2. CURRENT STATUS OF NATIONAL FORESTRY STATISTICS RELATED TO WOOD PRODUCTS.

The Central Statistics Office compiles National Statistics, including statistics related to Wood Products. The Wood Products statistics is compiled from secondary sources. Data sources include reports of, and personal communications, with various Government Departments, Non Governmental Organisations, Parastatals, Organisations, Private Companies and Research Institutions.

The national statistics are collected under the Statistics (Surveys of Industrial Production) regulations, 1989 for industrial production. The Environment Statistics Unit of the CSO collects the trade statistics.

Data on industrial production includes: quantity and value of products, number of people employed and their remuneration, changes in the activity of the timber establishment, and differences in production between different quarters of the year. The remuneration data is based on gross wages and salaries. It also includes overtime, rations, housing, allowances, and gratuities and leave pay. The Harmonised System has been adopted for wood products, but in the present publications, the Timber Industry data are grouped under two broad headings, "Wood and Wood Products", and "Paper and Paper Products." It is envisaged that the next publication (due before the end of year 2000) would have properly coded data compatible with the Harmonised System. Thus, various wood products would be coded independently.

The wood products trade statistics comprises of export and import data. This data includes value and quantity data and lists countries involved in trade with Botswana regarding wood products. The wood product data for trade statistics is coded using the Harmonised System. Various wood products are coded and classed in a comprehensive way. These data is collected through invoices received from the Department of Customs and Exercise, and through collaborations with other stakeholders as stipulated in paragraph 2.0. of this report. The United Nations Framework for the Development of Environmental Statistics (FDES) is used in the drafting of publications for these data.

2.2.1 Methodology for collection of data on industrial production of wood products.

Survey coverage: The Industrial Unit of the Central Statistics Office (CSO) collects data on wood products by sending out questionnaires to all the operating establishments in the Manufacturing Sector involved in wood products. After one month of dispatch of questionnaires, CSO officers make follow-ups to non-respondents. This is usually continued until the response rate is deemed satisfactory, which would be at least 50 percent.

The sampling frame for the survey of industrial production of Wood Products is the Enterprise and Establishment Register (EER). The maintenance of the EER is also the mandate of the Industrial Statistics Unit of the CSO. The EER is maintained on a regular basis and updated using various sources i.e. Registrar of Companies, the National Licensing Authority, Department of Consumer Affairs, Commercial Affairs Units of the Tender Board, telephone directories and newspaper advertisements.

Editing, Coding and Data Capture: Once the Industrial Statistics Unit of the CSO receives the completed questionnaires, they are scanned for any errors and inconsistencies. The goods are then coded using the Harmonised System. A computer based database program is used to enter data from the questionnaires. The database program is designed in a way that when data is entered, it goes into two files. One file is for general information (particulars of operating establishments, employment and earnings), the other file is for the product information (product details i.e. product value and quantity, product specifications, and month of production).

Estimation Procedure: The Wood Products data is analysed and tabulated using the product information file. The variable examined is the value of production. After determination of total production by industry, the Industrial Statistics Unit of the CSO compares these with data derived from the census of Manufacturing and Construction. The Census of Manufacturing and Construction is undertaken by the Industrial Statistics Unit and analysed by the National Accounts Unit. The comparison takes into account the fact that the survey of Industrial Production information refers only to the characteristic output, while the National Accounts output includes other sales and takes into account changes in stock.

Limitations: This methodology employed only the value of production because it is the only variable deemed reliable. The methodology does not use quantity data as it is subject to a lot of errors. The questionnaires also reveal that the operating establishments have the propensity to report values only without giving quantities.

The way quantity data is provided in the questionnaires is not compatible with the Harmonised System. Consequently, it is usually difficult to convert the given quantities to the harmonised System. As a result, it is difficult to produce reliable quantity data.

2.2.2. Forest Inventory Monitoring Programme Methodology for Fuelwood / Woody Biomass Assessment.

This methodology was developed by Natural Resources and People (PTY) Ltd (NRP PTY Ltd) which was commissioned by the Ministry of Minerals, Energy and Water Affairs in January 2000 to develop a Fuelwood Inventory and Monitoring Programme (FIMP).

The methodology is based on the following main design parameters:

Rapid methodology

Repeatable methodology

High frequency of sampling

Sample contained within one vegetation class

Giving an index or ranking of fuelwood quality

Estimation of dead fuelwood availability

Accurate reflection of biomass of multi-stemmed bushes

Easily documented and simple to train surveyors

The methodology developed under the FIMP is thus led by two main factors.

the available information upon the relationships between measurable parameters in the field, gleaned from an extensive literature review,

the need to be rapid, reliable and repeatable, amongst different observers and in different areas.

The method of woody biomass assessment / estimation entails laying / demarcating transects of 50m x 10m in the field using measuring tapes and survey poles. The stem diameters at ankle height of all the individual stems that are rooted within the transect are measured to the nearest centimetre. The measurements are recorded by species in a standard data sheet. The standard sheet is linked to a database. This constitutes a data capture; processing and analysis system that has been developed based on the Microsoft Access database software.

In other words the information system contains a data capture form which mirrors the field data sheet used to collect woody biomass measurements in the field. The data capture process automatically stores all the field data in a master data table containing class size measurements of all the woody species measured in the field. A series of queries calculate woody biomass for all the survey sites by applying the regression equation developed by Tietema (1993). The system produces summary data on the biomass per site, biomass range and averages for each vegetation type and biomass the proportion of each woody species.

Individuals whose woody canopies overhang the transect, but whose stems are rooted outside it, are not measured. The stem diameter classes shown on the data sheet were derived from absolute woody biomass estimates for twenty study sites and result in less than one- percent error across all of the sites.

Additional items are also assessed in the field (refer to table 1).

Table 1. Other items assessed in the field.

A 0-5 scale is used, where 5 refers to the factor operating at maximum and 0 means that there is no sign of the factor.

Additionally, the stem diameters of the stumps within the transect are recorded, and also used to assess the level of harvesting pressure. Coppicing is also assessed, but the number of cut branches is not explicitly recorded.

It is recommended that the selected sites should be located in ‘homogenous vegetation units’ that sample the strata identified by a preliminary unsupervised Landsat 7 classification. Thus, transitions between ecological zones are avoided.

2.2.3 Methodology for the inventory of natural woodlands.

Two resource inventories were carried out by the government of Botswana, Ministry of agriculture, on natural forest resources (Miombo Woodlands) in Ngamiland and Chobe (Northern Botswana). The methodology used was developed during the Chobe Inventory project. The objective of these forest inventories was to: determine the resource base and make recommendations on future management plans, identify potential timber concession areas for commercial timber harvesting and to determine the richness and diversity of forest resources.

Sampling design: An unstratified systematic sampling design with one sampling phase is used. Stratification of the resource is recommended when the area inventoried is not homogenous. The recommendation on whether to stratify or not is based on assessment of aerial photographs and topographic maps joined with reconnaissance surveys. Sampling plots are clustered into groups of four making up a ‘Tract’. A tract constituted a "Primary Sampling Unit". One plot is retained as a Permanent Sample plot, which is measured at five-year intervals. Circular plots of 30m radius are used and cited at 300m distance in each tract. Tracts to each line are placed at 2.1km apart running North-South, and 2.4km East West.

Thus,

The sampling intensity = (Area of each plot) x Total # of plots) x 100

Total sample area

Enumeration design: Each tract (a square of 300m sides) is orientated along North, South, East and West i.e. 0, 090, 180,and 270 degrees true north. Each plot is located at the corner of a tract as described in the sampling design. The first plot is located at the southwestern corner (near the UTM grid intersection). The intersections are located using satellite Geographical Positioning Systems (GPS).

In each tract, the first plot is retained as a permanent sample plot for the purpose of future determination of growth rates, timber yield, data and monitoring of forest health overtime. In each permanent sample plot the trees are given identities (metal tags / painted). The inventory plan stipulates a sample point with 30m and 15m concentric plots. In the 30m radius plot trees with diameter at breast height (dbh) of at least 9cm are recorded and in the 15cm radius plots trees with dbh less than 9cm are counted and recorded as regeneration. All veld product species are identified and recorded. Tree damage is also recorded.

Assessment of Growing Stock per hectare: Growing stock per hectare is assessed for the commercial species i.e. Baikiaea plurijuga, Burkea africana colophospermum mopane, and Pterocarpus angolensis etc. The following tree parameters are measured. Tree diameter (cm) (dbh = 1.3m), Total tree height (m), Timber height (m). Volume of stand to 5cm and 15cm top diameters are computed using volume tables derived from the Chobe Forest Inventory project. The numbers of live stems, regeneration counts and tree damage are also assessed.

The Information on soils in Botswana is obtained from the Land Utilisation Division computerised Geographical Information System, Ministry of Agriculture. A socio-economic survey is recommended if timber concessions are to be granted.

Stand Parameters: Fuelwood is assessed to 5cm top diameter while timber is assessed to 15cm top diameter for the commercial harvestable trees. A calibrated diameter tape is used for measuring dbh. A hypsometer is used to measure tree height. Stems per hectare is calculated using the formula:

Stems/ha = å (No. of live stems in a plot)¸ Area of the plot

A regeneration count for commercial species in a 15m radius is determined by dividing the number of regenerations in a 15m radius by 0.0707ha to give the number of regenerations per ha. Tree damage is assessed and coded into nine codes ranging from slight damage to dead. The inventory also recommends generation of a commercial species distribution map and the general vegetation map for the area being inventoried.

Volume of Timber and Basal Area per hectare:

Volume of fuelwood to 5cm and timber volume to 15cm top diameters is calculated using information collected on growing stock (dbh) and volume table (Chobe Forest Inventory project, 1992).

Volume of Timber/ha is calculated from the following formulae:

Volume/tree = å (Volume of all trees) ¸ number of trees/plot

Volume /ha = å (Volume/tree)¸ area of plot

The overall volume / ha = å (Volume/ha) ¸ number of plots

Two techniques could be used for calculation of basal area per hectare. One method entails the use of a relascope. The number of trees/stems counted within the range of the relascope when completing a 360-degree sweep from the centre of each plot is multiplied by the relascope factor. Thus;

Basal area/ha = å (stems within relascope sweep) x relascope factor

The other method involves adding the cross sectional area of each tree at breast height in a plot and dividing this area by area of plot.

Basal area/ha =å cross sectional area of each tree) ¸ area of the plot

Ultimately, the basal areas/ha for each plot is then divided by the number of plots to give the overall average area per ha.

Regression analysis is recommended to explore relationships between the variables

Volume/ha

Basal Area/ha

Dominant height

A simple linear regression equation takes the form:

Y = ax + b

Where Y = Volume/ha (dependent variable)

a = Slope from "Line of best fit"

x = basal area/ha x Dominant height (independent variable)

b = Interception of y axis

The parameter "a" and "b" can be estimated by feeding the variables "y" and "x" into the simple linear equation. A statistical computer programme on lotus 1.2.3. could be used for this purpose and for determination of r² (the coefficient of determination). The coefficient of determination (regression r²) measures the precision of the regression equations developed for each species. If the precision is high, there is a strong relationship between the variables: Vol./ha, Basal Area/ha and dominant height.

The data are compiled in computer databases and reports as described in Sections 2.2.1, 2.2.2, and 2.2.3. Various Forums, Workshops, Seminars and meetings involving different stakeholders are convened to discuss and formulate reports. The reference groups are derived from various stakeholders discussed earlier in this document. Consultants are often engaged for technical support in updating, acquisition of new technology, and imparting of new knowledge to personnel engaged in data collection, compilation and validation.

Various government departments do collection, analysis and dissemination of data related to wood products. These include the Environment and Trade Statistics Unit of the Central Statistics Office, the Industrial Statistics Unit of the Central Statistics Office, Energy Affairs Division in the Ministry of Minerals Energy and Water Affairs, Division of Forestry and Range Ecology in the Ministry of Agriculture, Department of Planning and Statistics in the Ministry of Agriculture, National Archives. The NGOs, Media (Press), Private Sector, Parastatals Bookstores and other Retailers are also involved.

The Southern African Development Community Forestry Co-ordinating Unit based in the Republic of Malawi is also involved in Forestry issues affecting member states. Botswana, like other developing countries has benefited from foreign assistance, e.g. from the European Union, Swedish International Development Agency (SIDA), United States Agency for International Development (USAID) in issues pertaining to Forestry and Natural Resources.

The most important wood products consumed and traded are industrial roundwood, wood fuel, wood charcoal, wood chips and particles, wood residues, sawn wood, wood based panels, plywood, particle board, fibreboard, recovered paper, paper and paper board. The information on local consumption of woodfuel at national level is insufficient and fragmentary. However Energy Affairs Division, National Institute of Research (University of Botswana) and some NGOs have conducted various studies in some parts of the country. Information on trade and consumption of wood pulp and other pulp is not available. It is assumed that the country does not trade in wood pulp (refer to annexes 2, 3, 4a, and 4b).

The most important Wood Industries presently active in the country are presented in table 2. These industries rely mainly on imported raw materials and are often characterised by lack of capacity to meet local demand. However, due to time constraints it was difficult to determine the production capacity of each establishment. Botswana is a vast country and these enterprises are scattered throughout the country. It is also difficult to obtain reliable quantity of production data from these establishments.

Table 2. Establishments involved in the manufacturing of wood products.