Application of Adaptive Cluster Sampling for Non-timber Forest Products Assessment in Training and Model Forest, Faculty of Forestry, National University of Laos

Main Article Content

Khamphet Phomphoumy
Weeraphart Khunrattanasiri
Santi Suksard

Abstract

The aim of this study were to assessment the quantity of non-timber forest prodwcts namely: rattan (Calamus vimina|is willd.), sugar pqüm (Tao) (Armnga westeòhou\ii Griff.), and  Hem (Goscinium fernestratum Colebr.) by employed adaptive cluster sampling (ACS) and simple random wampling (SR), The size of sample area sas fixed at 1.25 percent of the total area. The sampling plot was a rectangle with measuring sire of 20 m x 25 m. THe sampling without replacement consist 100 initial sampling plots were used tï collect data. Statistical methods used in the analysir were frequency, percent, mean, variance, sampling errors, coefficaent variation.


The result of ACS study, there"were 57 network consisting 179 plots in which 224 clumps or 1,085 culms of rattan were found, making up 38.14 percent. The average value was 176.69 culms per ha and its variance value was 1.46. For Tao, there are 152 clumps or 667 culms, which was about 23.44 percent. The average value was 164.69 culms per ha and its variance value was 5.32. For Hem, there were 198 clumps or 1,093 culms with the percentage of 38.42. The average value was 214.31 culms per ha and its variance value was 5.57. The result of SRS consisted of 100 plots in which 115 clumps or 539 culms of rattan were found, making up 40.31 percent. The average value was 171.13 culms per ha with the variance value of 42.77. The analysis also revealed that 71 clumps or 331 culms of Tao were found, making up 24.76 percent. The average value was 165.50 culms per ha with the variance value of 47.18. The analysis also found  89 clumps or 467 culms of Hem, making up 34.93  percent. The average value was 203.06 culms per ha with the variance value of 41.73.


The relative efficiency (RE) of ACS and SRS in the assessment of the amount of NTFPs of three types found that RE was greater than 1 indicating that the estimator derived from the Horvitz-Thompson, which primary random sampling was done with SRS under ACS was more efficient in term of accuracy than the traditional sampling of SRS.


 


Keywords: Adaptive cluster sampling, Non-timber forest products, Training and model forest

Article Details

How to Cite
Phomphoumy, K., Khunrattanasiri, W., & Suksard, S. (2022). Application of Adaptive Cluster Sampling for Non-timber Forest Products Assessment in Training and Model Forest, Faculty of Forestry, National University of Laos. Thai Journal of Forestry, 33(1), 36–46. Retrieved from https://li01.tci-thaijo.org/index.php/tjf/article/view/255403
Section
Original Articles