Compared to regressions obtained using the traditional logarithmic transformation method, the above regressions had a much better fit. The original biomass data of the sample trees were also used to calculate the proportion of total biomass apportioned to trunk, branches, leaves and roots.
JOURNAL OF SCIENCE OF HNUE Chemical and Biological Sci., 2013, Vol 58, No 9, pp 91-103 This paper is available online at http://stdb.hnue.edu.vn ALLOMETRIC RELATIONSHIP AND BIOMASS PARTITION OF Kandelia obovata LIU & YONG PLANTED IN NAM DINH PROVINCE Pham Hong Tinh1 and Mai Sy Tuan2 Mangrove Ecosystem Research Center, Hanoi National University of Education Faculty of Biology, Hanoi National University of Education Abstract Measurements were made of 101 Kandelia obovata Liu & Yong trees growing in Giao Lac Commune, Giao Thuy District, Nam Dinh Province, and, based on the diameter (cm) at 30 cm above the widening of the trunk base, allometric regressions were done to estimate the biomass (kg) of the entire tree, above-ground biomass, trunk biomass, leaf biomass and below-ground biomass, found to be W = 0.10316D 1.85845 (R2 = 0, 86), W = 0.09012D 1.78752 (R2 = 0, 84), W = 0.04975D 1.94748 (R2 = 0, 79), W = 0.00899D 1.7643 (R2 = 0, 63) and W = 0.01420D 2.12146 (R2 = 0, 73), respectively Compared to regressions obtained using the traditional logarithmic transformation method, the above regressions had a much better fit The original biomass data of the sample trees were also used to calculate the proportion of total biomass apportioned to trunk, branches, leaves and roots While for small trees (D ≤ cm), 20% is leaves, 20% is branches, 35% is trunk and 25% is roots, for larger trees (D ≥ 10 cm), only about 5% is leaves, 5% is branches, 65% is trunk and 25% is roots The allometric equations and biomass partition obtained in this study could be used effectively to account for biomass and carbon in Kandelia obovata Liu & Yong mangroves Keywords: Allometric relationship, Kandelia obovata Liu & Yong, Biomass, Vietnam Introduction Mangrove forest is an important ecosystem in tropical and subtropical coastal regions However, this ecosystem is being degraded and damaged at an alarming rate by human activites, e.g land use change, unsustainable exploitation and human population increase [6] Moreover, the value of mangrove forests in terms of carbon accumulation and other factors are not well understood Received September 26, 2013 Accepted December 18, 2013 Contact Pham Hong Tinh, e-mail address: phamtinhsp@yahoo.com 91 Pham Hong Tinh and Mai Sy Tuan Kandelia obovata Liu & Yong is a key species that has been planted in coastal areas of Northern Vietnam, mainly Quang Ninh, Nam Dinh and Thai Binh Provinces In a natural development and ecological succession of mangrove forest, this species and others grow together in national mangrove forests [2] Therefore understandings ecological characteristics of this species such as biomass and carbon accumulation could significantly contribute to sustainable management of the mangrove forests Scientists have developed a number of methods to estimate biomass (dry weight) of both inland and mangrove forest These methods are divided into three categories: 1) the harvest method, 2) the mean-tree method and 3) the allometric method The harvest method is not practical for measuring mature forests of the large amount of time and manpower needed Moreover, the harvest method is not reproducible because the trees are destroyed The mean-tree method is utilized only in forests that are relative homogeneous in terms of tree size The allometric method uses allometic equations to estimate the whole and partial weight of a tree referring to measured tree dimensions, including trunk diameter and height This method is useful for estimating temporal changes in forest biomass by subsequent measurements (Komiyama et al.) [5] However, allometric equations are site- and species-specific Little research done using allometric equations to determine Kandelia obovata Liu & Yong biomass has been published or cited The research presented in this paper is one of the first studies in Vietnam that makes use of allometric equations to determine the total and component biomass of Kandelia obovata Liu & Yong In addition, we compared two ways of building allometric equations, i.e determining parameters of equations In term of carbon accumulation, the roots of mangrove trees are probably the most important component However, an understanding of carbon accumulation in roots as well as in other components is limited because very few studies have been done on biomass proportion of mangrove trees Ong et al [7] have shown that in 20-year-old Rhizophora apiculata mangrove trees, carbon is accumulated in the roots at a rate of 0.42 tC/ha/year, whereas accumulation in the canopy is at a rate of 0.52 tC/ha/year We have not found published data on carbon accumulation and biomass proportion of Kandelia obovata Liu & Yong Therefore, in this paper we are the first to announce the biomass proportion of Kandelia obovata Liu & Yong in Vietnam The results of this study are expected to significantly contribute to biomass and carbon accounting for Kandelia obovata Liu & Yong mangroves in Nam Dinh Province as well as other areas of Northern Vietnam 2.1 Content Methodology 2.1.1 Study site The present study was carried out on Kandelia obovata Liu & Yong that was planted 10 - 20 years ago in Giao Lac Commune, Giao Thuy District, in Nam Dinh Province Giao 92 Allometric relationship and biomass partition of Kandelia obovata Lac is located at 200 13’-200 15’ North, 106015’-1060 30’ East (Figure 1) and is one of the buffer zones of Xuan Thuy National Park It is the first Ramsar Site in Vietnam In Giao Lac there are about 407.7 hectares of mangrove forest The Giao Lac mangrove trees received a lot of sediment from the Ninh Co and Red River and it is a relatively flat area with a thick layer of alluvial sediment Mangrove plantations may also be grown in clay mud and sand Giao Lac has a diurnal tide with sea level of 0.1 - 3.9 m, a temperature of 240 C, rainfall of nearly 1500 mm/year, air humidity of about 82% and salinity 18.0 28.3% Figure Location and map of the study site 2.1.2 Materials A sampling of 101 trees was chosen in the study site having diameters ranging from 0.5 to 15 cm The trees were cut down and the roots dug out in 2008, 2009 and 2013 to measure trunk diameter, total and component dry weight (above ground, stem, branch, leaf and below ground) All sample trees was selected using the stratified sampling method: 1) divide the study area into sections corresponding to the year the trees were planted (from 1998 to 2009); 2) randomly select two or three sample plots (10 × 10 m) within each compartment (a total of 34 sample plots were established) and 3) select three sample trees (smallest, largest and average) within each sample plot Before felling the sample trees, we measured and recorded the diameter at 30 cm above the widening at the base of the trunk The sample trees were cut and separated into four components: stems, branches, leaves and roots Then we weighed and recorded the fresh weight of each component We took about - 10 g samples of each component, and weighed and recorded the exact fresh weight of each sample The samples were then taken to the laboratory and dried to a constant weight The dry samples were weighed again to get the dry weight of each sample The ratio of fresh weight to dry weight of each sample was used to calculate the dry weight of each component The total dry weight of the stems, branches and leaves is the above ground biomass The total biomass of a tree is the sum of the dry weights of all four components The diameter and dry weight of the sample trees are presented in Table 93 Pham Hong Tinh and Mai Sy Tuan Table Diameter and dry weight of sample trees Dried weight W (kg) No 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 94 Diamter D (cm) 0.7 0.7 0.7 0.7 0.8 0.8 1.5 2.7 2.7 2.9 3.1 3.1 3.2 3.3 3.5 3.7 3.8 3.8 3.8 3.9 4.1 4.3 4.4 4.4 4.6 4.8 5.2 5.2 5.2 5.3 5.3 5.3 5.3 5.4 5.4 5.6 5.6 5.6 5.7 5.7 5.7 5.7 5.7 Total Above ground Trunk 0.02 0.55 0.03 0.08 0.02 0.02 0.02 1.41 0.23 0.67 0.30 0.15 0.19 0.29 0.39 0.49 1.24 1.21 0.99 0.92 1.68 0.87 1.21 1.58 2.13 3.07 1.76 2.85 1.17 2.45 3.19 3.87 0.57 2.78 2.14 3.25 2.25 2.92 1.92 3.46 2.81 2.43 3.15 0.02 0.51 0.02 0.04 0.01 0.02 0.02 0.83 0.13 0.59 0.21 0.14 0.18 0.24 0.29 0.32 0.73 0.76 0.90 0.75 1.14 0.77 0.90 1.13 1.60 1.55 1.24 1.61 1.11 1.74 1.72 3.27 0.47 2.15 1.70 2.37 1.79 2.20 1.57 2.58 2.28 2.00 2.22 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.19 0.08 0.10 0.12 0.14 0.17 0.19 0.17 0.21 0.51 0.33 0.28 0.23 0.59 0.45 0.51 0.62 0.88 0.81 0.70 0.93 1.08 1.40 1.03 2.64 0.33 1.81 1.44 1.90 1.01 1.01 1.08 1.66 1.15 1.51 1.41 Branches 0.46 0.03 0.43 0.01 0.02 0.07 0.03 0.04 0.01 0.59 0.29 0.23 0.28 0.32 0.36 0.45 0.40 0.32 0.49 0.18 0.15 0.33 0.04 0.16 0.13 0.18 0.60 0.78 0.29 0.50 0.76 0.36 0.58 Leaves 0.01 0.05 0.01 0.04 0.00 0.00 0.01 0.60 0.04 0.05 0.08 0.00 0.01 0.03 0.05 0.08 0.18 0.42 0.04 0.23 0.32 0.05 0.07 0.14 0.27 0.34 0.22 0.19 0.04 0.16 0.54 0.30 0.10 0.18 0.13 0.29 0.19 0.40 0.20 0.41 0.37 0.13 0.23 Below ground (roots) 0.01 0.04 0.01 0.04 0.01 0.01 0.01 0.58 0.10 0.08 0.09 0.01 0.01 0.05 0.10 0.17 0.47 0.44 0.09 0.17 0.52 0.10 0.30 0.43 0.50 1.49 0.49 1.21 0.06 0.69 1.43 0.57 0.10 0.60 0.41 0.87 0.44 0.71 0.35 0.82 0.53 0.38 0.91 Allometric relationship and biomass partition of Kandelia obovata 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 5.7 5.9 6.0 6.0 6.0 6.0 6.0 6.1 6.1 6.1 6.4 6.4 6.4 6.4 6.4 6.5 6.7 6.7 6.7 6.7 6.8 7.0 7.0 7.0 7.1 7.2 7.2 7.3 7.5 7.7 7.8 7.9 7.9 8.0 8.0 8.0 8.3 8.3 8.4 8.6 8.6 8.6 8.8 8.8 8.8 8.9 9.0 0.63 4.33 3.43 2.23 2.46 2.92 1.00 3.37 2.37 2.98 3.32 3.02 3.78 3.49 3.03 3.24 2.92 2.45 2.83 2.78 4.09 3.37 3.45 3.38 2.64 3.91 1.69 4.97 4.75 4.28 3.39 5.05 4.89 4.66 3.57 2.32 5.94 6.04 4.65 7.99 3.20 2.51 7.82 7.46 7.56 7.87 5.03 0.43 2.48 2.89 1.84 1.93 1.86 0.73 2.69 1.84 2.28 2.90 2.48 2.31 2.32 1.80 2.33 1.90 2.07 2.13 2.24 3.14 2.61 2.45 2.08 2.16 2.65 1.40 3.61 3.92 3.49 2.58 2.76 4.22 3.21 1.83 1.99 5.01 4.52 3.90 6.61 2.31 1.73 6.37 6.26 6.44 6.67 4.55 0.25 1.64 2.08 1.49 1.62 1.17 0.51 2.26 1.36 1.24 2.33 2.03 1.35 2.09 1.58 1.69 1.59 1.60 1.79 2.19 2.13 2.06 1.97 1.82 1.70 1.68 1.18 2.77 3.13 3.05 2.45 2.49 3.72 2.40 1.16 1.66 4.61 3.97 3.08 5.36 1.62 1.22 5.16 5.51 5.51 5.61 4.14 0.04 0.24 0.65 0.20 0.06 0.19 0.08 0.09 0.27 0.50 0.30 0.15 0.40 0.30 0.16 0.22 0.02 0.73 0.25 0.22 0.24 0.63 0.06 0.47 0.24 0.16 0.09 0.17 0.25 0.36 0.02 0.24 0.24 0.41 0.74 0.40 0.31 0.74 0.28 0.46 0.31 0.28 0.14 0.60 0.16 0.16 0.25 0.51 0.14 0.33 0.21 0.54 0.28 0.30 0.56 0.23 0.21 0.34 0.16 0.24 0.32 0.05 0.28 0.30 0.26 0.27 0.22 0.34 0.16 0.37 0.54 0.27 0.13 0.18 0.32 0.55 0.32 0.31 0.16 0.31 0.41 0.50 0.30 0.21 0.47 0.47 0.47 0.75 0.13 0.20 1.84 0.49 0.37 0.52 1.02 0.28 0.62 0.52 0.64 0.42 0.52 1.45 1.18 1.23 0.90 0.98 0.38 0.71 0.54 0.94 0.70 0.99 1.30 0.46 1.24 0.22 1.36 0.82 0.78 0.80 2.26 0.65 1.37 1.68 0.33 0.81 1.48 0.73 1.30 0.82 0.77 1.37 1.12 1.12 1.16 0.46 95 Pham Hong Tinh and Mai Sy Tuan 91 92 93 94 95 96 97 98 99 100 101 9.1 9.2 9.2 9.3 9.4 9.6 10.5 12.1 13.0 13.7 13.7 8.28 8.46 6.57 4.74 3.47 5.52 9.73 12.64 11.22 13.33 8.66 6.76 5.90 5.25 3.54 2.76 4.23 5.94 7.90 9.61 8.44 6.50 5.80 5.29 4.64 2.91 2.32 3.52 5.39 7.07 8.39 7.52 5.37 0.46 0.30 0.13 0.28 0.27 0.31 0.26 0.34 0.25 0.34 0.53 0.51 0.32 0.48 0.35 0.18 0.40 0.29 0.49 0.96 0.59 0.60 1.48 2.52 1.31 1.20 0.67 1.22 3.79 4.72 1.59 4.85 2.14 2.1.3 Allometry and variables of allometric equations Allometry is based on the fact that there is proportionality between the relative growth rates of two different parts of the plant [3] The relationship between the two variables can be expressed by the following equation: ( dy dx ):( )=k ydt xdt (1) where x is the independent variable (e.g stem diameter, tree height or both), y is the dependent variable (e.g biomass) and b and k are the allometric constants The equation can be simplified as follows: dy dx =k y x dx dy =k ⇔ y x ⇔ ln y + c1 = k (ln x + c2 ) (1) ⇔ ⇔ ln y = ln xk + ln b ⇔ y = bxk (2) In this study, we used diameter at 0.3 m height (D) as the independent variable because it is easy to measure in the field by using a caliper or tape measure Tree height was not used in the equation because measuring the height of mangrove trees does not in an accurate and it is time consuming In the latest guidelines on carbon assessment of mangroves, CIFOR (Center for International Forestry Research) recommended that the allometric equation should be developed and used with one independent variable – the diameter [4] 2.1.4 Allometric constant determination As the allometric equation form and variables of equations were determined, we used the statistical analysis software JMPIN with the input data presented in Table to 96 Allometric relationship and biomass partition of Kandelia obovata find the allometric constants b and k The equations obtained by JMPIN were compared with the corresponding equations developed using the traditional approach For the traditional method, the equation (2) was represented on logarithmic coordinates as a linear relationship of the form: lny = klnx + b (3) The measured date (Table 1) was also converted from arithmatic to logarithmic form in order to correspond with the equation (3) Then the least squares method was used to find the equation parameters b and k 2.1.5 Allometric equation evaluation The obtained allometric equations were evaluated using the the coefficient of determination (R2 ), sum squared error of estimate (SE) and mean error of estimate (bias) of the allometric equations developed by both approaches Bias, R2 and SEE were calculated by equation (4), (5) and (6) respectively R2 is a statistic that would give information about the goodness of fit of the allometric equations (i.e R2 is a statistical measure of how well the regression curve/line approximates the real data points) An R2 of indicates that the regression curves/lines fit the data perfectly SSE and bias are errors of estimates when using allometric equations compared with the true values N R2 = − i=1 N i=1 (yi − yi, )2 yi − N N yi i=1 2 (4) N (yi − yi, )2 SSE = (5) i=1 N bias = i=1 (yi − yi, ) (6) N where yi is biomass of ith tree measured in the field, yi′ is biomass of ith tree using the allometric equation and N is total number of sample trees 2.2 Results and discussions 2.2.1 Total biomass allometry Table Allometric equations for total biomass Log transformed Yes No Equation lnW = 2.31487lnD – 3.26991 (W = 0.03801D 2.31487 ) W = 0.10316D 1.85845 D (cm) N R2 SSE bias – 15 101 0.82 154.49 0.13 – 15 101 0.86 113.43 -0.07 97 Pham Hong Tinh and Mai Sy Tuan Table presents the allometric regressions obtained for total biomass with log-transformation and non-transformation The coefficients of determination (R2 ) for both equations are greater than 0.8 They indicate that a large proportion (over 80%) of the total biomass could be determined when referring to a diameter at a height of 0.3 m However, the errors of estimates (SE and bias) show that the non-transformed equation fits the data better than the log-transformed equation because their SSEs are 113.43 and 154.49 respectively, and their biases are -0.07 and 0.13 respectively Moreover, the non-transformed and log-transformed regressions fitting the actual data shown in Figure also indicate that the fit was better with the non-transformed data than with the log-transformed data For trees with a diameter less than cm, two equations overestimate the actual biomass For trees with a diameter greater than 10 cm, the log-transformed equation also overestimates the actual biomass, but the non-transformed equation yields a better estimate Figure Total biomass against diameter Figure Total above-ground biomass against of actual and two fitted models diameter of actual and two fitted models 2.2.2 Above-ground allometry Total above-ground biomass The allometric regressions developed for total above-ground biomass with log-transformation and non-transformation are presented in Table Table Allometric equations for total above-ground biomass Log transformed Yes No 98 Equation lnW = 2.33585lnD – 3.59103 (W = 0.02757D 2.33585 ) W = 0.09012D 1.78752 R2 SSE bias D (cm) N – 15 101 0.76 111.82 0.09 – 15 101 0.84 76.28 -0.06 Allometric relationship and biomass partition of Kandelia obovata The coefficient of determination (R2 ) for a non-transformed equation is 0.84 and for a log-transformed equation it is 0.76 This indicates that their proportions of total above-ground biomass when referring to a diameter at 0.3 m height are 84% and 76% respectively On the other hand, SSE and bias for the log-transformed equation are 111.82 and 0.09 while the values for the non-transformed equation are 76.28 and -0.06, respectively Thus, it is clear that the non-transformed equation fits the data better than the log-transformed equation A detailed consideration of the graphs of total above-ground biomass against the diameter of the actual and two fitted models (Figure 3) also indicates that for trees with a diameter greater than cm, the non-transformed equation fits the data better than the other Trunk biomass The allometric regressions developed for trunk biomass with log-transformation and non-transformation are presented in Table Table Allometric equations for trunk biomass Log transformed Yes No Equation lnW = 2.22714lnD – 4.26584 (W = 0.01404D 2.22714 ) W = 0.02485D 2.04924 D (cm) N R2 SSE bias - 15 101 0.76 67.66 0.19 - 15 101 0.80 60.83 -0.04 Figure Trunk biomass versus diameter Figure Branch biomass versus diameter of actual and two fitted models of actual and two fitted models Both equations have a coefficient of determination (R2 ) of around 80% indicating that about 80% of the trunk biomass could be determined referring to the diameter However, SSE and bias show that the non-transformed equation fits the actual data better than the log-transformed equations because SSE and bias for the log-transformed equation are 67.66 and 0.19 while the values for non-transformed equation are 60.83 and -0.04, respectively The graphs of trunk biomass against diameter of the actual and two fitted models (Figure 4) also indicate this for trees with a diameter that is greater than 10 cm For 99 Pham Hong Tinh and Mai Sy Tuan trees with a diameter less than 10 cm, both equations have an almost equivalent goodness of fit Branch biomass Branches were not separated from the total biomass in 15 of the 101 trees sampled The allometric regressions developed for branch biomass with log-transformation and non-transformation are presented in Table Table Allometric equations for branch biomass Log transformed Yes No Equation lnW = 1.35289lnD – 3.17169 (W = 0.02429D 1.35289 ) W = 0.00209D 2.56871 D (cm) N R2 SSE bias - 15 86 0.57 2.56 1.40 - 15 86 0.73 1.83 -0.38 The calculations of coefficient of determination (R2 ) and errors of estimates (SSE and bias) indicate that the non-transformed equation fits the actual data much better than the log-transformed equation The values for the non-transformed equation are 0.73, 1, 83 and -0.38, respectively, while the values for the log-transformed equation are 0.57, 2.56 and 1.40, respectively The graph of branch biomass against diameter of the actual and two fitted models (Figure 5) also clearly shows that the log-transformed equation does not fit the actual data well for all of the sampled trees; it overestimates actual biomass when trunk diameter is less than cm and underestimates actual biomass when trunk diameter is greater than cm Leaf biomass Table presents the allometric regressions obtained for total biomass with log-transformation and non-transformation Table Allometric equations for leaf biomass Log transformed Yes No Equation lnW = 1.250798lnD – 3.61637 (W = 0.02688D 1.250798 ) W = 0.00899D 1.7643 D (cm) N R2 SSE bias - 15 101 0.55 1.57 0.01 - 15 101 0.63 1.30 0.00 Both equations have coefficients of determination (R2 ) that are relatively low, 0.63 for the non-transformed equation and 0.55 for log-transformed equations This means that only about 63% of the leaf biomass can be determined when referring to trunk diameter at 0.3 m height when using the non-transformed equation and only about 55% can be determined using the log-transformed equation However, the non-transformed equation fits the actual data better than the log-transformed equation This conclusion is confirmed in light of a comparison between the errors of estimate (SSE and bias) and the graph of leaf biomass against the diameter of the actual and two fitted models (Figure 6) 100 Allometric relationship and biomass partition of Kandelia obovata Figure Leaf biomass versus diameter of actual and two fitted models Figure Below-ground biomass versus diameter of actual and two fitted models 2.2.3 Below-ground (root) allometry The allometric regressions developed for root biomass with log-transformation and non-transformation are presented in Table Table Allometric equations for root biomass D Log-transformed Equation N R2 (cm) lnW = 2.22579lnD – 4.57658 Yes - 15 101 0.71 (W = 0.01029D 2.22579 ) No W = 0.01420D 2.12146 - 15 101 0.73 SSE bias 20.28 0.08 19.34 -0.01 The result indicates that a rather small proportion (about 70%) of root biomass could be explained by trunk diameter (the coefficients of determination (R2 ) for log-transformed and non-transformed equations are 0.71 and 0.73, respectively) Errors of estimate (SSE and bias) (Table 5) and graphs of below-ground biomass versus diameter of actual and two fitted models (Figure 7) show that the non-transformed equation does not fit the actual data better than the log-transformed equation 2.2.4 General discussion on allometry for Kandelia obovata Liu & Yong The coefficients of determination for the allometric equations developed for Kandelia obovata, Liu & Yong, in this study are not high (R2 ≤ 0.86) However, small errors of estimate (bias ≤ 0.15) for all of the obtained equations confirm earlier observations that power curve or allometric function (y = bxk ) can accurately describes the relationship between biomass and stem diameter [9] and the relationship can be used to correctly estimate the biomass of trees based on stem diameter of mangrove species trees [7, 8] This study also confirms that the allometric constants (b and k) of allometric equations for mangrove trees could be effectively determined using a non-transformation approach and statistical software (e.g JUMPIN) The equations developed making 101 Pham Hong Tinh and Mai Sy Tuan use of non-transformation fit the actual data better than the log-transformed equations Baskervilles [1], Whittaker and Marks [9] explained the inefficiency of the log-transformation approach They concluded that logarithmic transformation requires input data to be converted from arithmatic form to logarithmic form and the input data conversion generates errors The amount of errors is proportional to the magnitude of converted values 2.2.5 Biomass partition of Kandelia obovata Liu & Yong Figure Biomass proportions in all components (a) and in above-ground components only (b) In this study, we calculated the biomass partition of Kandelia obovata Liu & Yong from actual measurements of total biomass and biomass composition of 101 sample trees The results are shown in Figure The proportion of total biomass (Figure 8a) is not consistent and is independent of the size of tree The variability regarding leaf and branch biomass is high for small trees (i.e trees with a diameter of less than 10 cm) Each branch and leaf component comprises no more than 20% of the total biomass for trees with a diameter less than cm and this fell to about 5% for trees 10 cm in diameter For root and trunk biomass, although there is great variability, on average about 35% of the total biomass consists of trunk matter and some 25% is root matter in trees less than cm in diameter The trunk biomass of trees with a diameter greater than cm reached 65% for trees greater then 10 cm in diameter Root biomass is relatively stable, even when trees which are quite large are included in the data The above observations indicate that the biomass apportioned to roots is relatively stable at about 25% and there is a high variability for trunk and leaf biomass (especially for trees less than 10 cm in diameter) This was again demonstrated by the calculation of biomass allocation in above-ground biomass components only, i.e the proportion 102 Allometric relationship and biomass partition of Kandelia obovata of above-ground biomass (Figure 8b) The variability in proportion of above-ground biomass is understandable because the above-ground components change in response to environmental conditions and/or environmental change In addition, our results for Kandelia obovata Liu & Yong are relatively consistent with the observations of Ong et al [6] on the biomass proportion of Rhizophora apiculata Conclusion The allometric equations for total and component biomass of Kandelia obovata Liu & Yong are thought to be (relatively) highly accurate Allometric equations determined using the non-transformation approach fit the actual data better than the equations determined by log-transformation Therefore, we recommend the non-transformation approach with statistical analysis software (e.g JMPIN software) to determine allometric constants of regressions For trees less than cm in diameter, both branches and leaves comprise about 20% of total biomass, the trunk comprises 35% and roots comprises 25% of the total biomass However, for trees which are more than 10 cm in diameter, the biomass proportion of branches and leaves drops to about 5% and the trunk biomass goes up to 65% The biomass proportion of roots is stable at around 25% The allometric equations and biomass partition obtained in this study could be used effectively to determine biomass and carbon stored in Kandelia obovata Liu & Yong mangrove trees in Nam Dinh Province and elsewhere in Vietnam REFERENCES [1] Baskerville G.L., 1972 Use of logarithmic regession in the estimation of plant biomass Canadian Journal of Forest Research 2, pp 49-53 [2] Hong P.N., Ba T.V., Nam V.N., San H.T Tang V.T., Tre L.T., Tri N.H., Tuan M.S., Tuan L.X, 1999 Mangroves of Vietnam Vietnam Agricultural Publishing House, Hanoi [3] Huxley J.S., 1972 Problems of Relative Growth (2nd ed.), Dover, New York [4] Kauffman J.B., Donato D., 2012 Protocols for the measurement, monitoring and reporting of structure, biomass and carbon stocks in mangrove forests CIFOR Working Paper 86, Bogor, Indonesia [5] Komiyama A., J.E Ong, S Poungtarn, 2008 Allometry, biomass, and productivity of mangrove forest: A review Aquatic Botany 89, pp 128-137 [6] Ong J.E., Gong W.K., C Wong.H., 2004 Allometry and partitioning of the mangrove Rhizophora apiculata Forest Ecology and Management 188, pp 395-408 [7] Ong J.E., Gong W.K., Clough B.F., 1995 Structure and productivity of a 20-year-old stand of Rhizophora apiculata Bl mangrove forest Journal of Biogeography 22, pp 417-424 [8] Putz F.E., Chan H.T., 1986 Tree growths dynamics Forest Ecology and Management 17, pp 211-230 [9] Whittaker R.H., Marks P.L., 1975 Methods of assessing terrestrial productivity of the biosphere Springer, New York 103 ...Pham Hong Tinh and Mai Sy Tuan Kandelia obovata Liu & Yong is a key species that has been planted in coastal areas of Northern Vietnam, mainly Quang Ninh, Nam Dinh and Thai Binh Provinces In a natural... the biomass proportion of Kandelia obovata Liu & Yong in Vietnam The results of this study are expected to significantly contribute to biomass and carbon accounting for Kandelia obovata Liu & Yong. .. determine the total and component biomass of Kandelia obovata Liu & Yong In addition, we compared two ways of building allometric equations, i.e determining parameters of equations In term of carbon