5.3 Assessment of Current Strategies To Control Fire in The Ramsar Site

EnviroNET Solutions Pty Ltd
5.3 Assessment of Current Strategies to Control Fire in the Ramsar Site
A requirement in the terms of reference was to assess the effectiveness of current

strategies (e.g. fire breaks, education, etc.) and policies (i.e. burning allowed with permits)
used to stem and control fire in the Ramsar site and suggest alternatives. The Department
of Forestry & Range Resources in Maun is the Government department responsible for the
suppression of wildfires and the administration of all regulations pertaining to fire in the
Ramsar Site. The assessment of the effectiveness of the Departments to suppress wildfires
and to administer the regulations pertaining to fire will be dealt with separately.
5.3.1 Fire Suppression
Currently all fire fighting equipment and staff are centralized in Maun and an inspection of
these resources showed that there were 16 fire fighting units referred to as “slip-on’s” or
“bakkie sakkies” and these were assessed to be adequate for the task of suppressing
wildfires in the Ramsar Site. However, a significant number of the units were not in working
order and available for instant use if required. Also the units stored at the offices of the
Department of Forestry and Range Resources were not suitably housed under cover and
were exposed to the elements of the weather and were showing signs of deterioration due
to the effects of rain, high day temperatures and harsh sunlight. While these are robust
forms of machinery well suited to the rigours of field use plastic and rubber hoses need to
be protected from rain and extreme sunlight when the units are not in use. An assessment
of the number and condition of the knapsack sprayers and fire beaters showed that there
were an inadequate amount for the task at hand and were generally in very poor non-
working condition being stored in the open between two buildings at the departmental
office. Again this type of equipment is adequately robust for field use during fires but unless
it is adequately stored under cover it deteriorates and becomes non-functional very rapidly.
The inspection also revealed that the staff involved in fire suppression were inadequately
trained and equipped with suitable clothing and footwear for fighting fires. The policy of
having all the fire fighting equipment and staff centralized at the office in Maun was
assessed to be inappropriate for being able to respond timeously and effectively for
wildfires occurring over such an extensive and often difficult to access area like the Ramsar
Site. Also the existing system of using local people on a volunteer basis to suppress fires
has not been effective due to the volunteers requiring some form of compensation for their
time and expenses during these fire incidents. Finally It was concluded that the
decentralization of responsibilities, staff and equipment is a pre-requisite for the
development of an effective fire suppression capability in the Department of Forestry and
Range Resources and this will be dealt with in detail in the section on recommendations for
controlling fires in the fire management plan – see section 6.4.
The main method used in Botswana for preventing and controlling wildfires is the
construction of strategic fire breaks approximately 30 metres wide. This is done using
graders and other mechanical equipment. After inspecting and assessing the existing fire
breaks in and around the Ramsar Site, as well as other fire breaks in Botswana during a fire
study tour in December 2005, the following conclusions were drawn:
• Many of the firebreaks were planned not taking local circumstances and weather into
consideration. For instance, some fire breaks were constructed parallel to the
Okavango Delta Management Plan – June, 2006 99
prevailing easterly winds and would therefore not prevent a wildfire from spreading in
a westerly direction. Aerial surveys conducted during April 2006 also showed that
existing strategic fire break systems have not been maintained and unless attended
to immediately would not be effective in controlling the spread of wildfires during this
fire season.
• Greater use could be made of natural barriers in the landscape as firebreaks and
with limited use of strategic burning potential buffer zones could be created with
minimal impact on the sensitive environment in many areas of the Ramsar Site.
5.3.2 Herbage Preservation Act
The Herbage Preservation Act promulgated in 1977 is the legal framework administering
the management of fire in Botswana. A detailed study of this legislation indicates that
certain changes are needed to enable the effective management of fires both in Botswana
and in the Okavango Delta Ramsar Site. These changes necessitate the development of
guidelines for managing fuel loads and regulatory structures to mitigate and better manage
the threats associated with wildfires.
5.3.2.1 Current situation
After analyzing the existing Act and extensively investigating the situation on the ground in
the Ramsar Site, it was found that the existing Herbage Preservation Act caters for a
complex and well structured set of rules for a potential integrated fire management system
in Botswana and the Okavango Delta Ramsar Site. However the Act requires additions to
the existing content to assist the State Departments and land users in the implementation of
the recommendations that are proposed within the ODMP Fire Management Plan. These
include a system to evaluate the fuels before burning to determine the necessity to burn,
and what method would be desirable for a specific situation. The implementation of a Fire
Danger Rating system would also create a safer environment to carry out the prescribed
burning and would greatly reduce the threat of fires to humans, infrastructure and the
environment as a whole. The Fire Danger Rating system would also allow for better early
warning of adverse weather conditions and being better prepared for possible wild fires.
5.3.2.2 Proposed changes to Herbage Preservation Act
5.3.2.2.1 Establishment of District Fire Committee and Sub-District Wards
The Act provides for existing District Committees to form sub committees as prescribed in
the Herbage Preservation Act. The establishment of a sub-committee within the Ramsar
Site to implement and manage all fire management within the Ramsar Site is essential. An
existing committee could be used or a new committee constituted for this function. The
membership of this fire sub-committee (District Fire Committee) must be made mandatory
to all land users within the Ramsar Site. This must include all stakeholders, both state and
private.
The District Fire Committee’s Executive Committee and Chairman must be assisted by a
“fire ranger” as described in the Herbage Preservation Act with the objective of
implementing and managing fire within the Ramsar Site. The “fire ranger” would be a full
time employee of the District Fire Committee and would be remunerated by funds raised
through a levy on members of the fire sub-committee. The District Fire Committee must
also be empowered to levy members to raise the required funds to effectively manage fire
within the Ramsar Site. The committee’s responsibility and objectives would be
implemented through a system of sub-district wards within the Ramsar Site.
To facilitate the implementation of the Herbage Preservation Act, the issuing of burning
permits for controlled burning, the co-ordination of suppression of wildfires and the timeous
response to these incidents with trained crews and equipment, the EnviroNET consultants
recommend the de-centralization of the existing infrastructure currently based at the
Department of Forestry and Range Resources in Maun. It is recommended that sub-district
wards be created and controlled by the District Fire Committee in Maun and that they be
based in Seronga, Shakawe and Tsau. A sub-district ward will also be based in Maun to
service the southern portion of the Ramsar Site and to act as additional back-up in the
event of a major wildfire in any of the sub-district wards (see Fire Prevention Map). The
staff and equipment could be based at existing Agricultural Offices, or in the event of there
being no Agricultural Office in a sub-district ward, could share facilities where Community
Trusts operate, for example. The existing centralized Government fire crews must be de-
centralized and supplemented where necessary, to these sub-district wards as this will
allow for crews to react to the initial attack on wildfires on government and community land.
It is recommended that 11-person fire crews be established and crews stationed at the sub-
district wards where they will be closer to assist with extended attack on fires within their
wards and be available to assist with the burning of existing firebreaks and to perform any
controlled burning required by the communities. In the case of Maun it is recommended that
there be two 11-person fire crews because of the extended nature of this ward extending
from Toteng up to Linyanti. They would also be available to other land users for
prescribed/controlled burning. The proposed structure establishing sub-district wards and
the issuing of burning permits is illustrated in Figures 35 and 36.
DISTRICT FIRE COMMITTEE

MAUN
Fire Related Responsibilities
SERONGA SHAKAWE MAUN TSAU

SUB SUB SUB SUB
DISTRICT DISTRICT DISTRICT DISTRICT
WARD WARD WARD WARD
2 x 11
1 x 11 1 X 11 1 x 11
Person Fire
Person Fire Person Fire Person Fire
Crew +
Crew + Crew + equipment Crew +
equipment equipment equipment
Coordinated by Fire Ranger from District Fire

Committee
Figure 35. Proposed decentralization of fire related responsibilities of the District

Fire Committee in Maun to Sub-District Wards in Seronga, Shakawe, Tsau
and Maun as provided for under the amended Herbage Preservation Act.
5.3.2.2.2 Delegation of responsibilities for issuing burning permits
To facilitate the implementation of the Herbage Preservation Act with respect to issuing
burning permits for controlled burning the EnviroNET consultants recommend the de-
centralization of the existing infrastructure currently based at the Department of Forestry
and Range Resources in Maun to offices in the sub-district wards. Easy access to offices
issuing burning permits and burning permit plan forms (see Table 13) will facilitate and
encourage stakeholders, in particular those from rural communities, to adhere to the
requirements of the Herbage Preservation Act. Burning permits should be printed in
triplicate and bound in book form to ensure that copies of permits are not lost and that data
capture can be facilitated. Two copies, signed by the delegated issuing officer may be
removed from the book - one signed copy is to be issued to the applicant, and the second
signed copy is to be sent to the Maun Head Office at the Department of Forestry and
Range Resources on a monthly basis, for data capture and permanent filing. The third copy
of the permit must remain in the book and all permit books are to be safely stored so that
reliable records are kept for future reference. The decentralization of issuing of permit is
illustrated in Figure 36.
DISTRICT FIRE COMMITTEE

MAUN
Permit Issuing Responsibilities
SERONGA SHAKAWE MAUN TSAU

SUB DISTRICT SUB DISTRICT SUB DISTRICT SUB DISTRICT
WARD WARD WARD WARD
COMMERICAL COMMERCIAL COMMERCIAL COMMERCIAL

FISHING & FISHING & FISHING & FISHING &
TOUR TOUR TOUR TOUR
OPERATORS OPERATORS OPERATORS OPERATORS
CONCESSION CONCESSION CONCESSION CONCESSION

HOLDERS HOLDERS HOLDERS HOLDERS
LOCAL LOCAL
COMMUNITY COMMUNITY LOCAL
LOCAL
COMMUNITY
COMMUNITY
Duplicate copy of permits sent to District Fire Office - Maun
Figure 36 Infrastructure showing delegation of responsibilities for issuing permits

for controlled burning by the District Fire Committee in the Okavango
Delta Ramsar Site.
5.3.2.2.3 Fire Danger Rating System
Currently no Fire Danger Rating system is used in Botswana or the Okavango Delta
Ramsar Site for assessing the potential fire danger on a daily basis as a preventative
measure for planning and coordinating fire fighting measures. Information from such a
system is also essential for deploying staff and equipment when fighting fires and for
selecting the most appropriate weather conditions to use when applying controlled burns.
The recommendation is for the introduction of an existing system that has been in use for
the past 20 years in Southern Africa.
Lizette and Johan Heine of the Forest Fire Association, in Nelspruit, South Africa,
developed the Fire Danger Index (FDI) system being recommended for use in Botswana.
The Fire Danger Index system was initially developed in Australia, further used and
modified in Zimbabwe, and finally adapted in South Africa for assessing the fire danger in
the commercial forestry industry as a means of facilitating fire suppression. A recent
adaptation and development has been the inclusion of the assessment of the flammability
of grass fuels, as influenced by the degree of curing of the grass sward, thus making it a
valuable tool for controlled burning and also the suppression of wildfires in African
Grasslands and Savanna ecosystems (Trollope, 2005).
The Fire Danger Rating system is colour coded and provides clear guidance in terms of
when to burn, or not to burn, and on the level of readiness for controlled burning or fire
suppression activities. For instance, on a “green” day with a low FDI of 21 – 45 it is suitable
to apply controlled burns for valid reasons and for fire suppression readiness nobody is
required to be on standby, whereas on a “red” day with a high FDI of 75 – 100 it is too
dangerous and unsuitable for controlled burning and all available resources should be
placed on standby as the probability of wildfires is high.
Prescribed burning is inherently dangerous as one uses fire to manage a certain situation.
The use of a Fire Danger Rating system will reduce the threat posed by prescribed fires as
it includes reasons for applying controlled burns and guide lines as when to burn and when
not to burn. It is also a very valuable tool for assessing the danger to personnel when
engaged in fire suppression activities and the degree of preparation that should be in place
under specific weather conditions.
The details of the recommended Fire Danger Rating System are presented in Table 12.
Table 12 Fire Danger Rating System for controlled burning and fire suppression in
African grasslands and savannas.
5.3.2.2.4 Prohibition on burning vegetation
No burning of any vegetation will be permitted without an assessment being completed of

the condition of the grass sward using the ecological criteria that have been developed for
the Acacia, Burkea, and Mopane Woodlands and the Seasonal Swamps in the Okavango
Delta Ramsar Site and the relevant documentation presented to the officer delegated to
issue burning permits.
5.3.2.2.5 Duties to extinguish fires
Fire suppression should be the responsibility of the concession holder or the land user. The
Government responsibility will be to supply assistance in the case of a wildfire outstripping
a concession holder or land users resources.
5.3.2.2.6 Readiness for fire fighting
The Herbage Preservation Act must require all land users to have sufficient, trained
personnel, equipment and personal protective clothing to accomplish their obligations in
terms of fire suppression and/or when required to implement a prescribed burning program.
The District Fire Committee would be responsible for determining the specific, minimum
requirements for personal tools and personal protective clothing.
Land users are obliged to ensure that during times of adverse weather conditions (high Fire
Danger Indices) persons are available to respond to fires threatening their property, the
environment and neighboring land users. Assisting others must be obligatory and failure to
do so may lead to prosecution.
5.3.2.2.7 Burning permit
The current burning permit in the Herbage Preservation Act requires an addition to the
existing permission-to-burn permit in the form of a burning plan that must be completed by
the person requesting to burn. The burning plan must be completed before the permit will
be issued and before controlled burning takes place. The details of the burning plan are
presented in Table 13 and must appear on the reverse side of the current burning permit.
BURNING PERMIT PLAN
1 1.1 DATE 1.2 IGNITION TIME 1.3 FINISH TIME

2 AREA TO BE BURNED 2.1 LOCATION/REGION
2.2 PROPERTY 2.3 PROPERTY
3 REASON FOR BURNING
3.1 REMOVE MORIBUNG GRASS: 3.2 CONTROL ENCROACHING PLANTS:
4 RANGE CONDITION 4.1 GRASS FUEL LOAD = kg/ha
4.2 % Decreasers & Increaser I spp = % 4.3 % Increaser II spp = %
5 SIZE OF AREA TO BE BURNED - HA
6 FIRE BREAKS OR TRACES IN PLACE ? YES NO
7 RESOURCES / MANPOWER
7.1 VEHICLES 7.2 BAKKIE SAKKIES 7.3 TRACTOR UNIT
7.4 EXTRA WATER 7.5 FIRE BEATERS 7.6 KNAP SACKS
7.7 BASIC FIRE FIGHTERS 7.8 FIRE FIGHTERS 7.9 RADIOS
7.11 DISTRICT COORDINATOR
7.10 PROTECTIVE CLOTHING YES NO
INFORMED
ST ND
7.12 1 NEIGHBOUR 7.16 2 NEIGHOUR
YES NO YES NO
INFORMED INFORMED
7.13 DATE INFORMED 7.17 DATE INFORMED
7.14 NAME 7.18 NAME
7.15 TEL/CELL/FAX 7.19 TEL/CELL/FAX
8 WEATHER & FDI ON DAY OF BURNING 10H00 14H00
BLUE GREEN YELLOW ORANGE RED
0 – 19 20 – 44 45 – 59 60 – 74 75 - 100
CONDITIONS:
1. Burning will be allowed when the forecasted FDI is less than 55 for slash burns and block burns,
burning of fire breaks will be allowed if FDI is less than 55, unless special exemption is required.
2. Adequate precautions are taken in terms of available manpower and minimum equipment.
3. Adequate liaison with surrounding land users and issuing officer or fire ranger has taken place.
4. The area to be burned is surrounded by an adequate firebreak/s or tracer lines are in place.
5. No burning on a holiday or on weekends from Friday, special burning permit must be obtained
to be signed by all adjacent neighbours if burning to take place on these days.
6. You are responsible to prevent a fire from spreading to neighbouring properties.
7. Do not leave a fire unguarded / unattended before it is properly extinguished, and particularly
if the next day the forecasted FDI is organge or red.
8. According to the Herbage Preservation Act, carelessness with fires is considered a criminal offence.
9. Report back after successful burning operation is completed to all relevant parties.
10. Responsibility for controlled fire remains with the landowner / applicant.
COMMENTS
APPLICANT NAME Tel:

SIGNATURE
ISSUING OFFICER NAME Tel:
SIGNATURE
The current misinterpretation of the Herbage Preservation Act in conveying the message to
the public that all controlled burning is against the law has instilled a culture of fear of
prosecution for the use of fire, with the resultant illegal use of fire to achieve various
objectives. In addition the difficulties experienced by stakeholders, notably rural
communities (see section 5.1.2), to obtain permits as required by the current official
Herbage Preservation regulations, requires that the current system for issuing permits be
changed in order to facilitate the process. In order to change attitudes and encourage
stakeholders to adhere to the legal requirements for burning in the Ramsar Site, the
aforementioned decentralization of issuing permits is necessary. This requires that
authorized officials in the sub-district wards may issue permits once suitable training and
monitoring systems have been developed and implanted.
5.3.3 Proposed amended Herbage Preservation Act
In accordance with the recommendations made in section 5.3.2 amendments to the current
Herbage Preservation Act have been presented in Appendix 5. The relevant Government
Committee should review these proposed amendments to the Act and any changes to the
Act should only be made upon the advice of the relevant legal advisory body to the Minister.
5.4 Identification of Deficiencies in Current Knowledge on Fire Ecology of the

Ramsar Site
In accordance with the requirement in the terms of reference of identifying gaps in the fire
ecology of the Okavango Delta Ramsar Site the following deficiencies in knowledge
became apparent and require both immediate and long term research attention and the
development of scientific services necessary for implementing the proposed fire
management plan for the Ramsar Site. The recommended research will be divided into two
categories, namely applied research and basic research.
5.4.1 Applied Research
The following applied research projects need to be conducted:
• The development of a simple and practical vegetation map of the Ramsar Site that
can be used for the implementation of the fire management plan. It is recommended
that the vegetation map that has been developed for this current study be taken as a
starting point as the five vegetation units comprising the Acacia, Burkea and Mopane
Woodlands and the Permanent and Seasonal Swamps have proved to be a very
practical classification of the vegetation for management purposes. However, the
boundaries and the ecological dynamics of the dominant plant communities within
each of these vegetation units need to be investigated, analyzed and presented in a
form that will be useful for the different forms of land use that are practiced in the
Ramsar Site. It must be stressed that the current classification of the vegetation in
the Ramsar Site by Jellema, Ringrose & Matheson, (2002) at the Harry
Oppenheimer Okavango Delta Research Centre in Maun has been a very valuable
initial step in the classification and development of a vegetation map for the Ramsar
Site. However, experience gained during the development of the fire management
plan immediately indicated that the ten vegetation units that had been consolidated
from the 45 units that had been originally identified, were still too numerous and
complex for formulating management recommendations on controlled burning. It
must also be stressed that the proposed vegetation map would not only be of use for
managing fire in the Ramsar Site. It would also be of value and applicable to the
other range management practices like stocking rates, grazing and browsing
management and animal ratios of which controlled burning is an integral part of in an
overall range management program.
• It is recommended that a fire research program be initiated to determine
quantitatively the effects of type and intensity and season and frequency of burning
on the botanical composition, productivity, sustainability and biodiversity of the
different dominant plant communities occurring in the different vegetation units
identified in the fire management plan for the Ramsar Site. It is also recommended
that the program be conducted according to the following order of priority for the
different vegetation units, namely, Permanent Swamps, Seasonal Swamps, Burkea
Woodlands, Acacia Woodlands and Mopane Woodlands. This information is required
for the formulation of management recommendations for controlled burning in the
different vegetation units. In the case of the Permanent Swamps no information is
currently available on the fire ecology of the different dominant plant communities in
this vegetation unit and therefore priority must be given to research in this vegetation
unit. Cassidy (2003) has provided excellent information on the knowledge of local
communities on the effects of fire on the vegetation in the Permanent Swamps and
this knowledge could serve as the basis of hypotheses that could be tested using
adaptive management as the most effective means of conducting this research.
While fire research is necessary in the other vegetation units a less intensive
investigation is necessary and could comprise monitoring and testing the current
recommendations made in this report using a program of adaptive management with
appropriate collaborators in the different systems of land use in the Ramsar Site.
• It is recommended that the development of a simplified technique for assessing the
condition of the vegetation, particularly the grass sward, using key grass species.
The current recommendations for assessing range condition contained in this report
apply to the grass sward using a full species technique that requires a knowledge
and ability to be able to identify all the different grass species occurring in the
different vegetation units in the Ramsar Site. Simplified techniques based on key
grass species have been developed and are being used in South Africa, Namibia,
Tanzania and Kenya (Trollope et al, 2005, Trollope, 1990, Trollope & Trollope, 1999,
Trollope et al, 2000). It is recommended that procedure describe in these
publications be use to develop a simplified technique for the Ramsar Site. Regarding
techniques for assessing range condition it is also recommended that this be
extended to include the assessment of the condition of the tree and shrub
component of the vegetation in the Ramsar Site. A recent technique has been
developed by Trollope et al (2005) for this purpose in South Africa and was tested
during the current project at Tsodilo and proved very successful. It is therefore
recommended that this technique be tested and developed for use in the different
vegetation units in the Ramsar Site;
• It is recommended that the calibration for the disc pasture meter be tested in the
different vegetation units in the Ramsar Site. This technique for determining the
standing crop of herbaceous vegetation in the grass sward has proven to be very
successful in the current study but it would be scientifically appropriate to test the
general calibration that has been developed for African grasslands and savannas by
Trollope & Potgieter (1986) if this technique is going to be extensively used in the
Ramsar Site in the Future;
• Using the aforementioned vegetation map it is strongly recommended that the
condition of the grass sward and the tree and shrub vegetation be comprehensively
assessed in all the different vegetation units in the Ramsar Site. Currently this will
only be possible in the Acacia, Burkea and Mopane Woodlands and the Seasonal
Swamps because a technique needs still to be developed for the Permanent
Swamps. It must be recognized that the current assessment of the condition of the
vegetation in the aforementioned vegetation units was preliminary and was limited to
the grass sward because of the limitations of time allocated to this project. It is
therefore recommended that a more intensive assessment be conducted in the
different vegetation units and that it be extended to include an assessment of the
condition of the tree and shrub component of the vegetation. This information is
necessary to identify areas in the Ramsar Site where fire should be avoided for the
benefit of the productivity, sustainability and biodiversity of the current vegetation
and areas where controlled burning is necessary for the aforementioned reasons.
5.4.2 Basic Research
The following basic research projects need to be conducted:
• It is recommended that detailed autecological studies be conducted on the key

grass, non-grass herbaceous species and tree and shrub species in the Ramsar
Site. For effective and comprehensive fire management programs to be formulated it
is essential that these be based on a knowledge of how the key plant species, both
herbaceous and non-herbaceous grow from germination to maturity. Currently
recommendations are based on general principles of how plants grow and react to
fire but in many cases this is not sufficient. A good example of this is the deficiency
in knowledge about the autecology of papyrus (Cyperus papyrus) and reeds
(Phragmites australis) that currently precludes recommendations being made on the
appropriate fire regime that should be applied for the management of these
communities;
• Mycorrhizal fungi are ubiquitous in soil and their beneficial effects on the growth of
plants have been extensively documented. As an example: plant growth is severely
constrained by the limited availability of nutrients in the soils of Mediterranean
ecosystems. To cope with this adverse situation, plants have developed strategies
such as increased efficiency of nutrient uptake that in most cases is achieved by
means of specialized root systems. One abundant type of specialized root is the
mycorrhiza. Mycorrhiza is the term used to describe the symbiosis between a plant
and a fungus in soil. There are several types of mycorrhiza but in general they
consist of three main components, the root of the host plant itself, the hyphae that
grow inside or between the cells of the root and the external mycelium ramifying
through the soil. This external mycelium acts as an extension of the root, exploiting a
greater volume of soil beyond the nutrient depletion zone of the root. After
mycorrhiza formation, the plant often benefits from improved nutrition, particularly of
phosphorous, as a result of which growth is stimulated. At the same time the plant
supplies the fungus with the carbohydrates that are essential for its growth.
Mycorrhizas are ubiquitous geographically and may occur in most plants (Giannakis,
1997).
Since phosphorous is a limiting factor in plant growth in the Delta an investigation

into whether a symbiotic relationship between Cyperus papyrus, Phragmites,
mauritianus, Miscanthus junceus, Echinochloa stagnina and Vossia cuspidata and
mycorrhiza exists, and the role that mycorrhiza may play in the uptake of nutrients,
particularly phosphorous, may provide an insight into the ecology also the
functioning of the ecosystem in general. Fire is a significant factor in the ecological
functioning of the Okavango Delta, as in the Mediterranean-type ecosystems.
Research into the effects of fire on the biogeochemical cycles and the availability of
nutrients, particularly of phosphorous, in a relatively nutrient deficient system could
add valuable insight into the role of fire in the Delta and its effects on the functioning
and sustainability of this important wetland.
• It is recommended that research be undertaken to investigate the impact of
frequency and season of burning of Cyperus papyrus on aggradation, deposition of
bed-load sediments and channel flow in the Okavango Delta as these functions are
vital to the dynamics, functioning and survival of the Okavango Delta as one of the
most important wetlands in Africa.
• It is recommended that the flammability of Cyperus papyrus be investigated to test
the hypothesis that this species contains volatile substances that enhance its
flammability. This knowledge will assist in understanding the fire behaviour
associated with this species and enable the formulation of more effective control
measures for reducing the threat wildfires pose to tourist lodges and rural
communities located on the periphery and within the Permanent and Seasonal
Swamps in the Ramsar Site;
• It is recommended that research on the effect of frequency of burning on water
quality as related to fish die off should be conducted. What role does fire play in the
release of nutrients into the waters of the Delta, particularly in the seasonal swamps?
Does fire have any effect on the pH of the water? If so this may partially explain the
fish die-off experienced at Guma Lagoon and if burning does in fact cause increased
pH this would not only occur at Guma, but maybe the high incidence of fire in that
area may exacerbate it. What nutrients are released into the system by the
combustion process and how this does affect the breeding success of the various
fish populations? What will be the effect on oxygen levels in the water if fire is
excluded from the papyrus for extended periods? What effect does ash leaching into
the water after a fire have on oxygen levels and pH and what effect do any such
changes have on fish survival? Answers to these questions need to be investigated
and determined to test the hypothesis that these products of fire are responsible for
the die-off of fish populations after fires in the Delta. This information will provide
more clarity on the effects of fires in the Permanent and Seasonal Swamps
particularly when used by fisherman to assist in clearing channels for improved
access and setting of fishing nets in these plant communities.
• Ellery (1987) investigated peat and sudd formation in the Manchuria River system.
Floating sudds are formed from the decomposition of, and resultant gas formation in
the organic detrital layer in the rooting zone of Nymphaea cerulea and Typha
capensis which leads to increased buoyancy of the submerged detrital mat. Portions
of this mat become detached and rise to the surface to form mobile sudds that then
become colonized by several plant species. At an advanced stage in the plant

succession process species such as Miscanthus junceus may become dominant.
Rural communities in certain parts of the Delta have ascribed some channel
blockages to the reduction or complete suppression of fire whereas it could in fact be
a combination of changes in water flow, mobile papyrus islands lodging in narrow
channels, plus sudd formation and possibly a change in the fire regime that causes
these blockages. Research into the impact of fire on such plant communities and
whether in fact fire plays a role in preventing channel blockage is required.
• No information could be found on the effect of season, frequency and fire intensity
on the reptile populations, notably pythons, in the Permanent swamps. Since several
stakeholders mentioned pythons in particular, the effect of fire on species diversity in
the swamps should be investigated.
5.4.3 Scientific services
It is recommended that the following scientific services be expanded and made more
readily available to the general public in the Ramsar Site:
• It is strongly recommended that climatic and weather data be made more available
and readily accessible to the general public in the Ramsar Site. Data on rainfall,
temperature, relative humidity and wind is essential for the formulation of an effective
fire management and fire prevention program. However, extreme difficulties have
been experienced in obtaining comprehensive data sets on these climatic
parameters in Maun where much of the data like temperature and relative humidity
should be available on a daily basis. This is because it is continually needed for
assessing the current fire danger and for selecting appropriate weather conditions for
the application of effective and safe controlled burning programs. A case in point is
that it has been impossible to obtain a set of rainfall records from official sources that
extend to beyond 2004 in this study and use has had to be made of data kept by
private individuals in the community for which one has been most grateful;
• It is also strongly recommended that the number of weather stations in the Ramsar
Site be significantly expanded to provide a more comprehensive coverage of the
climatic conditions in the Ramsar Site. Currently it would appear that from the
Tawana Land Board Report: Land Use And Management Plan (2006) that weather
data are only available for Maun and Shakawe. It is recommended that consideration
be given to establishing additional weather stations for recording rainfall,
temperature, relative humidity and wind at Seronga, Gumare, Sehithwa and North
Gate on the periphery of the Delta and at appropriate and practical sites within the
Delta. It is recommended that in the latter case cooperation with the tourist lodges in
the Delta be investigated as a means of obtaining greater and more comprehensive
coverage of climatic conditions in this region of the Ramsar Site.
CHAPTER 6
6. RECOMMENDED FIRE MANAGEMENT PLAN FOR THE OKAVANGO DELTA
RAMSAR SITE
6.1 Introduction
Africa is referred to as the “Fire Continent” (Komarek, 1965) and prescribed burning has
become recognized as an important range management tool in the grassland and savanna
ecosystems of the continent. As indicated in the review of literature on the fire ecology of
these ecosystems, research investigating the effects of the fire regime on the biotic and
abiotic components of the ecosystem has been conducted in these ecological regions since
the early period of the twentieth century. This has led to a general understanding of the
effects of type and intensity of fire and season and frequency of burning on the grass and
tree components of the vegetation i.e. the effects of the fire regime. This in turn has clarified
the use of fire as a range management practice and viable prescribed burning programs
have been developed for the grassland and savanna areas used for livestock production
and for various forms of wildlife management. An important conclusion revealed in the
review of literature in this report, was the conclusion by Heinl (2005) that based on research
results the vegetation in the dryland vegetation communities in the Ramsar Site generally
reacted similarly to other grassland and savanna communities in southern Africa. The
participating authors in this report agree with this conclusion and based on their own
research and field experience on the effects and use of fire in African grasslands and
savannas, firmly believe that fire is a natural and often essential environmental factor in the
ecology of the Acacia, Burkea and Mopane Woodlands. Similarly based on assessments of
the condition of the grass sward in the Seasonal Swamps and observations on the rapid
recovery of papyrus communities after fire at Seronga and Shakawe in the Permanent
Swamps, it was concluded for these vegetation units that fire is also a natural and often
essential environmental factor of the environment. Extensive practical experience has been
gained on the use of fire in southern and east African grasslands and savannas (Trollope,
1983; Trollope, 1989; Trollope & Potgieter, 1986; van Wilgen, Everson & Trollope, 1990;
Tainton, 1999; Trollope, et al. 2000) This experience together with the available information
on the fire ecology and assessed condition of the vegetation in relation to fire in the Ramsar
Site will form the basis for the recommendations on controlled burning in the proposed fire
management plan.
The most important factors to consider when planning a controlled burning program are the
reasons for burning, the ecological and physical condition of the vegetation and the
appropriate fire regime to be applied.
6.1.2 Reasons for burning
In section 5.1.1.2 the following reasons were identified for burning in the Okavango Delta
Ramsar Site:
• to improve the quality of grazing for domestic livestock and wildlife (Cassidy,
2003, Heinl, 2005; Tacheba, 2002 and Tlotlego, 2004);
• to attract wildlife to green grazing for improved viewing by tourists (Heinl, 2005);
• to attract wildlife to green grazing for improved hunting (Cassidy, 2003; Tlotlego,
2004);
• to improve the quality of thatch grass and reeds by removing plant debris after
harvesting (Cassidy, 2003);
• to construct burnt firebreaks to safeguard property (Heinl, 2005).
• to increase fish populations in the Permanent and Seasonal Swamps by
stimulating new shoots palatable to fish and promote better nesting conditions for
fish when the flood arrives (Cassidy, 2003);
• to improve access to the Permanent and Seasonal Swamps for fishing and
setting of nets (Cassidy, 2003; Tlotlego, 2004);
• to improve the quality of papyrus for weaving of mats by removing old dead
shoots and stimulating new growth (Cassidy, 2003);
• to improve access in the Seasonal Swamps for harvesting bulbs of water lilies
(Cassidy, 2003; Tlotlego, 2004);
• to clear land in preparation for establishment of crops (Tlotlego, 2004);
The first five reasons cited for burning all involve the removal of moribund and/or unacceptable
grass or other plant material like reeds as a means of rejuvenating the plant community. This
reason is a widely recognized and accepted reason for controlled burning in African
grasslands and savannas (Tainton, 1999). Therefore whether it is applied in areas used for
domestic livestock production, wildlife management, harvesting thatch grass or reeds, it can
be regarded as an ecologically acceptable reason for burning in the Acacia, Burkea and
Mopane Woodlands and the Seasonal Swamps in the Ramsar Site This is provided the
vegetation is in a condition suitable for burning.
Another widely recognized reason for burning in African grasslands and savannas is to control
and/or prevent the encroachment of undesirable plants (Trollope, 1989) such as the
encroachment of bush in savanna areas in Botswana (Field, 1978). Therefore this will also be
regarded as an ecologically accepted reason for burning in the Ramsar Site.
These are the basic reasons for burning grassland and savanna vegetation in Africa. It has
been suggested that fire can be used to control ticks that cause tick borne diseases in
livestock but this reason is generally discounted because ticks persist in areas that are
frequently burnt. However, Stampa (1959), in a study of the Karroo Paralysis Tick in the
Karroid Merxmuellera Mountain Veld in South Africa, has shown that this parasite can be
successfully controlled by altering the micro-climate at soil level and thereby creating an
unfavourable habitat for this organism resulting in its disappearance. Similar evidence has
been obtained by Trollope & Trollope (2001) in the Ngorongoro Crater and Serengeti
grasslands in Tanzania where controlled burning by nomadic Masai pastoralists has
resulted in a significantly lower incidence of ticks where this practice is applied. The
incidence of ticks can be high when the grass sward is in a moribund and unacceptable
condition for grazing by livestock or when it is encroached by excessive densities of trees
and shrubs therefore using fire for the aforementioned permissible reasons will have the
added benefit of minimizing the incidence of ticks in areas used for livestock husbandry.
Regarding the use of fire in the Permanent Swamps, as stated earlier it has been concluded
that fire is a natural factor of the environment in this type of vegetation Both the literature and
personal observations indicate that burning is not harmful to widespread plant communities
dominated by species like Cyperus papyrus (papyrus) and the two Phragmites spp. (reeds)
because the growing points of these plants are either inundated by water or are in moist soil
protecting them from fire. Therefore burning for the aforementioned reasons in the Permanent
Swamps is ecologically acceptable in principle. However, as stated there is very little
quantitative information available on the effects of the fire regime on these plant communities
and that autecological research is required to provide information for formulating ecologically
acceptable fire management practices for the Permanent Swamps. It must also be stressed
that burning in the Permanent Swamps also has secondary effects on fauna which inhabit this
vegetation unit e.g. slaty egret, and therefore any fire management program for the
Permanent Swamps must also take these factors into consideration and not only the effect of
the fire on the plants per se.
Finally burning to clear land in preparation for establishment of crops can be regarded as an
acceptable reason for using fire as a management tool as long as it is conducted in a safe
and responsible manner.
6.1.3 Ecological criteria for prescribed burning
In section 5.2 the condition of the vegetation was assessed using the ecological criteria that
have been developed in southern and east Africa based on the general response of
grassland and savanna vegetation to season and frequency of burning and type and
intensity of fire. The condition of the grass sward was assessed according to its botanical
composition, ecological status, basal cover and standing crop of herbaceous material. This
procedure proved to be very successful in terms of being able to quantitatively determine
whether fire will have a desirable or deleterious effect on the grass sward in terms of its
productivity, sustainability and biodiversity and the impact on grazing for domestic livestock
and wildlife. It is therefore strongly recommended that this procedure for assessing range
condition be used in the Okavango Delta Ramsar Site as means of being able to determine
quantitatively what effect fire will have on the vegetation and its associated fauna in the
Acacia, Burkea and Mopane Woodlands and the Seasonal Swamps. Therefore it is
recommended that the following ecological criteria be used to assess the condition of the
grass sward in the aforementioned vegetation units to decide whether controlled burning is
necessary or not in these vegetation units in the Ramsar Site:
• Burning is ecologically acceptable if the grass sward is in a climax and/or sub-climax

stage dominated by Decreaser and/or Increaser I grass species as a means of
maintaining the potential of the grass sward to produce grazing for both domestic
livestock and wildlife. When in this ecological condition the grass sward is resistant
to the negative effects of this extreme form of defoliation. Conversely burning should
not be applied when the grass sward is in a pioneer condition dominated by
Increaser II grass species in order to allow it to develop to a more productive stage
dominated by Decreaser grass species (Trollope, 1999).
• Burning is ecologically acceptable, if the grass sward is in a moribund and/or
unpalatable condition, as a means of restoring the vigour of the grass sward and
allowing new nutritious regrowth to occur. Field experience indicates that when the
standing crop of grass >4 000 kg/ha in African grasslands and savannas then the
grass sward has become moribund and/or unacceptable to grazing animals and
needs to be defoliated by burning or some other means (Trollope, 1999).
It is recommended that the botanical technique presented in Table 6 be used for assessing
the necessity for controlled burning in the Acacia, Burkea and Mopane Woodlands and
Seasonal Swamps in the Ramsar Site. This technique has the added advantage of being
able to assess the grazing potential of the grass sward for grazing animals. In assessing
the condition of the grass sward in areas used for harvesting grass thatching material it is
recommended that the botanical technique presented in Table 7 be used.
It is recommended that the moribund and/or unpalatable condition of the grass sward be
assessed using a disc pasture meter to estimate the standing crop of grass. The use of this
apparatus is illustrated in Figure 37.
Figure 37. The Disc Pasture Meter developed by Bransby & Tainton (1977) used to
estimate the standing crop of herbaceous plant material in a grass sward.
It is recommended that the calibration developed in the Kruger National Park in South Africa
and recommended for use in estimating grass fuel loads in African grasslands and
savannas be used for this purpose in the Okavango Delta Ramsar Site. The relationship
between the settling height of the disc and the standing crop of grass material is presented
in Table 14.
Table 14. Calibration for the Disc Pasture Meter developed in the Kruger National
Park in South Africa and recommended for use in estimating the grass
fuel load in African grasslands and savannas for management purposes
(Trollope & Potgieter, 1986).
X Y X Y X Y X Y X Y X Y X Y X Y
Cm Kg/ha cm kg/ha cm kg/ha cm kg/ha cm kg/ha cm kg/ha cm kg/ha cm kg/ha
2.0 177 6.0 2517 10.0 4128 14.0 5437 18.0 6569 22.0 7581 26.0 8505 30.0 9360
2.1 256 6.1 2563 10.1 4163 14.1 5467 18.1 6596 22.1 7605 26.1 8527 30.1 9380
2.2 333 6.2 2608 10.2 4199 14.2 5497 18.2 6622 22.2 7629 26.2 8549 30.2 9401
2.3 408 6.3 2654 10.3 4234 14.3 5527 18.3 6649 22.3 7653 26.3 8571 30.3 9421
2.4 482 6.4 2698 10.4 4269 14.4 5557 18.4 6675 22.4 7677 26.4 8593 30.4 9442
2.5 554 6.5 2743 10.5 4304 14.5 5587 18.5 6702 22.5 7701 26.5 8615 30.5 9462
2.6 625 6.6 2787 10.6 4339 14.6 5616 18.6 6728 22.6 7725 26.6 8637 30.6 9483
2.7 695 6.7 2831 10.7 4374 14.7 5646 18.7 6754 22.7 7749 26.7 8659 30.7 9503
2.8 763 6.8 2874 10.8 4408 14.8 5675 18.8 6780 22.8 7772 26.8 8681 30.8 9523
2.9 830 6.9 2918 10.9 4442 14.9 5705 18.9 6806 22.9 7796 26.9 8703 30.9 9544
3.0 895 7.0 2960 11.0 4477 15.0 5734 19.0 6832 23.0 7820 27.0 8724 31.0 9564
3.1 960 7.1 3003 11.1 4511 15.1 5763 19.1 6858 23.1 7843 27.1 8746 31.1 9584
3.2 1024 7.2 3045 11.2 4544 15.2 5792 19.2 6884 23.2 7867 27.2 8768 31.2 9605
3.3 1086 7.3 3087 11.3 4578 15.3 5821 19.3 6910 23.3 7890 27.3 8789 31.3 9625
3.4 1148 7.4 3129 11.4 4612 15.4 5850 19.4 6935 23.4 7913 27.4 8811 31.4 9645
3.5 1209 7.5 3170 11.5 4645 15.5 5879 19.5 6961 23.5 7937 27.5 8833 31.5 9665
3.6 1269 7.6 3211 11.6 4678 15.6 5907 19.6 6986 23.6 7960 27.6 8854 31.6 9685
3.7 1328 7.7 3252 11.7 4711 15.7 5936 19.7 7012 23.7 7983 27.7 8876
3.8 1387 7.8 3293 11.8 4744 15.8 5964 19.8 7037 23.8 8006 27.8 8897
3.9 1444 7.9 3333 11.9 4777 15.9 5993 19.9 7063 23.9 8030 27.9 8918
4.0 1501 8.0 3373 12.0 4810 16.0 6021 20.0 7088 24.0 8053 28.0 8940
4.1 1557 8.1 3413 12.1 4842 16.1 6049 20.1 7113 24.1 8076 28.1 8961
4.2 1613 8.2 3453 12.2 4875 16.2 6077 20.2 7138 24.2 8099 28.2 8982
4.3 1667 8.3 3492 12.3 4907 16.3 6105 20.3 7164 24.3 8122 28.3 9004
4.4 1722 8.4 3531 12.4 4939 16.4 6133 20.4 7189 24.4 8145 28.4 9025
4.5 1775 8.5 3570 12.5 4971 16.5 6161 20.5 7214 24.5 8167 28.5 9046
4.6 1828 8.6 3609 12.6 5003 16.6 6189 20.6 7239 24.6 8190 28.6 9067
4.7 1881 8.7 3647 12.7 5035 16.7 6217 20.7 7263 24.7 8213 28.7 9088
4.8 1932 8.8 3685 12.8 5067 16.8 6244 20.8 7288 24.8 8236 28.8 9109
4.9 1984 8.9 3723 12.9 5098 16.9 6272 20.9 7313 24.9 8258 28.9 9130
5.0 2035 9.0 3761 13.0 5130 17.0 6299 21.0 7338 25.0 8281 29.0 9151
5.1 2085 9.1 3799 13.1 5161 17.1 6327 21.1 7362 25.1 8304 29.1 9172
5.2 2135 9.2 3836 13.2 5192 17.2 6354 21.2 7387 25.2 8326 29.2 9193
5.3 2184 9.3 3873 13.3 5223 17.3 6381 21.3 7411 25.3 8349 29.3 9214
5.4 2233 9.4 3910 13.4 5254 17.4 6408 21.4 7436 25.4 8371 29.4 9235
5.5 2281 9.5 3947 13.5 5285 17.5 6435 21.5 7460 25.5 8393 29.5 9256
5.6 2329 9.6 3983 13.6 5315 17.6 6462 21.6 7485 25.6 8416 29.6 9277
5.7 2377 9.7 4020 13.7 5346 17.7 6489 21.7 7509 25.7 8438 29.7 9297
5.8 2424 9.8 4056 13.8 5377 17.8 6516 21.8 7533 25.8 8460 29.8 9318
5.9 2471 9.9 4092 13.9 5407 17.9 6543 21.9 7557 25.9 8483 29.9 9339
Relating the use of the aforementioned ecological criteria for assessing the condition of the
vegetation for controlled burning in the Ramsar Site the following conclusions can be drawn.
The results in Figure 27 in section 5.2.3.1 indicate that only in the Seasonal Swamps and in
the burnt area at Tsodilo in the Burkea Woodlands was the grass sward dominated by
Decreaser and/or Increaser I grass species and therefore in an ecological condition that
can be considered for controlled burning. All the other vegetation units were dominated by
pioneer, annual, grass species and were therefore in an ecologically sensitive condition that
precludes the general use of fire in these vegetation units. It must be added though that the
dominance of the grass sward by annual grass species and forbs in the Acacia Woodlands
has become for all practical purposes a permanent condition. This has presumably been
caused by excessively heavy grazing on a continuous basis with domestic livestock when
these sedentary populations of domestic animals replaced the migratory wildlife populations
of the past. This conclusion is based on the results of grass surveys conducted in controlled
hunting area CT 1 located north east and adjacent to the Nxai Pan National Park where in
Acacia Woodlands similar to those in the Ramsar Site, there was a significant proportion of
perennial grass species in the grass sward. A similar situation was observed in the
Makgadikgadi Pans National Park situated immediately east of the Acacia Woodlands in
the Ramsar Site, where a significantly greater proportion and in many areas, a dominance
of perennial grass species were present in the grass sward. However, the current apparent
permanent dominance of annual grass species and forbs in the Acacia Woodlands does not
invalidate the ecological criterion excluding burning when the grass sward is dominated by
Increaser II grass species. This is because, as noted earlier, investigations by Trollope &
Trollope (2005) in Concession Area No 34 located north east of Maun, showed that
wildfires that had occurred during 2005 in an area with a sparse cover of pioneer grass
species had been very damaging. In this case the grass sward had been converted into
extensive areas devoid of herbaceous vegetation and prone to wind erosion.
Regarding the criteria related to the standing crop the results in Figure 28 section 5.2.3.1
indicate that it was only in the Seasonal Swamps that the grass fuel load was >4 000 kg/ha
and therefore could be considered for controlled burning. These results clearly illustrate the
ability of these ecological criteria to determine what areas in the Ramsar Site should be
considered for burning and what areas should be excluded from burning based on the
condition of the grass sward. These results have also attended to the requirement in the
Terms of Reference of identifying ecologically sensitive zones in the Ramsar Site where
controlled burning should be excluded and measures taken to prevent and control wildfires.
In conclusion it must be cautioned though that the results from the assessment of the
condition of the grass sward in the different vegetation units are preliminary because time
did not permit a more exhaustive assessment of the overall condition of the grass sward in
the Ramsar Site. However, it is firmly believed that the results do provide a general
description of the condition of the vegetation in relation to fire in these vegetation units.
6.1.4 Fire regime
The following fire regime in terms of type and intensity of fire and the season and frequency of
burning is recommended for the Ramsar Site.
6.1.4.1 Type of fire
It is recommended that fires burning with the wind either as surface head fires in grassland
or a combination of surface head fires and crown fires in tree and shrub vegetation be used
in controlled burning. This is because surface head fires cause least damage to the grass
sward and crown fires can cause maximum damage to woody vegetation when fire is used
to control bush encroachment (Trollope, 1999).
6.1.4.2 Fire intensity
Research on fire behaviour in the Eastern Cape Province and Kruger National Park in South
Africa has shown that fire can be classified into the following categories according to fire
intensity (Trollope, 1983; Trollope and Potgieter, 1985).
Fire Intensity (kJ/s/m Description

<500 Very cool
501 – 1 000 Cool
1 001 – 2 000 Moderately hot
2 001 – 3 000 Hot
>3 000 Extremely hot
When burning to remove moribund and/or unacceptable grass material a cool fire of <1 000
kJ/s/m is recommended. This can be achieved by burning when the air temperature is <20°C
and the relative humidity >50 %. When burning to control undesirable plants like encroaching
bush, a hot fire of >2 000 kJ/s/m is necessary. This can be achieved when the grass fuel load
is >4 000 kg/ha, the air temperature is >25°C and the relative humidity < 30 %. This will cause
a significant topkill of stems and branches of bush species up to a height of 3 m. In all cases
the wind speed should not exceed 20 km/h.
6.1.4.3 Season of burning
Research in southern Africa has clearly indicated that the least damage is caused to the grass
sward if controlled burning is applied when the grass is dormant. Relating this principle to the
different vegetation units in the Ramsar Site it is recommended that if it is necessary to burn
areas based on the condition of the vegetation in either the Burkea, Acacia or Mopane
Woodlands (currently most probably the Burkea Woodlands) where plant growth is dependent
only on rainfall then these areas should be burnt at the end of the dormant winter season in
approximately October. When burning is used to remove moribund and/or unpalatable grass
material then preferably it should be applied after the first spring rains of >13mm to ensure
adequate moisture for regrowth to take place in the grass sward. Burning after rain will also
ensure that the fire intensity and danger are low, reducing the potential for the fire to escape
and become a wildfire. When burning is used to control the encroachment of undesirable
plants like bush encroachment, a high intensity fire is required and it is recommended that this
be applied before the first spring rains in August/September when it is extremely hot and dry.
This will ensure the development of a high intensity fire required to control the encroaching
plants.
In the case of burning in the Seasonal Swamps a different scenario exists in that the growth of
the vegetation is generally influenced to a greater degree by the annual flood waters in the
Delta, entering the Okavango River at Mohembo, than by rainfall. This generally occurs
between March and May (see Figure 6) causing the Seasonal Swamps to be flooded by June
and July. The ideal burning window in this vegetation unit for removing moribund and/or
unpalatable grass material is therefore to burn during the period May to July, applying the fires
when the grass sward is dormant before the flood waters start rising. This will ensure that the
burnt areas are subsequently flooded and the grass sward will recover rapidly when the flood
waters recede after July, stimulated by the high levels of moisture in the soil. A safety factor of
burning during this early winter period in the Seasonal Swamps is that there will be a
significant number of areas that are wet and where the vegetation is still green. This will
prevent extensive contiguous areas being burnt resulting in the development of a diverse fire
mosaic of burnt and unburnt areas which is ideal for the maintenance and promotion of
biodiversity. If it should be necessary to reduce the growth of trees and shrubs in the Seasonal
Swamps then this can be achieved by burning later in the winter during August/September
when it is extremely hot and dry thereby ensuring high intensity fires necessary to control
encroaching trees and shrubs. Extreme care will have to be exercised with these controlled
burns to avoid the fires escaping and becoming wildfires.
6.1.4.4 Frequency of burning
When burning to remove moribund and/or unacceptable grass material the frequency of
burning will depend upon the accumulation rate of excess grass litter (Trollope, 1999). Field
experience indicates that this should not exceed 4 000 kg/ha and therefore the frequency of
burning should be based on the rate at which this phytomass of grass material
accumulates. This approach has the advantage that the frequency of burning is related to
the stocking rate of grazers and to the amount of moisture the area receives. Therefore in
the dryland areas like the Burkea Woodlands, the required frequency of burning based on
the accumulation of excessive grass fuel >4 000 kg/ha loads, will be significantly lower than
in the Seasonal Swamps that receive markedly greater levels of moisture from the annual
flood waters in the Delta.
6.1.4.5 Post-fire range management
Grazing after burning in the Ramsar Site will be difficult to control because in both the areas
used for domestic livestock and wildlife there is open access to grazing after a fire. In order
to prevent overgrazing it is important to ensure that the burnt area exceeds the short term
forage requirements of the grazing animals that are attracted to the highly palatable and
nutritious regrowth that develops after a burn i.e. burn relatively large areas at any one time
(Trollope, 1990). Another strategy that has been successfully used in southern Africa is to
apply a series of patch burns at regular intervals throughout the duration of the burning
window during the dormant season. This has the effect of attracting the grazing animals to
the newly burnt areas after the different fires thereby spreading the impact of grazing over
the entire burnt area and avoiding the detrimental impacts of heavy continuous grazing after
the burns (Brockett et. al.2001).
6.2 Application of a Controlled Burning Program
6.2.1 Fire Danger Index for controlled burning
Assuming that the ecological requirements for applying a controlled burn have been met
and the reasons for burning are clearly understood the first step in the application of a
controlled burn is to assess the weather conditions in relation to being able to apply an
appropriate and safe controlled burn. For this purpose Johan and Lizette Heine from the
Forest Fire Association at Nelspruit in Mphumulanga Province in South Africa, have
developed a practical and effective Fire Danger Rating System for assessing the weather
conditions relative to the potential fire behaviour that can be expected during a burn. This
system was developed for use in controlling and suppressing wildfires in commercial
forestry operations in South Africa. However, as part of the recommendations for controlled
burning in the Serengeti National Park this system has been adapted for use in the
application of a controlled burning program. During the conducting of the fire workshop in
June 2005 in the Park, the system proved very effective in describing the potential fire
behaviour of the controlled burns that were applied during the field exercises. The ranger
staff attending the workshop readily mastered the procedures for calculating the Fire
Danger Index (FDI) in the field, clearly demonstrating that it was a practical and valuable
innovation to adopt and apply during controlled burning in the Serengeti National Park.
The Fire Danger Index (FDI) is calculated by considering the combined effects of air
temperature, relative humidity, wind speed, rainfall and the degree of curing of the grass
fuel on fire behaviour as described by the Burning Index. The following steps are used to
calculate the Fire Danger Index for controlled burning:
Step 1: Estimate the Burning Index using the following chart to relate the effects of
Air Temperature and Relative Humidity on the Burning Index.
AIR TEMPERATURE/ RELATIVE HUMIDITY vs BURNING INDEX
Step 2: Adjust the Burning Index using the following chart that relates the effect of wind
speed expressed as the Wind Correction Factor on the burning index. Add this
correction to the Burning Index.
WIND CORRECTION FACTOR vs BURNING INDEX
WIND WIND
CORRECTION CORRECTION
SPEED SPEED
FACTOR FACTOR
km/h
km/h
0 0 24 15
1 0 25 16
2 4 26 19
3 5 27 20
4 5 28 20
5 6 29 20
6 9 30 20
7 10 31 20
8 10 32 21
9 10 33 25
10 10 34 25
11 10 35 25
12 10 36 26
13 10 37 29
14 10 38 30
15 10 39 30
16 11 40 30
17 14 41 31
18 15 42 34
19 15 43 35
20 15 44 35
21 15 45 36
22 15 46 40
23 15
Step 3: Adjust the Burning Index using the following chart that relates the effect of the
degree of curing of the grass fuel expressed as the Grass Curing Factor on the
burning index. Multiply the Burning Index with this factor.
PERCENTAGE CURING OF GRASS SWARD GRASS CURING FACTOR

0 0
10 0.1
20 0.2
30 0.3
40 0.4
50 0.5
60 0.6
70 0.7
80 0.8
90 0.9
100 1.0
Step 5: This final adjustment of the Burning Index using the Grass Curing Factor then
becomes the Fire Danger Index (FDI) used for controlled burning for different
management purposes i.e.
FDI = Burning Index [(Air Temperature/Relative Humidity) + (Wind Correction

Factor)] x (Grass Curing Factor).
The Fire Danger Index (FDI) together with related fire intensities can be used to provide
guidelines for controlled burning when used for:
• removing moribund and/or unacceptable grass material;
• creating and/or maintaining an optimum balance between herbaceous and woody
vegetation;
• constructing burnt firebreaks.
The guidelines for controlled burning using different Fire Danger Indices are presented in
Table 15.
Table 15 Fire Danger Rating System using Fire Danger Indices (FDI’s) as a means for
selecting suitable burning conditions for controlled burning.
6.2.1.1 Practical example of the calculation of the Fire Danger Index (FDI)
i) Calculate the FDI for the following weather and grass fuel conditions:
• Air temperature = 280C;
• Relative humidity = 24%;
• Wind speed = 12 km/h;
• Grass curing = 100%.
FDI = Burning Index [Air Temperature/ Relative Humidity) + (Wind Correction Factor)] x
(Grass Curing Factor);
FDI = [51 + 10)] x (1);

FDI = 61;
FDI = Orange;
FDI = Very dangerous;
FDI = 2000-3000 kJ/s/m;
FDI = Burn to control bush encroachment.
ii) Calculate the FDI for the following weather and grass fuel conditions:
• Air temperature = 22 0C;
• Relative humidity = 44%;
• Wind speed = 10 km/h;
• Grass curing = 30%.
FDI = [37 + 10)] x (0.3);

FDI = 11;
FDI = Blue;
FDI = Safe;
FDI = <500 kJ/s/m;
FDI = Too cold, humid or wet for controlled burning.
6.2.1.2 Vegetation monitoring
In the proceedings of the ODMP Inception Report Workshop held in November 2004 under
the Vegetation Resources Management Component the overall objective of this component
was stated as “sustainable management of the Okavango vegetation, initiated and
supported by providing accurate data and assisting in resolving vegetation management
conflicts”. As stated by Bonyongo et al (2000) vegetation description is the starting point of
both small and large scale vegetation research. Vegetation description aims to enable
people, other that the observer, to build a mental picture of the area and its vegetation and
to allow comparison and classification of different vegetation units. The description of
vegetation, with or without concurrent recording of factors of the environment, has played a
major part in the development of plant ecology and continues to be important. Vegetation
description is, therefore, an essential and integral part of vegetation science as it provides a
scientific inventory for conservation and monitoring purposes as well as for future research.
To achieve these objectives it is recommended that monitoring sites be established in the

four main vegetation units as identified by the consultants as being practical units in
management terms, ie the Acacia, Burkea and Mopane Woodlands and the Seasonal
Swamps. Annual Range Condition Assessment surveys, using the rapid, simple but
effective technology introduced in this consultancy, would facilitate the objective of
sustainable vegetation management in the Okavango Delta Ramsar Site and provide
accurate data on the annual “vegetation production”. This is an achievable objective as
such a monitoring system was instituted in the Kruger National Park in South Africa in 1987
where 22 game rangers monitor 530 sites located in the different vegetation units in the
Park on an annual basis. This monitoring program has been collecting data since its
inception ie. 19 years, and the Kruger National Park now has one of the most extensive and
integrated vegetation databases, certainly in Africa, that is analyzed annually and reports
are generated using a specially designed computer program, for management purposes.
The reports are very valuable in planning fire management strategies and such information
could be extrapolated for use in the sustainable use and management of the rangeland for
livestock and wildlife purposes and will also provide information for strengthening
institutional and stakeholder capacity.
6.2.2 Burning procedure
There are basically two types of ignition procedures used in the application of controlled burns
viz. Block Burns and Patch Mosaic Burns.
6.2.2.1 Block burns
Block burns are the most widely used procedure where the area to be burnt is ignited around
the perimeter and allowed to burn out as a so-called block burn. The procedure to be followed
when applying a block burn is illustrated in Figure 38.
Starting Point
A
Back Fire Back Fire
Head Fire
Head Fire Back Fire
Head Fire
Head Fire
Head Fire Back Fire

C
Wind Direction
Figure 38. The procedure for using a perimeter ignition to apply a block burn.
The procedure for applying block burns illustrated in Figure 38 involves the following steps:
• Initiate simultaneously two surface back fires burning against the wind along two lines
commencing at the starting point and proceeding to corners A and B of the burn area.
The resultant back fire will be a slow moving burn characterized by low flames and is
relatively easy to control;
• Allow the back fires to burn back until an adequate fire break has been established;
• Initiate a surface head fire burning with the wind, by proceeding from points A and B to
point C, simultaneously and as swiftly as possible. The resultant head fire will be a fast
moving burn characterized by high flames that on meeting the back fire will cause both
flaming fronts to assume a vertical angle thereby preventing the fire from spreading
further.
This method of applying a controlled burn is the most frequently used ignition procedure in the
application of controlled burns. It is widely used in extensive land management operations
such as national parks, where it is necessary to limit controlled burns to pre-determined areas
and greatly reduces the risk of losing control of controlled burns.
6.2.2.2 Patch mosaic burns
The patch mosaic burning system was developed and is being successfully used in the
summer rainfall areas of the moist savannas in the Pilansberg National Park in South Africa
and has now become the preferred and recommended burning procedure in the Kruger
National Park. The basic idea of patch mosaic burning originates from Australia where it is
applied in Uluru and Kakadu National Parks (Brockett et. al. 2001). The assumption in this
burning procedure is that the burning patterns of fires are an effective surrogate measure of
biodiversity i.e. the more diverse the burning patterns the higher the biodiversity. Diverse
burning patterns are developed by applying point ignitions rather than perimeter ignitions in
the area to be burnt. The resultant fires increase in size in all directions spreading as head
fires with the wind, back fires against the wind and flank fires burning as intermediate fires
between head and back fires in relation to the direction of the wind (see Figure 39). It is
assumed that the resultant diverse fire mosaic will result in the development of a
heterogeneous mosaic of species and structural diversity in the vegetation thereby
promoting and/or maintaining biodiversity (Brockett et al, 2001; van Wilgen et. al. 1998).
Figure 39. Illustration of head, back and flank fires resulting from a point ignition
(Adapted from Cheney & Sullivan (1997)).
The general procedure for initiating a patch mosaic burn from a point ignition is illustrated in
Figure 40.
Figure: 40A Figure 40B
Figure 40. The procedure for using a point ignition for applying a patch mosaic burn;
Figure 40A: When the point ignitions are initiated at suitable intervals one
third of the distance down from the most up wind point of the area being
burnt; Figure 40B: When the point ignitions are initiated at suitable
intervals along the perimeter of the up wind portions of the area being
burnt.
The different steps involved in the application of a patch mosaic burn are:
• Ensure that the down wind side of the area to be burnt has a suitable fire break in the
form of either a constructed fire break, a road or natural features of the landscape like a
stream, river or cliff that can serve as an effective fire break during the application of
the controlled burn;
• Commencing from the most down wind point on the perimeter of the area to be burnt
(point A), initiate simultaneously two surface back fires burning against the wind along
two lines stopping at the points on the perimeter where the fires are becoming flank
and/or head fires;
• The resultant back fire will be a slow moving burn characterized by low flames and is
relatively easy to control;
• Allow the back fire to burn back until an adequate fire break has been established;
• Once the down wind fire break is deemed adequate apply one or more point ignition
fires at suitable intervals in a straight line at right angles to the direction of the wind
across the unburnt area. These point ignitions can be initiated at suitable intervals
commencing approximately one third of the distance down from the most up wind point
of the area being burnt. Alternatively the one or more point ignitions can be initiated at
suitable intervals along the perimeter of the up wind portions of the area being burnt.
This latter method is the safest procedure for the field staff to use because it involves
always initiating the fire from an established fire break;
• The application of patch mosaic burns is significantly more dangerous to apply and
more difficult to control than block burns. This is because the patch fires burn for a
longer period of time thereby increasing the probability of the direction of the wind
changing significantly. Such changes in the wind direction can render the original fire
break established with the initial back fire ineffective thereby requiring additional back
fires to be applied on the perimeter to ensure the security of the area being burnt. The
longer burning period can also provide the opportunity for radical changes in the
weather conditions involving air temperature, relative humidity and wind speed
necessitating additional control measures to be resorted to;
• The number and distance between the point ignitions to be applied will vary according
to the size of the area being burnt and the weather conditions under which the
controlled burns are being applied. Generally speaking, the point ignitions should not
be initiated less than 100 metres apart and patch mosaic burning should ideally be
applied preferably when the fire danger index is 21 - 45 indicating a green day.
6.2.3 Firebreaks
Firebreaks are natural or constructed barriers which are used to stop or check fires and/or
provide a control line from which to apply fires (Teie, 2003). As a fundamental general rule
controlled burning must not be conducted without adequate firebreaks having been prepared
prior to initiating the controlled burn. The lack of effective firebreaks is one of the most
important reasons for fires getting out of control during controlled burning. Furthermore the
practice of combining controlled burning with the simultaneous burning of a firebreak is very
dangerous and must be avoided. This is because if the wind direction changes it is not readily
possible to apply a back fire to contain the already initiated fire front resulting in an
uncontrolled burn with possible disastrous results.
There are basically three types of firebreaks viz. natural firebreaks like roads, streams, rivers
and cliffs, burnt firebreaks and clean cultivated firebreaks. However, in the case of the
Okavango Delta Ramsar Site it is recommended that where possible only natural and burnt
firebreaks should be considered because clean cultivated fire breaks are ecologically unstable
and prone to wind erosion.
6.2.3.1 Dry and wet line firebreaks
Burnt firebreaks comprise creating a burnt strip of vegetation around the perimeter of the area
to be burnt and can be constructed in several ways. The most common method is to initiate a
back fire by igniting a continuous line of fire for 4 to 5 metres along the perimeter of the area to
be burnt. Then with the aid of beaters or water sprayers extinguish the head fire portion of the
line of fire on the leeward side, allowing the back fire to slowly spread back against the wind.
Once the back fire has spread back adequately, a 4 to 5 metre long head fire is ignited 2 to 3
metres parallel to the back fire resulting in a rapid burning off of the intervening area. The
remaining back fire on the upper side of the burnt strip is then extinguished before repeating
the whole procedure. The resultant firebreak can be widened if necessary by igniting
additional head fires parallel to the previous head fire. This process is repeated until a burnt
strip has been created around the area to be burnt or for the distance required for applying a
controlled burn. When this type of fire break is constructed using only beaters it is called a
“Dry Line Fire Break” and when water sprayers are used it is referred to as a “Wet Line Fire
Break”. The aforementioned procedure for constructing dry and wet line firebreaks is
illustrated in Figure 41.
Figure 41. Procedure for constructing dry and wet line firebreaks.
The advantages of this method are that it requires a minimum of equipment and no prior
preparation. The wet line procedure is the safest to use for constructing burnt fire breaks
compared to the dry line procedure but has the disadvantage of requiring an adequate and
constant supply of water and is labour demanding.
6.2.3.2 Cut line firebreaks
Another very effective method of constructing burnt fire breaks is to cut two strips around the
perimeter of the area to be burnt with a mower, leaving an uncut strip in the centre which is
then burnt. This is a relatively safe procedure because the cut portions on either side of the
uncut strip reduce the intensity of the fire, making the fire easier to control. This is a more rapid
and less labour demanding method of constructing burnt fire breaks but requires adequate
machinery, prior preparation in the form of mowing and is limited to flat terrain.
6.2.3.3 Tracer line firebreaks
A similar method to the cut line firebreaks is to spray two strips of the grass sward around the
perimeter of the area to be burnt with a foliar weedicide towards the end of the growing
season. The two sprayed strips are then burnt when the aerial portions of the grass sward die
off and become flammable. The unsprayed area between the spayed strips is then burnt off
during the dry season when the grass sward is dormant and flammable. The advantages of
this method are that it is relatively safe and requires less labour. However, it is a time
consuming method requiring several discontinuous operations. This method is not

recommended for use in the Delta because of the possibility of polluting the water and the
resultant negative effects on the ecosystem, however it may be considered for use in the
dryland portions of the Ramsar Site away from the Delta itself.
6.2.3.4 The Botha Fire-Box
This method was developed by a South African farmer Mr Andries Botha, the apparatus was
imaginatively adapted and improved by Mr Fungo, the Workshop Manager at the
headquarters of the Serengeti National Park at Seronera.
The technique comprises constructing an open box-like structure with four sheets of
corrugated iron fitted with four wooden handles (see Figure 42).
“The Botha Fire-Box”
Hinges
Metal
rod
Wooden
Handles
Corrugated Iron
Figure 42. The Botha Fire-Box comprising four sheets of corrugated iron fitted with
four wooden handles.
The grass material is set alight around the inside perimeter of the "fire-box” causing a vortex of
hot air to rise resulting in a hot, clean burn. At the completion of the burn the "fire-box" is
moved forward and the same process repeated. The advantages of this method are that it
permits burning fire breaks under extremely windy and dangerous conditions, it is easy and
inexpensive to construct, it is well suited to broken topography and stony terrain and is very
labour efficient. The main disadvantage of this method is that it is slow in comparison to the
other burning techniques, but this is offset by the greater degree of safety.
6.2.3.5 `Width of firebreak
It is not possible to state how wide a firebreak should be because its effectiveness is
influenced by factors such as topography, the intensity of the fire and the velocity of the wind.
Furthermore it must be appreciated that the role of a firebreak is to provide a safe point from
which to lay a line of fire and not to check the advance of a rapidly moving fire front. Therefore
based on this principle, the firebreak need only be wide enough to contain the flames during
the application of the back fire during the initial stages of applying a prescribed burn. This is
because when the head fire is applied the flames are blown away from the firebreak into the
area to be burnt. Generally when burning grassland a firebreak of 2 - 3 m wide is adequate.
However when burning highly inflammable shrub vegetation a firebreak of 4 - 6 m is
necessary.
6.2.4 Burning equipment
In the application of a safe and effective controlled burn there are certain basic items of
equipment necessary for the operation. The equipment can be divided into two categories viz.
equipment for initiating and containing a fire.
6.2.4.1 Equipment for initiating a fire
Numerous techniques are used for initiating fires ranging from a matchstick to maize cobs
attached to a length of wire and saturated with diesel fuel. The most effective method is the
"drip torch" illustrated in Figure 43.
Figure 43. A drip torch used for laying fire lines.
A fuel mixture comprising two parts diesel fuel and one part petrol (2/3 + 1/3) is allowed to drip
out of a slit type aperture at the end of the pipe. The dripping fuel is ignited and when it falls to
the ground or onto vegetation it burns for approximately 20 seconds setting the vegetation
alight with a continuous line of fire. The advantages of this technique are:
• A continuous line of fire can be set very rapidly, significantly reducing the time required
to apply a controlled burn;
• It is very efficient and effective for swiftly laying a back fire during an emergency;
• It is very effective when setting shrub vegetation alight because the flaming drops of
fuel burn for a sufficient period for the shrubs to ignite.
6.2.4.2 Equipment for controlling and fighting fires
Fire fighting crews are as effective as the equipment they are issued with. Fires radiate heat
and when a fire fighter comes into contact with the heat he/she is adversely affected, the
personal protective clothing that he/she is wearing will determine whether he/she is able to
extinguish the fire or whether the fire will continue spreading uncontrollably. Tools play a big
role in extinguishing the fire and if the fire fighter is poorly equipped the fires will also
continue spreading. (The following recommendations are based on the Working on Fire
program Best Operating Procedures).
6.2.4.2.1 Personal protective clothing (PPC)
The safety and effectiveness of the fire fighter or fire manager is to a great extent
dependant on the clothing worn when fighting fires. If, for example, fire fighters are sent to
fight fires wearing polyester type overalls, the chances of suffering from burns is far greater
than if they are equipped with 100% cotton uniforms. Long sleeves and long pants offer
more protection and will allow fire personnel to get closer to a fire. Helmets with visors will
also allow fire personnel to be more effective in fighting fires as they will be able to get
closer to the fire and not feel the direct effects of the heat on their faces. Communities and
stakeholders during the PRA exercise reported damage to their eyelids when fighting
wildfires and the level of protection provided by the visors will reduce the discomfort and
injuries.
The colour of the uniform plays a vital role during fire fighting. Yellow is the colour used
internationally for shirts as it is the most visible along the fire line. Red and orange are too
“fire” in colour and persons easily become lost in all the flames and smoke on a fire line.
Visible fire fighters are safer fire fighters.
i) Shirt and Pants

Will be made from 100% cotton material. The shirt will be long sleeve and yellow
in colour and the pants will be long and will be either blue or green in colour.
Each fire fighter will wear a cotton t-shirt as a second layer under the long
sleeved shirt.
ii) Leather Boots

The boots will be 100% leather and have no steel, rubber or plastic inclusions as
these are affected by heat and will result in injury to the fire fighter.
iii) Helmet and Visor

Veld fire helmet with heat shield.
iv) Leather Gloves:

Leather gloves containing no rubber or plastic.
v) Water Bottle and Webb Belt:

Minimum of a 1 liter water bottle to provide fresh drinking water while on fire
fighting operations and guard against dehydration.
vi) First Aid Kit:

Each crew requires a first aid kit that has a supply of Burn Shield medication
which is very effective in reducing the effects of a burn injury.
6.2.4.2.2 Fire fighting tools:
Fire fighters are as effective as the tools they are provided with. A branch from a tree will be
as effective as the time it takes for the leaves to fall off and a sack is as effective as the
water available to keep it moist. The recommended tools have been developed over time
and have proven to be the most effective for fighting fires in predominantly grass and shrub
vegetation.
i) Fire Beater
This is used to extinguish a fire by removing the oxygen available to the fire. This
tool is manufactured from conveyor belt and has a wooden or a metal handle. It is
a very effective tool to control grass fires.
ii) Knap Sack Pump

Metabi back pack type water pump.20 litre capacity. The hand pump is double
action type system.
iii) Hand Held Weather Device

This device measures the on site wind speed, relative humidity, and temperature,
measurements that are necessary for calculating the Fire Danger Index.
iv) Fyre Fix Wetter

This is a wetting agent applied to all water used for fire fighting units, hand
operated or slip on vehicle drawn units. Active ingredient: Anionic acrylamide
copolymer dispersed in a light mineral oil 500gm/L.
v) Vehicle Slip on Fire Fighting Units

There are two types of fire fighting units available with capacities varying from
350L, 750L to 1000L, these are transported on the back of light delivery vehicles
and/or trucks. The units are also configured to be drawn behind tractors. They
have a fiberglass or plastic water container with a pump system that either
delivers low pressure high volumes (the impeller type pump) or high pressure low
volumes (the piston type pump). Both these units are powered by either a 2
stroke or 4 stroke petrol or diesel engines. Water delivery is through a single or
double 25mm hose with a length of 50m. The Southern African versions are
known as “Bakkie Sakkies”.
6.2.4.2.3 Communications:
Communications play an important role in fire fighting or fire management and fire safety
during operations. All crews must be in communication with the relevant Incident
Commander as well as the other crews and resources at the incident.
i) Radios:
All crews need radios to be able to communicate between the crews as well as
with the Incident Commander. All mobile resources need to have radio
communications.
ii) Frequencies:
All the radios being used for fire management within the Ramsar Site require one
common fire frequency. This will enable all the crews to operate effectively and
safely. This frequency would be administered by the District Committee.
iii) Land Telephone Line:

This form of communication may be used if available to notify the Coordinating
Centre of the fire and to request assistance.
iv) Cellular Phones:

These may also be used to communicate with the District Coordination Centre
and other land users for assistance. Cellular phones have been used where a
signal exists instead of hand held radios along fire lines. The limited signal in
certain areas of the Ramsar Site would limit this method.
6.3 Application of Fire Management Plan for the Okavango Ramsar Site
Fire management in the Ramsar Site needs to be implemented in a coordinated manner

and include all the role players within the Ramsar Site. This could be achieved through a
District Fire Committee that would cater for all the diverse requirements for fire
management and which would facilitate pooling of resources and skills for the benefit of all
role players working together with the goal of effective fire management.
6.3.1 District Fire Committee
This Committee would be created through the existing District Committee and it would be
delegated with the function of fire management and other related management deemed
necessary within the Okavango Delta Ramsar Site. Membership of this Committee would
be mandatory for all the role players within the Ramsar site and include local communities,
concessionaires, Government Departments, Botswana Defense Force, commercial fishing
organizations, tourist operators, Botswana police and other land users within the Ramsar
site.
The Committee would be responsible for developing the following:

• Determining the minimum rules for members in respect of equipment and personal
for fire management
• Placement strategic of fire breaks
• Providing for a “Fire Ranger” to implement the Committee’s objectives
• Creating and implementing a District Fire Coordination Centre
• Providing for extended attack fire fighting resources through the responsible
government departments
• Supplying a daily Fire Danger Index to members during the fire season
• Levying a seasonal membership fee on members to finance the required fire
management
• Delegating the issuing of fire permits
• Coordinating training for members
• Ensuring members are kept abreast of any technology developments in the field of
fire management
6.3.2 District fire coordination centre
The coordination would be performed by a person appointed for the duration of the fire
season by the District Fire Committee within the Ramsar Site and who would be available
for the months of April – October. This person would be based in Maun and be available 24
hours a day, either at the coordination center, or after hours on a mobile phone or radio.
The appointed coordinator would be responsible for the coordination of all resources to
assist with member’s fire suppression or application of controlled burning where necessary.
They would also source and coordinate any additional resources required through the
disaster management system in Botswana. There are existing facilities in Maun that could
be utilized as a coordination center, the possibility of using the current community
emergency system or the local police station in Maun would have to be investigated.
The coordinator would also be responsible for compiling a fire database and reporting to the
District Fire Committee and the Department of Forestry and Range Resources.
6.3.3 Fire Prevention Plan
The remoteness, lack of road infrastructure and sparse population holds numerous
challenges to any fire management plan for the Okavango Delta Ramsar Site. These
challenges will require a single plan which all role players subscribe to. Land users will
need to work together in a coordinated way to overcome the remoteness and the lack of
large numbers of resources. The emphasis must be on quality and not quantity. Five well-
trained and equipped fire fighters are better than 50 untrained and poorly equipped
persons. Two land users pooling their 6 fire fighters would create a 12-person crew which is
an effective size hand crew for fighting veld fires.
6.3.3.1 Strategic fire breaks
The majority of fire breaks within the Ramsar Site are the Government fire breaks which
have been constructed throughout Botswana and are maintained mechanically on an
annual basis. A number of these fire breaks serve no purpose in stopping the spread of
fires as they are situated in a east-west direction and should rather be in a north-south
direction so as to cut across the prevailing direction of the winds which drive the fires in the
season.
The method of maintaining these breaks could be altered from mechanical maintenance to
the use of fire. In order to be environmentally “acceptable” burning strip firebreaks could be
rotated annually. This method is environmentally friendly and could also be used as a
poverty relief exercise whereby a number of employment opportunities could be created.
These existing fire breaks could be augmented where threats exist by creating buffers
through the early burning of fuels, the buffers would be burnt out using game trails and
anchoring onto areas which lack fuels. The Kalahari is known to experience high levels of
relative humidity at night during certain times of the year. These high levels of humidity
would assist in allowing the planned fires to go out naturally at night. This method is being
used in the Serengetti National Park in Tanzania and the Kruger National Park in South
Africa.
The responsibility for these breaks should remain department responsible. Any changes to
existing breaks should be through a process of consultation with the local District Fire
Committee. This will allow for local knowledge to guide any changes.
6.3.3.2 Asset protection fire breaks
All assets and structures within the Ramsar Site must have adequate fire breaks prepared
to protect against the threat of fire. The respective owners of these structures will be
responsible for preparing the required fire breaks. Thatched lodges should have some form
of water sprinkler system to protect the thatch from catching fire during fire incidents. Under
adverse fire weather conditions grass fires are recorded to have spotted up to 100m (Bush
Fire Safety Webster). No fire break is wide enough to stop a fire under adverse weather
conditions. For increased safety buildings should be kept clear of all overhanging bush and
vegetation. Populations of papyrus and Phragmites in close proximity to buildings should
regularly be burnt every 200 days, where possible, to reduce the fuel load.
6.3.4 Fire fighting capabilities
6.3.4.1 Coordination of fire fighting
The remoteness and lack of adequate road infrastructure and the sparseness of the
population within the Okavango Delta Ramsar Site pose numerous challenges to any fire
management plan. To overcome these challenges a single plan to coordinate all actions
and role players within the Ramsar Site is required. All land users would need to work
together for the fire management plan to be effective and to over come the lack of
resources. The development of a centralized coordination system at district level in Maun
through the District Fire Committee appointed by the Herbage Preservation Act would
alleviate this situation. The Herbage Preservation Committee would divide the Ramsar Site
into fire wards. These sub district wards would operate under the Herbage Preservation
Committee authority based at the Department of Forestry and Range Resources in Maun.
Sub district wards would be responsible to develop local capacity for local fire management
and fire suppression and for reporting back to the District Fire Committee who would report
to the Department of Forestry and Range Resources. The sub district wards would also
coordinate resources and actions within the ward boundaries. Membership of these wards
would be mandatory to all land users. The strategic redeployment of existing centralized
Government fire fighting resources and personnel to sub district wards would facilitate rapid
and efficient initial attack on wildfires. The sub district wards must have a designated
person or persons trained to control large fire incidents within the sub district ward and to
maintain all fire fighting equipment that must be on stand by in good working order during
the fire season. See map indicating the sub-division of the Ramsar Site into fire sub-district
wards at Seronga, Shakawe, Tsau and Maun and the location of the Fire Coordination
Centre in Maun.
6.3.4.1.1 Initial Attack: (Rapid Response)
This is the first response to a fire once it has been detected. It would be the responsibility of
the individual land user, lease holder and/or concession holder who would provide the initial
personnel and resources to attack the fire in its early stages. This capacity would be
augmented by the immediate neighbors who would be coordinated through the district
coordination centre and the local sub district fire ward. Local lease holders and/or
concession holders would be required to have a minimum of 6 persons trained as fire
fighters and they would be equipped with the necessary equipment to enable them to
perform their function as initial attack fire fighters on a wildfire. The emphasis would be on
quality and not quantity.
The sub district Government fire crews would be responsible for fire fighting on all
government land.
6.3.4.1.2 Extended Attack:
An extended attack fire evolves from an initial attack fire situation. This is when the fire
outstrips the existing resources and additional resources are required. Extended attack fires
burn for longer than 8 hours, possibly for many days and require multiple resources.
Extended attack resources in the Ramsar Site would be primarily provided by the various
sub district fire fighting crews and crews from other sub fire district wards coordinated
through the district coordination centre in Maun.
6.3.4.1.3 Aerial Coordination:
The existence of the extensive network of runways and the availability of aircraft within the
Ramsar Site makes it viable to create a capacity to assist with fire suppression operations
and fire management in the Ramsar Site. The landscape is relatively flat, other than the
Tsodilo Hills area. The flat terrain creates problems when trying to coordinate fire fighting
operations from the ground. The use of a fixed wing aircraft would greatly assist the ground
crews during this time by directing operations during daylight hours. It is recommended that
3 - 4 persons qualified as pilots and operating in the Ramsar Site be also trained in the
tactics of aerial fire fighting. EnviroNET Solutions would supply this training at a cost to be
determined. The cost of the aircraft would be covered by the annual levy on all members of
the District Fire Committee. Aircraft that have been used for aerial spotting in South Africa
over the past 20 years are the Cessna 206, Cessna 182, Cessna 185. Helicopters are also
used but are excessively expensive.
6.3.4.2 Government sub-district ward fire crews
The existing centralized fire crews must be decentralized and stationed in the sub district
wards. They will then be on hand to react to the initial attack on wildfires on government
and community land. Crews stationed at the sub district wards will also be closer to assist
with the extended attack fires within their wards. The fire crews will be stationed at the
following centers and be equipped with the following personnel and equipment. These fire
crews will assist with the burning of fire breaks as recommended in section 6.3.3 and
perform any prescribed burns required by the communities. They would also be available to
other land users for prescribed burning.
Shakawe
• 1 x 4x4 fire truck equipped with a 2 000 litre Slip-on fire fighting unit
• 1 x 11-person fire crew consisting of 10 fire fighters and 1 crew leader fully
trained and equipped with the required personal protective equipment
• 50 fire beaters – the extra beaters will be used by volunteers from the community
when required
• 10 back pack fire pumps
• 2 drip torches
• 1 first aid kit with Burn Shield
• 3 hand held radios
• The truck must be equipped with a radio – all radios to have a common District
Fire Committee frequency
• 1 hand held weather measuring device
Seronga
when required
• 2 drip torches
Tsau
when required
• 2 drip torches
Maun
• 2 x 11-person fire crew consisting of 10 fire fighters and 2 crew leaders fully
• 100 fire beaters – the extra beaters will be used by volunteers from the
community when required
• 4 drip torches
• The trucks must be equipped with a radio – all radios to have a common District
• 2 hand held weather measuring devices
6.3.4.2.1 Personnel
Personnel are as effective as their training. The remoteness and lack of numbers within the
Ramsar Site requires a few highly trained (see section 6.5), strategically placed and
motivated fire fighters. The emphasis must be on quality and not quantity.
• Fitness is one of the pillars of a successful fire fighting crew. Fit fire fighters are
effective and safe. It is essential to maintain a minimum level of fitness during the
fire season. All fire fighters should be tested at least once before the season to
ensure that they are at the standard fitness level.
• All fire fighters are required to perform strenuous tasks when fighting fires – the
following criteria should be followed when selecting fire fighters.
Male
¾ must complete 2.4km run in under 12 minutes
¾ must complete 20 push ups in 1 minute
¾ must complete 20 sit ups in 1 minute
Females
¾ must complete 2.4km run in under 14 minutes
¾ must complete 20 push ups in 1 minute
¾ must complete 20 sit ups in 1 minute
• Training is the second pillar of an effective fire fighter. See section 6.5 for training
courses that are the minimum recommended standards required for a fire fighter to
operate safely and effectively.
• All fire fighters must pass an Occupational Health and Safety medical examination.
6.3.4.3 Communications
Communication between all fire crews within the Ramsar Site is critical for them to operate
efficiently and safely during operations. The District Fire Committee must apply through the
relevant authorities for the allocation of a common radio frequency which must be made
available to all members of the District Fire Committee and the sub district ward fire crews.
6.3.5 Guidelines for minimum requirements for commercial/private stakeholders
The following minimum requirements in terms of equipment and personnel are provided as
a guide for stakeholders in the Ramsar Site. These guidelines are designed to enable the
stakeholder to have on hand sufficient equipment and personnel during a fire emergency to
manage the initial attack until additional help arrives in the event of a wildfire. The
objectives in a fire emergency would be to protect lives and property in the form of lodges,
homes and other assets threatened by a wildfire. In the event of a fire threatening the
environment the various lease holders and/or concession holders and land users would join
their resources together under a unified command system. (Lowveld and Escarpment Fire
Protection Association Standards).
Small prescribed burns and/or the construction of firebreaks may be accomplished using
the following list of equipment. Large, planned prescribed burns however, would require
that a number of land users and lease holders and/or concession holders pool their
resources to perform these burns. This will reduce the number of personnel and equipment
required per lease holder and/or concession holder.
6.3.5.1 Per Single Tourist Lodge/Tented Camp Site
All lodges should have the following minimum personnel and equipment:
• Vehicle slip on fire fighting unit (Bakkie Sakkie), minimum 750L water capacity
with the ability to draw water from a river or pool of water. This would be placed
on a 4x4 utility vehicle
• 6 persons trained and equipped to fight fires
• 20 fire beaters
• 4 back pack fire fighting double action pumps
• 1 drip torch (ignition tool)
• 2 hand held radios to communicate with the sub district fire frequency
• 1 first aid kit with burn shield
6.3.5.2 Per prescribed burning operation
All land users/owners who intend performing a prescribe burn must ensure that they have
personnel who are trained and equipped to perform the burn safely. They must also ensure
that the weather is within the allowed fire danger rating parameters.
6.4 Advocacy and Fire Awareness Education
Advocacy and awareness strategies and programs are cross cutting communication
functions, using specially crafted messages to disseminate knowledge. In this case fire
awareness campaigns and programs would aim to change errant behaviour and
perceptions, and encourage good land management practice, based upon the desired
outcomes of the Okavango Delta Management Plan (encompassing the fire management
plan) for the ODRS. As such, research to ascertain current perceptions about fire, including
traditional community practice, would be a forerunner to the design of any in-depth fire
awareness program.
Fire can and should be used as a beneficial and cost effective tool to address certain land
management problems, while the incorrect use of fire, potentially impacting negatively upon
ecological patterns and processes, needs to be addressed. Both aspects require public
communication and education programs as well as support for these from leadership,
municipal authorities, conservation and agricultural extension agencies and practitioners
operating in the area.
6.4.1 Advocacy
Before engaging in any widespread education program it is important to engender support

at a leadership and sectoral management level. There may need to be demonstrated
political willingness before mandated Government Department Officials will adopt and
integrate any program that is suggested. Below are two examples of the kind of advocacy
work that can be envisaged prior to a behaviour change program. All identified sectors will
require a similar strategic approach and co-operation between departments is a pre-
requisite:
Example 1:
Effect of Prescribed Burning on Tourism Sector:

It would be important to engage the tourism sector in developing material that can be
dispersed to tourists and tour operators through mandated tourism authorities. This might
entail a series of meetings, understanding the tourism sector’s view, mitigating negative
implications of burning, understanding how prescribed burning might affect the tourism
economy in the short and longer term, building a knowledge base within the mandated
tourism authority so that there is understanding about the beneficial impacts of fire, initiating
a system whereby tour operators are informed about scheduled burning operations, how to
effectively integrate messages into existing material for inbound tourists and the distribution
of material through Regional Tourism offices and operators.
Example 2:
School Education Program:

The mandated regional education authority would be engaged, in order to develop and
integrate suitable material into the school curriculum. Teacher training may be necessary,
and this would need to be developed and scheduled in collaboration with the Department of
Education. The program might need to be delivered by teachers at times appropriate to
seasonal and prescribed burning and simultaneously not disrupt normal teaching or annual
school calendar events. A high degree of joint planning is therefore required, possibly
leading to a regional task team.
These examples illustrate the importance of advocacy, consultation, participation and

partnership if a behavior change program is to be successfully adopted and sustained.
6.4.2 Awareness campaigns and programs
An awareness campaign or program should aim to use the different (and appropriate)
communication mediums available to increase public knowledge and understanding.
Knowledge is power. Knowledge about a subject allows people to start thinking, talking and
engaging on the subject, have an opinion about what should be done and a better
understanding of how changes will affect them personally. Communication is therefore an
important stage in “empowerment”.
A ”Campaign” is generally a short–term, very focused, high impact communication project.
A “Program” is generally a longer-term initiative, needing sustained intervention and

integration of combined skill sets.
Both can be used effectively, but sustained behavior change normally requires the benefit
of a ‘programmatic’ approach. According to Velez (1997) programs should endeavour to
educate the public. Naturally education is a matter of time, so wildfire prevention programs
call for continuity and repetition year after year.
Communication mediums include print, newspapers, radio and television, and most
importantly for rural communities, word of mouth, tribal authorities and government
agencies. In the rural scenario, meeting places such as schools, clinics and council
meetings play a critical role in disseminating knowledge amongst the community.
The messages contained within a fire awareness campaign or program will be based upon
the research findings applicable to the area, however some key sectors and audiences are
already apparent, allowing the suggestions below:
6.4.2.1 Tourism sector
Preparation and dissemination of material to tour operators, guides, lodges, conservation

officials and inbound tourists:
6.4.2.2 Community–based fire management
• Identify and consult rural communities

• Apply accepted best practice principles adopted for community based natural
resource management. For sustainable behavior change to occur there should be
a tangible benefit accruing to the community, for them to want to adopt any
proposed land management changes.
• Assist the community in drawing up a community fire management plan that is
simple, practical, achievable and measurable. Respect traditional customs and
subsistence practices. The plan should be a result of community consultation and
“owned” by the community.
• Assist with practical training of community members to carry out community –
based fire management tasks, including monitoring compliance with a community
designed fire management plan. The Department of Wildlife and National Parks
in collaboration with NASCO (Namibian Association of CBNRM Organizations)
consultants from Namibia have adopted the very effective Management

Orientated Monitoring System for Natural Resource Management (MOMS). It is
recommended that a MOMS type monitoring system be initiated at community
level related to the use of fire plus possible causes and incidence of wildfires.
Once communities own the system and are requested to supply data to the
District Fire Committee and finally to the Department of Forestry and Range
Resources changes in attitudes to the use and misuse of fire in the long term will
be facilitated.
• Expertise and advisory capacity should be available to communities on an
ongoing basis, either through government agency extension capacity, NGO ‘s or
consultants working at grass root level. This might also include capacity–building
education of community members to take advantage of opportunities by applying
for grants and financial aid, agriculture and eco-tourism projects and additional
capacity–building training.
• Reward communities that have achieved successful compliance with their land
management plan.
• Community–based fire management definitely requires a long term
‘programmatic’ approach to make real, sustained impact. Velez (1997) reports
that the prevention of man-made fires is a truly complex task calling for the use of
multiple techniques to encourage certain types of behaviour and discourage
others with the aim of preventing wildfires. This complexity shows that the
objective cannot be achieved if the action planned is not commensurate with the
magnitude of the problem.
• Persuasion programs or campaigns are usually developed by means of publicity
and the education system and their aims are basically:
- to inform the public of the existence of wildfire risk and the damage that can
result,
- to explain what action people can take (Velez, 1997).
• Fire prevention programs should be directed at all sectors of the public, this calls
for a stratification study. A general program may possibly leave broad sectors
without suitable information whereas in fact these sectors might be the source of
the fires (Velez, 1997).
6.4.2.3 School Education Program
• Research has proven that engaging and educating school children between the
ages of 8 - 14 gives a good return on investment as they are most receptive to
new ideas at this age. Not only are these tomorrow’s citizens, but also the
messages taught at school are carried home.
• Practical fire awareness messages can be introduced into lesson plans and in this
way, an evaluation of the retention of information is possible. e.g in language
comprehension, the story to be read can be about fire and its uses. In testing the
child’s comprehension of the story, we are able to ascertain absorption of the fire
message. The same principle can be applied to most subjects, but especially
language, science, mathematics and natural science. For best effect, lessons
should draw on local experience that the child can identify with and learning must
be ‘fun’!
• Modules are prepared in conjunction with the Department of Education, for the
different grades.
• Practical group projects, such as annually preparing firebreaks on the school
perimeter, regular emergency fire drills, fire and life safety training can be
incorporated into the school’s program.
• Visits from firefighters, fire focused drama and art competitions can further
enhance the uptake of fire awareness education. Regional ‘best school’ awards
encourage teachers to take ownership of the fire awareness program. Using art
prepared by scholars or local artists, relating to the wise use of fire and the
negative impacts of the misuse of fire, in education programs is a very effective
vehicle for knowledge transfer.
• Ideally, a programmatic approach is required – long term institutional funding for
reproduction of school materials, integration into text books, etc. is advisable. A
two to three year evaluation period within pilot schools should be undertaken
appraise the efficacy of the program and the return on investment.
6.5 Recommendations for Future Training and Capacity Building for Improved Fire
Management
6.5.1 “On-The-Job” Training Of Staff
At the commencement of the consultancy, 29 March 2006, a workshop was held at HOORC
where lectures on fire ecology were presented to staff from the Department of Forestry and
Range Resources, the Department of Wildlife and National Parks, HOORC, ODMP, and
other interested stakeholders, interested Botswana students and visiting international
students also attended. Illustrated lectures on The Effects and Use of Fire in African
Grasslands and Savannas and The Practical Application of the Assessment of the
Condition of the Grass Sward for Controlled Burning were presented during the morning
and field training was conducted in the afternoon. Participants were involved in practical
training on rapid but simple methods for assessing the condition of the vegetation. The use
of the Disc Pasture Meter for measuring biomass and the calculation of the Fire Danger
Index using hand-held weather kits was demonstrated.
Staff from the Department of Forestry and Range Resources were involved in further
training in the assessment of veld condition using the Disc Pasture Meter and the Point to
Tuft Method of measuring basal cover as a means of assessing the suitability of vegetation
for controlled burning. DFRR staff were involved in field trips to Seronga, Shakawe, Tsodilo
Hills, Tubu, Gumare and Sehithwa where numerous veld condition surveys were conducted
in various vegetation types. During the surveys DFRR staff were involved in the
identification of different grasses and trees encountered during the surveys and diagnostic
characteristics for the different species were highlighted. The Point Center Quarter Method
of surveying the shrub and tree component of the vegetation was also included in the
survey training along with a rapid method for calculating the height of trees. Training in
calculating the FDI at all survey sites was involved.
On 17 – 18 May 2006 a Fire Workshop was held at the ODMP and HOORC. Lectures were
presented on fire ecology on 17 May and the practical application on the use of fire was
demonstrated in the field at HOORC on 18 May. Practical training for participants on veld
condition assessment techniques was conducted by Professor Trollope. Mr Alex Held dealt
with calculating the FDI and the measures involved in planning for controlled burning. Mr
Chris Austin delivered lectures and demonstrations on the various hand tools available for
fire suppression. The workshop participants were involved in practical demonstrations on
burning firebreaks using the tracer belt method, the wet line method and the dry method.
Controlled burning was demonstrated on 6 000m2 area where a perimeter fire was initiated
and spot burning was also demonstrated. All participants were involved in burning
firebreaks, fire suppression and mopping up activities to ensure that the fire would no pose
a threat to the surrounds after the workshop.
6.5.2 Future training
6.5.2.1 Fire fighting/management training
The following training courses are based on internationally recognized training courses
developed by the International Wild Fire Working Group. Training in these courses is
available through Working on Fire International based in Nelspruit, South Africa
i) Basic fire fighting training course

Introduction
Herbage Preservation Act
Standard Operating Procedures & Organization at Fires
Understanding Fire
Types of Fires and Their Causes
Safety at Fires
Survival at Fires
The Use of Hand Tools for Fighting Fires
Mopping Up
ii) Crew Leader training course

Introduction
Fire Terminology
Radio Procedures
Fire Behaviour
Fire Danger Rating
Organization at Fires
Initial Attack Strategy & Tactics
Fire Extinguishing Methods
Fire Line Safety
Fire Fighting
First Aid
iii) Fire Boss / Incident Commander Level 3 training course

Introduction
Fire Terminology
Incident Command System
Fire Behaviour
Fire Weather
Initial & Extended Attack Strategies

Fire Simulation
First Aid
iv) Training course for prescribed burning

Introduction
Fire Management
Fuels
Types of Fires
Fire Behavior
Weather Planning a Prescribed Burn
Control Principals & Burning Methods
Prescribed Burning Equipment
Mopping Up
Fire Line Safety
6.5.2.2 Training in monitoring
As mentioned in section 6.4.2.2 practical training of rural communities and other interested
stakeholders in the Management Orientated Monitoring System for Natural Resource
Management (MOMS) is recommended to facilitate capacity building for increased fire
management. The Natural Resource Working Group of NASCO (Namibian Association of
CBNRM Organizations) consultants from Namibia conduct training workshops in the use of
MOMS and it is recommended that a MOMS type monitoring system be initiated at
community level related to the use of fire plus possible causes and incidence of wildfires.
Awareness programs and education will be required to change attitudes and get buy-in to
the program but once this is achieved, communities will become a valuable source of data
for the District Fire Committee and the Department of Forestry and Range Resources. The
MOMS system could also be introduced to facilitate the sustainable use of veld products
and thereby in the long-term promote the sustainable use and management of the ODRS
and secure the biodiversity of this amazing wetland to the benefit of all stakeholders.
Workshops and training courses are also recommended for DFRR staff and other
stakeholders in Range/Veld Condition Assessment techniques in order that the
recommended vegetation monitoring program can be implemented. Only five DFRR staff in
Maun received “on-the-job” training due to commitments within the Department during the
period of the consultancy. To establish a comprehensive and reliable vegetation monitoring
program Agricultural Officers and other staff need training to enable them to accomplish
their commitments. Regular staff transfers to other centers and Departments also depletes
the small core of trained staff.
6.6 General Discussion and Conclusions
The motivation for the development of a fire management plan for the Okavango Ramsar
Site was the perception that generally wild fires are perceived to be an increasing problem
in terms of their frequency, severity and uncontrolled nature in the Ramsar Site.
Regarding the frequency of fires in the Ramsar Site, Heinl (2005) and Tacheba (2002)
concluded that the main ignition source of fires in the Okavango Delta is anthropogenic
because fires ignited by lightning are rare. Heinl (2005) concluded further that the high
frequency of anthropogenic fires in the Okavango Delta is associated with fires being used
intentionally and frequently to provide fresh grazing for wildlife and cattle, improved viewing
for safari-tourism and safeguarding property. However, the affected areas are very limited
and the overall frequency of burning in the Delta is highly variable indicating no large
regulating and homogenizing effect on the vegetation at this stage. This conclusion by Heinl
(2005) is supported by the results in Figures 16 to 21 showing the number of fire incidents
recorded by the Modis satellite during the period 2000 to 2005. These indicate clearly that
the occurrence of fires varies significantly from year to year as influenced by variations in
rainfall in the Acacia, Burkea and Mopane Woodlands and rainfall and flood levels in the
Seasonal and Permanent Swamps. The results in Figure 24 indicate that the highest
number of fires occurred in the Permanent Swamps and Burkea Woodlands with a
significant but lower number of fires in the Seasonal Swamps and very few in the Acacia
and Mopane Woodlands. However, none of these sources of information are able to
indicate whether the overall frequency of fires in the Ramsar Site has changed over time.
The only information in this regard is provided by Cassidy (2003) who reported that from
interviews with local communities in the Panhandle region of the Ramsar Site the frequency
of fires in this region had decreased but had resulted in the significant accumulation of
greater fuel loads causing less frequent but more extensive and intense fires. The lower
frequency of fires was also associated with protecting areas that were being harvested for
thatching material. Therefore there is very limited evidence for the perception that the
frequency of fires in the Ramsar Site has increased and further in depth research over a
longer period of time is required to determine whether there has in fact been an increase in
the frequency of fires.
It is not possible to assess whether the overall severity or the intensity of the fires has
increased over time in the Ramsar Site because there are no data available on the current
and past intensity of fires. Suffice it to repeat that Cassidy (2003) reported from interviews
with local communities in the Panhandle that it was felt that due to the decrease in the
frequency of burning in that region there had been an increase in the accumulation of
greater fuel loads that had resulted in the occurrence of more intense fires.
Finally there are no definitive quantitative records available to indicate whether the
perceived increase of fires in the Ramsar Site are uncontrolled wildfires. Records obtained
from the Department of Forestry and Range Resources in Maun indicate that of the 123
fires reported to the Department during the period 2000 to 2005 the overwhelming majority
(93%) were of unknown causes and the remainder were caused by children (3%), road
workers (2%) and lightning (2%) suggesting that the majority of fires reported to the
Department are uncontrolled wildfires. Also considering that there are numerous
ecologically acceptable reasons for using fire in the Ramsar Site but that because there is a
widespread public misunderstanding that it is illegal to burn, combined with the difficulties of
obtaining permission to burn from the centralized system in Maun, the majority of fires
occurring in the Ramsar Site are illegal and are therefore generally regarded as
uncontrolled wildfires. This is obviously not true as undoubtedly a significant proportion of
these fires are being ignited according to well tested procedures developed over the
millennia and while they maybe illegal in the eyes of the law they are not totally of an
uncontrolled nature.
An important result from the review of literature was the conclusion by Heinl (2005) that the
vegetation in the dryland vegetation communities in the Ramsar generally reacted similarly
to fire as the other grassland and savanna communities in southern Africa. Therefore these
results on the effects of fire from elsewhere in Africa have been able to be used in the
development of the fire management plan for the Ramsar Site. This does not mean that no
further fire research is required in the Ramsar Site but it significantly reduces the amount of
research that needs to be conducted. While the results of the effects of fire in vegetation
types similar to the Acacia and Mopane Woodlands conducted elsewhere in southern Africa
does not have to be repeated, there is an urgent need for determining the effects of the fire
regime in the Burkea Woodlands and the Seasonal and Permanent Swamps in the Ramsar
Site, particularly the latter two vegetation units. Of significant interest in the Burkea
Woodlands is the Tsodilo area where high intensity fires have resulted in a significant
mortality of large trees greater than 10 metres in height. First impressions gained of the
area are that fire has had a major negative impact on the vegetation because of the high
profile effect on the large trees. However, detailed assessments of the condition of the
vegetation conducted in this area indicated that the density of large trees was in fact very
low (<100 trees per hectare) and that the overall mortality of trees and shrubs of all sizes
was only 1.8 %. Furthermore the assessment of the condition of the grass sward showed
that it was in good condition providing excellent grazing for both domestic livestock and
wildlife. Therefore the question about controlling wildfires in this area is what type of
vegetation is best suited to the system of land use in this area. If the priority is providing
grazing for domestic livestock and wildlife then the current fire regime is best left alone and
allowed to continue. Conversely if greater priority is given to developing the structural
diversity of the tree and shrub vegetation then the current fire regime needs to be altered to
one having less high intensity fires. An excellent analysis of the current fire regime in the
Tsodilo area is provided in the Integrated Management Plan for the Tsodilo Hills World
Heritage Site by Ecosurv (2005) and it is strongly recommended that the proposed fire
management plan for this area be considered for implementation in the future.
One of the requirements in the Terms of Reference was to consider the phenomenon of
peat fires in the Ramsar Site. The general consensus in the literature by experienced
scientists like Ellery et al (1989) is that peat fires play a role in nutrient cycling in the
Okavango Delta. Furthermore the area affected by peat fires is limited and besides active
peat fires being potentially dangerous and harmful to humans, livestock and large wild
ungulates and considering their beneficial effects on soil fertility, they are of limited
ecological concern in the Ramsar Site.
Another requirement in the Terms of Reference was to consider the problem of cross
border fires entering the Ramsar site from neighbouring countries, particularly from the
Caprivi region in Namibia which forms an extensive northern boundary with the Ramsar
Site. An inspection of the condition of the firebreaks on either side of the international
border at Mohembo showed that they were well maintained and in excellent condition for
preventing the occurrence of cross border fires from either side. However, an aerial visit to
the Linyanti region in the northern sector of the Ramsar Site showed that there were no
constructed firebreaks on either side of the Kwando River which forms the international
boundary between Botswana and Namibia. Apparently the Kwando River is assumed to
provide an effective firebreak against cross border fires. However, in Figure 17 the map of
the fire events recorded by satellite in 2001 indicates a high frequency of burning on the
western side of the Kwando River in the Linyanti region during this year. Considering that
the dominant direction of the prevailing winds in the Ramsar Site are easterly then it is
highly probable that the origin of some of these fires in this region during 2001 was from
fires ignited on the Namibian side which “jumped” the Kwando River spreading into this
section of the Ramsar Site. A practical solution to this problem would be to burn broad
firebreaks annually on the Botswana side of the western sector of the Kwando River so as
to provide an effective firebreak against cross border fires originating from the eastern side
of the river that are driven by the prevailing easterly winds. This would have to be done
during the low tourist season in order not to affect the tourism potential in this wilderness
area. The development of a graded firebreak along this western sector of the Kwando River
is not recommended because of the remoteness of the area and the ecological
undesirability of physically constructing firebreaks that involve excessive soil disturbance in
an ecologically sensitive, arid environment.
It is firmly believed that a major contribution in the report to fire management in the
Okavango Delta Ramsar Site has been the finding that the ecological criteria used
elsewhere in African grasslands and savannas can be used very effectively to assess the
condition of the vegetation in the Acacia, Burkea and Mopane Woodlands and Seasonal
Swamps for controlled burning. These criteria have been incorporated into a technique for
assessing the condition of the grass sward in the aforementioned vegetation units and has
the added advantage of also being able to identify areas where controlled burning should
not be applied and wild fires suppressed as far as possible. The assessment of the
condition of the vegetation in the different vegetation units using this technique resulted in
providing a preliminary description of the overall condition of the vegetation in the Ramsar
Site and clearly identified the Seasonal Swamps as having the highest potential for
producing grass fuel. It also identified this vegetation unit as being in a condition where
controlled burning should be considered as a means of removing the current stand of
moribund and unpalatable grass material thereby significantly improving its grazing capacity
for domestic livestock and wildlife. The technique also identified areas in the Burkea
Woodlands that had a high potential for producing grass fuel and could be considered for
controlled when necessary. Conversely both the Acacia and Mopane Woodlands were
identified as having a low fuel production potential and dominated by pioneer Increaser
thereby generally excluding these vegetation units from being considered for controlled
burning and emphasizing the necessity to suppress fires as far as possible in these areas.
Finally the fire management plan provides a detailed procedure for the planning and
implementation of a fire management program for the Ramsar Site including a
recommendation on the necessity for providing training in controlled burning and fire
fighting for the different stakeholders. It is gratifying to see that this recommendation is
already being implemented in the form of the current and future training courses that are,
and will be conducted for members of the Department of Forestry and Range Resources
and the Department of Wildlife and National Parks.
REFERENCES
Alonso, L.E. & Nordin, L. (Eds.) 2003. A Rapid Biological Assessment of the Aquatic
Ecosystems of the Okavango Delta, Botswana: High Water Survey. RAP
Bulletin of Biological Assessment 27. Conservation International, Washington,
DC: 248.
Anonymous, 2004. Terms of Reference Okavango Delta Management Plan Vegetation

Resources Component: Subcomponent 1: Development of a Fire Management
Plan in the Okavango Delta Ramsar Site, Botswana. Agricultural Resources
Board & Division of Forestry in the Ministry of Environment, Wildlife and
Tourism, Government Botswana: 12.
Anonymous, 2005. The Okavango Delta Draft Framework Plan: 346.
Anonymous, 2006. Okavango Delta Ramsar Site Land Use And Land Management Plan
(2005 – 2029). Final Report. Twanana Land Board, Ngamiland District Land
Use Planning Unit, Dept. Environmental Affairs, Okavango Delta Management
Plan Project Secretariat: 245.
Banda, A. 2004. The influence of range fire on soil microbial biomass and activity, fungal
population and diversity along the Boro route in the Okavango Delta. MSc
(Applied Microbiology) Thesis. University of Botswana: 115.
Bendsen, H. 2003. The dynamics of land use systems in Ngamiland. Changing livelihood
options and strategies. Harry Oppenheimer Okavango Research Center,
University of Botswana: 28.
Bendsen, H. 2004. Report prepared for the ODMP: Hot Spots identified by the traditional
land users. Okavango Delta Management Plan – Introductory Kgotla meetings.
Harry Oppenheimer Okavango Research Center, University of Botswana: 39.
Bond, W.J. & van Wilgen, B.W., 1996. Fire and plants. Chapman & Hall, London: 263.
Bond, W.J. 1997. Fire. In: Vegetation of southern Africa. Eds. Cowling, R.M., Richardson,
D.M. & Pierce, S.M. Cambridge University Press: 615.
Bonyongo, M.C. 1999. Vegetation ecology of the seasonal floodplains in the Okavango
Delta, Botswana. MSc Thesis, University of Pretoria, South Africa.
Bonyongo, M.C., Bredenkamp, G.J. & Veenendaal, E. 2000. Flood plain vegetation in the
Nxaraga Lagoon area, Okavang Delta. S. Afr. J. Bot. 66 (1):15 – 21.
Boultwood, J.N. & Rodel, M.G.W., 1981. Effects of stocking rate and burning frequency on
Brachystegia/Julbernadia veld in Zimbabwe. Proc. Grassl. Soc Sth. Afr. 16:
111-115.
Bothma, J. du P., 1996. Game ranch management. Van Schaik Publishers, Pretoria, South
Africa: 639.
Bransby, D.I. & Tainton, N.M. 1977. The disc pasture meter : possible applications in grazing
management. Proc. Grassl. Soc. Sth. Afr. 12 : 115-118.
Brockett, B.H., Biggs, H.C., & van Wilgen, B.W. 2001. A patch mosaic burning system for
conservation areas in southern African savannas. Int. J. Wildland Fire.
Brown, A.A. & Davis, K.P., 1973. Forest fire: control and use. McGraw Hill Book Co., New
York.
Byram, G.M., 1959. Combustion of forest fuels. In: Forest fire: control and use. Ed. K.P. Davis.
McGraw Hill Book Co., New York.
Carson W.P. & Abbiw, D.K., 1990. The vegetation of a protection site on the northern Accra
Plains, Ghana. Afri. J. Ecol. 28: 143-146.
Cassidy, L. 2003. Anthropogenic burning in the Okavango Panhandle of Botswana:

Livelihoods and spatial dimensions. MSc Thesis, University of Florida, United
States of America: 211.
Cheney, N.P. 1981. Fire behaviour. In: Fire and the Australian biota. Eds. A.M. Gill, R.H.
Groves, I.R. Noble. Australian Acad. Sci., Canberra. 151-175.
Cottam, G. & Curtis, J.T. 1956. The use of distance measures in phytosoiciological
sampling. Ecol. 37: 451 – 460.
Cowling, R.M., Richardson, D.M. & Pierce, S.M. (Eds.) 2003. Vegetation of Southern Africa.
Pub. Cambridge University Press. Cape Town, South Africa: 615.
Dillon, R.F., 1980. Some effects of fire in the Tall Grassveld of Natal. M.Sc. (Agric.) Thesis,
Univ. Natal. Pietermaritzburg.
Dixon, A.B. 2003. Indigeneous management of wetlands. Ashgate Publishing Limited,

England: 243.
Dublin, Holly. T., 1995. Vegetation dynamics in the Serengeti-Mara Ecosystem; The role of
elephants, fire and other factors. In: Serengeti II: Dynamics, Management and
Conservation of an Ecosystem. Ed: A.R.E. Sinclair & Peter Arcese. Univ. Chicago
Press, Chicago, London. 71-90.
Dugan, P.J. 1992. Wetlands management: a critical issue for conservation in Africa.
Proceedings of the SADCC Wetlands Conservation Conference for Southern
Africa. Pg 1 – 8.
Du Plessis, W.P. 1997. Refinements to the burning strategy in the Etosha National Park,
Namibia. Koedoe 40,1: 63 – 76.
Ellery, K. 1987. Wetland plant community composition and succession processes in the
Manchuria River system of the Oksavango Delta. MSc Thesis. University of the
Witwatersrand, South Africa: 105
Ellery, W.N. 1988. Channel blockage and abandonment in the north east Okavango Delta:
The role of papyrus. MSc Thesis. University of the Witwatersrand, South Africa:
98.
Ellery, W.N., Ellery, K., McCarthy, T.S., Cairncross, B. & Oelofse, R. 1989. A peat fire in the
Okavango Delta, Botswana, and its importance as an ecosystem process.
African Journal of Ecology Vol. 27: 7 – 21.
Ellery, W.N., McCarthy, T.S. & Smith, N.D. 2003. Vegetation, Hydrology and Sedimentation
Patterns on the Major Distributary System of the Okavango Fan, Botswana.
Wetlands, Vol 23, No. 2: 357-375.
Everson, T.M., Everson, C.S., Dicks, H.M. & Poulter, A.G., 1988. Curing rates in the grass
sward of the Highland Sourveld at Cathederal Peak in the Natal Drakensberg. S
Afr. Forestry J. 145: 1-8.
Field, D. 1978. A handbook of basic ecology for range management in Botswana. Land
Utilization Division, Ministry of Agriculture, Gaberone, Botswana: 1-117.
Forbes, R.G. & Trollope, W.S.W., 1990. Veld management in the rural areas of Ciskei.
J.Grassl. Soc. South Afr. 8, 4: 147-152.
Giannakis, N. & Arianoutsou, M. 1997. Forest fire risk and management. Proceedings of the
European School of Climatology and Natural Hazards. Report EUR 16719EN:
383.
Giovannini, G. 1992. The effect of fire on soil quality – physical and chemical aspects. In:
Forest fire risk and management. Proceedings of the European School of
Climatology and Natural Hazards Course: 443.
Heinl, M., 2005. Fire and its effects on vegetation in the Okavango Delta , Botswana.
Doctoral Thesis Nature Conservation, Technical University of Munich,
Germany: 115.
Høberg, P., Lindholm, M., Ramberg, L & Hessen D.O. 2002. Aquatic food web dynamics on
a floodplain in the Okavango Delta, Botswana. Hydrobiologia 470: 23 – 30.
Howard, G.W. 1992. Floodplains: utilization and the need for management. Proceedings of
the SADCC Wetlands Conservation Conference for Southern Africa: 15 – 25.
Kannenberg, N. 1992. Grass-biomasse in savannenbiomen des Etosha National Park:

anwendungdes Disc Pasture Meter (DPM).In K. Weiss & J. Goldammer (ed).
Feuer in der Umwelt. Albert-Ludwig-Universitat Freiburg, Germany: 76 – 91.
Kennan, T.C.D., 1971. The effects of fire on two vegetation types of Matopos. Proc. Tall
Timbers Fire Ecology Conf. 11: 53-98.
Komarek,E.V. 1965. Fire ecology – grasslands and man. Proc. Tall Timbers Fire Ecology
Conf. 4:169 – 220.
Komarek, E.V., 1971. Lightning and fire ecology in Africa. Proc. Tall Timbers Fire Ecology
Conf. 11: 473-511.
Leigh, J.H. & Noble, J.C., 1981. The role of fire in the management of rangelands in Australia.
In: Fire and the Australian biota. Eds. A.M. Gill, R.H. Groves, I.R. Noble.
Australian Acad. Sci., Canberra. 471-495.
Luke, R.H. & McArthur, A.G., 1978. Bush fires in Australia. Australian Govt. Pub. Serv.,
Canberra.
Main, M. 2000. Kalahari – life’s variety in dune and delta. Publ. Southern Book Publishers
(Pty) Ltd, Johannesbburg, South Africa: 265.
McIntyre, C. 2003. Botswana, the Bradt travel guide. Pub. Bradt Travel Guides Ltd.
Buckinghamshire, England: 532
Mendelsohn, J. & el Obeid, S., 2004. Okavango River The flow of a lifeline. Struik
Publishers Pty Ltd., Cape Town, South Africa: 176.
Menaut, J.C., 1977. Evolution of plots protected from fire since 13 years in a Guinea savanna
of Ivory Coast. Actas Del IV Simposium Internacional De Ecologia Tropical Tomo
II, Universidad De Panama, Panama: 543-558.
Moe, S. R.,Wegge, P. & Kapela, E.B., 1990. The influence of man-made fires on large wild
herbivores in Lake Burungi area in northern Tanzania. Afr. J. Ecol. 28: 35-43.
Mosepele, K. 2002. Trends in fisheries development and fish utilization in the Okavango
Delta. Unpublished report. Harry Oppenheimer Okavango Research Center.
Munthali, S.M. & Banda, H.M., 1992. Distribution and abundance of the common ungulates
of Nyika National Park, Malawi. Afr. J. Ecol. 30: 203-212.
Phillips, J.F.V., 1965. Fire - as a master and servant: its influence in the bioclimatic regions
of Trans - Saharan Africa. Proc. Tall Timbers Fire Ecology Conf. 4: 7-10.
Plowes , D.C.H., 1957. The seasonal variation of crude protein in twenty grasses at
Matopos , Southern Rhodesia, and related observations. Rhod. Agric. J. 54: 33
– 55.
Pratt, D.J. & Gwynne, M.D., 1977. Rangeland Management & Ecology in East Africa.
Hodder & Stoughton, London.
Robinson, E.T., Gibbs-Russel, G.E., Trollope, W.S.W. & Downing, B.H., 1979. Assessment
of short term burning treatments on the herb layer of False Thornveld of the
Eastern Cape. Proc. Grassld. Soc. Sth. Afr. 14: 79-83.
Roggeri, H.1995.Tropical Freshwater Wetlands – A Guide to Current Knowledge and

Sustainable Management. Kluwer Academic Publishers, Dordrecht,
Netherlands: 349.
Ross, K. 1987. Okavango – Jewel of the Kalahari. Pub. BBC Books, London: 256.
Schackleton, C.M., 1992. Area and species selection by wild ungulates in coastal sour
grasslands of Mkambati Game Reserve, Transkei, southern Africa. Afr. J. Ecol.
30: 189-202.
Scott, J.D. 1970. Pros and cons of eliminating veld burning. Proc. Grssld Soc. Sth. Afr. 5:
23 – 26.
Scott, J.D., 1971. Veld burning in Natal. Proc. Tall Timbers Fire Ecology Conf. 11: 33-51.
Shea, S.R., Peet, G.B. & Cheney, N.P., 1981. The role of fire in forest management. In: Fire
and the Australian biota. Eds. A.M. Gill, R.H. Groves. I.R. Noble. Australian Acad.
Sci. Canberra. 443-470.
Siegfriied, W.R. 1981. The incidence of veld-fire in the Etosha National Park, 1970 – 1979.
Madoqua 12: 225 – 230.
Sinclair, A.R.E., 1995. Serengeti Past and Present. In: Sinclair A.R.E. & Peter Arcese (eds.)
Serengeti II: Dynamics, Management and Conservation of an Ecosystem. Univ.
Chicago Press, Chicago & London: 3-30
Smit, G.N. & Rethman, N.F.G 1998. Root biomass, depth distribution and relations with leaf
biomass of Colophospermum mopnae. . S.Afr. J. Bot. 64: 38 – 43.
Stampa, S., 1959. Tick paralysis in the Karroo areas of South Africa. Onderstepoort J. Vet.
Res. 28, 2. Govt. Print., Pretoria.
Sweet, R.J., 1982. Bush control with fire in Acacia nigrescens/ Combretum apiculatum
savanna in Botswana. Proc. Grassl. Soc. Sth. Afr. 17: 25-28.
Tacheba, B. 2002. Fire occurrence and distribution in the floodplains of the Okavango
wetlands (2000 – 2001 burning season). Harry Oppenheimer Okavango
research Centre, University of Botswana, Maun: 41.
Tainton, N.M., 1999. Veld burning. In: Veld management in South Africa. University Natal
Press, Pietermaritzburg, South Africa: 217-245.
Tainton, N.M., Groves, R.H. & Nash, R., 1977. Time of mowing and burning veld: short term
effects on production and tiller development. Proc. Grassl. Soc. Sth. Afr. 12: 59-
64.
Tainton, N.M. & Mentis M.T., 1984. Fire in grassland. In: Booysen, P.de V. & Tainton, N.M.
(eds). Ecological effects of fire in South African ecosystems. Ecological Studies
No 48: 115-197.
Teie, W.C., 2003. Fire manager’s handbook on veld and forest fires – strategy, tactics and
safety – South African edition. ISBN 1 931301 08 5. Intrepid Printers (Pty) Ltd,
Pietermaritzburg, South Africa: 1-488.
Thompson, K. 1974. A report on the interactions between environmental factors and

aquatic and semi-aquatic plants in the Okavango delta with recommendations
for further lines of investigation. Project BOT/506. Department of Biological
Sciences, University of Exeter, Wales: 32 –
Thomson, R. 1992. The wildlife game. Nyala Wildlife Publications Trust, Westville, South
Africa. ISBN 0 620 163151.
Tinley, K.L. 1975. An ecological Reconnaissance of the Moremi Wildlife Reserve, Northern
Okavango Swamps, Botswana. Pub. Okavango Wildlife Society: 147.
Tlotlego, S. 2004. Remote sensing and geographic information system and veldt fire
management: exploring possibilities for an interface for the rural communities.
Thesis, University of Botswana: 62.
Trollope, W.S.W., 1978. Fire behaviour - a preliminary study. Proc. Grassl. Soc. Sth. Afr. 13:
123-128.
Trollope, W.S.W., 1983. Control of bush encroachment with fire in the arid savannas of
southeastern Africa. PhD thesis, University Natal, Pietermaritzburg.
Trollope, W.S.W., 1984. Fire in savanna. In: Booysen, P.de V. & Tainton, N.M.(eds).
Ecological effects of fire in South African ecosystems. Ecological Studies No
48: 149-175.
Trollope, W.S.W., 1986. Land use surveys: Assessment of Veld Condition in Ciskei. In:
Republic Of Ciskei National Soil Conservation Strategy, Volume 1. Publ. Dept.
Agric. & Forestry, Ciskei: 1-37.
Trollope, W.S.W., 1987. Effect of season of burning on grass recovery in the False
Thornveld of the Eastern Cape. J. Grassl. Soc. South. Afr. 4, 2:74-77.
Trollope, W.S.W., 1989. Veld burning as a management practice in livestock production. In:
Veld Management in the Eastern Cape. (eds) J E Danckwerts & W R Teague.
Govt. Printer, Pretoria: 196.
Trollope, W.S.W., 1990. Development of a technique for assessing veld condition in the
Kruger National Park. J. Grassld. Soc. South. Afr.7,1:46-51.
Trollope, W.S.W., 1990. Veld management with specific reference to game ranching in the
grassland and savanna areas of South Africa. Koedoe 33, 2:77-86.
Trollope, W.S.W., 1999. Veld Burning In Different Vegetation Types: Savanna. In: Veld
Management In South Africa. Ed. N.M. Tainton. University Natal Press,
Pietermaritzburg. 236 - 242.
Trollope, W.S.W., 2005. Fire danger rating system for controlled burning in African
grassland and savanna ecosystems. Dept. Livestock & Pasture Science, Univ.
Fort Hare, Alice, 5700, South Africa:
Trollope, W.S.W., Biggs, H., Potgieter, A.L.F. & Zambatis, N., 1995. A structured versus a
wilderness approach to burning in the Kruger National Park in South Africa. In:
West N.E. (ed.). Rangelands in a sustainable biosphere. Denver: Society for
Range Management: 574-575.
Trollope, W.S.W & Booysen P de V., 1971. The eradication of macchia (fynbos) vegetation
on the Amatola Mountains of the Eastern Cape. Proc. Grassld. Soc. Sth Afr. 6:
28-38.
Trollope, W.S.W., Hines, C.J.H. & Trollope, L.A., 2000. Simplified techniques for assessing
range condition in the East Caprivi region of Namibia. Final Report. Directorate
of Forestry, Namibia-Finland Forestry Programme, Windhoek, Namibia: 1-54.
Trollope, W.S.W. & Potgieter, A.L.F., 1985. Fire Behaviour in the Kruger National Park. J.
Grassl. Soc. Sth. Afr. 2.2: 17-22.
Trollope, W S W and Tainton, N M., 1986. Effect of fire intensity on the grass and bush
components of the Eastern Cape thornveld. J. Grassld. Soc. Sth Afr. 2: 27-32.
Trollope, W.S.W., Potgieter, A L F and Zambatis, N., 1989. Assessing veld condition in the
Kruger National Park using key grass species. Koedoe 32, 1: 67-94.
Trollope, W.S.W., Potgieter A.L.F. & Zambatis, N., 1990. Characterization of fire behaviour
in the Kruger National Park. Final report. National Parks Board, Kruger National
Park, Skukuza.
Trollope, W.S.W & Tainton, N M., 1986. Effect of fire intensity on the grass and bush
components of the Eastern Cape thornveld. J. Grassld. Soc. Sth Afr. 2: 27-32.
Trollope, W.S.W., Trollope, L.A. & Bosch, O J H., 1990. Veld and pasture management
terminology in southern Africa. J. Grassld. Soc. South Afr. 7,1:52-61.
Trollope, W.S.W. & Trollope, L.A., 1999. Report on the assessment of range condition and
the fire ecology of the savanna vegetation on the Lewa Wildlife Conservancy in
Kenya – 1998. Final Report. Department Livestock & Pasture Science, Faculty
of Agriculture, University Fort Hare, Alice: 1-54.
Trollope, W.S.W. & Trollope. L.A., 2002. Fire behaviour: a key factor in the fire ecology of
African grasslands and savannas. Forest Fire Research & Wildland Fire Safety,
Millpress, Rotterdam, www.millpress.com :1-15.
Trollope, W.S.W. & Trollope, L.A., 2005. Assessment of the need for controlled burning in
Concession Area No 34 in the Okavango Delta in Botswana. Department
Livestock & Pasture Science, University Fort Hare, Alice, South Africa: 1-11.
Trollope, W.S.W., van den Broeck, D., Brown, D., Webber, L.N. & Nibe, S., 2005. Assessment
of veld condition in the thicket communities of the Great Fish River Reserve in the
Eastern Cape Province of South Africa. African J. Range & Forage Sci. in press.
van der Heiden, L.J. 1992. The Okavango Delta: current state of planning and
conservation. Proceedings of the SADCC Wetlands Conservation Conference
for Southern Africa: 109 – 124.
Van Wilgen, B.W., Everson, C.S. & Trollope, W.S.W. 1990. Fire management in Southern
Africa: Some examples of current objectives, practices and problems. In: Fire in
the Tropical Biota – Ecosystem Processes and Global Challenges. (ed). J G
Goldammer. Springer-Verlag, Berlin.
Van Wilgen, B.W., Biggs, H.C. & Potgieter, A.L.F. 1998. Fire management and research in
the Kruger National Park, with suggestions on the detection of thresholds of
potential concern. Koedoe 41,1: 69 – 78.
van Wilgen, B.W., Biggs, H.C., O’Reagan, S.P. & Mare’, N., 2000. A fire history of the
savanna systems in the Kruger National Park, South Africa between 1941 and
1996. South African J. Science 96: 167-178.
Vetter, S., 2003. What are the costs of land degradation to communal livestock farmers?
The case of the Herschel district, Eastern Cape. PhD Thesis, Univ. Cape Town.
West, O., 1965. Fire in vegetation and its use in pasture management with special
reference to tropical and sub-tropical Africa. Mem. Pub. Commonw. Agric. Bur.,
Farnham Royal, Bucks., England. No.1.
Willis, M.J. & Trollope, W.S.W. 1987.The use of key species for assessing veld condition in
the eastern Cape. J. Grssld, Soc, Sth. Afr. 4: 113 – 115.
Wright, H.A. & Bailey, A.W., 1982. Fire Ecology: United States and Southern Canada. John
Wiley & Sons, New York.
Vélez, R. 1997. Forest fire risk and management. Proceedings of the European School of
Climatology and Natural Hazards. Report EUR 16719EN: 443.
Zacharias, P.J.K. 1994. The fire/grazing interaction in Döhne Sourveld . DSc Agric. Thesis.
University of Fort Hare, Alice.
Appendix 1:
Terms of Reference
Department of Forestry & Range Resources in the
Ministry of Environment , Wildlife and Tourism
And
Range Ecology Section in the Ministry of Agriculture
And
Okavango Delta Management Plan, Secretariat
TERMS OF REFERENCE
ODMP Vegetation Resources Component
Understanding and managing fire in the Okavango Delta Ramsar Site
May 2005
1. BACKGROUND
1.1 On the 4th April 1997, Botswana became a contracting party of “The Ramsar
Convention” and listed the Okavango Delta as the world’s largest wetland of
international importance.
1.2 In order to ensure the Delta’s conservation and wise use, The Okavango Delta
Management Plan (ODMP) project proposal was drawn up in 2002 as a means “to
integrate resource management for the Okavango Delta that will ensure its long term
conservation and that will provide benefits for the present and future well being of the
people, through sustainable use of its natural resources” (ODMP, 2002).
1.3 The strategy that will be utilised to realise the implementation of the ODMP is to,
amongst other things, collectively create a greater sense of responsibility and
accountability amongst communities and in existing institutions with a mandate to
manage the Delta and its resources. In doing so, 9 components and their respective
responsible institutions were identified as follows:-
- Policy, Planning and Strategy by the Department of Environmental Affairs
- Communication by the Department of Environmental Affairs
- Research, data management and participatory planning by Harry
Oppenheimer Okavango Research Centre (HOORC)
- Hydrology and Water Resource by Department of Water Affairs (DWA).
- Fisheries by Department of Wildlife and National Parks (DWNP), Fisheries

Division
- Wildlife Management by Department of Wildlife and National Parks
(DWNP).
- Sustainable Tourism and CBNRM by the Department of Tourism (DoT),
Maun and the North West District Council (NWDC)
- Settlement development planning by the NWDC Physical Planning Unit
and the Department of Town and Regional Planning (DTRP).
- Sustainable Livestock Management by the Department of Animal Health
and Production (DAHP)
- Vegetation Resources by the Department of Crop Production (DCP),
Ministry of Agriculture and the Agricultural Resources Board
(ARB) and Forestry Division, Ministry of Environment, Wildlife
and Tourism
- Landuse Planning and Land Management by Tawana Land Board (TLB) in
association with the District Land Use Planning Unit (DLUPU).
- Waste management – NWDC Environmental Health Department
There will also be active stakeholder participation in the main stages, an association
of international stakeholders and an integrated planning process.
1.4 The ODMP requires the Department of Crop Production (DCP), Ministry of
Agriculture and the Agricultural Resources Board (ARB) and Forestry Division,
Ministry of Environment, Wildlife and Tourism, collectively known as the Vegetation
Resources component “to ensure sustainable management of the Okavango Delta
vegetation initiated and supported by providing accurate data and assisting in
resolving vegetation management conflicts” (ODMP Project Proposal 2002)
1.5 Wild fires are perceived to be an increasing problem in the Okavango Delta, both in
terms of their frequency, severity and uncontrolled nature. This issues has been
raised by communities, the tourism sector, scientists and other stakeholders during
consultation meetings. Their concerns are echoed by the large areas of the Delta,
both in the wetland itself but also in the dry sand veld surrounding the wetland, that
are seen to be burnt each year
1.6 It is apparent from this that existing policy and structures for managing fire are
neither adequate nor effective
1.7 An important element in the sustainable management of vegetation in the Delta is

thus the development of a more effective fire management strategy to better manage
the frequent wild fires occurring within the Ramsar site.
2.0 AREA OF STUDY
2.1 The Okavango Delta is situated at the northern most edge of the Kalahari Desert in
north western Botswana, below the Caprivi Strip in Namibia. It is the largest
designated inland wetland in the world and is fed by the water of the Okavango River
with between 8-15 thousand million cubic metres of water per annum from the river’s
headwaters in Namibia and especially Angola.
2.2 The proposed study is to be conducted within the newly proposed boundaries of the
existing Okavango Delta Ramsar site (see Map 1).
Map 1 Proposed new boundaries of the existing Okavango Delta Ramsar site –
boundary for the proposed contract
3.0 OBJECTIVES AND ACTIVITIES
3.1 Overall objective
The overall goal and objective is to develop a fire management strategy
3.2 Specific activities

a) Collect, assimilate, analyse and evaluate all existing information on fire (where, why,
how often, effects of etc.), including traditional knowledge, feedback from tourism
operators as well as scientific studies within the Ramsar site, as well as information
from elsewhere in southern Africa and in wetlands globally.
b) Assess effectiveness of current strategies (e.g. fire breaks, education etc.) and
policies (i.e. burning allowed with permits) used to stem and control fire in the
Ramsar site and the rest of Botswana and suggest alternatives.
c) Identify gaps in existing information (e.g. outline additional habitats in the Ramsar
site that require study, such as permanent swamp, dry land areas etc., as well as
other areas of study) and make recommendations for future work.
d) Using the available information, develop an initial fire management strategy for the
Okavango Delta Ramsar site. This should contain clear recommendations on how
fire can be better managed, on how existing policy can be adapted and adjusted to
achieve this and on the sensitivity of different ecological zones and habitats to fire.
e) Produce a report that clearly outline findings from analysis of information, e.g. on
why fires start, where they start, when they start, how often they occur and what their
effects are. This should be aimed at developing an enhanced understanding on the
causes and effects of fire and provide a useful tool to both GoB institutions as well as
stakeholders.
f) Deliver “on-the-job” training to staff of the Vegetation Component in delivery of the

consultancy
g) Make recommendations for future training and capacity building with the appropriate
departments and stakeholders to facilitate improved fire management
4.0 REPORTING SCHEDULE AND TIMING OF WORK
4.1 The study shall take 130 days
4.2 The client for this contract is the Vegetation Resources component of the ODMP.
The consultant engaged to deliver this contract will report directly to the staff of the
Vegetation Resources component and the Okavango Delta Management Plan
Project Secretariat.
4.3 The consultant will be expected to be based in Maun for the duration of the contract.
Office space will be provided by the ODMP Project Secretariat. All meetings will be
held in Maun.
4.4 A “Task Force” responsible for technical input and steering of the contract has been
established and the consultant will be expected to report their progress to that group.
At each milestone during delivery of the contract. The composition of the Task Force
is as follows:
(a) Vegetation component staff – Mr Kemoreile, Forestry; Mr Maswabi, ARB; Mr

Dithapo, Range Ecology (Chair)
(b) Physical Planner, DTRP/NWDC – Mr Fanuel Kibakaya
(c) ODMP Project Coordinator – Portia Segomelo
(d) ODMP Chief Technical Advisor – ?
(e) Appointed ODMP officer) Tawana Land Board – Mr Shimane Mongati
(f) District Officer (Development), North West District – Mr. G. Macha
(g) District Officer (Lands), North West District – Mr Ramsden
(h) Dept. of Tourism – Felicity Rabolo
(i) Dept. of Wildlife and National Parks – Regional Wildlife Coordinator (S.
Mosojane), CE&O officer (Claudia Zuze), Research (Lettie Pitlagano)
(j) NGO reps, e.g. KCS, CI, People and Nature Trust
(k) Harry Oppenheimer Okavango Research Centre – Hannelore Bendsen,
Mike Murray-Hudson, Caspar Bonyongo etc.
(l) Dept. of Animal Health and Production – Mr Gaebope
(m) Other members to be co-opted as and when necessary
4.5 The Consultants shall commence work not later than two weeks after award is made
and the instruction to proceed is given. The consultant will be expected to prepare
and submit to the Vegetation Resources component and the Task Force the
following reports:-
a) An Inception Report (5 digital and 15 hard copies) within 2 weeks of signing of

the contract. This report will detail proposed methodologies, timing, required
resources and a detailed plan for the delivery of training as an integral part of
the delivery of this subcomponent
b) Brief progress reports to be delivered at progress meetings on a monthly
basis. Progress reports should outline provisional results, challenges faced
etc
c) A Draft Final Report (5 digital and 15 hard copies) to be produced within 5
months of the signing of the contract
d) A Final Report (5 digital and 15 hard copies) to be produced within one month
after discussion of the Draft Final Report
4.6 The exact timing for each of these reporting stages should be aligned with the
schedule of reporting for the ODMP in order to ensure integration of this contract with
all other aspects of the wider ODMP project.
4.7 The consultant will be expected to establish a relationship with other consultants
working under the ODMP and to work with them wherever necessary and
appropriate in the achievement of their aims and the overall aim of the ODMP. It will
also be important for them to establish a thorough understanding of the ODMP and
its institutions.
4.8 The consultant will be expected to consult with tourism operators in the Delta to
establish their position on fire in relation to activity a) above. Feedback from
communities has already been established through ODMP project consultation and
by work done at the HOORC.
4.9 All reports will also ultimately be subject to consideration and approval of the ODMP
structures, i.e. the Project Management Group, the Okavango Wetland Management
Committee and the ODMP Project Steering Committee. Reports from this contract
will be incorporated into the ODMP Framework Management Plan, the Draft
Management Plan and the Final Management Plan.
5.0 OTHER PERTINENT INFORMATION
5.1 A desktop study to ascertain all existing information and the full background to this
area of research should be conducted. This should make reference, but not be
limited, to the following:
¾ Social and ecological study, Every River Project

¾ Southern African Red Data List (IUCN), Report No.14, 2002)
¾ Extensive fire study including effects and causes (2000 – 2003), SAFARI 2000
(Vol 1-3)
¾ SAFNET – Southern African Fire Network (www.safnet.net)
¾ Okavango Delta Management Plan, Project Proposal (April 2002)
¾ Vegetation Resources component Draft Inception Report for the Okavango Delta
Management Plan
¾ CITES checklist of endangered and threatened plant taxa (2003)
¾ Fire research documents from the Harry Oppenheimer Okavango Research
Centre and the University of Botswana and partners, including but not limited to
the following:
Heinl, M., Reconstructing the fire history of the Okavango Delta, Botswana
(southern part) 1989-2003. 2003, Vegetation Ecology, TU Muenchen,
heinl@wzw.tum.de
Fire and its effects on vegetation in the Okavango Delta, Botswana
(www.weihenstephan.de/vegoek/forschun/feuchte/engheinl_DA.html)
Fire Regime and vegetation response in African savannah ecosystems
(Model site: Okavango Delta, Botswana)
(www.wzw.tum.de/vegoek/personen/heinl/engindex.html)
¾ Other similar fire studies, either published or grey literature, from southern Africa,
especially Kruger National Park, RSA
¾ Herbage Preservation Act (1974)
5.2 Some fieldwork to collect and verify information may be required; if so, the consultant
will be expected to provide all their own fieldwork equipment and vehicles
5.3 These Terms of Reference, any proposed amendments and/or the consultants
proposed modus operandi may be discussed with the potential client prior to
submission of the tender and commencement of the study
5.4 The consultant should undertake consultation with other government institutions,
especially those directly involved in the delivery of other aspects of the ODMP, as
well as NGOs etc. The consultant should be aware that this contract is being
delivered as just one small part of a much larger, integrated and comprehensive
management plan for the Okavango Delta and should be prepared to work with this
in mind.
6.0 QUALIFICATION AND EXPERIENCE OF CONSULTANT
6.1 The consultant will require proven expertise in fire ecology, botany and taxonomy
and should have experience of work either directly in the Okavango Delta, Botswana
or the Southern African region. Experience of work with communities and other
stakeholders, knowledge of Geographical Information Systems (GIS), remote
sensing and mapping, experience in model development and an ability to deliver
comprehensive on-the-job and other training would be a distinct advantage.
6.2 The consultant must submit documents including a detailed resume showing
capacity to undertake the contract as well as their proposed methodology to deliver
the contract objectives.
7.0 FINANCIAL DETAILS
7.1 The consultant should include a full financial bid as part of their overall submission.
This should include:
a) Transport (mode and kilometres expected)

b) Accommodation (number of days expected)
c) Subsistence (number of days expected)
d) Printing and Documentation (based on the schedule of reports) and;
e) Any other costs (specify accordingly)
8.0 SCOPE OF THE PROPOSAL
7.1 Those tendering shall submit five sealed copies of two separate proposals covering
Technical and Financial Proposals.
7.2 The Technical Proposal should outline the following: -
• Proposed methodology, including a detailed plan for the delivery of training

• Details of relevant qualifications and experience of the consultant and any
other staff to be employed on each the subcomponent.
7.3 The Financial Proposals shall be based on personnel and time inputs for the
subcomponent delivery as detailed in the technical proposals, with a complete
breakdown of time allocated and rates charged for each personnel input.
8.0 CURRENT ON-GOING WORK
8.1 Those tendering shall disclose all current on-going similar or related work and
anticipated completion dates, for the client to assess the consultant’s capability of
giving total commitment and timely delivery of the contract, given the short time span
and urgent need for this work.
8.2 Failure to disclose this information shall result in disqualification of the tenderer.
9.0 PAYMENT
9.1 All payment in relation to this project shall be in Botswana Pula. Payment for
staff input and associated reimbursable costs will be made upon the receipt and
subsequent Task Force consideration and approval of the milestone reports as
already outlined.
9.2 The proposed payment schedule is as follows:
a) 20% upon approval of the Inception Report

b) 30% upon completion of three months contract
c) 20% upon approval of the Draft Final Report and;
d) 30% upon approval of the Final Report
10.0 SUBMISSION OF PROPOSALS
10.1 Five copies of both the Technical and Financial Proposals shall be submitted
to Agricultural Resources Board Secretary in Maun. The proposals shall be
submitted in separate, sealed envelopes and clearly marked:
a) “ODMP Vegetation Resources Component. Fire contract – Technical

Proposals” for the Technical Proposals, and;
b) “ODMP Vegetation Resources Component. Fire contract – Financial

Proposals” for the Financial Proposals.
Annual and monthly rainfall recorded at Maun and Shakawe in the Okavango Delta Ramsar
Site in Botswana. Rainfall expressed in millimetres. (Department Meteorological Services,
Botswana)
Appendix 2
Maun – Rainfall 1922 to 2006
YEAR JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TOTAL
1922 126 24 48 0 18 0 0 3 0 24 24 95 362
1923 120 118 95 9 12 0 0 0 0 1 26 36 418
1924 14 47 189 0 0 0 0 0 3 36 47 0 337
1925 167 244 159 48 62 2 0 0 14 1 10 83 789
1926 63 28 79 14 9 0 0 0 0 17 39 102 350
1927 38 42 11 18 0 0 1 1 0 67 3 39 219
1928 162 40 0 0 0 0 0 0 1 4 72 87 365
1929 113 54 14 0 0 0 0 0 1 1 23 108 314
1930 48 66 10 91 0 0 0 0 0 2 13 107 335
1931 89 103 84 36 0 0 0 0 0 11 47 42 411
1932 25 174 274 14 0 0 0 0 0 3 14 51 555
1933 90 20 15 0 0 0 0 0 0 0 116 57 297
1934 123 111 45 48 3 0 0 0 0 3 57 49 438
1935 74 57 16 13 0 0 0 0 0 4 57 64 285
1936 107 48 270 29 34 0 0 0 0 3 15 84 589
1937 60 169 39 10 0 0 0 0 0 0 17 44 339
1938 139 46 20 48 0 0 0 0 0 33 115 100 501
1939 62 216 43 1 0 1 0 0 10 22 36 101 492
1940 69 57 153 91 0 0 0 0 4 11 15 44 443
1941 151 60 0 10 0 0 0 0 0 41 36 57 355
1942 32 72 121 10 14 0 0 0 0 31 1 97 376
1943 93 11 58 82 18 0 0 0 0 5 24 72 364
1944 146 319 16 7 0 3 0 0 0 16 38 38 583
1945 18 43 162 1 2 0 0 0 0 15 48 81 370
1946 396 95 0 0 4 0 0 0 0 10 22 16 544
1947 84 32 125 0 0 0 0 0 0 1 92 102 436
1948 111 175 172 68 0 0 0 0 0 23 62 20 630
1949 82 8 128 1 0 10 0 0 0 5 62 133 429
1950 107 119 40 38 28 0 0 0 0 0 10 130 472
1951 101 73 51 37 23 0 0 0 0 61 70 89 505
1952 92 108 10 0 10 0 0 0 0 33 170 78 501
1953 93 157 77 21 1 0 0 0 0 5 77 137 568
1954 142 97 87 16 0 0 0 0 0 9 12 223 586
1955 183 246 168 20 2 0 0 0 0 32 48 82 782
1956 54 129 53 26 3 0 0 0 12 4 42 64 387
1957 72 105 92 15 0 2 0 0 2 29 28 96 441
1958 231 125 89 3 0 0 0 0 1 54 36 102 641
1959 133 71 66 13 5 2 1 0 0 2 18 80 391
1960 31 73 21 71 16 3 0 0 0 3 54 49 322
1961 112 133 145 12 24 0 5 0 0 4 30 35 501
1962 155 36 8 41 0 0 0 4 0 4 80 114 442
1963 209 58 67 13 5 0 0 0 0 30 99 209 689
1964 79 70 18 0 0 0 0 0 0 16 50 50 284
1965 33 54 6 58 0 0 0 0 3 9 36 53 251
1966 107 160 124 60 0 17 0 0 29 0 3 108 607
1967 137 186 21 120 0 0 0 0 0 9 112 64 650
1968 110 104 82 66 34 0 0 0 0 2 70 60 527
1969 50 247 37 15 0 0 0 10 21 13 68 13 473
1970 123 21 9 1 0 0 0 0 11 0 60 160 386
1971 108 12 33 45 0 0 0 0 2 10 67 94 371
1972 260 70 193 20 0 0 0 0 0 11 2 47 602
1973 81 68 22 15 0 0 0 0 0 101 44 270 600
1974 342 366 18 50 0 0 0 0 11 7 116 31 940
1975 169 61 191 71 4 0 0 0 0 1 11 71 578
1976 113 68 108 4 3 0 0 0 12 25 74 38 444
1977 115 126 127 19 4 0 0 1 32 3 77 227 731
1978 92 226 37 12 14 12 0 0 0 11 18 73 496
1979 125 30 26 7 0 0 0 0 0 27 20 37 272
1980 91 291 43 1 0 0 0 0 2 3 74 28 532
1981 126 161 93 5 6 0 0 0 0 8 75 61 535
1982 31 17 11 15 0 0 0 0 0 101 84 35 294
1983 99 11 25 20 16 1 0 0 0 20 54 153 398
1984 20 18 82 8 1 0 0 0 1 20 93 13 255
1985 77 42 55 2 0 0 0 0 0 22 8 135 341
1986 54 70 35 52 11 0 0 0 19 50 49 57 397
1987 32 60 27 0 0 0 0 0 4 9 6 92 231
1988 38 107 80 19 0 0 0 0 8 17 15 70 353
1989 216 180 44 96 0 0 0 0 0 0 7 40 584
1990 84 110 33 17 0 0 0 0 0 17 6 42 309
1991 178 134 123 0 0 0 0 0 6 29 13 78 560
1992 50 7 87 5 0 0 0 0 1 7 41 55 254
1993 67 122 15 47 0 0 1 0 8 14 24 155 453
1994 275 50 6 0 0 0 0 0 0 0 44 15 390
1995 27 14 48 0 2 0 0 0 22 11 17 54 195
1996 139 261 9 0 3 0 0 0 1 4 105 46 567
1997 154 19 65 2 5 0 0 0 13 27 25 68 377
1998 150 19 28 33 0 0 0 0 0 2 51 109 391
1999 110 57 41 4 10 0 0 0 0 8 14 64 308
2000 255 193 33 22 2 3 0 0 0 6 44 4 561
2001 34 104 48 88 4 0 0 0 0 22 126 41 467
2002 53 45 15 6 0 0 0 13 0 31 69 55 288
2003 28 52 6 40 0 3 0 0 0 0 31 73 232
2004 147 101 151 14 0 0 0 0 0 0 4 15 432
2005 95 46 51 20 0 0 0 0 0 0 124 81 416
2006 139 247 75 10 8 - - - - - - - -
MEAN 110 99 67 24 5 1 0 0 3 16 47 77 446
SDEV 70 79 61 28 10 3 1 2 7 20 36 50 143
CV -% 64 80 91 114 203 366 689 475 218 125 76 65 32
MAXIMUM - 1977 396 366 274 120 62 17 5 13 32 101 170 270 940
MINIMUM - 1995 14 7 0 0 0 0 0 0 0 0 1 0 195
Shakawe – Rainfall 1932 to 2004

1922 - - - - - - - - - - - - -
1923 - - - - - - - - - - - - -
1924 - - - - - - - - - - - - -
1925 - - - - - - - - - - - - -
1926 - - - - - - - - - - - - -
1927 - - - - - - - - - - - - -
1928 - - - - - - - - - - - - -
1929 - - - - - - - - - - - - -
1930 - - - - - - - - - - - - -
1931 - - - - - - - - - - - - -
1932 137 39 269 11 0 0 0 0 0 2 16 92 566
1933 131 33 29 0 0 0 0 1 0 4 53 129 379
1934 222 227 131 18 1 0 0 0 0 28 42 136 806
1935 - - - - - - - - - - - - -
1936 194 29 319 24 16 0 1 0 0 2 19 76 680
1937 - - - - - - - - - - - - -
1938 66 223 25 9 0 0 0 0 0 43 152 162 681
1939 - - - - - - - - - - - - -
1940 - - - - - - - - - - - - -
1941 - - - - - - - - - - - - -
1942 - - - - - - - - - - - - -
1943 - - - - - - - - - - - - -
1944 - - - - - - - - - - - - -
1945 101 35 125 0 9 0 0 0 0 13 93 120 495
1946 348 82 0 12 0 0 0 0 0 0 23 33 499
1947 56 49 105 0 0 0 0 0 0 12 21 114 357
1948 181 402 148 22 0 0 0 0 0 0 35 35 822
1949 39 18 112 0 0 0 0 0 0 1 39 113 321
1950 - - - - - - - - - - - - -
1951 - - - - - - - - - - - - -
1952 107 196 0 0 0 0 0 0 0 28 27 85 443
1953 - - - - - - - - - - - - -
1954 98 142 135 11 5 0 0 0 0 67 292 186 935
1955 139 190 167 39 3 11 0 0 0 15 30 144 737
1956 - - - - - - - - - - - - -
1957 100 182 101 3 0 0 0 0 3 51 87 89 614
1958 160 225 62 25 0 0 0 0 15 0 39 201 727
1959 267 158 114 0 8 0 0 0 0 6 11 129 691
1960 65 227 31 51 18 0 0 0 0 6 69 64 530
1961 105 128 149 88 2 0 0 0 0 0 19 29 520
1962 238 139 25 12 1 0 0 2 0 0 57 111 585
1963 87 75 105 33 0 0 0 0 0 13 148 110 570
1964 56 12 38 4 0 0 0 0 0 8 118 102 339
1965 80 51 66 3 0 0 0 0 0 0 62 49 309
1966 128 237 94 68 0 4 0 1 44 9 26 55 665
1967 159 91 82 130 1 0 0 0 0 10 122 103 697
1968 85 90 51 100 17 0 0 0 1 9 65 78 495
1969 54 69 78 18 1 0 0 0 1 19 11 64 316
1970 88 98 5 14 0 0 0 0 0 12 34 62 312
1971 275 120 57 10 0 0 0 0 0 2 81 153 697
1972 245 53 202 20 0 0 0 0 0 4 0 80 602
1973 85 170 39 1 0 0 0 0 0 42 93 226 656
1974 369 204 27 134 7 0 0 0 3 15 38 108 905
1975 97 66 119 33 5 0 0 0 0 0 23 93 436
1976 172 104 88 13 0 0 0 0 29 15 90 54 566
1977 85 355 117 29 1 0 0 0 2 0 54 257 900
1978 181 218 110 11 0 0 0 0 1 30 55 46 652
1979 147 78 11 34 0 0 0 4 0 19 51 55 398
1980 45 173 61 25 0 0 0 0 3 17 114 157 595
1981 165 94 68 0 1 0 0 0 0 6 88 51 473
1982 64 87 58 25 0 0 0 0 0 10 98 51 393
1983 105 28 78 11 4 0 0 0 0 20 88 237 571
1984 23 86 131 21 1 0 0 0 0 13 47 27 348
1985 177 190 18 2 0 0 0 0 0 28 13 76 504
1986 175 67 47 23 0 0 0 0 28 66 87 53 547
1987 20 91 21 0 0 0 0 0 6 16 42 32 228
1988 80 161 56 68 0 0 0 0 10 26 30 73 504
1989 330 241 113 47 0 0 0 0 0 3 28 76 838
1990 235 33 78 46 0 0 0 0 0 5 23 100 520
1991 124 140 46 0 0 0 0 0 0 15 29 216 569
1992 108 0 157 24 0 0 0 0 1 1 84 159 534
1993 181 136 37 16 0 0 0 0 4 7 29 74 485
1994 263 56 1 34 0 0 0 0 0 0 40 77 472
1995 12 142 53 1 5 0 0 0 0 8 1 102 324
1996 126 109 23 1 0 0 0 0 2 1 34 73 368
1997 173 117 163 0 3 0 0 0 4 18 11 51 539
1998 91 75 8 1 0 0 0 0 0 11 105 61 352
1999 134 12 80 0 1 0 0 0 0 4 19 151 401
2000 283 285 39 0 0 0 0 0 0 0 12 141 760
2001 46 111 38 0 0 0 0 0 0 12 60 146 412
2002 60 34 44 10 0 0 0 0 9 0 17 79 253
2003 55 40 12 21 0 3 0 0 0 9 9 64 213
2004 131 107 134 122 0 0 0 0 0 0 40 86 620
2005 - - - - - - - - - - - - -
2006 - - - - - - - - - - - - -
MEAN 137 122 80 24 2 0 0 0 3 13 55 101 536
SDEV 83 85 63 32 4 2 0 1 8 15 48 54 174
CV -% 61 69 79 133 225 508 603 434 288 119 87 53 32
MAXIMUM - 1954 369 402 319 134 18 11 1 4 44 67 292 257 935
MINIMUM - 2003 12 0 0 0 0 0 0 0 0 0 0 27 213
Appendix 3
Responses from Interviews - Rural Communities
1 Reasons for burning – get rid of old grassland – both areas, swamps and sandveld - no
uncontrolled burning – only with permission
2 Poachers burn to track animals – they cause most of the fires in May, August & September. Hunt
buffalo
3 Both hunters and poachers burn to attract wild animals
4 Suspect photographic concessionaires also burn
5 Incidence of lightning fires is very low
6 Burning for crops is controllable because people get permits
7 The Government, not others, must construct firebreaks and look into forming committees
8 Dry areas – burning for livestock? – no
9 Wildfires burn often during the year – mostly trans-boundary fires so they burn from the same
direction
10 Thatching grass – used to burn to improve thatch grass but now they have education so they
harvest at the correct time so burning is not necessary
11 Conservation Act is a source of information on correct and wise land use
Okavango Community Trust and Okavango Polers Trust have regulations re wise use of
12
resources
Harvest reeds for subsistence and commercial uses but use conservation guidelines so only
13 harvest mature reeds. People used to believe it was necessary to burn reeds but now they know
better and also it takes longer for the reeds to regenerate if they are burnt
14 Chief doesn’t have authority to issue burning permits – have to go to Maun
15 Ways of making it easier to get permits – Agriculture demonstrators are based in most areas so
should be given authority to issue authority
16 There is knowledge transfer through workshops, Kgotla meetings and Trusts and the people are
already becoming aware so they are using fire less
17 Funds to conduct workshops more frequently should be made available
18 They report wildfires (trans-boundary fires) to the DFRR in Maun
19 Access is very difficult so fire fighting is very difficult and there is only one firebreak
20 Lots of poachers come from Namibian side so lots of fires originate from that side although they
blame Botswana
21 Prevailing winds are from the North East
22 Equipment is available in Maun plus one unit in Shakawe
23 Drip torches are not used because of time
24 Main effort is fire suppression, no back burning for wildfires
25 Knapsack sprayers too heavy to use in the sand
26 Have lots of beaters in Maun otherwise use branches
27 No uncontrolled fires in the area since 1983
28 Fires originate in the Delta and come South East into chief's area of jurisdiction
29 Community and Government fight fires
30 Don’t use fire currently but if need to use fire will ask permission from the chief
31 Chief mobilises people to control any wildfires
32 Use fire to clear fields after harvest
33 Don’t burn to improve thatching grass quality following year
34 Don’t burn because value range resources especially for their livestock
35 Do they use fire to prepare areas for cropping? – they do pile up the cleared bush and burn only
with Chief’s permission and the assistance of the Community
36 Use 20 –30 people as required when burning fields

37 Burn July to September
38 Burn the same time to get rid of residual material
39 Burn for draft power ploughing – graze fields first with cattle
40 Burn only when above normal rainfall
41 Burn early morning – 7am – 9am and late afternoon – 4.30pm – 6pm
Communities are aware of the Herbage Preservation Act – knows the conditions of the Act and
42
that they have to report if they want to burn
43 According to the Act the Chief has authority to grant permission to burn
44 There are lots of firebreaks in the Shakawe area because fire is a problem
45 Have livestock, fishing, thatching and hunting by those given quotas
46 Don’t use fire because if burn veld livestock will perish
47 Fire is a big problem because destroys forage + wild foods + trees
48 Would like to see fires stopped because of destruction
49 Culprits are not caught
50 Chief believes in the past hunters used to burn veld but now all farmers burn for is ticks etc
51 Realize careless handling of fire causes wildfires
52 Cigarettes, wild honey harvesting and campers cause wildfires
53 Chief has no idea of what causes fires in Tsodilo area
54 Use fire for clearing fields – stump, pile and burn but if fuel load is too high fire might get out of
control
55 Burn June to August. Only when expanding fields or if allocated a new field
56 Burn from 9am and 3pm
57 Fishing – no idea if fishermen use fire but does accept that if fishermen do use fire wildfires may
result
58 Thatching – burning thatch depends on the attitude. Jealousy may cause people to burn thatch
because someone has over harvested the thatch
59 No burning for thatch management
60 Herbage Preservation Act – aware that burning veld is an offence and also that burning is
detrimental
61 Chief is aware that permission for stumping is allowed
62 Would he like devolvement of power from Maun to district level? – Yes
63 When someone in Shakawe area burns the person should be accompanied by a police officer
and the Agricultural officer but this does not really happen
People help fight fires because it is an offence to refuse, those refusing are dealt with by the
64
Kgotla
65 No equipment for fighting wildfires
66 Do those burning for clearing fields have any fire fighting equipment? No – people are alert to the
fact that they should not allow wildfires to develop because they will be liable for prosecution
67 There are there stipulated steps available and before issuing permit letter the Chief informs the
applicant what steps should be taken to control wildfires
68 Who ever volunteers to fight wildfires should be paid. Currently they are only given food and water
69 Government employees are given an allowance if they are out fighting fires so the Government
should give an allowance to the communities too
70 In below average rainfall if they burn the veld the cattle suffer because the veld is less
71 Cattle do better in areas where there is no fire
72 Cattle in Tsodilo area OK – haven’t suffered too much from frequent fires
73 Cyperus swamps – no idea where fires are from or why swamps burn

74 Chief is worried about the fires in the swamps – caused by people hunting in the swamps but the
culprits are not caught
75 Locals use fire for robbing and controlling bees and poaching
76 90% of fires are deliberate
77 Locals cut reeds then deliberately burn so others can't cut then they have to buy from the person
who has already cut reeds
78 Poachers burn to distract authorities from where they are actually poaching
79 If a snake goes into a tree people will burn to get rid of it
80 There is no controlled burning
81 Not only locals responsible for burning - commercial hunters, in particular, are also responsible for
fires
82 Subsistence poaching has not really stopped - legal allocation of 100 rounds of
ammunition/rifle/year
83 Most of the fires are June & July
There is no fire equipment in Seronga and no people trained in fire fighting, they just use
84
branches
85 Voluntary fire fighting will not work, people will not operate without any formal reward
86 Fire in the papyrus does have an effect on pythons, sititunga and lechwe
87 1950 - 1958 there were very few fires in the Delta
88 In Etsha 6 there is controlled burning on floating islands to burn the reeds so that sugar cane can
be planted
89 There are areas in the Delta where methane gas has been bubbling out for 5 - 6 years - if a
wildfire burns into one of those places it will burn for years
90 Sititunga feed off green shoots of papyrus and breed in October
91 Burnt Papyrus shoots within one week
92 Safari companies should have 5 people/lodge trained in fire fighting/management but they must
have a dynamic person co-ordinating the fire management plan for the Delta
93 Pilots should be co-opted for an early warning system for wild fires
94 Seronga could be a base for the eastern side of the Panhandle for community co-operation in fire
management/fire fighting
95 There are very few trans-boundary fires from Namibia
96 Thatch grass cutting starts on the flood wane and before the first rains in July/Aug/Sept then fires
start
97 BDF is used for anti-poaching. They could also be used for training for fire co-ordination but the
Top Management must buy-in and also Ian Khama for it to be effective
98 To control poaching and hunting increase the price of ammunition and also be very strict with
issuing of licenses
99 Review firebreaks and realign them
100 Channel money spent on private contractors constructing and maintaining firebreaks to local
communities
In the Tsodilo area the Bushmen burn a lot. Gudigwa Bushmen also burn, mainly for honey
101
hunting
102 In August it is very windy - exacerbates wildfires
103 In 1958 there were very few sititunga in the Delta, crocodile hunting started upsetting the food
chain which led to an increase in sititunga
104 Squeakers die from ash not bigger fish
105 Rate of spread of fire in papyrus is 3 - 4 seconds on one spot

106 Papyrus will burn annually but will not burn 2x in one season
107 The biggest threat to the Delta is water extraction in Angola and Namibia, not fire
108 The Ramsar Site is densely populated and there will be impacts on the ecosystem
109 The urbanization shift is starting to reverse as more people are starting to stay in the rural areas
110 Don't put natural resources and wildlife before people who lived here long before wildlife
management areas were declared
111 Cattle numbers declined in the area since the lung disease outbreak, cattle do graze on the flood
plains until it rains then they go out into the back country (drier areas) to graze
Appendix 3
Responses from Interviews - Commercial Stakeholders
1 Burning permits usually have no criteria for burning
2 Some applicants say why they want to burn based on personal knowledge and visual assessment
3 Need scientific criteria for burning the permanent and seasonal swamps
4 Papyrus swamp – fast regenerator whereas vegetation on islands regenerates much slower
5 Hence base assessment for burning on most the fragile component of the vegetation
6 Grasslands – use moribund status
7 Annually swamped flood plains get moribund and this prevents new growth
8 More regularly flooded areas have less moribund grass
9 Rate at which new regrowth can access light – 3 variations in flood plains
10 Tremendous amount of fire used in the Delta without applications for permits
Photographers + hunters – tremendous amount of fire used by hunters – not professional hunters
11
tho
12 Trackers will purposefully burn without permits (can’t blame them)
13 Fire may also be used by the Photographic industry
14 The ecosystem is very resilient and has a tremendous ability to recover
15 There is not an over-utilization of fire in the Delta
16 Time of use is a big problem – spatial + time
17 Southern end of Delta burnt more often than Central part
18 Moremi very protected – does not need burning
19 Panhandle – more fires because more people
20 Wildfires due to neglect – people leave fires burning
21 Peat fires in the Boroga area
22 Dry areas – more utilization by people therefore there are less fires
Elephants are replacing fires with regard to structural balance of the vegetation. At the Savuti
23 Channel the Acacias have not moved in because of the elephants, in the Boteti there are no
elephants so the Acacias have grown into the river
24 Purpose for use of fire outside of permitting – lots of people burn
25 There is a loose allocation of causes of wildfires – if the villagers are cutting reeds then the fire is
ascribed to the nearest village
26 In the drier areas – in the 1990’s flood areas were small. Early burns were in May and black
stumps could be seen in the loose sand
27 Winter winds blow the sand away and also the grasses. Since there is lots of grass seed there is
recovery
28 Fire causes bottleneck dangers when there is lots of game that needs grazing
29 Fire in the Linyanti will be a problem because it is much drier
30 Fire in May, June & July leaves the vegetation exposed to winds until September to October
31 Fire is part of the ecosystem and the fire frequency is not as high as people think
32 Need to add figures to fire occurrence
33 Anthropogenic fires are part of the system
34 The current perception is that burning is not allowed
35 In management plans it is stated that burning is not allowed
36 Guidelines in management plans are specifically to eliminate fire
37 There is no longer a committee issuing permission to burn
38 It is necessary to add criteria to the Herbage Preservation Act because even if there is a committee
issuing permits they may not always be informed or aware of the criteria

39 It would be better to have a responsible person appointed to issue permits who is informed of all
the criteria and can take responsible decisions
40 There is a lot of research being done on burning – is there any collaboration?
41 South African National Parks have a pretty defined policy re burning – it may also be applicable to
the Delta
42 There are no long term data sets available for the Delta
43 The Delta with its water patterns would be ideal for patch mosaic burning and where it needs to
burn it will burn
44 Africa has a fire climax vegetation – need criteria for burning
45 No research available on burning in actual permanent & semi-permanent swamps
46 There is concern that frequent fire is affecting endangered species habitats
47 Reduced habitats are rather due to reduced swamps and elephant damage rather than due to fire.
Reed beds were destroyed in 3 years by elephants
48 Blockages may also be a cause of loss of habitats
49 Climate change and swamp dynamics are the main effects on the Delta
50 Must have goals – start with how much management is enough?
51 Apriorie is significant – how long a time span is acceptable?
52 Landuse plan is a priority – tourism/wildlife then goals to aim for
53 Is the current status acceptable? What is the trend and what pattern accompanies change?
54 Use adaptive management to achieve the objective
55 No need to control wildfires – need to assess whether they damage and alter the vegetation
56 Population density may cause a higher frequency
57 The change from rural to commercial ventures especially fishing and reed cutting have resulted in
more fires
58 Spatial fire patterns have changed
59 Adaptive management needs to be applied not only to the whole Delta but also to specific areas
60 Use fire for adaptive management ie. To open up water channels
61 The issue is how fires start
62 Total control is debatable
63 A fire management plan may have some influence on when and how fires start
64 It is very difficult to assess what actually happens with fire in the permanent swamps
65 It is important to establish where fires are coming from
66 Need adaptive management but communities will not easily accept or understand adaptive
management
67 Individual perceptions re whether fire is good or bad – not a community decision whether fire is
beneficial or not
68 Had education program with the ARB but it was not successful and it fell flat
69 Must have input from stakeholders to get buy-in
70 Need long term research on effects of fire on permanent swamps
71 It is a complex system and effects maybe confounded by hydrology and not just fire
72 Papyrus area did burn and has regenerated after 9 years
73 Trans-boundary fires from Namibia in dry years is a big problem
74 Over past 20 years the annual onset of fires has been early – mid April
75 NORAD provided nets and boats to fisherman and this has resulted in an increase in the number
of fires

76 Small boat propellers get stuck in the weeds so the fishermen burn for access and they can sink
their nets deeper if they burn
77 If have a set of criteria for burning may eliminate randomness and start trends
78 There are more penetrating and more damaging fires later in the season
79 If there are early burns close to water where the animals come to drink they will keep the grass flat
80 If burn big enough area the animals will survive. If burn small patches they tend to get hammered.
The recovery tho is remarkable
81 Fires don’t have a major effect on the grass but trees are often affected
82 Specific area exclusions should be included in criteria ie. Slayt Egret, Wattled Crane and Sititunga
habitats
83 Control of wildfires is very difficult and graded firebreaks are very damaging to the ecosystem
84 Tourism training – each concession must have dedicated fire officer with training in order to get
permission to burn
85 Most Lodges have fire training and capacity to protect Lodges and have agreements for combined
fire fighting if threatened
Working-on-Fire system will be the only way to manage and control fire in Botswana. Would need
86
a central pool of equipment and lots of money to move teams and equipment to make it effective
87 Training and capacity needs to be included in BWTI curriculum
88 Need for training course for guides in Jan/Feb
89 Camp managers and guides should be trained on an annual basis
90 Dept of Agric have 16 fire fighting units but need training in use and maintenance
91 Need for training for DWNP
92 If Botswana Wildlife + Concessionaires have fire fighting plans and training it will mean that 70% of
the Delta will be starting to implement fire control and management
93 Can burn papyrus every 6 weeks
94 Papyrus has volatile component so burns when green
95 Cannot control burn because access to papyrus beds is problematic – they are thick, one sinks into
the water, can’t cut access channels and once alight the fire escapes quickly
96 The heat from the fire is also prohibitive to fire fighting
97 Locals set continuous line of spot fires when burning the papyrus
98 Fires originate locally and do not necessarily originate from the other side of the Delta
99 Check Ngarange area on the satellite images for fires - lots originate from there
100 Fire is a natural part of the ecosystem
101 The Delta is a dynamic system and is never the same from year to year
102 The dynamics are water dependant and the area of the Delta depends on the amount of rainfall in
the area and the level of the floods coming in from Angola
103 When it burns in the papyrus it helps with opening up the channels
104 Fires in the papyrus only burn to the water level and the roots are not affected
105 Burnt papyrus regenerates within 3 weeks
106 There is spotting during papyrus fires as one often sees an explosion of the papyrus and the fire is
thrown forward
107 Sititunga graze the new regrowth
108 When it burns in the bushveld the fires are usually quick and burn mostly grass
109 Fires do not have very drastic effects on the Riperian woodlands
110 Lightning is often the cause of wildfires both in the woodland and in the permanent swamp

111 If there were no regular fires the fuel wood would build up and when there eventually was a fire
from some cause there would be a very hot fire that would cause a lot of damage to the trees
112 Elephants damage the bigger trees and along with fire may cause die-off
113 The increase in the population density does not mean there has necessarily been an increase in
the frequency of fires
The younger generation has urban inclinations so moves to the villages and town so the population
114 in the actual Delta itself has declined rather than increased so population increase doesn't mean
more fires
115 The prevailing wind is North East so trans-boundary fires are more of a problem for the Namibians
rather for Botswana
116 Fire every 3 – 4 years is good
117 Bayeyi culture is to burn even if they don’t use the area
118 Pompom – used to burn every year – burnt by the community & concessionaire
119 Fire is needed in Moremi at North Gate area
120 West Gate – flood plain desperately need fire – hasn’t burnt for years
121 Historically Moremi was the hunting area for the Bayeyi – they used to burn every year
122 When this was stopped the channels blocked up and Ficus vericosa started growing. The flood
plains didn’t get water
123 Moremi and Savuti occasionally need fire to keep channels open
124 When Savuti had a burn the game increased greatly
125 There was also a need to burn for aesthetic reasons too
126 Fires came from the north with the swing in the winds before the rains
127 Need to keep important channels open ie Mubabe
128 Papyrus fires – locals burn reed beds and the fires spread to the papyrus
129 Papyrus burns every year but not annually, different areas burn – depends on flood levels
130 Tubu – east of Gumare – agric, cattle + grass cutting + accessiblity - reasons for burning
131 Xaxaba Camp – fly – Nsxehgha village behind camp – hunters that burn willy nilly
132 Hunters like to burn from the beginning of the dry season
133 Whole Delta burns
134 Moremi hasn’t had a good burn due for one
135 No plan re burning
136 The Delta will burn naturally every 4th year depending on the rains – maybe every 3 – 4 years
137 Lodges use corrugated iron because of fire
138 Prevailing wind is SE, Wind blows from end of July, Aug & Sept. Wind velocity increases during
wildfires
139 As soon as water levels drop, papyrus dries then fires start - Oct & Nov
140 Are fires natural? They kill everything - pythons, sititunga, birds
141 Lightning fires 20%, anthropogenif fires 80%
142 Local tribes always burnt to open channels and feed the fish.
143 Burning kills fish, it causes die-off because floods pick up the soot - toxic to fish
144 Scientists say it is lack of oxygen but locals say it is the soot that gets in the fishes eye and gills,
this leads to fish death
145 Fishing has declined because of overfishing - not fires
146 Burning papyrus does improve channel access
147 Subsistence fishermen burn for poaching too
148 The Delta is not in a stable state - it is going backwards

149 The population has increased 10x in the last 20 years - more fires
150 Must stop fires in the Delta
151 Reporting fires in the Delta does not help
152 Reporting fires - often only gets a response 3 days later because there was no transport
153 Reed cutters burn reeds for regeneration - does fire improve quality?
154 Can't have prescriptive burning because can't control fire
155 A community based fire management system might work if it is done properly
156 Time of burning will vary because of differing flood levels - high water areas won't burn
157 Papyrus fronds die and fall, as soon as water level drops below mat it will burn
158 Need heat to make green papyrus burn
159 Can see oily film on the water when papyrus burns
160 Mbukushu live in area, not Batawana - burning is part of their lifestyle
161 Fire awareness projects in schools would work. Kgotla meetings may have some effect
162 BDF patrol have had an effect on poaching - the people are now scared of being caught poaching
163 In 1968 hunting was the reason to burn - it attracted game and made tracking easier
164 Fire frequency has not changed
165 Water levels affect fire frequency - in the drier years there is more fire
Fire does impact on tourism, especially mokoro safaris - but burnt swamps regenerate within 6
166
months
167 Farmers burn when they want to plant
168 Molopo farmers don't worry about the fires
169 There is no information on the effect of fires on pythons
170 Sititunga numbers are quite good, they breed in Sept so late fires impact on breeding
171 Locals say burning for fish increases the numbers of fish
172 4 - 5 000 fish die off - rotting vegetation and ash causes salinity levels to increase from 4 - 7%.
Fish die from ash in their gills
173 Community Trusts do not function well - too much bribery and corruption, Trusts do not always
help the actual communities
174 If farmers want to burn for cropping rather get the ARB to burn safely for them
175 Report wildfires to Mr Maswabi at ARB
176 Can burn annually - areas burnt this year won't burn next year
177 Fires in the Delta cause their own meteorological conditions
178 Children burn for poaching in school holidays
179 Fishermen smoke fish on islands - cut trees down to smoke fish because green palms don't burn.
These fires are also a source of wildfires
180 Use a helicopter to monitor wildfires and find source of fires - can't catch guys in mokoros who are
responsible for setting fires
181 Impose heavy fines for people caught setting fires
182 Pete Smith said there was no damage to papyrus if it burnt because it regenerates within 90 days
183 It would be a good idea to have controlled burns in the papyrus, if possible, to reduce the fuel loads
184 Burning papyrus is a problem because the fires are uncontrollable
185 The time of day that management burns to be applied in the papyrus is very important
186 Stakeholder agrees with the proposed rule that if in a sample of 10 papyrus fronds 4 are dead then
controlled burning would be recommended
187 The existing fire regime is wrong because people burn in the same place every year, they should
rotate areas that are burnt
188 There should be prescribed burning to control huge wildfires

189 Fighting huge wildfires in remote areas is impossible. Rather have a fire mosaic to break up huge
run-away fires
190 Must manage wetlands because there is a lot of ground moisture so fire is feasible
191 Concessionaires must have a management plan that includes fire and they must be responsible for
implementation with monitoring by the Government or some other agency
192 The perception that fire is damaging the Delta is not based on information, rather it's a perception
193 Fire is necessary in the Delta – people must be responsible
194 The DFRR should be the focus for implementation of fire management
195 Central co-ordination for fire management is very important because season of burn is critical
196 Season of burn depends on land use
197 Consultants must recommend training for various stakeholders - private, Government,
Communities
198 CBRM are setting criteria re pristine sites - fire management plan must take these into
consideration
199 People must submit a burning plan in order to get a permit to burn
200 Must have decentralized burning equipment
Burning channels will not necessarily open channels because the papyrus on the edge is moist
201 and lodges rather than burns off thus more likely to block channels. If burn the sawgrass behind
the papyrus - that will open channels
202 Papyrus that has not burnt for 7 - 10 years does not look very different to that recently burnt
203 Most burning in the Panhandle is in October - this is not a problem for tourism because in 3 weeks
it is green again
204 In the fan part of the Delta where lots of grass is burnt and there are lots of tourists it is a problem
205 Fire awareness program must be through schools
206 Awareness at Kgotla meetings not so effective because news gets mostly only to those who attend
the meeting
207 The 2004 fire awareness program caused much less fires but the mosquito problem was huge.
Also there was no rain so the locals won't stop burning because they think there won't be any rain
208 Papyrus forms a mat up to 1 meter thick, 150 days is the life of a frond, if it burns it takes 150 -180
days to get back to a burnable status
209 Phragmites - 200 days before can burn again
210 Can burn under the right conditions to control papyrus - definitely no wind
211 It would be preferable to burn with the right atmospheric conditions and control fires than to have
wildfires
212 Burning Papyrus creates its own whirlwinds
213 In the Delta at 17h30 a wind comes up and this creates problems with burning in the late afternoon
214 The season of burning is very, very important in the Delta, in the seasonal and permanent swamps
due to the water, they can burn every year because of the regeneration capacity of the vegetation
215 The hunting season opens 16th April, in dry years burning starts then
216 In occasionally flooded swamps moribund material can burn once in 4 -5 years
217 Because of elephant trampling there is very little vegetation left to burn in dry areas
218 Lechwe love the green flush of burnt Imperata cylindrica
219 Ditshiping area is the only area in the Delta that has too much burning
220 Reed cutters burn - they don't intend to burn large areas but the fires get out of control
221 The fence against the northern boundary prevents elephant migration into Namibia so they walk up
and down the fence and create a firebreak

222 Fires only come into Botswana from Namibia on the western border in wet years
223 The fire management plan must include who will manage the implementation of the ODMP plan at
various levels
The Botswana Government, IUCN or Ramsar committee should provide funding to appoint a
224 manager who will check eg vegetation levels etc and organize everything in order to apply
management
225 District Wetland Committee is purely advisory, they have no funds to implement anything
226 Use the Herbage Preservation Act to form sub-committees to manage fire
Currently the ARB has the responsibility of fire management but budget is limited and also
227
manpower
228 Form a Fire Management committee of all Lodges under the auspices of the Herbage Preservation
Act
HATAB has Lodges Division - this may be a means of forming a Fire Management Committee with
229 5 people trained in fire management/fighting per Lodge, this will mean that there will be 200 people
in the Delta with relevant fire training
230 The change in land use from hunting to photographic safaris has resulted in a reduction in the
incidence of fire
If fishing quotas were issued and there were concession rights for harvesting reeds and thatch
231 grass where commercial harvesters would have to pay a fee for the concession area, it may
reduce the incidence of wildfires because people would value and protect their areas
Appendix 3
Responses from Interviews - Government Departments
1 Fire occurrence – extensive and broken up, air will be the only way to get an over view
2 In the Park – chunks of permanent Delta burn annually. Important to get the size and where
3 Papyrus & Phragmites – impacts minimal, recovers quickly.
4 Impacts on the sititunga – negative because fire reduces shelter in the drier parts of the year
resulting in them moving to the open plains which results in increased predation
5 With no burning Papyrus gets moribund – needs to burn every 3 – 5 years, not too large an area
and in a patch mosaic
6 Seasonal Plains – annual burning has negative impact on the species composition & basal cover.
Palatable species are declining.
7 Current Park policy is fire suppression
8 With annual burning species composition changes to Imperata cylindrical and to pioneers
9 Seasonal Flood Plain – depends on management objectives of the DWNP
In the dry season there are huge increases in buffalo, wildebeest, zebra & elephant, the area is
10 heavily used - there is lots of grazing and trampling. There are too many animals at the most
sensitive time of the year.
11 In the wet season the grass is shoulder high but there are no animals, 4 months later it is over
populated
12 The fringes of the more permanent swamps are burnt and grazed
13 In the seasonal swamps there is nothing left to burn because of seasonal pressure of the animals
– nothing gets moribund
14 Biodiversity is declining but fire is not the primary cause – it is the high animal numbers
15 Quantitative criteria would be practical for Delta grasslands
Ngamiland has the same constraints as the Kalahari in South Africa – low and erratic rainfall. Fire
16
is not desirable
17 NB – Working-on-Fire must be non-departmental to be effective because government
departments are not functional
18 Policy of the Parks with regard to fire – fire is not seen as a management tool
19 Policy is control and suppression
20 Fence + fire causes animal deaths
21 Anthropogenic fires considered destructive
22 No burning policy for the Ramsar site
23 DWNP takes no responsibility for fires – that is the responsibility of ARB
24 No structures are in place for control – the Govt involves all to help fight fires
25 Try to construct fire breaks – done by the ARB
26 Firebreaks in the Delta are not possible – only in the adjacent areas
27 Ecologically sensitive sites – ask ODMP
28 Don’t see a necessity for controlled burning in natural areas
29 Allow lightning fires because they are natural
30 Concerned about germination of certain seeds
31 DWNP has policy of “no intervention” management
32 Lodges burn some places each year to attract game
33 Depends on management objectives of the area
Some areas should be burnt as the animals are under nutritional stress even tho the grass is
34
shoulder high
35 The Delta evolved with fire but fire should be prescribed and controlled
36 There should be a management policy for Moremi

37 DWNP has its own roads & firebreak crew but because the roads are so bad they are spending
more time on fixing the roads and won't get to preparing firebreaks this season
38 They should have a fire management plan for Moremi
39 There was a fire in Mboma Island in 2005 - cause unknown
40 Communities used to burn annually in Kwhai area
41 Wild fires in Moremi come from Pandamatenga, they burn through Nxai Pan and on into Moremi
42 In June/July huge areas burnt by these wildfires
43 Parks personnel and communities fight fires for weeks, they get sick and cough up soot and their
eyelids get burnt
44 No fire fighting equipment in Moremi
45 When fighting wildfires they just use branches
46 Botswana law says people have to help fight wildfires - only old and infirm are exempt
47 Recommend Moremi gets fire fighting equipment
48 Has departmental training in fire fighting – department was USA Forest Service trained in 1996
49 Burn for access for fishing and attracting fish – also makes floods flow in the channels
50 Sandveld – Gudigwa to Tobera – lots of fires, to scare away animals (lions & hyenas), poaching
51 Tsodilo area – fire is a big problem
52 Firebreaks are constructed East-West and prevailing winds are East-West so the firebreak is not
effective
53 Tsodilo community have a fire management plan
54 Tsodilo area fire is used for agriculture- burn the bush for access when the bush has dropped its
leaves, then cut down bush and burn piles on the sides of the fields
700 000ha burnt in the Tsodilo fire in 2005 from Xau Xau. August and November fires together
55
burnt 700 000ha
DFRR want to know the effect of fire in permanent swamps and rangeland and what actions to
56
take
57 Need fire policy that contains prescriptions
58 Need a management framework for the user so that officials can work with the people
59 There is a misinterpretation of the Herbage Preservation Act
60 The only office issuing permits for burning is Maun
61 There are Herbage Committee and Conservation Committee members in the villages but they
can’t issue permits, only the Secretariat in Maun issues burning permits
62 The DFRR is a new department but if they would have extension staff in the villages they would
the people to train and then give them the authority to issue permits
63 Conservation Committees are volunteers and they are not always effective so they would not be
the right committee to give authority to
64 Suggestion that the Delta be divided into Districts with responsible officers for each district
65 Problem is that even if issue permits the villagers don’t bother to get permits
66 Could be that villagers don’t know about burning and conditions attached to obtaining permits
67 Interpretation of the Herbage Preservation Act by both parties leads to misconceptions
Lack of firebreaks maybe related to getting permits eg Gomoti River area used to burn and they
68 have been stopped so now the channel has dried up and the villagers are now going to burn
without permits
The Range Ecologist in Gaborone is driving the monitoring programme for Botswana – they
69 measure canopy cover, species, forage quantity. It is done in March/April after the rains - not
used for burning tho
70 Fires in the Delta – very difficult to trace the cause or the person responsible

71 They charge persons who overstep the prescriptions of the Herbage Preservation Act
72 Chief also has authority to grant permission for burning fields – they burn late in the afternoon or
early in the morning
73 Villagers burn the seasonal flood plains in the dry season for grazing and for poaching red lechwe
74 Main fire season – end July to October
75 There is no fire officer in Seronga
Because of the merger with Forestry there is an office in Shakawe – does not know if there is a
76
post for a fire officer
77 Tsodilo area burns annually
78 National Heritage Area has been fenced only for one year – fence completed July 2005, concern
about lack of firebreaks
79 Tsodilo area - manage fire on a crisis basis
80 Tsodilo, 2004 – very dry – less fire
81 Tsodilo, 2005 – huge fire. Burnt in valley between Mother & Child - concern about destruction of
heritage site
82 Tsodilo area is 30 sq km – 3000ha fenced off round Heritage site
83 buffer zone <30m x 22 km around Tsodilo sanctuary
84 Because of the high fuel load and high fire hazard in the Tsodilo camp site and sanctuary, it was
important to introduce grazing animals to reduce the fuel load and the fire hazard
85 Fire in Tsodilo started 2 October 2005 and burnt for 3 weeks
86 There were three prongs of fire attack on the 2005 Tsodilo fire
87 First time fire burnt with a south wind, then fire direction changed and three days later there were
88 back fires
89 At night the wind was not too strong
90 At the time of the fire it was very very hot + 410 C, before the rains
91 They have ideas for fire management in the Tsodilo National Heritage Management Plan for the
area but they are not put in place - linked to ARB plan
92 Firebreaks – Ikoga and Samochima fences plus firebreak further south
93 Would like to have intensive fire management guidelines to deal with fire problems
94 To a cultural site fire is a problem but in the ecosystem it is part of the checks and balances
95 The 2005 fire impacted very badly on the Mukwa trees – there is no seedling regeneration
96 Would like a fire management plan that is practical – not just theoretical. Fire is a big problem
because of minimal human resources
97 No fire fighting equipment, little training. 12 staff, 3 stay in camp. 6 males and 3 females. Females
not expected to assist with fire fighting
98 Would be very keen to have training on fire fighting and management
99 No livestock losses in 2005 fire
100 People very lax in fighting fire
101 Conflict between community and ARB staff – community think that Government should be solely
responsible for fire fighting or else the community should be paid to fight fires
102 August to September people extend their fields through slash and burn
103 The population in the area is small but the fires start annually.
104 No burning by livestock farmers – no biomass problems because grass is available throughout
season
105 Little overgrazing because cattle numbers are fairly low - is grass to burn
106 Mukwa die-off is progressive – not only from 2005 fire, multiple fire impact
107 Local people come from Angola and Namibia and it might be part of their culture to burn

108 2005 – first time there has been a fire threat to the villages
109 NEED FIREBREAKS AROUND THE HERITAGE SITE
110 P1.3 million was spent on the fence and there is no provision for firebreaks – the current high fuel
load presents a major problem
111 Heritage Site has a big problem with budgets for maintenance
112 Communities believe that if they burn the channels it will keep them open - believes this notion is
wrong
113 In Guma & Etsha there are lots of fires - people burn areas and plant crops, maize, bananas, &
sugar cane - think that this is wrong
114 Channel blockages are a big issue - how can fire be used to unblock channels?
115 Frequency of fires? Papyrus regenerates quickly so can be burnt on an annual basis
116 Burning annually causes shifts in vegetation species
117 With regard to a forage score for seasonal swamps? - they are not good grazing because grass
quality is not high and also diseases in seasonal swamps are high.
118 Wildlife only graze seasonal swamps if there is no free standing water in the drier areas - they
prefer to graze dry areas
119 Concern about Slaty egret habitat - good stands of Phragmites no longer there
120 What desired state of the vegetation is important? What do you want? Slaty Egret or Sititunga
habitat or habitats favouring the communities?
121 Desired state of the vegetation depends on land use
122 Must have a fire mosaic
123 Socio-economic issues vs wildlife preservation should be considered, balance biophysical needs
with socio-economic needs
124 There was a high frequency of fire prior to 1982 when the buffalo fence was erected, in the
Tsodilo area
125 One can classify seasonal swamps as highly grazed
126 Can the whole Delta be classified as highly grazed? Tubu & Shorobe areas often overgrazed &
there is wind erosion
127 A community project, with 100 Polers, take mokoro trips from the Buffalo Fence to Chief's Island
128 Fishermen burn to get young shoots for bream
129 Polers also burn for tourism - it increases game and visibility
130 Polers burn in May/June before the new flood arrives when the swamps start drying out
131 Seasonal flood plains burn every year, if dry
132 Buffalo, elephant, hippo and antelope graze on the seasonal flood plains
133 The semi-permanent swamp areas at the southern tip of Chief's Island, under normal
circumstances burns almost every year
134 Cigarettes, hunters and trans-boundary fires from the north west are the main causes of wildfires
135 Photographic safaris - don't burn
136 Fishermen use fire to dry fish
137 Reed cutters burn to get good reeds, they harvest 6 months after the reeds have seeded
138 Lightning fires occur in the semi-permanent swamps
139 Reeds also provide habitat for fish
140 People don't take cognizance of traditional knowledge
141 Must institute laws for cutting grass and reeds and harvesting forest products so locals don't burn
at the wrong time and this will also prevent conflicts
142 Kgotla meeting are a means of knowledge transfer

143 Local people refuse to put out fires because they say the Government must pay them to assist -
politics is interfering
144 Kgotla meetings are not well attended because young people are not interested
145 Kgotla meetings for information transfer in the mornings are good - in the afternoon they are not
so good as people are very disruptive and disorderly
146 Use pamphlets, radio, TV, National radio and theatre for fire awareness programs. Theatre is a
very effective means of knowledge transfer but must get the people involved for most effect
147 Use out-of-school youths for actors in theater program - contact Botswana Youth Council
148 Local people burn big trees to make mokoros, especially in Ditshiping area
149 Local People used to burn thatch grass but this practice stopped because of species
disappearing, also due to cutting too green before grass seeded
150 The people are starting to understand the ecology and are making their own laws re harvesting –
Tubu
151 For co-operative fire fighting should have bases in Shakawe and Seronga, there are ARB
representatives in those two towns
152 The people want to form small groups to learn, and maybe form co-operative fire fighting teams
153 Game control is very effective by the anti-poaching squad and the BDF wildlife team so hunters
are not the main reason for wildfires
154 People report fires to the ARB but they won't identify the culprits for fear of retribution
155 Lots of fields destroyed by wildfires but not too many stock losses
156 Tsodilo fires come from Namibia
157 Livestock farmers do not burn
158 In winter the grass dies off, termites and wind take the grass so there is very little to burn
159 People will kill you if you burn livestock grazing
160 No fire in the Mopane areas
161 Fire came with development, first safaris came in 1962, fires started in the Delta from the 1970's
because of hunting
162 Hunters burn the swamps they are here to destroy and go
163 With the advent of photographic safaris there is less fire
164 In the 1970's there was little fire, in the 1980's it increased because of hunters but fire incidence
is still increasing
165 There are not more wildfires in the school holidays, children are taught about the dangers of fire
166 New Phragmites beds are developing again after being burnt
167 Reeds regenerate at the rate of 300cms in height per month
168 Nesting Quelias and elephants also destroy reed beds
169 We need to know if the reed beds burnt just before the new flood regenerate better
Appendix 3
Responses from Interviews - NGO's etc.
1 Thatch project in 4 villages – Tobera, Mogotho, Ngarange, Shakawe. Have problems with fire,
cutline serves as firebreak but there is no maintenance
2 Namibian side not maintained but Botswana side is better
3 Cutline is wide and they have no idea how it jumps
4 Trans-boundary fires plus those from Seronga and Beetsha are a problem
5 People use fire and allow escaped wildfires to burn
6 There is a cattle post at Shakawe and people living there use fire but they are not concerned
about it escaping
7 They need training in fire fighting and also a vehicle to take the wounded to hospital
8 There are no radios for communication to aid with co-ordination and call for help in case of injuries
9 The clinics are scattered and distant so treatment for any injuries sustained during fire fighting is
minimal
10 There are no firebreak lines in Shaikarawe, Tobera or Ngarange
11 Thatch areas are far from the village communities so the thatch areas are burnt before anyone is
aware of the wildfire – this is a problem
12 Need to harvest when the grass is mature – June to July
13 Fire in June when the thatch is dry is a big problem because the community can’t easily control
the wildfires
14 Jealousy is a big problem because those not getting income resort to arson
15 Fires set by hunters and poachers plus trans-boundary fires threaten the thatch
16 Because of getting income the communities now try to protect the thatch areas
17 They harvest an area every alternate year
18 Those who used to traditionally burn the thatch areas no longer do it partly because of the
information via the Kgotla that it is an offence to burn
19 Traditional fire management was to burn in November just before the rains
20 Also burn in November for the traditional plants
21 Previous fires that affected the thatch were caused by hunters
22 If the thatch is burnt they have to wait 2 years before they can harvest mature thatch
23 There is a big problem with fire and low rainfall years
24 Cutlines are put in by the Government, Forestry and the drought relief programme
25 DFRR cutlines on the Border are hired out to private companies – legal implications with payment
caused clearing of the firebreaks to be late
26 The cutlines were constructed in 2001 and were maintained from 2002
27 In 2004 there was no maintenance because of lack of funds
28 They need an area allocated just for thatch harvesting and a mandate for controlling these areas
– currently the thatch areas are grazed by the communities – ie they are multi-purpose areas
29 If they get a mandate to control certain areas they would:
1. fence them
2. build cutlines around the areas
3. have 5 representatives per area to alert the community if there is a fire threat
4. 15 people to stay at thatching areas for one month and the Trust will pay them
5. they will take people from each of the 4 communities to patrol the thatch areas so the will
need training and fire fighting equipment
30 The village management committees used to have fire management training from the DFRR
31 Communities need to approach the DFRR and also the ODMP may assist
32 DFRR now needs to address the communities themselves
Responses from Interviews - NGO's etc.

33 Communities need equipment
34 DFRR started to purchase equipment but the lady responsible for organising it has left so the
progress has halted
35 TOCADI has one tractor that is not strong enough to cart water or people in the heavy sand, they
also have another tractor and trailer
36 The meteorology Services in Shakawe have access to satellite imagery and can predict hot spots
but the information flow is not adequate for warning purposes
37 The information is available but they have no extension arm so the communities have to go to the
met office to get information on weather conditions and probable hot spots or wildfires?
38 Harvest thatch grass from end of July to beginning of November - don't want early fires
39 Could have central control office at Ngarange for co-operative fire fighting centre
40 Fishermen do burn in October in the Upper Panhandle
41 Can harvest reeds for two years then need to burn otherwise size and the quality is poor
42 Burn for reeds in October when the water levels drop
43 Trust is run by Board of Trustees – 4 villages each have two representatives on Board – no
particular fire message through Trust Workshops
44 Most of the fires are illegal – villagers burn after harvesting
45 Burning is not good because the area is very dry and fire does lots of damage - the bigger trees
suffer damage
46 Most of burning for livestock
47 Villagers also burn because they believe they will have better rains the following season
48 Also burn to open channels
49 Burn August to September
50 4 – 5 fires per month in different areas – not always the same area but different areas in the same
locality
51 Most of the fires burn in the late afternoon and burn through the night
52 August is very windy so it is impossible to control fires. Winds are North to North East
53 Winds are strong – 20 – 30 kms/hr
54 Flood plains burn August to September when dry
55 The regenerated grass provides good grazing as the soils are probably still moist
56 Burn Sandveld areas – Eretsha to Namibia
57 Only grade firebreak along the Border and it is graded once or twice per annum
58 Firebreaks in between the villages are not maintained
59 Not lots of hunting in Seronga area – not sure if photographic safaris burn
60 Lots of houses burn in the villages – mostly due to lightning
61 Both Delta and dryland areas burn too frequently
62 There is the perception that if one is responsible for management then one is responsible for
everything
63 The area is populated by 3 ethnic groups with 6 000 people in 5 villages
64 Bukhagwe Bushmen in Gudigwa don’t burn as much as the other ethnic groups
65 Livestock is very important to the people so they look after their cattle
Fire impacts negatively on photographic areas – tourists don’t like to see burnt areas. These
66 areas are 30 kms away from the villages so don’t have big fire threat in the area. Wilderness
Safaris reacts quickly to fires and have fire fighting equipment
67 Photographic concession is next to a hunting concession so the game moves into these areas
Workshops are not an effective means of knowledge transfer because only a portion of the village
68 attends eg a Kgotla meeting in a village of 2 000 residents will only attract 60 people. Villagers
say they are busy harvesting crops before the elephants come
Appendix 4
Proposed Amendments to the Herbage Preservation Act
1. Short Title
This Act may be cited as the Herbage Preservation (Prevention of Fires) Act.
2. Interpretation
In this Act, unless the context otherwise requires –
“boundary firebreak” means a firebreak never less than 6m in width one edge of which,
throughout the length of the firebreak, is contiguous to a land boundary;
“department” means the department responsible for the implementation of the herbage
preservation act;
“firebreak” means a strip of land cleared of inflammable matter to prevent the spread of
fire;
“Herbage Prevention Committee” means the Principal Herbage Preservation Committee

or a Subordinate Herbage Preservation Committee established by or under section 3;
“vegetation” means growing or standing vegetation, and includes any tree or part thereof
and any bush, shrub, brushwood, undergrowth, grass, crops or stubble.
“strategic fire break” are fire breaks prepared by the Department annually which have a
minimum width of 30m
“concessionaire” a person who has a concession over a piece of land for a specific
period.
“land user” a person who utilizes the land and its products for a livelihood.
3. Establishment of Herbage Prevention Committees
3.1 There is hereby established a body called the Principal Herbage Preservation
Committee which shall consist of such persons or body of persons as the Minister
shall, by notice published in the Gazette, appoint.
3.2 The Minister may, by notice published in the Gazette, establish such number of
bodies called Subordinate Herbage Preservation Committees as he thinks
appropriate which shall each consist of such persons or body of persons as the
Minister shall, by such notice, appoint.
3.3 The area of authority of the Principal Herbage Preservation Committee shall be the
whole of Botswana and the area of authority of a Subordinate Herbage
Preservation Committee shall be such area as the Minister shall, by order

published in the Gazette, specify.
3.4 The Minister shall appoint a chairman of each Herbage Preservation Committee
from among the members of that Committee.
3.5 The Principle Herbage Preservation Committee may give directions to a Subordinate
Herbage Preservation Committee, or to such Committees generally, as to the
carrying out of its or their functions under this Act and every such Committee to
whom such directions have been given shall comply with those directions.
3.6 The Principle Herbage Preservation Committee may appoint any person to be a fire
ranger.
3.7 The Herbage Committee or a committee appointed by the Herbage Committee has
the following duties:
3.7.1approve fire management plans for each land user or concession holder
3.7.2make rules which bind its members
3.7.3provide the strategy for agreed mechanisms for the cooperation of actions
with adjoining Herbage committee boundaries in the event of a fire
crossing boundaries
3.7.4regularly communicate the fire danger rating to its members
3.7.5organize and train its members in fire fighting, management and preservation
3.7.6develop coordinate and implement a fire awareness program for the members
3.7.7develop and implement a system to coordinate all fire fighting and fire
management activities for members
3.7.8Exercise the powers and perform the duties delegated to it by the Herbage
Preservation Act
3.7.9Identify the ecological conditions that affect the fire danger
3.7.10 Appoint a “fire ranger”
3.7.11 Furnish regular information to the Department of Forestry and Range
Resources on fires at least monthly
3.7.12 Inform its members of equipment and technology available for preventing
and fire fighting of fires
4. Fire Danger Rating System
The Department of Forestry and Range Resources must supply a Fire Danger Rating system for the
entire country and issue warnings about high fire risk periods based on assessments of prevailing
weather conditions during the months of April to the end of November in consultation with the
committees appointed in terms of the Herbage Preservation Act.
5. Fire Management Plans
All lease holders and/or concession holders and land users must have a fire management plan for
his/her property.
5.1 All fire management plans must be lodged with the District Fire Committee for
approval.
6. Prohibition of Burning Vegetation
6.1 No person shall, without the permission in writing of the Herbage Preservation
Committee, set fire to any vegetation on land which he/she is not the lease holder
and/or concession holder, or in lawful occupation.
6.2 No person shall wilfully or negligently light a fire, which by spreading damages or
destroys, or threatens to damage or destroy, the property of another person.
7. Duty to Extinguish Fires
All persons shall carefully and properly extinguish any fire lit or used by themselves for any purpose
on any land, irrespective of whether he/she is the lease holder and/or concession holder or in lawful
occupation.
7.1 Readiness for Fire Fighting
Every lease holder or concession holder on whose land a fire may start or burn, or from
whose land it may spread must -
a) have the required equipment, protective clothing and trained personal for extinguishing fires
b) ensure that in his/her absence, responsible persons are present on or near

his/her land who will, in the event of a fire notify the neighboring owners,
concession holders and
c) extinguish the fire or assist in doing so
7.2 Actions to Fight Fires
Any lease holder or concession holder who believes that a fire on his/her land, or the
land of an adjoining lease holder and/or concession holder, may endanger life, property
or the environment must immediately –
a) take all necessary steps to notify the fire ranger, Fire Committee coordination
center and adjoining land user; and do everything in his or her power to stop
the spread of the fire
8. Notice to be Given Before Burning Vegetation
8.1 every person, before burning vegetation on land of which he/she is the lease
holder or concession holder, or on which he/she is permitted or authorized to burn
vegetation, shall give reasonable notice of his/her intention to do so and, as nearly
as possible, of the time at which the burning is to begin to all lease holders or
occupiers of adjoining land and, where reasonably practicable, to a police officer or
headman.
8.2 In subsection (1) “headman” has the same meaning as in the Chieftainship Act.
8.3 If a fire lawfully lit after notice has been given in accordance with this section
spreads to adjoining land, the fact that such notice was given shall -
a) to the person who lit or was responsible for lighting the fire, be a sufficient
defence to any charge of contravening section 4, unless it is proved that he
wilfully or by the negligence of himself, his employee/s or agent caused or
permitted the fire to spread across the boundary of the land on which the fire
was lit to the adjoining land; and
b) not affect the right of any person aggrieved to recover damages by civil action
for any loss thereby sustained by him.
9. Power to Prohibit Burning of Vegetation
9.1 A Herbage Preservation Committee may, by order, prohibit the burning of

vegetation within its area of authority either generally or subject to such exceptions
as the order may specify.
9.2 An order under this section shall -
a) apply to the whole area of authority of the Committee or to such area or place
as the order may specify
b) specify the date on which the order is to come into operation; and
c) be published in the Gazette and by public exhibition at an office of the District

Commissioner or of the Local Authority, if any, situated within the area or place
to which the order applies, not less than 14 days before the date of
commencement of the order
d) Prohibited the burning of vegetation other than if a assessment of the

vegetation has been done
10. Prohibition of Burning of Vegetation by Employees
No employee/s of the lease holder and/or concession holder, or occupier of any land shall
burn any vegetation on the land except with the consent and under the personal direction of
the lease holder and/or concession holder, or occupier, or with the consent and under the
personal direction of some other person authorized by the lease holder and/or concession
holder, or occupier in that behalf.
11. Firebreaks
11.1 Any lease holder and/or concession holder, or occupier of land who desires to
guard against fires spreading beyond the boundaries of the land may, by notice in
writing served upon the lease holder and/or concession holder, or occupier of any
adjoining land on or near the common boundary of which sufficient firebreaks
have not in his opinion been constructed and maintained, require the and/or
concession holder, or occupier of the adjoining land -
a) to construct and maintain on the adjoining land a boundary firebreak along

the whole length of the common boundary or along such stretches of the
common boundary as the notice may specify; or
b) to contribute one half of the cost or labour necessary to construct or

maintain –
i) on both sides of the common boundary, boundary firebreaks along the

whole length of the common boundary or corresponding boundary firebreaks
along such stretches of the common boundary as the notice may specify;
and
ii) on either side of the common boundary such other firebreaks as may have
been agreed upon.
11.2 Every notice under subsection (1) (b) shall be deemed to contain a requirement
that the lease holder and/or concession holder, or occupier upon whom the notice
is served shall permit all reasonable access to the adjoining land to enable the
lease holder and/or concession holder, or occupier serving the notice to construct
and maintain or to contribute to the construction and maintenance of such of the
firebreaks referred to in the notice as are, or are to be, on the adjoining land.
11.3 Any lease holder and/or concession holder, or occupier upon whom a notice has
been served under subsection (1) may appeal in writing against the requirements
contained or deemed to be contained in the notice, within 30 days after its service
upon him/her, to the Principal Herbage Preservation Committee:
Provided that where the land in question lies wholly within the area of authority of
a Subordinate Herbage Preservation Committee the appeal shall lie to that
Committee.
11.4 The Herbage Preservation Committee to whom an appeal has been made under
subsection (3) shall give every party to the appeal the opportunity to make written
or oral submissions to it and shall then either –
a) allow the appeal in which case the requirements contained in the notice shall be
void; or
b) dismiss the appeal and, subject to subsection (5), order the lease holder and/or
concession holder, or occupier on whom the notice was served to comply with
the requirements contained or deemed to be contained in the notice.
11.5 Before making an order under subsection (4) (b) the Herbage Preservation
Committee may alter or vary the requirements contained or deemed to be
contained in the notice.
11.6 The decision of a Herbage Preservation Committee on appeal under this section
shall be final.
11.7 Every lease holder and/or concession holder, or occupier upon whom a notice
has been served under subsection (1) shall comply with the requirements
contained or deemed to be contained in the notice unless he/she appeals against
the requirements in accordance with subsection (3), in which case he/she shall
comply with any order made by the Herbage Preservation Committee under
subsection (4) (b).
11.8 Where the lease holder and/or concession holder, or occupier of adjoining land
fails to comply with the requirement to construct and maintain or to contribute one
half of the cost or labour necessary to construct and maintain a boundary
firebreak on that land or to permit reasonable access to that land in respect of the
construction and maintenance of a boundary firebreak, in contravention of
subsection (7), the lease holder and/or concession holder, or occupier who
served the notice in question may –
a) enter upon the adjoining land, without thereby incurring any liability for trespass,
and there construct and maintain the boundary firebreak; and
b) recover from that lease holder and/or concession holder, or occupier the
reasonable cost of such construction and maintenance.
11.9 Where a Herbage Preservation Committee is of the opinion that a lease holder
and/or concession holder, or occupier has failed to comply, either wholly or in
part, with a requirement to construct and maintain a boundary firebreak, in
contravention of subsection (7), it may, by notice in writing served upon that lease
holder and/or concession holder, or occupier, give him/her such directions as are
in the opinion of the Committee necessary for the purpose of remedying the
default.
11.10 Every lease holder and/or concession holder, or occupier upon whom a notice
has been served under subsection (9) shall comply with the directions contained
in the notice.
11.11 Every lease holder and/or concession holder who has structure or property to
protect is required to prepare a fire break to protect against fires.
12. Herbage Preservation Committees May Order Compulsory Firebreaks
12.1 A Herbage Preservation Committee may, by order, require the construction and
maintenance of firebreaks by the lease holder and/or concession holder, or
occupiers of land within its area of authority.
12.2 An order under this section shall –
a) apply to the whole area of authority of the Committee or to such area or place
as the order may specify;
b) specify the extent and location of the firebreaks;
c) specify the date on which the order is to come into operation; and
d) be published in the Gazette and by public exhibition at an office of the District

Commissioner or of the local authority, if any, situated within the area or place
to which the order applies, not less than 14 days before the date of
commencement of the order.
12.3 Where a lease holder and/or concession holder, or occupier of land fails to
construct or maintain a boundary firebreak on that land in contravention of an
order made by a Herbage Preservation Committee under this section, the
Committee, or the lease holder and/or concession holder, or occupier of any
adjoining land, with the permission in writing of the Committee, may –
a) enter upon that land, without thereby incurring any liability for trespass, and
there construct and maintain the boundary firebreak; and
b) recover from that lease holder and/or concession holder, or occupier the
reasonable cost of such construction and maintenance.
13. Power to Require Assistance in Putting Out Fire
13.1 Where a fire is burning and a member of a Herbage Preservation Committee, an

administrative officer, an agricultural officer, a fire ranger, a forest official, a police
officer, a traditional authority, or a lease holder and/or concession holder, or
occupier of the land on which the fire is burning, or such lease holder’s and/or
concession holder’s or occupier’s employee/s, reasonably believes that the fire
may become a danger to life or property, he/she may require any person present
at or in the vicinity of the fire to render assistance or to do any act or perform any
service he/she reasonably considers necessary or expedient to control,
extinguish or prevent the spread of the fire.
13.2 In subsection (1) –
“fire ranger” means a person appointed to be a fire ranger under section 3 (6);
“forest official” has the same meaning as in the Forest Act.
13.3 Any person acting in good faith may, either alone or with persons under his
control, enter upon any land for the purpose of controlling, extinguishing or
preventing the spread of a fire which he/she reasonably believes is not under
control or may become a danger to life or property.
14. Protection of Life, Person or Property by Counter-Firing
Nothing in this Herbage Preservation Act shall prohibit any person, when life, person or
property is in danger of loss or injury from an approaching fire, from lighting and burning
vegetation, in the manner commonly known as counter-firing, in order to prevent such loss
or injury; but he/she shall take reasonable care that the fire so lit does not spread beyond
the limits necessary to prevent such loss or injury.
15. Liability of Employee/s and Employer
Where an employee/employees contravene/s any provision of this Act or of an order made

by a Herbage Preservation Committee under section 7 or section 10 while acting under the
directions or command of his employer, the employer shall be deemed also to have
contravened that provision and may be prosecuted under this Act for that contravention
either alone or jointly with his/her employee/s.
16. Offences and Penalty
16.1 Any person who -
a) contravenes any provision of the Herbage Preservation Act or of an order

made by a Herbage Preservation Committee under section 7 or section 10; or
b) fails, without reasonable excuse, to comply with a requirement made under

section 11 (1), shall be guilty of an offence and liable to a fine not exceeding P1
000 or to imprisonment for a term not exceeding 12 months, or to both and to
corporal punishment.
16.2 In any prosecution for failure to comply with a requirement made under section
11 (1) the court shall presume the absence of a reasonable excuse on the part of
the person charged unless the contrary is proved.
17. Civil Remedy Preserved
Nothing in this Herbage Preservation Act shall affect the right of any person aggrieved to
recover damages by civil action for any loss sustained by him/her unless it is provided to
the contrary.
18. Regulations
The Minister may make regulations -
a) to regulate the procedure of Herbage Preservation Committees; and
b) to facilitate the implementation of the purposes and provisions of this Act.
Appendix 5
Forage and fuel factors representing the forage and fuel production potential of the
Decreaser, Increaser I and Increaser grass and non-grass species recorded in the
Okavango Delta Ramsar Site in Botswana.
DECREASER SPECIES FORAGE FACTOR FUEL FACTOR

Anthephora pubescens 8 8
Brachiaria nigropedata 8 8
Brachiaria dura 5 5
Brachiaria sp 5 5
Bulbostylis sp 4 4
Cenchrus ciliaris 8 8
Digitaria eriantha 8 8
Echinochloa pyramidalis 7 8
Panicum maximum - perennial 8 8
Panicum maximum - annual 4 6
Panicum repens 8 8
Schmiditia pappophoroides 6 6
Setaria sphacelata 8 8
Sporobolus fimbriatus 7 7
Sporobolus iocladus 3 3
INCREASER I SPECIES FORAGE FACTOR FUEL FACTOR

Aristida stipitata 2 6
Aristida meridionalis 1 4
Cyperus sp. (Medium) 3 4
Cyperus sp. (Tall) 5 6
Eragrostis pallens 2 8
Eragrostis cimicima 4 4
Hermarthria altisima 10 10
Imperata cylindrica 1 10
Leersia hexandra 5 5
Miscanthus junceus 2 10
Oryza longistaminata 4 10
Pennisetum glaucocladium 2 10
Sporobolus sp (Tall) 3 4
INCREASER II SPECIES FORAGE FACTOR FUEL FACTOR

Arisitida congesta 2 2
Aristida stipoides 1 2
Bare Ground 0 0
Forbs 1 2
Brachiaria deflexa 2 2
Chloris virgata 1 2
Cenchrus incertus 1 1
Cynodon dactylon 5 6
Enneapogon desvauxii 2 2
Eragrostis aspera 1 2
INCREASER II SPECIES FORAGE FACTOR FUEL FACTOR

Eragrostis echinochloidea 2 2
Eragrostis inamoena 2 3
Eragrostis lehmanniana 3 3
Eragrostis rigdior 3 6
Eragrostis pseudosclerantha 1 1
Eragrostis viscosa 0 1
Eragrostis sp. 2 2
Eragrostis superba 2 2
Eragrostis trichophora 1 2
Dactyloctenium giganteum 4 5
Digitaria velutina 1 1
Melenis repens 1 2
Microchloa indica 0 0
Perotis patens 1 1
Pogonarthria fleckii 1 3
Schmidtia kalihariensis 2 4
Setaria verticilliata 2 2
Sporobolus africanus 2 4
Sporobolus panicoides 1 1
Sporobolus spicatus 1 1
Stipagrostis uniplumis 5 5
Tragus berteronianus 1 1
Tricholeana monachne 1 2
Urochloa panicoides 1 1
Urochloa trichopus 1 2

5.3 Assessment of Current Strategies To Control Fire in The Ramsar Site

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

5.3 Assessment of Current Strategies To Control Fire in The Ramsar Site

Hochgeladen von

Copyright:

Verfügbare Formate

EnviroNET Solutions Pty Ltd

5.3 Assessment of Current Strategies to Control Fire in the Ramsar Site

A requirement in the terms of reference was to assess the effectiveness of current

5.3.1 Fire Suppression

5.3.2 Herbage Preservation Act

5.3.2.1 Current situation

5.3.2.2 Proposed changes to Herbage Preservation Act

5.3.2.2.1 Establishment of District Fire Committee and Sub-District Wards

DISTRICT FIRE COMMITTEE

Fire Related Responsibilities

SERONGA SHAKAWE MAUN TSAU

Coordinated by Fire Ranger from District Fire

Figure 35. Proposed decentralization of fire related responsibilities of the District

5.3.2.2.2 Delegation of responsibilities for issuing burning permits

DISTRICT FIRE COMMITTEE

Permit Issuing Responsibilities

SERONGA SHAKAWE MAUN TSAU

COMMERICAL COMMERCIAL COMMERCIAL COMMERCIAL

CONCESSION CONCESSION CONCESSION CONCESSION

Duplicate copy of permits sent to District Fire Office - Maun

Figure 36 Infrastructure showing delegation of responsibilities for issuing permits

5.3.2.2.3 Fire Danger Rating System

5.3.2.2.4 Prohibition on burning vegetation

No burning of any vegetation will be permitted without an assessment being completed of

5.3.2.2.5 Duties to extinguish fires

5.3.2.2.6 Readiness for fire fighting

5.3.2.2.7 Burning permit

BURNING PERMIT PLAN

1 1.1 DATE 1.2 IGNITION TIME 1.3 FINISH TIME

APPLICANT NAME Tel:

5.3.3 Proposed amended Herbage Preservation Act

5.4 Identification of Deficiencies in Current Knowledge on Fire Ecology of the

5.4.1 Applied Research

The following applied research projects need to be conducted:

5.4.2 Basic Research

The following basic research projects need to be conducted:

• It is recommended that detailed autecological studies be conducted on the key

Since phosphorous is a limiting factor in plant growth in the Delta an investigation

become colonized by several plant species. At an advanced stage in the plant

5.4.3 Scientific services

6.1.2 Reasons for burning

6.1.3 Ecological criteria for prescribed burning

• Burning is ecologically acceptable if the grass sward is in a climax and/or sub-climax

6.1.4 Fire regime

6.1.4.1 Type of fire

6.1.4.2 Fire intensity

Fire Intensity (kJ/s/m Description

6.1.4.3 Season of burning

6.1.4.4 Frequency of burning

6.1.4.5 Post-fire range management

6.2 Application of a Controlled Burning Program

6.2.1 Fire Danger Index for controlled burning

AIR TEMPERATURE/ RELATIVE HUMIDITY vs BURNING INDEX

WIND CORRECTION FACTOR vs BURNING INDEX

PERCENTAGE CURING OF GRASS SWARD GRASS CURING FACTOR

FDI = Burning Index [(Air Temperature/Relative Humidity) + (Wind Correction

FDI = [51 + 10)] x (1);

FDI = [37 + 10)] x (0.3);

6.2.1.2 Vegetation monitoring

To achieve these objectives it is recommended that monitoring sites be established in the

6.2.2 Burning procedure

6.2.2.1 Block burns

Head Fire Back Fire