submitted by:
Fire Management Specialist
United States Forest Service
P.O. Box 3
Kernville, California 93238