Biophysical Spectral Response Modeling Approach for Forest Density Stratification
P.S. Roy, S. Miyatake* and A. Rikimaru*
Indian Institute of Remote Sensing, (NRSA),
Department of Space, Dehra Dun- 248001, India
Tel: (91) -135 – 744518 Fax : (91)
E-mail: [email protected]
*Japan Overseas Forestry Consultants Association
Rinyuu-Blgd : 1-7-12 Koraku, Bunkyo-Ku, Tokyo, Japan
Tel: 81-3-5689-3435 Fax : 81-3-5689-3439
E-mail: [email protected]
The present paper suggests biophysical spectral response modeling approach using a three way crown density stratification utilizing advanced vegetation index, bare soil index, shadow index and temperature for the classification of forest crown density in two different study sites viz., evergreen forests of Andaman and dry deciduous forest of central India. The present study is part of field test of International Tropical Tmber Organization (ITTO) and Japan Overseas Forestry Consultants Association (JOFCA) project on Rehabilitation of logged over forest in Asia/Pacific region (PD 32/93 Rev 2 (F)).
Satellite remote sensing provides a mean to obtain a synoptic view of forest and their condition on real time basis. The standardization of ground sampling methods, understanding of spectral and temporal responses of vegetation, coupled with the recent advancements in the digital image processing techniques have brought about a profound acceptance of the application of satellite remote sensing data in forest inventory and mapping (Malingreau, 1991). Forest density expressing the stocking status constitutes the single major stand physiognomic characteristic of the forest. Satellite remote sensing provides a measure of the canopy closure. The percent crown closure is a measure of area occupation rather than stand density. However, it is an important parameter used to stratify the forests. Ver few attempts have been reported to stratify the forest density using satellite remote sensing digital data (Roy et al. 1990).
International Tropical Timber Organisation (ITTO) and Japan Overseas Forestry Consultants Association (JOFCA) while working on project entitled Utilization of Remote developed a methodology wherein biophysical spectral indices were developed to stratify forest density (Anon., 1993 and Rikimaru, 1996). In the present study the methodology has been validated in two different Indian sites viz, evergreen forests of Andaman and dry deciduous forest of central region.
2. Study area
Two study sites having different forest types, structure and undergrowth conditions have been selected for the present study.
2.1 Test area in Evergreen forests
South Andaman Forest Division situated in Andaman islands is one of the test area. The terrain of the area is irregular and undulating. The main hill range runs through north and south, however, minor ridges run in all directions. The slope ranges from steep to moderate. The climate is wet tropical and is characterized by high rainfall (2750-3400 mm), high relative humidity (63-91%) and equable temperature (21-33oC). The forests are mainly of Andaman tropical evergreen forest, Andaman semi evergreen forest, Andaman moist deciduous forest, Tidal mangrove forest and Bamboo brakes (Champion and Seth, 1968).
2.2 Test area in Dry deciduous forests
Shivpuri is the second test site for the present study. It is located in central India occupying the Vindhyan landscape. The climate of the area is hot and dry. It is classified under semi-arid to arid region. The annual rainfall is 895.5 mm. The relative constitutes low hills, plateau and plains, mostly of the upper Vindhyan. The elevation varies between 250 m to 510 m above msl topography. The forests are mainly of northern tropical dry and moist deciduous forest (Champion and Seth 1968).
3 Materials and methods
3.1 Data used
Landsat Thematic Mapper data (Path-Row 134-052) of March 1984 has been used for the digital analysis of evergreen forests of South Andaman Forest Division. The aerial photographs of 1:64,000 scale of December 1968 were also used during ground data collection. The stratified random sampling approach has been the basis of distributing sample points in the study area. Thirty sample plots of 0.1 ha. Size have been selected in various forest cover types of the study area. The total enumeration was done in the sample plots, for parameters like, crown density, height (stratwise), number of trees/species, and basal areas were measured.
Landsat Thematic Mapper (TM) data (path-Row 146-042 ) of January 1990 has been used for the digital analysis of dry deciduous forest stock maps prepared by the state forest departments have been used as reference.
The digital image processing for evergreen forests of South Andaman Forest Division has been done using PC based MAI (Modular GIS Advanced Imager) of Intergraph package on Windows NT. The Landsat TM bands TM bands (except bands 6) were normalized using linear transformation. The temperature calibration using coefficients for Landsat 5 was done to estimate ground temperature. The temperature data has only been used to separate soil and non-tree shadow. The colour images produced from Landsat TM raw bands 4,3,2 and 5,4,3 provide valuable information on the forest cover type distribution.
The digital image processing for dry deciduous forests of central India has been performed on IBM RS-6000 series using EASI PACE software. The spectral dataset is subjected to physical transformation using enhancement techniques. Attempts have been made to isolate vegetation cover, soil background influence and canopy shadow been made to isolate vegetation cover, soil background influence and canopy shadow
pattern from Landsat TM data. The vegetation feature space data was stratified based on the ‘texture’ of the data as influenced by the canopy shadow. Finally a rule based logic is implemented to achieve land cover and forest density classification. In both the case the results have been improved by using water mask from Landsat TM band7.
The process and steps involved in calculating Biophysical Spectral Indices are given as below (Anon., 1993 and Rikimaru, 1996):