10 February 1998
Scope/Coverage: Management inventory of eight Forest Divisions: The SRF, 4 C/A Divisions and Sylhet, Chittagong and Cox's Bazar Forest Divisions.
Objectives: (a) To generate forest resources (trees, poles, saplings, seedlings, bamboo, and Golpatta) statistics for forest management planning purposes, (b) to provide "abstract time-series" data for plantation yield/growth modelling, whenever possible, and (c) to set up a system or at least provide a basis for setting up a system of "hidden"/unmarked sample plots for continuous monitoring and assessment of change on the forest/other resources of the target forest divisions.
Target precision/sampling errors of the estimates: The sampling design specifications were guided by the following target precision of estimates at the division and stratum levels: upper limit of 5% sampling error (division level) based on tree volume; upper limit of about 10 to 15% sampling error for each of the more important strata and species groups, and about 20% for the other strata. These target precision levels apply to the major divisions: SRF and the 3 Hill Forest Divisions (Sylhet, Chittagong and Cox's Bazar). In the case of the 4 C/A Divisions, the upper limit for the sampling error, also based on tree volume, is 10% at the division level and about 15 to 20% at the stratum level at least for the main stratum and tree species. To try and attain the higher precision levels would double the time and resource requirements of the field sampling work in the C/A Divisions, time and resources that were not available.
Sample size (n): The following sample sizes were determined to achieve the target precision levels. Also shown are the actual samples sizes, na, and samplings errors that were attained at the Division level.
Division |
SRF |
Sylhet |
Cox'x Bazar |
Chtg |
Chtg C/A |
Noak'li |
Bhola |
Patuak'li |
N |
1277 |
1162 |
1017 |
1120 |
374 |
716 |
328 |
293 |
Est'd area |
401632 |
32173 |
49319 |
53266 |
12535 |
24027 |
11022 |
9835 |
Na |
1143 |
1084 |
1158 |
1088 |
408 |
289 |
225 |
168 |
Area, ha (Forest) |
399470 |
41565 |
49482 |
73822 |
20042 |
34223 |
12420 |
9848 |
SE % |
2.5 |
4.8 |
3.4 |
9.0 |
18.3 |
7.1 |
10.5 |
10.7 |
SE = sampling error
SRF = Sundarban Research Forest
Sampling design and the sampling unit: For 3 reasons, systematic sampling was adopted for the FRMP forest inventories (a) stratified random sampling was not feasible because forest mapping was not yet completed during the design phase, hence, definite/specific strata could not be identified and precise stratum areas were not available; (b) a set of systematic sample plots provides a good base for a continuous/recurrent change monitoring and assessment system; and (c) systematic sampling is simpler to implement in the field. Equal probability sampling was used in the SRF (sampling units were in one-minute grid) and C/A plantations (20" grid) while unequal probability sampling was considered more suitable to the hill forest divisions, primarily due to the large differences in areas of the forest types. The basic grid was 40" x 40" with different strata having different grids, e.g. 10" x 10", 20" x 10", 20" x 20" and 40" x 20", depending upon the area, relative importance and variability of the stratum. The sampling unit was a plot cluster made up of 5 plots/sub-plots, a set of centre plots/sub-plots and 4sets of plots/sub-plots situated 100 m from the centre plot along the 4 cardinal directions (see below). Instead of 100 m, 50 m was used in the bamboo forest, SRF and C/A plantations due to the relative difficulty of movement in these areas. A set of plots/sub-plots consists of several plots/sub-plots, one each for seedlings (1 m radius), saplings (2 m), poles (5 m), trees (11 m), bamboo Muli (5 m), plantation bamboo (11 m), other/small steam bamboo (2 m), rattan with less than 3 m long stem (2 m) with 3 m or longer stem (5 m), medicinal plants (2 m), Golpatta seedlings (2 m) and Golpatta (5 m).
Sampling procedure/instructions: After the sampling design was completed and approved, the detailed sampling procedures and instructions were prepared in June/July 1995.
Field sampling: Training of field crews started in October in 1995 with ACFs are crew leaders recruited for the field work. Thereafter, field sampling started in Sylhet, SRF, Chittagong and Cox's Bazar. The field sampling activities were under the direct supervision of the concerned DFOs-WP. The regular field sampling activities were completed in May 1997.
Data entry/validation: A data entry and validation program (DEVP) was designed starting in December 1995. The first working prototype was ready in early May 1996 for entry and initial validation of the data gathered up to the end of the field work season in May 1996. Data entry and initial validation activities were also the responsibility of the DFOs-WP. Data entry and initial validation activities were finally completed starting with Sylhet in May 1997, Surdarbans in August and the other Divisions in September 1997.
Tree volume equations studies: In cooperation with BFRI Researchers, the volume equations for Acacia auriculiformis (Akashmoni), Acacia mangium, and Eucalyptus camadulensis were improved through (a) the collection of additional data particularly bigger DBH trees that were not available when the original equations were developed and (b) re-specification of the volume equations. In addition, new tree volume equations were also developed for SRF trees: Sundri, Gewa, Koera, Baen, Kankra, Passur and Dhundul due to the apparent bias of the equations used in ODA inventory in 1983-85. In addition, volume and weight equations were derived for Goran as well as weight equations for Golpatta leaves.
Data processing: A fielding data processing program (FDPP) was also designed and developed specially for the FRMP forest inventories. FDPP was ready to process valid data in May 1997. Data processing started as soon as data entry and initial validation was completed, the FI databases were received from the DFO-WPs concerned, and after final check/validation of the database by the FIS and Computer Programme. Data processing to generate the required statistics and other information needed in integrated forest management planning that can be provided by the forest inventories have been completed.
The forest resources statistics and databases: The draft final reports on the forest inventories of the 8 Forest Divisions have been submitted. These include the detailed stand and stock tables for trees by DBH class and species/species group: no. of trees, basal area and volume per ha and for the whole division as well as by stratum, forest type, compartment, block, and range. The detailed statistics on seedlings, saplings, and pole as well as Golpatta (in SRF) and bamboo area also included.
In addition to the results of the forest inventories being produced in the traditional hard copies, they are also available in digital form as dbf files which are consistent with the RIMS-GIS database, hence, these can be integrated into the RIMS-GIS quite easily. All the forest inventory data that were collected in the field and entered into the Forest Inventory database are also stored in the computer and are available for future reference and use.
The results, in general terms: The following tables give in broad terms the major findings of the FRMP forest inventories.
Statistics |
SRF |
Noakhali |
Ctg C/A |
Bhola |
Patuak'li |
Area, ha |
399470.00 |
34223.00 |
20042.00 |
12420.00 |
9848.00 |
Sample size |
1143.00 |
289.00 |
408.00 |
225.00 |
168.00 |
NT/ha, 15+cm dbh |
144.70 |
172.00 |
10.00 |
58.00 |
156.00 |
BA/ha (sqm), 15+cm dbh |
5.38 |
5.26 |
0.29 |
1.96 |
5.86 |
Vol/ha (cum) 15+cm dbh |
27.40 |
25.69 |
1.02 |
9.31 |
36.40 |
SE % |
2.50 |
7.40 |
18.30 |
10.50 |
10.70 |
Seedlings/ha |
33120.00 |
13360.00 |
15228.00 |
641.00 |
5840.00 |
Saplings/ha |
7469.00 |
3151.00 |
2202.00 |
185.00 |
967.00 |
Poles/ha |
1120.00 |
1696.00 |
700.00 |
462.00 |
511.00 |
Golpatta/ha |
7955.00 |
- |
- |
- |
- |
Wt. Leaves/ha (tons) |
14.60 |
- |
- |
- |
- |
Areas of Golpatta, ha |
7797.00 |
- |
- |
- |
- |
Statistics |
Sylhet, NF |
Sylhet Plant'ns |
Cox's B, NF |
Cox's B, Plant'ns |
Chittagong NF |
Chittagong Plant'ns |
Area, ha |
23693.00 |
17871.00 |
30398.00 |
19084.00 |
59679.00 |
14143.00 |
N |
279.00 |
726.00 |
544.00 |
444.00 |
548.00 |
251.00 |
NT/ha |
56.00 |
133.00 |
80.00 |
67.00 |
13.00 |
42.00 |
BA/ha |
5.39 |
6.91 |
8.97 |
3.52 |
1.32 |
2.25 |
Vol/ha |
53.20 |
37.64 |
74.36 |
19.68 |
10.87 |
11.32 |
SE % |
7.10 |
3.60 |
3.80 |
6.70 |
10.90 |
11.60 |
Seedlings |
5790.00 |
2940.00 |
4888.00 |
4288.00 |
735.00 |
615.00 |
Saplings |
737.00 |
607.00 |
1401.00 |
1092.00 |
323.00 |
362.00 |
Poles |
241.00 |
602.00 |
339.00 |
696.00 |
151.00 |
379.00 |
NT = No. of trees
NF = Natural forest
Conclusion and recommendation: The FRMP forest inventories of the eight forest divisions have generated the desired results as evidenced by the sampling errors of the tree volume and other estimates. The division-wide sampling errors of the volume estimates are less than the target precision of 5% sampling error in SRF, Sylhet and Cox's Bazar. The sampling error in Chittagong is higher (9.0%) because the forests there are unusually fragmented and have very high variability. In the C/A plantations, the sampling errors (7.4 to 10.7%) are also within the targets, except in Chittagong where the stands are highly variable. In general, the sampling errors at the stratum level are also within the designed targets of the forest inventory. The ENRS (Extended Natural Resources Survey) which independently collected, handled, entered and analyzed their own data has confirmed the results of the FRMP inventory. More convincingly, in SRF, a validation re-survey of 56 plot clusters that were drawn at random with emphasis on the Gewa and Gewa Sundri forest types, showed a difference of less than one-half percent in the total volume of the 56 plot clusters when compared with the results of the regular enumeration. The validation re-survey which was participated in by representatives of the DFO (Khulna) and KNM (Khulna Newsprint Mill) measured only three plots per cluster while the regular sampling enumerated five plots per cluster. Thus, it can be concluded that the statistics presented in this report are valid and represent the true picture on the ground. They are also quite adequate for forest management planning purposes.
The tree resources of the SRF had decreased dramatically over the last 37 years from the FORESTAL inventory in 1959 to the ODA inventory in 1983 and the FRMP inventory in 1996. Estimates of the three inventories show that Sundri had decreased from 211 trees/ha in 1959 to 125 in 1983 and 106 in 1996 in the whole SRF based on 15-cm+ dbh trees, or about 50% over the 37-year period. In the case of Gewa, the number of trees/ha had decreased from 61 in 1959 to 35 in 1983 and only 20 in 1996, or a decrease of about 67% for the same period. In terms of all tree species, the decrease had been from 296 in 1959 to 180 in 1983 and 144 in 1996, or about 51% over the 37-year period. The picture is a little worse if the 10-cm+ dbh trees are considered. The decrease in number of Sundri trees/ha in the SRF in this case had been from 511 in 1959 to 296 in 1983 and 215 in 1996, or a decrease of about 58% over the 37-year period. The case of Gewa and all species is similar. Gewa decreased from 345 in 1959 to 224 in 1983 and 153 trees/ha in 1996, or a decrease of 55% over the same period. For all tree species, the decrease was from 952 in 1959 to 557 in 1983 and 398 trees/ha in1996, or a decrease of about 58% for the 37-year period. From the standpoint of sustainable production of the two major species, the trend over the 37-year period is certainly a cause for alarm and has to be addressed immediately and effectively. This is a major challenge for the on-going forest management planning activity and implementation of said plan by the concerned authorities. From the standpoint of vegetative cover, 144 trees/ha (15-com+ dbh) or 398 trees/ha (10-cm+ dbh) plus more than 2500 small poles/ha, 7500 saplings/ha and 33200 seedlings/ha do not yet present a grim deforestation picture, but obviously, the trend cannot be allowed to continue?
It should be worth repeating here that the decrease in the number of trees in the SRF from 1959 to 1996, 51% if based on the 15-cm+ dbh trees or 58% if based on the 10-cm+ dbh trees, does not mean that 58% of the SRF forest is gone. It simply means that the tree density of the forest has been reduced to about half of what it used to be 37 years ago. It also means that if sustainable management of these forest requires increasing or maintaining the present density or even changing the stand structure of the forests, then these have to be addressed by the forest management system. In Sylhet, the bamboo (Muli) resources have decreased from about 17,000 stems to 11,000 stems per hectare over the last eight years.
The fact remains that people, mostly from the less privileged segment of the country's population, some 1.2 million or more in/around the SRF, depend upon the forest resources for their livelihood, directly or indirectly. On the other hand, the forest resources had decreased dramatically over the years such that there is imminent danger of degrading the resources to a situation where they would lose their capacity to provide desirable levels of goods and services. Before these resources reach that critical stage, it is imperative that the users, managers and the people, notably the leaders, provide the needed concerned efforts (political will, policies, strategies, programmes, support systems, resources, ...) to manage and conserve the forest resources so that they remain productive on a sustainable basis. This also requires the eradication of pervasive poverty as a minimum necessary condition, for, there can be no sustainable management/conservation of renewable resources under conditions of widespread poverty.
The integrated forest management plan for the 8 forest divisions are being prepared by FRMP. More projects to conserve the forest resources including the generation of more livelihood opportunities for the people in the neighbouring areas are being proposed for funding by international financing institutions. All these efforts have noble objectives and could help, but in the end, the conservation or degradation of these resources will depend primarily upon the leaders and people of Bangladesh!!!
It is recommended that the other forests of the country are inventoried including the line/strip plantings along embankments, roads/highways and railroads. A simple design that was prepared for the CCF's Office more than a year ago could be improved along the concept that was applied in the determination of the area of Golpatta strips along the river banks of the SRF.
A continuing resources change assessment system (CRCAS) for the country's forests is hereby strongly recommended. CRCAS must be designed, supported and implemented to provide timely (at least annually) resource change statistics for the Forest Managers, the leaders and people to respond effectively to any aggravating circumstances. The basic components of CRCAS have now been set in place by FRMP at the RIMS/GIS Wing of FD. What more are needed include: a) staff to operate and maintain the system, b) annual field check/enumeration/measurement of one-tenth to one-fifth of the one-minute grid plot clusters so that all plot clusters would have been re-visited/re-enumerated in five to ten years, c) annual/biennial acquisition of appropriate satellite imageries covering portions (sensitive portions) of the SRF to detect resource changes, d) ground monitoring system to check areas identified on the satellite imagery to have unusual activities/changes, and e) RIMS/GIS personnel to conduct necessary studies including strategies studies to manage and conserve the SRF and other forest resources for the maximum benefit of the people of Bangladesh.
List of Working Papers already released
APFSOS/WP/01 Regional Study - The South Pacific
APFSOS/WP/02 Pacific Rim Demand and Supply Situation, Trends and Prospects: Implications for Forest Products Trade in the Asia-Pacific Region
APFSOS/WP/03 The Implications of the GATT Uruguay Round and other Trade Arrangements for the Asia-Pacific Forest Products Trade
APFSOS/WP/04 Status, Trends and Future Scenarios for Forest Conservation including Protected Areas in the Asia-Pacific Region
APFSOS/WP/05 In-Depth Country Study - New Zealand
APFSOS/WP/06 In-Depth Country Study - Republic of Korea
APFSOS/WP/07 Country Report - Malaysia
APFSOS/WP/08 Country Report - Union of Myanmar
APFSOS/WP/09 Challenges and Opportunities: Policy options for the forestry sector in the Asia-Pacific Region
APFSOS/WP/10 Sources of Non-wood Fibre for Paper, Board and Panels Production: Status, Trends and Prospects for India
APFSOS/WP/11 Country Report - Pakistan
APFSOS/WP/12 Trends and Outlook for Forest Products Consumption, Production and Trade in the Asia-Pacific Region
APFSOS/WP/13 Country Report - Australia
APFSOS/WP/14 Country Report - China
APFSOS/WP/15 Country Report - Japan: Basic Plan on Forest Resources and Long-Term Perspective on Demand and Supply of Important Forestry Products
APFSOS/WP/16 Country Report - Sri Lanka
APFSOS/WP/17 Forest Resources and Roundwood Supply in the Asia Pacific Countries: Situation and Outlook to Year 2010
APFSOS/WP/18 Country Report - Cambodia
APFSOS/WP/19 Wood Materials from Non-Forest Areas
APFSOS/WP/20 Forest Industry Structure and the Evolution of Trade Flows in the Asia-Pacific Region - Scenarios to 2010
APFSOS/WP/21 Decentralization and Devolution of Forest Management in Asia and the Pacific
APFSOS/WP/22 Commentary on Forest Policy in the Asia-Pacific Region (A Review for Indonesia, Malaysia, New Zealand, Papua-New Guinea, Philippines, Thailand, And Western Samoa
APFSOS/WP/23 Asia Pacific Forestry Sector Outlook: Focus On Coconut Wood
APFSOS/WP/24 Ecotourism And Other Services Derived From Forests In The Asia-Pacific Region: Outlook To 2010
APFSOS/WP/25 Technology Scenarios in the Asia-Pacific Forestry Sector
APFSOS/WP/26 In-Depth Country Report - India
APFSOS/WP/27 People and Forests: Situation and Prospects
APFSOS/WP/28 Non-Wood Forest Products Outlook Study for Asia and The Pacific: Towards 2010
APFSOS/WP/29 Opportunities for Forestry Investment in Asia and the Pacific Through Carbon Offset Initiatives
APFSOS/WP/30 Country Report - The Maldives
APFSOS/WP/31 Country Report - Vietnam
APFSOS/WP/32 Country Report - Nepal
APFSOS/WP/33 Country Report - The Philippines
APFSOS/WP/34 Regional Study on Wood Energy Today and Tomorrow in Asia
APFSOS/WP/35 The Status, Trends and Prospects for Non-Wood and Recycled Fibre Sources in China
APFSOS/WP/36 Outlook, Trends and Options with Special Reference to Legislation, Institutions and Capacity Building (A Review for Bangladesh, Bhutan, China, Myanmar, Japan and Vietnam) (Draft)
APFSOS/WP/37 Perspectives of Environmental Civil Society Organizations on Forestry in the Asia-Pacific Region: Outlook To 2010
APFSOS/WP/38 Country Report - Laos
APFSOS/WP/39 ICIMOD
APFSOS/WP/40(A) FAO Outlook Study On Wood Based Panels Production, Consumption and Trade in the Asia Pacific Region 1996 to 2010
APFSOS/WP/40(B) FAO Outlook Study On Wood Based Panels Production, Consumption And Trade In The Asia Pacific Region - 1996 To 2010 - China Section Study On China's Wood-Based Panel Market Outlook For The Years 2000-2010
APFSOS/WP/41 Scenarios For Extra- And Inter-Sectoral Developments Of Forestry Outlook Study For Asia And The Pacific
APFSOS/WP/42 Country Report - Forestry Of Mongolia
APFSOS/WP/43 Statistical Profile
APFSOS/WP/44 Urban Forestry
APFSOS/WP/45 Country Report - Indonesia
APFSOS/WP/46 Country Report - Thailand
APFSOS/WP/47 Country Report - Papua New Guinea
APFSOS/WP/48 Country Report - Bangladesh
