Distribution of apple grown area in China. The regions without colour indicate non-apple grown areas. doi:10.1371/journal.pone.0038883.g001 

Context in source publication

Context 1

... A vector ( B i ) was used to assign weighting factors for each subset in U i . The values in the vector reflect the contribution of the factors in the overall assessment for orchard i . The elements in B i include: weighting factors of long residence woody, leaf, fruit, fine root, pruning, soil respiration, irrigation and fertilizer application in order. The matrix V i derived from the product of the matrix and its corresponding vector was the outcome of the C capture and emission from the orchard i : The measurement of the potential assessment factors described were as below. Three representative apple trees from each orchard were identified in September, 2009 and the increments in stem diameter at 20 cm from the ground of the trees were measured in October, 2009 to October, 2010. Increment of stem diameter measurements one year apart were used to estimate the biomass increment for the experimental year. The trees were divided into leaves, branches, main stems and coarse roots, and all fresh weights were measured in the field. Then sub-samples from each fraction were dried to determine total dry matter of each fraction. The sub-samples were oven-dried at 65 u C until constant mass was reached. The dried sub-samples were ground after weighting before total C analysis was made by EA 1108 elemental analyser (Italy, Carlo Erba) to determine C content. The data were used to fit allometric equations to quantify the relationship between the biomass of different parts (leaf, stem, branch, coarse root) and stem diameter. Because of the life cycle of an apple tree, fruit production would begin to decline in the over-mature phase year by year. Therefore, a parabola equation was used to quantify the relationship between fruit production and stem diameter. Statistical analysis was made using SPSS for Windows (Rel. 11.5.0, 2002. Tokyo: SPSS Inc.). The turnover rate of fine roots is an important parameter to indicate the contribution of the fine roots to C sequestration. Minirhizotron and soil coring method were used to estimate the turnover rate. In July, 2009 before the experiment started, nine 90 cm-long minirhizotron tubes of 5 cm diameter were inserted into the soil at 45 u angle with the horizon in the second quadrat of each orchard to allow fine roots to settle in the soil surrounding the tubes. Minirhizotron images were collected every ten days using a BTC- 10 minirhizotron microscope (Bartz Technology, USA) from October 2009 to October 2010. This generated 64 images from each minirhizotron tube. The length of fine roots was calculated for all the images using the I-CAP software. The collected information was used to calculate the length of fine roots (cm) per unit area. A fine root turnover index, defined as the ratio of fine root mortality in a year (cm cm-1) to initial fine root length (cm cm ) within the minirhizotron window [26], was calculated using the data with the reported method. Within a quadrat from each orchard, ten soil cores of 4 cm in diameter and 80 cm in depth were sampled. The soil columns were separated into depths 0–20, 20–40, 40–60 and 60–80 cm. The soil samples were transferred in plastic bags. Roots were manually picked out from the samples, washed and sorted ( , 2 mm), then they were oven-dried at 80 u C until constant mass were reached. Fine root dry matter density (mg DM cm 2 3 ) was estimated based on the relationship between the weight and the surface area of sampled fine roots, which was used to estimate the biomass of fine roots. Soil respiration (and temperature) at 10 cm soil depth in the three quadrats in each orchards were measured every ten days between June 2009 and June 2010 (excluding first three months in 2010 when the top soil was frozen) using developed closed gas- exchange system (LiCor 6400 Portable Photosynthesis System with 6400-09 soil CO 2 flux chamber; LiCor, Lincoln, NE, USA). At all three orchards, 27 replicate LiCor soil collars in total were installed. The collars remained in place throughout the experiment period, allowing repeated measurements. All measurements were carried out between 10:00 to 11:30 am. Measurements were not made on days of rain. At the beginning of September 2010, 20 g fresh leaves collected from the apple trees was put into a nylon bag with a mesh size of 2 mm. Six bags were randomly placed on the surface of the orchard soil (after removing the litter layer). At the end of each month for the following three months, two litterbags were collected from each orchard to calculate weight loss and C content of litterfall. The data was used to determine decomposition rate by fitting the exponential function [28]: where X 0 is initial C content of the litterfall (g C), X t is C content at time t and k is the decomposition rate (d 2 1 ). Data processing and allometric equations. There are three geographical regions for apple production in China: the western, the central and the eastern regions. This study covered major apple regions in China (Figure 1), biomass data of apple trees at various ages from geographic locations was used to validate the allometric equations described in 2.2. The dynamics of the net C sink and C storage of apple orchards in China was calculated based on cultivation areas and tree age groups (Figure 2). The groups were set 0–7, 8–18 and 19–30 years old at national level. C capture of long residence woody, leaf, fruit. Carbon storage from a part of an apple tree is estimated based on dry matter and C content of the part. The average stem diameters at 20 cm from the ground were 3.2 6 0.2, 12.9 6 0.8 and 14.2 6 0.6 cm for the 5-, 18- and 22-year-old trees. The parameter values for the allometric equations to estimate dry matter from different parts of an apple tree are presented in Table 2. In order to validate the allometric equations which can be applied to the apple production regions in China, biomass data of an apple tree were collected dated from 1990 in major apple production regions (Table 3). The observed biomass and estimated values from the equations were compared (Table 4 and Figure 3). For the trees with stem diameter between 4 and 6 cm, the equations underestimated biomass of trees due to a large standard deviation in the collected data. For the other two groups at stem diameters of 0–4 cm and 10–15 cm, the estimated values confirm to the actual ones. These results indicated that the allometric equations should be suitable for biomass estimation of apple trees which were subjected to managed agricultural production systems. The annual biomass increments of living organs of apple trees at different ages were calculated using the equations. The increments of an individual tree were converted into the increments per unit land area through multiplying tree density. The results showed that the 5-year-old tree has a much higher growth rate than the other two groups although its standing biomass is low (Figure 4). The growth rates of all parts (except stem) for the trees older than 18 years began to slow down, which may be caused by reduced growth of a mature tree. As fruit trees get older, the proportion of the long residence woody biomass in total standing biomass production also increases. C capture and turnover of fine roots. The analysis of minirhizotron images indicated that the net growth rate of fine roots had apparent seasonal changes for the trees of all the ages. Two growth peaks appeared, in early summer and late fall for the 5- and 18-year-old trees throughout the observed soil profile. However, no significant seasonal change in growth rates was observed in the 22-year-old trees. The active growing zone of fine roots fell between 20 and 60 cm of the soil profile (Figure 5). Based on the observation and the regression analysis between fine root weight and surface area ( W fr = 0.634 + 0.7689 S area , R 2 = 0.8119 and n = 5), the annual growth rate of fine roots was calculated as 33.4 6 16.8, 41.7 6 19.0, and 17.7 6 6.8 g DM m 2 2 for the 5-, 18- and 22-year-old trees, respectively. The indices of annual fine root turnover were 7.7, 6.8 and 1.5 for the 5-, 18- and 22-year-old trees, respectively. The proportion of appeared and disappeared fine roots in the minirhizotron window for the 5-year-old trees remained at a similar value for the longest period compared to those with other two ages (Figure 6), which indicated that the 5-year-old trees not only produced a large amount of fine roots, but their root systems also had a high metabolic rate. C emissions of pruning and soil respiration. In China, each individual orchard follows its own guidelines for the disposal of pruned branches based on the age of trees. In general, more branches are pruned with tree ages. Biomass in pruned branches at different ages was shown in Table 5. Soil respiration from the 18-year-old orchard showed the strongest seasonal changes among the orchards, and two peaks in a year existed (Figure 7). We observed that the highest rate for the 5-year-old orchard occurred in spring and early summer, and that for the 22-year-old orchard occurred in the middle of summer. Based on the samples, annual soil respiration rates were 1.3 6 0.3, 1.6 6 0.6 and 1.2 6 0.5 kg C m 2 2 in the 5-, 18- and 22-year-old orchards, respectively. C emissions of litterfall. The data from the litterfall bags showed that the decomposition of litterfall followed an exponential equation: The total decomposition rate was equivalent to 312 days of turnover time. i.e., complete decomposition of litterfall would occur within a year either through emission to the atmosphere as CO 2 , or transformation into a stable organic matter pool in the soil. sequestration. Annual C increment rate for each part of the trees in an orchard was treated as a C sink. The value of the assessment elements related to tree biomass was calculated by the allometric equations, plant density and the conversion fraction of dry matter to C content. The fraction was 0.46 g C g 2 1 DM based on ...