ASSESSMENT OF FOREST LANDSCAPE

RESTORATION OPPORTUNITIES IN GHANA

Final Report

SEPTEMBER 2011
List of Acronyms

CBO Community Based Organization

CERSGIS Centre for Remote Sensing and Geographic Information Services
CREMA Community Resource Management Areas
DBH Diameter at Breast Height/Base Diameter
EI Economic Index
EU European Union
FC Forestry Commission
FLEGT Forest Law Enforcement, Governance and Trade
FORIG Forestry Research Institute of Ghana
FSD Forest Services Division
GHI Genetic Heat Index
GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit
HFZ High Forest Zone
ITTO International Tropical Timber Organization
IUCN International Union for Conservation of Nature
MTS Modified Taungya System
NCRC Nature Conservation Research Centre
NGO Non Governmental Organization
N-REG Natural Resources and Environmental Governance
PI Pioneer Index
PROFOR Program on Forests
REDD Reducing Emissions from Deforestation and Forest Degradation
RMSC Resource Management Support Center of the Forestry Commission
SDSU South Dakota State University
TZ Transition Zone
UNFCCC United Nations Framework Convention on Climate Change
UNESCO United Nations Educational, Scientific and Cultural Organization
WD Wildlife Division of the Forestry Commission
WRI World Resources Institute
2
Table of Contents

List of Acronyms ............................................................................................................................. 2

Table of Contents ............................................................................................................................ 3
1. Introduction ............................................................................................................................. 5
2. Approach ................................................................................................................................. 6
2.1 The global assessment method ........................................................................................... 6
2.2 An assessment method for Ghana ...................................................................................... 7
2.3 Constraints ......................................................................................................................... 8
2.4 Strategy............................................................................................................................... 8
3. Methodology and Implementation ......................................................................................... 10
4. Conclusions ........................................................................................................................... 13
5. References.............................................................................................................................. 15
Appendix I. The Forest and Policy Context of Ghana.................................................................... 16
1. Forests in Ghana ................................................................................................................. 16
2. Forest and Landscape Restoration in Ghana ....................................................................... 17
3. Collaborative Landscape Restoration in Ghana – An example ............................................ 19
4. Forest Policy Framework .................................................................................................... 21
4.1 Historical Perspective .................................................................................................... 21
4.2 Current Framework ..................................................................................................... 21
4.3 Restoration of Degraded Lands ................................................................................... 23
4.4 Reversing Land and Natural Resources Degradation through Investments .................. 23
4.5 Biodiversity Conservation ............................................................................................ 23
4.6 Equitable Benefit Sharing ............................................................................................ 24
4.7 Implications of Participating in REDD+ ..................................................................... 25
Appendix II. National Stakeholder Workshop................................................................................ 27
1 Workshop Objectives .......................................................................................................... 27
2 Expected Outcomes ............................................................................................................ 27
3 Organization and Participation ............................................................................................ 28
4 Presentations ....................................................................................................................... 28
4.1 Assessing opportunities for forest landscape restoration: Why? What? How? .............. 28
4.2 Restoration in Ghana: What are the experiences to date? What are the needs, options,
benefits, and constraints? ....................................................................................................... 29
4.3 Assessing Restoration Opportunities in Ghana: Strategy and Process .......................... 30
4.4 Sample comments made after the presentations ........................................................... 30
5 Break-Out Sessions ............................................................................................................. 31
6 Guidelines for Group Work ................................................................................................ 31
6.1 Group 1: Criteria for degradation scoring of reserves................................................... 31
6.2 Group 2: Criteria for mapping of sacred groves, mangroves ........................................ 32
6.3 Group 3: Options for restoration of reserved forests: costs, consequences, and
constraints.............................................................................................................................. 33
6.4 Group 4: Restoration issues off reserves ...................................................................... 33

3
 7 Summary of Group Presentations ....................................................................................... 33
7.1 Group 1: Criteria for degradation scoring of reserves................................................... 33
7.2 Group 2: Criteria for mapping of sacred groves, mangroves ........................................ 34
7.3 Group 3: Options for restoration of the above categories and their costs, consequences,
and constraints ....................................................................................................................... 35
7.4 Group 4: Restoration issues off reserves ...................................................................... 39
7.4.1 Land and Tree Tenure.............................................................................................. 39
7.4.2 Crop Types .............................................................................................................. 39
7.4.3 Forest fires and land uses ......................................................................................... 39
7.4.4 Benefit sharing ......................................................................................................... 39
7.4.5 Physical development ............................................................................................... 39
7.5 Existing efforts to restore forests in Ghana.................................................................. 40
7.6 Some Practical Approaches .......................................................................................... 40
7.7 Map Requirements. ..................................................................................................... 40
7.8 Short Term vrs. Long Term Benefits .......................................................................... 40
7.9 Potential Users of FLR Map ........................................................................................ 41
8. Workshop Recommendations ............................................................................................. 41
9. List of Participants .............................................................................................................. 42
10. Program .......................................................................................................................... 43
Appendix III. Method .................................................................................................................... 44
1. Ecological Zones in Ghana ................................................................................................. 44
1.1 Classification of Ecological Zones ............................................................................... 45
1.2 Extent of Ghana’s forest and Trends over the last half century.................................... 45
1.3 Forest Reserves ............................................................................................................ 46
1.4 Unreserved Forest Areas (Off-reserves) ....................................................................... 47
1.5 Extent of Ghana’s Savannah & Trends over the Last Half Century ............................. 48
2. Forest Condition Scoring .................................................................................................... 52
2.1 Savannah Condition Scoring ........................................................................................ 54
3. Criteria and Indicators for Ecological Zone Classification ................................................... 56
3.1 Basis for Scoring Description....................................................................................... 57
4. Forest and Woodland Cover Mapping ................................................................................ 58
4.1 Image acquisition and pre-processing........................................................................... 58
4.2 Image classification ...................................................................................................... 59
4.3 Field Validation and Accuracy Assessment .................................................................. 61
4.4 Map composition and presentation .............................................................................. 61
4.5 Software used .............................................................................................................. 61
Appendix IV. Criteria for scoring condition and restoration opportunities ..................................... 62
Appendix V. Forest and Woodland Cover Classification Maps ...................................................... 79

4
1. Introduction

The importance of forest landscape restoration in addressing climate change and other societal
needs was brought home in November 2009 when the Global Partnership on Forest Landscape
Restoration (GPFLR) concluded that more than 1 billion hectares of lost forests and degraded lands
worldwide present restoration opportunities that could sequester significant amounts of carbon. The
large magnitude of the estimate suggests that ongoing restoration efforts are insufficient and not at
scale. They are not consistent with the opportunities, not to mention the need. This suggests that
restoration is an overlooked opportunity for improving the lives of rural people in degraded
landscapes and for enhancing carbon stocks and resilience to climate change.

While the global assessment suggested that huge areas provide opportunities for restoration, it also
had important limitations. It was coarse, building exclusively on global datasets. It was made without
consultation with stakeholders, again due to its global scope. To overcome these limitations, more
refined assessments at the national scale are needed.

No methodology for national level restoration opportunity assessments exists, however. A

methodology for national level assessment therefore needs to be developed and tested. Ghana
volunteered to provide a “laboratory” in which such a methodology could be developed and
PROFOR agreed to support this initiative. Later, the German International Climate Initiative has
also decided to provide follow-up support.

This report presents the first results of PROFOR supported project to develop and apply a
methodology for assessing restoration opportunities in Ghana and their potential contribution to
REDD-plus strategies for climate change mitigation, poverty alleviation and sustainable forest
management.

The report presents a methodology for restoration opportunity assessment, using forest reserves as a
starting point, and first results of its application to forest reserves. It also reviews the existing
context in Ghana with respect to national forest policy, legal and technical framework for forest
resource management within the forest sector agencies, and future plans for reform.
5
2. Approach

2.1 The global assessment method

The global assessment was carried out by posing a sequence of questions and then answering them
in terms of maps, using globally consistent datasets. This sequence goes as follows:

A. Set the stage

1. What is the area of analysis? Delineate the lands to be examined.

B. What has been lost or degraded?

2. Where can forests potentially grow? Map the potential extent of forests and woodlands if
the only constraints were climate and soils.
3. Where are forests growing today? Map the current extent of forests and woodlands.
4. Where have forests been lost or degraded? Contrast the potential with the current extent
of forests in light of current land use.

C. What can be restored?

5. Where are the constraints on restoration? Map human pressure.
6. Where are the opportunities for restoration? Map the areas where human pressure is low
enough to provide opportunities for restoration of different types.

D. What are the carbon benefits of restoration?

7. What are the opportunities for carbon sequestration through restoration? Assign
carbon values to the restoration opportunities.

6
2.2 An assessment method for Ghana

The global method is not appropriate for a direct application to Ghana.

• A higher spatial resolution is needed. At the same time, there is no need to use global
datasets or globally relevant definitions. National datasets can and must be used to the extent
possible.
• Richer information is needed. The global method was severely constrained by lack of data on
potentially relevant institutional and social factors such as land ownership and conflicts.
• National stakeholders and experts must be engaged. The global method does not include
such engagement for feasibility reasons but a national method can and must do so.

A new or modified method must therefore be developed. The following underlying principles for
the Ghana assessment were adopted:

1. Stakeholder participation is a key feature of the project and will ensure that the analysis and
recommendations are appropriate and credible;
2. Visualization techniques that communicate landscape characteristics to all stakeholders need
to be employed;
3. Governmental ownership of the process through the Forestry Commission is needed to
ensure that results are incorporated into government programs;
4. The most promising starting point for developing a national restoration opportunity
assessment methodology for Ghana is the forest reserves; and
5. The final product should incorporate field validation and verification by major stakeholders.

The methodology developed and applied in Ghana therefore included two major components:
• stakeholder participation from all related sectors, and
• the use of remote sensing tools for the geospatial assessment.

7
2.3 Constraints

Ghana is not a data-rich country. At the start of the project, Ghana did not have a recent land
cover/land use map of sufficient detail, nor was a recent Landsat mosaic available. The technical
infrastructure for acquiring and processing data was also a major constraint and remains so.

2.4 Strategy

A strategy was adopted that consists of two parallel and mutually supporting streams of activity:
- A consultation process, and
- A mapping process.
The idea is that the consultation process be informed and constrained by maps, so that it is
underpinned by good and realistic information, while the mapping process is informed and inspired
by the consultations.

Ghana is a highly diverse country in terms of natural geography (see Appendix I). The different
forest zones present very different challenges for the opportunity assessment. While canopy density
is a useful criterion of degradation in the high forest zone, it is problematic in the savannah zone
where the trees are naturally far apart. Ghana is also diverse in terms of available data and
information. The high forest zone is much better known than other zones and forest reserves are
much better known than off-reserve lands.

The forest reserves were selected as a spring board for a step-wise process of constructing and
applying an assessment method that is applicable across the whole country. There are several
seasons for this decision:
• The condition of the forest reserves was described in detail by Hawthorne and Abu-Juam in
1995. This study provides an excellent point of reference. In general, the information about
the reserves is much richer than for other areas in Ghana.
• The Ministry for Lands and Natural Resources and the Forestry Commission are very
interested in information on the current condition and restoration opportunities of the forest
8
 reserves. There is a concern that they have suffered significant degradation since 1995,
although no new condition assessment has been undertaken since then. There is a strong
demand for a re-assessment and for a cost-effective method that can be applied regularly at
affordable cost for monitoring purposes.
• The immediate opportunities for restoration that would lead to carbon sequestration were
deemed to be the greatest in the reserves, given their protected status.

A consultation strategy with several steps was adopted:

1. An initial national consultation

2. Subsequent regional consultations, tentatively four
3. A concluding national workshop, at which the results for the whole country would be
presented and discussed.

A multi-step strategy was also adopted for the mapping part of the assessment methodology:

1. Develop and apply a method for assessment/scoring of condition and restoration

opportunities in the forest reserves in the high forest zone. The method would use satellite
imagery as the main source of information in order to be cost effective and use a scale for
scoring of condition that is consistent, if not identical, with the scale used by Hawthorne and
Abu-Juam (1995), so that comparisons can be made.
2. Apply this method to off-reserve areas in the high forest zone.
3. Construct condition scoring scales that are appropriate for other ecological zones of Ghana
and can be used with satellite imagery.
4. Apply these scales first to reserves and then to off-reserve areas.
5. Construct degradation maps and assess opportunities for restoration.

9
3. Methodology and Implementation

3.1 National Stakeholders’ Workshop:

A national stakeholder workshop was held in Accra on March 22 (a detailed account is given in
Appendix II).

The purpose of the workshop was to create a platform for open and objective discussion on the
need for forest restoration at landscape scale.

The stake holders were invited from the forestry sector, civil society organizations working in the
forestry sector, community based organizations, Development Partners, research institutions, and
the Ministry responsible for Forestry.

Break –out sessions were held on the following themes:

- Criteria for degradation scoring of forest areas (reserves)

- Criteria for mapping of sacred groves, mangroves
- Options for restoration of reserved forests, sacred groves and mangroves and their costs,
consequences, and constraints
- Restoration issues within off reserve areas

The workshop registered significant interest in restoration from the participants. Nobody disputed
that the need and the opportunities for restoration in Ghana are very significant and strong interest
in the output of the assessment project was expressed.

3.2 Develop/Review Criteria for Forest Condition Scoring

A scale for scoring vegetation condition and restoration opportunities that can be used together with
satellite imagery was developed for the three key ecological zones in Ghana: the high forest zone,
the transition zone, and the savannah zone (see Appendix III for details).

It is essential that forest landscape restoration at national scale be preceded by a characterization of

vegetation types over the landscape. Furthermore, a landscape approach to restoration also takes
10
into account the various land uses and ecosystems (forest, savannah, forest-savannah transitional
belt, mangrove, etc). A multiple-use landscape is often a mosaic of land use types, in which some
parts may be formally protected (e.g. forest reserves) while others are sustainable-use communal
areas, private farmland and/or other types of productive use (e.g. off-reserve areas serving as
agricultural landscapes).

3.3 Use of visualization Techniques

A coarse restoration opportunity map of Ghana was developed and used during the national
workshop (see Appendix II).

Satellite images have been used in this assessment to cost effectively show levels of deforestation
and degradation within reserved areas in both the forest and savannah ecological zones. Map
outputs from classified satellite images have been used by the institutional stakeholders as field
validation tools for preliminary assessment of landscape restoration potential, and possibly used to
guide forest investment programs. The maps could be used for preliminary cost computations, as
well as advocacy tools for forest landscape restoration at the local and community levels, by
depicting in “real world” terms the extent of deforestation and its spatial distribution.

Consensus building is key in any participatory process, thus appropriate tools for facilitating this
process should be identified from the onset. Participatory mapping and the use of GIS overlays as
negotiation support tools are well tried and tested in rural development at local scales. (Lewis et al
2005, Sheppard, 2006)

3.4 Preliminary costs for FLR Implementation

This phase of the assessment did not perform a detailed economic analysis. Preliminary cost figures
could be obtained by applying per unit area costs for implementation options to areas identified in
forest/woodland cover maps as potential restoration sites.

Figure1 attempts to represent in the form of a flow chart the different steps that we have used in
this assessment for forest landscape restoration.

11
Figure 1: Methodological Framework

NATIONAL FOREST POLICY FRAMEWORK AND REVIEW PROCESS

National Stakeholder’s
Workshop

Develop/Review Criteria for National

Forest Condition Scoring Forest/Woodland
Cover Definitions

Image-based Community
Forest/Woodland Cover Participation in Field
Visualization Validation

Economic Analysis Forest/Woodland Investment

Planning

Monitoring Outcomes Forest/Woodland Community

Restoration Participation and
Learning

12
4. Conclusions

4.1 Difficulties

The work has been more difficult than anticipated.

The project has revealed serious shortcomings in the geodata infrastructure of Ghana. Lack of data
and processing capacity has turned out to be a significant obstacle that has slowed down progress.
The project has created a mosaic of cloud-free Landsat images but this mosaic has limitations,
particularly for assessment of the savannah zone, as it does not take seasonal variations into account,
and the resolution (30m) is insufficient to identify individual trees. For assessment of the high forest
zone, these limitations are less critical.

The common habit of thinking about landscapes from the viewpoint of forests and ecosystems has
also been difficult to overcome. Restoration in terms of farmer managed natural regeneration in
northern Ghana and agroforestry system has so far received less attention than classical tree
planting.

In general, the need for restoration in Ghana is, at the same time, both obvious and overlooked.
Little attention is being paid to the opportunities for restoration in relation to the need to stop
deforestation and degradation. This balance seems to be inconsistent with the realities on the ground
in Ghana, where decades of deforestation and degradation have left huge areas in need of
restoration.

4.2 Next steps

The project will continue with the goal of presenting a national map of restoration opportunities and
an assessment of the associated opportunities for carbon sequestration.

4.3 Recommendations

Ghana should take immediate measures to improve the availability of geodata, as the current lack of
such data is a severe limitation on planning and development far outside the theme of restoration. A
mosaic of recent Landsat imagery or similar should be created and made available to organizations
and to the public. The web-based interactive atlas of DRC is a good example of something that

13
could easily be done. Such data, particularly if they are made broadly available, can help substantially
in planning of monitoring of REDD plus related activities. Having access to an accurate and detailed
geodata platform would also help actors in Ghana implement countless of other projects that
depend on good map information.

14
5. References

1. Annette den Hollander H.J. (2008): Precipitation influence on savannah vegetation: cover,
biomass, leaf area index and c3:c4 ratio. ESS-80436, Wageningen UR, The Netherlands.
47pp.
2. C.S.A (1956): Phytogeography. Cons.Scient. Afr. Sud Sahara, Comm. Coop. Tech. Afr Sud
Sahara. London.
3. Forestry Commission Compilation Reports: Forest Inventory Project of 1986-1990; Multi-
resource inventory program of 2001 and RMSC updated inventory data of 2009.
4. Hall and Swaine (1981): Hall, J.B., Swaine, M.D. 1981. Distribution and vascular plants in a
tropical Rain forest (forest vegetation in Ghana). Geobotany1. W. Junk, The Hague. 383 pp.
5. Hawthorne, W.D., Abu-Juam, M. (1995): Forest protection in Ghana. IUCN, Gland,
Switzerland.
6. House and Hall (2001): House, Jo I.; Hall, David O., (2001) Productivity of Tropical
Savannahs and Grasslands (16), Terrestrial Global Productivity (363-400).
7. Keay, R.W.J. (1949): An example of Sudan zone vegetation in Nigeria.
8. Ntiamoah-Baidu, Y. 2001: Indigenous versus introduced biodiversity conservation strategies:
the case of protected areas in Ghana. In Weber, W., White, J.T.L., Vedder, A., Treves, L.N.
(eds.): African Rain Forest Ecology and Conservation. Yale Univ. Press, New Haven. pp.
384−394.
9. Scholes and Hall (1996): Scholes, R.J., and Hall, D. O. (1996). The carbon budget of tropical
savannahs, woodlands and grasslands. In "Global Change: Effects on Coniferous Forests
and Grasslands, SCOPE," (A. I. Breymeyer, D. O. Hall, J. M. Melillo, and G. I. Agren, eds.),
Vol. 56, pp. 69-100. Wiley, Chichester.
10. UNESCO (1973): International Classification and Mapping of Vegetation. UNESCO, Paris.
11. White, F. (1983): The vegetation of Africa: a descriptive memoir to accompany the
UNESCO/AETFAT/UNSO vegetation map of Africa.
12. World Bank (2007): Ghana Country Environmental Analysis. Report No. 36985-GH:
Environment and Natural Resources (AFTERN) Africa Region.

15
Appendix I. The Forest and Policy Context of Ghana

1. Forests in Ghana

Forests play a crucial role in the socioeconomic wellbeing of the people of Ghana. They are
particularly important to the rural population, which constitutes about 60% of Ghana’s population
(Ghana Statistical Services, 2000). In addition to their role in protecting and enriching soils, forests
also provide food items, fuel wood, raw materials for building as well as site for eco-tourism and
recreation.

Ghana was richly endowed with forest resources vital for her development. Records indicate the
existence once of relatively undisturbed forests, which preserved abundant biodiversity (Alpert
1993), protected fragile soils (FAO, 1999; FAO, 2007; UNEP, 2002), and regulated the supply of
scarce water resources (Glantz & Katz 1985,). However, deforestation and forest degradation have
accelerated loss of biodiversity in recent decades. The degradation of forests and the loss of
biodiversity in Ghana have increased sharply in recent decades (Dixon et al., 1996). The reserved
forest is made up of 11,590 km² of production forests; 4,323 km² of protection forests; and about
1,980 km² of game production reserves (Siaw, 2001; Ghana Forestry Commission, 1995).

Ghana is currently experiencing one of the highest deforestation rates in the World (~2%) and had
between 1990 and 2010 a net forest loss of 125,000 ha/yr. It is estimated that vegetation in forests in
Ghana contains 103-200 tC/ha and 90% of the carbon in vegetation is released when forests are
cleared (FAO 2010b). Most of the forests have lost their pristine interior habitats that are critical for
the protection of vulnerable species (FAO 2001; Forest Services Division of Ghana, 1995). The
Government took steps to address the startling deforestation by introducing the Ghana National
Plantation Project to plant 20,000 ha per annum (Domson et al., 2007; Ghana Forestry Commission,
2005; IUCN, 2006).

The forests in Ghana cover an area of 4,939,958 ha (FAO 2010b) and are divided into on-reserve
and off-reserve areas, where the on-reserve are legally demarcated forested areas vested in the
traditional land-owning communities but set aside to be managed in trust for the nation by the state
(Osafo 2005) and covered ~1.6 million ha in 1996 (Kotey et al. 1998). The off-reserve forests are on
land vested in and controlled by local communities who, along with the landowners have the right of
usufruct to land but not the rights to timber (Osafo 2010).

The unsustainable exploitation of forest resources, coupled with the lack of environmental
awareness, has devastated the country's forests. Subsistence agriculture and fuel wood exploitation is
common throughout Ghana and have been increasing over the years due to a population growth.
Logging and mining activities in certain areas have also contributed to the degradation of the
country's forest resources. The consequence of forest loss in Ghana has been droughts, bushfires
and loss of livelihood to communities that depend on eco-services provided by the forest.
Desertification is encroaching on some deforested lands and soil erosion is rampant. The continuous
incidence of forest degradation and deforestation is contributing to climate change with its
concomitant adverse effects on the biophysical environment as well as social and economic lives of
the people.

In view of the negative effects, there is the need for mitigation strategies to reverse the looming
dangers associated with forest degradation and deforestation on such an unprecedented scale. This
requires a paradigm shift from the traditional site level forest restoration efforts to a landscape level.
This entails restoring forests and landscapes to reinstate ecological integrity and enhance human
wellbeing in deforested or degraded forest landscapes, recognition of the rights of local as well as
institutional stakeholders, the identification and active participation of such stakeholders in the
entire process of learning and adapting, and policy reviews which provide mutually beneficial
incentives to all stakeholders, thereby creating an enabling environment for the successful
implementation of forest restoration at landscape scale, and ensuring the resolution of competing
claims on forests.

2. Forest and Landscape Restoration in Ghana

Forest restoration efforts are not new in Ghana. However restoration based on the principles of
Forest and Landscape Restoration (FLR) differs from the conventional restoration approaches
employed in Ghana, due to its goal to enhance human well-being through restoration of forest
functionality and productivity at the landscape level.

17
The concept of FLR was introduced in Ghana, when the Government of Ghana supported the
establishment of a National Working Group on Forest Landscape in 2004 to develop a National
Plan of Action on FLR and encourage innovative restoration efforts. In addition, a National
workshop on ITTO guidelines for the restoration, management and rehabilitation of degraded and
secondary tropical forest was held in Ghana in 2006.

There was consensus subsequently at a national workshop, where participants acknowledged the
need for building the capacity of the stakeholders in practical use of the techniques. They also
reiterated the need for the production of Ghana specific manual on Restoring Forest Landscapes,
which can be used as a guide and a reference material. This was followed by a participatory resource
assessment to determine relevant principles and actions that could form the basis for forest
restoration guidelines specific to Ghana. Four workshops were held at which stakeholders from six
degraded forest areas were invited to validate and finalize the responses gathered from the
stakeholders. Official approval of the guidelines is being sought from the Ministry of Lands and
Natural Resources.

The following guiding assumptions were proposed by the stakeholders for forest landscapes
restoration in Ghana:

• Forest resources provide the direct source of livelihoods for the majority of the rural
population and that poverty reduction and wealth creation in the country are dependent upon
effective management of these resources;

• However due to different anthropogenic activities the forest resources are gradually being
degraded and hence need to be restored;

• Restoration requires collaboration among stakeholders;

• District Assemblies (DAs), Traditional Authorities (TAs), opinion leaders and local community
groups, including women and youth groups are important actors in the restoration process;

• Restoration activities at all levels will be carried out based on effective and efficient planning
and networking; and

18
 • Capacity of communities and community structures will be developed in Forest Landscape
Restoration and this will be sustained through the provision of adequate and appropriate
logistic and technical support by taking into consideration their indigenous knowledge.

3. Collaborative Landscape Restoration in Ghana – An example

A typical forest restoration success story carried out in Ghana was the ITTO/FORIG Community
Collaborative Restoration Project in the Pamu Berekum forest area situated in the Brong-Ahafo
Region of Ghana. The area had been severely disturbed after years of unsustainable logging,
rampant bushfires and shifting cultivation, resulting in nutrient-deficient soils, reduced primary
productivity and low biodiversity.

Based on forest landscape concepts and techniques, restoration efforts were directed at plantation
establishment, involving communities at various locations in the reserve. Plantations include
Taungya, monocultures of Tectona grandis and Cedrela odorata, also mixed with Terminalia
superba, Terminalia ivorensis and Cedrela odorata. Successes attributable to the project include
increased food production by the local farmers involved in the project, which translated into
increased income to the farmers and improvement in rural livelihoods. Besides these positive social
developments, a general improvement in biophysical resources of the area has also been observed.

Name of landscape: Pamu Berekum Forest Reserve

Location: Brong Ahafo Region, Ghana

Figure2 Two-data satellite image classification of project area Before FLR Four years after FLR

19
Landscape Characteristics

An area of about 189 sqkm had been severely disturbed after years of unsustainable logging and
rampant bushfires Size, degraded area, nutrient-deficient soils, reduced primary productivity and
low biodiversity.

Response

ITTO and FORIG initiated restoration programme

Point of significance

Involvement of community stakeholders should lead to agreement on roles and responsibilities, as

well as benefit sharing arrangements.

Envisaged outcomes:

Planting target of 1800 ha/year and an annual planting target of 20 ha for farmers involved in the
ITTO/FORIG community project.

Results

Availability of land for farming by land-hunger communities, increased food production, increased
income to farmers and an improvement in biophysical resources

Lessons Learnt

1. Participatory planning and implementation ensures success;

2. Importance of addressing the needs of local communities to achieve stakeholder
collaboration; Conflict resolution should be an integral part of any Restoration project;
3. Mainstream remote sensing and GIS tools to do initial assessment and monitor progress.

20
4. Forest Policy Framework

4.1 Historical Perspective

Policies for resources management in Ghana dates back to 1906 when legislation was enacted to
control the felling of commercial tree species. The demarcation of forest reserves was in the main
completed by 1939 and a forest policy was adopted in 1948 (Ghana Forestry Commission, 1994).
Since then, a consistent policy of demarcation, reservation, protection of water supplies, and the
promotion of research and public education have been vigorously pursued. However, most of the
early forest policies mainly emphasised a sustained supply of timber for the wood industry and
promoted over-exploitation and an eventual demise of unreserved forests. As a consequence, by the
end of 1978, the Government placed about 3,267,250 ha of forests under permanent forest estate.

In addition, quite a number of policies and attempted remedies were initiated by government and its
agencies such as Forest Commission Act of 1960; forest improvement fund Act of 1960;
Concessions Act of 1962; Forest ordinance for the protection of forests including reserves of 1972;
Trees and timber (chain saw operation) regulation of 1983; Administration of land (amendment)
degree of 1984; Forest products inspection Bureau Law of 1985; Forest protection (amendment)
Law of 1986; Control and prevention of bushfires Law of 1990 and Trees and timber (chain saw
operation) regulation of 1991 as guides for forests resources management in the country Forest
(Ghana Forestry Commission 1994). These policies and related laws were contained in various
official documents and vested in specific Ministries and state agencies for implementation. The
agency responsible for forest resources management in Ghana is the Forestry Commission which
was established under Act 405 - Ghana Forestry Commission Act, 1980 to coordinate the activities
of the forestry sector institutions, namely: the Forestry Department, Department of Game and
Wildlife, Forest Products Research Institute and Ghana Timber Marketing Board. Section 6 of the
Act mandated the Commission to regulate and manage the utilization of all forestry and wildlife
resources of Ghana and also coordinate the policies in relation to forest resources (Forest and
Wildlife Policy, 1994)

4.2 Current Framework

Ghana’s Forest and Wildlife Policy was drawn up in 1994, with the aim to conserve the nation’s
forest and wildlife resources for maintenance of environmental quality and perpetual flow of

21
optimum benefits to all segments of society. The Forestry Development Master Plan (1996-2020)
was subsequently prepared as a basis for implementing the policy.

The specific objectives of the policy are:

• Management and enhancement of Ghana’s permanent estate of forests and wildlife

resources;
• Promotion of viable and effective forest based industries, particularly in secondary and
tertiary processing;
• Promotion of public awareness and involvement of rural people in forest and wildlife
conservation so as to maintain life-sustaining systems, preserve scenic areas, and enhance
the potential of recreation;
• Promotion of research-based and technology-led forestry and wildlife management,
utilization and development, to ensure resource sustainability, socio-economic growth
and environmental sustainability;
• Development of effective capability at national, regional and district levels for effective
forest and wildlife management.

A number of implementation strategies including restoration of degraded forest estate; regulation of

uncontrolled harvesting, protection of forest reserves and wildlife protected areas; encouragement of
local community initiatives to protect natural resources for traditional, domestic and economic
purposes; encouragement of value-added processes in the timber sector to maximize income and
related benefits; increasing public awareness and people’s involvement in the conservation of forests
and wildlife; public participation through the development of consultative participatory mechanisms
to enhance farmers’ land and tree tenure rights; and ensuring access to traditional use of natural
products have been clearly spelt out in the policy. These notwithstanding, Poku et al, cites fair
access to forest resources, fair benefit sharing, corporate exploitation, and greater participation in
forest policy-making and management as major governance challenges that has a direct bearing on
the ability of mandated institutions to enforce the laws that restrict industry’s access to resources,
develop rules that will facilitate greater communal access to timber and non-timber forest products,
and to enforce the revenue sharing schemes set out in the Constitution of Ghana. Furthermore,
stakeholders have pointed out that there is a disconnect between official policy and legislation, citing

22
collaborative forest management as an important component of the 1994 Policy, which has still not
been captured in legislation (Tropenbos, 2005).

4.3 Restoration of Degraded Lands

About 69% of the total land surface area is estimated as being prone to severe or very severe soil
erosion. Farming practices and the removal of vegetation cover are some of the direct causes of soil
degradation, with deforestation and mining, especially illegal mining activities, contributing the most
serious forms of natural resource degradation. Land degradation has a significant impact on the
environment and natural resource base of the country. Key policy objectives identified to address
this, include reversing land degradation through investments and encouraging appropriate land use
(GSGDA, 2010).

4.4 Reversing Land and Natural Resources Degradation through Investments

Various strategic actions have been identified to reverse natural resource degradation. Appropriate
agriculture intensification techniques that provide irrigation infrastructure and promote correct soil
conservation techniques will be applied, while afforestation of degraded forests and off-reserve areas
would be encouraged, including the adoption of a medium to long-term plan for public and private
programmes. Investments would be encouraged in industrial scale tree farming in specific depleted
Forest Reserves and on degraded land, and in commercial forestry outside forest reserves and along
dried up and flowing streams and rivers. Other strategies include the promotion of plantation and
woodlot development to meet the needs of society; human- centred biodiversity conservation
initiatives; the use of Lesser Used Species (LUS), particularly for the construction industry on the
domestic market; and the utilization of non-traditional tree species such as rubber-wood, coconut
and bamboo to supplement raw material supply from natural forests.

4.5 Biodiversity Conservation

REDD+ goes beyond deforestation and forest degradation, and includes the role of conservation,
sustainable management of forests and enhancement of forest carbon stocks. Although Ghana is a
signatory to the Convention on Biodiversity (CBD), a national CBD Forest Programme of Work
does not exist. What does currently exist is Ghana’s 2002 National Biodiversity Strategy (NBS),
which neglects a number of CBD’s programs of work including that on forest biological diversity. In
order to bring the country’s strategy up to the current requirements of the Convention, and to

23
include the CBD Expanded Programme of Work on Forest Biodiverity, the NBS is currently
undergoing a review. Additionally, the existing NBS does state that Ghana’s policy relating to the
conservation of biodiversity and the sustainable utilization of biological resources are guided by the
three objectives of the CBD Convention, namely:

• Conservation of biological diversity

• Sustainable utilization of biological resources; and
• Fair and adequate sharing of the benefits arising from the use of genetic resources.

Ghana’s Medium-Term National Development Policy Framework: 2010 -2013 recognizes that the
loss of biodiversity is proceeding at an alarming rate, and has thus become a major area of concern.
The policy framework proposes the setting up of a Steering Committee with the following mandate:

• Mainstream biodiversity issues into sector programmes to facilitate the development of

relevant sector biodiversity policies. Another drawback is the lack of research,
• Promote public education and awareness on biodiversity and ecosystem services.

The Ministry of Environment, Science and Technology will during the period facilitate the
collaboration and harmonization of the biodiversity-related agreements, and establish monitoring
mechanisms for biodiversity activities.

4.6 Equitable Benefit Sharing

Access of local communities to resources, such as forests, already exists in Ghana, and form the
basis for common law in certain traditional jurisdiction. Benefit sharing for natural forest timber
revenue, including royalties, and social responsibility agreements, which give forest communities a
financial stake in commercial timber operations exist, and timber permit holders are also required to
negotiate social responsibility agreements with local communities. In practice however, fair access to
resources, fair benefit sharing from corporate exploitation, greater participation in policy-making
processes, and enforcing the laws are the main governance challenges (Amelia et al, 2007).

Although, equitable benefit sharing in the forest sector is seen globally to be one factor that could
contribute to sustainable forest management practices, and policy frameworks have been proposed
and implemented in many parts of the world, some of the mechanisms seem not be working in

24
Ghana, payment of royalties and social responsibility agreements, the main incentive frameworks
between resource extractors, Forestry Commission and forest fringe communities were evaluated
and found not to be accomplishing the policy objectives.(Green Dove, 2007). This has been shown
by the clashes between communities and the Forestry Commission and or logging companies. There
have also been eruptions between rival forest communities over entitlements of royalties (Amelia et
al, 2007).

4.7 Implications of Participating in REDD+

Ghana’s participation in the REDD+ provides a good opportunity to align its forest sector policy,
legislation and governance, towards achieving low carbon economy, in the medium to long term.
Accordingly, new legislation on carbon rights and the rights of people to their own natural resources
would have to be formulated to create the enabling environment for implementing the nationally
proposed REDD+ strategies. Currently, the laws on forestry as well as the National Biodiversity
Strategy are being reviewed, and a new Wildlife Bill is being drafted. The Government also intends
to constitute a multi-sectoral body to upgrade and possibly separate wildlife policy from forest policy
as contained in the 1994 policy (GSGDA, 2010).

The introduction of forest carbon and carbon rights into the country’s sector policies and legislation
will provide legal backing for carbon trading in Ghana, thereby providing the legal basis for slowing
its rate of greenhouse gas emissions, in principle, and possibly having access to multilateral support
to meet her development objectives. It is worthy to note however that the introduction of such new
laws and pieces of “carbon-friendly” legislation with emphasis on forest carbon could transform the
forest management landscape in a manner that diminishes all the other important roles that forests
play in Ghana economic development. Pilot projects under the REDD+ should therefore be used
as test beds to assess the risks and opportunities of any “carbon-friendly” policy. In cases where
related sector policies such as biodiversity and agriculture are incongruous to the forest policy they
should be aligned to make them compatible in responding to the objectives of the REDD+.

The 1994 forestry policy of Ghana recognises the need for collaboration and decentralisation, with
the expectation that involvement of all stakeholders could contribute to sustainable forest
management and improvement in forest governance at the community level. As a result, there has
emerged a number of CBOs and NGOs working the sector ministry, the forestry commission and

25
the donor community to execute programs such as forest boundary cleaning with forest fringe
communities, Modified taungya systems etc) all aimed at encouraging good forestry governance and
livelihood enhancement of forest fringe communities.

These programs, and by extension, the 1994 forestry policy however failed to address underlying
causes of conflict with regards to tree tenure and tree related livelihoods experienced by the local
stakeholder. A critical review of the forest policy does not only require the inclusion of sound forest
governance principles, but the recognition of indigenous ownership and registration of land and tree
tenure at the community level with the aid of Geographic Information Systems. Certification of
such title to land and trees by respective owners in the presence of certified Government Surveyors
could provide an objective documentation of these tenural arrangements and promote the use of
evidence-based conflict resolution mechanisms and thus enhance pro-poor forest governance.

26
Appendix II. National Stakeholder Workshop

The involvement of key stakeholders in any landscape restoration program is essential for the
successful implementation of the program. Sectoral and cross-sectoral coordination should therefore
be established with formal and informal institutions in the forest and related sectors.

The restoration opportunity assessment process begun with organising a national stakeholder
workshop to sensitize participants and to receive commitment from institutions represented during
forest restoration implementation efforts.
1 Workshop Objectives

1. Sensitize participants on the project objectives and FLR principles

2. Receive commitment from institutions represented by participants to collaborate with

project partners during project implementation and beyond

3. Discuss general condition and trends affecting forest areas across Ghana and review criteria
for forest condition scoring defined by Hawthorne and Abu-Juam (1995)

4. Develop suitable forest condition scoring criteria for reserves within and outside the high
forest zone (including savannah and mangrove areas) based on Landsat-type satellite imagery
(30 meter resolution).

5. Identify and develop assessment criteria for sacred groves as important starting points for
landscape restoration.

6. Determine options for restoration of reserved forests: costs, consequences, and constraints

7. Identification of restoration issues, off reserved areas and propose practical approaches to
dealing with such issues

2 Expected Outcomes

1. Participants sensitized on the project objectives and FLR concepts, and principles

2. Institutional commitment received to collaborate with the project

3. Participant understanding of the concept of forest landscape restoration and its components

4. Forest condition scoring proposed for reserves within and outside the high forest zone
based on current forest condition

27
 5. Standard benchmark for Forest condition scoring proposed

6. Options for restoration of reserved forests: costs, consequences, and constraints proposed

7. Forest landscape restoration issues outside reserved areas identified and practical approaches
proposed

3 Organization and Participation

The one-day workshop was held on March 22, 2011, from 9am to 5pm, with snack and lunch
breaks. The agenda was prepared in collaboration with the International Partners, WRI and IUCN.
The welcome address was delivered by a representative of CERSGIS, followed by three
presentations by the International Partners. Four break-out sessions were organized after the
presentations to discuss and report to a plenary four thematic areas identified by the project partners
as relevant to the objectives of the project. This ensured the full participations of each participant,
and provided the platform for extensive discussions on each thematic area.

Major stakeholder organizations, agencies, departments and government ministries as well as civil
society, NGOs and the private sector actors working on forest management and rehabilitation of
degraded forests were invited to the workshop to share knowledge and discuss the feasibility and
options for institutionalizing forest landscape restoration in Ghana.

Twenty-six (26) participants from the Ministry of Lands and Natural Resources, Forestry
Commission, Non-Governmental Organizations, the private sector, and representatives from
International Bilateral Agencies took part in the workshop. Most of the participants were of the
senior or middle management level, involved in forest management activities related to the subject
matter of the workshop. The list of Participants is in Section 9, and the agenda for the workshop is
in Section 10.
4 Presentations
4.1 Assessing opportunities for forest landscape restoration: Why? What? How?

This first presentation was made by Lars Laestadius of WRI. He gave a background to deforestation
on a global scale using examples from countries such as Sweden, Germany and USA in the 1620s,
1850s and 1920s. Maps displayed later in the presentation showed restored lost forests after many
years. Dr Lars pointed out that 5.5 billion hectares of forest has been lost globally and thus the need

28
for forest landscape restoration. At the local scale, the presentation showed the spatial distribution
of degraded forest areas extracted from the Global Landscape Restoration Opportunities dataset,
and indicated that Ghana, and by extension Africa has immense opportunity for forest landscape
restoration. The graph in the figure below illustrates the statistics for forest landscape restoration
opportunities globally.

Opportunities for Forest Landscape Restoration

More than a Billion Hectares

Africa

Asia

South America

Europe & Russia

North & Central America

Oceania

0 100 200 300 400 500

Million hectares
France Wide-scale
To forest restoration Mosaic-type
To mosaic restoration

Opportunities for Forest Landscape restoration

4.2 Restoration in Ghana: What are the experiences to date? What are the needs, options,
benefits, and constraints?

This second presentation was done by Wale Adeleke of IUCN. The presentation centred on
combating climate change through forest restoration. He explained forest landscape restoration and
the stakeholder participation process required as a necessary step for making restoration projects
feasible. This includes bringing people together to identify projects, collective decision making on
projects and negotiating strategies for implementing such projects. He also explained the principles
of forest restoration such as the need to improve ecological integrity; restore ecosystem balance and
collaboration among stakeholders.

Wale enumerated the challenges to forest restoration in Ghana as illegal logging, encroachment on
farms and forest fires, and gave graphic examples of how these driving forces have contributed to
forest cover loss in the Afram Headwaters from 1990 – 2000. The presentation also listed initiatives
by the Forestry Commission to restore lost/degraded forests as, implementation of the national

29
forest plantation development project and the modified ‘Taungya System’ – A Community-based
forest management system designed to alleviate poverty at community level.

4.3 Assessing Restoration Opportunities in Ghana: Strategy and Process

This third and final presentation made by Lars Laestadius of WRI, focused on the objectives of the
forest restoration project in Ghana, the process of implementation, and how it feeds into other
national initiatives such as N-REG, FLEGT, and REDD+.

4.4 Sample comments made after the presentations

1. “Major drivers of deforestation vary spatially. For example in the Afram plains, charcoal
burning is the main driver whereas in the mining areas ‘galamsay’ activities are common
drivers of deforestation”.

2. “The District Chief Executives of Afram Plains Nkoranza North, Kintam,po North
District has been focusing on charcoal production for revenue generation in their his district
without considering the negative effects of the act on the environment. All district
assemblies have to be educated on negative effects of deforestation on the environment.
Charcoal production could be allowed but it has to be sustainable. This requires planting
woodlots for charcoal since its production is a basic source of livelihood for the people”.

3. “Restoration along river banks may require a different restoration approach from forest and
savannah landscapes hence the methods of landscape restoration have to be specific to the
issues at stake”.
4. “There is the need to integrate socio-economic, environmental and legislative instruments
into the forest restoration process in Ghana”.

5. “Savannah areas make up two-thirds of Ghana’s land mass hence when talking about
restoration in Ghana the savannah must be considered”.

6. “The Government of Ghana will need these maps for its forest investment project decisions.

7. “Apart from the UNFCCC arena the map should be finished for showcase at the 2013 UN
Forest Forum”

30
 8. “Current baseline data on forest resources in Ghana is non-existent, thus most figures are
quoted based on outdated information. This project provides us with an opportunity to
produce current maps and thus provide current data for decision makers in the forestry
sector”

9. “National statistics on forest condition are out-dated. The credibility of data concerning
Ghana can be in question. It is therefore necessary that reputable institutions such as the
stakeholders assembled at the workshop make the right inputs so that the project results can
become credible and relevant for Ghana”.

5 Break-Out Sessions

Four work groups were constituted with guidelines to address the following issues:

Group1: Criteria for degradation scoring of reserves

Group 2: Criteria for mapping of sacred groves and mangroves
Group 3: Options for restoration of the above categories and their costs, consequences, and
constraints
Group4: Restoration issues concerning off reserves

6 Guidelines for Group Work

6.1 Group 1: Criteria for degradation scoring of reserves

The most immediate opportunities for restoration of forest landscapes in Ghana are in the forest
reserves. A country-wide assessment of the forest quality in the reserves was made in the mid 1990’s
(Hawthorne and Abu-Juam, 1995). This assessment has two limitations:

• It has not been updated

• It is limited to the high forest zone

31
Updating and expanding the degradation assessment across the whole country is a high priority.
New satellite images will help make this feasible.

Tasks for the group:

1. What is the general condition and trends affecting reserves across Ghana? Please state your
opinion by forest type.

2. How can degradation in reserves be scored using Landsat-type satellite imagery (30 meter
resolution)? Please propose criteria that are suitable for:
a) Reserves within the high forest zone
b) Reserves outside the high forest zone

3. What are the significant issues and options, and what are the reasons for your suggestions?
Please list for each set of criteria.

6.2 Group 2: Criteria for mapping of sacred groves, mangroves

Sacred groves and mangroves provide important starting points for restoration of an otherwise
degraded landscape. The location and condition of these should therefore be known.

Tasks for the group:

1. What is known about sacred groves and mangroves? Please list known sources of
information on their location and condition, e.g. maps, databases, books, people who know,
etc.
2. How can we map the location of sacred groves and mangroves? Please propose ways,
including criteria that would work with satellite imagery of Landsat type (30 meter
resolution) or better
3. Where do mangroves need to be restored? What areas are the most important to restore?
What are the constraints? Which areas provide the best opportunities and why? Please
indicate your answers by drawing on the map and provide commentary.

32
6.3 Group 3: Options for restoration of reserved forests: costs, consequences, and constraints

Forest reserves can be restored in many ways, from community based fire control to intensively
managed plantations. Different conditions require different treatments.

Tasks for the group:

1. What are the constraints on restoration in reserves? List by forest type.
2. What are the realistic options for restoration of reserves? List by forest type.
3. What are the typical costs of each option? Please list by forest type.
4. Where is restoration most likely to be possible and what kind of restoration is realistic?
Please show on the map.

6.4 Group 4: Restoration issues off reserves

Most of the degraded lands in Ghana are off reserves. Restoring these lands is obviously difficult. It
is important to understand the issues concerning off reserves in order to systematically map the
opportunities for restoration.

Group Task: Identification of factors militating against forest restoration in Ghana.

7 Summary of Group Presentations

7.1 Group 1: Criteria for degradation scoring of reserves

Hawthorne and Juam (1995) applied a single score to each forest reserve in the high forest zone, to
summarise its general condition. The condition score run from 1 to 6, with condition 1 for forest
reserves with minimal signs of disturbance and condition 5 forests being seriously degraded, with
few upper canopy trees. Condition 6 described reserves with no significant forest left.

The group agreed that the Hawthorne and Juam (1995) condition scores be reviewed to reflect
current forest conditions, taking into consideration the new definition for “forest” proposed by the
Forestry Commission of Ghana. The Group also proposed the definition of woodland condition
scores for the savannah zone to take account of woodland patches in reserved and non-reserve
areas. Detailed description of the format for defining condition scores for forest and savannah
vegetation types can be found in Appendix III.

33
Where intact forest can be found in the landscape, this could be used as reference for forest
condition scores. Forest relics such as sacred groves, which constitute areas of intact forests
reserved by traditional authorities due to religious and traditional beliefs, should for instance be used
as benchmark for defining Excellent Condition or Condition Score 1 for the forest or woodland
reserves. This is because such sacred groves are usually kept intact than are poorly represented in
state-managed Forest Reserves due to a variety of underlying beliefs and prohibitions against felling
of trees which add considerable genetic value to forests or woodland within such groves.

Proposed criteria for Condition Scoring:

1. Canopy cover
2. Human disturbance
3. Natural regeneration
4. Logging damage
5. Fire damage

7.2 Group 2: Criteria for mapping of sacred groves, mangroves

The group defined sacred groves as forest relics preserved by traditional norms customs and taboos.
These groves are of religious and cultural importance to the people living around it. Widely
distributed across the country and found in small patches, they used to be burial grounds for
traditional rulers. Some of these groves are getting degraded and need to be restored. Mapping and
inventory of sacred groves are required to identify the areas to generate digital maps of the sites

Sources of information were identified as community and traditional authorities, Forest Services
Division, UNDP and Wildlife Division of the Forestry Commission.

The following specific sources of information were identified:

1. Forestry Commission - List and status of sacred groves in Ghana

2. Resource Management Support Center - Map of sacred grove distribution in Ghana.

34
The group identified the following constraints for restoration within mangrove areas:
1. Accessibility: The mangroves are easily accessible to the people due to the location of
settlements close to them as such they are cut for domestic wood energy and for commercial
fish smoking.
2. Resources/logistics: It requires a lot of financial and logistical resources to restore
mangroves but such needed resources are not easily available.
3. Technical Capacity: Mangrove restoration requires much technical knowledge which is
limited.
4. Policy initiatives: For mangrove restoration to be successful government policy and laws to
protect them have to be in place.
5. Urbanization: Urbanization of coastal mangrove areas is contributing to development of
settlements and other land use activities that result in the rapid loss of mangroves.

7.3 Group 3: Options for restoration of the above categories and their costs,
consequences, and constraints

The group’s analysis was done for each ecological zone namely, high forest, transition and savannah
eco-zones. The results are presented by ecological zone in the 3 tables below.

Cost elements such as social, environmental and economic are associated with landscape restoration
for each ecological zone. The tables also give information on suggested areas for restoration, as well
as some relevant information sources.

35
 HIGH FOREST Costs
ZONE Relevant
Proposed Projects/Sources of
Options Constraints Social Environmental Economic Restoration Information
Conversion to plantations Compaction of soil by Lack of access and Loss of biodiversity Land-preparation Cultivation of GSBAs
e.g. through Modified logging machinery affects ownership of land by especially under pegging, ploughing, multipurpose
Taungya System and use of natural regeneration migrant farmers thus plantations planting trees in admitted
indigenous species. Avoid affecting tree tenure farms
monoculture to improve
survival rates
Enrichment planting (filling Nutrient availability Introduction of alien Cost of seedlings Carbon map
in the gaps) and invasive species
Natural regeneration Inadequate capacity and Loss of access to land and Sometimes sources of Plantation Maintenance FLR Map
motivation/ thus connected to livelihood drinking water are Fire protection
remuneration for Forest activities disturbed under
Commission staff plantations
Sustainable logging Indigenous cultural systems Fire protection The current Plantation
may be adulterated or lost Development Programme
due to influx of migrants under the FC
Time and other resources Compaction of soil Transportation High Forest Biodiversity
invested in restoration by Programme
farmers/ communities
Cost of maintaining Climate Change Unit,
feeder roads FC

Average cost of
establishing a plantation/
Ha. is GH¢2030 (Ref:
Plantation Dept-FC)
Logging damage control
 TRANSITION Costs Relevant
ZONE Social Environmental Economic Proposed Projects/Sources of
Options Constraints Restoration Information
Conversion to plantations Lack of Maintenance of Lack of access and Loss of Irrigation facilities e.g. Around ecologically Carbon map
e.g. through Modified trees by farmers especially ownership of land by biodiversity dug outs sensitive areas e.g. water
Taungya System (MTS) to under MTS migrant farmers thus especially under bodies (enrichment
avoid monoculture to affecting tree tenure plantations planting)
improve survival rates
Enrichment planting (filling Breach of contract by Introduction of Land preparation- Law enforcement to back FLR map
in the gaps) farmers e.g. under MTS alien and invasive pegging, ploughing, restoration
species planting
Natural regeneration under Competition from food Loss of access to Sometimes sources Cost of seedlings Woodlots especially The current Plantation
controlled fire and grazing crops; once farmers are land and thus of drinking water (charcoal endemic areas) Development Programme
given access to reserves, connected livelihood are disturbed under under the FC
they give priority to the activities plantation
food crops especially in the
savannah where soil
moisture is very low
Fire and grazing Indigenous cultural Maintenance Cultivation of
systems may be multipurpose trees in
adulterated or lost admitted farms
due to influx of
migrants
Illegal logging Time and other Fire protection Because a lot of the forests The FC Climate Change
resources invested in have been lost, the Unit
restoration by transition zone provides a
farmers/ good opportunity for
communities restoration
Inadequate capacity and Transportation
motivation/ remuneration
for FC staff
Increased wear and
tear of roads

37
 Costs Relevant
SAVANNAH Projects/
ZONE Environ Proposed Sources of
Options Constraints Social mental Economic Restoration Information
Conversion to Lack of Maintenance of trees by Influx of migrants into the Loss of Irrigation facilities e.g. dug Around Carbon map
plantations e.g. farmers especially under Modified community with different biodiversity outs ecologically
through Modified Taungya System (MTS) motives which might lead especially under sensitive areas e.g.
Taungya System and to social vices such theft, plantations water bodies i.e.
better to avoid promiscuity, etc ; the black, white
monoculture to implications on and Red Volta
improve survival rates reproductive health and (enrichment
population growth planting)
Enrichment planting Breach of contract by farmers e.g. Introduction of Land preparation- pegging, Law enforcement FLR map
(filling in the gaps) under MTS alien and invasive ploughing, planting to back restoration
species
Natural regeneration Competition from food crops; once Sources of Cost of seedlings Woodlots Savannah resource
under controlled fire farmers are given access to reserves, community especially (charcoal management
and grazing they give priority to the food crops drinking water are endemic areas) project
especially in the savannah where soil disturbed under
moisture is very low plantations
Fire and grazing Loss of access to land and Indigenous Maintenance Cultivation of The current
thus connected to cultural systems multipurpose trees plantation Dev’t
livelihood activities may be adulterated in admitted farms programme under
or lost due to the FC
influx of migrants
Customs and taboos which prevent Fire protection
planting of certain tree species may
be a constraint. The fire festival in
the north may attest to how fire use
as a customary practice can affect
restoration
Fire is sometimes used as catalyst Time and other resources Transportation
for initiating buds in trees such as invested in restoration by
shea and dawadawa. Poor control of farmers/ communities
the fires my result in destruction of
vegetation
Droughts Increased wear and tear of
roads
Illegal mining Inadequate capacity and Average cost of establishing a
motivation/ remuneration plantation/ HA is 2030
for FC staff (plantation dept)

38
7.4 Group 4: Restoration issues off reserves

The group identified the following as factors militating against forest restoration in Ghana.

7.4.1 Land and Tree Tenure

Problems of who owns the land, and who owns the trees on the land do exist. This negatively
affects restoration efforts. Fragmented land use systems also do not encourage forest
restoration as land parcels are too small for forest restoration projects.

7.4.2 Crop Types

Crops such as maize, rice, beans cannot do well under shade hence when such crops are
grown in an area it will be difficult to practice forest restoration using agroforestry method.
Tree crops like hybrid cocoa does not do well under much shade hence it will not be
practicable for forest restoration in such areas. Establishment of jatropha plantations for bio-
fuel may constrain forest landscape restoration in that land preparation practices for such
plantations may not be useful for forest restoration.

7.4.3 Forest fires and land uses

Particular areas within Ghana are susceptible to frequent forest fires hence forest restoration
will have challenges to deal with. Sand wining, ‘Galamsey’ (Illegal mining), charcoal
production and wood fuel exploitation are further threats to forest restoration.

7.4.4 Benefit sharing

Once there are laws that declare trees on individual farm lands as Government property,
farmers deliberately destroy timber species on their farms since they know they are not the
beneficiaries when the trees mature to timber size.

7.4.5 Physical development

Physical development such as, road construction and the expansion of communities, towns
and cities, could militate against restoration efforts. The Accra plains for instance has been
affected by massive estate developments around the Shai Hills National Park.

39
7.5 Existing efforts to restore forests in Ghana

1. Plantation development and restoration of degraded forests by the Forestry

Commission
2. Private sector initiatives such as Samatex Ghana Forestation
3. Establishment of Community Resource Management Areas (CREMA) in Wasa Amenfi
District
4. Collaborative Mangrove Restoration Projects in Muni Ramsar Site, Songor and Ada
Ramsar Sites
5. In an effort to protect wildlife some protected areas such as Boabeng Fiema and Tafi
Atome monkey sanctuaries have been created and expanded.

7.6 Some Practical Approaches

1. In the savannah areas, planting of trees along rivers and other water bodies will be
useful.
2. Community participation is very important. For example, the Densu Restoration
Project was not successful due to lack of adequate community participation.
3. Natural regeneration should be encouraged in savannah ecological zones.
4. Social approaches: Preservation of burial grounds by turning them into forest as they
mostly have some existing tree cover.

7.7 Map Requirements.

1. Land Use and Land Cover

2. Land suitability
3. Land ownership: stool lands and family lands.

7.8 Short Term vrs. Long Term Benefits

Traditional Rulers who benefit from land use activities that contribute to deforestation will not
be willing to stop such activities in favor of forest restoration. For example, chiefs in the
Gonja Traditional Area who benefit from charcoal revenue in their traditional areas will not be
willing to enforce laws banning charcoal production.

Conversely, traditional areas where sacred groves are found welcome restoration efforts to
expand the groves.

40
7.9 Potential Users of FLR Map

1. The group identified the following as potential users of the FLR Map:
2. Climate Change Unit of the FC
3. District Offices of the FC
4. Ministry of Lands and Natural Resources
5. Ministry of Environment Science and Technology
6. District Assemblies
7. Environmental NGOs (National and International)

8. Workshop Recommendations
1. Consultation and participation of all stake holders is necessary before any landscape
restoration project is initiated. This is to ensure the participation and commitment of
communities to the project.
2. The integration of indigenous knowledge into proposed Forest Landscape Restoration
(FLR) concepts and approaches is very essential for the acceptance of restoration
projects in communities.
3. There is the need to build on already existing initiatives regarding forest landscape
restoration. This will ensure the harmonization of project outcomes for national
development thus avoiding duplication of efforts.
4. Restoration of forest landscape in Ghana has to conform to global standards to be
considered credible and accepted internationally.

41
9. List of Participants

Name Institution Position

1 Musah Abu-Juam Ministry of Lands and Natural Resources Technical Director
2 Raphael Yeboah Forestry Commission -FSD Executive Director
3 Kwakye Ameyaw Forestry Commission -FSD Operations Manager
4 Roselyn F. Adjei Forestry Commission Assistant Manager
5 David Kpelle Forestry Commission Consultant
6 Charles Amankwa Forestry Commission-WD Wildlife Co-odinator
7 Richard Gyimah Forestry Commission Manager
8 Sulemana Adamu Forestry Commission Asst. Programme Manager
9 Nana Poku Bosompim Forestry Commission Administrative Officer
10 Yakubu Mohammed Forestry Commission -RMSC Manager GIS
11 Grace Gyabaah Forestry Commission -RMSC Administrative Officer
12 James Agyei-Ohemeng NCRC Comm. Rel. Officer Tech.
13 Winston Asante NCRC Research Officer
14 Aristotle Boaitey Kumasi Wood Cluster Nat. Service Personnel
15 Alex K. Dadzie Ghana Timber Association Vice President
16 Osofo Quarm Wassa Amenfi West District Executive Director
17 Justice Odoi USAID-Ghana Environmental Specialist
18 Susanne Bouma Royal Netherlands Embassy Intern
19 Clare Bragon EU Delegation Advisor
20 Evyon Pfeil GIZ Team Leader
21 Vincent Awotwe-Pratt Conservation Alliance Int. Programme Manager
22 Lars Laestadius World Resources Institute Senior Associate
23 Wale Adeleke IUCN Coordinator
24 Saadia Bobtoya IUCN REDD Project Officer
25 Opoku Pabi CERSGIS Landscape Ecologist
26 Selase Adanu CERSGIS Environmental Systems
Spec.
27 Kwadwo Kyeremateng CERSGIS Remote Sensing/GIS
Spec.
28 Foster Mensah CERSGIS Executive Director

42
10. Program

Dialogue on
REDD+ OPPORTUNITIES THROUGH FOREST LANDSCAPE RESTORATION

Date: Tuesday 22 March, 2011

Venue: Alisa Hotel, Accra

9:00 Welcome and opening (Foster Mensah)

9:15 Assessing opportunities for forest landscape restoration: Why? What? How? (Lars
Laestadius)

9:45 Forest landscape restoration in Ghana: What is it? What are the experiences to date?
Are there any bright spots in Ghana or elsewhere? What are the needs, options,
benefits, and constraints? (Wale Adeleke)

10:45 Assessing restoration opportunities in Ghana: strategy and process (Lars, Foster)

11:15 Breakout group session

Introduction of break out groups
Criteria for degradation scoring of reserves
Criteria for mapping of sacred groves, mangroves
Options for restoration of reserved forests, sacred groves and mangroves and their
costs, consequences, and constraints
Restoration issues off reserves

12:15 Lunch

13:00 Breakout group sessions (continued)

13:30 Breakout group reports and discussion

15.30 Plenary discussion

16:30 Conclusions, next steps

17:00 Closing Remarks

43
Appendix III. Method

1. Ecological Zones in Ghana

It is essential that forest landscape restoration at national scale be preceded by the

characterization of vegetation types over the landscape which the restoration is intended and
the criteria that qualify an area for restoration. Furthermore, landscape approach to restoration
also takes into account the various land uses and ecosystems (forest, savannah, forest-savannah
transitional belt, mangrove, etc). A multiple-use landscape is often a mosaic of land use types,
in which some parts may be formally protected (e.g. forest reserves) while others are
sustainable-use communal areas, private farmland and/or other types of productive use (e.g.
off-reserve areas serving as agricultural landscapes).

A country-wide assessment of the forest condition in Ghana’s reserves was made in the mid
1990’s (Hawthorne and Abu-Juam, 1995). Currently, reliance on this assessment has two
limitations in the sense that it has not been updated since 1995 and it is also limited in scope
(i.e. only for forest reserves in the high forest zone). Updating and expanding degradation
assessment across the whole country is of a high priority. This therefore requires a review of
the Hawthorne and Abu-Juam forest condition that takes into consideration the major
vegetation types and hence ecological zones of Ghana.

44
In order for such broad ecological zone classification to be relevant, there is the need to define
the scope and limits of the major vegetation types and develop suitable condition scoring
criteria for the vegetation types within and outside the high forest zone (including savannah
and mangrove areas) based on 30 meter resolution Landsat satellite imagery.

1.1 Classification of Ecological Zones

Ghana, which is in the western part of

the African continent (0−3° W and
5−11° N), has a land area of
approximately 239,000 km². Identifiable
ecological zones in the country include
rain forest (3 %), moist forest (31 %),
interior savannah (57 %), coastal
savannah (5 %) and swamp vegetation (4
%) (Ntiamoah-Baidu, 2001).

Broadly two major kinds of vegetation

cover the surface of Ghana: forest and
savannah; these and the various
formations into which they may be
subdivided were defined at a Pan-African
conference at Yangambi (C.S.A. 1956),
UNESCO (1973) and the most recent by Figure 1: Distribution of key vegetation types in Ghana

Hall and Swaine (1981).

1.2 Extent of Ghana’s forest and Trends over the last half century

Forest is dominated by trees at least 5 m high, whose crowns interlock, and lacks a continuous
grassy ground layer. The area of forest zone in Ghana is about 82,000 km2, representing 34%
of the total area of Ghana. The forest zone is occupied by various forest types as described in
detail in Table 2 (Hall and Swaine, 1981).

45
Table 2: Areas occupied by the various forest-types in Ghana

Area (km2)
Percent of total
forest area
Types and subtypes subtype type

Wet Evergreen 6,570 8.1

Moist Evergreen 17,770 21.9

Upland Evergreen 292 0.3

Moist Semi-deciduous 32,890 40.4

South-east 18,460

North-east 14,430

Dry Semi-deciduous 21,440 26.4

Inner Zone 8,630

Fire Zone 12,810

Southern Marginal 2,360 2.9

South-East Outlier 20 0.03

Total Forest Area 81,342 100

1.3 Forest Reserves

Although the vegetation within forest reserves is less heterogeneous it is by no means uniform.
It is on record that forest quality and quantity in the reserves have decreased over the years.

Forest ordinances were established from 1927 for the reservation and subsequent management
of forests. The then Forestry Department had powers to reserve areas for management by the
state, in some (but not all) cases in agreement with chiefs to whom the land belong. With these
powers, reserves were defined in all the major hills and watersheds. Furthermore, barrier and
shelterbelt reserves were established to hold back wildfires, and to maintain local rainfall and
humidity levels respectively. Shelter belts are corridor strips 1-2 km wide stretching northwest
to south east. The barrier reserves fringe the savannah zone. Some of these included savannah
woodland even when reserved, and were (some still are) managed with early burning in the

46
belief that eventually forest would return if the severity of the annual fires was kept to a
minimum (Hawthorne and Abu-Juam, 1995).

The boundaries between northern savannah woodland and the southern high forest used to be
abrupt but the frequent interventions of wildfires have changed extensive portions of the
northern part of the high forest into semi-forest and savannah and therefore has decreased the
forest resource diversity gradient between the northern savannah and the high forest zone. The
gazetted forest and wildlife reserves of Ghana stands as 266 forest reserves, 15 Wildlife
Reserves and Parks and Ramsar Sites. The total area covered by these reserves is approximately
3,800,000 ha and constitute about 16% of the total land area of Ghana.

The present state of the reserved forests is in a worse condition compared to the time of their
reservation. The reserves have had a long history of human interference and its cumulative
effect today, is obvious. Several factors have contributed to the bad or secondary phases of the
forest mosaic. For instance, excessive logging and uncontrolled fire have destroyed some
reserves in the Brong-Ahafo region which were originally kept as shelter belt reserves
protecting the forest zone in the south from the advancing savannah from the north. It is on
record that the original area (82,000 km2) of forest zone in Ghana at the turn of the century
has shrunk to about 16,000 km2 this does not include plantation forest and others due to
natural regeneration.

1.4 Unreserved Forest Areas (Off-reserves)

Off-reserve areas are lands that are currently (or were) forests but where the forest policy
presumes these lands would be converted to other use—in particular, agriculture. This includes
about 5,000 km² of unreserved forests, approximately 60,000 km² of bush fallow,
approximately 36,000 km² of unimproved pasture, and about 29,000 km² of tree and annual
cropland World Bank (2007).

The rate of forest clearing outside forest and wildlife reserves is such that intact forest will
soon have virtually disappeared except within reserves and in small patches of sacred forest
near villages. The proportion of forest outside reserves declined from 20% in 1995 to 5% in
1972 according to estimates in annual reports of the Forestry Department for those years (Hall
and Swaine, 1981).

47
In most cases forest clearing in unreserved areas is for agricultural purposes. In cash crop
farms such as cocoa, more shade trees are left standing over the cocoa farms whereas in food
farms most trees are cleared and burnt. Swamp vegetation often dominated by raphia palms
occur in isolated areas across the High Forest Zone but is common in some parts of the
Western Region (e.g. Tarkwa), where they are sometimes cleared for rice-farming. The
unreserved landscape is therefore characterized by a mosaic of agricultural landscape and
secondary vegetation showing different stages in regrowth on old farms. When a forest is
cleared, tall herbaceous or sub-shrubby weeds emerge (“forb regrowth”) which may be
followed by thicket with abundant woody climbers and young trees which shade out and kill
the light demanding weeds. The pioneer trees are quick growing and short-lived species which
form the early succession phase of the secondary forest. Taller and long-lived species take over
after 10-20 years to give older secondary forest from which the mature forest gradually
develops. Thus the unreserved forest areas are greatly influenced by anthropogenic factors
compared to the reserved forest areas.

1.5 Extent of Ghana’s Savannah & Trends over the Last Half Century

The boundary between forest and savannah is usually described as a transitional zone. This
zone has been greatly influenced by combined effect of geology, topography and rainfall. The
stability of the forest-savannah boundary is enhanced by annual savannah wildfires which tend
to prevent the spread of forest (Hall and Swaine, 1981). The crowns of trees in the savannah
are not continuouses but are separated by a distance depending on the stand density, or in
extreme cases trees may be absent; there is always a seasonally dense under storey of grasses
(Hall and Swaine, 1981). Approximately 66% of the total area of Ghana is savannah. Savannah
ecosystems and dynamics are currently poorly understood because little attention has been
paid to these areas in the past.

In Ghana, vegetation trends for savannah ecosystems have not been well been documented
and national inventory programmes have only been restricted to forest zones. Data on the
dynamics of savannah vegetation is therefore non-existent and poorly described. However, it
is widely acknowledged that rainfall amount, soil types, fire history, grazing and farming
practices have influenced the savannah vegetation over the years in Ghana to develop into
what is commonly termed “fire climax” vegetation. Hawthorne and Abu-Juam, (1995)
observed that a number of reserves in the savannah zone of Ghana never had forest in them,
but rather are savannah or open woodland reserves. Examples of these could be cited in forest

48
reserves north of Sunyani in the Brong-Ahafo Region and in the Volta Region.

Most of the degraded reserves in the transitional zone were once forested or partly forested at
the time of reservation, but which have subsequently been destroyed. Some were burnt,
farmed, and then suffered failed Taungya with subsequent repeated wildfires. Most of the
transitional zone vegetation contains scattered fragments of forest that require attention for
further enhancement and/ or protection. The ecological significance of the transitional zone
forming a barrier protecting the forest zone from the rapid advancement of the savannah from
the north requires that their forest restoration potential is carefully considered and uniquely
managed. Hence the transitional zone is considered as a key ecological zone in this assessment.

49
Figure 2: land cover classification from mosaic of Landsat images

50
 Burkina Faso
V
&
BAWKU
ZEBILLA
BONGO
NAV RONGO
TUMU V
&
V
&
V
&
V
&

SANDEMA V
& BOLGATANGA
V
&
LAWRA
V
&
JIRAPA GAMBAGA
V
& V
& N
NADAWLI
WALEW ALE
V
&
V
&

WA
V
&

GUSHIEGU
V
&

SABOBA
V
&
SAVELUGU
V
&

TOLON YENDI
TAMALE
V
& V
&
V
&
ZABZUGU
V
&

DAMONGO
BOLE V
&
V
&

BIMBILA
V
&

Republic
of
SALAGA Togo
V
&

NKWANTA
Republic V
&

of
Cote D'ivoire KINTAMPO
V
&

KETE-KRACHI
WENCHI ATEBUBU
KWAME DANSO V
&
V
& V
&
V
&

DROBO TECHIMAN NKORANZA

V
& V
& V
&
KADJEBI
V
&
BEREKUM
V
& JASIKAN
EJURA
SUNYANI V
&
V
&

DORMAA AHENKRO V
&
V
&

HOHOE
BECHEM
V
&
V
&
DONKORKROM
TEPA KPANDU
KENYASE NO. 1 V
&
V
& OFINSOAGONA AKROFOSO V
&
V
&
V
& V
&
EFFIDUASE
GOASO MANKRANSO
V
&
V
&
V
& V
&
KUMASI
NKAWIE V
&
MAMPONG ASHANTI V
& EJISU
V
&
MPRAESO HO
V
&
V
& JUASO
KUNTENASE V
&
V
& V
&

MANSO NKWANTA V
&
BIBIANI BEKWAI
V
& V
&
V
& BEGORO
JUABESO NEW ABIREM V
&
V
& V
&
ATIMPOKU
SEFWI WIAWSO OBUASI
V
&
V
& KIBI
V
& AKATSI
KADE V
& KOFORIDUASOMANYA V
&
DENU
NEW EDUBIASE ODUMASE KROBO ADIDOME V
&
V
& SUHUM V
& V
& V
&
V
& V
& SOGAKOPE
DUNKWA-ON-OFIN V
&
AKIM ODA V
&
KETA
V
& V
& AKROPONG AKWAPIM
V
& ASAMANKESE DODOWA V
&
ENCHI ASANKRAGUA V
& NSAWAM V
&
ADA FOAH
V
& V
& V
&
V
&
ASSIN FOSU AMASAMAN
V
& V
&
TWIFO PRASO TEMA MUNICIPAL AREA
BREMAN ASIKUMA V
&
V
& ACCRA
V
& AGONA SWEDRU
V
& V
&
AJUMAKO
V
&
ABURA DUNKWA WINNEBA
TARKWA V
& Towns
V
& APAM V
&
V
&
V
&
SALTPOND
National Park
DABOASE V
&
V
& ELMINA Forest Reserve
HALF ASSINI V
& CAPE COAST
V
&
V
& a
in e Waterbody
Gu
SEKONDI
AGONA NKWANTA V
&
AXIM
V
&
V
&
G u lf of

Figure 3: Spatial Distribution of Forest Reserves and National Parks in Ghana

51
2. Forest Condition Scoring

Hawthorne and Abu-Juam (1995) applied a single score to each forest reserve, to summarize
its general condition as indicated in Table 2. The condition score runs from 1 to 6, with
condition 1 forest reserves showing minimal signs of disturbance and condition 5 forests being
seriously degraded, with few upper canopy trees. Condition 6 reserves have no significant
forest at all. Scores 1-3 are low to moderate, and ecologically tolerable, with healthy, vital parts
of the mosaic in the ascendance and regeneration of timber trees and other forest plants
usually abundant. Well-managed selective logging should generate condition 2-3 forest by the
end of a felling cycle. Scores 4-6 are applied to forests that have suffered slight degradation
over more than half the reserve or heavy degradation over more than one-quarter of it. Very
often significant parts of the mosaic have poor regeneration of timber trees and other forest
plants, partly because of lack of ‘good’ parts of mosaic nearby. Badly managed logging typically
produces condition 4-5 forest. In dry areas these forests are susceptible to repeated wildfires
because of the density of flammable, young pioneer plants in the dry season, and this has often
lead to deterioration of forests to condition 5 or 6, particularly where Taungya has been
encouraged.

The scores applied in the work of Hawthorne and Abu-Juam (1995) were based on
observation, conversations with forest staff working in the area, the inventory results and use
of satellite imagery in decreasing order of importance. A combination of inventory data from
a 0.25% Temporary Sample Plot (TSP) gathered around 1986 and botanical surveys of the
early to mid 90s were used in the forest condition scoring. From the botanic survey canopy
trees (exposed to sun,>30 cm dbh) recorded were used to estimate Pioneer Index (PI),
Economic Index (EI) for those trees and Genetic Heat Index (GHI). The PI is a weighted
average with pioneers scoring 2, non-pioneer light demanders scoring 1, and others scoring
zero. The sum of these scores (i.e. over all species for PI over all tree stems > 30cm dbh for
PI (trees) is multiplied by 100 and divided by the number of stems or species in the sample.
The EI (trees) is similarly a weighted average of stem scores multiplied by 100, where each
stem of a scarlet species scores 3, Red star stems score 2, pink star stems score 1, other stars
score zero. The GHI is a similar index for species content, relating to global conservation
concern. Indices based on GHI were included if enough species (>30) were included in the
botanic survey sample. Additionally, ‘average forest type’ was estimated from average
ordination score as used by Hall and Swaine (1976; 1981).
According to Hawthorne and Abu-Juam (1995) in many cases the scores allocated is open to

52
debate. Nevertheless, they ctaught the general trend and reflected the true situation. Although
the pioneering work by the authors is commendable, the level of subjectivity is fairly high.
Vide Table 3 for condition score used to summarize forest reserves (Hawthorne and Abu-
Juam, 1995). This assessment is most critically limited to only forest reserves in the southern
part of Ghana. Since 1995, there has not been any attempt to evaluate and re-assess the forest
condition scores put forward by the authors mentioned above.

Table 3: Scale used by Hawthorne and Abu-Juam (1995) to Condition scores used to summarize the condition of
forest reserves

Score Definition
1 EXCELLENT with few signs (<2%) of human disturbance (logging/farming) or fire damage,
with a good canopy and virgin or late secondary forest throughout.
2 GOOD with <10% heavily disturbed. Logging damage restricted or light and well dispersed. Fire
damage none or peripheral.
3 SLIGHTLY DEGRADED: Obviously disturbed or degraded and usually patchy, but with good
forest predominant. Max 25% with serious scars and poor regeneration; max. 50% slightly
disturbed, with broken upper canopy.
4 MOSTLY DEGRADED: Obviously disturbed and patchy, but with bad forest predominant; 25-
50% serious scars but max. 75% heavily disrupted canopy. Or forest lightly burnt throughout.
5 VERY POOR: Forest with coherent canopy < 25% (more than three-quarters disturbed), or more
than half of the forest with serious scars and poor or no forest regeneration; or almost all heavily
burnt with conspicuous Eupatorium and other pioneers throughout. Not, however, qualifying as
condition 6.
6 NO SIGNIFICANT FOREST LEFT: Almost all deforested with savannah, plantation or farm
etc; <2% good forest; or 2-5% v. disturbed forest left; or 5-10% left in extremely poor condition
e.g. as scattered trees or riverine fragments remnants with little chance of surviving 10 years.

53
2.1 Savannah Condition Scoring
Globally, the Savannah biome is diverse, including formations ranging from almost pure
grasslands to closed woodlands with considerable variation in plant composition, biomass and
net primary productivity (House and Hall 2001). The distinction between what is forest,
grassland, and different structural savannah types can only be set with arbitrary limits and
various descriptions such as those defined by Scholes and Hall (1996):

• Forests: complete tree canopy cover and three or more overlapping vegetation strata,
• Woodlands: 50-100% tree canopy cover, and a graminaceous layer,
• Savannahs: 10-50% cover by woody plants and well-developed grass,
• Grasslands: Less than 10% tree cover.

The Dry Semi-deciduous type forest is a fairly good starting point for tracing the path of
savannah and hence, scoring the different types of savannah vegetation that exists in Ghana.
The Dry Semi-deciduous type forest exists under a wide range of environmental conditions
and it forms a peripheral band around the moister forest types and is adjacent, in the north, to
the Guinea savannah zone. Two sub-types of this forest (i.e. Fire Zone subtype and inner
Zone subtype) provide a useful vegetation clue of a transition from forest to savannah
vegetation. Vegetation in the Volta region and Afram Plains in the Eastern region indicate
such trends. Southern Marginal type also exists as coastal thicket and scrub. Relatively little of
this type is today covered by forest (e.g. Amoawa hill between Winneba and Mankesim. The
land is mostly covered by thicket, farms and savannah (partly derived from more woody
vegetation by farming). Forests occur as small patches mostly on rocky hills not favored for
farming. The south-east outlier is forest at its driest extreme in Ghana with the lowest mean
annual rainfall reaching about 650 mm. Canopy trees in this forest are short. This type of
forest occupies a small area of Ghana. Example is the Shai Hills Game Production Reserve.

Unlike the forest zone of Ghana, the savannah zone lacks detail inventory studies that could
be relied upon for savannah condition scoring. The relative dominance of trees versus grasses
in savannah systems of Ghana varies widely according to rainfall amount, soil type and fire
history, as well as grazing intensity and the overall extent of past disturbance.
Annette den Hollander H.J. (2008) conducted a study in savannah sites in Mali, Burkina Faso,
Ghana, and Cameroon all in the African continent. The aim of this study was to make
observations concerning other driving variables in the savannah ecosystem. The Gentry
methodology used in this study consist of an inventory of the vegetation:

54
 1. Estimating percentage cover of the lower vegetation layer (<1.5 m).
2. Weighing grass ‘clipping’ sample for biomass.
3. Measuring crown of the shrubs (>1.5 m) and DBH1.3 of trees.
4. Measuring crown separation ratio of the shrubs and trees (>1.5 m).
5. Estimating percentage canopy cover.

The Mole National Park in Ghana which constitutes northern Guinea savannah was used as
the study site for Ghana. The long history of this park as a conservation area in Ghana and
also a typical savannah ecosystem makes it a good proxy as far as savannah inventory data is
concerned. The study revealed that the total lower vegetation layer (shrub, grass and herb)
covered about 35% of the area whereas woody vegetation layer by way of canopy trees and
crown cover was 51 and 23%, respectively. The canopy cover measured with the spherical
densitometer and varies between 4 and 60%. Data from the Mole National Park was relatively
high in canopy cover because of the relatively abundance of woody vegetation. The crown
cover is estimated from the crown separation ratio, which is estimated from the crown of
individual trees and the gap between the crowns of trees. From this crown cover the savannah
can be categorized in different savannah vegetation types (see Table 4).

In open savannah the trees are more widely separated; trees are fairly closely and evenly spaced
but do not form a closed canopy in the case of open savannah woodland. In savannah
woodland the trees form a more or less complete canopy. Canopy Cover underestimates the
crown cover compared to the Australian crown separation method. An explanation for this
could be the method of measuring the canopy cover with the spherical densitometer, which is
hold at breast height. Because of this the lower canopy is not included, while in the crown
separation ratio method, this is included (Annette den Hollander H.J. 2008).

Table 4: Savannah Classification with Crown Cover for the sampled area in Mole National Park [Annette den
Hollander H.J. (2008)]

% Crown Cover Cover Class Vegetation Type According to

in Mole National described by
Hnatiuk et al. 2009
Park
Scholes & Hall Kaey 1949 White 1983
1996
23 Sparse Savannah Open Wooded
Savannah Grassland
Woodland

55
The woody vegetation included shrubs < 2.5 cm DBH1.3, small trees of DBH1.3 between 2.5-
10 cm and, relatively large trees > 10cm DBH1.3 cm. Those trees> 10 cm DBH1.3 further had
a subdivision between10-20, 20-30, 30-40, and 40-50 (see Table 5).

Table 5: Number of stems per ha for the sample areas in Mole National Park [Annette den Hollander H.J. (2008)]

DBH Class Vegetation Layer life-form Type # of Trees/ha in Savannah

< 2.5 cm Shrubs 16

2.5-10 cm Small trees 75
> 10 cm Trees 22

3. Criteria and Indicators for Ecological Zone Classification

Criteria and indicators that may be used to define the condition score for any ecological zone
type could be formulated alongside the work of Hawthorne and Abu-Juam (1995). However,
in an effort to reduce the level of subjectivity of any scoring description, measurable indicators
that can be verified with relative ease and also show consistency in application must be aimed
at.

Firstly, stratifying the ecological zones into forest, savannah and forest-savannah interface
(otherwise known as transitional zone) will be a fair approximation for the main eco-zones of
Ghana. More especially as an attempt for assessing forest landscape restoration potential in Ghana to
contribute to REDD-Plus strategies for climate change mitigation, poverty alleviation and sustainable forest
management.

Secondly, with the availability of imaging systems for satellite image acquisition such as
Landsat with 30 meter resolution or better, corroborated with ground truthing and field
observations, requirements for vegetation condition scoring can be reasonably achieved.
Criteria worth considering include but not limited to the following:
• Canopy cover
• Human disturbance including logging damage and settlements/farming
• Tree stems number estimates (stems/unit area)
• Basal area
• Natural regeneration (seedlings and sapling count /unit area)
• Wildfire damage

56
3.1 Basis for Scoring Description
Adopting a country-wide scoring description that encompasses a mosaic landscape of different
ecological zones requires careful consideration of the scoring criteria to be adopted. The
following measures were adopted:

• Scoring criteria used shall be applicable to all the key ecological zones identified

• For each ecological zone, a scoring description on a gradation scale from 1-4 shall be
uniformly applied to all the criteria listed above. This shall be done to reflect the
inherent ecological attribute(s) of that particular zone (see Appendix IV).

• Score 1 shall be the most ideal condition score; score 2 shall be deemed ecologically
tolerable score whereas score 3 shall be described as degraded and 4 shall be an
undesirable score.

• In determining the vegetation condition score for a particular eco-zone the average
score for all the individual criteria scores (i.e. canopy cover, human disturbance, tree
stem number estimates, basal area, natural regeneration and fire damage) shall be used
(see Appendix IV).

• In general the lesser the gradation scoring the more likely the element of subjectivity
will be minimized. However, the precaution here is that the criteria for ecological zone
description should be distinct and measurable with less ambiguity.

• Forest condition scoring adopted by Hawthorne and Abu-Juam (1995) remains an

important reference point. However, whereas Hawthorne and Abu-Juam (1995)
applied the scores based on observation, conversation with forest staff working in the
area, the inventory results and satellite imagery in decreasing order of importance, this
current approach gives Landsat-type satellite imagery (30 meter resolution or better)
prominence followed by ground truthing and field observations.

57
4. Forest and Woodland Cover Mapping

The satellite image-based mapping was done at the national large scale for all reserves and
protected areas in the forest and savannah ecological zones. Landsat TM, ETM+ and ASTER
satellite image scenes spanning 2003 – 2010 were used for the analysis. Image classification was
done to determine the nature and extent of forest cover loss within the reserves. The spatial
framework adopted for the mapping was based on the two broad vegetation types namely,
forest and savannah. This is because these vegetation zones serve as a spatial framework for
environmental resource management in Ghana. They denote areas that contain a
geographically distinct assemblage of environmental conditions, natural communities, and
plant species.

The image processing focussed principally on processing Landsat Thematic Mapper (TM)
Enhanced Thematic Mapper Plus (ETM+) satellite images, and in certain cases where good
Landsat scenes were not available, ASTER Images. Digital topographic data layers including
settlement locations, road and stream network were used as ancillary information sources.

The methodology involved the following:

1. Image acquisition and pre-processing
2. Image processing classification and field validation
3. Map composition and presentation

4.1 Image acquisition and pre-processing

Landsat data between 1999 and 2010 images were acquired and examined to judge any missing
scenes or data gap including cloud cover. Scenes selected for the mapping were almost cloud
free except some patches in some scenes. The vegetation bands (Bands 3 and 4 of Landsat)
were further examined and used for the classification after the necessary geometric and
radiometric corrections.

Band 3 (Red) is a useful band for discriminating among different kinds of vegetation, and band
4 (Near infrared) is especially responsive to the amount of vegetation biomass present in a
scene. It is also useful for distinguishing between vegetation and soil, and for seeing the
boundaries of bodies of water.

All images have been georeferenced to UTM Zone 30, WGS 84.

58
 Figure 4: Vegetation band (Band 4) of Landsat scenes covering Ghana

4.2 Image classification

Satellite image analysis can be carried out in two ways: visual or automatic. Visual
interpretation is purely subjective interpretation by a human interpreter. Automatic methods
are generally preferable where possible because the interpretation is repeatable and efficient,
furthermore large areas can be classified much faster than with visual analysis. Automatic
classification can be further divided into two categories: supervised and unsupervised
classification. In unsupervised classification, the classification software is allowed to sort pixels
according to their reflectance values. Unsupervised classification is good in the classification of
pixels within a target area without human interference.

59
An unsupervised classification with an ISODATA (Iterative Self-organizing Data Analysis)
algorithm was performed on image scenes to group pixels with similar spectral response into
unique clusters of different classes. The homogeneous clusters corresponding to specific
forest cover types were assigned appropriate cover classes based on reference data from color
composite images, high resolution images where available and expert ground knowledge.
Colour composite images helped identify the different forest cover types, thus clusters of
homogeneous reflectance were merged together.

The spectral variability during the classification procedure produced ‘salt-and-pepper’

appearance of the classified image as a result of the pixel by pixel classification. The classified
image was therefore resampled with a 3 x 3 majority smoothing filter.

Four cover classes, as described in Tables 6 and 7 were assigned to the satellite image
classification as closed canopy, intermediate canopy, open canopy, degraded area.

Forest map unit refers to continuous, multi-storied stand of trees at least 5m high with
interlocking crowns, usually lacking a grass ground cover. Forest is sub-categorized at into
“closed canopy” where the canopy cover exceeds 50% (>50% canopy cover), and “open
canopy” where the canopy cover is less than 50% (<50% canopy cover).

Savannah has been mapped as an area of land comprising a mixture of woodland (a single
storied of trees less that 5m high), and/or bushes (woody plant with multiple stems, usually
over 2m tall), and/or shrubs (woody plant with multiple stems, usually less than 2m tall),
and/or grassland with or without scattered cultivation. Savannah is sub-divided into closed
woodland, open woodland, and grassland.

Table 6: Forest Cover Classes

Closed Canopy Land with a minimum tree crown cover (or equivalent
stocking level) of 50% and a minimum area of 1.0 ha

Open canopy Land with tree crown cover (or equivalent stocking level)
between 50%<15% and a minimum area of 1.0 ha

Deforested Area Land with tree crown cover (or equivalent stocking level)
of less than 15% and a minimum area of 1.0 ha

60
Table 7: Savannah Cover Classes
Closed Woodland Land with tree crown cover of more than 30% and a
minimum area of 1.0 ha

Open Woodland Land with tree crown cover of between 30<10% and a
minimum area of 1.0 ha

Grassland Land with predominantly covered by grass with scattered

trees with crown cover below 10% and a minimum area of
1.0 ha

4.3 Field Validation and Accuracy Assessment

The mapping was validated in the field, in order to assess mapping accuracy. Reserves were
randomly sampled for each vegetation zone for field validation. Sample sites easily assessable
were selected within each of the sampled Reserves for ground-truthing. The validation
concentrated on areas where the accuracy was known to be lower. These areas were edited to
reflect the ground-truth, so as to improve mapping accuracy.

4.4 Map composition and presentation

Maps have been produced for each Reserve and statistics for cover classes have been reported.

4.5 Software used

ERDAS IMAGINE 2011 was used for image processing and ArcGIS 9.3, for the map
preparation. All maps are in Appendix VI.

61
Appendix IV. Criteria for scoring condition and restoration
opportunities

A set of criteria has been developed for scoring of the condition of the vegetation in
Ghana and associated restoration opportunities. These criteria are shown below in four
tables:

1. Criteria and associated scales for condition scoring of high forest, transition zone,
and savannah (Table 1)

2. Comprehensive scale for scoring the condition and restoration opportunities in

key ecological zones
a. the forest zone (Table 2)
b. the transition zone (Table 3)
c. the savannah zone (Table 4)

Forest is defined according to FAO (2000) as:

• “Land with tree crown cover (or equivalent stocking level) of more than 10
percent and an area of more than 0.5 hectares (ha). Ghana recently defined its
forest as land with tree crown cover of more than 15 percent and a minimum
area of 1.0 hectare (ha).

• The trees should be able to reach a minimum height of 5 meters (m) at maturity
in situ.

• May consist either of closed forest formations where trees of various storey and
undergrowth cover a high proportion of the ground or open forest formations
with a continuous vegetation cover in which tree crown cover exceeds 10
percent.

• Young natural stands and all plantations established for forestry purposes which
have yet to reach a crown density of 10 percent or tree height of 5 m are included
under forest,

• are areas normally forming part of the forest area which are temporarily
unstocked as a result of human intervention or natural causes but which are
expected to revert to forest.”

62
The classification of savannahs based on crown cover is adopted from Annette den
Hollander H.J. (2008) where crown cover is estimated from the crown of individual trees
and the gap between the crowns of trees. Thus the crown cover is calculated from the
crown separation ratio which is different from that used for forests.

63
Table 1: Criteria and associated scales for condition scoring in key ecological zones in Ghana

CRITERION SCALE FOR CONDITION SCORING VERIFICATION

Forest Transition Savannah
Canopy cover Land with a minimum tree crown Land with a minimum tree Land with tree crown cover Satellite imagery +
cover (or equivalent stocking crown cover (or equivalent of more than 30% and a ground truth data/
level) of 50% and a minimum area stocking level) of 50% and a minimum area of 1.0 ha = 1 survey
of 1.0 ha = 1 minimum area of 1.0 ha = 1 Land with tree crown cover [protocols/methods
Land with tree crown cover (or Land with tree crown cover (or of between 20-30% and a to be defined
equivalent stocking level) between equivalent stocking level) minimum area of 1.0 ha = 2 including sampling
30-50% and a minimum area of between 30-50% and a Land with tree crown cover design]
1.0 ha = 2 minimum area of 1.0 ha = 2 between 10-20% and a
Land with tree crown cover (or Land with tree crown cover (or minimum area of 1.0 ha= 3
equivalent stocking level) between equivalent stocking level) Land with tree crown cover
15-30% and a minimum area of between 15-30% and a below 10% and a minimum
1.0 ha= 3 minimum area of 1.0 ha= 3 area of 1.0 ha= 4
Land with tree crown cover (or Land with tree crown cover (or
equivalent stocking level) of less equivalent stocking level) of less
than 15% and a minimum area of than 15% and a minimum area
1.0 ha = 4 of 1.0 ha = 4
Human Very few signs (<2%) of human Minimal signs1 (<5%) of human Minimal signs (<10%) of Satellite imagery +
disturbance disturbance attributable to disturbance attributable to human disturbance field observations
(including logging/farms; virgin or late logging/farms; late secondary attributable to agricultural
logging damage, secondary forest visible forest visible throughout = 1 practices including grazing;

1 Minimal signs of disturbance at that percentage range are conservative average figures judged from observations made in sacred groves dotted throughout the
transitional zone. The sacred groves (remnant forest) serve as proxies of near original state of the transitional zone forests

64
 CRITERION SCALE FOR CONDITION SCORING VERIFICATION
Forest Transition Savannah
farms, etc) throughout = 1 Between 5-15% of forest late-stage savannah
Between 2-10% of forest mosaic mosaic heavily disturbed. woodland vegetation2 visible
heavily disturbed. Logging Logging damage is evident = 2 throughout = 1
damage quite visible and Between 15-50% of forest Between 10-25% of
extensive= 2 mosaic heavily disturbed. savannah woodland is heavily
Between 10-50% of forest mosaic Logging damage is evident = 3 disturbed especially through
heavily disturbed. Logging More than 50% of forest mosaic fire and over-grazing, and
damage quite visible and visibly disturbed and patchy; approaching an “open
extensive= 3 logging damage is extensive; savannah”. = 2
More than 50% of forest mosaic forest is heavily burnt with Between 25-50% of
visibly disturbed and patchy; conspicuous Eupatorium and savannah woodland is heavily
logging damage is extensive and other pioneers throughout = 4 disturbed especially through
intensive; forest is heavily burnt fire and over-grazing, and
with conspicuous Eupatorium and approaching “grassland”. = 3
other pioneers throughout = 4 More than 50% of savannah
vegetation is visibly disturbed
and patchy; damage due to
fire and over-grazing is
extensive; Ground cover
predominantly grasses or
other herbaceous plants; tree
cover is almost absent= 4
Tree stem Average stem frequencies3 (>450 Average stem frequencies(>250 Average stem frequencies4 (> Repeated ground
number stems per ha) estimated for all stems per ha) estimated for all 200 stems per ha) estimated truth data/ survey

2 Late stage savannah wood land is considered here as climax vegetation which is more stable than other forms of derived savannah

65
 CRITERION SCALE FOR CONDITION SCORING VERIFICATION
Forest Transition Savannah
estimates trees down to 10 cm dbh in the trees down to 10 cm dbh in the for all trees and shrubs down required
different forest types = 1 Dry Semi-deciduous forest to 2.5 cm dbh in the
Average stem type= 1 savannah vegetation types =
frequencies(between 300-450 Average stem 1
stems per ha) estimated for all frequencies(between 150-250 Average stem frequencies
trees down to 10 cm dbh in the stems per ha) estimated for all (between 150-200 stems per
different forest types = 2 trees down to 10 cm dbh in the ha) estimated for all trees and
Average stem Dry Semi-deciduous = 2 shrubs down to 2.5 cm dbh
frequencies(between 200-300 Average stem in the savannah vegetation
stems per ha) estimated for all frequencies(between 100-150 types = 2
trees down to 10 cm dbh in the stems per ha) estimated for all Average stem frequencies
different forest types = 3 trees down to 10 cm dbh in the (between 100-150 stems per
Average stem frequencies below Dry Semi-deciduous = 3 ha) estimated for all trees and
200 stems per ha estimated for all Average stem frequencies below shrubs down to 2.5 cm dbh
trees down to 10 cm dbh in the 100 stems per ha estimated for in the savannah vegetation
different forest types = 4 all trees down to 10 cm dbh in types = 3
the Dry Semi-deciduous = 4 Average stem frequencies
(<100 stems per ha)
estimated for all trees and
shrubs down to 2.5 cm dbh
in the savannah vegetation
types =4

3
Data estimated from Forest Inventory Project of 1986-1990; Multi-resource inventory program conducted in 2001 and more recently RMSC updated inventory
data of August, 2009. Stems per ha values (down to10 cm dbh) reflect average values ranging from the highest in Wet Evergreen to the lowest in Dry Semi-
deciduous forest type in Ghana
4 Adopted from Annette den Hollander H.J. (2008) see Table 4

66
 CRITERION SCALE FOR CONDITION SCORING VERIFICATION
Forest Transition Savannah
Basal area Average basal area5 (>24 m2/ ha) Average basal area (>18 m2/ ha) Data not available Repeated ground
(m2/ha) estimated for all trees down to 10 estimated for all trees down to truth data/ survey
cm dbh in the different forest 10 cm dbh in the Dry Semi- required
types = 1 deciduous forest type= 1
Average basal area (between 15- Average basal area (between 10-
24 m2/ ha) estimated for all trees 18 m2/ ha) estimated for all
down to 10 cm dbh in the trees down to 10 cm dbh in the
different forest types = 2 Dry Semi-deciduous = 2
Average basal area (between 10- Average basal area (between 5-
15 m2/ ha) estimated for all trees 10 m2/ ha) estimated for all
down to 10 cm dbh in the trees down to 10 cm dbh in the
different forest types = 3 Dry Semi-deciduous = 3
Average basal area below10 m2/ Average basal area below 5 m2/
ha estimated for all trees down to ha estimated for all trees down
10 cm dbh in the different forest to 10 cm dbh in the Dry Semi-
types = 4 deciduous = 4
Natural Natural regeneration evident Natural regeneration evident Natural regeneration7 evident Repeated ground
regeneration through average counts6 greater through average counts greater through average counts truth data/ survey
than 350/ha of seedlings/ than 300/ha of seedlings/ greater than 150/ha of required
saplings (10-29.9 cm dbh) in the saplings (10-29.9 cm dbh) in the shrubs (<2.5 cm dbh) and
forest floor =1 forest floor =1 small trees (2.5-10.0 cm dbh)

5
Data estimated from Forest Inventory Project of 1986-1990; Multi-resource inventory program conducted in 2001 and more recently RMSC updated inventory
data of August, 2009. Basal area per ha values (down to10 cm dbh) reflect average values ranging from the highest in Wet Evergreen to the lowest in Dry Semi-
deciduous forest type in Ghana
6 Data estimated from Forest Inventory Project of 1986-1990; Multi-resource inventory program conducted in 2001 and more recently RMSC updated inventory

data of August, 2009. Stems per ha values (10-29.9 cm dbh) reflect average ranges estimated for the various forest types in Ghana
7 See Table 4 for similar inferences

67
CRITERION SCALE FOR CONDITION SCORING VERIFICATION
Forest Transition Savannah
Natural regeneration evident Natural regeneration evident in the lower vegetation layer
through average counts between through average counts between =1
200-350/ha of seedlings/ saplings 200-300/ha of seedlings/ Natural regeneration evident
(10-29.9 cm dbh) in the forest saplings (10-29.9 cm dbh) in thethrough average counts
floor =2 forest floor =2 between 100-150/ha of
Natural regeneration evident Natural regeneration evident shrubs (<2.5 cm dbh) and
through average counts between through average counts between small trees (2.5-10.0 cm dbh)
100-200/ha of seedlings/ saplings 100-200/ha of seedlings/ in the lower vegetation layer
(10-29.9 cm dbh) in the forest saplings (10-29.9 cm dbh) in the=2
floor =3 forest floor =3 Natural regeneration evident
Poor or no forest regeneration; Poor or no forest regeneration; through average counts
counts less than100 seedlings/ counts less than100 seedlings/ between 50-100/ha of shrubs
saplings (10-29.9 cm dbh) in the saplings (10-29.9 cm dbh) in the(<2.5 cm dbh) and small
forest floor = 4 forest floor = 4 trees (2.5-10.0 cm dbh) in the
lower vegetation layer =3
Natural regeneration evident
through average counts less
than 50/ha of shrubs (<2.5
cm dbh) and small trees (2.5-
10.0 cm dbh) in the lower
vegetation layer =4
Fire damage No signs of fire damage = 1 Fire damage is visible (< 20% of Evidence of annual fires in Satellite imagery +
Fire damage is visible (< 10% of forest mosaic) but not extensive recent period (past decade) ground truth data/
forest mosaic) but not extensive = = 1 but damage not extensive survey and past fire
2 Between 20-30% of forest probably due to better fire history record if
Between 10-30% of forest mosaic mosaic with scars of fire management control available
with serious scars of fire damage damage = 2 regimes= 1
=3 Between 30-50% of forest Evidence of annual fires with

68
CRITERION SCALE FOR CONDITION SCORING VERIFICATION
Forest Transition Savannah
More than 30% of forest mosaic mosaic with serious scars of fire a medium term history
with serious scars of fire damage damage = 3 (within the last 50 years)
=4 More than 50% of forest mosaic exacerbated by
with serious scars of fire anthropogenic factors
damage; visible evidence of including improper fire
derived savannah due to management regime = 2
repeated fires = 4 Evidence of annual fires with
a medium to long-term
history (within the last 50-75
years) exacerbated by
anthropogenic factors
including improper fire
management regime = 3
Evidence of annual fires with
a relatively longer history
(about 100 years or so)
exacerbated by
anthropogenic factors
including improper fire
management regime = 4

69
Table 2: Comprehensive scale for scoring the condition and restoration opportunities in the forest zone

Avg. Definition Indicative Suitability for Potential Area Remarks

Condition Vegetation Area Restoration (km2) Available
Score Restoration
1 Land with tree crown cover (or equivalent Example is Not suitable for Very few isolated Wet Evergreen
stocking level) of more than 50% and a Ankasa Forest restoration pockets of slight (WE) forest type.
minimum area of 1 ha; very few signs (<2%) Reserve in the except few human Remnants of a
of human disturbance; excellent regeneration Western Region pockets of areas disturbance that typical rain forest ;
recording over 350 seedlings/saplings per ha; requiring should be very few of such
average stem frequencies for all diameter enhanced natural allowed to forest type is left in
classes well above 450 stems/ha (very good regeneration or undergo natural Ghana
stock density); basal area > 24 m2/ha; no enrichment regeneration.
evidence of fire damage detected or recorded planting Approx. < 10%
of WE forest
type (600 km2)

2 Land with tree crown cover (or equivalent Examples of Some areas may About 30% Commonly occurs
stocking level) of between 30-50% and a forest reserves be suitable for (5,300 km2) of in parts of the forest
minimum area of 1 ha; between 2-10% of with similar restoration but the total forest area classified as
forest mosaic heavily disturbed through condition are requires careful area classified as Moist Evergreen
logging and other human disturbance; good Bonkoni, Bonsa intervention Moist Evergreen
regeneration (200-350 seedlings/saplings per Ben, Bura River, strategies that
ha; average stem frequencies for all diameter Tano Anwia, Tano make use of
classes in the range of 300-450 stems/ha Nimiri, etc and native tree
(good stocking density); basal area between 15 their adjoining species originally
- 24 m2/ha; fire damage is visible but not off-reserve areas present in the
extensive over the defined forest area within 20 km area and other
radius ecological
considerations

70
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
3 Land with tree crown cover (or equivalent Examples of Suitable for About 50% Commonly occurs
stocking level); of between 15-30% and a forest reserves restoration (27,000 km2) of in parts of the Moist
minimum area of 1 ha; between 10-50% of with similar through the total forest Semi-Deciduous
forest mosaic heavily disturbed; logging condition are enrichment area classified as and Dry Semi-
damage is visible and extensive; natural Desiri, Dome planting and Moist Semi- Deciduous forest
regeneration (100-200 seedlings/saplings per River, South plantation Deciduous and types
ha; average stem frequencies for all diameter Formangsu, programs. Dry Semi-
classes between100-200 stems/ha (low Krochua, Requires careful Deciduous forest
stocking density); basal area between 10 - 15 Mpameso, Tain intervention types potentially
m2/ha; fire damage is visible. Between 10- Tributaries II, strategies that: qualify for
30% of forest mosaic with serious scars of fire (i) make use of restoration
damage over a defined forest area. native tree
species originally
present in the
area
(ii) ensures
effective fire
prevention
measures
4 Land with tree crown cover (or equivalent Examples of Suitable for About 30% of Commonly occurs
stocking level); of less than15% and a forest reserves restoration Dry Semi- in parts of the Dry
minimum area of 1 ha; More than 50% of with similar through Deciduous (inner Semi-Deciduous
forest mosaic heavily disturbed and patchy; condition are particularly large- & fire zone type) forest type,
logging damage is extensive and intensive; North Bandai scale plantations. + 50% of especially the inner
forest is heavily burnt with conspicuous Hills, Pamu- Requires careful Southern zone type and the
Eupatorium and other pioneer species Berekum, Tain intervention marginal forest Southern marginal
throughout; very low regeneration (less than Tributaries I, strategies that: (7,580 km2) types forest types

71
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
100seedlings/saplings per ha; average stem Bodi, Kronwam, (i) make use of potentially
frequencies for all diameter classes below 100 native tree qualify for
stems/ha (very low stock density); basal area species originally restoration
below 10 m2/ha; fire damage is visible. More present in the
than 30% of forest mosaic with serious scars area
of fire damage over the defined forest area. (ii) ensures
effective fire
prevention
measures

72
Table 3: Comprehensive scale for scoring the condition and restoration opportunities in the transition zone

Avg. Definition Indicative Suitability for Potential Area Remarks

Condition Vegetation Area Restoration (km2) Available
Score Restoration
1 Land with tree crown cover (or equivalent Examples of Suitable for About 10% of Commonly occurs
stocking level) of more than 50% and a forest reserves restoration but Dry Semi- in parts of the Dry
minimum area of 1 ha; minimal signs (<5%) with similar requires careful Deciduous (inner Semi-Deciduous
of human disturbance; very good regeneration condition are intervention & fire zone type) forest type,
recording over 300 seedlings/saplings per ha; Opimbo 1, 2, strategies that: + 10% of especially the inner
average stem frequencies for all diameter Ongwam 1,2,3 (i) make use of Southern zone type and the
classes well above 250 stems/ha (good stock and identifiable native plant marginal forest Southern marginal
density); Average basal area (>18 m2/ ha) sacred groves species originally (2,400 km2) types forest types
estimated for all trees down to 10 cm dbh; fire within Dry Semi- present in the potentially
damage is visible (< 20% of forest mosaic) Deciduous (inner area qualify for
but not extensive & fire zone type) (ii) ensures restoration
effective fire
prevention
measures
2 Land with tree crown cover (or equivalent Examples of Suitable for About 20% of Commonly occurs
stocking level) between 30-50% and a forest reserves restoration but Dry Semi- in parts of the Dry
minimum area of 1 ha; minimal signs with similar requires careful Deciduous (inner Semi-Deciduous
(between 5-15%) of human disturbance; good condition are intervention & fire zone type) forest type,
regeneration recording between 200-300 Nsemere and strategies that: + 10% of especially the inner
seedlings/saplings per ha; average stem Bomfuom, Bandai (i) make use of Southern zone type and the
frequencies for all diameter classes between Hills (South) native plant marginal forest Southern marginal
150- 250 stems/ha (good stock density); species originally (4,500 km2) types forest types
Average basal area (between 10-18 m2/ ha) present in the potentially
estimated for all trees down to 10 cm dbh; area qualify for
Between 20-30% of forest mosaic with scars (ii) ensures restoration

73
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
of fire damage. effective fire
prevention
measures
3 Land with tree crown cover (or equivalent Examples of Suitable for About 10% of Commonly occurs
stocking level) of between 15-30% and a forest reserves restoration but Dry Semi- in parts of the Dry
minimum area of 1 ha; about 20% of human with similar requires careful Deciduous (inner Semi-Deciduous
disturbance; low-medium natural regeneration condition are intervention and fire zone forest type,
recording between 100-350 seedlings/saplings Kpandu Range strategies that: types) +30% especially the fire
per ha; average stem frequencies for all (Dayi block) (i) make use of Southern zone type and the
diameter in the range of 100-400 stems/ha; native plant marginal forest Southern marginal
more than 10% of forest mosaic with scars of species originally types (2,800 km2) forest types
fire damage present in the potentially
area qualify for
(ii) ensures restoration
effective fire
prevention
measures
4 Land with tree crown cover (or equivalent Examples of Suitable for About 30% of Commonly occurs
stocking level) of less than 15% and a forest reserves restoration but Dry Semi- in parts of the Dry
minimum area of 1 ha; more than 50% of with similar requires careful Deciduous (inner Semi-Deciduous
human disturbance in a defined transitional condition are intervention and fire zone forest type,
zone; very poor natural regeneration Kpandu strategies that: types) +30% especially the fire
recording less than 100 seedlings/saplings per RangeWest, and (i) make use of Southern zone type and the
ha; average stem frequencies for all diameter Pru Shelterbelt native plant marginal forest Southern marginal
classes well below 100 stems/ha; More than species originally types (8,000 km2) forest types
50% of forest mosaic in a defined area with present in the potentially
serious scars of fire damage; visible evidence area qualify for

74
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
of derived savannah due to repeated fires (ii) ensures restoration
effective fire
prevention
measures

75
Table 4: Comprehensive scale for scoring the condition and restoration opportunities in the savannah zone

Avg. Definition Indicative Suitability for Potential Area Remarks

Condition Vegetation Area Restoration (km2) Available
Score Restoration
1 Land with tree crown cover of more than Examples of Suitable for Further Commonly occurs
30% and a minimum area of 1.0 ha; minimal savannah restoration but validating studies in parts of the Dry
signs (<10%) of human disturbance vegetation with requires careful required to Semi-Deciduous
attributable to agricultural practices including similar condition intervention confirm present forest type,
grazing; late-stage savannah woodland occurs in parts of strategies that: conditions on the especially the fire
vegetation visible throughout; average stem Bomfobiri (i) make use of ground zone type and the
frequencies (> 200 stems per ha) estimated Wildlife Sanctuary native tree Southern marginal
for all trees and shrubs down to 2.5 cm dbh in and Shai Hills species originally forest types
the savannah vegetation types; natural Game Production present in the
regeneration evident through average counts Reserve area
greater than 150/ha of shrubs (<2.5 cm dbh) (ii) considers
and small trees (2.5-10.0 cm dbh) in the lower wildlife habitat
vegetation layer; evidence of annual fires in (iii) ensures
recent period (past decade) but damage not effective fire
extensive probably due to better fire prevention
management control regimes measures

2 Land with tree crown cover of between 20- Examples of Suitable for Further Commonly occurs
30% and a minimum area of 1.0 ha; between savannah restoration but validating studies in parts of the Dry
10-25% of savannah woodland is heavily vegetation with requires careful required to Semi-Deciduous
disturbed especially through fire and over- similar condition intervention confirm present forest type,
grazing, and approaching an “open could be found in strategies that: conditions on the especially the fire
savannah”; average stem frequencies (between parts of Mole (i) make use of ground zone type and the
150-200 stems per ha) estimated for all trees National Park in native tree Southern marginal
and shrubs down to 2.5 cm dbh in the the northern species originally forest types

76
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
savannah vegetation types; natural savannah zone present in the
regeneration evident through average counts area
between 100-150/ha of shrubs (<2.5 cm dbh) (ii) considers
and small trees (2.5-10.0 cm dbh) in the lower wildlife habitat
vegetation layer; evidence of annual fires with (iii) ensures
a medium term history (within the last 50 effective fire
years) exacerbated by anthropogenic factors prevention
including improper fire management regime measures
3 Land with tree crown cover between 10-20% Examples of Suitable for Further Commonly occurs
and a minimum area of 1.0 ha; between 25- savannah restoration but validating studies in parts of the
50% of savannah woodland is heavily vegetation with requires careful required to Northern Region of
disturbed especially through fire and over- similar condition intervention confirm present Ghana
grazing, and approaching “grassland”; average are found in strategies that: conditions on the
stem frequencies (between 100-150 stems per Odonkokrom on (i) make use of ground
ha) estimated for all trees and shrubs down to the Afram Plains; native tree
2.5 cm dbh in the savannah vegetation types; Yilo Krobo area species originally
natural regeneration evident through average between Huhunya present in the
counts between 50-100/ha of shrubs (<2.5 and Asesewa area
cm dbh) and small trees (2.5-10.0 cm dbh) in where intensive (ii) considers
the lower vegetation layer; evidence of annual farming have wildlife habitat
fires with a medium to long-term history replaced the forest (iii) ensures
(within the last 50-75 years) exacerbated by by a kind of effective fire
anthropogenic factors including improper fire secondary prevention
management regime grassland measures
4 Land with tree crown cover below 10% and a Examples of Suitable for Further Commonly occurs
minimum area of 1.0 ha; more than 50% of savannah restoration but validating studies in parts of the
savannah vegetation is visibly disturbed and vegetation with requires careful required to Northern and

77
Avg. Definition Indicative Suitability for Potential Area Remarks
Condition Vegetation Area Restoration (km2) Available
Score Restoration
patchy; damage due to fire and over-grazing is similar condition intervention confirm present Upper Regions of
extensive; Ground cover predominantly may be typical of strategies that: conditions on the Ghana
grasses or other herbaceous plants; tree cover the Upper East (i) make use of ground
is almost absent; average stem frequencies and West Regions native tree
(<100 stems per ha) estimated for all trees of Ghana where species originally
and shrubs down to 2.5 cm dbh in the the Sudan present in the
savannah vegetation types; natural savannah area
regeneration evident through average counts predominates. (ii) considers
less than 50/ha of shrubs (<2.5 cm dbh) and wildlife habitat
small trees (2.5-10.0 cm dbh) in the lower (iii) ensures
vegetation layer; evidence of annual fires with effective fire
a relatively longer history (about 100 years or prevention
so) exacerbated by anthropogenic factors measures
including improper fire management regime

78
Appendix V. Forest and Woodland Cover Classification Maps

79