- Tytuł:
- Non-Linear Regression Models for Volume Estimation of Gmelina arborea (Roxb.) in Uyo Ravine Plantation, Akwa Ibom State, Nigeria
- Autorzy:
-
Oyebade, B. A.
Aigbe, H. I.
Eguakun, F. S.
Edem, M. A. - Powiązania:
- https://bibliotekanauki.pl/articles/1030779.pdf
- Data publikacji:
- 2020
- Wydawca:
- Przedsiębiorstwo Wydawnictw Naukowych Darwin / Scientific Publishing House DARWIN
- Tematy:
-
Gmelina arborea
Models
Non-linear Regression
Volume estimation - Opis:
- This research elucidates non linear relationship for volume estimation of Gmelina arborea in Uyo Ravine plantation, Akwa Ibom State Nigeria. Series of functional models were developed and the estimates of measurement of stand parameters such as diameter at breast height (DBH), stump diameter (Dst), total height (THT), merchantable height, (MTH), merchantable length (MLT) were used for modeling procedures for best fit models for stand volume estimation. Twenty temporary sample plots of 20m x 20m were randomly established without replacement and all trees in each plot were measured. Quantitative data collected were subjected to correlation and regression analyses for determination of empirical relationship between the growth variables. The developed models for volume estimation were evaluated by confirming the goodness of fit of the model and the statistical significance of the parameters using statistical relevant fit indices and criteria. The results of the study showed there were significant associations between the Gmelina arborea growth characteristics both at the individual and stand levels. The correlation analyses revealed a significant association with coefficient of correlation (r) ranging from 0.576 – 0.836 among the Gmelina arborea growth characteristics at the stand level while the association between the volume and diameter at breast height gave r-value of 0.978 which was very significant P<0.05. Similarly, the results of the study in testing for relationship among the growth parameters using volume models showed that the non linear regression volume models significantly fulfilled the criteria for model selection or goodness of fit among cubic, logarithm and quadratic non linear regression models with their coefficient of determination (R2) ranging from 0.640-0.881 and low values of standard error of the estimate (SEE) at P< 0.05. These models were recommended for volume estimation of Gmelina arborea in the study area and other Gmelina arborea plantations for effective plantation forest management in Nigeria.
- Źródło:
-
World Scientific News; 2020, 145; 46-61
2392-2192 - Pojawia się w:
- World Scientific News
- Dostawca treści:
- Biblioteka Nauki
Artykuł