Geo-informatics modeling on tropical seasonal forest dynamics in Mae Klong Watershed Research Station, Kanchanaburi province

Name: Torlarp Kamyo

LCSH: Kasetsart University -- Dissertations. Ph.D. (Forestry) 2016

Classification :.LCCS: SD381.5

LCSH: Kasetsart University. Faculty of Forestry -- Dissertations

LCSH: Forests and forestry -- Data processing

LCSH: Rorests and forestry -- Mae Klong Watershed Research Station

LCSH: Geoinformatics

LCSH: Information storage and retrieval systems -- Forestry

Abstract: This study was conducted at Mae Klong Watershed Research Station, MKWRS, during 2011.Three objectives were set up as 1) to classify vegetation cover based on multi date variation in the MKWRS using satellite imagery interpretation, 2) to clarify the forest structure and species composition, included distribution pattern of dominant species in the abandoned areas and 3) to establish a spatial model for suitability site identification of dominant species. The normalized difference vegetation index, NDVI, and different normalized difference vegetation index, DNDVI, analysis were used for vegetation cover changes based on satellite images of Landsat 5 TM which 4-year intervals from 1992 to 2012. While, 16-ha permanent plot was established in 2011 and species investigation was done to clarify the rest objectives. The result shows that the changes from degraded forest into forest areas fluctuated and varied among years (7.85±7.35 km2 4y-1 ). During the first 4-year period (1992 to 1996), degraded forest areas rapidly recovered to forest areas (61.33 % of total changed). Some changed areas from forest into degraded forest were also detected (4.2±1.27 km2 4y-1 ), indicating forest areas were still being disturbed. High species number was found, 176 species in 120 genera and 47 families, in 16-ha plot since abandoned in 1985. Negative exponential growth from based on diameter class distribution was detected. Indicating good regeneration process was detect. Dominance family based on number of species was Fabaceae, Euphobeaceae and Moraceae, respectively. Lagerstroemia tomentosa, Garuga pinnata, and Toona ciliata were the dominant species based on basal area (BA) and constituted 8.14 % of total BA. Distribution patterns based on morisita (Iδ) analysis across life stages showed that most species in saplings stage had clumped pattern, 92.59 % of total 54 species. The reduction in spatial clumping in going from younger to older life stages and the neutral or negative sapling-adult association imply density dependent mortality in the neighborhood of adult trees. In addition, the spatial distribution model based on logistic regression of 23 dominant tree species showed that it varies among species up on topographic characters. Total 23 species had positive effected to spatial model (P<0.05), aspect (9 species), convexity (12 species), elevation (6 species) and slope (16 species). In contrast the rest species, 23 species, had negative effected, aspect (14 species), convexity (11 species), elevation (17 species) and slope (7 species).Indicting, spatial distribution model is a useful tool not only predict where tree species can be coexisted in the natural forest but also apply to select the suitable species for forest restoration.

Kasetsart University. Office of the University Library

Address: Bangkok

Email: tdckulib@ku.ac.th

Name: Dokrak Marod

Role: Thesis Advisor

Name: Sura Pattanakiat

Role: Thesis Co-Advisor

Created: 2016

Modified: 2023-06-14

Issued: 2023-06-14

วิทยานิพนธ์/Thesis

application/pdf

URL: https://kasetsart.idm.oclc.org/login?url=https://www.lib.ku.ac.th/KUthesis/2559/torlarp-kam-all.pdf

CallNumber: SD381.5.T67

eng

DegreeName: Doctor of Philosophy

Descipline: Forestry

Grantor: Kasetsart University

Level: Doctoral Degraee

©copyrights Kasetsart University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	torlarp-kam-all.pdf	6.44 MB

ใช้เวลา

0.040198 วินาที