GCP/RAS/162/JPN The status of the forest resources assessment in the south-asian sub-region and the country capacity building needs   Proceedings of the GCP/RAS/162/JPN Regional Workshop  held in Dehradun, 8-12 June 1998

Table of Contents

Abbreviations

1. Introduction

1.1Purpose

1.2 Resource Organization

1.3 Regional Representation

1.4The Proceeding

2. Overview of workshop activities

2.1 Inaugural session

2.3 Working sessions

2.3 Concluding session

3. Thematic deliberations

3.1 Area of Forest and Other Wooded Land

3.2 Protection Status

3.3 Forest Ownership

3.4 Forest By Ecological Zone

3.5 Wood Supply Potential

3.6 Changes Over Time

3.7 Volume and Biomass

3.8 Felling and Removal

3.9 Forest Fires

3.10 Non Wood Goods and Services

3.11 Additional Parameters

3.12 Other Items discussed at the workshop

References

Appendix 1: Resource persons and participants

Appendix 2: Daywise session profiles of the workshop

Appendix 3: An outline for country report on FRA 2000

Appendix 4: Information for electronic networking

APPENDIX 5: Technical papers

Section 1: Deforestation and Forest Accessibility

SECTION 2: The Area Production Model (APM)

SECTION 3: Forest Resources Assessment

SECTION 4: Workshop on country capacity building in Forest Resources Assessment in South Asian Countries

SECTION 5: CCB - FRA 2000 - The Biodiversity Assessment

SECTION 6: Forest Cover Change - Assessment Methodology

SECTION 7: Estimation of forest cover for standard reference years using a modelling approach

SECTION 8: Forest Fire - Problems related to the assessment of forest area -The country perspective

SECTION 9: Tool for assessment of forest biological diversity

SECTION 10: Problems Related to the Assessment of Forest Area: The Country Perspective - GRAZING

SECTION 11: Global Forest Resource Assessment 2000 - a user's perspective

SECTION 12: Forest Naturalness

SECTION 13: Non Wood Forest Products

SECTION 14: Forest Ownership

SECTION 15: Assessment of Forest Plantation Areas

SECTION 16: Planning A Protected Area Network

SECTION 17: Shifting Cultivation in Nagaland : A Pilot -Study

SECTION 18: Purpose Minimum Protected Area Network for Low-Country wet zone of Sri Lanka in the Districts of Matara, Galle, Kalutara and Ratnapura.

The Forest Resources Assessment Programme

Forests are crucial for the well-being of humanity. They provide foundations for life on earth through ecological functions, by regulating the climate and water resources, and by serving as habitats for plants and animals. Forests also furnish a wide range of essential goods such as wood, food, fodder and medicines, in addition to opportunities for recreation, spiritual renewal and other services.

Today, forests are under pressure from expanding human populations, which frequently leads to the conversion or degradation of forests into unsustainable forms of land use. When forests are lost or severely degraded, their capacity to function as regulators of the environment is also lost, increasing flood and erosion hazards, reducing soil fertility, and contributing to the loss of plant and animal life. As a result, the sustainable provision of goods and services from forests is jeopardized.

FAO, at the request of the member nations and the world community, regularly monitors the world's forests through the Forest Resources Assessment Programme. The next report, the Global Forest Resources Assessment 2000 (FRA 2000), will review the forest situation by the end of the millennium. FRA 2000 will include country-level information based on existing forest inventory data, regional investigations of land-cover change processes, and a number of global studies focusing on the interaction between people and forests. The FRA 2000 report will be made public and distributed on the world wide web in the year 2000.

The Forest Resources Assessment Programme is organized under the Forest Resources Division (FOR) at FAO headquarters in Rome. Contact persons are:

Robert Davis FRA Programme Coordinator [email protected]

Peter Holmgren FRA Project Director [email protected]

or use the e-mail address: [email protected]

Paper drafted by Kailash Govil & edited by Patrizia Pugliese

Abbreviations

1. Introduction

1.1Purpose

FAO has established a regional project office (GCP/RAS/162/JPN) at Dehradun, UP, India in April 1998, to build institutional capacities of the South Asian countries in the field of Forest Resource Assessment and Planning. The first regional activity under this project was a workshop "The status of Forest Resource Assessment in the South Asian Sub-region and Country Capacity Building needs", June 8 to 12, 1998 at Forest Survey of India, Dehradun, UP, India.

This document provides not only the proceeding of the workshop but also the copies of the material exchanged during the deliberations. This document provides the core information of FRA, 1990, about six South Asian countries and Myanmar and recapitulates the relevant definitions of the core items to present complete picture and to serve as a reference document. The document also presents any new information relating to FRA 2000 provided by the participants.

1.2 Resource Organization

FAO, Forest Survey of India, Indira Gandhi National Forest Academy, Indian Institute of Remote Sensing and French Institute of Pondicherry provided the technical resource persons and material.

Country Capacity Building "CCB" project (GCP/RAS/162/JPN) provided total financial support for the workshop. The project provided money to respective FAOR to meet travel and other expenses of the participants. The project also provided $9,000 to FSI-Dehradun to cover the local transportation, lodging and boarding, resource material, and remuneration to resource persons.

1.3 Regional Representation

Six countries (Bangladesh, Bhutan, India, Nepal, Sri Lanka and Myanmar) participated in this regional workshop. Each country sent two delegates to participate in this workshop. Delegates from Pakistan could not attend it due to some political constraints. The participants from Myanmar desired that, in future, Myanmar should be included in this group of South Asian countries for deliberations on forestry issues. This group of twelve participants included foresters and ministry officials dealing with forestry subjects in their countries.

1.4The Proceeding

The proceeding is organized in three sections. First section has the main report. The second section consists of annexes. The third section reproduces the technical papers or transparencies presented and provided by the participants during the workshop.

This main section of the proceeding is divided in two parts. The first part provides an overview and the second part concentrates on the thematic sessions at the workshop.

2. Overview of workshop activities

The workshop was organized in eleven sessions comprising one each of inaugural, foundation, and concluding session and eight working sessions on core items of FRA.

The participants deliberated on technical issues in the eight working sessions over the period of 5 days. Each session focussed on selected items of the "core information." The workshop was participative and consultative in nature and provided excellent opportunities to develop the capacities and enhance co-operation in field of forest resource assessment amongst the countries of the region.

2.1 Inaugural session

Dr. B. N. Gupta, Director General, ICFRE, the Chief Guest, formally inaugurated the five day workshop. While welcoming the delegates in his keynote address, he explained the unique socio-economic importance of forest resources at micro levels in the South Asian Region. He stressed on the need for capacity building of institutions dealing with forest resources and planning. He specially emphasized the need for enhancing the capacities of local communities for increasing their efficiency in participation for planning, protection and management of local forest resources.

Mr. A. Marzoli, FAO welcomed the participants and deliberated on the FAO's thinking on GFRA 2000 and the need for enhancing the capacities of the countries in order to decentralize collection, storage, retrieval and utilization of information about forest resources. He outlined the objectives of the workshop; defining strategies for CCB, establishing mechanism for sharing information about FRA 2000 and providing scenarios of status of forest resources in respective countries.

Mr. P. C. Tyagi, Joint Director-FSI and the course director informed about the agenda, schedule and organization of the workshop. Dr. D. Pandey, Director FSI introduced the audience to the activities of FSI, genesis of the workshop and the need for supplying quality information for the GFRA 2000. Dr. K. C. Govil, Regional Project Coordinator-FAO, explained the purpose of the workshop and made a wish for very productive deliberations. Mrs. Ranjana Gupta, FSI, proposed the vote of thanks for the inaugural session. 2.2 Foundation session

Revelation of the expectation from the workshop organized by the FAO and the views of the participants, coupled with two baseline presentations by FAO staff, laid a well-structured foundation for the workshop.

2.2.1 Expectations

The FAO representatives expected that during the workshop, the data needs of FRA 2000 will become more clear to the participants and that FAO will have a better appreciation of the problems faced by different countries with respect to the collection, compilation and sharing of information for FRA 2000. They also expected that the workshop would help in developing strategic activities under CCB to address the aforesaid problems through training, studies, workshop and equipment.

The participants from Bangladesh, Bhutan and Nepal expected that after the workshop they will have a better understanding of FRA 2000 needs, objectives, methodology and management and will be in a better position to contribute to the development of FRA 2000.

The Indian representatives expected the emergence of a clear perspective about the data requirements for planning and assessment at the ground level and the national level and the standardization in the format of collection and storage of the data.

The Sri Lankan representatives expressed their desire to understand their role and to learn more about the experience of different countries in developing proper strategies to support this project.

The Myanmar delegates expressed their hope that this workshop will improve exchange of information, technology, and support from FAO and other institutions like ICIMOD.

2.2.2 Forest assessment activities

Following sub-section provides a brief description on the forest assessment activities presented by the participants of the six countries.

Bangladesh

In 1980, the forest department with UNDP/FAO assistance conducted a detailed inventory of forest areas including a countrywide inventory of village forests. Recently (1995-97), the forest department completed a massive inventory program in hill and coastal forest areas under a loan agreement with International Development Agency. This is a unique inventory in the sense that a socio-economic survey has also been conducted along with the forest inventory to understand the behavioral pattern of the users. The forest inventory design is based on systematic sampling with grid intervals of 40x40 feet in combination with smaller grids 20 or 10 feet, to cover the required number of cluster plots. Forest statistics are generated with continuous resource change assessment. A Resource Information Management System (RIMS) is in place which uses SPOT data for generating signatures of different kinds of forest vegetation.

Bhutan

The PISFR-India designed and conducted first inventory for Bhutan in 1976-1981. Bhutan has 72% forest cover (with 8% scrub forest) amounting to 2.9 million hectares. About 39% of total "Reserve Forest" is considered operable. Forests are managed through 39 Forest management units. Social forestry and Agro forestry divisions raise and monitor the plantations. Currently about 27% of land is under "Protected Forest area" category.

India

Forest Survey of India is doing assessment utilizing satellite imageries, aerial photographs and ground inventories. Vegetation maps on the scale of 1:250000 are prepared for the whole country on a 2-year cycle. Thematic maps using aerial photographs on 1:50000 scale are being produced on a 10-year cycle. Approximately, 70% of India has been covered on thematic maps. Similarly ground inventory has covered about 80% of the area. State of Forest Report is published by FSI on a two-year cycle. Forest cover assessment is done partially (28%) in digital mode and rest (72%) by visual interpretation. FSI has adopted a mandate to digitally assess the vegetation cover of the whole of India in the near future.

Myanmar

A FAO/UNDP project has helped Myanmar to assess forest resources through use of satellite data from LANDSAT in 1970. The National Forest Survey and Inventory Project (1981-91), funded by UNDP/FAO, has provided reliable information on forest resources and aerial photographs for about 90% of the area. During this period, information has been collected at three levels; pre-investment, reconnaissance and management covering 12 million ha of forest land. Since 1991, Myanmar has been conducting forest inventory every year, covering 2 million ha using remote sensing and GIS technology. The available information is used to generate maps of 1:25000 and 1:50000 scale. Forest cover assessment of 1980 used Landsat imagery (scale 1:1000000). The 1989 forest cover estimation utilized Landsat TM data (scale 1:500000). Interpretation of the collected information is done visually.

Nepal

Nepal has established Forest resource Information System Program (FRISP) and is currently carrying out hierarchical grid inventory at the regional and the district level. Inventory data on forest type, crown diameter type and density class at both the levels is being collected. Recent maps based on aerial photographs were produced in 1996. Nepal faces an uphill task in forest resource assessment and inventory due to difficult terrain, non- availability and high cost of aerial photographs, problems in adopting new technologies, and untrained manpower.

Sri Lanka

First National level Inventory was carried out in 1956 with the Swiss and the Canadian help. Next inventory was done in 1983 with the help of a FAO/UNDP project. The total forest cover with closed canopy is 23.9%, sparse forest is 7% and forest plantation is 2%. Their resource base consists of natural forests, forest plantations, and non-forest tree resources from home garden and the rubber and the coconut plantations. Forest Cover Assessment Maps using LANDSAT imagery were produced in 1991-92. Indicatory inventories of non-forest land and detailed periodic inventories of plantation are carried out for assessing resources and their potential of supply. Forest Resource Assessment is done using the aerial photographs of scale 1:20000 for natural forest and 1:10000 and 1:20000 for plantations. Ground survey is used for management planning. Problems are faced in the acquisition of data, manpower and training.

2.2.3 Foundation papers

Two presentations by FAO staff members, Mr. A. Marzoli, on "Global Forest Resources Assessment: Introduction to the project Objectives, Activities and Definitions", and M. Lorenzini, on "FAO strategy for Country Capacity Building" laid down an excellent foundation for workshop deliberations.

Mr. Marzoli explained the definitions, intent and responsibilities regarding the data for FRA 2000, as decided in Expert Consultation at Kotka III in Finland and about the FAO responsibility for collecting, compiling and processing of information of the developing countries.

Mr. Lorenzini presented the concept of Country Capacity Building "CCB". The CCB expects the countries to provide information to FAO on the status of their forest resources, identification of problems, if any, related to the development of baseline statistical and spatial geo-referenced databases, and utility of the data for strategic integrated planning for forest resources. At the same time, countries can expect that FAO, through CCB, will support countries in solving their problems in the acquisition, homogenization and integration of data, and strategic planning, including access to the data bases available with the FAO.

2.3 Working sessions

The following table provides an overview of the coverage of "core items" during the eight working sessions.

Table 1. Coverage of the Core items of FRA 2000 at June 1998 Workshop

The workshop was enriched by the deliberations on items, in addition to the "core", required by FRA 2000. The additional issues included Shifting Cultivation, Biodiversity, Networking and MIS. The following table provides an overview of such deliberations.

Table 2: Coverage of Additional Items during June 1998 workshop.

2.3 Concluding session

This session reviewed the deliberations of the last four days and wrapped up the workshop. Each delegation was requested to provide a feed back on the utility of the workshop and the activities that, according to them, are necessary to enhance the capacities of forestry institutions in their countries. All delegates valued the workshop and requested for immediate provision of a website, internet, email facilities to them in addition to financial and technical help for FAO to develop the material for FRA 2000. All the members were requested to expedite the submission of information for FRA 2000 as explained in the workshop.

Bangladesh delegates considered this workshop very useful for sharing knowledge and experience amongst the South Asian countries. They emphasized that the strength and the weakness of each country are different, therefore, each country will need different support from FAO under this project. They wanted financial and technical help for internet and access to web-site to prepare material for FRA 2000 and forest planning.

Bhutan representatives considered the workshop a great source of education and exposure. They also felt the need for technical and financial help from the FAO to provide information for FRA 2000.

The Indian delegates supported the workshop for improving the efficiency of the efforts in FRA 2000 and wanted that a web-site be established and internet facilities extended to them to share information and problems. They wanted quicker switching over to the digital mode of analysis and continuous field inventory methodology. The Indian delegates wanted that software like Area Production Model should be made available to all the institutions.

The Myanmar delegates were interested in training to enhance their capacities and in internet and email facilities to share information. They requested the FAO to provide financial and technical help. While appreciating the utility of the workshop, the Nepalese delegates informed that facilities, funds, manpower, and infrastructural facilities were not sufficient in Nepal to serve FRA 2000 and needed FAO support.

The representatives from Sri-Lanka highlighted the utility of the workshop and emphasized on more access to relevant and useful data with forestry-related national, regional and global institutions. They requested for an early establishment of a web-site and extension of Internet facilities among all the member countries.

3. Thematic deliberations

This part provides description of the technical deliberations on "core" and "additional" items of FRA 2000 in various working sessions. Annex "A" contains copies of papers, transparencies or other material provided by the participants in the workshop.

Forest Resource Assessment (Kotka III) identifies ten "core items" on which information is required for the assessment; Area of forest, Protection Status, Forest Ownership, Forest by Ecological zones, Wood Supply Potential, Changes over time, Volume and Biomass, Felling and Removals, Forest Fires, and Non-wood Goods and Services. The "additional" items of FRA 2000, deliberated at the workshop include shifting cultivation, biodiversity, networking and MIS.

This part also contains new information, if any provided by the participants over and above the information included in FRA 1990. Such new information was only provided by India (SFR, 1997), Sri-Lanka (OSFSMP,1995), Bangladesh (Ahmad, 1998) and Myanmar (SFAR, 1998). To complete the picture on each of the core item, this section provides past information contained in FRA 1990 and current definition of related terms for FRA 2000. This helps to identify gaps in the information for FRA 2000.

3.1 Area of Forest and Other Wooded Land

The delegates shared information on the present status of forest resource assessment in their countries. India (SFR, 1997), Sri-Lanka (OSFSMP, 1995), Bangladesh (Ahmad, 1998) and Myanmar (SFAR, 1998), and some of them provided new information over and above what is contained in the FRA 1990. India presented the recent information on forest cover and growing stock assessment contained in State of Forest Resources 1997, that is based on 1993 to 1995 satellite data. Myanmar provided the information on the classification of forest cover types according to the 1995 assessment based on TM digital data from LANDSAT. Similarly, Sri Lanka provided new information on the area of forests based on the 1992 assessment.

Mr. A Marzoli, in his baseline presentations titled "Problem related to the assessment of forest area" and "Estimating Forest Cover for Standard Reference Years", explained in detail about the establishment of a computerized database and development of an adjustment function to compute forest cover for the standard reference years. The data collected on forest cover from different countries, have different reference years and therefore, need to be brought up to the standard years, for example, to the year 1990 and year 2000 for GFRA 2000, to facilitate aggregation, analysis and reporting. The forest area adjustment function or model, which is used to accomplish this difficult task, is based on the correlation between the forest cover and ancillary variables, like, population density over time.

Three main aims of this model-based approach are:

(i) to adjust the sub-national units inventory data to standard reference years,

(ii) to provide a methodology to correctly expand, to all the units, the forest area change data, and

(iii) to develop an understanding of the causes and the dynamics of change in forest area.

This function or a "General" model, that is applicable to a wide range of localities, is computed through a regression analysis using forest change database, together with population time series and ecological zone maps. This analysis during FRA 1990 has provided a good understanding of the relationship between explanatory variable and the forest cover. The analysis indicates a logarithmic relationship along a "S" shaped curve between the population density and the forest cover. The ecological conditions mediate this relationship across their boundaries.

The latest available forest inventory is used as a baseline for each unit, to estimate the forest cover and the changes in forest cover during the standard reference years (for example 1980 and 1990 for FRA 1990), under any one of the following conditions:

(i) Reliable multi-date inventories are available,

(ii) Reliable single-date inventory is available, or

(iii) No reliable inventory is available

If reliable multi-date inventory is available then the model utility is limited to an adjustment function. With single-date inventory, the model is used in its general form. When no reliable inventory is available, then the baseline forest cover areas are extracted from the calibrated vegetation maps existing in the GIS database of the project. The estimated values at sub-national level are aggregated at national, regional and global levels.

Dr. D Pandey in his paper "Assessment of Forest Plantation Areas", discussed about the problems faced in the assessment of plantation areas. One of the main problems is in assessment of the net area of the plantation. Dr. Pandey emphasized on the need of developing a methodology for inventory of plantations, which also provides the necessary information to develop reduction factors for estimating the net areas from gross areas of forest plantations. He also presented values of such reduction factors at global, regional and country levels. These factors vary from 0.47 9(Laos) to 0.92 (Nicaragua).

Dr. Pandey listed eight main problems in assessment of the plantation area in India; ambiguous definition, non-deletion of failed or felled areas, cumulative reporting, non maintenance of proper plantation records, over reporting, corrupt practices, lack of proper monitoring, and difficulty in the assessment of the area through remote sensing. The last problem is technological in nature because linear features such as roadside plantations of size less than the resolution scale could neither be recorded nor delineated.

Mr. P C Tyagi in his presentation on "Forest by Naturalness", discussed about different levels of classification under FRA 2000 guidelines. He also apprised the participants about various difficulties in collecting the information at these levels for FRA 2000. Most of the forests in India have been worked for timber and only few forest tracts may be in their natural state and these are located in remote and difficult areas. High cost of exploitation may have checked their commercial use, but local population or tribals living close to the forests may have used them for meeting their domestic needs. When requested by GOI in 1981, very few states could identify patches of such forests. Therefore, considering the difficulties in identifying natural forest tracts in India, Mr. Tyagi suggested use of special studies to assess the forest and other wooded land according to naturalness categories.

3.1.1 Recapitulation of Terms and Definitions

The deliberation on "Area" was more focussed around the definition, methodologies and difficulties in providing information for FRA 2000. The following lines briefly recapitulate the relevant definitions:

Total area: Total area of country, including inland water bodies but excluding offshore territorial waters.

Inland water: Area occupied by major rivers, lakes and reservoirs.

Land Area: Total area excluding inland water. It may consist of Forest or Non-Forest lands.

Forest: Land with tree cover of more than 10%, area of more than 0.5 hectares and minimum height of 5 meters at

maturity but excludes land predominantly under agriculture. The forest area is divided into three categories:

Undisturbed by man: These are classified into "Closed forest" and "Open forest."

Closed forest: It consists of areas with high proportion of area covered by trees and which do not have continuous dense grass layer. The "closed forest" category is further classified into three categories; Broad-leaved, Coniferous, Bamboo/Palm.

Open forest: These are forest tracts with tree cover of at least 10 percent and generally have continuous grass layer. The forests under belong to two main groups;

Broad-leaved and Coniferous.

Plantations: Forest stands established by planting or /and seeding in the process of afforestation and reforestation.

Semi-Natural: Forest or other wooded land, which is neither "forest/other wooded land undisturbed by man" nor "plantation"

Non-Forest: It includes "other wooded land" and the rest "other land" defined below,

Other wooded land: Land with any of the following three attributes for crown cover and height,

This category contains shrubs and fallow lands. The shrub is defined as the woody vegetation dominated by shrubs, with height between 0.5 m to 5 m and without a definite crown. The forest fallow is defined as complexes of woody vegetation deriving from the clearing of natural forest for shifting agriculture.

Other land: Land that is not classified as "forest" or as "other wooded land."

3.1.2 Information on "Forest Area" with FAO and that provided by the participants

Bangladesh

Bangladesh is in the process of publishing new information of 1997 therefore participants shared following FRA 1990 information only on the area of forest and other wooded land.

Table 3. Area of Forest and Other wooded land in Bangladesh (FRA, 1990)

Source: FAO, 1995.

Bhutan

Bhutan has completed its new assessment but participants could share information contained in FRA 1990 only.

Table 4. Area of Forest and Other wooded land in Bhutan (FRA, 1990)

Source: FAO, 1995.

India

FSI provided a copy of the "Status of Forest Resources "SFR", 1997" to each of the participants which contains new information over and above the FRA 1990 report. Following provides details of information of "Area" available in FRA 1990 and SFR 1997.

Table 5: Area of Forest and Other wooded land in India (FRA, 1990)

Source: FAO, 1995.

Table 6: New information on Forest and Non Forest Cover in India (SFR, 1997)

Source: SFR, 1997. (FSI)

Table 6. Extent of Forest area undisturbed by man in India, 1981

From 1991-92, FSI has also taken up inventory of trees growing outside conventional forest area in the country. Such plantations have been classified according to their locational characteristics like Farm forestry, Village woodlot, Block Plantation, Roadside, Pond, Railside and Canalside Plantations, etc. An inventory for trees planted outside forest areas has been completed by FSI for one of the states (Haryana) in India (SFR 1997).

Myanmar

Table 7. Area of Forest and Other wooded land in Myanmar (FRA, 1990)

Source: FAO, 1995.

Nepal

Table 8. Area of Forest and Other wooded land in Nepal (FRA 1990)

Source: FAO, 1995.

New information is under process. The forest cover of Nepal, on the basis of 1944 information from NOAA satellite data, reveals that the mixed and coniferous forests exist over 41.6% of the area of the country, excluding 14.5% of degraded forests. (Myint, 1998)

Sri Lanka

Sri-Lanka provided 1992 information over and above FRA 1990 for the benefit of the participants.

Table 9. Area of Forest and Other wooded land in Sri-Lanka (FRA, 1990)

Source: FAO, 1995.

Based on the new information, the total land area including all inland water bodies is 6,616,628 ha. The area of natural closed canopy forest is 1,582,756 ha and the sparse and open forest occupies a total of 463,842 ha. The total area under mature and well-established forest plantations is 72,340 ha (Legg and Jewell, 1998).

Table 10. Areas of Natural Forest in Sri- Lanka in 1992

Source: Legg and Jewell, 1998.

Table 11. Areas of Forest Plantations in Sri-Lanka, 1992

Source: Legg and Jewell, 1998.

In 1992, Home gardens covered 860,000 ha and rubber and coconut plantations together covered about 500,000 ha. (OSFSMP, 1995).

3.2 Protection Status

Mr. P C Tyagi motivated the participants to deliberate on the issue of "Protection status" in their countries, through his paper on "Protected Areas in India". He explained the important biological and practical parameters necessary in planning protected areas, available information on the status and statistics of the existing protected areas. He also dealt with the bio-geographical classification, developed by the Wildlife Institute of India, for conservation planning in India. During the workshop, only India and Sri-Lanka provided some new information on this item of FRA 2000.

3.2.1 Recapitulation of Terms and Definitions

Information is required for protection status of the "Forest" and "Other wooded lands" by following six IUCN categories under two groups; (a) Strictly Protected Areas (Categories I and II), and (b) Protected Areas with integrated management (Categories III to VI).

(A) Strictly Protected Areas

I. Strict Nature Reserve/Wilderness Area: Protected area managed for science or wilderness protection.

II. National Parks: Protected area managed mainly for ecosystem protection and recreation.

(B) Protected Areas with integrated managements

III. Natural Monument: Protected area managed mainly for conservation of specific natural features.

IV. Habitat / Species Management Area: Protected area managed mainly for conservation through management intervention.

V. Protected Landscape/Seascape: Protected area managed mainly for landscape /seascape conservation and recreation.

VI. Managed Resource Production Area: Protection area managed mainly for sustainable use of natural ecosystem.

3.2.2 Information on "Protection Status" with FAO and that provided by the participants.

Bangladesh

Information not available

Bhutan

Information not available

India

Analysis of protected areas is done within the boundaries of the bio-geographic zones and provinces. Information on biome within each protected area or within each bio-geographic unit is not readily available. As an alternative, presence or absence data was used for analysis of biome coverage. The data analysis shows that the PA's are not distributed uniformly across the states or across the bio-geographic Zones or provinces of the country.

In India, the conservation status of forests is directly linked to the legal category of protection under Forest and Wildlife Acts.

The Wildlife Protection Act, 1972, recognizes four categories of protected areas; National parks, Sanctuaries, Game Reserves, and Closed Areas. The IUCN categories I and II are almost equivalent to the National parks and categories IV and VI of IUCN cover the Sanctuaries.

Accordingly, India has a network of 85 National parks covering 36,619 Sq. Km of area conforming to the group comprising of category I and II of IUCN. Further, 450 Wildlife sanctuaries in India are spread over 113,169 Sq. Km and come under the group consisting of categories IV and V of IUCN. National Parks and Sanctuaries, as a whole, cover some 149,788 sq. km or 4.70% of the countries land surface. (Tyagi, 1998).

Table 12. Protected Areas in India by IUCN categories

Source: Tyagi, 1998.

Myanmar

At present, 18 Wild life Sanctuaries and 3 National parks totaling 7,731 Sq.km or 1.14% of the total area of the country have been established.

Nepal

Information not available

Sri Lanka

In Sri-Lanka, the total protected area is about 9,300 Sq. km., which is about 14% of the total land area. (OSFSMP, 1995).

Table 13. Protected areas in Sri-Lanka

Source: FAO- APFSOS/WP/16, 1997

3.3 Forest Ownership

Mr. R K Upadhyay initiated the discussion on "Ownership", through presenting his paper titled "Forest Ownership in India". He focused on statutory and legal nature of forest ownership in India. All property in India is vested in the Constitution of India, which respects all types of ownership for forests, ranging from private to community, to corporate and to the government. He provided information, which may help to classify ownership of forests on the basis of rights, concession and possession (legal transfer or encroachment). The Paper provides details on statewise forest cover in India, which has been notified as forest, forest owned by departments other than forests, corporate bodies, community, and private individuals. The Paper also deals with the system of annual reconciliation of ownership and possession of forestland in one of the states (Tamil Nadu) of India.

3.3.1 Recapitulation of Terms and Definitions

The following lines provide the agreed terms and their definitions (Kotka III) for the above core information item, upon which the information is needed for GFRA 2000.

Public Ownership: Belonging to State or Other public bodies. It is categorized in following three groups:

State Ownership: Owned by National, State and Regional governments or by Government owned

Corporations; Crown forests

Owned by other public institutions: Belonging to cities, municipalities, villages and communities. It includes all publicly owned forest and other wooded lands that are not under "State ownership"

Owned by indigenous and tribal people: Owned by indigenous and tribal councils in independent countries.

Private ownership: It is, also, categorized in following three types of ownership:

Individual ownership: By individuals and families

Owned by forest industries: By private forest or wood processing industries

Owned by other private institutions: By private corporations, co-operatives or institutions.

3.3.2 Information on "Forest Ownership" with FAO and that provided by the participants.

Bangladesh

No information is available.

Bhutan

No information is available.

India

According to 1988 National Forest Policy, forest lands or lands with tree cover are treated as a national asset and the Constitution Of India safeguards them for providing sustained benefits to the entire community. The level of protection provided to them varies with their legal status or differentiation. First major differentiation is on the basis of control by the government forest department and others. The forests with forest departments are further legally differentiated into reserved, protected and unclassed forests, which define the degree of their access and use, by local communities (Table 14).

Table 14. Forests with Government Forest Departments

Source: Upadhyay, 1997.

The forests, other than those with government forest department, are further classified on the basis of ownership and control (other government departments, public corporate bodies and communities, and private individuals), as summarized below (Table 15).

Table 15. Forest not with Government Forest Departments

Source: Upadhyay, 1998.

Myanmar

Information was not available.

Nepal

Information was not available.

Sri Lanka

Information was not available.

3.4 Forest By Ecological Zone

Mr. A K Myint laid down the basis for deliberations on "Ecological zoning" of forests through his paper titled "Forest Resources of Hindu Kush Himalayas (HKH)" covering regional view of the part of the forests of Pakistan, India, Nepal, Myanmar, China, Bhutan, Bangladesh, and Afghanistan. Following Kawosa (1988), he divides the Himalayan forests of India, Bhutan and Bangladesh into two zones; Eastern and Western. Major forest types of the Eastern Himalayas zone are further differentiated as Tropical Evergreen forests, Sub-Tropical forests, Eastern Temperate forests, Sub-Alpine forests, Alpine vegetation, Stony deserts. The Western Himalayas zone is similarly described as Tropical Deciduous forests, Sub-Tropical Pine forests, Himalayan Moist Temperate forests, Himalayan Dry Temperate forests, Sub-Alpine forests, or Alpine Pastures and Scrub.

The forests in HKH region roughly cover about 24% of the land. Forest cover estimates are available individually for countries as a whole in the HKH region from many sources like FAO, WCMC, and WRI. However, it is difficult to get a precise picture of forest cover in this region due to non-availability of even district level estimates for the areas falling in the HKH region. International co-operation can help in this field by providing necessary information for such study.

3.4.1 Recapitulation of Terms and Definitions

Kotka III deals with ecological zoning through following terms and classification. Forests are divided in two main domains:

(i) Tropical and Sub-tropical, and

(ii) Temperate and Boreal.

The first domain "Tropical and Sub-tropical" consists of following seven ecological zones:

Tropical wet and very moist zone,

Tropical moist zone,

Tropical dry zone,

Tropical very dry zone,

Desert (Hot and Cold) zone,

Tropical and sub-tropical hill and montane zone, hot and moist, and

Tropical and sub-tropical hill and montane zone dry.

The second domain "Temperate and Boreal" comprises of following five ecological zones:

Dry (Mediterranean type) temperate zone,

Temperate zone,

Temperate boreal hill and montane zone,

Boreal zone, and

Tundra zone.

3.4.2 Information on "Ecological Zoning" with FAO and that provided by the participants.

Bangladesh

Table 16. Land area and Forest cover by Ecological zone in Bangladesh (FRA 1990)

Source: FAO, 1993

Bhutan

Table 17. Land area and Forest cover by Ecological zone in Bhutan (FRA, 1990)

Source: FAO, 1993

India

Table 18. Land area and Forest cover by Ecological zone in India (FRA 1990)

Source: FAO, 1993

Myanmar

Table 19. Land area and Forest cover by Ecological zone in Myanmar (FRA 1990)

Source: FAO, 1993

The Japanese Forestry Technical Association (JAFTA) using Landsat TM digital data in 1995, has identified following forest cover types:

Table 20. Forest Cover Types in Myanmar

Source: SFAR, 1998.

Myint (1998) recognizes the following 10 standard forest types in Myanmar:

Tidal, beach and dune and swamp forests,

Tropical evergreen forests,

Tropical semi-evergreen of Riverain-evergreen forests,

Mixed deciduous forests,

Lower mixed deciduous,

Moist upper mixed deciduous forests,

Dry upper mixed deciduous forest,

Deciduous Dipterocarp forests,

Dry and thorny forests, and

Sub-tropical hill and alpine forests.

Nepal

Table 21. Land area and Forest cover by Ecological zone in Nepal (FRA 1990)

Source: FAO, 1993

The 1997 national forest inventory (FRSC/FINNIDA 1997) differentiates the forests of Nepal by dominant species into following 10 forest types.

Table 22. Forest Types in Nepal

Source: Myint, 1998.

Sri Lanka

Table 23. Land area and Forest cover by Ecological zone in Sri-Lanka (FRA 1990)

Source: FAO, 1993

Forest categories interpreted from Satellite Imagery and derived by GIS techniques in Sri-Lanka are as under

Table 24. Forest types in Sri-Lanka

Source: Legg and Jewell, 1998.

3.5 Wood Supply Potential

Mr. M Lorenzini led a discussion on the basis of his paper on "Deforestation and Accessibility" which evaluates use of GIS information as a potential tool to improve the knowledge of human-induced "deforestation" or forest "access" process. Mr. Lorenzini explained it with the help of a study in the Central Africa region, which includes five countries (Cameroon, Central African Republic, Congo, Equatorial Guinea, Gabon and Zaire). This study attempts to use new global 30 ARC seconds (approx. 1 km resolution) DEM GTOPO30 data sets with other global digital data sets like the Digital Chart of the World and the Global Land Cover characterization data set. Besides these, this study uses a number of selected multi-temporal observations (full LANDSAT scenes) for the region, to develop the maps for accessibility.

The paper hypothesizes that, people convert forests into other land-use due to two main factors; (a) easy accessibility, and (b) income generating potential of tree and forest land. The model is based on a cost distance algorithm available within ARC/INFO GIS package. Two surfaces were generated independently from two data sets i.e. urban settlements and roads. These were used in the final cost map as a weighing factor, so that total cost increases when people move far away from the towns/villages and roads.

3.5.1 Recapitulation of Terms and Definitions

The expert consultation held at Kotka, Finland in 1996 (Kotka III) desires that the forests should be divided into available/not available for wood supply, where availability for wood supply is defined according to legal, economic, technical or environmental restrictions.

3.5.2 Information currently available with FAO and that provided by the participants is given below.

Bangladesh

Table 25A. Forest Available for Wood Supply

Source: GFSS Estimate

Table 25B. Forest Not Available for Wood Supply

Source: GFSS Estimate

Bhutan

Table 26A. Forest Available for Wood Supply

Source: GFSS Estimate

Table 26B. Forest Not Available for Wood Supply

Source: GFSS Estimate

India

Table 27A. Forest Available for Wood Supply

Source: GFSS Estimate

Table 27B. Forest Not Available for Wood Supply

Source: GFSS Estimate

Myanmar

No information is available.

Nepal

Table 28A. Forest Available for Wood Supply

Source: GFSS Estimate

Table 28B. Forest Not Available for Wood Supply

Source: GFSS Estimate

Sri Lanka

Table 29A. Forest Available for Wood Supply

Source: GFSS Estimate

Table 29B. Forest Not Available for Wood Supply

Source: GFSS Estimate

3.6 Changes Over Time

Mr. S Tripathi, FSI & Mr. P K Pathak, FSI presented a paper on a methodology of FSI to assess the changes in forest cover. FSI assesses forest cover periodically on a 2-year cycle. FSI uses IRS-1B Linear Imaging Self-Scanning (LISS-II) Sensor satellite data for the assessment. National Remote Sensing Agency (NRSA) supplies this data in the form of False Color Composites (FCC). FSI staff interpret most (72 percent) of the information visually and the rest (28 percent) digitally. Ground verification helps in post interpretational validation. More than 2,000 spots are checked during the ground verification in each assessment. Manual dot-grid method provides estimates of an area. FSI is slowly adopting the Computer Aided Cartography (CAC) and digital mode of forest cover assessment in a phased manner. The paper elaborates the four steps of Digital Image Processing, i.e. Image preprocessing, Classification, Area calculation, and Ground truthing.

Mr. Marzoli, FAO presented Mr. Drigo's paper the "Monitoring of Forest Resources at district level using Multi-Date Satellite Data" to explain FAO methodology to deal with monitoring of forest resources at district level in order to introduce reliable information on the processes of change into the forestry planning system by applying interdependent interpretation of multi-date satellite data methodology.

Mrs. Ranjana Gupta (FSI) presented a paper on "Forest Grazing". She provided information on the estimated and projected livestock population, based on cumulative growth rates. She also presented the projected annual requirement of green fodder and dry fodder based on NCA (1976) norms. FSI assesses the "incidence of grazing" at the time of ground inventory or validation of forest cover. The basis of assessment is ocular estimation.

FSI classifies forest into five categories, based on the extent of grazing incidence. The paper provides such assessment of grazing incidence by the FSI. Indian forests bear heavy grazing in 43587 Sq.km. (18.85 %), medium grazing in 54638 Sq.km. (23.63 %) light grazing in 64406 Sq.km (27.85%), no grazing in 64810 Sq.km. (28.03 %) and unrecorded incidence of grazing in 3796 Sq.km. (1.64 %) area.

3.6.1 Recapitulation of Terms and Definitions

According to the expert consultation on FRA 2000 held at Kotka, Finland in 1996 (Kotka III), following are agreed terms and their definitions in respect of "Forest Change."

FRA 2000 requires three types of information on forest change- Deforestation, Forest degradation and Forest plantations. The deforestation is defined as change of land use where tree crown cover has depleted to less than 10 percent. The Forest degradation is assessed through reduction in biomass changes in species composition and soil degradation. Data on area and species composition of Forest plantations established either through afforestation or through reforestation is required to assess the changes.

3.6.2 Information on "Change over time" with FAO and that provided by the participants.

Bangladesh

Table 30. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

Bhutan

Table 31. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

India

Table 32. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

FSI provides following additional information regarding change in forest cover over time.

Table 33. Change in forest cover

Source: SFR 1997.

Myanmar

Table 34. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

Nepal

Table 35. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

Sri Lanka

Table 36. Annual changes in Forest and Other wooded land, 1980-90

Source: FAO, 1995.

Rate of Deforestation in Sri-Lanka averages around 20,000 ha per year. Legg and Jewell (1998) provide additional information about change in forest cover.

Table 37. Changes in Forest Cover, 1992

Source: Legg and Jewell, 1998.

3.7 Volume and Biomass

Mr. S K Chakrabarti (FSI) presented the methodology of FSI for assessing volume of the forest growing stock in India. This information is collected at the time of ground validation of the forest cover. The methodology, therefore, consists of combining the maps of forest density and forest types into a forest "density-type" map. This "density-type" map at 1:50,000 scale of all the States and Union Territories is divided into grids of 2.5' x 2.5' (latitude x longitude). Each grid is allocated one of the four density categories; Very dense forest, dense forest, open forest, no forest. The grids are stratified on the basis of the species associations (forest type) and density categories. For example, in 1995, the grids were stratified into 21 strata of unique forest composition. For each of these grids, the secondary data on "volume per hectare" is drawn from standard national forest inventory on the basis of two sample plots of 0.1 ha in each forested grid. The volume of growing stock in a forested grid is obtained by multiplying the average volume per hectare information from the concerned sample plots with the area of the grid (1800 ha). The grid volumes are added to get an estimate of growing stock for each stratum. The volumes of each strata are added at the state or national level to estimate the volume of the forest growing stock.

Mr. M Lorenzini presented FAO methodology of estimating actual forest volume using forest inventory and climatological data. This methodology is based on GIS and helps to improve the knowledge of spatial distribution of forest volume. FAO develops a model to produce an "Actual Volume" map considering that available data from forest inventories were insufficient to extrapolate volume estimates across the region.

The data used in his paper were derived from the forest inventory reports of the Tropical South Asia: India, Pakistan, Nepal, Bhutan, Bangladesh and Sri Lanka. The vegetation maps and administrative unit maps were combined in an overlay and linked to the forest inventory database to generate Forest Inventory Plot map. The population data of the year 1980 were considered and a Population Density map was generated. The Vegetation map was drawn by using the map proposed by the International Institute for Vegetation Mapping, France. The Forest map was derived from a reclassification of the vegetation map on the basis of statistical analysis.

Weck's Climatic Index -WPIB (calculated with Brown's formula) and an Accessibility map is prepared, based on the assumption that the current distribution of the forest biomass is defined by the potential amount that a landscape can support under the prevailing climatic, edaphic and topographic condition and the cumulative impact from human activities.

Different data layers like Inventory Plot Map, Accessibility Map, Population Density Map and Productivity Index Map(WPIB) help to generate the model that provides the Volume Density Map. The Volume Density map is finally clipped with the Forest Map to obtain Forest Volume Map.

3.7.1 Recapitulation of Terms and Definitions

The expert consultation on FRA 2000 held at Kotka, Finland in 1996 (Kotka III), desires information on forest volume in terms of "Volume" and "Biomass" as defined below.

The growing stock is defined as Gross "Volume Over Bark" of free bole (from stump or buttress to crown point or first main branch) of all living species more than 10-cm dbh.

The "Biomass" is defined as the totals above ground organic matter in trees, expressed in oven dry tons per unit area.

3.7.2 Information on "Volume and Biomass" with FAO and that provided by the participants.

Bangladesh

Table 38. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

Bhutan

Table 39. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

India

Table 40. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

An attempt to estimate the growing stock of forest resources was made by the FSI in 1995 under a UNDP project. The growing stock of the country was assessed using information available in vegetation maps, thematic maps and ground inventories of forests carried out from time to time. The growing stock of the country is estimated to be 4740 million cu.m with an average volume of 74.42 cu.m per ha.

Table 41. Forest Growing Stock in India by Broad Forest Strata, 1995

Source: Chakrabarti,1998.

Myanmar

Table 42. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

Nepal

Table 43. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

Sri Lanka

Table 44. Forest area, Volume and Biomass, 1990

Source: FAO, 1995.

3.8 Felling and Removal

The workshop did not deliberate on this core item of FRA 2000. According to the expert consultation held at Kotka, Finland in 1996 (Kotka III), "felling" and "removals" indicate the intensity of wood utilization in a given period (e.g., annual) as described below.

"Felling" indicates the average over bark volume of all trees upto 10 cm (dbh), living or dead that are felled, whether removed or not.

"Removals" means the annual under bark volume actually cut and removed that generate revenue for the owner of the trees. It includes wood used for domestic and industrial purposes but does not include "fuelwood."

3.9 Forest Fires

Mrs. Ranjana Gupta, FSI, India presented a paper on FSI methodology for assessing forest fire. She explained the sampling design and the methodology for assessing the extent of forest fires. The information on fire incidence is collected at the time of ground validation and forest inventory, which is done on 1:50,000 scale topographic sheet divided into 36 grids of 2 �' x 2 �' of latitudes and longitudes. In each of such grids, two sample points are marked. The fire incidence information is collected for the sampled area from secondary sources as well as from visual examination of about 2 ha area around the sample square plot of .1 ha. Qualitative data such as land use, crop composition and its density, fire and grazing incidence etc. are recorded in the Plot Description Form (PDF).

FSI uses six categories for reporting forest fire incidence in the country. Five of these categories relate to the forest area for which recorded information on forest fires is available and the sixth category represents rest of the forest area for which no such information exists. The five categories are forest areas with very heavy, heavy, frequent, occasional, and no fire.

3.9.1 Recapitulation of Terms and Definitions

According to the expert consultation on FRA 2000 held at Kotka, Finland in 1996 (Kotka III), information is needed on area (size), damage and time trend of forest "fire". The expert consultation describes the forest "fire" as given below,

The "fire" includes fire that breaks out and spreads on or to "forest" and "other wooded land" but does not include "prescribed" or "controlled" burning of "forest" or "other wooded land."

3.9.2 Information on "Forest Fires" with FAO and that provided by the participants.

Bangladesh

No information is available.

Bhutan

No information is available.

India

Forest fires may be natural or man made. The number of fires that arise due to natural causes are not more than 5 percentage of the total number of fires in one year. Men cause about 95 percent of the fires. Table 44 demonstrates assessment of the extent of forest fire under different categories.

Table 45. Extent of forest fire under different categories

Source: Gupta, 1998.

Myanmar

No information is available.

Nepal

No information is available.

Sri Lanka

No information is available.

3.10 Non Wood Goods and Services

Mrs. Ranjana Gupta, FSI, India presented a paper on FSI conception of Non Wood Timber Products. She dealt with classification, commercial significance, importance, and constraints in the collection and marketing of NTFP. Statistics on statewise contribution to the forest revenue, potential and production, and quantity and value of export of NTFP are presented in the paper. The paper also deals with the proposed methods for the assessment of NTFP's.

3.10.1 Recapitulation of Terms and Definitions

The expert consultation on FRA 2000 held at Kotka, Finland in 1996 (Kotka III), classified non-wood goods and services into six major groups. The goods are divided in three categories; (a) Products for human consumption, (b) Fodder and Forage and (c) Other non-wood goods. Similarly, the services have been grouped into three classes; (a) Providing protection values (b) Providing Social and Economic values and (c) Providing Aesthetic, Cultural, Historical, Spiritual and Scientific values. The non-wood goods include all goods of biological origin derived from forest or any land under similar use, and exclude wood in all its forms.

FRA 2000 needs the following information on each "Non Wood Good" or "Service":

� description,

� indication of relative and absolute importance,

� changes in supply and demand, and

� quantity and value supplied.

3.10.2 Information on "Non Wood Goods and Services" with FAO and that provided by the participants.

Bangladesh

No information is available.

Bhutan

No information is available.

India

In India, Non Timber Forest Products (NTFP) of commercial importance have been divided into the following (Table 46) nine classes:

Table 46. Categories of NTFP in India

Myanmar

No information is available.

Nepal

No information is available.

Sri Lanka

No information is available.

3.11 Additional Parameters

Biodiversity

Mrs. C Elouard, FIP, provided exposure on a field method to assess biodiversity through her paper "Tools for Assessment of Forest Biological Diversity". The data used in this paper was collected during the field work from three forest types, e.g., evergreen, moist deciduous, dry deciduous forests in Karnataka state, India. The structural analysis in the paper uses the frequency data to understand the distribution of species and individuals, and the application of the relationship between height and girth, to understand the structural stratification of the forest. The paper also explains the diversity assessment through computation of various indices like Chao, Simpson, Shannon -Wiener, and Evenness.

Mr. M Lorenzini explained the FAO methodology to assess biodiversity at ecosystem level by presenting his paper "Biodiversity Assessment: FAO perspective". The combination of Eco-floristic layers and Vegetation type layers lead to a number of unique combinations of actual vs. potential land-cover conditions. Dominant or characteristics woody species of the flora and their successional patterns within major ecological regions define eco-floristic zones. The vegetation cover/condition layers are expressions of vegetational characteristics such as density, continuity of plant cover, height, etc. and human impact on the vegetation. The vegetation map together with eco-floristic zones may be used to capture global biodiversity. Relating biodiversity with human pressure indicators provides capacity to predict or fill gaps to assess biodiversity at any spatial or temporal location.

3.11.1 Recapitulation of Terms and Definitions

FRA 2000 utilizes the following definition of biodiversity provided by UNCED 1992:

"Biological diversity refers to the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are a part; this includes diversity within species, between species and of ecosystems"

FRA 2000 assesses biodiversity at ecosystem level and defines it as "the different forest or vegetation types or species associations in an area unit" (Singh, 197)

Shifting Cultivation

Dr. V. N. Pandey, FSI, introduced the subject of assessment of forestland, under shifting cultivation, through his paper on "Shifting Cultivation." The shifting cultivation is one of the major factors, that define forest change, not only in India but other countries. In India, it extends over Andhra Pradesh, Bihar, Madhya Pradesh, Orrissa, Tamil Nadu and North-Eastern states. Currently, FSI assesses these areas, as a part of the regular forest cover assessment, and classifies such areas as non-forest areas.

The shifting cultivation is the single most important factor, that explains not only the forest change in North-Eastern states but also the social and economic conditions of the dependent communities. Recognizing the unique significance of shifting cultivation in forest assessment, the FSI has initiated a special study on this subject in Nagaland. The paper provides a brief review of past assessments of shifting cultivation areas and a summary of the proposed study.

3.12 Other Items discussed at the workshop

3.12.1 Electronic Networking

Dr. Kailash Govil, FAO presented a paper on electronic networking including webpage. The paper deals with country, regional and global node for electronic networking. The Country Capacity Building Regional Node "CCBRN" aims to build country capacities for better assessments and effective strategic planning for participative sustainable management of forests through exchange and sharing of information and knowledge with the help of electronic networking. The CCBRN will serve as a repository of regional data and publisher of useful information. In addition, CCBRN will have an Answer to Question "AQ" service. This network will encourage the development of joint activities between countries, institutions and individuals in the region. In each country, network activities will be coordinated by one of the member institutes or organization, which will act as national node CCBNN. These national nodes will maintain contacts with all the regional nodes CCBRN. The regional node, in turn, will maintain similar relations with global node CCBGN maintained by Director CCB at FAO HQs. All the participants were excited about it and wanted "Internet" and "Web" facilities as early as possible.

3.12.2 Planning

Andrea Kutter, FAO deliberated on her paper "An integrated approach to planning and management of Landscape resources as promoted by FAO", and explained the framework for land evaluation (1976) and its major problems including the new approach for integrated land-use planning.

Mr. A Marzoli, FAO presented a very useful and potential paper on "Area Production Model (APM)" which is a simulation model for long-term land use changes and prediction of yields from Agriculture and Forestry. It aims to offer the possibility of harmonizing the use of sectoral inputs and balancing the supply and the demand through the simulation of consequences under various conditions. The APM needs data on land use, biomass/energy and forestry to develop alternative development scenarios.

3.12.3 Forest Inventory Management System

Mr. Grewal presented his paper "Global Forest Resource Assessment 2000 - a user's perspective", to initiate discussion on Forest Inventory Management System (FIMS) in India. He pointed out that, almost all the state forest departments in India are regularly carrying out very intensive micro level forest inventories, for working plan and other purposes. This inventory data is generally kept, by the forest departments, in hard copies, and therefore, is very voluminous and cannot be easily harmonized and used. The Forest Department of Maharashtra State has taken a lead in this matter by computerizing such forest inventory data for its easy and quick use in planning, management and monitoring of forest resources. Mr. Grewal presented a live demonstration of such a FIMS, which is being used for online inventory data processing in Maharashtra State for the last one year. This system is very user friendly with pull down menu for easy data entry and browsing. The potential utility of this FIMS has been increased through its integration with a Geographical Information System.

References

Ahmad, Ishtiaq Uddin, 1998. Forest Resource Assessment and Introduction of Geographical Information System (GIS) for Resource Information Management in Bangladesh. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12,1998.

Chakrabarti, S. K, 1998. Growing Stock Estimates: The Country Perspective - FSI contribution. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

FAO, 1993. Forest Resources Assessment 1990 -- Tropical Countries. FAO Forestry Paper 130. Food and Agriculture Organization of the United Nations, Rome.

FAO, 1995. Forest Resources Assessment 1990 - Global synthesis. FAO Forestry Paper 124. Food and Agriculture Organization of the United Nations, Rome.

GFSS, 1998. Draft of Global Fibre Supply Study, Forest Products Division, Forestry Department, Food and Agriculture Organization of the United Nations, Rome.

Gupta, Ranjana, 1998. Forest Fire. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

Legg, Christopher and Jewell, Nicholas, 1998. A 1:50,000 Scale Forest Map of Sri Lanka: the basis for a National Forest GIS. Paper circulated at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

Myint, A. K, 1998. Forest Resources of Hindu Kush - Himalaya. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

OSFSMP, 1995. An Overview of the Sri Lanka Forestry Sector Master Plan,

Forestry Planning Unit, Ministry of Agriculture, Lands and Forestry, Battarmulla, Sri Lanka.

Pandey, D, 1998. Assessment of Forest Plantation areas. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

SFAR, 1998. State of Forest Assessment Report, Forest Department, Ministry of Forestry, Myanmar.

SFR, 1997. State of Forest Report, Forest Survey of India, Ministry of Environment and Forests, Dehradun, India.

Tyagi, P. C., 1998. Protected Areas in India. Paper presented at FAO organized workshop on "The Status of Forest Resources Assessment in the South Asia Sub-region and Country Capacity Building Needs" at Dehradun, India. June 8 to 12, 1998.

Appendix 1: Resource persons and participants

The resource persons

FAO

Mr. A. Marzoli, Forestry Biometrician/Ecologist, FAO

Dr. M. Lorenzini, GIS Expert, FAO

Dr. Kailash C Govil, Regional Project Coordinator

French Institute of Pondicherry

Miss. C. Elouard, French Institute of Pondicherry

Forest Survey of India

Dr. D Pandey, Director, FSI

Mr. P C Tyagi, Joint Director, FSI

Mr. V N Pandey, Joint Director, FSI

Mr. S K Chakrabarti, Deputy Director, FSI

Mrs. Ranjana Gupta, Deputy Director, FSI

The participants

Bangladesh

Mr. Anwarul Huq, Deputy Secretary, Ministry of Environment and Forest

Mr. Ishtiaq Uddin Ahmed, Deputy Conservator of Forest

Bhutan

Mr. G K Pradhan, Joint Director, Forestry Survey Division, Ministry of Agriculture

Mr. Pasang Wangchen Norbu, DFO, Forest Resources Development, Ministry of Agriculture

India

Mr. J S Grewal, Conservator of Forests, Working Plans, Nagpur

Mr. Ashok Ramteke, AIGF, Ministry of Environment and Forests

Myanmar

Mr. U Shwe kyaw, Deputy Director General, Forest Department

Mr. U Tint Lwin, Staff Officer, Forest Department

Nepal

Mr. Sharad Rai, Divisional Forest Officer, Department of Forest

Mr. Surya Prasad Adhikari, Legal Officer, Department of Forest

Sri Lanka

Mrs. P A Kumaradasa, deputy Director, Ministry of Forestry & Environment

Mr. P M A de Silva, Dy. Conservator of Forests, Department of Forests

Appendix 2: Daywise session profiles of the workshop

Appendix 3: An outline for country report on FRA 2000

I. Identification of policy problems and information needs.

Satisfying present and future needs of fuelwood and fodder

Satisfying present and future needs of watershed protection and rehabilitation of degraded land

Satisfying present and future needs of industrial wood

Sustainable forest management

Conservation of biological diversity

Joint forest management

II. Information

II.A. Ecological setting

Country land area and administrative units

Ecological zones

Physiography

Rainfall

Productivity index

II.B. Socio-economic setting

Demography

Land use

Agriculture (area/productivity)

Livestock

Energy requirements

GNP

Occupation

Ownership

II.C. Forest Resources

Description of the natural vegetation types

Present forest area by administrative unit and ecological zone.

Potential versus actual forest resources

Volume and Biomass

Forest available for wood supply

Forest protection and Conservation

Forest plantations

Forest management

Increment

Minor forest produces

II.D. Forest production / consumption studies

Wood production and consumption (roundwood, fuelwood) balance

Fodder production/consumption studies

III. Trends and prospectus (Future scenarios)

Deforestation

Plantation development (area and increment)

Land use changes

Forest degradation

Biomass change

Wood supply and demand future balance

Biodiversity loss risk

Soil/water conservation

Appendix 4: Information for electronic networking

Country: ____________________ Name of person filling this form _________________

Position or post: ______________________________________________________________

(1) Has someone been nominated to coordinate this project at National level YES / NO

(2) Address of the person for correspondence regarding the Country Capacity Building Project.

His/Her Name: ____________________________

Position: _________________________________

Mailing Address: ___________________________________________________________

___________________________________________________________

___________________________________________________________

Telephone: ___________________________________________________________

Fax: ___________________________________________________________

Email: ___________________________________________________________

(3) More information about the computer and Electronic Network Facilities

Macintosh Yes / No

IBM Yes / No

Speed _______Mhz

Type Motorola / Pentium or other _______________

RAM _________Mega bytes

Hard disk _________Mega / Giga bytes

Floppy 5(1/4)", 3(1/2)", SD / HD

Modem Type ___________ and Speed less than 14.4 kbps/14.4/more than 14.4

Type of Internet TCP / IP or shell _____________________________________________

Operating System Macintosh/OS2/Unix/Windows3.14/Windows95/WindowsNT3.5/Windows 4.0

Word Processing: Word / Word Perfect and other __________________________________

GIS ___________________________________________________________

Remote sensing ___________________________________________________________

(Please fill the form and return it personally to Dr. Kailash C. Govil, Regional Project Coordinator, Country Capacity Building Project, Room No. 124, Forest Survey of India Building, Dehradun)

APPENDIX 5: Technical papers

Section 1: Deforestation and Forest Accessibility

by Massimiliano Lorenzini, FAO

Introduction

The word deforestation indicates the conversion of forests to other land cover types. These changes can be produced by natural events like fire, tornado, and avalanches as well as by human beings. In this paper a methodology based on the use of GIS information, is evaluated as a potential tool to improve the knowledge of the human-induced deforestation processes.

The problem, that many researcher are nowadays approaching is related to the following question:

Where is the deforestation taking place and

Where is likely going to take place in the future?

On a global scale this is a very challenging problem since local scale deforestation models, usually based on a great number of variables, in most of the cases cannot be applied on a global scale due to the lack of information for such variables all over the world.

Simple models have been implemented in the past 10 years, which parameterize in a quite simple but robust way the obvious observation that people are the cause of the deforestation; the population pressure explains very well the depletion of the forests.

Nevertheless tests are in progress attempting to evaluate where the population pressure is likely to affect the forest.

Study area and material

The present study was carried out using global coverage GIS layers limited the analysis to the Central Africa Region. This includes the following countries Cameroon, Central Africa Republic, Congo, Equatorial Guinea, Gabon and Zaire.

The idea of attempting an accessibility analysis become feasible few months ago when EROS DATA CENTER released the new global 30 arc seconds (approx. 1 km resolution) D.E.M. GTOPO30. This linked to other global digital datasets like the Digital Chart of the World and the Global Land Cover Characterization Dataset, represents a formidable toolkit for global studies related to forestry applications.

Aside these wall-to-wall spatial databases a number of selected multitemporal observation (full Landsat scenes) were available for the region (see next map) These were used to assess the reliability of different accessibility maps.

The hypothesis

The hypothesis to be tested was the following:

People convert forests in other land -use covers due to two main reasons:

� There are forests available and either the forests or the land are of good quality for clearings. Obviously the forested lands accessibility is a key factor to determine the convenience in selecting on place or another.

� People need to either convert forested lands to other, more profitable, land-use practices or log them because of the value of the trees. This value can be determined in term of energetic need (fuelwood) or trade income (commercial logging). The weight of these needs is determined by a number of factors, mainly related to the socio-economic context. Large areas of the Brazilian Amazon have been converted from forest to grassland to support cattle grazing. Shifting cultivators in Tropical South East and Insular Asia rely on burn and slash practices for their villages agricultural sustainability. Trees felling can be caused by commercial or self-sustainability needs. The demand of forested land is, in both the cases, influenced by their accessibility. It is reasonable to think that people move from their villages/towns to convert forested lands into other land-use cover types. It is realistic to assume that this action is not isotropic but will take place along preferential paths. These were considered, in this study, to be function of the road network and the landscape morphology.

First phase Modelling

The model tested in the Central Africa study is based on a cost distance algorithm available within the ARC/INFO GIS Package. Two surfaces were generated independently from two datasets urban settlements and roads. These were used in the final cost map as weighting factors so that the total cost increases when people move far away from both the towns/villages and roads.

Step-1- a slope map was generated from the D.E.M. map expressed as percent rise.

Step-2- a first raster map was generated as the least accumulative cost surface resulting from the function sum cost* distance from villages/town) estimated from the villages\town mapped in the DCW. The resulting map looks like a D.C.m. where the peaks (white) indicate high cost regions and the valleys (black) low cost regions centered in the urban settlements.

Step-3- a second raster map was generated as the least accumulative cost surface resulting from the function sum (cost* distance-from-roads) estimated from all the roads mapped in the DCW.

Step4- The final cost map, shown below, was produced combining the two maps obtained during step 1 and step2 with the following formula:

Final cost = int(ln(cost-urban*cost-road) )+1

Accessibility vs. Deforestation

In order to assess the value of the accessibility map information derived from multi temporal remote sensing data was used. 23 sites in the Central Africa region were analyzed in the past year with the scope of monitoring the processes of change in the Central Africa forests.

For each site, one of the final outputs was a change map depicting 14

9 out of the 100 possible) categories of land cover changes. Two of them were relevant for the present study) DEFORESTATION and b) STABLE FOREST.

The first category was defined as change of land cover from forest (closed, open or long fallow.

With the change maps available within the GIs it was possible to overlay each of the 23 sites with the accessibility map. This process produced individual as well as a cumulative partial/statistical breakdown of the DEFORESTATION and STABLE FOREST category by accessibility belts.

Ranking forested areas on the basis of accessibility belts could be a viable approach for surveys based on sampling techniques where stratification, oriented to minimize the errors in estimating deforestation, could improve the final results.

These findings, if confirmed by similar studies in other regions, could be useful in setting quantitative parameters to the likelihood of the forest to be depleted by mankind.

Second phase Modelling:

The first phase modelling represented a preliminary exploratory analysis oriented to evaluate in a semi-quantitative way the relationship between accessibility and deforestation. The initial results authorized a more in-depth validation phase, which was carried out analyzing more than 100 satellite scenes throughout the Tropics. The accessibility map was hence produced to cover most of the tropical countries.

The analysis, performed on a sub-regional basis, clearly indicates with different threshold, deforestation patterns and accessibility surfaces are highly related.

Applications

The preliminary analysis indicated the potentiality of the accessibility map as indicator human induced disturbance on forests. An application of this map was the assessment of forest volume densities in the Sub-region 44 (Continental South Asia) where it was found that observed volumes could be predicted using climatological indicators (Weck's Productivity Index) and disturbance/degradation indicators (accessibility).

Accessibility belts were grouped into 3 major zones to highlight, within each of the original TRESS category, areas where the combined presence of human beings (indicated by town/ villages and roads) and favourable topographic conditions can be interpreted (and, on a sampling basis, validated by high resolution satellite observation) as forest zones at high, medium and low risk of deforestation/degradation.

SECTION 2: The Area Production Model (APM)

by Walter Antonio Marzoli, Forest Biometrician/Ecologist, FAO

A CASE STUDY: ADILABAD

Example of a case study developed for one district of Andhra Pradesh - India

1. Introduction

The APM is a simulation model for long-term land use changes and prediction of yields from agriculture and forestry. APM aims to offer the possibility of harmonising the use of sectoral inputs and balancing supply and demand through the simulation of consequences under various conditions (scenarios). Such model can be used by decision-makers at the policy level to examine possible alternatives through identifying consequences of various strategies under given conditions.

A case study has been developed for the district of Adilabad using APM in order to show one example of practical utilization of the data collected described in Part 1. The data required for the APM are the following:

The data set required for APM has been collected for all district of Andhra Pradesh (see Appendix 1), only minor adjustments are needed to prepare an input data set for a given district. The case study for Adilabad was worked out as example.

In the context of Andhra Pradesh the APM can be used for the forecast of the following variables:

Estimates the demand for new agricultural land and the resulting deforestation.

Project the supply and demand of biomass for energy and their balance. This is tightly related to forest degradation since the biomass deficit is mainly filled by over exploitation of natural forests.

Simulate the development of new forest plantation and their impact on the future energy balance.

The Adilabad case study was carried out using alternative development scenarios. The reference scenario chosen is based on the assumption of no significant changes in development patterns. Basically the observed development during 1981-91 has been projected over a 30-year period (1991 to year 2021). This basic scenario will be used as a reference to compare the results of several development alternatives.

The data used as input were extracted from the Districts Information Summary listed in Annex 1 and organized into standard input form for APM as described in the next paragraph.

2. Input data

Input data set for basic scenario: APM run 00

STATE = ANDHRA PRADESH DISTRICT = ADILABAD

STARTING YEAR: 1991

NO OF FIVE- YEAR PERIODS: 6

TOTAL AND RURAL POPULATION (*1000): 2083. 1601.

TOTAL LAND AREA (HA* 1000): 1644.

DISTRIBUTION OF TOTAL LAND AREA (HA*1000):

AGRICULTURE LAND

SUBSTANCE FOOD 356

MARKETED CROPS 262.

TOTAL 618.

In Adilabad subsistence food crops include Rice, Millets, Other cereals and pulses. Marketed crops are Cotton, Groundnut, Seasamum, Castor, Sugarcane and Tobacco.

FARM FORESTS

NATURAL FORESTS 74.0

TOTAL 74.0

The area under this categories represents forest blanks (i.e. designated forest land not actually covered with trees), assuming that part of this areas should be available for Joint Forest Management.

INDUSTRIAL FORESTS

NATURAL FORESTS 627.0

`PLANTATIONS 22.0

TOTAL 649.0

Area of natural forests and plantations.

OTHER LAND

POTENTIAL AGRICULTURE LAND 23.00

POTENTIAL FOREST LAND 179.00

UNPRODUCTIVE LAND 101.00

TOTAL 303.00

The culturable waste land was classified as potential forest land while agricultural fallows where classified as potential forest land (for afforestation). Unproductive land refers to barren and other lands.

1*** INPUT SPECIFICATION SET NO 2 ***

GROWTH FACTORS PER YEAR FOR 6 FIVE-YEAR PERIODS:

TOTAL POPULATION 1.025 1.024 1.023 1.023 1.023 1.022

RURAL POPULATION 1.019 1.018 1.017 1.017 1.017 1.016

GROSS DOM PROD 1.000 1.000 1.000 1.000 1.000 1.000

PROD SUBSIS FOOD 1.006 1.006 1.006 1.006 1.006 1.006

PROD MARKET FOOD 1.006 1.006 1.006 1.006 1.006 1.006

The total population growth during 1981-91 was 2.5%, in this basic scenario a slight decrease in population growth, from 2.5% to 2.2% in year 2021, is assumed. The rural population growth will be 1.9% and 1.6% respectively.

In the present run the development of GDP was assumed to be stable in absence of reliable data and also due to some inconsistencies in the APM.

The agricultural productivity growth was purposely kept to the level of 1981-92 (+0.6% per year) for the whole simulation period.

ENERGY DEMAND (GIGACALORIES/CAPITA/YEAR) AND CHANGE/YEAR

RURAL POPULATION 2.68 1.00

URBAN POPULATION 2.29 1.00

The energy requirements for rural and urban population were carefully estimated using several data sources. The estimate accounts for the reported daily energy consumption and the consumption patterns like the share of commercial fuels and the biomass sources. In particular in the district the animal residues (cow dung) are generally marketed as source of income and not used for energy production. Therefore fuelwood and agricultural residues are the main sources of energy, besides a marginal share of commercial fuels (Kerosene, gas etc.), especially in urban areas.

AUTOPRODUCTION (M3/HA) BY LAND CLASSES WITH 3 SUB-CLASSES EACH:

AGRICULTURE LAND . 2 . 2 . 0

FARM FOREST LAND . 2 . 0 . 0

INDUSTR FOR LAND . 6 1.0 . 0

ENVIRON FOR LAND . 0 . 0 . 0

OTHER LAND . 2 . 2 . 1

AUTOPRODUCTION (TON/HA) OF RESIDUES ON 3 CLASSES OF AGRICULTURE LAND:

.2 .3 .0

Autoproduction refers to the production of biomass (wood or resident) suitable for energy as a by-product of the principal crop. Autoproduction includes (i) wood production of scattered trees and shrubs on agricultural and other land including forest blanks and ( ii ) production of agricultural residues suitable for energy (cotton sticks etc.). It also includes (iii) the production of biomass of natural forests and (iv) the fuelwood extracted from timber plantations.

It is difficult to give correct estimates of autoproduction since very few studies exist on the subject. Using the available literature the wood production on agricultural land was estimated at 0.2 m3/ha/year. The yield of natural forest was roughly estimated at around 1% of the biomass or 0.6 m3/ha/year. The production of residues is proportional to the crop yield and it is estimated between 0.2-0.3 tons per hectare.

Finally the priority for land use transfer were estimated as follows:

PRIORITY ORDER FOR TRANSFER OF LAND TO AGRICULTURE

1. OTHER LAND, POTENTIAL AGRICULTURE LAND

2. FARM FOREST LAND, NATURAL

3. OTHER LAND, POTENTIAL FOREST LAND

4. INDUSTRIAL FOREST LAND, NATURAL

5. ENVIRONMENTAL FOREST LAND, NATURAL

The demand for new agricultural land will be met first by culturable waste land and then by natural forests, when potential agricultural land is over.

3. Simulation results (basic scenario)

3.1 Land use changes

(all areas in thousands of hectares)

The table above shows the expected increase of agricultural land obtained from culturable waste land for the first 4 year and afterwards the demand is met by converting natural forests to agriculture. The natural forests to agriculture. The natural forests would decline to 627 000 ha in 1991 to 508 000 in year 2021 with an annual deforestation of 4000 ha.

The results are just indicative and should be used for further investigation, but are still significant to demonstrate the existing trends.

3.2 Energy Balance

The second issue of APM deals with present and projected biomass for energy balance. The balances are calculated using Teracalories (calories * 10^9) and can be converted into cubic meters of wood equivalent using a conversion factor of 2.6 gigacalories to one m3 of wood. The balance between demand and supply shows a deficit of 1 300 000 m3 of wood equivalent in 1991, rising to 3 300 000 m3 in year 2021. This gap is mainly filled by over exploitation of natural forests through illicit fellings. If the amount of wood required is taken from natural forests the degradation rate can be estimated at 2 m3 per hectare in 1991, rising to 6 m3/ha in 2021.

The projected demand and supply are as follows

All columns in teracalories except last column in million m3

4. Development scenarios

4.1 Land use changes

That approach to land use changes is relatively simple. Adilabad district suffers from a chronic lack development. It has traditionally been a `forest' district with several peculiarities in socio-economic and ethnic characteristics. In the course of time the high speed of population growth has led to an increasing demand for agricultural land and in most cases the land converted into agriculture was carrying forests on medium to steep slopes. The hazards related to soil denudation and erosion as well as habitat destruction are well evident in the district.

In order to balance the growing pressure of population it is necessary to improve the yield and the production of agricultural land. Adilabad rank last among the Andhra Pradesh districts for food crops productivity with only 602 kg/ha compared to over 2300 kg in Hyderabad district. It is interesting to note that even drier zones of the State (e.g. Ananthapur and Mahabunagar districts) achieve higher yields. The soil and climate conditions are generally favourable for agriculture in Adilabad and a consistent increase of land productivity and socio-economic development should be established as urgent targets. According to the basic scenario an increase of 2.5% per year would be sufficient to balance the growing population pressure. Such increase would lead to a productivity level of slightly more than double in the next 30 years. This appears to be feasible considering that the present level is very low as shown in the next table.

4.2 Energy balance

As described earlier the projection of energy balance under present conditions show a very serious negative trend. Even today the pressure on the natural forest is well evident and according to the simulation results it is estimated to become three time higher over the next 30 year. There is a considerable risk of heavy growing stock reduction and land degradation and there is an urgent need for corrective measures.

In order to estimate the effects of the possible corrective measures some alternative scenarios were formulated and tested. The development scenarios for energy balance are more complex than for land use changes. The necessary actions can be subdivided into (i) demand side; and (ii) supply side.

Among the possible scenarios the following have been tested:

DEMAND SIDE

d1. Decrease in population growth from 2.5% to 2.0 in year 2021

d2. Decrease of per capita energy consumption due to increased energy efficiency and to use of alternative energy sources. In this scenario the annual decrease of per capita demand was estimated at - 1% per year, or a reduction of 26% by year 2021

d3. Same as d2 but with a more consistent decrease in per capita demand, set to -2% per year or a reduction of 45% by year 2021

d4. Combined effect of d1 and d3

SUPPLY SIDE

s1. Improvement in wood production on agricultural land through tree planting and agro-forestry. The estimated wood production from agricultural land was expected to double from 0.15 m3/ha/year to 0.3/m3/ha/year.

s2. Development of fuelwood plantations. A total amount of 100 000 ha of fuelwood plantations established over 30 years is simulated in this scenario. The selected target is only indicative and should be formulated together with APFD, keeping in mind the financial and technical constrants. The land designated for reforestation is represented by forest blanks and from degraded (open) forests. The species to be planted should be high yielding. AMAI of 10 m3/ha was assumed, given the relative site fertility and presuming a suitable species selection, with a rotation period of 20 years and 4 coppicing cycles 1 every 5 years.

s3. Combined effect of s1 and s2.

The individual and the combined effect of each scenario were tested through successive runs of the APM program. In total 20 combination were tested and for each scenario the projected fuelwood deficit was computed. The results are tabulated thereunder in a matrix format where each cell represent the combined effect of two actions, one on the demand and one on the supply side.

The analysis provides 4 main scenarios i.e. present conditions (00), maximum effort on the supply side only (04), maximum effort on the demand side only (30), maximum effort on both supply and demand (34).

Scenario	Supply side	Demand side
00	No change	No change
04	1. Improvement of wood production on agricultural land + 2. Establishment of 100 000 ha of fuelwood plantations	No change
30	No change	1. Reduction of population growth rate + 2. Reduction of 2% per year of woody biomass demand per capita
34	1. Improvement of wood production on agricultural land + 2. Establishment of 100 000 ha of fuelwood plantations	1. Reduction of population growth rate + 2. Reduction of 2% per year of woody biomass demand per capita

From the table above the following consideration can be made;

- The absence of corrective measures will lead to a very severe degradation of the existing forest resources.

- Corrective measures are necessary from both sides i.e. demand and supply

- The comparative analysis of the scenarios shows that the proposed development on the demand side are the most effective. Even in absence of actions on the supply side the projected deficit is of 1.28 million m3 while which is less than the maximum result on the supply side alone (-2.20 million m3). In addition the actions on the demand side are likely to be more affordable: the targets of per capita demand reduction (-25% and -45% respectively) are within the range of the expected improved energy efficiency by use of closed chullas, coupled with the increase of alternative sources (e.g. bio-gas). According to the simulation results this is the most effective measure.

- The present output from forests is very low compared with their extent. Efforts in improving the wood production is needed to prevent further forest degradation and destruction.

5. Conclusions and recommendations

The case study of Adilabad shows the potentiality of APM as a tool for development planning. Using the standard data collected for all districts it was possible to identify the present trends for the district and to quantify the effects of some alternative scenarios. This should be useful to the planners and policy makers as a tool for decision making.

In particular the main conclusion for Adilabad can be summarized as follows:

- At present both deforestation and forest degradation are occurring in the district. Three is an urgent need of corrective measures to avoid a severe destruction of the forest resources in the near future.

- Deforestation is mainly due to the rural population growth combined with a substantial stagnation of agricultural productivity. Concerted actions between forest and agriculture department as well as other related development agencies are needed to reduce the demand of new agricultural land through improvement of agricultural crop yields.

- Forest degradation is the consequence of the energy deficit. Fuelwood is the main source of energy in the district and it is being widely used by both rural and urban population for cooking. In the absence of alternative sources fuelwood is collected by the villagers in the natural forests and also marketed to the towns. The disproportion between demand and supply is causing an unsustainable pressure on the remaining natural forests.

- While analyzing the possible actions to preserve the natural forests it was found that the most effective measures are to be taken on the demand side. The first priority is to identify and promote appropriate technologies that improve the energy efficiency. The solutions must be simple enough and well adapted to the local cultural and socio-economic environment. A considerable effort is justified in this direction since an increase of 50% in the energy efficiency over the next 30-year is essential to re-equilibrate the fuelwood balance.

-Besides the efforts on the demand side it is also necessary to intervene on the supply side. According to this preliminary simulation a target of 100 000 of fuelwood plantation is required over the next 30 years, or 3 300 ha per year.

The conclusions above are based on the analysis of a few scenarios selected by the consultant according to his experience and after discussion with forest Officers in Andhra Pradesh. The main effort here was to collect and organize data in a consistent way so that they could be used for modelling. Many other development scenarios can be built and tested using different assumptions. Moreover the APM can be applied to any other district of Andhra Pradesh. The main goal of the present case study is to stimulate the Forest Department to use APM as a tool for strategic planning.

APM has been used here at district level. The next step is to collect and organize more detailed data for running the model at mandal or forest division level. Ideally APM should be used at both levels (district and division/mandal). Data are being collected and organized to fulfill this objective.

District: ADILABAD

Average annual rainfall (mm): 1149 Area (thousand ha): 1612.8

POPULATION: (thousands)

FORESTS: (areas in thousands of hectares)

LANDUSE: (areas in thousands of ha)

AGRICULTURE: F00D CROPS (area in percent of total sown area)

AGRICULTURE: NON FOOD CROPS (area in percent of total sown area)

AGRICULTURE PRODUCTIVITY:

Average yield per ha of food grains: 527

FOOD CROPS: (yield in kg/hectare of principal crops)

NON FOOD CROPS: (yield in kg/hectare of principal crops)

LIVESTOCK: (number of heads in thousands)

SECTION 3: Forest Resources Assessment

by Shri S.K.Chakrabarti, ISS, Forest Survey of India, Dehradun

Introduction

The only source of assessing the growing stock in our country is the working plans at the divisional level of various state forest departments. The forest Survey of India had been conducting field inventory in the forest areas of the country ever since its inception as PISFR in 1965. PISFR has conducted field inventory in the selected areas covering 23 million ha upto 1981. After the upgradation and reorganization of PISFR as FSI in 1981, field inventory of the entire forest area of the country was included in the mandate. Since then an additional area of 45 million ha has been inventoried constituting about 80% forest area of the country. But there has been no attempt in the past for determining the growing stock of forest resources at the national level.

Methodology for Growing Stock Assessment at the Country Level

State/UT wise growing stock for the entire country (major forest strata-wise) was first assessed during 1995 as a part of National Forest Action Plan (NFAP), a project sponsored by FAO, using information available from the vegetation maps (density) thematic maps (forest types) and the ground forest inventory (average vol./ha) done by the FSI. For this purpose the maps of 1:50,000 scale of all the states and Union territories (I) density, (2) forest types and (3) average vol./ha for each combination of (density, type), were divided into grids of 2.5' x 2.5' (lat.x long). This exercise yielded information on the extent of forest cover, composition (21 species strata), density (3) and inventory data on growing stock. The following information for each grid was collected:

a. Density

The landuse category occupying more than 50% area of the grid was taken into account e.g. if more than 50% area is forested, the grid was marked as forested otherwise `non-forested'. Then in each forested grid only one major density class was taken as the density class of the entire grid. This information was collected from the vegetation maps prepared by FSI on 1:250,000 scale based on satellite imageries interpretation for the period 1989-91. Each grid was classified on either of the following three density classes;

Crown density Code

1. Very dense forest 70% and above D1

2. Dense forest 40% to 70% D2

3. Open forests 10 to 40% D3

Areas, which were having, less than 10% forest density were treated as scrub. Grids falling under non-forest and scrub were not used further. Those grids, which were spread over in more than one state, it was included in the state in which its maximum area fell.

b. Forest types/strata

The major forest type/strata in each grid was marked using information from the

following sources.

Thematic maps prepared by the FSI on 1:50,000 scale using aerial photographs were used for marking the major species composition of each grid.

For the areas for which thematic maps were not available in FSI, the information on species composition was collected from the stock maps of the State Forest Department

For the areas for which neither thematic maps nor the stock maps of State Forest Departments were available, the information on forest types was collected from the inventory field forms in which this information is recorded for each survey.

In case of areas for which none of the above sources of information was available on species composition, the information of adjoining areas was taken into consideration by taking into account same agro-ecological zone/sub-zone

The forest types obtained were:

1. Fir 11. Sal

2. Spruce 12. Bamboo (tree crop in bamboo)

3. Fir-spruce 13. Dipterocarpus

4. Blue pine 14. Khasi pine

5. Deodar 15. Khair

6. Chir-pine 16. Salai

7. Mixed conifers 17. Alpine Pastures

8. Hardwood mixed with 18. Miscellaneous

conifers

9. upland hardwoods 19. Western Ghat Evergreen

10. Teak 20. Western Ghat Semievergreen

21. Western Ghat Deciduous

c. Forest Inventory Data

Sampling Design - Forest Area

A national inventory design is followed for the whole country under which the SOI toposheet of 1:50,000 scale is divided into 2.5'x2.5' grids, each grid representing about 18-20 sq.km area. In each grid, two plots of 0.1 ha each are marked on the toposheets. The plots falling within the forest areas are inventoried. The inventory field form also gives the forest density and forest type under which the plot falls. From the inventory important parameters like growing stock, vol./ha etc. are estimated. The data of forest inventory done by the FSI in various states was used for determination of average vol./ha for each combination of (Forest density, Forest type) in each mapsheet at 1:250,000 scale.

Estimation of growing stock

In the present methodology, for estimation of the growing stock of any state, the number of grids for each combination of forest density and forest type was calculated. The areas of i.e. 1800 multiplied the number of grids ha to arrive at total area under each mapsheet for the same combination. Then the area was multiplied by the average vol./ha of the same combination. The growing stock was first estimated for each mapsheet and then summing up all the data in the mapsheets for each state, the total growing stock of each state was arrived at. Wherever the vol./ha was not available, it was borrowed from nearby areas considering the agro-ecological zones (i.e., areas falling under the same agro-ecological zone was taken into consideration).

Growing Stock

Adopting the methodology as described in the previous chapter, the Growing stock and the area under various density classes and forest stratum for various States/UT's has been worked out. The State/UT-wise details are given in Table 1 forest typewise growing stock is given in Table 2. The Growing Stock of the country is estimated to be 4740.858 million Cu. M which works out to be 74.42 Cu. M per hectare of the vegetation cover.

0

1 Resource Assessment: State of Arunachal Pradesh

1:250,000 sheet No. 78 M Quadrant No. 9 (1:50,000)

Table 1: Extent of Forest cover and Growing stock in India

Note: *-The volume of mangrove forests has been estimated separately.

@- The growing stock is insignificant.

Table 2: Distribution of growing stock over various forest strata of India

For the GFRA - 2000, it is proposed to estimate the area as well as growing stock under three major forest types

a. Conifers

b. Broad leaved

c. Bamboo

The growing stock assessed in 1995 as per the above three categories is in Table 3.

Table 3: Statewise Growing Stock by Broad Forest Strata and Density

(million cu.m)

Note: In the above Table Dense = D1 + D2, Open = D3

Annual Increment

Apart from estimating the growing stock of the country, an attempt was also made, in absence of any precise estimates, to have an idea of the annual increment of India's forests by using Von Mentals' formula (I=2 GS/ r, where GS= growing stock of the forest type and r = rotation year). Since the growing stock of various states/UT s has been calculated for different forest strata and there are various species which fall in one stratum, a general period of rotation was adopted and the increment was calculated on the basis of Von Mentals' formula given in Table 3 & 4 below:

Table 4: Distribution of Annual Increment over various states/UT s of India

Table 5: Distribution of Annual Increment over various forest strata of India

Followup of the last workshop of FAO-FSI

During the October workshop, this methodology was discussed in detail. It was agreed that the same methodology will be followed for the current assessment of forest cover i.e. SFR-1997 (based on satellite imagery of 1994-95) so that a picture of degradation of forest may emerge by comparing statewise and countrywise growing stock based on SFR-1993 and SFR-1997.

The work of assessment of the country is in progress. As a pilot study, the growing stock of one state i.e. Manipur has been completed based on same volume factors, which was used in 1995 assessment. The following tables 5 & 6 give a picture of the area of growing stock figures for two assessments.

Table 6: Comparison of Forest area in Manipur based on two assessments of SFR-1993 (satellite imagery of 1989-91) and SFR-1997 (satellite imagery of 1994-95)

(000 ha)

Table 7: Comparison of Growing Stock in Manipur based on two assessments of SFR-1993 (satellite imagery of 1989-91) and SFR-1997 (satellite imagery of 1994-95)

(million cu.m)

SECTION 4: Workshop on country capacity building in Forest Resources Assessment in South Asian Countries

by Ishtiaq Uddin Ahmad, Deputy Conservator of Forests

1. Introduction

Information on the status of the resource is the basis for its scientific management and sustainable planning. Forest inventory helps in collecting, collating and analyzing data on various parameters. Forest inventories at national, sub-national, pre-investment and operational levels are a pre-requisite for sound forestry development planning and for rational use of goods and services provided by the forests. Lack of reliable data on forests and land resources and lack of modern skills have hindered the formulation of sound forestry polices, strategic planning and sustainable management practices. The biotic and a biotic influence on forestry necessitates close monitoring of the forest resource, updating of information periodically and reiterative planning.

Forest assessment is done to capture information on various parameters which influences forest growth and management. This information base thus created helps in deciding adoption of management's options like silvicultural regime, tending operations, logging, transport or the like. Today forest management embraces, more than just stand information, which includes environment and social parameters. In context to tropical forestry in particular one can not think of forestry in isolation of the people or avoiding the issues relating to environments. Integration of these information into the forest management plan is very important. The idea is to bring about harmonized and balanced management planning towards participatory approach. In such planning, one must, on one hand takes into consideration the traditional forestry production targets, on the other hand, one also considers peoples involvement and social impacts.

So, forest management at present context needs a holistic approach. Timely and quickly data acquisition is important. Remote sensing now a days integrated with Geographic Information System (GIS) is a useful tool to face this problem. A wide variety of sensors is available with different spectral characteristics, geometric resolutions and periodicity. The advent of computer has revolutionized the databased management system. The combination of remotely sensed data and GIS gives not only an actrual images of the earth surface but may provide also a detailed base for planning management.

2.1 Historical background

In this part of the world first methodical inventory was done in Sundarban (Mangrove) forest in 1930 by a British Forester Mr. Curtis. He established transect lines all over the Sundarban forest. Finally he produced a forest cover map at a scale of 1:63360. After 1930 the forest was again inventoried in 1958 by Canadian forestal forestry. Aerial photographs were used in this inventory. Along with the stock and stand table a forest cover map of 1:15,840 was developed. This inventory was mainly to study the feasibility of establishing a newsprint mill be using Excoecaeria agallocha (Gewa) wood. The present news print mill at Khulna is the outcome of the 1958 inventory proposal. Forest department developed a working plan for a period of 10 years by using the result of that inventory.

Almost at the same time the Forestal Forestry of Canada undertook the inventory of the Chittagong Hill Tracts region which totals 0.67 million ha of forest land. Detailed forest composition had been interpreted and forest cover map at a scale of 1:15840 was developed based on aerial photographs of 1:15,000 (approx). A contour map and a land classification map were also developed for the same region.

After that, with the financial assistance from the UNDP/FAO the Forest Department with the collaboration of FAO experts carried out inventories of different forest areas including a countrywide inventory of village forests to assess the village wood resources of the country. The inventories under the auspices of UNDP/FAO helped to build the country capability in the field of inventory by improving the skill of the manpower and logistics. For the first time in Bangladesh the status of the village wood resources has been determined through a detailed inventory in 1979. The inventory helped to determine the quantum of wood the village forest provided to the people beside the wood resource comes from states owned forest lands. This gave an opportunity to undertake different developmental programmes to improve the village wood resources of the country.

Cluster plot sampling has been followed so as to capture changing conditions within a stand/landscape unit. A cluster of 5 plots in a cross arrangement oriented along the cardinal directions with a centre plot and the other four plots situated 100 m (in some cases 50 m ) north, east, south and west of the centre plot, respectively. Each plot in a cluster consists of a few sub-plots.

2.2.2 Forest Resources Statistics & Database

Computer based data entry and validation program and field data processing

program were developed specially for this inventory. Data processing to generate the required statistics and other information needed in integrated forest management planning have been developed.

The final inventory report includes detailed stand and stock tables for trees by DBH class, species area and volume per hectare and for the whole Division as well as by stratum, forest type, compartment, block and range. The detailed statistics on seedlings saplings and poles as well as nypa and bamboo are also included.

2.2.3 Other data base

At the same time socio-economic survey has been conducted in the target forest divisions. This survey is to determine peoples attitude towards forestry, demand of forest produce, dimension of demand, number of population around forest reserves, rate of deforestation, rate of encroachment etc. These are all to develop an integrated management to arrest further degradation of the forest resources.

Besides socio-economic survey special inventory has also been conducted for the conservation areas. This survey has been conducted to determine further candidate areas to qualify to be declared as protected areas. Through this survey a data base has been developed on different parameters showing status of flora and fauna.

Following that, another UNDP/FAO forestry project along with a strong component to carryout inventory of the Hill forest & coastal plantation was launched in 1981. That project was entirely implemented by the working plans division of the forest department with the expert assistance from FAO. The inventory of these forests were carried out by using aerial photographs and SPOT satellite image in the area where photographs were not available. In 1984 Sundarban forest was again inventoried with the assistance from British ODA by the help of aerial photographs.

2.2 Forest Inventory of Recent Time

Forest department from 1995 to 1997 carried out a massive inventory program in the Hill & Coastal forest areas including Sunderban Reserved Forests under a loan agreement with International Development Agency. The inventory was carried out for forest resources management planning purposes.

2.2.1 Objectives and design of the inventory

The primary objectives of the inventory are threefold. The first to generate

information on the status of the forests of the target forest divisions and the second primary objectives is to provide "abstract" time series data for yield modelling purposes. The third major objectives is to set up a system of hidden recurrent same plots for continuous monitoring and assessment of change in the target forest areas. Such information will be used for forest resources management planning purposes taking into account not only the timber/wood resources but also other important resources like bamboo, rattan, nipa, medicinal plants, wildlife and tourism as well as social and environmental considerations.

Systematic sampling is used in all the target divisions. The sample plots were laid following a grid system. In the hill forest grid interval of 40" by 40" is used mostly in combination with smaller grids 20" or 10" to cover required number of plot clusters to attain the target precision of estimates. In the case of Sundarbans, a systematic sample of one minute by one minute grid is used. The standard error of the estimated total volume of trees in each of the management units should not exceed 5%.

Database and associated tabular database is entered into the computer, it can be used as tool for management purposes such as production of thematic maps, area calculation visualization of different scenarios etc.

3.2 RIMS-GIS-A scope for integration

It was mentioned earlier that RIMS database deals only with the stand related information based on mappable management units. The restriction with the RIMS database is that the link to the mappable units-normally forest sub-blocks-is not available in a suitable computerized form.

The combination of RIMS with a suitable GIS would provide the missing link between database and maps. The integration could enhance the capabilities as follows:

- enhance mapping process and map products by inclusion of non-forest features like topography, infrastructure and administrative boundaries, based on a consistent projection system.

- Allow up-to-date map presentation of current state of the forest.

- Enable simulation of different management scenarios and present them in map form.

- Allow inclusion of non-forest data, which is of importance for forest planning and management (e.g. thana-wise or village-wise demographic data, socioeconomic data, location and characteristics of consumers and markets for forest produce etc.).

- Produce forest maps on management plan level, which are suitable for the daily work of the forester at Range and Beat level.

- Produce thematic maps on higher level (e.g. division level), which can serve as overview map for general decisions and forest policy.

- Provide means for a reporting system, suitable to keep records at FD HQ up-to-date in order to provide them with a decent data and map base for forest policy and forest management decisions.

In order to facilitate the operation, maintenance and analysis of the comprehensive RIMS-GIS database, a Graphical User Interface (GUI) has been programmed. This is a mouse-oriented, user-friendly interface, which makes the

All these data bases are available in digital form as .dbf files which are consistent with the RIMS-GIS data base, to facilitate to be integrated into RIMS-GIS.

2.2.4 Continuous Forest Assessment

In the previous years forest inventories of different types were done for once only. There were hardly any scope to establish linkages with the inventory of the subsequent years. Therefore, it was difficult to assess absolute changes in certain strata of forest compositions. Though overall comparisons were done but degradation of specific types were never addressed properly. The design of this inventory strongly recommends to establish a "Continuous Resource Change Assessment System (CRCAS)". The basic components of CRCAS have now been set in place at RIMS-GIS. Out of total valid inventory plots annual measurement of one-tenth to one-fifth of the grid plot clusters so that all plot clusters would have been revised in five to ten years in revisited divisions.

3. Resource Information Management System

3.1 The Management System & GIS

The Resource Information Management System (RIMS) was originally established under a World Bank financed project. The system consists of combination of dBASE III command files and BASIC programs. It is designed for general wood resources management planning, country wide wood supply projections and strategic planning. The current database deals with stand related information including stand growth and yield functions and some silvicultural prescriptions.

GIS the abbreviation stands for Geographic Information System is a system of hardware & software which is designed to support the "Capture, Management, manipulation, analysis, modulation and display of spatially referenced data" in order to solve complex planning and management problems. GIS provides the means to capture of a certain area with all kinds of spatially related data for that area (e.g. data on forest types and growth, administration, demography, socio-economy, environment, infrastructures etc.). Once this combination of map system operation for the user much more comfortable, although this would in principle also be possible by using the standard functions of the software installed in the RIMS-GIS system. The GUI consists of a number of programs in the programming language Avenue (a programming language that comes with ArcView) and of some programs done in Visual Dbase.

This system also should integrate the functionality of the previous RIMS tabular database program and of the Environmental Information Management System (EIMS) and socio-economic data, which is also based on tabular data.

The general concept and interaction of RIMS-GUI with other important project components is depicted in the following figure.

Main components for data input are the forest inventory, the aerial photo interpretation and various map sources (e.g. forest cover maps, topographic maps, maps showing administrative boundaries, thematic maps etc.). On the other side, the system delivers outputs for those components, e.g. support for the stratification in the forest inventory, base map for transfer of updated photo interpretation results etc.).

The directives for RIMS-GIS have to come from the FD management, whereby the term `directives' under these circumstances stands mainly for general management rules and prescriptions.

� The General concept & interaction of RIMS-GUI.

The RIMS-GUI was programmed to fulfill the following tasks and outputs:

� Flexible use of the geographic data bases (e.g.enable overlays, intersections, buffering etc. of any spatial data base with any other spatial data base available in the system)

� Easy-to-use combination of geographic data and tabular data

� Allow selection of areas of interest (reporting or planning areas) in several ways:

� Based on administrative boundaries (e.g. Division, Range, Beat).

� Based on free-hand polygons defined by the user (simply by indicating the area of interest in a map on the screen).

� Enable data entry and data editing.

� Allow output of maps, table, charts and summarizing statistics.

� Allow scenarios and predictions.

� Easy-to-use updating/editing of necessary inputs as volume functions, yield tables and management prescriptions.

� Easy-to-understand graphical user interface.

�

4. Application of management information and planning system

This management system aims at better management planning, decision making and performance using a system of computerized geographical map information, GIS, traditional forestry data and also information on natural resources integrated with environmental and socio-economic data. The integration of the following variables are evident in RIMS-GIS e.g.

� Area size of different landuse, forest types, number of trees by species

� Silvicultural treatment

� Forest inventories for management purposes

� Socio economic data on settlements within or near Forest Reserves

� Environmentally sensitive information

Rims- GUI is a graphical user interface for RIMS-GIS. It is to facilitate the handling and analysis of above mentioned data base.

Integrated in RIMS-GIS are also programs for :

� Selection of area of interest

� Intersection (overlaying) of different spatial data sets.

� Loading of event themes (e.g. sample plot data, which can be georeferenced)

� Generation of buffers around polygons, lines or points

� Determination of co-ordinates of points of different projection system

� Generation of grids and graticules for map production

� Generation of latitude / longitude point grids (e.g. for planning of sample grids)

� Area calculation

� Volume calculation

� Editing yield tables and management prescriptions

� Generation of maps using standardized formats customized for the need of the Forest Department

� Generation of reports

GIS provides both, simple point and click query capabilities and sophisticated analysis tools to provide timely information to managers and analysis alike. This is not an automased decision making system but a tool to query, analyze and map data in support of the decision making process. The system assists in tasks such as presenting information at planning inquiries, helping resolve territorial disputes, and setting watch tower and check posts in such a ways as to optimize their distribution.

The system helps to reach a decision about the location of a new plantation, that is located on a suitable soil and is close to a population centre. The information can be presented in a shortest possible time in a form of map, tabular data and graphs. The greatest advantage is the capability to evaluate multiple scenarios or options efficiently and effectively

5. Conclusion

Forest management has too often been implemented by short sighted exploitation, independent of other natural resources management regimes and also too often not considering properly all environmental and socio-economic constraints.

The growing importance of forestry demands quick appraisal of forest status, people socio-economic condition and the parameter responsible for the environmental degradation. So forest management at present context needs a holistic approach. For that it is necessary to handle on multiple scenarios together to see their affect individually or interactively.

GIS has already proved to be an efficient management too particularly for the planners and decision makers. RIMS-GIS has been established in the Forest Department of Bangladesh with a great expectation and hope that a dynamism can be brought in the field of forest management of the country.

SECTION 5: CCB - FRA 2000 - The Biodiversity Assessment

by M.Lorenzini, FAO

According to the United Nations Convention on Biodiversity

"biological diversity refers to the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems".

Developing a methodology, to be applied on a global basis and based on the gap analysis experience, requires an accurate definition of the goals and, subsequently, the tools which can lead to the achievement of the same goals.

Scope of the document

The document presented by the author contains the outline of a methodology aimed at organizing a number of indicators, some of them recursively mentioned in earlier CCB-FRA2000 documents, with a "view" of the Biological conservation issue.

It is the author's feeling that, inspite of the many efforts to build-up an organic/systematic proposal clarifying CCB-FRA2000's objectives in the assessment of biological diversity, definitions, roles and integration of various proposed indicators have as of yet not been achieved.

In view of this need and on the basis of past experiences a number of publications were reviewed by the author related to the Gap Analysis, since its conceptual framework is very close to how we foresee the structure of an Ecosystem Report, Gap Analysis was developed in the early 80s and it has been adopted in a number of countries (USA, Australia, etc.) to support conservation programs.

Being the assessment of forest area and deforestation over the last decade the primary goal of FRA2000, the question to which FRA2000 will try to answer is:

"how will the current status and the changes in land-use, measured during the assessment, likely to affect the biological diversity associated to the forest ecosystems"

Adopting Gap Analysis to CCB-FRA needs

Developing a methodology, to be applied on a global basis and based on the gap analysis experience, requires an accurate definition of the goals and, subsequently, the tools which can lead to the achievement of the same goals.

The CCB-FRA assessment of biodiversity can be implemented adapting the Burley's approach to the actual FRA targets only if we take into account the major differences between the framework (biological conservation) within which gap analysis and related information were developed, and a realistic vision of what should be the main objective of the CCB-FRA assessment, say the identification of critical zones in terms of nature/mankind interaction, and the kind of information which can be generated on a global basis useful to achieve the goal.

On the basis of these assumptions the landmarks of the FRA Gap Analysis will be:

a) first identify and classify the various elements of biological diversity

being our current knowledge limited to

� broad eco-floristic zones (efz) delineation and

� broad actual vegetation cover types mapping

a combination of the two datasets should serve to classify and identify the landscape/biological elements to be analysed.

b) using additional information about human pressure on forests (population density/growth, deforestation likelihood, naturalness, fragmentation (?)), further qualify the elements for the analysis into risk categories.

The combination efz-Vegetation can be further classified into risk categories on the basis of

� population pressure indicators such as population density and growth and

� spatial indicators of the ecosystem naturalness such as accessibility (further information about this indicator can be found in the access.doc).

c) examine the existing system of protected areas

limit the analysis to the existing global protected area system as classified by IUCN and stored in the form of digital maps by WCMC.

d) using various classifications, determine which elements (e.g., major ecosystems, vegetation types) are under stress conditions (b) and unrepresented or poorly represented in the existing system of conservation areas (c)

Once a comprehensive database is built on the basis of the above mentioned variables it will be possible to identify, through spatial/statistical queries, areas with features considered critical in terms of ecosystem self-sustainability.

In-depth discussion of the proposed approach

The analysis of the status of forest ecosystems, carried out with the use of geographic information system (GIS), relies on a number of basic layers:

1. GIS layer -

a) Eco-floristic zones, vegetation types

b) administrative units boundaries with statistics, accessibility

c) protected areas.

2. Information content

a) Landscape/Biological elements to be analysed

b) Human pressure indicators (deforestation, population density and growth, naturalness, etc.)

c) Management status.

Landscape/Biological elements

Combining the Eco-floristic zones and Vegetation types maps will lead to a number of unique combinations of actual vs. potential land-cover conditions. EFZs and vegetation maps can be seen as the first two levels (ecosystem and community) of the hierarchy:

Ecosystem - EFZ Ia & IIa

Community - Dense almost primary forests

Guild - Dipterocarpus forest

Species - Dipterocarpus spp.

Organism - -

Gene - -

Following is a brief description of the information content of the two maps :

� Eco-floristic Zones

This data-set integrates ecological with the floristic information. The parameters applied in the ecological classification are ; climatic, physiographic, and edaphic. Climatic parameters include mean annual rainfall, rainfall regime, length of the dry season, relative humidity and temperature. The physiographic contours and the soils map are used to further subdivide the bio-climatic zones. Within the major ecological regions thus established, floristic zones are defined using the dominant or characteristics woody species of the flora, with attention given to their successional patterns.

Eco-regions have been mapped with a consistent methodology across all regions. Maps exist for Africa, Continental South and Southeast Asia, Insular Asia and South America. (Mexico and the countries of Central America are not covered). The floristic zones are continent specific.

� Vegetation cover/condition

The vegetation cover/condition have been defined by using remote sensing images and existing country maps. They are expressions of vegetational characteristics such as density, continuity of plant cover, height, etc. and human impact on the vegetation. The legend includes classes like dense forests, secondary forests, woodlands, thickets, savanna. The vegetation map together with eco-floristic zones, may be used to map the different stages of the regressive series or, in very rare cases, the evolutionary tendency of the vegetation within a zone.

Human pressure indicators

FRA 2000 goal is not so much related to define conservation strategies but to depict the status of forest ecosystems. In view of this, ancillary information have to be integrated with the primary elements described earlier in order to build up a complete overview of the state of the ecosystems.

Due to time and financial constraints potential contributors will be extracted from the current FRA data holdings:

� Population time-series at sub-national level (+ related administrative units map);

� Statistics about deforestation at sub-national level (+ related administrative units map);

� GIS layers related to urban settlements and other human infra-structures (road/railroads);

� A product called accessibility map derived from the combination of the above mentioned layers and DTM;

� The first two parameters, demography and forestry statistics, can be associated to each landscape/biological element on the basis of the areas or fraction of areas of the administrative units reporting such statistics. Data are stored in a statistical database named FORIS: Forest Resources Information System.

� A second group of statistics, associated to each landscape/biological element, is derived from spatial database reflecting, according to existing cartography, the spatial patterns of people distribution and forest disturbance preferential paths on the basis of urban settlements, roads and topography. Each landscape/biological element is hence further described in terms of areas under various levels of likely disturbance (high, medium, low no risk).

Management Status

A world-wide overview of the status of conservation of biological diversity can be obtained using a global database developed by the World Conservation Monitoring Center (WCMC) which, sponsored by the International Union for Conservation of Nature (IUCN) has been collecting, over the past decade, georeferenced information related to protected areas. These are classified according to the IUCN definitions of :

I Strict nature reserve/wilderness area. Protected area managed mainly for science or wilderness protection.

II National Park. Protected area managed mainly for ecosystem protection and recreation.

III Natural monument. Protected area managed mainly for conservation of specific natural features.

IV Habitat/species management area. Protected area managed mainly for conservation through management intervention.

V Protected landscape/seascape. Protected areas managed mainly for landscape/seascape conservation and recreation.

VI Managed resource protection area. Protected area managed for the sustainable use of natural ecosystems.

Following the Kotka III recommendation the information required is protection status of (I) Forest and (ii) Other wooded land could be two main management status categories :

(1) IUCN categories I and II; and

(2) IUCN categories III, IV, V and VI.

Analysis and expected output

A major function of gap analysis is to present, in tabular as well spatial format, statistics evaluating the current status of each landscape/biological element in terms of past/present degradation processes and its current level of protection. Aggregating landscape/biological elements into higher hierarchical levels, say EFZs or Vegetation types, or rearranging statistics on a national and/or sub-national level will allow to generate information which will serve us the precursor to additional analysis designed to in-depth studies for specific areas.

The biodiversity spatial/statistical information system relies on a number of integrated data sources, which, as detailed in the previous chapter, will enable analysts to perform complex queries.

On the basis of such information it will be feasible to select subsets of the main database with critical features e.g.:

� More than 70% of the element falls in the naturalness category "high risk" (topography, urban settlements and roads indicate easy access to the natural resources)

� No protected areas are recorded within the boundaries of the element.

� Most of the sub-national units shows high population density and population growth rate values.

� Deforestation rates are high in all the sub-national units.

The Ecosystem report

Following the proposed guidelines the ecosystem report will become an information system including maps and linked statistics. The report will be fully documented in each of its components (both spatial and statistical database). Worked-out examples will show methods to

� identify,

� map and

� produce statistical reports

Concerning biological elements to be considered at risk.

Additional studies : Study of species-richness

The study will be performed using FRA 1990 approach. The main steps in assessing the risk of biological diversity loss will be the following:

(i) Estimation of the forest area at 1990 and losses since 1991, by ecological region and sub-national unit; and

(ii) Estimation of species richness associated with forest area and risk of loss due to deforestation: both at the sub-national unit based on a model.

It is proposed to carry our special studies for improving the existing knowledge on the extent of biological diversity associated with forests of different types and sizes.

The following groups of available information is proposed to be collected :

(i) Inventory projects where species have been carefully identified;

(ii) Protected area records with reliable inventory of plant and animal species;

(iii) Dated herbarium records providing evidence of historical occurrence of a species.

SECTION 6: Forest Cover Change - Assessment Methodology

by Sandeep Tripathi & P.K.Pathak

OBJECTIVES

� BIENNIAL FOREST COVER ASSESSMENT OF INDIA

� TOTAL GEOGRAPHICAL AREA OF INDIA: 328 million ha.

� TOTAL FOREST COVER:63.34 million ha.(19.26% OF GEOGRAPHICAL AREA)

� VEGETATION MAPS AT 1:250000 SCALE

DETAILS OF SATELLITE DATA

� DATA PERIOD : OCTOBER - NOVEMBER

� DATA SOURCE : NATIONAL REMOTE SENSING AGENCY

� SENSOR : IRS - IB (LISS-II)

� RESOLUTION : 36.25 m

� MEDIA : HARD COPY /

COMPUTER COMPITABLE TAPES (CCT)

I.VISUAL INTERPRETATION

� DATA PROCUREMENT IN HARD COPY SATELLITE IMAGERY

� DRAWING OF BASE MAP ON1:250000 SCALE

� VISUAL INTERPRETATION OF IMAGERY

� COMPARISION WITH PREVIOUS CYCLE VEGETATION MAP

� GROUND VERIFICATION

� DISTRICT WISE AREA ESTIMATION

II. DIGITAL INTERPRETATION

VAX-11/780 DIGITAL IMAGE PROCESSING SYSTEM HARDWARE:

-3 GRAPHIC WORKSTATION

-5 ALPHA NUMERIC TERMINALS

-A0 SIZE DIGITISER

SOFTWARE:

VIPS-32

DIP STEPS

� A. IMAGE PREPROCESSING

� B. CLASSIFICATION

� C. AREA CALCULATION

� D. GROUND TRUTHING & MAP PLOTTING

IMAGE PREPROCESSING

1. RADIOMETRIC CORRECTION:

CARRIED OUT AT NRSA

2. GEOMETRIC CORRECTION:

IMAGE TO MAP RECTIFICATION SELECTING GCP

3. NORMALISED DIFFERENCE VEGETATION INDEX (NDVI)

APPLICAION OF NDVI ON FCC

CLASSIFICATION

� MASKING OF FOREST AREAS INTERACTIVELY

� EXTRACTION OF MASKED FOREST AREAS FROM NDVI IMAGE

� DENSITY SLICING

� SHADOW EXTRACTION

� MOSAICING THE CLASSIFIED IMAGE

� SHEET EXTRACTION AT 1:250,000 SCALE

AREA CALCULATION

� OVERLAYING CLASSIFIED IMAGE ON DISTRICT BOUNDARIES

� IMAGE TO IMAGE RECTIFICATION

� DISTRICT WISE AREA ESTIMATION OF DIFFERENT FOREST CLASSES

GROUND TRUTHING & MAP PLOTTING

� POST CLASSIFICATION GROUND TRUTHING

� POST GROUND TRUTHING CORRECTION

� PLOTTING OF FINAL MAP THROUGH VERSATEC ELECTROSTATIC PLOTTER

FUTURE PROSPECTS

� VEGETATION MAPPING-SWITCH OVER TO DIGITAL MODE

� UPGRADATION OF DIP CAPABILITIES

� ACQUISITION OF HIGH RESOLUTION DATA

� FUSION OF HIFH & COARSE RESOLUTION DATA

� ACQUISITION & DIGITIZATION OF DIVISION ADMINISTRATIVE/FOREST BLOCK BOUNDARIES

� G.I.S. DATABASE

� FOREST ATLAS

SECTION 7: Estimation of forest cover for standard reference years using a modelling approach

Estracted from: FAO Forest Resources Assessment, Forest Resources Information System (FORIS)

Concepts and methodologies for estimating forest state and change using existing information

Report prepared by Walter Antonio Marzoli, Forest Biometrician/Ecologist

Country Capacity Building Project, FAO-Forest Resources Assessment Programme

ESTIMATION OF FOREST COVER FOR STANDARD REFERENCES YEARS FOR TROPICAL COUNTRIES

The technique of assessment consists of the following three steps (i) establishment of a computerised database, (ii) development of deforestation model (or and adjustment function), and (iii) computation of results for the standard reference years.

1. Establishment of a database

The Forest Resources Assessment 1990 was not intended to reproduce the available official statistics but to make an objective appraisal of the current state of forest resources and of the rates of changes, following a standard methodology and classification system.

The data compiled by the Project consists of three categories;

1.1. Tabular data:

It includes forest resources, population and socio-economic data at the subnatonal level (province, state). A thorough search was made of information sources related to the following items;

� area of land use and vegetation types (if possible multi-date observations),

� forest function, land ownership and legal status,

� forest management and forest plantation,

� forest volume, biomass and production,

� forest harvesting and utilization

The Guidelines for assessment based on existing data were prepared in the three official FAO languages to compile the country statistics and bring them in a framework of common concepts and classification. Using the guidelines, the country reports were reviewed and reliable information extracted, edited and stored as part of the Project database with the acronym FORIS which stands for Forest Resources Information System.

A special effort was made to collect information on forest cover area and its change in time. The data sources are very heterogeneous generally, and include recent and past forest inventories and maps based on different techniques (aerial photographs, satellite imagery and ground inventories). The data collected for each country will be re-appraised in the standard categories defined earlier for land use and forest types.

The following adjustments may be necessary:

(i) Conceptual adjustments: the classification system adopted in the existing data may not correspond to the standard classification. Basically, two types of adjustment may be necessary,

(a) Original classes must be grouped into one single broader category. This is the case for example when data are available by ecological zones within one forest formation (e.g. humid, moist, dry, lowland, mountain forests, etc. within broad-leaved closed forests). In this case no particular problems arise. However particular attention should be paid to the distinction between closed and open formations.

(b) One original category needs to be subdivided into two or more standard categories (i.e. inverse situation). This case is more difficult to handle and a combination of indirect methods and sound judgement has to be used depending upon the quality and quantity of information available.

(ii) Area adjustments. One of the following could arise in assessing forest area for the reference dates,

(a) Complete forest inventories or forest maps available.

If complete national forest inventory or map coverage is available, then the procedure of reappraising data into the standard classification described earlier should be used.

(b) Incomplete or missing forest inventory

If no forest inventory or forest map is available or if the country coverage is partial then area assessment will have to be carried out using existing vegetation maps. A description of the techniques used to handle missing inventory data is described in section

The tabular information was compiled for most of the countries at sub-national level (district, province, and state) and the original inventory data were incorporated in the FORIS database together with description of the sources and of the classification used. This database will be continuously updated as new information becomes available and constitutes the basis for estimating the state and change of the tropical forest resources.

1.2. Map data:

It includes vegetation types, ecofloristic zones and country and subnational boundaries.

Realising that deforestation is a location-specific process driven by, among other things, population pressure and environmental conditions (particularly the population carrying capacity of the area), demographic and ecological parameters were included in the database and integrated with the statistical data in the form of a GIS. From the outset it was realised that a GIS was essential to handle statistical and spatial data from a variety of sources, in a variety of formats and in different projections. Therefore, GIS development was given special attention by the Project.

In the course of time, more spatial data were added to the Project GIS including LANDSAT 1.5 Wold Reference System Grid, potential global and forest biomass, mean annual precipitation and biotemperature, topography, protected areas, and vegetation map from NOAA/AVHRR satellite imagery (for some regions).

1.3. Remote sensing data

A remote sensing based survey of the tropical forests using high resolution

Satellite data of two dates, one close to 1980 and the other close to 1990, was implemented at 117 sample locations. State and change information was interpreted and stored in the form of change matrices following the standard procedure developed. These data were analysed to obtained in-depth knowledge of the process of deforestation and forest degradation on the sample locations, by geographic sub-regions and ecological zone. As mentioned earlier the results of the sample survey will be presented in a separate paper. For the present study the forest cover change data derived from remote sensing have been used to perform a local calibration of the model parameters for Africa where the existing multi-date data were very scarce.

2. The Deforestation Model

2.1. Model building

The collected data on forest cover area have different reference year and need to be brought for reporting purposes to the standard years, namely 1980 and 1990. This was done with help of deforestation model (or forest area adjustment function) which correlates the forest cover change in time with ancillary variables by means of regression analysis. Aim of the model approach were (i) to adjust the sub-national units inventory data to 1980 and 1990, (ii) to provide a methodology to correctly expand to all units the available forest area change data, and (iii) to develop the understanding of the causes and of the dynamics of the deforestation process.

Model building: Step 1

The first step was to study the correlation between forest cover and some explanatory variables. Population density showed a significant correlation with forest cover as expected since deforestation is one of the effects of man-environment interaction Moreover the use of population density was interesting since population can be a substitute for time, so that units with different baseline years can be processed together by substituting population to the X variable.

A number of simple regression were performed with data of the same country or of different countries and years using cross-sectional analysis. The results of the regression analysis were encouraging and two major conclusions could be drawn: (I) In terms of mathematical formulation the transformation of population density into its logarithm gave the best fits; and (ii) in terms of the general shape of the curve an S-shaped curve concave in the early stages of development with low population (see the graph for Brazil on the next page) and convex at higher levels of population (see example for Thailand) seemed to describe the general trend.

SECTION 8: Forest Fire - Problems related to the assessment of forest area -The country perspective

by Ranjana Gupta, Dy. Director, Forest Survey of India

The ancients regarded fire, along with air, water, and land, as a basic element of the universe. Fire exerts a powerful fascination. An instinctive awareness and fear of it are deeply ingrained in mankind. Wild fires have covered large areas in the tropics. Next to climate and soil, fire has been the most important single factor affecting the extent, composition, and character of forest and other vegetative cover on wildlands. A bad master, fire is also a useful servant. Carefully and selectively used on a prescribed basis it can perform many useful tasks in forest management.

DEFINITION AND NATURE OF FOREST FIRES

Fire as general term, is applied to the common physical phenomenon resulting from the rapid combination of oxygen with a substance and is characterized by heat, light, and usually flame. A forest fire is unenclosed and freely spreading combustion which consumes the natural fuels of forest i.e. duff, frass, weeds, brush, and trees.

CAUSES OF FOREST FIRES

Fires may be natural or man made

The fires caused by man can be divided into classes-

(i) Unintentionally

(ii) Deliberately and intentionally.

1. Natural causes- may be lightning, rolling stones or rubbing of dry bamboos with each other. The number of fires that may be caused by all the three natural causes are not more than 5% of the total number of fires caused in any one year.

2. Man's Carelessness- About 95% of the fires in this country are caused by man.

(i) The accidental fires may be due to following causes-

(a) Smokers - Throwing of burning match stick or bidi or cigarettes stumps in the forest by graziers or other travellers.

(b) Campfire - Includes forest fire originating from fires built for cooking by cartman or travellers.

(c) Rail ways- throwing of glowing coal pieces by trains speeding through the forest.

(d) Debris burning- (Clearing)- Including forest fires which originate from fires set for clearing land for purposes like control burning, cultivation or pastoral purposes, disposal of such fires spread on ground.

(e) Illicit collection of forest produce- Throwing of fire carelessly ater collection of honey etc. During course of Mahua flower collection also people burn the area to make the surface clean.

(f) Control burning of fire line

(g) Control burning of grasses

(h) Falling of dry needles or leaves on electric poles giving out sparks.

(i) Throwing of torch wiid by villagers while passing through the forest in the night

(2) Those caused by man deliberately

(a) Burning of undergrowth and grass to collect MPF.

(b) Including new shoots of grass in summer by burning dry grass.

(c) Personal enmity of one wing to other wing of deptt.

(d) By kendu leaf staff during bush cutting (during course of bush cutting they used to burn the area instead of bush cutting to save money as well as getting flush.

(e) By villages aggrieved with forest official.

(f) During destroying or charring the stumps of illicitly felled trees in the forest.

CLASSIFICATION OF FOREST FIRES

(a) On the basis of causative factors-

(1) Natural fires,

(2) Accidental fires,

(3) Deliberate or intentional fires.

(b) On the basis of the place of their action-

(1) Creeping fire - is defined as a forest fire spreading slowly over the ground with low flame.

(2) Ground fire - is defined as a forest fire that burns the ground cover only i.e. the carpet of herbaceous plants and low shrubs, which cover the soil. E.g. in deodar forests as a results of slash disposal when it burns inside the humus for days without gice out flames and then causes a huge conflagration.

(3) Surface fire - is defined as forest fire which burns not merely the ground cover but also undergrowth. e.g. the fires in the plains

(4) Crown fire - is defined as a forest fire which spreads through the crowns of trees and consumes all or part of the upper branches and foliage, e.g. in coniferous forest. The above classes of fires are not exclusive, and independent of each other.

The Forest Survey of India has done an assessment of the extent of forest fires and the findings are given in the table below:

EXTENT OF FIRE INCIDENCE IN FOREST AREAS OF THE COUNTRY

SECTION 9: Tool for assessment of forest biological diversity

by Claire Elouard & Rani M.Krishnan, French Institute of Pondicherry

FAO-GOI Regional Project on " The status of Forest Resources Assessment in the South Asia sub-region & Country Capacity Building needs"

Introduction

Biodiversity can be assessed both at the landscape level and at the species level. Vegetation types are defined according to their species composition, in accordance with the bioclimatic conditions. Assessing biodiversity at the species level helps characterizing the stand studied as well as can be used as part of the factors to be taken into account for the identification of high conservation value areas which areas which need protection.

The study presented below was part of a FAO Training Programme entitled "Assessment of forest biological diversity". Conducted at the French Institute of Pondicherry in 1996. This Programme aimed at training foresters from three South-East Asian countries, Vietnam, Laos and Cambodia, to diversity assessment. Part of the programme was theoretical and field-work was conducted to apply the methods presented for assessing diversity.

The data used in this paper have been collected during the field-work with the aim of learning the manipulation of some tools for biodiversity assessment. Due to the limited time allocated to the field-work, only a small number of plots were laid. The results cannot therefore be used as a reference to characterize the forest types studied.

Material and Methods

Study area

The field work and data collection were undertaken in three forest types, e.g., evergreen, moist deciduous and dry deciduous forests, in Karnataka State, South India. The evergreen forest type is located in the Makut Reserve Forest, Kodagu (Coorg) District; the moist deciduous and dry deciduous forests are located in the Bandipur National Park, Mysore District.

The medium elevation evergreen forest studied belongs to the Mesua ferrea-Palaquium ellipticum type, as defined by Pascal (1984). The moist deciduous forest belongs to the Lagerstroemia microcarpa-Tectona grandis-Dillenia pentagyna type and the dry deciduous forest belongs to the Anogeissus latifolia-Pterocarpus marsupium-Terminalla spp. Type (Pascal 1984).

Data collection

The data presented below were collected using the following method, which consists of a cluster of 20x20 m plots.

Using compass, a square transect is established and measurement points are made every 20 m interval in evergreen forest and every 50 m in deciduous forest where variability and density are less. The increment of distance in between two plots allows then to capture higher species diversity. These measurements concern only the boundary of the square established. At each of the measurement points, a 20x20-m plot is established with the following measurements:

- The 20x20 m plot laid is divided into 10x10 m subplots ;

- All individuals with gbh > 30 cm (trees) are located (co-ordinates x and y) within the 20x20 m plot and measured for height and girth;

- All individuals with 30 >gbh >3 cm (saplings) are located within one 10x10 m subplot and measured for height and girth;

- In the same subplot, a smaller plot measuring 5x5 is laid and all individuals with gbh < 3 cm (seedlings) are measured for the height; heights are classified into intervals of 20 cm;

- Within this 5x5 m subplot, the percentage of the area covered by grasses, Strobilanthes, bamboo's and reeds is estimated in the deciduous forest;

- The slope is measured and spatial mapping of all individuals with gbh > 30 cm is undertaken in the 20x20-m plot.

The method was adjusted to fit to the field conditions and the time allocated for it. Therefore, the total number of measurement points- which should be 20 according to the method- were reduced to 9 in evergreen forest, to 9 in moist deciduous forest and to 4 in dry deciduous forest. The area therefore varied according to the forest types (Table1).

Table 1. Number of plots and total areas sampled in the forest types

Data analysis

Structural analysis

The structural analysis involved girth and height distribution among the trees, the relationship between the height and diameter with the representation of the past, present and future populations of trees for data obtained from both fixed-area plot and cluster methods. The basal area, biovolume estimation and the density can also be calculated and used to characterize the stand studied.

The data were analysed as follows:

- Frequency histograms for height and girth classes;

- Relationship between the height and diameter: plotting h=100d, following Oldeman (1984). The girth (gbh) was converted into diameter (dbh) using the formula

gbh= *dbh.

Diversity analysis

The data collected from plot and cluster methods have been used in this analysis. In order to understand the variations of the values obtained by the indices with varied sample size, the analysis has been undertaken at the quadrat or plot level and at the community level. Species richness was estimated with the total number of observed species and the Chao estimator. Species diversity was estimated with Simpson and Shannon-Wiener indices.

-Number of observed species

The number of species is represented as S_obs.

_-Cho estimator of species richness (Chao, 1987)

The Chao index, which corrects S_obs with additional terms by taking the role of rare species into account , was calculated using the formula :

where a = number of species represented by 1 individual

b = number of species represented by 2 individuals

S_obs = total number of species observed.

- Simpson index (Magellan 1988)

The Simpson index is defined from as the probability that two individuals randomly and independently selected belong to the same species. This index is calculated using the formula:

where n_i =individuals in the species i; N = total number of individuals.

The Simpson index can be defined in two ways:

� as =1-D, also called a dominance index; it varies between O(for a one-species community) and 1-1/S (for an even community composed of S species);

� as ' = 1 /D; it varies between 1 (for a one-species community) and S.

- Shannon index (Magurran 1988)

- For a given species richness S, the maximum value of the Shannon index is obtained when the species distribution is even (i.e., p_i =1/S, max(H)= lnS)

The Evenness value is therefore calculated as follows,

An example of indices calculation is given in Annex 1.The indices can be calculated for each 20x20m plot and for the total population recorded with all the 20x20m plots laid. The total population is obtained by adding all the plots species and individuals (the list is constituted and then sorted to get the number of individuals belonging to each species). The total values for the indices cannot be obtained by summing the values for each plot, but have to be calculated separately for the total population.

Results

Structural analysis

The analysis is illustrated here with the example of the tree data (gbh>30 cm) from the evergreen, moist deciduous and dry deciduous forests. The analysis uses the frequency data to understand the distribution of species and individuals across vertical and horizontal space and the application of the relationship between height and girth to understand the structural stratification of the forest.

Height distribution

The relative height of the ensembles in the forest is an important criterion for understanding the nature of the forest. The average height of the forest stand can vary depending upon the history of the site, the topography, the abiotic factors and the climatic climax of the area considered. In general, lowland dipterocarp forests harbour tall and emergent trees can reach up to a height of more than 35m. In the higher altitudes, there are stunted forests where the tallest trees will seldom be taller than 20m. Evergreen forests usually present higher and larger individuals than deciduous forests.

Evergreen forest

Maximum number of trees were encountered at the under-storey (<15 m) and subcanopy (15-25m) levels. The emergent level (>40m) is almost absent (less than2%). The tallest tree reaches 43 m. Normally, individuals with gbh>30 cm girth are taller than 5m. the representation of individuals in the lower height classes is due to the disturbance (natural and man-made); in cases of half broken or bent trees due to tree falls, or lopping ,the trees can have a girth>30cm, but be less than 5m in height.

Moist deciduous forest

Moist deciduous forests are usually not taller than 30m height, the first structural ensemble (canopy level) varying between 20 to 30 m (Pascal.1986). The tallest tree sampled is 30.5 m height. The canopy level, lower than in evergreen forests (reaching over 40 m height), is one of the structural differences between evergreen and deciduous forests. The number of structural ensembles is also reduced, the emergent level is absent in deciduous forests (Pascal, 1988)

Dry deciduous forest

The total height of the trees is lower in dry deciduous forests. Most of the trees belong to the height class 10-15m. Few tall trees (20-25m) are indicated for this forest type.

Girth distribution

The frequency of girth classes is also used to obtain the same kind of information as for height. In timber-oriented surveys, they formed important criterion for exploiting the species at an optimum girth class. The current studies on biodiversity use this kind of data available to understand the patterns of girth class distribution in an undisturbed forest.

Evergreen forest

This population presents trees belonging to all the girth classes, including the large ones. The presence of large trees, mostly belonging to timber species, indicated the absence of heavy exploitation and minimal disturbance and canopy species within the small girth classes shows that this forest population contains an important part of young trees, ensuring the regeneration and perennity of the forest stand.

Moist deciduous forest

The presence of trees of large girth (_100 cm) shows that the forest area sampled has had minimal disturbance. A large number of trees (29% of the population) belong to the small girth classes (30-50 cm). These girth classes comprise trees belonging to the different strata (Table 2). The presence of young trees of the canopy level shows that the population's future is ensured.

Dry deciduous forest

The girth classes range from 30 to 210 cm. This wider distribution and therefore the presence of large mature trees ensures the reproduction of the stand. Trees are poorly represented in the smaller girth classes (21% of the population). This can be related to fires occurring each in this dry ecosystem, injuring or destroying the saplings and young trees. Though large trees are found in this dry deciduous forest, the major part of the trees are distributed within the girth. Though large trees are found in this dry deciduous forest, the major part of the trees are distributed within the girth classes 50-100 cm (52%). This feature is common in dry deciduous forests, where very few species reach a large girth, with a preponderance of small and medium girth species. The small girth classes, though poorly represented, contains in a major part trees of the canopy level, constituting its future stand (Table 2).

Table 2. Distribution of the trees of the girth classes 30-50 cm within the strata

Height and girth relationship

The relationship between the diameter and the height is the expression of the growth programme that determines the successive architectural phases of the trees (Oldeman 1974). The simple average model for the expression of the relationship between height (h) and trunk diameter (d) is h=100d (in meters). Trees growing spontaneously are found to the right of the line or on the line and form the `set of the present'. The older trees are located right of the set of present and called as `set of the past'. They are characterized by a large diameter. Trees located to the left of the line form the `set of the future', these individuals have not yet expressed their full architectural development: they first grow in height to reach their strata level and once they get the maximum light available, they grow in diameter. The layering of the ensembles in the natural forest is not always distinct due to the presence of trees of the future, which fill all the gaps. The model has been found valid for tall evergreen forests of French Guiana (Oldeman 1974), Indonesia (Laumonier 1980), Africa (Devineau 1975), Western Ghats, India (Pascal 1984, Ramesh 1989, Pelissier 1995).

Evergreen forest

The set of the future is strongly represented, with accumulation of several trees ready to take over (0.1-0.3m diameter and 10-30 m height). The set of present and past are also present.

Moist deciduous forest

Relationship between diameter and height shows that a significant number of trees are seen in the set of the present and even in the past. Few individuals are present in the lower diameter classes, representing the set of the future and the potential for regeneration.

Dry deciduous forest

The set of the future is quite inexistant in this forest. Trees are distributed in the sets of the present and the past. This puts forward the question of the perennity of this forest.

Diversity analysis

The data entry, the data format required for this kind of analysis and the application of the necessary formulae used for estimating diversity were applied using Microsoft Excel spreadsheets.

Abbreviations used

- N: Total number of individuals

- S: Total number of observed species

- Chao: Chao estimator of species richness

- Simpson (D): Simpson index (species diversity)

- Shannon (H): Shannon index (species diversity)

- (a, b): number of species with only 1 individual (a) or only 2 individuals (b); Chao index cannot be calculated if b=0;

- * : no data.

Evergreen forests have been reported to be the richest forests in the Western Ghats, in comparison with deciduous forest. Along with the bioclimatic gradient is a species richness and diversity gradient from the evergreen forests to the moist deciduous and dry deciduous forests. Endemic species, largely present in the Western Ghats (Ramesh etal. 1997), are mainly distributed in the evergreen forests.

Evergreen forest

The number of trees species observed is 59 in 3600 m² area (Table 3). The compiled data (total of the plots) show a high Chao estimator: the stand is represented by many species with 1 to 2 individuals (15.3%). Dipterocarpus indicus (Dipterocarpaccae) dominates (39 individuals, 18%), followed by Knema attenuata (Myristicaceae; 14 individuals, 7%), and in a smaller extent Dimocarpus longan (Sapindaceae; 11 individuals, 5%) and Mesua ferrea (Clusiaccac; 11 individuals, 5%).

Though the stand is dominated by 4 species, diversity remains quite important as there are many species represented by few individuals. Evenness, quite high, reflects this feature, as well as the Shannon index (H'), relatively close to the maximum value (Hmax).

Moist deciduous forest

The number of tree species recorded with this method is 18, with 134 individuals (Table 3). The number of species is regular among the plots, and the number of individuals varies from 8 to 27. Dominance is clearly shown by Anogeissus latifolia (59 individuals, 44%), Terminalia crenulata (27 individuals, 20%) and Tectona grandis (15 individuals, 11%). Rare species are present, though 3-7 individuals represent most of species, after the dominant species.

The diversity is low both reflected by the low values of Simpson (1/D), Shannon (H) and Evenness (E) indices.

Dry deciduous forest

The number of tree species recorded with this method is 11, represented by 42 individuals. Though few individuals represent half of the species, the Chao estimator is quite low. Terminalia crenulata (13 individuals, 31%), Anogeissus latifolia (7 individuals, 17%) and Tectona grandis (6 individuals, 14%) are dominant. Diversity, though higher than in the moist deciduous forest, remains quite low, due to the high dominance of the three species mentioned above.

Reference

Chao A., 1987. Estimating the population size for capture-recapture data with unequal catchability. Biometrics, 43: 783-791.

Devineau J-L.., 1975. Etude quantitative des forests-galeries de Lamto (Moyenne Cote d'Ivoire). These 3eme Cycle, Paris.

Ganesh T, Bawa K.S Ganeshan R & Davidar P., 1996. Assessment of plant biodiversity at a mid-elevation evergreen forest of Kalakad-mundanthurai Tiger reserve, Western Ghats, India. Current Science, 71: 379-391.

Magurran A. E., 1988. Ecological diversity and its measurement. Croom Helm ltd, London 179 pp.

Olden R. A. A 1974. L'architecture de la forest guyanaise. Memoires O.R.S.T.O.M., 73, 204 p.

Pascal J.P. 1986. Explanatory booklet on the forest map south India. Institut Francais de Pondichery.

Pascal J. P. 1988. Wet evergreen forests of the Western Ghats in India: ecology, structure, floristic composition and succession. Travaux de la section scientifique et technique, 20bis Institute francais de Pondichery, 345 pp.

Pelissier R., 1997. Heterogeneite spatiale et dynamique d'une foret dense humide des Ghats occidentaux de l'Inde. Publications du Department d'Ecologie, n 37. Institut Francais de Pondichery, 150 pp.

ANNEX 1

SECTION 10: Problems Related to the Assessment of Forest Area: The Country Perspective - GRAZING

by Ranjana Gupta

Introduction

Forests are an important source of food for cattle in most parts in India.

Thelivestockpopulation of India accounts for about 15 percent of the total livestock

population of the world and for only 2 percent of the total world's geographical area. As

per the 1982 census, the livestock population in India was 415.9 million and is expected

to cross 500 million by 2000 AD. The Livestock population in India has shown a rising

trend over the years as shown in Table 1.

DAMAGE TO THE FOREST BY GRAZING ANIMALS-

Grazing animals graze seedlings making regeneration a problem.

(I) Seedlings are trampled and crushed or their roots get exposed by the hoof of moving animals.

(II) Uncontrolled heavy grazing compacts the soil, reduces its porosity and causes increased run-off and causes soil erosion.

(III) Depletion of nutritive palatable grasses which cannot even regenerate resulting in reduced grazing capacity.

(IV) Heavy continuous grazing results in dry and grazing resistant inferior species.

TABLE 1 : LIVESTOCK POPULATION IN INDIA 1951-1982

Category of livestock	Livestock population in (million) different years				% increase in 1982
	1951	1961	1972	1982 over 1951
Cattle	155.30	175.56	178.34	190.79	22.85
Buffaloes	43.35	51.21	57.34	69.00	59.17
Sheep	38.43	40.02	39.99	48.07	25.08
Goats	47.08	60.86	61.52	94.72	101.19
Horse & ponies	1.51	1.33	0.94	0.93	99.07
Pigs	4.42	5.18	6.90	9.58	90.42
Camels	0.63	0.90	1.11	1.03	63.49
Others	1.30	1.15	1.11	1.82	40.00
Total	292.02	336.21	353.34	415.94	42.43

Deputy Director, Training & Vegetation Mapping Unit, FSI, Dehradun.

The increase was not uniform in different States of the country. The density of livestock population in the states is given in Table 2. It is observed that almost in all the States the density of livestock is in excess of the carrying capacity of the land.

The NCA (1976) estimated the requirements for the projected livestock population for the year 2000 as given in the table 3.

TABLE 3 : REQUIREMENT OF FODDER FOR PROJECTED LIVESTOCK POPULATION IN 2000