Description and Composition of Tree Species in a Tertiary Institution Agricultural Faculty Arboretum, Ibadan, South-West Nigeria

: The Arboretum of the Agricultural Faculty of a tertiary institution in Ibadan is known for its rich diversity of trees. Therefore, the study investigates the tree growth variables in the arboretum such as diameter at breast height (dbh), diameter at the base, middle and top of the bole, total height, merchantable height and crown diameter. The basal area and volume were then calculated per species and per family. Several models were fitted for the height – diameter relationship and crown diameter – diameter relationship and crown diameter – dbh relationship. Positive linear relationships were observed among the growth variables. The fitted models showed that cubic models exhibit a more reliable function than quadratic and linear models for crown diameter – dbh predictions as it has R 2 above 0.75. Endangered species were observed too and this was indicated through the diversity index obtained. The highest basal area encounter belongs to myrtaceae family (9.61m 2 ) while the lowest belongs to pinaceae family (0.24m 2 ). The total basal area obtained at (31.72m 2 ) from the faculty trees indicates that they are exhibiting better growth and yield.

A tree inventory is a collection of data for description and analysis of the status, quantity, quality or product of a resource and provides information such as tree health, species, size and location; this information is used to generate reports that can help urban foresters in their strategic planning, such as the development of maintenance plans and management plans, as well as to help educate residents about their urban forest. Kangas et al., (2006) and Michael et al., (2008) is of the opinion that for sustainable forest management to be attained there is need for up-to-date forest inventories to assess the composition, structure, and distribution of forest vegetation that, in turn, can be used as base information for management decisions. Hence, this paper is aimed at describing the composition of tree species in a tertiary institution agricultural faculty arboretum in Ibadan, South-West Nigeria.

Study Area:
The study was carried out at the Arboretum in the Faculty of Agriculture and Forestry at The University of Ibadan. Oyo State, Nigeria.
Tree Growth Variable Measurement: Measurement was limited to woody plants of 20 cm diameter at breast height and above as done by FORMECU (1997) while identification was limited to woody plants of 10 cm diameter at breast height and above (Okali andOla-Adams, 1987, Swaine andHall, 1987). The following tree data were collected in the study area for further analysis: i. Diameter at breast height: Diameter at breast height is the stem diameter at a position of 1.3m above the ground level. ii.
Diameter at the base iii.
Diameter at the middle iv.
Diameter at the top v.
Crown length vi.
Total height of all trees using Spiegel relascop Data Analysis Basal Area Calculation: The basal area of each tree in the enumerated plots was calculated using the formula: BA = 2 4 ⁄ Where BA = Basal area (m 2 ), D = Diameter at breast height (cm) and л = pie (3.142).
Volume Calculation: The volume of each tree was calculated in every plot using the Newton's formula of Husch et al (1982).
However, all data collected are been subjected to descriptive statistics, regression analysis, analysis of variance etc. Using various statistical packages like excel, SPSS and Statistical.
Species diversity indices: Species diversity indices according to Dearth and Winterborn (1995), the choice of index is complicated by the fact that diversity comprises two main components namely species richness and specie evenness. However, the diversity indices were calculated from mathematical formula that takes into account of both species richness and relative abundance of each species in the community. Relative abundance refers to the number of individuals of a given species divided by the total number of individual of all species found.

= 100 ⁄
Where, RF is the relative frequency or abundance; ni is the number of individual in the entire population; N is the total number of the entire population Also, using Shannon Weiner diversity index to calculate habitat diversity. H = ∑ Where 'H' is the Shannon diversity index; ∑ is the total number of species in the habitat; Pi is the relative abundance i.e number of specie divided by the total number of individual in the habitat.
In is natural logarithm.
For specie evenness we used Magurran (1988)   also be deduced that the soil in this location supported the growth of both plantation and natural forest. The total families of trees encountered (table 2) were eight in all, from which Myrtaceae family dominated by having the highest frequency followed by Papilonaceaea, Meliaceae, Verbanaceae, Caesalpindaeae, Bignontaceae, Pinaceae and Sapindaceae with frequency of 18, 15, 7, 7, 5, 1, 1 and 1 respectively. The total tree volume encountered were 965.45m 3 while the total merchantable height was found to be 448.85m 3 and the total basal area were found to be 31.72m 2 . However, it was discovered that the least total volume belongs to Pinus caribea (0.39m 3 ) while the highest total volume belongs to Eucalyptus camadulensis (21.53m 3 ). The least merchantable volume were found to be 0.22m 3 and the highest merchantable belongs to Eucalyptus camadulensis with volume of 18.35m 3 , all these values were found to be in accordance with the findings of Adekunle (2000) who noted the least volume to be 0.42m 3 in the natural forest and further justify that the forest was supported by good soil that is rich in nutrients needed for plant growth.     (9) 7.288 0.0085 Figure 1 shows the graph of the relationship between total height, total merchantable height and total basal area for all the families encountered. The graph further reveals that the basal area were found to be significantly low compared to total height and total merchantable height. The highest basal area belongs to Myrtaceae family while the least belong to Pinaceae family. However, the height observed shows that there is variation in species and families observed, as it is normally observed in a typical uneven aged forest as this was also observed in Fig 2, which shows the graph of the relationship between mean total height, mean merchantable height and mean basal area. Figure 3 this shows the regression model for the relationship between the total volume and the merchantable volume. Meanwhile, the coefficient of determination (R 2 ) at 95% significant level shows that there is a strong relationship between total volume and merchantable volume with R 2 =0.76 and this explains how best the data fit in to the model also the equation obtained shows that the two volume were not statistically different from each other as similar findings were observed by Onyekwelu and Akindele (1995) who reported that there is a strong relationship between the total volume and merchantable volume that it is an indication of how well the data fit into the model and how suitable the model is for further use.  In addition, Fig 5 and 6 shows the regression equation between the total volume and basal area, and between the merchantable volume of the basal area. Meanwhile, the model proved very strong between total volume and basal area (R 2 =0.68) while there is a very weak relationship between merchantable volume and basal area (R 2 =0.35). However, this is an indication that the former is in accordance with the findings of Adekunle (2000) who reported that basal area is a function of volume which follows the assumptions of normality , also similar findings were observed by Onyekwelu and Akindele (1995) when predicted volume were validated for Gmelina arborea plantation in Oluwa forest reserve who reported that a very small bias value (less than 20%) obtained as this is an indication for the reliability of the model for further use. The regression for the relationship between the crown diameter and dbh (fig 7).  Table 6, this shows the model generated for the total height, merchantable height, total volume and the merchantable volume in equation 1,2,3, and 4 respectively for all the species. Model 1,3&4 indicates a significant different (p≥ 0.05) while model 2 shows that there is no significant different between the growth variables. However, based on the level of coefficient of determination (R 2 ) and standard error of estimates it can be seen that all the models generated exhibits a positive relationship and are linearly stable. This is an indication that all the variables fit well into the model and can also be deduced that the trees were well formed. Consequently, according to the value of statistics used to compare the models in the fitting phase model 4 was found to be more suitable for volume prediction Mvol = -3.37 +0.49 MHT+4.79BA while the coefficient of determination and standard error of estimates were found to be R 2 =0.78 and SEE =1.49 respectively. The diversity index carried out for the trees family encountered (table 7) shows that the highest diversity was found in Myrtaceae family followed by Papilonaceae, Meliaceae, Verbanaceae and Caesalpiniaceae (2.81, 2.70, 1.94,1.94 and 1.60) respectively and others with zero indices. From the diversity index it can be deduced that there is high diversity among five families while the others show zero diversity due to the fact that they have a very low frequency (1). However, the evenness carried out indicates that there is high evenness among the five families while the other three shows zero values. This was in accordance with the findings of Adeduntan et al (2006) who reported that the frequency and abundance is a function of diversity as this is an indication that the higher the frequency the higher the diversity level and vice-versa. The result of the correlation analysis carried out to examine the relationship among growth parameter is presented in table 8. There is generally a positive linear relationship between the variables. The highest correlation coefficient was obtained between diameter at the middle and diameter at the top (0.97) followed by db and dbh, db and dm, dbh and dm, THT and DT and CD (0.97,0.82,0.81,0.79,0.71,0.71 and 0.50) respectively. However, the least relationship existed between MTH and dbh (0.19). Consequently, this shows a positive relationship among all the growth variables as it was also supported by the findings of Adekunle (2000) who also noted a positive linear relationship between the growth variable measured in Omo and Ala forest reserves. In addition, this is an indication that tree growth variable exhibits a positive linear relationship when correlated. The models generation for the tree height and diameter distributions (table 9 -13) for the family of tress up at least five in frequency. The predication models used were linear, quadratic and cubic models respectively. However, for the five differently selected families all the models proved in adequate for myrtaceae and verbanaceae family. While such models were very adequate for meliaceae, papilonaceae and caesalpiniaceae families.  Considering the coefficient of determination and standard error of estimates it can be deduced that cubic regression models were seen to be very much adequate for the prediction of the height diameter relationships.

RESULTS AND DISCUSSION
Consequently, the work of Turan (2009) noted that cubic model gives best performance according to the value of test statistics (R 2 and SEE).     diameter at breast height relationships as be noted that cubic regression model is more suitable for crown diameter predictions. However, the regression models between dbh and crown diameter variables were found to be statistically significant (P<0.05) that the R 2 value is more than 0.75 (Turan, 2009) in the models estimated. This indicates that dbh and crown diameter. From the results of this study it was seen that dbh and crwon diameter could be estimated by means of dbh which is easy to measure for the studies in ground based forest inventory.     Conclusion: The result of the inventory carried out in the faculty of Agriculture and Forestry University of Ibadan reveals that there is no significant difference among the growth variables. Sustainability, as an objective of forest management demands that forest inventories be planned to look far beyond the assessment of marketable timber volume only. The assessment of a baseline data for the continuous monitoring of forest condition is absolutely essential. Therefore, as a way of recommendation, remote sensing through the use of modern technology should also be incorporated into forest inventory practical as this will widen the scope of the students and further strengthen their horizon.