Home > Geospatial Application Papers > Environment > Conservation & Monitoring


Forest working plan in new millennium


Enumeration of growing Stock (Stock Maps)



Project objectives
The primary objectives of the present study are
  1. Develop methodologies for creation of forest administrative maps in GIS environment.
  2. Develop methodologies for rapid monitoring of forest vegetation at Mouza/Compartment level, including generation of classified forest stock maps on real time basis.
  3. Developing in-house capabilities for use of computer-based GIS and Satellite Imageries in preparation of updated Working Plans of various Forest Divisions.
The secondary objectives of the project are
  1. Use of GIS tools for evaluating the success of Joint Forest Management in South West Bengal.
  2. Monitoring the bio-diversity status of the National Parks and Sanctuaries of the State.
  3. Generating database to substantiate the principles of forestry operation in South West Bengal and North Bengal.
Methodology
One GIS Cell has been made operative in the office of the Conservator of Forests, Working Plan and GIS from the July, 1999 onwards. The complete GIS solution package of Arc/Info and Satellite Imagery data processing software ERDAS have been installed on Windows NT Platform.

ARC/INFO GIS Software on Windows NT and ERDAS image processing software, professional version on Windows NT were used for the project.

Though the basic forest stock maps are prepared at the Mouza level (1:3960) in South West Bengal, or forest Block/Compartment level (1:15840) in North Bengal, these maps are not geo-referenced. In order to geo-reference, these maps, the following strategy was adopted
  1. The Police Station (PS)/ Block maps, in 1:63,360 scale, were procured from BLLRO/DLLRO. These maps have latitude/ longitudes reference and contain the Mouza boundaries, as well as information on road/rail network and rivers. These maps were digitised and the three coverages containing Block / Mouza boundaries, road/rail and rivers were created under geographic projections.

  2. These coverage were then projected into real world co-ordinates. [ Projection Polyconic, with spheroid Modified Everest, and Central Meridian / Projection Origin at 88° / 24° ].

  3. Topology was built for the coverage, containing Mouza boundary after correcting the digitisation errors and information on each mouza was attached to the database after adding the relevant items.

  4. IRS ID, LISS 3 Satellite imageries, pertaining to the period December, 1998 was obtained from NRSA, Hyderabad on CD and a sub set for the Area of Interest was created through ERDAS image processing software.

  5. The satellite imagery was geometrically rectified with reference to the geo-referenced, digitised Road / Rail coverage and River coverage.

  6. Various options of image enhancement techniques were tried out to get the best image for visual interpretation. A number of Ground Control Point (GCPs) were chosen for Ground Truth Verification.

  7. Supervised classification, using maximum likelihood classifier algorithm, was carried out with the help of collected Ground Truth Information (GTI) for the Area of Interest. The final classified output contained vegetation classification on the basis of species (Sal, eucalyptus, cashew) and density (more than 40% = Dense, 40% to 10% = open, less than 10% = Degraded), age (young Sal coppice and matured Sal), agriculture, water bodies, habitations and wasteland.

  8. The classified raster image was converted into a grid theme using image to grid utility and then the grid theme into a poly coverage (vector file) using grid to poly utility in ArcInfo.

  9. The classified output vector coverage was overlayed on to digitised PS/Block coverage containing Mouza boundaries. Each mouza, with superposed classified vegetation output containing forest types, spp, wasteland and habitations, was extracted and inter-mouza boundary junction points were identified and 4 to 5 such control points were created for this coverage.

  10. The individual mouza maps were scanned.

  11. The scanned image was geo-referenced with the existing PS/Block coverage, using the known control points (mouza boundary intersections).

  12. The individual mouza maps containing plot boundaries were digitised with the registered scanned maps as back environment.

  13. Topology was built and relevant data for the mouza added.

  14. The individual mouza coverage was overlayed on the polygon coverage obtained from the classified image using the Intersect Command in ArcInfo and a new mouza map was created as the updated stock map.
Page 3 of 6
| Previous | Next |