Keywords
Principal Component
Modeling
Volume
Forest Reserve
How to Cite
Abstract
The study evaluated the application of clustering and principal component analysis for modeling tree volume in Shasha forest reserve. Information on the present status of the resource is required for sustainable forest resource management. The data set used in this study was collected from 15 temporary sample plots of 30 m x 30 m dimensional square plots were established. Clustering Analysis was applied to distribute the observations into four different groups using k-means method. Four distance measures and methods were used to group the data from Shasha rain forest. The results of model selection indices on each cluster revealed that in cluster one, model three had the lowest model selection indices of RMSE (0.1404), MAE (0.0952), BIC (-10.2377) and AIC (-16.1279) when compared to models one, two and four (4). In cluster two, model three with RMSE (0.1932), MAE (0.1604), BIC (2.3226), AIC (-5.3092) was selected over models one, two and four as it had the smallest residuals of RMSE (0.1404) and MAE (0.0952). In cluster three, model four had the lowest model selection indices with RMSE (0.2642), MAE (0.203), BIC (14.0543) and AIC (10.3977) when compared to models one, two and three. In cluster four, model 3 had the lowest model selection indices with RMSE (0.2014), MAE (0.1440), BIC (6.2121) and AIC (8.1649) when compared to models one, two and four. The results of the model selection indices on the fifteen (15) sample plots showed that model three had the lowest model selection indices with RMSE (0.1992), MAE (0.1602), AIC (-23.8572) and BIC (-11.5277) when compared to models one, two and four. Hence, model three is most suitable for tree volume predictions in the study area. The application of this management tools in forest modeling aids in prescribing sustainable management strategies for shasha forest reserve.
References
Abayomi JA (1983). Volume tables for Nauclea diderrichii in Omo forest reserve, Nigeria. Nigerian Journal of Forestry, 1983, 13 (1–2): 46–52.
Aigbe HI, Akindele SO, Onyekwelu JC (2013). Volume equation for Oban tropical forest reserve, Cross River, Nigeria. Nigerian Journal of Forestry, 2013, Vol 43 Nos 1&2
Akindele SO, LeMay VM (2006). Development of Tree Volume Equations for Common Timber Species in the Tropical Rain Forest Area of Nigeria. Forest. Ecology and Management, 2006, 226: 41-48.
Avery ET, Burkhart HE (2002). Forest measurements Mc Graw Hill New York U.S.A 5th Edition. 2002, 456 pp
Bischof AL and Selb A (1999). MDL Principle for Robust Vector Quantization. Pattern Analysis and Applications, vol. 2, 59-72
Clutter JL, Fortson JC, Piennnar LV, Brister, GH, Bailey RL (1983). Timber management; A Quantitative Approach. John Wiley and Sons, New York, USA, 1983, 333pp.
Edminster CB, Beeson RT, Metcalf GE (1980). Volume tables and point-sampling factors for ponderosa pine in the Front Range of Colorado. U. S. Forest Service, Rocky Mountains Forest and Range Experimental Station, Research Paper RM-218, 1980, 14Pp
Husch B, Beers TW, Kershaw JA Jr (2003). Forest Mensuration. 4th Edn., John Wiley and Sons Inc., New Jersey, USA., 2003, 949 Pp
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J (2000). Biodiversity hotspots for conservation priorities. Nature, 2000, 403:853–858
Oliveira-Filho AT, Fontes MA (2000). Patterns of floristic differentiation among Atlantic forests in Southeastern Brazil and the influence of climate. Biotropica, 2000, 32: 793–810
Osho JSA (2016). Matrix model for tree population projection in a tropical rain forest of south-western Nigeria. Ecological Modelling, 2016, 59 (3/ 4), 247–255.
Phillips P, yasmani B, Van Gardingen Pr (2002). Grouping tree species for analysis of forest data in Kalimantan (Indonesia Borneo) Forest Ecology and Management, 2002, 157: 205 – 216Pp
Picard N, Bar-Hen A, Franc A (2002). Modelling forest dynamics with a combined maxtrix/individual-based model. Forest Science, 2002, 48: 643–52Pp
Plotkin Jb, Chave J, Ashton PS (2002). Cluster analysis of spatial patterns in Malaysian tree species. The American Naturalist, 2000, 160: 629–644Pp
Qin J, Cao QV, Blouin DC (2007). Projection of a diameter distribution through time. Canadian Journal of Forest Research, 2007, 37, 188–94.Pp
Jain MM and Flynn P (1999). Data Clustering: A Review. ACM Computing Surveys, vol. 31 (3): 264-323
Jain R and Duin JM (2000). Statistical Pattern Recognition: A Review. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22 (1): 4-37
Schumacher FX, Hall FS (1933). Logarithmic expression of timber-tree volume. Journal of Agricultural Research, 1993, 47 (9): 719-734.Pp
Souza and Souza dr (2006). Análise multivariada para estratificação volumétrica de uma floresta ombrófila densa de terra firme, Amazônia Oriental. Revista Árvore, 2006, 30: 49–54Pp
Trincado GV, Quezada PR, von Gadow K (2003). A comparison of two stand table projection methods for young Eucalyptus nitens (Maiden) plantations in Chile. Forest Ecology and Management, 2003, 180: 443– 51Pp
Zien A, Qui˜nonero J (2005). Candela. Large margin non-linear embedding. In Luc De Raedt and Stefan Wrobel, editors, Machine Learning, Proceedings of the Twenty-Second International Conference (ICML 2005), Bonn, Germany, August 7-11, 2005, volume 119 of ACM International Conference Pro-ceeding Series, pages 1060–1067. ACM, 2005.